journal of allergy and clinical immunology

THE JOURNAL OF

AllergyANDClinicalImmunologyVOLUME 116 d NUMBER 2

OFFICIAL JOURNAL OF THE AMERICAN ACADEMY OF ALLERGY, ASTHMA AND IMMUNOLOGY

The editors’ choice 239Donald Y. M. Leung, MD, PhD, Harold S. Nelson, MD, and Stanley J. Szefler, MD

Reviews and feature articles

Current reviews of allergy and clinical immunology

Innate immune responses to infection 241Michael F. Tosi, MD, New York, NY

Continued on page 7A

2005 American Academy of Allergy, Asthma and Immunology

About the cover

This month’s theme feature examines the fascinating interaction ofinfection and immunity. Our cover displays two exquisite images of theneutrophilic leukocyte’s response to infection in a mouse model. Theseleukocytes pass through the endothelium of blood vessels in response toa chemoattractant such as created by tissue infection. In the cover image,the left panel is a low power view of a small inflamed venule (pink) withnumerous leukocytes (green) adhering to the endothelial lining. Theright panel is a higher power view of a transmigrating leukocyte whosecell body lies beneath the pink endothelium, its amoeboid shape beingeasily appreciated. The leukocyte’s trailing tail or uropod (green) has notyet passed through the endothelium. Other articles in this issue that focuson the topic of infection and immunity are noted in the Table of Contentsby the ‘‘theme’’ icon.

Our sincere appreciation to Alan Burns, PhD, and C. Wayne Smith,MD, of the Departments of Pediatrics and Medicine, Baylor College ofMedicine, who have captured these EM scanning images and made themavailable to us for presentation on our cover.

Y This month’s theme: Infection and immunity

wCME

Y

The Journal of Allergy and Clinical Immunology (ISSN 00917-6749) is published monthly (12 issues per year) by Elsevier Inc., 360 Park Avenue South, NewYork, NY 10010-1710. Business and Editorial Offices: 1600 John F. Kennedy Boulevard, Suite 1800, Philadelphia, PA 19103-2899. Accounting andCirculation Offices: 6277 Sea Harbor Drive, Orlando, FL 32887-4800. Periodicals postage paid at Orlando, FL 32862 and additional mailing offices.POSTMASTER: Send address changes to The Journal of Allergy and Clinical Immunology, Elsevier Periodicals Customer Service, 6277 Sea Harbor Drive,Orlando, FL 32887-4800.

J ALLERGY CLIN IMMUNOL August 2005 5A

Contents

CONTENTS

CONTINUED

Continuing Medical Education examination: Innate immune responsesto infection

250

Molecular mechanisms in allergy and clinical immunology

EBV the prototypical human tumor virus—just how bad is it? 251David A. Thorley-Lawson, PhD, Boston, Mass

Continuing Medical Education examination: EBV the prototypicalhuman tumor virus—just how bad is it?

262

Editorial

Infection versus immunity: What’s the balance? 263William T. Shearer, MD, PhD, Houston, Tex

Asthma diagnosis and treatment

Rostrum

The role of rhinovirus in asthma exacerbations 267Samuel L. Friedlander, MD, and William W. Busse, MD, Madison, Wis

Perspectives in asthma

Perspectives on the past decade of asthma genetics 274Carole Ober, PhD, Chicago, Ill


dEC Editors’ Choice (p 239)

dOR Online Repository material

Y Theme issue

wCME CME examination article available online at www.mosby.com/jaci

The Journal of Allergy and Clinical Immunology posts in-press articles online in advance of their appearance in the

print edition of the Journal. They are available at the JACI Web site at www.mosby.com/jaci at the ‘‘Articles in

Press’’ link, as well as at Elsevier’s ScienceDirect Web site, www.sciencedirect.com. Each print article will

acknowledge the e-publication date (the date when the article first appeared online). As soon as an article is

published online, it is fully citable through use of its Digital Object Identifier (DOI). Please visit the JACI Web site

and view our hot-off-the-wire articles through the ‘‘Articles in Press’’ link.

Y

wCME

Y

Contents


CONTENTS

CONTINUED

Original articles

dEC Is it traffic type, volume, or distance? Wheezing in infants living neartruck and bus traffic

279

Patrick H. Ryan, MS, Grace LeMasters, PhD, Jocelyn Biagini, MS, David Bernstein, MD,Sergey A. Grinshpun, PhD, Rakesh Shukla, PhD, Kimberly Wilson, MS, Manuel Villareal, MD,Jeff Burkle, BS, and James Lockey, MD, Cincinnati, Ohio

dEC Effect of low-dose ciclesonide on allergen-induced responses in subjectswith mild allergic asthma

285

Gail M. Gauvreau, PhD, Louis Philippe Boulet, MD, Dirkje S. Postma, MD, PhD, Tomotaka Kawayama, MD,Richard M. Watson, BSc, MyLinh Duong, MD, Francine Deschesnes, BSc, Jan G. R. De Monchy, MD, PhD,and Paul M. O’Byrne, MD, Hamilton, Ontario, and Quebec City, Quebec, Canada, andGroningen, The Netherlands

Roflumilast, an oral, once-daily phosphodiesterase 4 inhibitor, attenuatesallergen-induced asthmatic reactions

292

Emmerentia van Schalkwyk, MBChB, K. Strydom, MBChB, Zelda Williams, RN, Louis Venter, MSc,Stefan Leichtl, PhD, Christine Schmid-Wirlitsch, PhD, Dirk Bredenbroker, MD, and Philip G. Bardin, FRACP, PhD,Cape Town and Rivonia, South Africa, Melbourne, Australia, and Konstanz, Germany

Duration of postviral airway hyperresponsiveness in children withasthma: Effect of atopy

299

Paraskevi Xepapadaki, MD, PhD, Nikolaos G. Papadopoulos, MD, PhD, Apostolos Bossios, MD, PhD,Emmanuel Manoussakis, MD, Theodoros Manousakas, MD, and Photini Saxoni-Papageorgiou, MD, PhD,Athens, Greece

Mechanisms of asthma and allergic inflammation

Dissecting asthma using focused transgenic modeling and functional genomics 305Douglas A. Kuperman, PhD, Christina C. Lewis, PhD, Prescott G. Woodruff, MD, Madeleine W. Rodriguez, BS,Yee Hwa Yang, PhD, Gregory M. Dolganov, PhD, John V. Fahy, MD, and David J. Erle, MD,Chicago, Ill, and San Francisco, Calif

Endobronchial adenosine monophosphate challenge causes tachykininrelease in the human airway

312

Fionnuala Crummy, MD, MRCP, Mark Livingston, PhD, Joy E. S. Ardill, PhD, FCRPath,Catherine Adamson, MSc, Madeleine Ennis, PhD, and Liam G. Heaney, MD, MRCP,Belfast, Northern Ireland, United Kingdom

Rat tracheal epithelial responses to water avoidance stress 318Hiroshi Akiyama, PhD, Hiroo Amano, MD, PhD, and John Bienenstock, MD, Tokyo andMaebashi, Japan, and Hamilton, Ontario, Canada

Allergen-induced substance P synthesis in large-diameter sensory neuronsinnervating the lungs

325

Benjamas Chuaychoo, MD, Dawn D. Hunter, PhD, Allen C. Myers, PhD, Marian Kollarik, MD, PhD,and Bradley J. Undem, PhD, Baltimore, Md


dOR

dOR


Contents

CONTENTS

CONTINUED

Differential effects of (S)- and (R)-enantiomers of albuterol in a mouseasthma model

332

William R. Henderson, Jr, MD, Ena Ray Banerjee, PhD, and Emil Y. Chi, PhD, Seattle, Wash

Rhinitis, sinusitis, and ocular diseases

Comparison of test devices for skin prick testing 341Warner W. Carr, MD, Bryan Martin, DO, Robin S. Howard, MA, Linda Cox, MD, Larry Borish, MD, andthe Immunotherapy Committee of the American Academy of Allergy, Asthma and Immunology, Silver Spring, Md,Fort Lauderdale, Fla, and Charlottesville, Va

Allergen-specific nasal IgG antibodies induced by vaccination with geneticallymodified allergens are associated with reduced nasal allergen sensitivity

347

Jurgen Reisinger, MSc, Friedrich Horak, MD, Gabrielle Pauli, MD, Marianne van Hage, MD, Oliver Cromwell,PhD, Franz Konig, Rudolf Valenta, MD, and Verena Niederberger, MD, Vienna, Austria, Stockholm, Sweden,Strasbourg, France, and Reinbek, Germany

Levocetirizine: Pharmacokinetics and pharmacodynamics in childrenage 6 to 11 years

355

F. Estelle R. Simons, MD, FRCPC, and Keith J. Simons, PhD, Winnipeg, Manitoba, Canada

Striking deposition of toxic eosinophil major basic protein in mucus:Implications for chronic rhinosinusitis

362

Jens U. Ponikau, MD, David A. Sherris, MD, Gail M. Kephart, BS, Eugene B. Kern, MD, David J. Congdon, MD,Cheryl R. Adolphson, MS, Margaret J. Springett, BS, Gerald J. Gleich, MD, and Hirohito Kita, MD,Rochester, Minn, Buffalo, NY, and Salt Lake City, Utah

Intranasal tolerance induction with polypeptides derived from 3 noncross-reactive major aeroallergens prevents allergic polysensitization in mice

370

Karin Hufnagl, PhD, Birgit Winkler, MD, Margit Focke, PhD, Rudolf Valenta, MD, Otto Scheiner, PhD,Harald Renz, MD, and Ursula Wiedermann, MD, PhD, Vienna, Austria, and Marburg, Germany

Environmental and occupational respiratory disorders

dEC Prevalences of positive skin test responses to 10 common allergens in theUS population: Results from the Third National Health and NutritionExamination Survey

377

Samuel J. Arbes, Jr, DDS, MPH, PhD, Peter J. Gergen, MD, MPH, Leslie Elliott, MPH, PhD, andDarryl C. Zeldin, MD, Research Triangle Park, NC, and Bethesda, Md

Airborne endotoxin in homes with domestic animals: Implications forcat-specific tolerance

384

James A. Platts-Mills, BA, Natalie J. Custis, BA, Judith A. Woodfolk, MD, PhD, andThomas A. E. Platts-Mills, MD, PhD, Charlottesville, Va


dOR

dORdOR


Contents

CONTENTS

CONTINUED

Food allergy, dermatologic diseases, and anaphylaxis

dEC COX-2 inhibition enhances the TH2 immune response to epicutaneoussensitization

390

Dhafer Laouini, PhD, Abdala ElKhal, PhD, Ali Yalcindag, MD, Seiji Kawamoto, MD, PhD,Hans Oettgen, MD, PhD, and Raif S. Geha, MD, Boston, Mass

dEC Responsiveness to autologous sweat and serum in cholinergic urticariaclassifies its clinical subtypes

397

Atsushi Fukunaga, MD, Toshinori Bito, MD, Kenta Tsuru, MD, Akiko Oohashi, MD, Xijun Yu, MD,Masamitsu Ichihashi, MD, Chikako Nishigori, MD, and Tatsuya Horikawa, MD, Kobe, Japan

Lack of detectable allergenicity of transgenic maize and soya samples 403Rita Batista, BSc, Baltazar Nunes, MSc, Manuela Carmo, Carlos Cardoso, PharmD, Helena Sao Jose,Antonio Bugalho de Almeida, MD, PhD, Alda Manique, MD, Leonor Bento, MD, PhD, CandidoPinto Ricardo, PhD, and Maria Margarida Oliveira, PhD, Lisboa, Oeiras, and Alges, Portugal

Basic and clinical immunology

Advances in Asthma, Allergy, and Immunology Series 2005

Basic and clinical immunology 411Javier Chinen, MD, PhD, and William T. Shearer, MD, PhD, Bethesda, Md, and Houston, Tex

Current perspectives

The gastrointestinal tract is critical to the pathogenesis of acute HIV-1 infection 419Saurabh Mehandru, MD, Klara Tenner-Racz, MD, Paul Racz, MD, PhD, and Martin Markowitz, MD,New York, NY, and Hamburg, Germany

Editorial

Are you immunodeficient? 423Francisco A. Bonilla, MD, PhD, and Raif S. Geha, MD, Boston, Mass

Rostrum

From idiopathic infectious diseases to novel primary immunodeficiencies 426Jean-Laurent Casanova, MD, PhD, Claire Fieschi, MD, PhD, Jacinta Bustamante, MD, Janine Reichenbach, MD,Natasha Remus, MD, Horst von Bernuth, MD, and Capucine Picard, MD, PhD, Paris and Creteil,France, and Frankfurt, Germany

Original articles

dEC Infant home endotoxin is associated with reduced allergen-stimulatedlymphocyte proliferation and IL-13 production in childhood

431

Joseph H. Abraham, ScD, Patricia W. Finn, MD, Donald K. Milton, MD, Louise M. Ryan, PhD,David L. Perkins, MD, and Diane R. Gold, MD, Boston, Mass


dOR

dOR

Y

Y

Y

Y


Contents

CONTENTS

CONTINUED

Does early EBV infection protect against IgE sensitization? 438Caroline Nilsson, MD, Annika Linde, MD, PhD, Scott M. Montgomery, BSc, PhD, Liselotte Gustafsson,Per Nasman, Ph Lic, Marita Troye Blomberg, PhD, and Gunnar Lilja, MD, PhD, Stockholm, Sweden

Biased use of VH5 IgE-positive B cells in the nasal mucosa in allergic rhinitis 445Heather A. Coker, PhD, Helen E. Harries, MBiochem, Graham K. Banfield, FRCS, Victoria A. Carr, RGN,Stephen R. Durham, MD, Elfy Chevretton, FRCS, Paul Hobby, MSc, Brian J. Sutton, PhD, andHannah J. Gould, PhD, London, United Kingdom

Antibody responses against galactocerebroside are potential stage-specificbiomarkers in multiple sclerosis

453

Til Menge, MD, Patrice H. Lalive, MD, Hans-Christian von Budingen, MD, Bruce Cree, MD, PhD,Stephen L. Hauser, MD, and Claude P. Genain, MD, San Francisco, Calif, and Zurich, Switzerland

Letters to the Editor

Perilesional GM-CSF therapy of a chronic leg ulcer in a patient with commonvariable immunodeficiency

460

Ammar Z. Hatab, MD, Deanna McDanel, PharmD, BCPS, and Zuhair K. Ballas, MD, Iowa City, Iowa

Asthma caused by cyanoacrylate used in a leisure activity 462Mona-Rita Yacoub, MD, Catherine Lemiere, MD, MSc, and Jean-Luc Malo, MD, Montreal, Quebec, Canada

Correspondence

Cystic fibrosis gene mutations and chronic rhinosinusitis 463Clement L. Ren, MD, Rochester, NY

Leukotriene receptor antagonists are not as effective as intranasalcorticosteroids for managing nighttime symptoms of allergic rhinitis

463

Robert A. Nathan, MD, Colorado Springs, Colo

Efficacy of ant venom immunotherapy and whole body extracts 464Simon G. A. Brown, MBBS, PhD, FACEM, Robert J. Heddle, MBBS, PhD, FRACP, FRCPA, Michael D.Wiese, BPharm, MClinPharm, and Konrad E. Blackman, MBBS, FACEM, Fremantel, Bedford Park, andHobart, Australia

Reply 465David B. K. Golden, MD, Baltimore, Md

Images in allergy and immunology

Toll-like receptors and atopy 467Pierre Olivier Fiset, BSc, Meri Katarina Tulic, PhD, and Qutayba Hamid, MD, PhD, Editors


Y

Y

14A August 2005 J ALLERGY CLIN IMMUNOL

Contents

CONTENTS

CONTINUED

Chronic active Epstein-Barr virus infection of natural killer cells presentingas severe skin reaction to mosquito bites

470

Susan E. Pacheco, MD, Stephen M. Gottschalk, MD, Mary V. Gresik, MD, Megan K. Dishop, MD, TakayukiOkmaura, MD, and Theron G. McCormick, MD, Guest Editors

Beyond our pages 473Burton Zweiman, MD, and Marc E. Rothenberg, MD, PhD, Editors

Correction

Physical activity and exercise in asthma: Relevance to etiology and treatment 298(Lucas SR, Platts-Mills TAE. 2005;115:928-34)

Reader services

Instructions for authors www.mosby.com/jaci and July 2005, pages 15A-22A

Information for readers 19A

Newsview—American Academy of Allergy, Asthma and Immunology 25A

CME calendar—American Academy of Allergy, Asthma and Immunology 30A

CME activities information 32A

Professional opportunities 35A

Change of address 354

Y

The Editors of The JACI are pleased to announce that Continuing Medical Education (CME) credit is now offered to

readers who successfully complete examination questions accompanying monthly review articles in the Journal’s

Current Reviews of Allergy and Clinical Immunology and Molecular Mechanisms in Allergy and Clinical Immunologyseries. This CME opportunity furthers the joint educational goals of the Journal and its sponsoring foundation, the

American Academy of Allergy, Asthma and Immunology (AAAAI). Learning objectives, examination questions, and

full details appear in each review article in the print and online Journal. The self-directed examinations can be taken at

the JACI website (www.mosby.com/jaci). Credit is administered by the AAAAI.

Complimentary 1-year subscriptions to The Journal of Allergy and Clinical Immunology are availableto AAAAI member FITs in the United States through an unrestricted educational grant from AlconLaboratories, Inc.

Statements and opinions expressed in the articles and communications herein are those of the author(s) and not necessarily those of the Editor, publisher, or theAmerican Academy of Allergy, Asthma and Immunology. The Editor, publisher, and the American Academy of Allergy, Asthma and Immunology disclaimany responsibility or liability for such material and do not guarantee, warrant, or endorse any product or service advertised in this publication, nor do theyguarantee any claim made by the manufacturer of such product or service.


Contents

THE JOURNAL OF

AllergyANDClinicalImmunology

Editor in Chief DONALD Y. M. LEUNG, MD, PhD Denver, Colo

Deputy Editors HAROLD S. NELSON, MD, AND STANLEY J. SZEFLER, MD Denver, Colo

Associate Editors ANDREA J. APTER, MD, MSc Philadelphia, Pa BRUCE BOCHNER, MD Baltimore, MdROBERT K. BUSH, MD Madison, Wis FRED FINKELMAN, MD Cincinnati, Ohio

QUTAYBA HAMID, MD, PhD Montreal, Quebec, Canada DAVID B. PEDEN, MD Chapel Hill, NCWILLIAM T. SHEARER, MD, PhD Houston, Tex SCOTT SICHERER, MD New York, NY

AND DONATA VERCELLI, MD Tucson, Ariz

Guest Editors BURTON ZWEIMAN, MD Philadelphia, Pa AND MARC E. ROTHENBERG, MD Cincinatti, Ohio

Editorial Board

LARRY BORISH, MD Charlottesville, Va 2006 CEZMI A. AKDIS, MD Davos, Switzerland 2009

REDWAN MOQBEL, PhD, FRCPath Edmonton, Alberta, Canada 2006 WILLIAM J. CALHOUN, MD Pittsburgh, Pa 2009

SANTA JEREMY ONO, PhD London, United Kingdom 2006 VERNON M. CHINCHILLI, PhD Hershey, Pa 2009

ZUHAIR K. BALLAS, MD, PhD Iowa City, Iowa 2007DANIEL L. HAMILOS, MD Boston, Mass 2009

THOMAS BIEBER, MD, PhD Bonn, Germany 2007ANTHONY A. HORNER, MD La Jolla, Calif 2009

PEYTON A. EGGLESTON, MD Baltimore, Md 2007HANS C. OETTGEN, MD, PhD Boston, Mass 2009

DAVID P. HUSTON, MD Houston, Tex 2007 DEVENDRA K. AGRAWAL, PhD Omaha, Neb 2010

PEDRO AVILA, MD Chicago, Ill 2008JOSHUA A. BOYCE, MD Boston, Mass 2010

DENNIS LEDFORD, MD Tampa, Fla 2008JAVIER CHINEN, MD, PhD Bethesda, Md 2010

DAVID B. PEDEN, MD Chapel Hill, NC 2008GURJIT K. KHURANA HERSHEY, MD, PhD Cincinnati, Ohio 2010

HARALD RENZ, MD Marburg, Germany 2008JOHN M. KELSO, MD San Diego, Calif 2010

HUGH A. SAMPSON, MD New York, NY 2008HANS-UWE SIMON, MD, PhD Bern, Switzerland 2010

ERIKA VON MUTIUS, MD, MSc Munich, Germany 2008

Board of Directors of the American Academy of Allergy, Asthma and Immunology

President F. ESTELLE R. SIMONS, MD, FAAAAI

Winnipeg, Manitoba, Canada

President-Elect THOMAS A. E. PLATTS-MILLS, MD, PhD, FAAAAI

Charlottesville, Va

Vice President THOMAS B. CASALE, MD, FAAAAI

Omaha, Neb

DAVID H. BROIDE, MD, FAAAAI La Jolla, Calif

THOMAS A. FLEISHER, MD, FAAAAI Bethesda, Md

SANDRA M. GAWCHIK, DO, FAAAAI Upland, Pa

STANLEY GOLDSTEIN, MD, FAAAAI Rockville Centre, NY

PAUL A. GREENBERGER, MD, FAAAAI Chicago, Ill

REBECCA S. GRUCHALLA, MD, PhD, FAAAAI Dallas, Tex

Secretary/Treasurer HUGH A. SAMPSON, MD, FAAAAI

New York, NY

Immediate Past President MICHAEL SCHATZ, MD, MS, FAAAAI

San Diego, Calif

Past Past President LANNY J. ROSENWASSER, MD, FAAAAI

Denver, Colo

RICHARD W. HONSINGER, MD, FAAAAI Los Alamos, NM

DENNIS K. LEDFORD, MD, FAAAAI Tampa, Fla

DONALD Y. M. LEUNG, MD, PhD, FAAAAI Denver, Colo

ARNOLD I. LEVINSON, MD, FAAAAI Philadelphia, Pa

JAMES T. LI, MD, PhD, FAAAAI Rochester, Minn

DENNIS R. OWNBY, MD, FAAAAI Augusta, Ga

Executive Vice President KAY WHALEN, CAE

Address of Executive Office

AMERICAN ACADEMY OF ALLERGY, ASTHMA AND IMMUNOLOGY555 East Wells Street, Suite 1100, Milwaukee, WI 53202-3823 (414) 272-6071 fax: (414) 272-6070

e-mail: [email protected]


THE JOURNAL OF


Information for Readers

COMMUNICATION

Communications regarding original articles and editorial man-agement should be addressed to Donald Y. M. Leung, MD, PhD,Editor in Chief, The Journal of Allergy and Clinical Immunology,National Jewish Medical and Research Center, 1400 Jackson St(J324), Denver, CO 80206; phone 303-398-1963; fax 303-270-2269.

CUSTOMER SERVICE(Orders, claims, online, change of address)

Elsevier Periodicals Customer Service, 6277 Sea Harbor Drive,Orlando, FL 32887-4800. Phone 800-654-2452 (US and Canada);407-345-4000 (outside US and Canada). Fax 800-225-6030 (US andCanada); 407-363-9661 (outside US and Canada). [email protected] changes must be submitted four weeks in advance.

YEARLY SUBSCRIPTION RATES

United States and possessions: Individual $211.00; Institution$465.00; Student/Resident $106.00. All other countries (pricesinclude airspeed delivery): Individual $270.00; Institution $529.00;Student/Resident $136.00. Single issues $45.00. To receive Student/Resident rate, order must be accompanied by name of affiliatedinstitution, date of term, and signature of program/residencycoordinator on institution letterhead; order will be billed at theIndividual rate until proof of status is received. Current processes arein effect for back volumes and back issues.

Further information on this journal is available from the Publisher orfrom this journal’s Web site (http://www.mosby.com/jaci). Infor-mation on other Elsevier products is available through Elsevier’sWeb site (http://www.elsevier.com).

Advertising information.Advertising orders and inquiries can be sentto: US, Canada, and South America, Cunningham Associates, 180Old TappanRoad, Old Tappan, NJ 07675. Classified advertising ordersand inquiries can be sent to Sia Battle, Elsevier Inc, 360 Park AvenueSouth,NewYork,NY10010; phone 212-633-3721; fax 212-633-3820.Europe and the rest of the world, Julie Toop; phone +44 (0) 1865843016; fax +44 (0) 1865 843976; e-mail [email protected].

Author inquiriesFor inquiries relating to the submission of articles (including elec-tronic submission where available), please visit Elsevier’s AuthorGateway at http://authors.elsevier.com. The Author Gateway alsoprovides the facility to track accepted articles and set up e-mail alertsto inform you of when an article’s status has changed, as well asdetailed artwork guidelines, copyright information, frequently askedquestions, and more. Please see Information for Authors for indi-vidual journal. Contact details for questions arising after acceptanceof an article, especially those relating to proofs, are provided afterregistration of an article for publication.

‘ The paper used in this publication meets the requirements ofANSI/NISO Z39.48-1992 (Permanence of Paper).

Reprints. For queries about author offprints, e-mail [email protected] reprints for educational, commercial,or promotional use, contact the Commercial Reprints Department,Elsevier Inc, 360 Park Avenue South, New York, NY 10010-1710;fax: 212-462-1935; e-mail [email protected] of single articles available online may be obtained bypurchasing Pay-Per-View access for $30 per article on the journalWeb site, www.mosby.com/jaci.

2005 American Academy of Allergy, Asthma andImmunology. All rights reserved.

This journal and the individual contributions contained in it areprotected under copyright by the American Academy of Allergy,Asthma and Immunology, and the following terms and conditionsapply to their use:

PhotocopyingSingle photocopies of single articlesmay bemade for personal use

as allowed by national copyright laws. Permission of the Publisherand payment of a fee is required for all other photocopying, includingmultiple or systematic copying, copying for advertising or pro-motional purposes, resale, and all forms of document delivery.Special rates are available for educational institutions that wish tomake photocopies for nonprofit education classroom use.

Permissions may be sought directly from Elsevier’s RightsDepartment in Philadelphia, PA, USA: phone 215-239-3804, fax215-239-3805, e-mail [email protected]. Requestsmay also be completed online via the Elsevier home page (http://www.elsevier.com/locate/permissions).

Usersmay clear permissions andmake payments in the US throughthe Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers,MA 01923, USA; phone 978-750-8400; fax 978-750-4744; and in theUK through theCopyright LicensingAgencyRapidClearance Service(CLARCS), 90TottenhamCourt Road, LondonW1P0LP,UK; phone44-20-7631-5555; fax 44-20-7631-5500. Other countries may havelocal reprographic rights agencies for payment.

Derivative WorksSubscribers may reproduce tables of contents or prepare lists of

articles including abstracts for internal circulation within theirinstitutions. Permission of the Publisher is required for resale ordistribution outside the institution.

Permission of the Publisher is required for all other derivativeworks, including compilations and translations.

Electronic Storage or UsagePermission of the Publisher is required to store or use

electronically any material contained in this journal, including anyarticle or part of an article.

Except as outlined above, no part of this publication may bereproduced, stored in a retrieval system or transmitted in any form orby any means, electronic, mechanical, photocopying, recording orotherwise, without prior written permission of the Publisher.

Address permissions requests to: Elsevier Rights Department, atthe fax and e-mail addresses noted above.

NoticeNo responsibility is assumed by the Publisher or the American

Academy of Allergy, Asthma and Immunology for any injury and/ordamage to persons or property as a matter of products liability,negligence or otherwise, or from any use or operation of any methods,products, instructions or ideas contained in the material herein. Be-cause of rapid advances in the medical sciences, in particular, inde-pendent verification of diagnoses and drug dosages should be made.

Although all advertisingmaterial is expected to conform to ethical(medical) standards, inclusion in this publication does not constitutea guarantee or endorsement of the quality or value of such product orof the claims made of it by its manufacturer.Indexed or Abstracted in Index Medicus, Science CitationIndex, Current Contents/Clinical Medicine, Current Contents/LifeSciences.

Available electronically fromOvidTechnologies (800-950-2035).Microform edition available from ProQuest Information andLearning, 300 N Zeeb Rd, Ann Arbor, MI 48106-1346.


J ALLERGY CLIN IMMUNOL

VOLUME 116, NUMBER 2

AAAAI Developing Researcher Award 23A

The most frequently cited allergy/immunology journal in thefield, The Journal of Allergy and Clinical Immunology, con-tinues to grow in prominence because of the excellent research

it showcases. The Academy is proud to announce this new JACI Awardwhich will recognize innovative research and outstanding scientificwriting by the new generation of allergy researchers and clinicians.

The AAAAI Award for Outstanding Research Published in the JACI bya Developing Researcher provides an unrestricted prize of $1,500, com-memorative plaque, and recognition during the AAAAI’s Annual Busi-ness Meeting. Up to 4 awardees will be chosen during one calendar year.

Objective• To encourage the submission to JACI of the highest quality articles

on our science’s frontier• To recognize ground breaking research and excellent writing by the

new generation of allergy researchers and clinicians

Award• Formal presentation of an unrestricted prize of $1500 and a plaque for each award (funded by the AAAAI) will be

made at the annual Academy Meeting• Announcement of the awards for the year will be featured in the published program for the annual Academy

Meeting• Announcement and commendation of winners will be published in the Journal

Title of the Award• The AAAAI Award for Outstanding Research Published in the JACI by a Developing Researcher

Judgment Process• A chairman will be appointed by the JACI Editors to coordinate a panel whose task will be to review and

evaluate nominated papers and reach a decision on recipients of the award• Judgement will be by a panel composed of the chairman of the award committee, one Editor, a member of the

AAAAI Research Advisory Council, and at least two researchers or clinicians who are knowledgeable in the areaof research being reported

Conditions of Eligibility• Candidates must meet the following criteria for eligibility

—can be from any training program world-wide—must have completed an MD and/or PhD or the equivalent within the 7 years prior to nomination for the

award—must have conducted the research and written the paper during the post-doctoral fellowship training and not

while holding a faculty position• The research must be considered outstanding and represent a conceptual advance in the treatment or pathogene-

sis of allergic disease• The article must have been published in JACI• The article must have appeared within 12 months of the nomination• The candidate must have conducted the majority of the research or have been the primary leader of the research

team and appear as the first author• A written nomination of the candidate and paper must be submitted by a sponsor (the training director, precep-

tor, research mentor, or a current Editor of the Journal.) The sponsor must submit a completed NominationApplication Form, a statement briefly discussing the impact of the scientific observations on the field of allergyand immunology and justification of why the published work is deserving of the award, and 3 letters of recom-mendation for the nominated paper, 2 of which are from individuals at institutions with which the Fellow hasnever been associated. (Letters should address the significance and importance of the work in the paper and itsrelevance to the Fellow’s body of work, the quality of the written presentation, the Fellow’s personal characteristicsand potential for a future in academic medicine.)

• Up to 4 awards may be given within each calendar year; if no candidates are deemed of sufficient merit, no awardwill be given

Criteria for Excellence in a Publication• The publication must present novel information that holds significant importance for the basic and clinical science

of allergy, asthma or immunology• The article must meet the highest standards for scientific writing• The publication must represent a significant portion of the candidate’s fellowship work

for Outstanding ResearchPublished in the JACI

by a Developing Researcher

AAAAI Award

*For application forms and further information, please contact the Editorial Office at 303 398-1963.

24A AAAAI Developing Researcher Award J ALLERGY CLIN IMMUNOL

AUGUST 2005

for Outstanding ResearchPublished in the JACI

by a Developing Researcher

AAAAI Award

Nomination for AAAAI Developing Researcher Award

SPONSOR Name Date

Relationship to FIT: ______TPD/Preceptor __________Research Mentor ________JACI Editor

DEVELOPING RESEARCHER Name

Mailing Address Phone Fax

Country E-mail

Academic affiliation

Academic degree(s) Date degree(s) awarded

Faculty appointments Appointment date(s)

Training program where work on paper was completed Date of Training ____ to ____

Article submitted for consideration (MUST HAVE BEEN ACCEPTED FOR JACI PUBLICATION)

(Provide title, authors & complete publication reference; if in press, provide planned publication date)

Attach 3 copies of the published article; if the article is “in press,” provide 2 copies of the manuscript, illustrations, and tables)

3 Letters of recommendation provided by Academic affiliation:(Name and address)

1. 1.

2. 2.

3. 3.

(Attach the original copy of each letter in a sealed envelope. At least 2 letters should be provided by individuals who are not asso-ciated with the nominee’s training center)

Attach a typewritten statement, no longer than 2 paragraphs, identifying the significance of the scientific observations inthis publication and explaining why this published work is deserving of the AAAAI Developing Researcher publication award

n Anaphylaxis: bridging the gap betweenresearch, clinical practice

In the next year, the AAAAI will work to bridge the gapbetween basic science research in anaphylaxis and clinicalpractice through a comprehensive public education outreachinitiative. The new Anaphylaxis Public Education Task Force

is developing this educational effort to raise awareness ofanaphylaxis as a killer allergy with many potential triggers.The initiative will also emphasize the importance of the role

of the allergist/immunologist in anaphylaxis managementand prevention.

The new anaphylaxis public education campaign is one of

the initiatives of PresidentF.EstelleR.Simons,MD,FAAAAI.The anaphylaxis initiative will include three major public

education campaigns:

d Back-to-school campaign, September 2005

Elements of the back-to-school campaign will includepublicity to national magazines, and national daily andweekly publications. The Task Force is also considering

a plan to reach school nurses, who may serve as a bridge toeducating a vast population of students. Regional, state andlocal allergy societies may be asked to work with their state

and local school nurse associations. The Food Allergy &Anaphylaxis Network (FAAN) also offers a quarterlynewsletter to school nurses, and the AAAAI Allergy and

Asthma Tool Kit for School Nurses will also be utilized.

d Holiday season campaign, late October through

December 2005

The holiday season campaign will focus on three main

holidays: Halloween, Thanksgiving and Christmas/Hanukkah. Outreach will begin in October with publicity tonational magazines, and daily and weekly media outlets.

d Great outdoors campaign, March through May 2006

The great outdoors campaign will conclude just before theMemorial Day weekend in May 2006. May is National Allergy

and Asthma Awareness Month, and also includes FoodAllergy Awareness Week. Anaphylaxis publicity efforts willwork hand-in-hand with these events.

Core messages for the public will include a brief descriptionof anaphylaxis, and answers to basic questions including:

d What is anaphylaxis?d Who is at risk?d When can anaphylaxis occur?d Where can anaphylaxis occur?d How is anaphylaxis treated?d Why is follow-up needed?

2006 Annual Meeting

The anaphylaxis public outreach effort will culminate

during the 2006 Annual Meeting in Miami Beach, FL, with

Sunday, March 5, 2006, designated as Anaphylaxis Day. OnAnaphylaxis Day, the Presidential Symposium will be held

concurrently with Sunday’s Plenary Session, and focus onadvances in research in basic and clinical science, relevant tothe diagnosis and treatment of anaphylaxis.

The AAAAI will also host a press conference highlighting

anaphylaxis-related abstracts selected for presentation at theAnnual Meeting.

2006 AAAAI Primary Care Symposium

The 2006 Primary Care Symposium ‘‘OptimizingTreatment of Your Allergic Patients: An Update for thePrimary Care Provider on Issues Affecting One-Third of Your

Practice,’’ co-chaired by Joann Blessing-Moore, MD,

FAAAAI and Richard F. Lockey, MD, FAAAAI, will alsoinclude a segment on anaphylaxis. The annual symposium isheld immediately prior to the Annual Meeting and designed

to update local primary care physicians about the mostrecent developments in allergic disease.

Simons will also serve as one of ten AAAAI representatives

at the Second Symposium on the Definition and Managementof Anaphylaxis, sponsored by the National Institutes of Health(NIH) and FAAN. The Symposium will build upon previous

efforts and work toward developing a universally accepteddefinition for anaphylaxis. It will further expand itsanaphylaxis research agenda, and identify educational needs

for health care professionals and patients. Hugh A. Sampson,

MD, FAAAAI, and Anne Munoz-Furlong will Co-Chair theSymposium.

Anaphylaxis Retrospective

The AAAAI is preparing a retrospective on anaphylaxis forthe 2006 Annual Meeting. If your research over the years hasmade a significant contribution to the understanding of

anaphylaxis, the AAAAI History and Archives Committee,chaired by Michael A. Kaliner, MD, FAAAAI, wants to hearfrom you. Please contact Audrey Mudek at the AAAAI

executive office, (414) 272-6071 ore-mail [email protected].

n American Academy of Allergy, Asthma& Immunology Lifelong Learner Bill ofRights

Dear Colleagues and Friends:

The American Academy of Asthma, Allergy &

Immunology provides outstanding continuing medical

education through The Journal of Allergy and ClinicalImmunology (citation impact factor now 7.205), the AAAAI

Annual Meeting, and other AAAAI programs, resources, and

materials.

In order to emphasize our commitment to continuing

medical education, we have recently developed the AAAAI

Lifelong Learner Bill of Rights. We pledge to maintain the

highest quality in all of our educational programs and to

NEWSVIEWnewsviewA Monthly Update of Developments

from the AAAAI

J ALLERGY CLIN IMMUNOL August 2005 Page 25A

fulfill our promise to provide you with a premier

educational experience.

Sincerely,

F. Estelle R. Simons, MD, FAAAAI

AAAAI President

The American Academy of Allergy, Asthma & Immunology(AAAAI) recognizes that you are a life-long learner who has

chosen to engage in continuing medical education to identifyor fill a gap in knowledge, skill or performance. As part of theAAAAI’s duty to you as a learner, you have the right to expect

that your continuing medical education experience with theAAAAI includes:

Content that:d promotes improvements or quality in healthcare;d is valid, reliable, and accurate;d offers balanced presentations that are free of

commercial bias for or against a product/service;d is vetted through a process that resolves any conflicts

of interests of planners, teachers, or authors;d is driven and based on learning needs;d addresses the stated objectives or purpose; andd is evaluated for its effectiveness in meeting the

identified educational needs.

A learning environment that:d supports learners’ ability to meet their individual needs;d respects and attends to any special needsof the learners;d respects the diversity of groups of learners; andd is freeofpromotional, commercial, and/or salesactivities.

Disclosure of:d relevant financial relationships planners, teachers,

and authors have with commercial interests related to

the content of the activity; andd commercial support (funding or in-kind resources) of

the activity.

n Free online CME courseThe Environmental Management of Asthma, a free online

Continuing Medical Education (CME) program, is available

on the AAAAI Web site, www.aaaai.org. The program provideshealthcare professionals and the insurance industry withinformation and resources to incorporate environmentalmanagement into clinical practices, and standards of care for

patients with asthma. AAAAI members and non-membershave an opportunity to earn 1.0 or 1.2 CME/CE credits.

The program addresses the impact and management of

environmental asthma triggers such as air pollutants, pollen,environmental tobacco smoke, mold, dust mites,cockroaches and warm-blooded pets. The course was

designed for physicians and allied health professionals,including nurse case managers, in primary care specialties.

The CME program consists of a 15-minute didacticpresentation with lecture notes highlighting the classification

of environmental triggers, followed by a series of pediatricand adult case studies with question and answer sections.The didactic portion is presented with real-time audiocapabilities.

The AAAAI designates this educational activity for amaximum of 1.0 category 1 credits towards the AMAPhysician’s Recognition Award. The course is also

approved by the California Board of Registered Nursing,Provider #10704, for 1.2 contact hours and by theCommission for Case Manager Certification (CCMC) for

1.0 contact hours.For more information, visit the Professionals Center of

the AAAAI Web site, www.aaaai.org. Choose theProfessionals Center, click on the Education for You

section and select the AAAAI Continuing EducationOpportunities link.

AWARDS AND GRANTS

n AAAAI/JACI Award for OutstandingResearch

David M. Fleischer MD, Denver, CO, has been awarded the2005 AAAAI/JACI Award for Outstanding Research.

Fleischer was honored for his November 2004 paper, ‘‘Peanutallergy: Recurrence and its management’’ (J Allergy ClinImmunol 2004 114(5):1195-201) designed and written during

his fellowship at Johns Hopkins, Baltimore MD.

n New FIT abstract awardsA new AAAAI Interest Section Fellow-in-Training Abstract

Award will honor seven fellows-in-training who submit thebest abstract in each Interest Section. The abstracts will

be selected by each Interest Section based on the highestscored abstract. The AAAAI will award the presenter with$500 and a plaque.

No application is necessary for this award. All abstracts

submitted by an FIT will be considered. To submit anabstract, visit the Annual Meeting Web site,www.annualmeeting.aaaai.org and choose the 2006 Annual

Meeting link. Award recipients will be honored throughoutthe 2006 Annual Meeting and during the Interest SectionForums.

For more information, please contact Mikelle Johnsonat the AAAAI executive office at (414) 272-6071 or [email protected].

n 2006 AAAAI abstract awards availableApplications and details about all abstract awards

are available on the AAAAI Annual Meeting Web site,

www.annualmeeting.aaaai.org.


from the AAAAI

Page 26A August 2005 J ALLERGY CLIN IMMUNOL

All applicationsaredueSeptember7with the exception

of the Fellow-in-Training (FIT) Travel Scholarship.

AAAAI Allied Health Travel Scholarship

Purpose: Four scholarships are awarded to AAAAI alliedhealth members who submit the best abstracts for

presentation at the 2006 Annual Meeting.Eligibility: AAAAI allied health members submitting

abstracts for presentation at the 2006 Annual Meeting

Award: Actual travel and hotel expenses incurred by theabstract presenter, up to a limit of $750 per recipient.

AAAAIInterestSectionFellow-in-TrainingAbstractAward

Purpose: This new award will honor seven fellows-in-training who submit the best abstract in each Interest Section.The abstracts will be selected by each Interest Section basedon the highest scored abstract.Noapplication is necessary for

this award. All abstracts submitted by an FIT will beconsidered. The award recipients will be honored throughoutthe Annual Meeting and during the Interest Section Forums.

Eligibility: AAAAI fellows-in-trainingAward: $500 and a plaque

AAAAI/Sepracor Research Excellence Awards

Purpose: Three awards are given to abstract presentersselected by the Abstract Review Committee. Abstracts will bejudged by the novelty of their research development and the

overall significance of their research toward the advancementof allergy, asthma and immunology, and optimal patient care

Eligibility: The abstract first author must be an allergy and

immunology trainee (pre or post-doctoral), and the traineemust have conducted the vast majority of the abstract relatedwork. The abstract co-author must be an AAAAI member.

The research should be investigator led and not sponsored byany for-profit entity.

Award: $5,000 per recipient

American Academy of Pediatrics Section on Allergy/

Immunology Outstanding Pediatric Allergy, Asthma and

Immunology Abstract Awards for Fellows-in-Training

and Junior Faculty

Purpose: Five awards are given, three to fellows-in-

training, and two to junior faculty members. All awardrecipients are eligible for a complimentary, one-yearmembership in the AAP and SOAI.

Eligibility: Fellows-in-training or junior faculty who

submit the best abstracts involving pediatric patients, birthto age 21 years, that are accepted for presentation at the 2006Annual Meeting

Award: $750 each to three fellows-in-training and $1,000each to two junior faculty members

Domestic Fellow-in-Training (FIT) Travel Scholarship

Purpose: A grant to cover travel expenses to the AnnualMeeting

Eligibility: AAAAI FIT members in the United States and

Canada

Award: $1,000 with an accepted abstract or $500 withoutan abstract

Deadline: TBAApplications will be available on the AnnualMeetingWeb

site this fall. Please direct questions to Reaca Pearl at theAAAAI executive office at (414) 272-6071 or

e-mail [email protected] ERT Allied Health Travel Grants

Purpose: The grants support education, research

and travel for three AAAAI allied health members who submitthe best abstracts for presentation at the 2006 AnnualMeeting.

Eligibility: AAAAI allied health members submittingabstracts for presentation at the 2006 Annual Meeting

Award: $750 per recipientInternational Fellow-in-Training (FIT) Travel

Scholarship

Purpose: The scholarships are designed to financiallyassist research and non-research fellows and residents-in-

training in allergy and immunology in attending the AAAAI2006 Annual Meeting.

Eligibility: International FIT AAAAI members, or

international FITs who have a completed membershipapplication on file at the time of scholarship applicationsubmission, are eligible. Applicants must be post-doctoral

trainees in allergy/immunology who, at the time ofapplication, are within seven years of their latest doctoraldegree, and are outside the United States and Canada.

Award:To ensure greater equity, financial support reflects

variable air travel costs based on location of award recipient.Awards generally vary from $850 to $1,300.

AAAAI MEMBER UPDATES

n 2006 Annual Meeting: call for abstractsThe 2006 AAAAI abstract submission site is now open.

Take this important opportunity to make a significantcontribution to the overall scientific content of the 2006Annual Meeting and share your findings with fellow Annual

Meeting delegates.For more information, visit the AAAAI Annual Meeting Web

site, www.annualmeeting.aaaai.org, or e-mail [email protected] abstract submission deadline is September 7.

n Strategic Training in Allergy Research(ST*AR) Program

The new AAAAI Strategic Training in Allergy Research

(ST*AR) Program will strengthen basic science researchin allergy/immunology and its translation to excellenceinclinical practice. Debuting at the 2006 Annual Meeting, the

ST*AR Program will offer 40-50 PhD or post-doctoral


from the AAAAI


students from the United States and Canada an in-depthintroduction to allergy/immunology research and an

opportunity to learn more about the specialty.Target audience: PhD trainees conducting research

related to allergic diseases, asthma and immunologyProgram overview: PhD trainees are invited to submit

abstracts for presentation at the 2006 AAAAI AnnualMeeting, March 3-7, 2006, in Miami Beach, FL. Up to 40selected participants will be invited to participate in the

ST*AR Program, which will include:

d Presentation of original research at the ST*AR Programd Attendance at AAAAI Annual Meeting postgraduate

program and symposiad Participation in a special ST*AR Program that includes

basic research presentations in small groups by AAAAImembers who are NIH funded researchers in allergy/immunology

d Information about career opportunities in allergy,asthma and immunology research

d Funds for travel (including a stipend for ground

transportation), hotel accommodation and meetingregistration

The goal of the ST*AR Program is to provide PhD traineeswith an opportunity to consider research opportunities in thefield of allergy and immunology. The AAAAI Annual Meeting

provides an excellent forum for those interested intranslating basic science advances into clinical practice tonetwork with basic scientists, translational clinical

researchers and clinical experts in allergy/ immunology.For more information, or to receive a ST*AR Program

application, contact Reaca Pearl at the AAAAI executiveoffice at (414) 272-6071 or e-mail [email protected].

n 2005 Annual Meeting Plenary Session onCD-ROM

Did you miss one of the six Plenary Sessions held duringthe 2005 Annual Meeting? You still have an opportunity tolearn about the newest research in the specialty and earn

continuing education credits. AAAAI members and non-members may order a CD of the six Annual Meeting PlenarySessions.

The Plenary Session CD will offer two types ofcontinuing education credit. Physicians may earn 10Continuing Medical Education (CME) credits, and nurses

may earn 11.9 Continuing Education (CE)contact hours for designated sessions.

The six Plenary Sessions are:

d Atopy: Nature and Nurtured New Paradigms in Cutaneous Inflammationd Optimizing Outcomes for Asthmad New Insights in Primary Immune Deficiencies

d New Insights in Food Allergyd New Paradigms in Upper and Lower Airways Diseases

The Plenary Session CD is $35. The CD may be purchasedfor up to two years following the 2005 Annual Meeting.

For more information or to order your CD, contact AliciaJosten at the AAAAI executive office at (414) 272-6071.

n AAAAI Annual Meeting dates and sites2006—Miami Beach, FL, March 3-72007—San Diego, CA, February 23-27

2008—Philadelphia, PA, March 14-182009—Washington, DC, March 13-172010—New Orleans, LA, February 26-March 2

AAAAI WEB SITE RESOURCESVisit the AAAAI Web site, www.aaaai.org, for a variety of

professional and patient education resources.

n Practice management tools availableVisit the AAAAI Web site, www.aaaai.org, for easy-to-use

resources that provide practice management assistance forestablished physicians, as well as those just starting out. Aseparate page will be developed for office managers and otherclinical staff that will include specific techniques and tools to

support them in their day-to-day work.Current resource center topics include:

d 2004 Practice Management Workshop presentationsd Financial performanced HIPAAd Medical office technologyd Patient satisfactiond Personal/time managementd Practice startupd Recruitment and staffingd Resources and sample forms

AAAAI practice management resources are located in theMembers Center of the AAAAI Web site. Visit the Members

Center often, additional practice management information isadded continuously.

n AAAAI mission statementThe mission of the American Academy of Allergy, Asthma

& Immunology is the advancement of knowledge and practice

of allergy, asthma and immunology for optimal patient care.For more information about the AAAAI, contact the executiveoffice, 555 E. Wells Street, Suite 1100, Milwaukee, WI 53202-

3823; phone (414) 272-6071; fax (414) 272-6070; [email protected]; or visit the Web site, www.aaaai.org.


from the AAAAI


CME Mission Statement(Approved by the AAAAI’s CME Committee, March 18,

2001. Approved by Board of Directors, June 23, 2001.)

The purpose of the AAAAI’s CME program is to provide

educational activities that will stimulate, maintain,

develop, and enhance the study and practice of allergy,

asthma and immunology. The content areas of the AAAAI’s

CME program include any topic within the field of allergy,

asthma and immunology which impacts the study or

practice of the specialty.

The target audience for the AAAAI’s CME program

includes allergists, immunologists, specialty and primary

care physicians, and allied health professionals. The types

of activities included as part of the AAAAI’s CME program

include a national Annual Meeting; regional, local, and

other live conferences; computer-assisted interactive

activities; and an array of written, audio, video and

multimedia enduring materials.

The AAAAI supports the concept of jointly sponsoring

activities with organizations whose goals are compatible

with those listed in the Mission Statement. The expected

results of the AAAAI’s CME program are to advance the

science and practice within the specialty of allergy, asthma

and immunology.

AAAAI SPONSORED ENDURINGMATERIALS

The activities listed below are directly sponsored by the

AAAAI and are available in print form or on the AAAAI Web

site, www.aaaai.org, as indicated by the descriptions below.

Participants may review the materials at their leisure for

credit; no attendance is necessary. CME credit will be

awarded upon completion of each activity’s evaluation or

post test. More specific instructions for claiming credits are

included in the individual enduring material.

Allergy-Immunology Medical Knowledge Self-Assessment ProgramSponsored by the AAAAI, in partnership with the

American College of Physicians (ACP). Funded through

an unrestricted educational grant from sanofi aventis.

For more information, e-mail [email protected]. Credits

available: 50.0 CME.

2003 Postgraduate Plenary Session WebcastAllergy, Asthma & Immunology: 60 Years of Progress

Sponsored by the AAAAI

Location: www.aaaai.org

More information: e-mail [email protected]

Credits available: 2.5 CME

Dinner Symposium 3906The Suspension is Over: New Solutions for the

Treatment of Asthma

Sponsored by the AAAAI and funded through an

unrestricted educational grant from sanofi aventis.




Dinner Symposium 4906 WebcastA Look to the Future: The Management of Allergic

Diseases

Sponsored by the AAAAI and funded through an

unrestricted educational grant from sanofi aventis.




The JACI monthly review articles:

Current Reviews of Allergy and Clinical Immunology

Molecular Mechanisms in Allergy and ClinicalImmunologyLocations: JACI and online at www.mosby.com\jaci



Respiratory Digest articlesJointly sponsored by AAAAI and Adelphi, Inc., and

funded through an unrestricted educational grant from

sanofi aventis.

For more information, contact Joan Weiss at Adelphi,

Inc. by phone at (646) 602-7060, Fax (646) 602-6071, or

e-mail [email protected].


Respiratory Digest, Volume 3, Issue 3Clinical Implications of the Allergic Rhinitis/Asthma

Connection by Thomas B. Casale, MD, FAAAAI

Respiratory Digest, Volume 3, Issue 4Sleep-Disordered Breathing in Adults by Philip L. Smith,

MD, and Alan R. Schwartz, MD

Respiratory Digest, Volume 4, Issue 1Evaluation of the Solitary Pulmonary Nodule by

E. James Britt, MD


CME ACTIVITIESCALENDAR

cme activitiescalendar

Respiratory Digest, Volume 4, Issue 2Management of Exercise-Induced Asthma by John M.

Weiler, MD, and Julie A. Grant, MSPA-C

Respiratory Digest, Volume 4, Issue 3Treatment of Severe Asthma by

Stephen P. Peters, MD, PhD

Respiratory Digest, Volume 4, Issue 4Management of Exacerbations of Chronic

Bronchitis by Victor M. Pinto-Plata, MD, and

Bartolome R. Celli, MD

Respiratory Digest, Volume 5, Issue 1Allergic Rhinitis and Its Comorbidities by

Eli O. Meltzer, MD, FAAAAI

Respiratory Digest, Volume 5, Issue 2Managing Antihistamine Impairment in Allergic Rhinitis

by Thomas B. Casale, MD, FAAAAI

Respiratory Digest Special Report 1 (2003)A Look to the Future: The Management of Allergic

Diseases by Thomas B. Casale, MD, FAAAAI, and

Erwin W. Gelfand, MD, FAAAAI

Emerging Trends in the Management of Asthma andOther Allergic Diseases: A Focus on Anti-IgE TherapyJointly sponsored by AAAAI and Synermed

More information: E-mail [email protected]


The Clinical Benefits of Anti-Ige Therapy: A TargetedTreatment for Patients with Asthma and OtherAllergic Airway DiseasesJointly sponsored by AAAAI and Synermed

More information: E-mail [email protected]



Executive Office555 East Wells Street

Suite 1100Milwaukee, WI 53202-3823

(414) 272-6071www.aaaai.org

THE JOURNAL OF


INFORMATION FOR CATEGORY 1 CME CREDIT

Credit can now be obtained, free for a limited time, by reading the review articles in this issue. Please note the followinginstructions.

Method of Physician Participation in Learning Process: The core material for these activities can be read in this issue ofthe Journal or online at the JACI Web site: www.mosby.com/jaci. The accompanying tests may only be submitted online atwww.mosby.com/jaci. Fax or other copies will not be accepted.

Date of Original Release: August 2005. Credit may be obtained for these courses until July 31, 2006.Copyright Statement: Copyright 2005-2006. All rights reserved.Overall Purpose/Goal: To provide excellent reviews on key aspects of allergic disease to those who research, treat, or

manage allergic disease.Target Audience: Physicians and researchers within the field of allergic disease.Accreditation/Provider Statements and Credit Designation: The American Academy of Allergy, Asthma and

Immunology (AAAAI) is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to providecontinuing medical education for physicians. The AAAAI designates these educational activities for up to 1.0 hour in category 1credit toward the AMA Physician’s Recognition Award. Each physician should claim only those hours of credit that he or sheactually spent in the educational activity.

CME article

‘‘Innate immune responses to infection’’(page 241)

List of Design Committee Members: Author:Michael F. Tosi, MD

Activity Objectives1. To achieve a greater and more current understanding

of the nature and scope of innate immune responses toinfection.

2. To develop an enhanced appreciation for the rangeof interactions among various components of innateimmunity.

3. To be able to appreciate the specific microbialtargets of specific effector mechanisms of innateimmunity.

Recognition of Commercial Support: This CMEactivity has not received external commercial support.

Disclosure of Significant Relationships with RelevantCommercial Companies/Organizations:

Michael F. Tosi has no significant relationships todisclose.

CME article

‘‘EBV the prototypical human tumor virus—justhow bad is it?’’ (page 251)

List of Design Committee Members: Author:David A. Thorley-Lawson, PhD

Activity Objectives1. To understand that EBV uses mature B cell biology to

establish latency, persist, and replicate.2. To understand that even though EBV is so widespread

and apparently benign, it is potentially life-threatening.3. To understand that EBV evolved the capacity to make

cells grow because it is an essential part of the mechanism forestablishing latency in resting cells that are not pathogenic.

4. To understand that EBV-associated tumors arise fromdifferent stages in the life cycle of latently infected B cellsand that disruption of the immune response is an importantcomponent in the development of all of the EBV-associatedlymphomas.

Recognition of Commercial Support: This CMEactivity has not received external commercial support.

Disclosure of Significant Relationships with RelevantCommercial Companies/Organizations:

David A. Thorley-Lawson has equity ownership inEBVax.


Ciclesonide and allergen-inducedasthmatic responses

New-generation inhaled steroids having low systemiceffects are being sought for the treatment of asthma.

Ciclesonide (Alvesco) is a once-daily, nonhalogenatedinhaled corticosteroid (ICS) that has recently beenintroduced into the United Kingdom and Germany forthe treatment of persistent asthma at doses of 80 lgand 160 lg. This ICS remains inactive until cleaved byesterases present in the airway, where its active metab-olite, desisobutyryl-ciclesonide, then binds glucocorticoid

receptors. In this issue of the Journal, Gauvreau et al(p 285) have investigated the effects of ciclesonide in amulti-center, randomized, crossover study, comparing

once-daily dosing of 40 lg, 80 lg, and placebo on allergen-induced airway responses of individuals with mild

asthma. This study demonstrates that ciclesonide 80 lgattenuates the allergen-induced early and late changes inFEV1 as well as serum eosinophil cationic protein and

sputum eosinophils measured at 24 hours after challenge(P < .025), whereas ciclesonide 40 lg attenuates the lateasthmatic responses and sputum eosinophils measured at24 hours after challenge (P < .025). This study provides

new information regarding the minimally effective dosesof inhaled ciclesonide for inhibition of allergen-inducedairway responses and the apparent local anti-inflamma-

tory effects on the airways. Further evaluation ofciclesonide will be required to address whether theselow doses are clinically effective.

Wheezing in infants—associated with‘‘stop-and-go’’ traffic?

Recent research has suggested a possible link between

diesel exhaust particulates (DEP) and respiratory andallergic diseases. In this issue of the Journal, Ryan et al(p 279) describe the association between wheezing in

infants less than 1 year of age and exposure to stop-and-gotruck and bus traffic. The investigators observed a dramaticincrease in wheezing in infants who reside less than 100 m

from stop-and-go truck and bus traffic compared withinfants who reside farther from all sources of DEP. Inaddition, African American infants had the highestprevalence of wheezing in comparison with Caucasian

infants, regardless of exposure category. These findingssuggest that living very close to stop-and-go traffic earlyin infancy is a significant risk factor for wheezing. It also

suggests that even within an urban environment, an infant’srisk for wheezing varies with exposure to different types and

amounts of traffic. This study is the first report from theongoing Cincinnati Childhood Allergy and Air PollutionStudy, the goal of which is to elucidate the environmental

and genetic contributions to the development of allergicdiseases.

Endotoxin and the pathway to allergy

Endotoxin, a part of the cell wall of gram-negativebacteria, is present at higher levels in households with largeanimals (livestock or dogs). Infant endotoxin exposure has

been proposed as a factor that might protect against allergyand early childhood immune responses (eg, productionof the cytokine IL-13) that increase IgE production to

allergens. Cross-sectional European studies have found thatelevated endotoxin levels are associated with allergyprotection in children of farmers, but the immunologicpathway to explain this association is uncertain and the

relevance of the finding to children in the more urbanUS setting is unclear. In this issue of the Journal, Abraham

et al. (p 431) assessed in a cohort of US children

household dust endotoxin at age 2-3 months andPBMC-proliferative and cytokine responses to cockroach,dust mite, and cat allergens and to the nonspecific mitogen

phytohemagglutinin at age 2-3 years. They found thatincreased endotoxin levels were associated with decreasedIL-13 in response to cockroach, dust mite, and cat allergen

but not in response to mitogen stimulation. An inverse,though nonsignificant, association was found betweenendotoxin and proliferative responses. Early-life endotoxin-

related reduction of IL-13 production might represent onepathway through which elevated endotoxin decreases therisk of allergic disease and allergy in later childhood.

Prevalence of wheeze (without cold) by distance from DEP source. Moving

exposure, residing within 400 m of a highway with >1000 trucks daily;

Stop-and-go exposure, residing within 100 m of either a bus route or an urban

state route.

The Editors’Choice

Donald Y.M. Leung, MD, PhD

Harold S. Nelson, MD

Stanley J. Szefler, MD

THE JOURNAL OF


VOLUME 116 NUMBER 2

J ALLERGY CLIN IMMUNOL August 2005 9

Cholinergic urticaria patients: Sometimesallergic to their own sweat

Cholinergic urticaria (CU) presents a characteristic

picture of pinpoint-sized, highly pruritic wheals withsurrounding erythema that occurs after sweating duringphysical exercise, bathing, or emotional stress. The path-ogenesis of CU is not well defined. However, it has been

reported that these patients have positive intradermalskin tests to their own sweat. In this issue of the Journal,Dr Fukunaga and colleagues (p 397) report on this sensi-

tization in 18 subjects with CU and 10 controls. Theyperformed intradermal skin testing and basophil histamine

release with autologous sweat and serum. Eleven of17 patients with CU had positive skin test results, and

10 of 17 had basophil histamine release with autologoussweat, with a significant correlation between the 2 re-sponses. Nine of 16 of the patients with CU had a positive

intradermal skin test result with autologous serum. Theauthors proposed that there are 2 distinct subtypes ofpatients with CU: (a) those who have strong reactions

to autologous sweat and negative reactions to autologousserum whose wheals are not associated with hair folliclesand (b) those characterized byweak reactions to autologoussweat and positive reactions to autologous serum whose

wheals are associated with hair follicles.

Sensitization to aeroallergens in theUS population

The National Health and Nutrition Examination Survey(NHANES) is a population-based survey undertaken

periodically by the National Center for Health Statisticsto determine the health and nutritional status of the USpopulation. Prick/puncture skin testing (PPST) was

performed in a subset of subjects to 8 aeroallergens inNHANES II (1976-80) and to 9 aeroallergens and peanuts inNHANES III (1988-94). The results of the skin testing in

NHANES III are reported by Dr Arbes and colleagues in thisissue of the Journal (p 377). Of 10,508 skin-tested subjectsaged 6 to 59 years, more than half had at least 1 positivePPST result to the 9 aeroallergens. Those skin tests most

commonly positive were to dust mite (positive in 27.5% ofsubjects), rye grass (in 26.9%), short ragweed (in 26.2%),and German cockroach (in 26.1%). The remainder—

Bermuda grass, Russian thistle, white oak, cat, andAlternaria alternata—were positive in 10% to 20% ofsubjects. The least commonly positive skin test was to

peanut (8.6%). The 3 most significant independentpredictors of a positive PPST result to aeroallergens wereage (maximal for the 20- to 29-years age group), male sex,

and minority ethnicity, especially non-Hispanic black.

Comparison of these results with the results of the skintesting performed in NHANES II was difficult because

of numerous methodologic differences between the 2studies. However, for the 6 aeroallergens tested in bothsurveys, a positive PPST result was 2.1 to 5.5 times more

common in NHANES III than in NHANES II. Although theauthors could not definitely conclude that these differencesrepresent an increased sensitivity in the US population,

they point out that such an increase would be consistentwith reports from other countries.

COX-2 inhibitors enhance allergicresponses

Mechanical injury to the skin by scratching is animportant feature of atopic dermatitis (AD). In this issueof the Journal, Laouini et al (p 390) show that mechanical

injury to mouse skin inflicted by tape stripping results inrapid induction of cyclooxygenase-2 (COX-2) mRNA andprotein and in accumulation of the COX-2 product prosta-glandin E2 (PGE2). The role of COX-2 was examined in a

mouse model of AD elicited by repeated epicutaneous (EC)sensitization with ovalbumin (OVA) and characterized byeosinophil skin infiltration and a systemic TH2 response to

antigen. Administration of the COX-2 selective inhibitorNS-398 during EC sensitization resulted in enhanced skin

infiltration by eosinophils and expression of IL-4 mRNA,enhanced OVA-specific IgE and IgG1 antibody responses,and increased IL-4 secretion by splenocytes following

OVA stimulation. COX-2–deficient mice also exhibited anenhanced systemic TH2 response to EC sensitization. Theseresults demonstrate that COX-2 products limit allergic skininflammation, and they are consistent with the recent

finding that engagement of the EP3 receptor by PGE2

inhibits allergic reactions (Nat Immunol 2005;6:524-31).More importantly, the work of Laouini et al suggests that

COX-2 inhibitors might worsen allergic skin inflammationand should be avoided in patients with AD.

Distribution of positive skin tests to 9 aeroallergens and peanuts in the

US population, age 6 to 59 years.

0 August 2005 J ALLERGY CLIN IMMUNOL

Current reviews of allergy and clinical immunology

Series editor: Harold S. Nelson, MD

Innate immune responses to infection

Michael F. Tosi, MD New York, NY

This activity is available for CME credit. See page 32A for important information.

The human host survives many infectious challenges in the

absence of preexisting specific (adaptive) immunity because of

the existence of a separate set of protective mechanisms that do

not depend on specific antigenic recognition. These antigen-

independent mechanisms constitute innate immunity.

Antimicrobial peptides are released at epithelial surfaces and

disrupt the membranes of many microbial pathogens. Toll-like

receptors on epithelial cells and leukocytes recognize a range of

microbial molecular patterns and generate intracellular signals

for activation of a range of host responses. Cytokines released

from leukocytes and other cells exhibit a vast array of

regulatory functions in both adaptive and innate immunity.

Chemokines released from infected tissues recruit diverse

populations of leukocytes that express distinct chemokine

receptors. Natural killer cells recognize and bind virus-infected

host cells and tumor cells and induce their apoptosis.

Complement, through the alternative and mannose-binding

lectin pathways, mediates antibody-independent opsonization,

phagocyte recruitment, and microbial lysis. Phagocytes migrate

from the microcirculation into infected tissue and ingest and

kill invading microbes. These innate immune mechanisms

and their interactions in defense against infection provide

the host with the time needed to mobilize the more slowly

developing mechanisms of adaptive immunity, which might

protect against subsequent challenges. (J Allergy Clin

Immunol 2005;116:241-9.)

Key words: Innate immunity, antimicrobial peptides, Toll-like

receptors, chemokines, natural killer cells, complement, phagocytes

It is traditional to organize host responses to infectioninto separate arms or compartments, such as complement,phagocytes, cytokines, cell-mediated immunity, andhumoral immunity. A more current approach has been toconsider 2 larger categories: innate immunity, incorporat-

ing the more rapid and phylogenetically primitive non-specific responses to infection, such as surface defenses,cytokine elaboration, complement activation, and phago-cytic responses,1 and adaptive immunity, involving moreslowly developing, long-lived, and highly evolvedantigen-specific protective responses, such as antibodyproduction and cell-mediated immunity, that exhibitextraordinarily diverse ranges of specificity.2,3 However,the components of innate and adaptive immunity engagein a range of interactions that is remarkably diverse andcomplex. This review attempts to provide an overview ofthe main innate responses to infection that are availableto the human host, including relevant examples of suchinteractions.

INNATE IMMUNITY

Epithelia, defensins, and otherantimicrobial peptides

The epithelium of skin and mucosal tissue functions asa mechanical barrier to the invasion of microbial patho-gens. In the last 2 decades, it has become clear thatepithelial cells also are a major source of antimicrobialpeptides that play important roles in local host defense.4,5

Studies of their structure, sources, expression, and actionsalso have revealed an unexpected range of immunologicactivities for these molecules, the functions of which oncewere considered mainly antimicrobial in nature.4

Abbreviations usedCXCL: CXC ligand

HBD: Human b-defensin

ICAM-1: Intercellular adhesion molecule 1

LFA-1: Lymphocyte function-associated antigen 1

MAC: Membrane attack complex

Mac-1: Macrophage antigen-1

MBL: Mannan-binding lectin

NADPH: Reduced nicotinamide adenine dinucleotide

phosphate

NF: Nuclear factor

NK: Natural killer

PMN: Polymorphonuclear leukocyte

TLR: Toll-like receptor

From the Department of Pediatrics, Mount Sinai School of Medicine,

New York, and the Division of Pediatric Infectious Diseases, Maimonides

Medical Center, Brooklyn.

Disclosure of potential conflict of interst: M. F. Tosi—none disclosed.

Received for publication May 17, 2005; accepted for publication May 18,

2005.

Available online July 5, 2005.

Reprint requests: Michael F. Tosi, MD, Division of Pediatric Infectious

Diseases, Maimonides Medical Center, 977 48th St, Brooklyn, NY 11219.

E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.036

241

Review

sand

featu

rearticles

mailto:[email protected]

Epithelial cells of mucous membranes of the airwaysand intestines, as well as keratinocytes, express the humanb-defensins (HBD-1 through HBD-4). These small cati-onic peptides are similar to the a-defensins stored in theazurophilic granules of neutrophils, and they display anti-microbial activity against a broad range of bacteria, fungi,chlamydiae, and enveloped viruses.4,5 Their productionby epithelial cells might be constitutive, as for HBD-1, orinducible, as for HBD-2, HBD-3, and HBD-4. For exam-ple, recent evidence indicates that epithelial cells of theairway or intestine can produce HBD-2 in response toactivation by bacterial products through toll-like receptors(TLRs) 2 or 4 (see below) on the epithelial cells.6,7

Stimulation of epithelium by cytokines, including IL-1 orTNF-a, also can induce defensin production.4 Defensinshave been reported to exert their antimicrobial actioneither through the creation of membrane pores or throughmembrane disruption resulting from electrostatic interac-tion with the polar head groups of membrane lipids, withmore evidence now favoring the latter mechanism.4,8

Some microorganisms have evolved mechanisms forevading the action of defensins. For example, bacterialpolysaccharide capsules might limit access of microbialpeptides to the cell membrane,9 and an exoprotein ofStaphylococcus aureus, staphylokinase, neutralizes themicrobicidal action of neutrophil a-defensins.10

Several immunoregulatory properties of defensins andrelated peptides, distinct from their antimicrobial actions,have been documented.4 Several such peptides have beenshown to facilitate posttranslational processing of IL-1b.11

Some of the b-defensins have been shown to function aschemoattractants for neutrophils, memory T cells, andimmature dendritic cells by binding to the chemokinereceptor CCR-6.5,12,13 Separately, HBD-2 has been shownto activate immature dendritic cells through a mechanismthat requires TLR4.14 The activation of immature dendriticcells by these mechanisms also promotes their maturation.Theb-defensins also act as a chemoattractant formast cellsthrough an undefined mechanism and can induce mast celldegranulation.15 HBD-2 and several other antimicrobialpeptides can interfere with binding between bacterial LPSand LPS-binding protein.16

Additional antimicrobial peptides of epithelial cellsinclude lysozyme and cathelicidin. Lysozyme, an antimi-crobial peptide also found in neutrophil granules, attacksthe peptidoglycan cell walls of bacteria and can be releasedfrom cells through mechanisms that involve TLR activa-tion.17 Cathelicidin, or LL37, like lysozyme, is releasedfrom both neutrophils and epithelial cells. It exhibits broadantimicrobial activity and can inhibit lentiviral replica-tion.5,18 Cathelicidin also exhibits chemotactic activity forneutrophils, monocytes, and T lymphocytes. This activityis mediated by a formyl peptide receptor-like molecule,FPRL1, rather than the chemokine receptor CCR6 boundby b-defensins.19

The release of defensins in response to activation ofTLRs and the many actions of these peptides, includingtheir direct antimicrobial activities, their chemoattractantactions for a wide range of immune cells, and their

activation of dendritic cell maturation, already suggest ahighly complex and regulatory role in the development ofhost defense and immunity. Recent genomic evidence forthe possible existence of as many as 25 additional humandefensins that have not yet been characterized suggeststhat current knowledge describes but a small sample ofthe overall contribution of these peptides to immuneresponses.20

TLRs

Mononuclear phagocytes, including circulating mono-cytes and tissue macrophages, other phagocytic cells, andmany epithelial cells, express a family of receptors that ishighly homologous to the Drosophila receptor calledToll.6,7,21 These receptors mediate a phylogeneticallyprimitive, nonclonal mechanism of pathogen recognitionbased on binding not to specific antigens but to structurallyconserved pathogen-associated molecular patterns.21-23

There are at least 10 human TLRs with a range ofmicrobial ligands, such as gram-negative bacterial LPS,bacterial lipoproteins, lipoteichoic acids of gram-positivebacteria, bacterial cell-wall peptidoglycans, cell-wall com-ponents of yeast and mycobacteria, unmethylated CpGdinucleotide motifs in bacterial DNA, and viral RNA.22-24

Gram-positive cell-wall components bind mainly toTLR2, and TLR2 also can bind components of herpessimplex virus.22-25 Gram-negative LPS activates TLR4indirectly by first binding to LPS-binding protein, whichbinds in turn to CD14 at the cell surface. The bound CD14has no transmembrane domain but associates directly withan extracellular domain of TLR4.23,24 TLR5 has beenidentified as the receptor for bacterial flagellin, TLR9recognizes CpG motifs of bacterial DNA, and TLR3 hasbeen shown to bind synthetic and viral double-strandedRNA.26-28

Signalling through TLRs occurs through a well-described pathway in which receptor binding generatesa signal through an adaptor molecule, MyD88, that leadsto intracellular association with IL-1 receptor-associatedkinase. In turn, this leads to activation of TNF receptor–associated factor 6, which results in nuclear translocationof nuclear factor kB (NF-kB).23,24 NF-kB is an importanttranscription factor that activates the promoters of thegenes for a broad range of cytokines and other proin-flammatory products, such as TNF-a, IL-1, IL-6, and IL-8.This signalling pathway, on the basis of studies withTLR4, is similar but not identical to the signallingpathways activated by other TLRs.24 The activation ofcytokine production by TLRs plays an important role inrecruiting other components of innate host defense againstbacterial pathogens. However, with large-scale cytokinerelease, the deleterious effects of sepsis or other formsof the systemic inflammatory response syndrome demon-strate that these pathways have both beneficial andpotentially harmful effects for the host.24 Genetic poly-morphisms in TLRs might play a role in determining thebalance of these effects in certain individuals respondingto the challenge of systemic infection.24,29,30


AUGUST 2005

242 Tosi

Review

sand

featu

rearticle

s

In addition to their first-responder roles in generatingan inflammatory response to invading pathogens, TLRscan network with other components of innate and adap-tive immunity. TLR4 function is suppressed by activa-tion of cells through the chemokine receptor CXCR4.31

Activation of some TLRs also can induce expression ofthe costimulatory molecule B7 on antigen-presentingcells, which is required for activation of naive T cells.21

Cytokines

A heterogeneous group of soluble small polypeptide orglycoprotein mediators, often collectively called cyto-kines, form part of a complex network that helps regulatethe immune and inflammatory responses. Included in thisgroup of mediators, the molecular weights of which rangefrom about 8 to about 45 kd, are the ILs, IFNs, growthfactors, and chemokines (see separately below).Most cellsof the immune system, as well as many other host celltypes, release cytokines, respond to cytokines throughspecific cytokine receptors, or both. The range of sourcesand effects of cytokines and their actions and interrela-tionships are of such complexity that they cannot beaddressed here in detail. A number of them will beaddressed individually in sections below, and severalexcellent reviews are available.32-34 However, 2 cyto-kines, IL-1 and TNF-a, are of such fundamental impor-tance in acute host responses to infection that they warrantspecific attention.

IL-1 and TNF-a are small polypeptides, each with amolecular weight of approximately 17 kd, that exhibit abroad range of effects on immunologic responses, inflam-mation, metabolism, and hematopoiesis.34,35 IL-1 origi-nally was described as ‘‘endogenous pyrogen,’’ referringto its ability to produce fever in experimental animals, andTNF-a, which produces someof the same effects producedby IL-1, was originally named ‘‘cachectin’’ after thewasting syndrome it produced when injected chronicallyin mice.34,35 Many of the physiologic changes associatedwith gram-negative sepsis can be reproduced by injectingexperimental animals with these cytokines in the absenceofmicroorganisms. Depending on the doses injected, theseeffects might include fever, hypotension, and either neu-trophilia or leukopenia.34,35 In the production of endotoxicshock caused by gram-negative sepsis, IL-1 and TNF-aare produced by mononuclear phagocytes in response toactivation of TLRs by bacterial LPS. They in turn activatethe production of other cytokines and chemokines, lipidmediators (eg, platelet-activating factor and prostaglan-dins), and reactive oxygen species. They also induceexpression of adhesion molecules of both endothelial cellsand leukocytes, stimulating recruitment of leukocytes byinducing release of the chemokine IL-8 and activatingneutrophils for phagocytosis, degranulation, and oxidativeburst activity.24,35 These all are important and usuallybeneficial host responses to infection. However, at highlevels of activation, there sometimes are pathophysio-logic effects of this proinflammatory cascade, includingvascular instability, decreased myocardial contractility,capillary leak, tissue hypoperfusion, coagulopathy, and

multiple organ failure.24 For some systemic actions,notably the production of hemodynamic shock, IL-1 andTNF-a are synergistic. Both IL-1 and TNF-a also induceproduction of IL-6, a somewhat less potent cytokine thatexhibits some of the actions of IL-1 and TNF-a.34 Thehuman host produces several soluble antagonists ofIL-1 and TNF-a that can modulate their effects, includingIL-1 receptor antagonist, soluble TNF-a receptor, andanti-inflammatory cytokines, especially IL-10.24

The importance of effects mediated by IL-1 and TNF-ain the pathophysiology of septic shock has promptedmuch active research aimed at blocking their effects toreduce morbidity and mortality. Monoclonal antibodiesagainst TNF-a and other inhibitors of TNF-a or IL-1 haveshowed early promise in vitro and in animal models ofseptic shock.24,34,36,37 However, they have been far moreeffective at preventing the effects of cytokines thanreversing them. More recent attempts to address the issueof the timing of intervention have been directed at theintracellular signaling mechanisms activated through theTNF-a receptor or at mediators that appear later thanTNF-a. Lipophilic inhibitors of protein tyrosine kinases,enzymes that propagate the cellular signals throughTNF-areceptors, have been found to enhance survival in exper-imental animals, even when administered 2 hours aftersystemic injection with endotoxin.38 Additionally, mAbsagainst a cytokine-like nonhistone nucleoprotein productof macrophages, high-mobility group B1 (which appearsmuch later than TNF-a or IL-1 after LPS stimulation),were found to rescue mice from endotoxin shock whengiven 2 hours after an otherwise lethal dose of LPS.39Morerecently, clinical trials with activated protein C, a regula-tory protein in the coagulation cascade, has demonstratedbeneficial effects in selected patients with septic shockthrough mechanisms that might involve inhibition ofNF-kB activation.40 To date, despite progress, clinicalstrategies to interfere with the cytokine-induced cascadethat leads to endotoxin-induced shock have continued,overall, to meet with limited success.

Chemokines

A specialized group of small cytokine-like polypep-tides, chemokines, which all share the feature of beingligands for G protein–coupled, 7-transmembrane segmentreceptors, plays an increasingly complex role in theimmune response as cellular activators that induce directedcell migration, mainly of immune and inflammatorycells.41-44 The chemokines and their receptors have beenclassified into 4 families on the basis of themotif displayedby the first 2 cysteine residues of the respective chemokinepeptide sequence. Each of at least 16 CXC chemokinesbinds to 1 or more of the CXCRs (CXCR1 throughCXCR6). Examples of CXC chemokines include IL-8and Gro-a. Similarly, at least 28 CC chemokines, such asmacrophage inflammatory protein 1a, RANTES, andeotaxin-1, 2, and 3, bind to one or more of the CCRs(CCR1 through CCR10). The sole CX3C chemokine,fractalkine, or neurotaxin, binds to CX3CR1, currentlythe only receptor in its family. The 2 XC chemokines,



Tosi 243

Review

sand

featu

rearticles

including lymphotaxin, bind to the sole receptor in thisfamily, XCR1. A new nomenclature has been proposed todesignate each of the chemokines as a numbered ligand forits respective receptor family. In this systemGro-a is CXCligand (CXCL) 1 (or CXCL-1), and IL-8 now becomesCXCL-8. Similarly, RANTES becomes CCL-5, fractal-kine is CX3CL-1, and lymphotactin is XCL-1.41,43 Anupdate of this nomenclature system recently has beenpublished, tabulating each of the families with theirrespective ligands and receptors, as well as the traditionalnames in both human and murine systems.43

Virtually every cell type of the immune systemexpresses receptors for one or more of the chemokines.The cells of most inflamed tissues can release a variety ofchemokines, and tissues infected with different bacteria orviruses release chemokines that recruit characteristic setsof immune cells.44,45 For example, whereas rhinovirusesinduce the release of chemokines that result mainly inrecruitment of neutrophils (early in the course of infec-tion), EBV induces a set of chemokines that result inrecruitment of B cells, natural killer (NK) cells, and bothCD41 and CD81 T cells.45 It is of interest that almostmutually exclusive sets of chemokines are induced bycytokines associated with TH1 (IL-12 and IFN-g) versusTH2 (IL-4 and IL-13) immune responses, indicating a tightinterplay between cytokines and chemokines in determin-ing the type of immune response to specific infectiouschallenges generated under different conditions.46 Thespecificity of such cellular responses is strongly influencedby the chemokines released from specific tissues, thevascular adhesion molecules expressed in those tissues,the chemokine receptors expressed by different popula-tions of leukocytes, and the specific adhesion moleculesexpressed by leukocytes.44-46

Modulation of chemokine function can occur throughseveral mechanisms. Chemokines themselves can bepotentiated or inactivated by tissue proteases, includingtissue peptidases and matrix metalloproteases.47 Heparinsulfate–related proteoglycans on endothelial cell surfacestether chemokines locally, where they can most efficientlyactivate circulating leukocytes for adhesion (see below).However, similar proteoglycans free in the extracellularenvironment can bind and sequester chemokines, keepingthem from interacting with their cellular receptors.48,49

Finally, in addition to the well-described use of chemokinereceptors as coreceptors for viral entry by HIV-1, otherviruses, especially members of the herpesvirus family,encode soluble decoy receptors that compete with nativehost receptors for chemokine binding, thereby disruptingnormal host responses.49,50

NK cells

NK cells are an important cellular feature of innateimmunity. They are lymphoid cells that do not expressclonally distributed receptors, such as T-cell receptorsor surface immunoglobulin, for specific antigens.51 Theyrespond in an antigen-independent manner to help containviral infections before the development of adaptive im-mune responses, and they aid in the control of malignant

tumors. NK cells are found in the peripheral circulationand in the spleen and bone marrow. Like many otherleukocytes, they can be recruited to sites of inflammationby chemokines and other chemoattractants. They appearto be important for the control of tumors in vivo and serve acritical function in host defense against viral infections,especially those caused by members of the herpesvirusfamily.51,52 Activated NK cells also are an importantsource of IFN-g, which limits tumor angiogenesis andpromotes the development of specific protective immuneresponses.51,52

Regulation of NK cell activity involves a balancebetween activating and inhibitory signals. Several cyto-kines can activate NK cell proliferation, cytotoxicity, orIFN-g production, including IL-12, IL-15, IL-18, IL-21,and IFN-ab.51 Activating signals through other receptorson NK cells, such as NKG2D, can lead to cytotoxicity,cytokine production, or both, depending on the receptor’sassociation with distinct intracellular adaptor proteins thatsignal through different kinases.51,53 Other molecules onNK cells can act as either costimulatory or adhesionreceptors, including CD27, CD28, CD154 (CD40 ligand),and lymphocyte function-associated antigen 1 (LFA-1)(CD11a/CD18).51-53 Additionally, FcgRIII (CD16) cancontribute to NK cell cytotoxic activity through mecha-nisms that include antibody-dependent cell cytotoxicity.51

NK cells are able to distinguish normal cells of self originthrough receptors that recognize specific MHC class Imolecules. Activation of such receptors provides aninhibitory signal that protects healthy host cells fromNK cell–mediated lysis. Virus-infected cells and malig-nant cells often express MHC class I molecules at reducedlevels and thus are less able to generate inhibitory NK cellsignals, rendering them more susceptible to attack by NKcells.51,52 NK cell inhibitory receptors, which are not wellcharacterized, appear to contain intracytoplasmic tyro-sine-based inhibition motifs and antagonize NK cellactivation pathways through protein tyrosine phospha-tases.51,54 Thus the regulation of the phosphorylation stateof specific tyrosine residues by activating kinases andinhibitory phosphatases appears to be a pivotal determi-nant of NK cell activation.

NK cells kill infected or malignant cells through therelease of perforin and granzymes from granular storagecompartments and through binding of the death receptorsFas and TRAIL-R on target cells through their respectiveNK cell ligands.51,55 The mechanisms by which perforinand granzymes mediate target cell death are not fullyunderstood. The best available evidence suggests thatperforin and one or more of 5 human granzymes releasedalong with perforin from cytotoxic granules of NK cellsassociate with the cell membranes of target cells, eitherby binding through the mannose 6-phosphate receptor orthrough another mechanism that remains to be defined.One or more of the granzymes appears to activate intra-cellular pathways leading to target apoptosis throughpathways that involve the mitochondria, caspases, orboth.56 Separately, binding of the death receptors alsoactivates caspases, causing target cell apoptosis.51 It is


AUGUST 2005

244 Tosi

Review

sand

featu

rearticle

s

notable that although some tumor cells do not express Fas,NK cells can induce Fas expression on these targets byreleasing IFN-g and then proceed to kill them by bindingto the newly expressed Fas.51

NK cells engage in several kinds of interactions withother cells of the immune system, including dendritic cellsand other antigen-presenting cells. Dendritic cells caninfluence the proliferation and activation of NK cells boththrough release of cytokines, including IL-12, and throughcell-surface interactions, including CD40/CD40 ligand,LFA-1/intercellular adhesion molecule 1 (ICAM-1), andCD27/CD70.57 In return, NK cells can provide signals thatresult in either dendritic cell maturation or apoptosis.51,52

The complement system

The complement system is made up of at least 30proteins in serum or at cell surfaces. Most of these proteinsare made in the liver or, to a lesser extent, by mononuclearphagocytes. Activation of the complement cascade by oneor more of 3 distinct pathways leads to the evolution ofitsmain effector functions:microbial opsonization, phago-cyte recruitment, and bacteriolysis. All 3 activation path-ways act at a microbial surface to assemble a convertasethat cleaves C3 to form C3b, which in turn binds to thetarget surface, either as an opsonin or to help activate C5and the remainder of the cascade.58-61

Complement activation pathways. The classical path-way ordinarily is activated by IgG or IgM bound tomicrobial surface antigens. Antigen binding makes a siteon the Fc portion of the antibody available to bind C1q,which in turn binds C1r and C1s. C1qrs catalyzes thecleavage and binding of C4 and C2, in the form of C4b2a,the classical pathway C3 convertase. This convertase thencleaves C3 to formC3a, which is released, and C3b, whichremains bound on the target surface, and the cascadebeyond C3 continues.59

The recently characterized mannan-binding lectin(MBL) pathway is similar to the classical pathway butdoes not involve antibodies.60 MBL is a serum protein ofthe collectin family with structural and functional simi-larities to C1q and binds to mannose-containing carbohy-drates on microbial surfaces. Subsequently, MBL and theMBL-associated serine proteases, which have structuraland functional similarities to C1r and C1s, form a complexthat activates C4, with sequential binding of C4b andC2a and formation of C4b2a, the classical pathway C3convertase.60 C3 is activated, and the cascade proceedsas described.

The alternative pathway is vital to the host as a separatemeans by which C3 can be activated before developmentof specific antibodies.59,61 A low constitutive level ofhydrolysis of the thioester of C3 in the fluid phaseproduces an activated form of C3, C3(H2O). This acti-vated form of C3 can bind factor B, and the latter iscleaved by factor D to form the fluid-phase C3 convertaseC3(H2O)Bb. The constitutive presence of small amountsof this convertase in the fluid phase ensures that someC3b always is available to initiate the alternative path-way at microbial surfaces.59,61 Factor B binds to surface

C3b and undergoes proteolytic cleavage by factor D toform C3bBb, the alternative pathway C3 convertase.Properdin stabilizes the convertase, which producesmore C3b, establishing the C3 amplification loop of thealternative pathway and activating the remainder of thecascade.59 Microbes that bear large amounts of surfacesialic acid are usually poor activators of the alternativepathway because factor H outcompetes factor B for C3bbinding at such surfaces.59,61 No convertase or amplifica-tion loop is created because factor H allows C3b cleavageby factor I to form iC3b, the only function of which isopsonic.59 Nonactivators of the alternative pathway aresome of the most successful pathogens in nonimmunehosts. They include K1 Escherichia coli, groups A andB streptococci, Streptococcus pneumoniae, Neisseriameningitidis, Haemophilus influenzae type b, and somesalmonellae.62

Complement effector functions. Opsonization (fromthe Greek ‘‘to cater or prepare’’) facilitates the removal ofmicroorganisms from the circulation by macrophages inthe liver and spleen and from tissue sites by neutrophilsand tissue macrophages.58,59 Recognition and attachmentof surface-bound C3b and iC3b on microbes by the type1 and type 3 phagocytic complement receptors, CR1(CD35) and CR3 (CD11b/CD18), respectively, activatesingestion and intracellular killing of the organisms.59

Antibodies, themselves important opsonins, direct thelocalization of C3b binding on microorganisms throughthe classical pathway. This is important for encapsulatedbacteria because the capsule’s presence as a barrier meansthat only C3b bound at the capsular surface by specificanticapsular antibody will be accessible to phagocytereceptors.63

The free cleavage fragments of C3 and C5 can promotehost inflammatory responses. C3a stimulates marrowrelease of granulocytes, and C5a serves as a potentchemoattractant for monocytes, neutrophils, and eosino-phils. C5a also stimulates expression of CR1 and CR3,aggregation, and microbicidal activity of phagocytes.C5a-induced neutrophil aggregation and stasis in thepulmonary circulation can contribute to the respiratorydistress syndrome associated with sepsis.64 C4a, C3a, andespecially C5a are anaphylotoxins that can induce releaseof histamine from mast cells and basophils, causingincreased vascular dilatation and permeability.64 In largeamounts, they can contribute to the pathophysiology ofseptic shock.64

When C5a is released by the classical or alternativepathway C5 convertase, C5b is bound at the target surface.The terminal complement proteins C5b, C6, C7, C8, andC9 assemble sequentially to form the membrane attackcomplex (MAC), which can kill and lyse target cells,especially gram-negative bacteria, by penetrating theirouter membranes.59,61 The C5b-C8 complex serves as apolymerization site for several molecules of C9.61,65

Although C9 is not essential to membrane penetration,its presence as poly-C9 allows it to proceed more effi-ciently.59,65 The MAC also can lyse certain virus-infectedhost cells and some enveloped viruses directly.66



Tosi 245

Review

sand

featu

rearticles

Complement fragments can modulate other parts of theimmune response, both directly by binding to CR1, CR2,and CR3 on the surfaces of T cells, B cells, and other cellsinvolved in antigen recognition and indirectly by stimu-lating the synthesis and release of cytokines.67 For exam-ple, the C3b cleavage product, C3dg, when covalentlybound to antigen, brings the antigen close to B cells bybinding to B-cell CR2 (CD21); C3 influences antigenlocalization within germinal centers, anamnestic re-sponses, and isotype switching; and C1, C2, C4, and C3are important for normal antibody responses.68,69

Phagocytes

The first recognized cellular mechanism of host defensewas the accumulation of phagocytic host cells around aforeign body in starfish observed by Metchnikoff.70

Polymorphonuclear leukocytes (PMNs), the most abun-dant circulating phagocytes in the human host, will serveas a model for discussing phagocyte functions. These cellsconstitute a major line of defense against invading bacte-rial and fungi. The proliferation of myeloid marrowprogenitors and their differentiation into mature progenyare regulated by specific growth factors and cyto-kines.71,72 The normal half-life of circulating PMNs isapproximately 8 to 12 hours.73 In the absence of activeinfection, most PMNs leave the circulation through thegingival crevices and the lower gastrointestinal tract,where the resident flora stimulate ongoing local extrava-sation of PMNs, a process that helps maintain the integrityof these tissues.74 In response to invasive bacterial infec-tion, circulating PMNs engage in 3 major functions: (1)migration to the site of infection, (2) recognition andingestion of invading microorganisms, and (3) killing anddigestion of these organisms.

Phagocyte recruitment to infected sites. Activationof endothelial cells that line the microvessels of acutelyinfected tissue occurs through locally produced cytokines,eicosanoid compounds, and microbial products.75 As aresult, the endothelial cells rapidly upregulate their surfaceexpression of P-selectin and then E-selectin.75,76 Theseselectins engage in lectin-like interactions with thefucosylated tetrasaccharide moiety sialyl Lewis X, whichis presented on constitutively expressed glycoproteins onPMNs, including L-selectin and P-selectin glycoproteinligand 1.76 These early interactions slow the PMNs in thisfirst adhesive phase of leukocyte recruitment, sometimesdescribed as ‘‘slow rolling.’’75,77 Within several hours,newly synthesized ICAM-1 is expressed at the endothelialsurface.75,77,78 The slowly rolling PMNs are activated bytransient selectin-mediated interactions and locally pro-duced mediators, especially endothelium-derived chemo-kines, such as IL-8.77 These chemokines are mosteffective in activating the PMNs when they are boundby complex proteoglycans at the endothelial cell surface.48

The activated PMNs then increase the surface expression,binding avidity, or both of the b2-integrins LFA-1 andmacrophage antigen-1 (Mac-1) that interact with endo-thelial cell ICAM-1 in this second, firm adhesion phasemediated by integrin-ICAM interactions, which is also

necessary for transendothelial migration of thePMNs.75,77,79,80 Other chemoattractants, such as C5a,N-formyl bacterial oligopeptides, and leukotrienes (eg,leukotriene B4) that diffuse from the site of infection fur-ther activate PMNs and provide a chemotactic gradientfor PMN migration into tissue.41,80,81 The receptors forthese chemoattractants, like the chemokine receptors, areG protein–coupled and share a 7-transmembrane domainstructure.41,81 They constitute important sensory mecha-nisms of the PMNs for activating adhesion, directionalorientation, and the contractile protein–dependent lateralmovement of adhesion sites in the PMN membranenecessary for migration.41,81-83 Although the specificstimuli and adhesion molecules might vary, this generalscheme applies to the local recruitment of virtually allcirculating cells of the immune system.44-46

Phagocytosis. After PMNs reach the site of infection,they must recognize and ingest, or phagocytose, theinvading bacteria. Opsonization, especially with IgG andfragments of C3, greatly enhances phagocytosis.58,63

Although nonopsonic phagocytosis can occur, only opso-nin-mediated phagocytosis is considered here. CR1 andCR3 are the main phagocytic receptors for opsonic C3band iC3b, respectively.79 When PMNs are activated bychemoattractants or other stimuli, CR1 and CR3 arerapidly translocated to the cell surface from intracellularstorage compartments, increasing surface expression up to10-fold.79 Note that CR3 is identical to the adhesion-mediating integrin Mac-1.79 CR1 and CR3 act synergis-tically with receptors for the Fc portion of antibodies,especially IgG.58,59 Phagocytic cells can express up to 3different types of IgG Fc receptors, or FcgRs, all of whichcan mediate phagocytosis.84 FcgRI (CD64) is a high-affinity receptor that is expressed mainly on mononuclearphagocytes.84 The 2 FcgRs ordinarily expressed on circu-lating PMNs are FcgRII (CD32) and FcgRIII (CD16).84

FcgRII is conventionally anchored in the cell membrane,exhibits polymorphisms that determine preferences forbinding of certain IgG subclasses, and can directly activatePMN oxidative burst activity.84,85 FcgRIII is expressed onPMNs as a glycolipid-anchored protein, although it isanchored conventionally on NK cells and macrophages.84

Most phagocytes also express IgA receptors. The bestcharacterized, CD89, binds monomeric IgA and promotesphagocytosis and killing of IgA-opsonized bacteria.86

The engagement of phagocyte receptors with opsoninsbound on microbes locally activates cytoskeletal contrac-tile elements, leading to invagination of the membrane atthe site of initial engagement and extension of pseudopodsaround the microbe. The ligation of additional opsonin-receptor pairs leads to engulfment of the microbe within asealed phagosome.87 This is followed by fusion of thephagosome with lysosomal compartments containing thephagocyte’s array of microbicidal products.Phagocyte microbicidal mechanisms. The microbi-

cidal mechanisms of PMNs usually are categorized aseither oxygen dependent or oxygen independent. Oxygen-dependent microbicidal mechanisms of phagocytes de-pend on a complex enzyme, reduced nicotinamide adenine


AUGUST 2005

246 Tosi

Review

sand

featu

rearticle

s

dinucleotide phosphate (NADPH) oxidase, which con-verts molecular oxygen (O2) into superoxide anion(O2

2).88 This enzyme is assembled at the activated cellmembrane from 6 or more components that include acytochrome (a- and b-subunits, designated gp91phox andp22phox, respectively), a flavoprotein, and a quinone, all ofwhich are associated with the cell membrane, and at least 2cytoplasmic proteins, p47phox and p67phox (‘‘phox’’ refersto phagocyte oxidase), that assemble with the membrane-associated components to form the active enzyme com-plex.88,89 Each of the main oxidant products derived fromNADPH activity and subsequent reactions exhibitsmicrobicidal activity, including the earliest products,O2

2 and H2O2, which are less potent than the downstreamproducts hypochlorite (OCl2) and chloramines (NH3ClandRNH2Cl), with chloramines being themost stable.90,91

Oxygen-independent microbicidal activity of PMNsresides mainly in a group of proteins and peptides storedwithin primary (azurophilic) granules. Lysozyme is con-tained in both the primary and the secondary (specific)granules of PMNs.92 It cleaves important linkages in thepeptidoglycan of bacterial cell walls and can act in concertwith the complement MAC.58 The primary granulescontain several cationic proteins with important microbi-cidal activity. A 59-kd protein, bactericidal/permeability-increasing protein, is active against only gram-negativebacteria.93 Smaller arginine- and cysteine-rich peptides,the a-defensins, similar to the b-defensins of epithelialcells, are active against a range of bacteria, fungi,chlamydiae, and enveloped viruses.6 Other related mole-cules include cathelicidin and a group of peptides calledp15s.4,94

SUMMARY

An overview of most of the main features of innateimmunity discussed above, along with some of theirimportant interactions, is diagrammed in Fig 1. Severallevels of interaction are depicted, from initial host-path-ogen contact, through a variety of activating signals, to theattack by host effector mechanisms on pathogenic targets.Initial contact between the host and microbes or theirproducts might result in viral infection of cells, activationof TLRs on macrophages and epithelial cells, and activa-tion of the alternative pathway or mannose-binding lectinpathway of complement. The resulting activation signals,including cytokines (eg, IL-12, TNF-a, and IL-1), che-mokines, and products of the complement cascade mobi-lize both cellular (NK cells and phagocytes) and humoral(antimicrobial peptides and MAC) effectors that attacktheir respective microbial targets.

REFERENCES

1. Janeway CA, Medzhitov R. Innate immune recognition. Annu Rev

Immunol 2002;20:197-216.

2. Padlan EA. Anatomy of the antibody molecule. Mol Immunol 1994;31:

169-217.

3. Garcia KC, Teyton L, Wilson IA. Structural basis of T cell recognition.

Annu Rev Immunol 1999;17:369-97.

4. Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat Rev

Immunol 2003;3:710-20.

5. Oppenheim JJ, Biragyn A, Kwak LW, Yang D. Roles of antimicrobial

peptides such as defensins in innate and adaptive immunity. Ann Rheum

Dis 2003;62(suppl 2):17-21.

6. Vora P, Youdim A, Thomas LS, Fukata M, Tesfay SY, Lukasek K, et al.

Beta-defensin-2 expression is regulated by TLR signaling in intestinal

epithelial cells. J Immunol 2004;173:5398-405.

FIG 1. Innate immunity: responses to first contact. Diagrammed are important host responses to infection that

are independent of specific cell-mediated immunity or antibodies. Initial contact between the host and

microbes or their products results in a range of activating signals that mobilize both cellular and humoral

effectors for attack on their respective microbial targets. Components of the host response are highlighted in

blue. MF, Macrophages; AP, alternative pathway; MBLP, mannose-binding lectin pathway.



Tosi 247

Review

sand

featu

rearticles

7. Hertz CJ, Wu Q, Porter EM, Zhang YJ, Weismuller KH, Godowski PJ,

et al. Activation of Toll-like receptor 2 on human tracheobronchial

epithelial cells induces the antimicrobial peptide human beta defensin-2.

J Immunol 2003;171:6820-6.

8. Hoover DM, Rajashankar KR, Blumenthal R, Puri A, Oppenheim JJ,

Chertov O, et al. The structure of human beta-defensin-2 shows evidence

of higher order oligomerization. J Biol Chem 2000;275:32911-8.

9. Campos MA, Vargas MA, Regueiro V, Llompart CM, Alberti S,

Bengoechea JA. Capsule polysaccharide mediates bacterial resistance

to antimicrobial peptides. Infect Immun 2004;72:7107-14.

10. Jin T, Bokarewa M, Foster T, Mitchell J, Higgins J, Tarkowski A.

Staphylococcus aureus resists human defensins by production of staph-

ylokinase, a novel bacterial evasion mechanism. J Immunol 2004;172:

1169-76.

11. Perregaux DG, Bhavsar K, Contillo L, Shi J, Gabel CA. Antimicrobial

peptides initiate IL-1 beta posttranslational processing: a novel role

beyond innate immunity. J Immunol 2002;168:3024-32.

12. Hoover DM, Boulegue C, Yang D, Oppenheim JJ, Tucker K, Lu W, et al.

The structure of human macrophage inflammatory protein-3alpha/

CCL20. Linking antimicrobial and CC chemokine receptor-6-binding

activities with human beta-defensins. J Biol Chem 2002;277:37647-54.

13. Niyonsaba F, Ogawa H, Nagaoka I. Human beta-defensin-2 functions as

a chemotactic agent for tumour necrosis factor-alpha-treated human

neutrophils. Immunology 2004;111:273-81.

14. Biragyn A, Ruffini PA, Leifer CA, Klyushnenkova E, Shakhov A,

Chertov O, et al. Toll-like receptor 4-dependent activation of dendritic

cells by beta-defensin 2. Science 2002;298:1025-9.

15. Niyonsaba F, Iwabuchi K, Matsuda H, Ogawa H, Nagaoka I. Epithelial

cell-derived human beta-defensin-2 acts as a chemotaxin for mast cells

through a pertussis toxin-sensitive and phospholipase C-dependent

pathway. Int Immunol 2002;14:421-6.

16. Scott MG, Vreugdenhil AC, Buurman WA, Hancock RE, Gold MR.

Cutting edge: cationic antimicrobial peptides block the binding of

lipopolysaccharide (LPS) to LPS binding protein. J Immunol 2000;

164:549-53.

17. Palaniyar N, Nadesalingam J, Reid KB. Pulmonary innate immune

proteins and receptors that interact with gram-positive bacterial ligands.

Immunobiology 2002;205:575-94.

18. Steinstraesser L, Tippler B, Mertens J, Lamme E, Homann HH,

Lehnhardt M, et al. Inhibition of early steps in the lentiviral replication

cycle by cathelicidin host defense peptides. Retrovirology 2005;2:2-13.

19. Yang D, Chen Q, Schmidt AP, Anderson GM, Wang JM, Wooters J,

et al. LL-37, the neutrophil granule- and epithelial cell-derived cath-

elicidin, utilizes formyl peptide receptor-like 1 (FPRL1) as a receptor

to chemoattract human peripheral blood neutrophils, monocytes, and

T cells. J Exp Med 2000;192:1069-74.

20. Schutte BC, Mitros JP, Bartlett JA, Walters JD, Jia HP, Welsh MJ, et al.

Discovery of five conserved beta-defensin gene clusters using a compu-

tational search strategy. Proc Natl Acad Sci U S A 2002;99:2129-33.

21. Medzhitov R, Preston-Hurlburt P, Janeway CA. A human homologue of

the Drosophila Toll protein signals activation of adaptive immunity.

Nature 1997;388:394-7.

22. Zarember KA, Godowski PJ. Tissue expression of human toll-like

receptors and differential regulation of toll-like receptor mRNAs in

leukocytes in response to microbes, their products, and cytokines.

J Immunol 2002;168:554-61.

23. O’Neill LAJ. TLRs: Professor Mechnikov, sit on your hat. Trends

Immunol 2004;25:687-93.

24. Cohen J. The immunopathogenesis of sepsis. Nature 2002;420:885-91.

25. Kurt-Jones EA, Belko J, Yu C, Newburger PE, Wang J, Chan M, et al.

The role of toll-like receptors in herpes simplex infection in neonates.

J Infect Dis 2005;191:746-8.

26. Hayashi F, Smith KD, Ozinsky A, Hawn TR, Yi EC, Goodlett DR, et al.

The innate immune response to bacterial flagellin is mediated by Toll-

like receptor 5. Nature 2001;410:1099-103.

27. Bauer S, Kirschning CJ, Hacker H, Redecke V, Hausmann S, Akira S,

et al. Human TLR9 confers responsiveness to bacterial DNA via species-

specific CpG motif recognition. Proc Natl Acad Sci U S A 2001;98:

9237-42.

28. Kulka M, Alexopoulou L, Flavell RA, Metcalfe DD. Activation of mast

cells by double-stranded RNA: evidence for activation through Toll-like

receptor 3. J Allergy Clin Immunol 2004;114:174-82.

29. Lorenz E, Mira JP, Cornish KL, Arbour NC, Schwartz DA. A novel

polymorphism in the toll-like receptor 2 gene and its potential association

with staphylococcal infection. Infect Immun 2000;68:6398-401.

30. Lorenz E, Mira JP, Frees KL, Schwartz DA. Relevance of mutations in

the TLR4 receptor in patients with gram-negative septic shock. Arch

Intern Med 2002;162:1028-32.

31. Kishore SP, Bungum MK, Platt JL, Brunn GJ. Selective suppression of

Toll-like receptor 4 activation by chemokine receptor 4. FEBS Lett 2005;

579:699-704.

32. Borish LC, Steinke JW. Cytokines and chemokines. J Allergy Clin

Immunol 2003;111:5460-75.

33. Rubinstein M, Dinarello CA, Oppenheim JJ, Hertzog P. Recent advances

in cytokines, cytokine receptors and signal transduction. Cytokine

Growth Factor Rev 1998;9:175-81.

34. Oppenheim JJ, Feldman M. In: Oppenheim JJ, Feldman M, Durum SK,

Hirano T, Vilcek J, Nicola NA, editors. Cytokine reference. San Diego:

Academic Press; 2001. p. 3-20.

35. Beutler B, Cerami A. The biology of cachectin/TNFa—primary medi-

ator of the host response. Annu Rev Immunol 1989;7:625-55.

36. Fisher CJ Jr, Dhainaut JF, Opal SM, Pribble JP, Balk RA, Slotman GJ,

et al. Recombinant human interleukin 1 receptor antagonist in the

treatment of patients with sepsis syndrome. JAMA 1994;271:1836-43.

37. Abraham E, Wunderink R, Silverman H, Perl TM, Nasraway S, Levy H,

et al. Efficacy and safety of monoclonal antibody to human tumor

necrosis factor alpha in patients with sepsis syndrome. JAMA 1995;273:

934-41.

38. Vanichkin A, Patya M, Gazit A, Levitzki A, Novogrodsky A. Late

administration of a lipophilic tyrosine kinase inhibitor prevents lipopol-

ysaccharide and Escherichia coli-induced lethal toxicity. J Infect Dis

1996;173:927-33.

39. Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J,

et al. HMG-1 as a late mediator of endotoxin lethality in mice. Science

1999;285:248-51.

40. Rice TW, Bernard GR. Therapeutic intervention and targets for sepsis.

Annu Rev Med 2005;56:225-48.

41. Rossi D, Zlotnik A. The biology of chemokines and their receptors. Annu

Rev Immunol 2000;18:217-42.

42. Baggiolini M. Chemokines in pathology and medicine. J Intern Med

2001;250:91-104.

43. IUIS/WHO Subcommittee on Chemokine Nomenclature. Chemokine/

chemokine receptor nomenclature. Cytokine 2003;21:48-9.

44. Kim CH. Chemokine-chemokine receptor network in immune cell traf-

ficking.CurrDrugTargets ImmuneEndocrMetabolDisord2004;4:343-61.

45. Glass WG, Rosenberg HF, Murphy PM. Chemokine regulation of

inflammation during acute viral infection. Curr Opin Allergy Clin

Immunol 2003;3:467-73.

46. Bisset LR, Schmid-Grendelmeier P. Chemokines and their receptors in

the pathogenesis of allergic asthma: progress and perspective. Curr Opin

Pulm Med 2005;11:35-42.

47. McQuibban GA, Butler GS, Gong JH, Bendall L, Power C, Clark-Lewis

I, et al. Matrix metalloproteinase activity inactivates the CXC chemokine

stromal cell-derived factor-1. J Biol Chem 2001;276:43503-8.

48. Kuschert GS, Coulin F, Power CA, Proudfoot AE, Hubbard RE,

Hoogewerf AJ, et al. Glycosaminoglycans interact selectively with

chemokines and modulate receptor binding and cellular responses.

Biochemistry 1999;38:12959-68.

49. Comerford I, Nibbs RJ. Post-translational control of chemokines: a role

for decoy receptors? Immunol Lett 2005;96:163-74.

50. Rosenkilde MM. Virus-encoded chemokine receptors—putative novel

antiviral drug targets. Neuropharmacology 2005;48:1-13.

51. Smyth MJ, Cretney E, Kelly JM, Westwood JA, Street SE, Yagita H,

et al. Activation of NK cell cytotoxicity. Mol Immunol 2005;42:501-10.

52. Cerwenka A, Lanier LL. Natural killer cells, viruses and cancer. Nat Rev

Immunol 2001;1:41-9.

53. Vivier E, Nunes JA, Vely F. Natural killer cell signaling pathways.

Science 2004;306:1517-9.

54. Regunathan J, Chen Y, Wang D, Malarkannan S. NKG2D receptor-

mediated NK cell function is regulated by inhibitory Ly49 receptors.

Blood 2005;105:233-40.

55. Sato K, Hida S, Takayanagi H, Yokochi T, Kayagaki N, Takeda K, et al.

Antiviral response by natural killer cells through TRAIL gene induction

by IFN-alpha/beta. Eur J Immunol 2001;31:3138-46.


AUGUST 2005

248 Tosi

Review

sand

featu

rearticle

s

56. Trapani JA, Smyth MJ. Functional significance of the perforin/granzyme

cell death pathway. Nat Rev Immunol 2002;2:735-47.

57. Degli-Esposti MA, Smyth MJ. Close encounters of different kinds:

dendritic cells and NK cells take centre stage. Nat Rev Immunol 2005;5:

112-24.

58. Joiner KA, Brown EJ, Frank MM. Complement and bacteria: chemistry

and biology in host defense. Annu Rev Immunol 1984;2:461-91.

59. Berger M, Frank MM. The serum complement system. In: Stiehm ER,

Ochs HD, Winkelstein JA, editors. Immunologic disorders in infants and

children. 5th ed. Philadelphia: Elsevier Saunders; 2004. p. 157-87.

60. Peterson SV, Thiel S, Jensenius JC. The mannan-binding lectin pathway

of complement activation: biology and disease association. Mol Immunol

2001;38:133-49.

61. Walport MJ. Complement. First of two parts. N Engl J Med 2001;344:

1058-66.

62. Joiner KA. Complement evasion by bacteria and parasites. Ann Rev

Microbiol 1988;42:201-30.

63. Hostetter MK. Serotypic variations among virulent pneumococci in

deposition and degradation of covalently bound C3b: implications for

phagocytosis and antibody production. J Infect Dis 1986;153:682-93.

64. Hugli TE. Structure and function of the anaphylatoxins. Springer Semin

Immunopathol 1984;7:193-220.

65. Bhakdi S, Tranum-Jensen J. Complement lysis: a hole is a hole. Immunol

Today 1991;12:318-20.

66. Cooper NR, Nemerow GR. Complement-dependent mechanisms of virus

neutralization. In: Ross GD, editor. Immunobiology of the complement

system. Orlando (FL): Academic Press; 1986. p. 139-62.

67. Erdei A, Fust G, Gergely J. The role of C3 in the immune response.

Immunol Today 1991;12:332-7.

68. Ochs HD, Nonoyama S, Zhu Q, Farrington M, Wedgwood RJ. Regu-

lation of antibody responses: the role of complement and adhesion

molecules. Clin Immunol Immunopathol 1993;67(suppl):S33-40.

69. Bottger EC, Bitter-Suermann D. Complement and the regulation of

humoral immune responses. Immunol Today 1987;8:261-4.

70. Metchnikoff E. Immunity in the infectious diseases. 1st ed. New York:

Macmillan Press; 1905.

71. Bainton DF. Developmental biology of neutrophils and eosinophils. In:

Gallin JI, Goldstein IM, Snyderman R, editors. Inflammation: basic

principles and clinical correlates. 2nd ed. New York: Raven Press; 1992.

p. 303-24.

72. Tenen DG, Hromas R, Licht JD, Zhang DE. Transcription factors,

normal myeloid development, and leukemia. Blood 1997;90:489-519.

73. Walker RI, Willemze R. Neutrophil kinetics and the regulation of

granulopoiesis. Rev Infect Dis 1980;2:282-92.

74. Anderson DC, Schmalsteig F, Finegold MJ, Hughes BJ, Rothlein R,

Miller LJ, et al. The severe and moderate phenotypes of heritable Mac-1,

LFA-1, p150, 95 deficiency: Their quantitative definition and relation to

leukocyte dysfunction and clinical features. J Infect Dis 1985;152:668-89.

75. Butcher EC. Leukocyte-endothelial cell recognition: Three (or more)

steps to specificity and diversity. Cell 1991;67:1033-6.

76. Ley K. The role of selectins in inflammation and disease. Trends Mol

Med 2003;9:263-8.

77. Ley K. Integration of inflammatory signals by rolling neutrophils.

Immunol Rev 2002;186:8-18.

78. Smith CW, Marlin SD, Rothlein R, Toman C, Anderson DC. Cooper-

ative interactions of LFA-1 and Mac-1 with intercellular adhesion

molecule-1 in facilitating adherence and transendothelial migration of

human neutrophils in vitro. J Clin Invest 1989;83:2008-17.

79. Berger M, O’Shea J, Cross AS, Folks TM, Chused TM, Brown EJ, et al.

Human neutrophils increase expression of C3bi as well as C3b receptors

upon activation. J Clin Invest 1984;74:1566-71.

80. Seo SM, McIntire LV, Smith CW. Effects of IL-8, Gro-alpha, and LTB4

on the adhesive kinetics of LFA-1 and Mac-1 on human neutrophils. Am

J Physiol Cell Physiol 2001;281:C1568-78.

81. Gerard C, Gerard N. C5a anaphylatoxin and its seven transmembrane-

segment receptor. Annu Rev Immunol 1994;12:775-808.

82. Anderson DC, Hughes BJ, Smith CW. Abnormal mobility of neonatal

polymorphonuclear leukocytes. Relationship to impaired redistribution

of surface adhesion sites by chemotactic factor or colchicine. J Clin

Invest 1981;68:863-74.

83. Stossel TP. The mechanical responses of white blood cells. In: Gallin JI,

Goldstein IM, Snyderman R, editors. Basic principles and clinical

correlates. New York: Raven Press; 1992. p. 459-75.

84. Unkeless JC, Shen Z, Lin CW, DeBeus E. Function of human Fc gamma

RIIA and Fc gamma RIIIB. Semin Immunol 1995;7:37-44.

85. van der Pol WL, van de Winkel JGJ. IgG receptor polymorphisms: risk

factors for disease. Immunogenetics 1998;48:222-32.

86. Hostoffer RW, Krukovets I, Berger M. Enhancement by tumor necrosis

factor-a of Fca receptor expression and IgA-mediated superoxide

generation and killing of Pseudomonas aeruginosa by polymorphonu-

clear leukocytes. J Infect Dis 1994;170:82-7.

87. Stossel TP. Phagocytosis. Prog Clin Biol Res 1977;13:87-102.

88. Clark RA, Leidal KG, Pearson DW, Nauseef WM. NADPH oxidase of

human neutrophils. Subcellular localization and characterization of an

arachidonate activatable superoxide-generating system. J Biol Chem

1987;262:4065-74.

89. Borregaard N, Heiple JM, Simons ER, Clark RA. Subcellular localiza-

tion of the b cytochrome component of the human neutrophil microbi-

cidal oxidase: translocation during activation. J Cell Biol 1983;97:52-61.

90. Root RK, Cohen MS. The microbicidal mechanisms of human neutro-

phils and eosinophils. Rev Infect Dis 1981;3:565-98.

91. Tosi MF. Immunologic and phagocytic responses to infection. In: Feigin

RD, Cherry JD, Demmler GJ, Kaplan S, editors. Textbook of pediatric

infectious diseases. 5th ed. New York: WB Saunders; 2004. p. 20-62.

92. Spitznagel JK, Dalldorf FG, Leffell MS, Folds JD, Welsh IR, Cooney

MH, et al. Character of azurophil and specific granules purified from

human polymorphonuclear leukocytes. Lab Invest 1974;30:774-85.

93. Weiss J, Victor M, Elsbach P. Role of charge and hydrophobic

interactions in the action of the bactericidal/permeability-increasing

protein of neutrophils on gram-negative bacteria. J Clin Invest 1983;

71:540-9.

94. Levy O, Weiss J, Zarember K, Ooi CE, Elsbach P. Antibacterial 15-kDa

protein isoforms (p15s) are members of a novel family of leukocyte

proteins. J Biol Chem 1993;268:6058-63.



Tosi 249

Review

sand

featu

rearticles

Continuing Medical Education examination

Innate immune responses to infection

Instructions for category 1 Continuing Medical Education credit

The American Academy of Allergy, Asthma and Immunology is accredited as a provider of Continuing MedicalEducation (CME) by the Accreditation Council for Continuing Medical Education.

Test ID no.: mai0063Contact hours: 1.0Expiration date: July 31, 2006

Category 1 credit can be earned by reading the text material and taking this CME examination online. For completeinstructions, visit the Journal’s Web site at www.mosby.com/jaci.

Learning objectives: ‘‘Innate immune responses to infection’’

1. To achieve a greater and more current understanding of the nature and scope of innate immune responses to infection.

2. To develop an enhanced appreciation for the range of interactions among various components of innate immunity.

3. To be able to appreciate the specific microbial targets of specific effector mechanisms of innate immunity.

CME itemsQuestion 1. The main host defense functions of C5a and

the membrane attack complex of the complement systemare most closely reproduced by —

A. TNF-a and phagocytes.B. chemokines and defensins.C. Toll-like receptors and NK cells.D. IL-12 and dendritic cells.

Question 2. NK cells are activated directly by virus-infected target cells —

A. via specific receptors for viral antigens.B. through specific receptors for IL-12.C. because these targets express fewer molecules that

bind to inhibitory receptors.D. in spite of increased MHC class I surface expres-

sion.

Question 3. The cellular nature of leukocytic infiltratesinto infected tissue is most influenced by —

A. the size of the microbial inoculum.B. preexisting specific antibodies.C. the chemokine receptors expressed by the infil-

trating leukocytes.D. The defensins released at the infected site.

Question 4. The innate immune mechanisms that bestepitomize rapid responses on first contact with microbesare —

A. Toll-like receptors and complement.B. chemokines and NK cells.C. phagocytes and defensins.D. chemokines and phagocytes.


Review

sand

featu

rearticle

s

http://www.mosby.com/jaci

Molecular mechanisms in allergy and clinical immunologySeries editors: William T. Shearer, MD, PhD, Lanny J. Rosenwasser, MD, and Bruce S. Bochner,MD

EBV the prototypical human tumorvirus—just how bad is it?

David A. Thorley-Lawson, PhD Boston, Mass

This activity is available for CME credit. See page 32A for important information.

EBV was the first candidate human tumor virus. It is found

in several human cancers, particularly lymphomas and

carcinomas, and has potent transforming activity in vitro. Yet

the virus persists benignly for the lifetime of more than 90% of

the human population. Thus it seems that EBV has the

potential to be highly pathogenic yet rarely manifests this

potential. Studies over the last several years show this is

because the virus actually persists in resting memory B cells

and not proliferating cells. EBV needs its growth-promoting

ability to gain access to the memory compartment but has

evolved to minimize its oncogenic potential. These studies also

reveal that the different EBV-associated tumors apparently

arise from different and discrete stages in the life cycle of

B cells latently infected with EBV. This raises the question of

how actively EBV participates in the development of human

tumors. Does the virus cause the disease, or is it simply a

passenger? In the case of immunoblastic lymphoma in the

immunosuppressed patient, the virus almost certainly plays a

causative role, but in other cases, such as Burkitt’s lymphoma,

the contribution of EBV remains less clear. (J Allergy Clin

Immunol 2005;116:251-61.)

Key words: Epstein-Barr virus, carcinoma, lymphoma, persistentinfection, latency, B cell, memory

EBV is well known because of its characteristic biol-ogy.1-3 If you define the success of a pathogen by thenumber and extent of hosts it infects, EBV is the mostsuccessful human pathogen because it latently infectsvirtually the whole human population and persists forlife.4 In tissue culture EBV is one of the most potenttransforming viruses,5,6 and it is found in several humancancers,1,3 yet for most of the population, it remains

benign. The collection of viral latent proteins expressed isdifferent in each tumor type (Table I). Sometimes all of theknown latent proteins are expressed, sometimes a limitedsubset, and sometimes only one.

Despite the apparent robustness with which the humanpopulation deals with EBV (>95% of all adults carry thevirus), the diseases caused by EBV indicate that thesituation is finely balanced. The first indication comesfrom X-linked lymphoproliferative disease.7 In this dis-ease persistent infection is not established because muta-tions in the SH2D1A gene8,9 cause acute EBV infection tobecome a fatal disease. Put melodramatically, a singlenucleotide change in the SH2D1A gene is all that preventsthe vast majority of the human race from dying of acuteEBV infection.

The second indication comes from the observation thatimmunologic disturbance, as a predisposing factor, is aunifying theme for all of the EBV B-cell lymphomas. Thisalso suggests that the regulation of EBV infection in Bcells is finely balanced. Disruptions can lead to deregula-tion and EBV-driven tumor development, even in other-wise healthy carriers of the virus. The clearest example ofthis is individuals who are immunosuppressed, such aspatients undergoing organ transplantation, who are iatro-genically immunosuppressed, or patients with AIDS, whoare immunosuppressed by HIV. These individuals are atrisk for EBV lymphomas that are aggressive and oftenfatal.10 This means that it is only courtesy of an activeimmune response that we are protected from fatal EBV-driven lymphoma. Yet there are some curious propertiesof these tumors that suggest the risk is not as high as mightbe expected. For example, not every immunosuppressed

Abbreviations usedBL: Burkitt’s lymphoma

CTL: Cytotoxic T lymphocyte

EBNA: EBV nuclear antigen

HD: Hodgkin’s disease

IM: Infectious mononucleosis

LMP: Latent membrane protein

NPC: Nasopharyngeal carcinoma

PTLD: Posttransplantation lymphoproliferative disease

From the Department of Pathology, Tufts University School of Medicine.

Supported by grants R01 CA65883, R01 A118757, and R01 A1062989.

Disclosure of potential conflict of interest: D. Thorley-Lawson has equity

ownership in EBVax.

Received for publication April 11, 2005; accepted for publication May 16,

2005.


Reprint requests: David A. Thorley-Lawson, PhD, the Department of

Pathology, Jaharis Building, Tufts University School of Medicine, 150

Harrison Ave, Boston, MA 02111. E-mail: David.Thorley-Lawson@

tufts.edu.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.038

251

Review

sand

featu

rearticles



patient has the tumors, and the tumors are frequentlyoligoclonal. This is not the expected outcome. If it weresimply a case of the immune system failing to control theEBV-infected cells, every immune-suppressed personshould fill up with multiple tumors because everybodycarries approximately 5 3 105 infected cells,11 andimmunosuppressed individuals carry perhaps 50 timesmore.12

Taken together, these observations raise several ques-tions. How does EBV persist benignly for the lifetime of ahuman despite its pathogenic potential? Why does EBVhave such potent and pathogenic properties if it hasevolved to persist for the lifetime of the human host itputs at risk by manifesting those properties? Where do theEBV-associated tumors come from, why do they havedifferent patterns of latent gene expression, and why doesdisruption of the immune system predispose to EBVlymphoma development? Lastly, what goes wrong in themaintenance of persistence that leads to EBV-associateddiseases?

The key to answering these questions comes from amodel of EBV persistence13,14 developed from the obser-vation that despite EBV’s transforming ability, it persistsin vivo in resting15 memory16 B cells that do not expressany viral proteins.17 This article will first briefly review thecomplete life cycle of EBV infection and then discuss howthe origins of EBV-associated tumors can be explained inthe context of thismodel, with special emphasis on the roleof an impaired immune response. Finally, the model willbe used to attempt to answer the questions posed above.

EBV PERSISTENCE IN VIVO

The essence of EBV’s behavior is that under normalconditions, it does not aberrantly deregulate the behaviorof infected B cells in vivo. It initiates, establishes, andmaintains persistent infection by subtly using virtuallyevery aspect of normal B-cell biology. Ultimately, thisallows the virus to persist within memory B cells for thelifetime of the host in a fashion that is nonpathogenic. Thethesis of this review is that EBV is not a natural tumor

virus and that it has developed strategies to minimize itspathogenic potential to the host.

Establishment

To understand EBV biology, it is first necessary tounderstand the biology of the B lymphocyte in themucosal lymphoepithelium of the tonsil (Fig 1). A sum-mary of normal mature B-cell biology and the proposedparallels with EBV is given in Figs 2 and 3, and a summaryof information on the different viral latency programs ispresented in Table I. The model has been described indetail elsewhere.13,14

The normal B-cell response. Environmental antigensentering the mouth are continuously sampled by theepithelium of the tonsil. Underneath the epithelium is abed of lymphoid tissue including large numbers of naivelymphocytes.18,19 If antigen is recognized by the antibodyon the surface of the naive B cell, it will bind and cause theB cell to become an activated blast and migrate into thefollicle to form a germinal center (Fig 2).20 Here the cellundergoes rounds of rapid proliferation associated withisotype switching and mutation of the immunoglobulingenes, followed by competitive selection for those withthe antibody that binds the antigen best. Those who lose inthe competition to bind antigen die by apoptosis.Ultimately, the surviving cells leave the germinal centeras memory cells primed to make a rapid response torechallenge with the antigen. This process requires, inaddition to the antigen, a signal to the B cell from anantigen-specific T helper cell.

The parallel with EBV. EBV also transits the epithe-lium and infects naive B cells21 in the underlying tissue,where it expresses a set of latent genes that cause the cellto become activated and proliferate as though it were re-sponding to antigen. This EBV transcription program (thegrowth program, Fig 2) involves 9 latent proteins, includ-ing nuclear antigens (EBV nuclear antigens [EBNAs]) andmembrane proteins (latent membrane proteins [LMPs]).2

These proteins have all the necessary activities to push theB cell to become an activated blast without any necessityfor external signaling. This cell migrates to the follicle,where the viral transcription program changes,22 such that

TABLE I. The EBV transcription programs in normal B cells and tumors

Transcription

program Genes expressed*

Infected normal

B-cell typey Function

Infected

tumor type

Growth EBNA1, 2, 3a, 3b, 3c, LP, LMP1,

LMP2a, and LMP2b

Naive Activate B cell Immunoblastic

lymphoma

Default EBNA1, LMP1, and LMP2a Germinal center Differentiate activated B

cell into memory

HD

Latency None Peripheral memory Allow lifetime persistence

EBNA1 only EBNA1 Dividing peripheral

memory

Allow virus in latency program

cell to divide

Burkitt’s

lymphoma

Lytic All lytic genes Plasma cell Replicate the virus in plasma cell

*Does not include the noncoding EBER and BART RNAs that are assumed to be ubiquitous but have not been rigorously identified in all of the

infected subtypes.

Except where indicated, the cell types are primarily restricted to the lymphoid tissue of the Waldeyer ring.


AUGUST 2005

252 Thorley-Lawson

Review

sand

featu

rearticle

s

only 3 of the latent proteins are expressed: EBNA1(required to replicate the viral DNA) and 2 membraneproteins, LMP1 and LMP2 (the default program, Fig 2).

The functions of LMP1 and LMP2 have evolved tosteer the latently infected B cell through the germinalcenter environment. LMP2 alone will push B cells to forma germinal center in the mucosal follicle23; LMP1 andLMP2 can drive immunoglobulin gene mutation23 andisotype switching24 (the defining markers of the germinalcenter), respectively, and LMP1 downregulates expres-sion of the germinal center regulatory transcription factorbcl-6,25 the signal for a memory cell to exit the germinalcenter.26 This implies coordinated expression of LMP1and LMP2, where LMP2 is turned on before and LMP1is turned on during the germinal center reaction. Thusconstitutive expression of LMP1 in the absence ofLMP2 blocks germinal center formation because the cellscan never turn on bcl-6, an essential step in germinalcenter formation.27

This explains why EBV has the ability to make cellsproliferate, despite the fact that this puts the host at risk forneoplastic disease. Essentially it has to because this isthe mechanism, activation followed by differentiation, bywhich a normal B cell enters the B-cell memory pool.

Maintenance

Once in the periphery, the latently infected cells shutdown all viral protein expression (the latency program)and appear to be maintained as normal memory B cells.17

In the early stages of acute infectious mononucleosis (IM;primary EBV infection in the adult), the number of suchcells in the blood can reach staggering proportions, with

50% or more of all memory cells being infected.28

However, the numbers decrease rapidly (half-life of 7days; Hadinoto and Thorley-Lawson, unpublished data)for the first 2 months and then more steadily after that,until by 1 year there are typically only about 1 in 105 to 106

infected memory B cells. After this time, the level ofinfected cells appears to be relatively stable over manyyears.11 This presumably represents a balance between thereplenishment of latently infected memory cells throughcell division17 and their loss through viral replication (seebelow). This cell division must be regulated as part ofnormal memory B-cell homeostasis because there are noviral proteins expressed that could cause the cell to divide.When they divide, they express EBNA1 (the EBNA1-onlyprogram, Fig 2),17 which is needed to allow the viral DNAto replicate with the cells.29 Perhaps not surprisingly,because EBNA1 represents the only point of immuneattack of the memory cells, EBNA1 has evolved to bepoorly recognized by the immune system.30

By gaining entrance to normal memory B cells andshutting down viral protein expression, the virus is safefrom immune surveillance. It is also benign because noneof the latent proteins that drive growth are expressed. Thisexplains why EBV is able to persist benignly in the vastmajority of human subjects: EBV infection in vivo doesnot drive limitless proliferation. Rather it drives transientproliferation so that the cells can become resting memorycells. The virus persists in nonpathogenic resting cellsnot proliferating blasts. This also explains why EBV-associated tumors do not arise in every infected individual,even when they are immunosuppressed; something mustgo wrong with the normal biology that takes the latently

FIG 1. The lymphoepithelium of the palatine tonsils from the Waldeyer ring. The tonsil consists of a highly

involuted epithelium, creating a large surface area with deep invaginations. The epithelial surface is at the top

of both micrographs. Antigen and EBV both enter through saliva and cross the epithelial barrier to activate or

infect, respectively, the naive B cells below. The mantle zone (MZ), containing naive B cells (dark blue), is

always facing the surface and is continuous with the epithelium. Naive cells enter the tonsil (black arrow)

through the high endothelial venules (orange cuboidal cells). Numerous follicles containing germinal center

B cells (GC) are arranged parallel to the surface. B cells leave the germinal center (red arrow) and enter the

circulation through the efferent lymphatics. A higher magnification (expanded box) reveals the sponge-like

structure of the epithelial cells in the lymphoepithelium that create spaces extending all the way to the mantle

zone that are filled with infiltrating lymphocytes such that there is frequently only a single epithelial cell

between the outer surface and the lymphocytes. (The micrographs were kindly provided by Dr Marta Perry).



Thorley-Lawson 253

Review

sand

featu

rearticles

infected cells into a resting state before EBV could beinvolved in tumor development.

Release

By accessing the memory compartment, EBV has asite for long-term persistence. However, it must replicateand be shed to spread to new hosts. The parallels between

normal B-cell biology and the mechanism of viral shed-ding are shown in Fig 3. Signals that cause the B cell todifferentiate into an antibody-secreting plasma cell willin turn reactivate the virus.31 Because antibody-secretingplasma cells migrate into the mucosal epithelium,18,32

such a cell will be perfectly placed to release virus ontothe mucosal surface, which, in the case of the tonsils,

FIG 2. A model of how EBV uses normal B-cell biology to establish and maintain persistent infection in

memory B cells. The response of a normal B cell to antigen, leading to the production of antigen-specific

memory cells in the peripheral circulation, is diagrammed to the left, and the parallel series of steps by which

EBV establishes latent infection in peripheral memory B cells is shown to the right. The specific viral

transcription programs are labeled in blue to the right. For details, see the text and Table I.

FIG 3. A model of how EBV uses normal B-cell biology to replicate and be shed into saliva. The pathway by

which antigen-specific B cells become activated and differentiate into antibody-producing plasma cells is

shown to the left, and the parallels that lead to shedding of EBV are shown to the right. The EBV transcription

program is indicated in blue to the right. For details, see text and Table I.


AUGUST 2005

254 Thorley-Lawson

Review

sand

featu

rearticle

s

is saliva. Thus infectious virus is spread through salivacontact.33

Epithelial cells and viral shedding

Although EBV is considered to be a B-lymphotropicvirus, it can also infect epithelial cells because it is foundin several important diseases of epithelial cells, includingnasopharyngeal34 and gastric35 carcinomas and oral hairyleukoplakia.36 What is less clear is whether epithelial cellsplay a role in the normal biology of EBV. Early reportsthat claimed to find EBV in healthy nasopharyngealepithelium have been discredited37; however, recentwork has revisited this possibility. There is now evidencethat normal epithelial cells in the nasopharynx express adistinct EBV receptor,38 that they can be infected in vitro,and that they are infected in vivo.39 However, it remainsundetermined whether this infection occurs fortuitouslybecause this epithelium is an area in which EBV happensto replicate or because it is an important component of theviral biology. The most likely role for epithelial cells is asa site for replication and amplification of the virus ratherthan as a site of persistent latent infection.40,41 Because thereceptor is only expressed on the basolateral surface ofepithelial cells, the virus can only infect from the lym-phoid tissue and not from saliva. Thus if epithelial cellsplay an amplification role, it is during viral shedding andnot primary infection. Perhaps the most compelling indi-rect evidence for epithelial cell infection comes fromsimple numbers. Estimates of the number of lymphocytesreplicating EBV in the tonsils42 indicates that there are notnearly enough to account for the rates of viral sheddingfound in saliva (Hadinoto and Thorley-Lawson, unpub-lished data). This suggests that there must be a location-mechanism for amplifying the virus shed from plasmacells. The obvious candidates are epithelial cells because,from studies on oral leukoplakia, we know that epithelialcells replicate EBV to high copy numbers.

Unresolved questions about persistencein memory cells

There are important unresolved questions relating toEBV persistence in memory B cells.

First, what is the relative contribution of reinfectionversus homeostatic cell division to the maintenance ofstable levels of latently infected cells? We know that thehost mounts a massive cytotoxic T-cell response againstcells replicating EBV and newly infected cells43 and aneutralizing antibody response against the virus.4 It istherefore unclear whether newly infected cells are pro-duced rapidly enough and survive long enough to con-tribute to the pool of latently infected memory cells oncethe immune response has begun. It is conceivable that newinfection is only critical in establishing the pool of latentlyinfected memory cells before the onset of the immuneresponse and thereafter plays no role. A clue that thismight be true comes from the observation that the epitopesrecognized by cytotoxic T cells on newly infected B cellsare conserved.44 Usually, a virus is continuously varyingits sequence to avoid the immune response (eg, HIV45),

but in the case of EBV, it seems to ensure that newlyinfected cells are rapidly destroyed. This suggests that thenew infection route might only be viable before theimmune response arises (ie, in acute infection). There-after, the virus depends on homeostasis of the pool oflatently infected memory cells for persistence and ensuresthat any new infected cells are rapidly killed because theymight pose a lymphoproliferative threat to the host.

Second, if EBV persists in normal antigen-selectedB cells (unpublished results), why does it have LMP1 andLMP2, which can replace all the signaling necessary toproduce a memory B cell? The answer to this is not yetclear. One possibility is that the role of LMP1 and LMP2might be to give a selective advantage to the virus-infectedcells in the highly competitive environment of the germi-nal center. This would give the latently infected cell abetter chance of making it into the memory pool.

Third, why does EBV not infect memory cells directly?A priori there seems no reason why EBV could not use thesame mechanism to drive an infected memory cell backinto memory; however, the evidence does not favor thisalternative. First, there is no evidence that direct infectionof memory cells occurs consistently in vivo.21,22 Second,when it does occur, it seems to lead to clonal prolifera-tion46,47 and not differentiation, and third, the pool oflatently infected memory cells is skewed (Sousa andThorley-Lawson, unpublished data), which would notbe expected if EBV infected memory cells at random.One possible explanation comes from the known biologyof B cells. Activation of naive B cells through the ger-minal center leads predominantly to the production ofmemory cells over plasma cells,48 whereas activationof memory cells leads predominantly to the production ofplasma cells.49 Therefore if the goal of EBV is to accessthe memory compartment, it will do so more efficiently byinfecting and activating naive B cells rather than memorycells.

EBV AND DISEASE

General considerations

EBV has been associated with a number of humandiseases. These generally fall into 2 categories: autoim-munity and cancer.1,3 The idea that EBV might beinvolved in autoimmunity stems from the knowledgethat the virus can infect any B cell and cause it toproliferate indefinitely in culture. This raises the possibil-ity that EBV could immortalize forbidden clones of B cellsin vivo, perhaps allowing them to produce autoimmuneantibodies in an uncontrolled fashion. This philosophicunderpinning for a role of EBV in autoimmunity can nowbe seen to be incorrect. We know that EBV does notpersist in vivo by immortalizing B cells but by establishinga true latency in normal resting memory B cells. There arealso technical difficulties to proving a causal role for EBVin these diseases. First, EBV persists in circulating mem-ory cells and therefore will be found in all tissues, irres-pective of disease causality. Second, it is now apparent



Thorley-Lawson 255

Review

sand

featu

rearticles

that EBV is extremely sensitive to the state of the immunesystem. This is because it relies on normal B-cell biologyto establish and maintain persistence and T-cell responsesto modulate the level of infection. Changes that affectthe functionality of the immune system affect EBV bychanging overall viral loads and states of infection.Because autoimmune diseases classically disrupt theimmune system, it will be extremely difficult to dissectout causality of EBV from the background noise ofchanges occurring in the virus because of the disease.For all of these reasons, it has been difficult to establish aclear connection between EBV and any autoimmunediseases. Currently, such associations remain speculative,controversial, or both.

The reason to believe EBV might cause cancer isapparent. EBV encodes genes that make B cells grow.Such genes will, of their nature, have potential as onco-

genic risk factors. However, as described above, EBV hasevolved to minimize the risk that an infected cell willproliferate out of control. Therefore something must gowrong with the normal viral biology for EBV to play acausative role in tumor development. The plausibility ofEBV as an oncogenic virus has led to claims of itsassociation with many human tumors. Some, such asbreast and hepatocellular carcinoma, have never beensubstantiated, but there are now several for whichstrong evidence exists, including immunoblastic lym-phoma in immunosuppressed patients, Burkitt’s lym-phoma, Hodgkin’s disease (HD), and nasopharyngealcarcinoma (NPC). The origins of all of these tumors canbe understood as arising from specific stages in the EBVlife cycle (Fig 4) and appear to be associated withdisturbances of the immune system. This begs the follow-ing question: How convincing is the evidence that EBV

FIG 4. The putative check points in the EBV life cycle that give rise to tumors. The events that occur normally in

healthy carriers are denoted in black. For details, see Figs 2 and 3. EBV normally infects naive B cells in the

Waldeyer’s ring, and these cells can differentiate intomemory cells and out of the cell cycle (thick arrows), and

therefore they are not pathogenic. PTLD: If a cell other than the naive B cell in the Waldeyer ring becomes

infected, it will express the growth program and continue to proliferate because it cannot differentiate out of

the cell cycle (thin dashed arrows). This is a very rare event, highlighting how carefully controlled EBV

infection is. Normally, these bystander B-cell blasts would be destroyed by CTLs, but if the CTL response is

suppressed, then they can grow into PTLD. Note: a bystander-type cell could also arise if a latently infected

germinal center or memory cell fortuitously switched on the growth program. Hodgkin’s disease arises from

an EBV-infected cell that is blocked at the germinal-center cell stage. This results in constitutive expression of

the default program. Burkitt’s lymphoma evolves from a germinal-center cell that is entering the memory

compartment but is stuck proliferating. Consequently, the cell expresses EBNA1 only. Nasopharyngeal

carcinoma is hypothesized to arise from a latently infected epithelial cell blocked from terminal differentiation

and viral replication. It is unclear why these cells would express the default program.


AUGUST 2005

256 Thorley-Lawson

Review

sand

featu

rearticle

s

plays a causative role in these tumors and is not simply apassenger in a tumor cell that arose from an infected celltype?

Lymphoma in the immunosuppressed

Individuals who are immunosuppressed are at risk fordevelopment of B-cell lymphoproliferative diseases, suchas the immunoblastic lymphomas in patients with AIDSand the posttransplantation lymphoproliferative diseases(PTLDs) in patients undergoing organ transplantation.10

These are a heterogeneous collection of disorders thatusually carry the virus and express the growth program(Table I).50 A wide range of factors (eg, organ type,immunosuppressive regime, location, and donor origin)influence the frequency with which these tumors arise.The explanation usually given for the origin of thesetumors is that immunosuppression of the cytotoxicT-lymphocyte (CTL) response to EBV allows uninhibitedgrowth of EBV-infected cells; however, it is not thatsimple.

From the discussion above on the mechanism of EBVpersistence, it is apparent that, under normal conditions,infected naive B cells in the tonsils do not give rise tolymphoma because they differentiate out of the cell cycleto become resting memory cells. For a cell to express thegrowth program, survive, and evolve into a neoplasm, 2events must occur: the EBV-infected cell must be unableto respond to signals that drive it to differentiate into aresting memory cell, and the CTL response must becrippled so that these lymphoblasts can continue toproliferate. This could occur if any B cell that is not anaive B cell in the tonsil is exposed to the virus bychance—bystander infection (Fig 4). It could also occur ifa latently infected germinal center or memory cell fortu-itously received signals that caused it to inappropriatelyturn on the growth program. These cells can not exit thegrowth program, and therefore they continue to prolifer-ate. Normally, they would be rapidly eliminated by CTLsbecause of the conserved CTL epitopes they express (seeabove); however, in the absence of effective T-cellimmunity (immunosuppression), they will continue toproliferate. Direct evidence that this is indeed the casecomes from studies of tonsils from acutely infectedindividuals. In these tonsils clonal expansions of directlyinfected germinal center51 and memory cells46 driven bythe growth program can be found. Because these arebystander-infected cells, they are unable to differentiateinto resting memory cells. Consequently, they proliferateuntil the immune response arises to eliminate them,explaining why such clones are never seen in healthycarriers of the virus but will appear if the immune responseis subsequently suppressed.

The origin of these tumors also explains their hetero-geneity. They are derived from a mixture of B-cell types52

consistent with arising from a variety of bystander B cellsthat get infected by chance and not a specific subset ofinfected cells. This also explains why the tumors arerelatively rare. The vast majority of infected cells differ-entiate into a resting memory state because they are naive;

they will not be a cancer risk. Only the rare, atypicalbystander infection is a risk for tumor development.

Hodgkin’s disease

Acute EBV infection in the adolescent-adult can giverise to IM, long known to be a risk factor for HD.However, the strongest evidence directly linking EBVwith HD came with the finding that approximately 40%of the tumors contain clonal EBV,53 which can approach80% in developing countries and up to 100% in AIDS-related HD.54 In addition, the tumor cells express thedefault transcription program (Table I),55-58 which in-cludes 2 proteins (LMP1 and LMP2) that deliver survivaland growth signals,59-61 at least one of which (LMP1) isknown to act as an oncogene.62 A characteristic of IM,compared with the subclinical infection seen in children,is profound disruption of the immune system.63 This in-cludes massive levels of virus-infected memory B cells(50%), a striking T-cell lymphocytosis caused, at least inpart, by a very aggressive cytotoxic T-cell response, andtissue damage in the lymph nodes. The disease is almostcertainly a product of an overreactive inflammatoryresponse, and B-cell function is so badly disrupted thatone of the characteristics of IM is the production of a broadrange of nonspecific, low-affinity, so-called heteropohileantibodies. This suggests that HD is the consequence ofderegulated EBV infection caused by the severe immu-nologic disturbance of IM. Nevertheless, the possibilitythat EBV is a passenger cannot be excluded. If theimmunologic disruption of IM alone is the risk factor forHD, it is possible that the premalignant B cell will haveEBV in it simply by chance.

There is good evidence that EBV-positive HD arisesfrom an infected germinal-center cell. As discussed above,one of the characteristics of germinal-center cells is thatthey actively mutate their immunoglobulin genes in aprocess termed hypermutation, which leaves a character-istic pattern of mutations. The immunoglobulin genes ofHRS cells have this pattern of mutation.64 In addition, thedefault transcription program is used by EBV in latentlyinfected germinal center B cells.22 Thus the immunoglob-ulin mutations and the viral gene expression data inde-pendently support the idea that EBV-positive HD arisesfrom an EBV-infected germinal center B cell (Fig 4).

Burkitt’s lymphoma

EBV was discovered in cultured tumor cells frompatients with the endemic form of Burkitt’s lymphoma(BL).65 It is sobering to realize that 40 years later, we stilldo not know how or even for sure whether EBV causesBL. This is despite the large volume of informationwe have acquired about EBV’s molecular and cellularbiology, immunology, virology, epidemiology, clinicalmanifestations, and disease associations.1-3 The mostcompelling evidence of EBV’s involvement in BL is thehigh frequency (98%) of tumors carrying the virus66 inendemic areas and the presence of clonal EBV in all of thetumor cells.67 However, none of the growth-promotinglatent genes are expressed. The only genes expressed



Thorley-Lawson 257

Review

sand

featu

rearticles

encode for EBNA168 and the untranslated RNAs calledEBERS and BARTS. It has been suggested that EBNA169

and the EBERS70 might have oncogenic potential, butthe findings remain unsubstantiated and are controversial.Consequently, there is currently no broadly acceptedunderstanding of the role of EBV in BL.71-74 What isapparent, however, is that malaria, which is chronicallyimmunosuppressive,75 is classically known to be a riskfactor for endemic (ie, EBV-positive) BL development.76

Once more, this supports the notion, discussed throughoutthis review, that EBV infection in the context of acompromised immune system is the risk factor for lym-phoma development.

Using the same arguments as for HD, we can surmisethat BL is a tumor cell of a proliferating, latently infectedmemory B cell (Fig 4). BL has the same pattern ofimmunoglobulin gene hypermutations as memory Bcells,77 and there is only one way known for producingan EBNA1-only phenotype in nontumor cells. This iswhen a latently infected memory cell expressing thelatency program divides as part of normal B-cell homeo-stasis (Fig 2).17 One property of BL inconsistent with thisidea is that the tumor cells have the surface phenotype ofgerminal-center cells.78 However, the cellular phenotypeof tumor cells can be misleading. This is exemplified byHD, which is generally thought to be derived from agerminal-center cell, although it bears no phenotypic ormorphologic resemblance to such cells. Thus it is difficultto know how directly the final cellular phenotype of BLrelates to the original infected precursor. Possibly, BL isderived from a germinal-center cell on its way to becom-ing a resting memory cell expressing the latency programbut through tumor-driven growth continues to proliferateand therefore expresses the EBNA1-only phenotype.

Nasopharyngeal carcinoma

Given the B lymphotropism of EBV, it is surprising thatone of the best candidates for a tumor caused by EBV isnot a lymphoma but a carcinoma, NPC, responsible for20% of all cancers in China and Taiwan79 and thereforean important world health problem. Virtually 100% ofundifferentiated NPCs worldwide contain clonalEBV.34,80 The tumors express the viral default transcrip-tion program.81-83 Although only a subset, approximately40%, express LMP1, it has been reported that the prema-lignant lesions of NPC all express LMP1.84 As with HD,the presence of LMP1 and LMP2 is additional evidencethat the virus is playing a part in the cause of the tumor.Because LMP1 and LMP2 are potently and specificallyevolved B cell–signaling molecules, their presence in theepithelial cells of NPC suggests the virus might be therefortuitously. An example of this is LMP2, which functionsto cause B cells to migrate into mucosal follicles.23 Thismigratory ability, expressed in epithelial cells, mightresult in the invasive and metastatic activity of NPC.

The potential role of EBV inNPC is clouded by our lackof certain knowledge about the role of epithelial cells inEBV biology. In Fig 4 the speculative assumption is madethat EBV latently infects epithelial cells that then proceed

to replicate the virus and shed it into saliva. NPC wouldderive from such a latently infected, undifferentiatedepithelial cell, which was blocked from switching to viralreplication and therefore continues to be latently infected.Why the default program, usually found in germinal centerB cells, is expressed in NPC is completely unclear.

CONCLUSION: ANSWERS TO THEQUESTIONS

In the introduction to this article, several questions wereraised about EBV. The answers to these questions can nowbe explained in light of the discussion above.

First, why does EBVmake cells proliferate when it putsthe host at risk for neoplastic disease?

Because it has to. The newly latently infected naiveB cell has to become an activated blast before it candifferentiate into a resting memory cell.

Second, where do the EBV-positive tumors come from,why do the different tumors express different viral latentgene transcription programs, and why is disruption ofthe immune system a risk factor? The virus uses thesedifferent transcription programs to manipulate the biologyof the infected B cell so that it can gain entry into and thenpersist in memory B cells. Any disruption of the immunesystem that interferes with the ability of the EBV-infectedcells to become a resting memory cell will increase the riskof tumor development. Each tumor derives from a differ-ent step in this process and represents a cell that is blockedfrom progressing into a resting state and therefore con-tinues to express the viral transcription program of itsprogenitor.

Third, why are there so few EBV-infected tumors in thehuman population, evenwith immunosuppression, despitethe large numbers of EBV-infected cells in each individualand the ability of EBV to make lymphocytes grow? This isbecause the viral biology is tightly regulated to ensure thatan EBV-infected naive B cell that becomes activated andstarts to proliferate will rapidly exit the cell cycle andbecome a resting memory cell expressing none of thedangerous growth-promoting genes. In addition, the virushas conserved the targets for CTLs to ensure that if a newlyinfected cell does not exit the cell cycle, it will be rapidlykilled.

CONCLUSIONS AND FUTURE DIRECTIONS

We now know the basic outlines of the EBV life cycleand have some understanding of where and why thetumors arise. For the basic scientist, the challenge remainsto understand, at the molecular level, how EBV negotiatesthe changes between the different latency states in thedifferent B-cell types. Because this is so dependent onB cells, it is likely that the mechanisms will only becomeclear when we learn how the processes are normallyregulated in B cells. There is still also much to be learnedabout the role EBV plays at the molecular level in


AUGUST 2005

258 Thorley-Lawson

Review

sand

featu

rearticle

s

tumorigenesis, particularly for BL. But perhaps the big-gest gap in our knowledge is understanding what isdifferent in the disruption of the immune response thatleads to immunoblastic lymphoma in some cases, HD inothers, and BL in yet others. Could timely immunologicintervention reduce the risk of subsequent development ofthese diseases?

The identification of EBV within tumors provides apotentially unique opportunity to develop tumor-specifictherapy targeted at the virus that will not hurt normal cells.A good example of this is recent work showing thatin vitro expanded, EBV-specific CTLs can be effectivetherapy against PTLD,85 although they hold less promisefor treatment of HD and NPC. An important and poten-tially fruitful area of clinical investigation will be thedevelopment of drugs specifically targeted against EBV.The best candidate latent protein might well be EBNA1,which allows replication of the viral DNA and therefore isessential for retention of the viral DNA in a proliferating(eg, tumor) cell. The crystal structure of EBNA1 bound toDNA is known, opening the path to the development ofdrugs that block this interaction. If the tumor requires EBVto grow, loss of the viral DNA should prevent tumorgrowth. Whether EBV truly plays a causative role in thesetumors is therefore not an esoteric question. If the virus isnot a key player, then therapies directed at the virus will beineffective against the tumors. A hint of this comes fromPTLD, in which restoration of the immune response leadsto tumor regression. Eventually, however, the tumorsbecome resistant. This raises the possibility that ultimatelytumor growth might not be dependent on the virus.

An interesting approach that does not require the virusto be essential for tumor growthwould be the developmentof drugs that efficiently cause EBV in the tumors to beginreplicating. Because replication of the virus kills the cell,this would be an indirect way to destroy EBV-positivetumors, irrespective of their dependence on the virus forgrowth. This would not need to lead to wholesaleproduction of virus, however, because drugs that blockthis are already available (eg, valacyclovir).

REFERENCES

1. Rickinson AB, Kieff E. Epstein-Barr virus. In: Knipe DM, Howley PM,

editors. Virology. 4th ed. New York: Lippincott Williams and Wilkins;

2001. p. 2575-628.

2. Kieff E, Rickinson AB. Epstein-Barr virus and its replication. In: Knipe

DM, Howley PM, editors. Virology. 4th ed. New York: Lippincott

Williams and Wilkins; 2001. p. 2511-74.

3. Thorley-Lawson DA. Epstein-Barr virus. In: Cantor H, editor. Sampter’s

immunologic diseases. New York: Williams and Wilkins; 2001. p.

970-85.

4. Henle W, Henle G. Seroepidemiology of the virus. In: Epstein MA,

Achong BG, editors. The Epstein-Barr virus. Berlin: Springer-Verlag;

1979. p. 61-78.

5. Pope JH, Horne MK, Scott W. Transformation of foetal human

leukocytes in vitro by filtrates of a human leukaemic cell line containing

herpes-like virus. Int J Cancer 1968;3:857-66.

6. Henle W, Diehl V, Kohn G, Zur Hausen H, Henle G. Herpes-type virus

and chromosome marker in normal leukocytes after growth with

irradiated Burkitt cells. Science 1967;157:1064-5.

7. Seemayer TA, Gross TG, Egeler RM, Pirruccello SJ, Davis JR, Kelly

CM, et al. X-linked lymphoproliferative disease: twenty-five years after

the discovery. Pediatr Res 1995;38:471-8.

8. Coffey AJ, Brooksbank RA, Brandau O, Oohashi T, Howell GR, Bye

JM, et al. Host response to EBV infection in X-linked lymphoprolifer-

ative disease results from mutations in an SH2-domain encoding gene.

Nat Genet 1998;20:129-35.

9. Sayos J, Wu C, Morra M, Wang N, Zhang X, Allen D, et al. The

X-linked lymphoproliferative-disease gene product SAP regulates signals

induced through the co-receptor SLAM. Nature 1998;395:462-9.

10. Hopwood P, Crawford DH. The role of EBV in post-transplant malig-

nancies: a review. J Clin Pathol 2000;53:248-54.

11. Khan G, Miyashita EM, Yang B, Babcock GJ, Thorley-Lawson DA.

Is EBV persistence in vivo a model for B cell homeostasis? Immunity

1996;5:173-9.

12. Babcock GJ, Decker LL, Freeman RB, Thorley-Lawson DA. Epstein-

Barr virus-infected resting memory B cells, not proliferating lympho-

blasts, accumulate in the peripheral blood of immunosuppressed patients.

J Exp Med 1999;190:567-76.

13. Thorley-Lawson DA, Gross A. Persistence of the Epstein-Barr virus and

the origins of associated lymphomas. N Engl J Med 2004;350:1328-37.

14. Thorley-Lawson DA. Epstein-Barr virus: exploiting the immune system.

Nat Rev Immunol 2001;1:75-82.

15. Miyashita EM, Yang B, Babcock GJ, Thorley-Lawson DA. Identification

of the site of Epstein-Barr virus persistence in vivo as a resting B cell.

J Virol 1997;71:4882-91.

16. Babcock GJ, Decker LL, Volk M, Thorley-Lawson DA. EBV persistence

in memory B cells in vivo. Immunity 1998;9:395-404.

17. Hochberg D, Middeldorp JM, Catalina M, Sullivan JL, Luzuriaga K,

Thorley-Lawson DA. Demonstration of the Burkitt’s lymphoma Epstein-

Barr virus phenotype in dividing latently infected memory cells in vivo.

Proc Natl Acad Sci U S A 2003;101:239-44.

18. Brandtzaeg P, Farstad IN, Haraldsen G. Regional specialization in the

mucosal immune system: primed cells do not always home along the

same track. Immunol Today 1999;20:267-77.

19. Perry M, Whyte A. Immunology of the tonsils. Immunol Today 1998;19:

414-21.

20. Liu YJ, Arpin C. Germinal center development. Immunol Rev 1997;156:

111-26.

21. Joseph AM, Babcock GJ, Thorley-Lawson DA. Cells expressing the

Epstein-Barr virus growth program are present in and restricted to the

naive B-cell subset of healthy tonsils. J Virol 2000;74:9964-71.

22. Babcock GJ, Hochberg D, Thorley-Lawson AD. The expression pattern

of Epstein-Barr virus latent genes in vivo is dependent upon the

differentiation stage of the infected B cell. Immunity 2000;13:497-506.

23. Casola S, Otipoby KL, Alimzhanov M, Humme S, Uyttersprot N, Kutok

JL, et al. B cell receptor signal strength determines B cell fate.

Nat Immunol 2004;5:317-27.

24. He B, Raab-Traub N, Casali P, Cerutti A. EBV-encoded latent membrane

protein 1 cooperates with BAFF/BLyS and APRIL to induce T cell-

independent Ig heavy chain class switching. J Immunol 2003;171:

5215-24.

25. Panagopoulos D, Victoratos P, Alexiou M, Kollias G, Mosialos G.

Comparative analysis of signal transduction by CD40 and the Epstein-

Barr virus oncoprotein LMP1 in vivo. J Virol 2004;78:13253-61.

26. Calame KL, Lin KI, Tunyaplin C. Regulatory mechanisms that determine

the development and function of plasma cells. Annu Rev Immunol 2003;

21:205-30.

27. Uchida J, Yasui T, Takaoka-Shichijo Y, Muraoka M, Kulwichit W,

Raab-Traub N, et al. Mimicry of CD40 signals by Epstein-Barr virus

LMP1 in B lymphocyte responses. Science 1999;286:300-3.

28. Hochberg D, Souza T, Catalina M, Sullivan JL, Luzuriaga K, Thorley-

Lawson DA, et al. Acute infection with Epstein-Bar virus targets and

overwhelms the peripheral memory B cell compartment with resting,

latently infected cells. J Virol 2004;78:5194-204.

29. Yates JL, Warren N, Sugden B. Stable replication of plasmids derived

from Epstein-Barr virus in various mammalian cells. Nature 1985;313:

812-5.

30. Levitskaya J, Coram M, Levitsky V, Imreh S, Steigerwald MP, Klein G,

et al. Inhibition of antigen processing by the internal repeat region of the

Epstein-Barr virus nuclear antigen-1. Nature 1995;375:685-8.



Thorley-Lawson 259

Review

sand

featu

rearticles

31. Laichalk LL, Thorley-Lawson DA. Terminal differentiation into plasma

cells initiates the replicative cycle of Epstein-Barr virus in vivo. J Virol

2005;79:1296-307.

32. Brandtzaeg P, Baekkevold ES, Farstad IN, Jahnsen FL, Johansen FE,

Nilsen EM, et al. Regional specialization in the mucosal immune system:

what happens in the microcompartments? Immunol Today 1999;20:

141-51.

33. Golden HD, Chang RS, Prescott W, Simpson E, Cooper TY. Leukocyte-

transforming agent: prolonged excretion by patients with mononucleosis

and excretion by normal individuals. J Infect Dis 1973;127:471-3.

34. Andersson-Anvret M, Forsby N, Klein G, Henle W. Relationship

between the Epstein-Barr virus and undifferentiated nasopharyngeal

carcinoma: correlated nucleic acid hybridization and histopathological

examination. Int J Cancer 1977;20:486-94.

35. Shibata D, Weiss LM. Epstein-Barr virus-associated gastric adenocarci-

noma. Am J Pathol 1992;140:769-74.

36. Greenspan JS, Greenspan D, Lennette ET, Abrams DI, Conant MA,

Petersen V, et al. Replication of Epstein-Barr virus within the epithelial

cells of oral ‘‘hairy’’ leukoplakia, an AIDS-associated lesion. N Engl J

Med 1985;313:1564-71.

37. Karajannis MA, Hummel M, Anagnostopoulos I, Stein H. Strict

lymphotropism of Epstein-Barr virus during acute infectious mono-

nucleosis in nonimmunocompromised individuals. Blood 1997;89:

2856-62.

38. Tugizov SM, Berline JW, Palefsky JM. Epstein-Barr virus infection of

polarized tongue and nasopharyngeal epithelial cells. Nat Med 2003;9:

307-14.

39. Pegtel DM, Middeldorp J, Thorley-Lawson DA. Epstein-Barr virus

infection in ex vivo tonsil epithelial cell cultures of asymptomatic

carriers. J Virol 2004;78:12613-24.

40. Gratama JW, Oosterveer MA, Zwaan FE, Lepoutre J, Klein G, Ernberg I.

Eradication of Epstein-Barr virus by allogeneic bone marrow transplan-

tation: implications for sites of viral latency. Proc Natl Acad Sci U S A

1988;85:8693-6.

41. Faulkner GC, Burrows SR, Khanna R, Moss DJ, Bird AG, Crawford DH.

X-Linked agammaglobulinemia patients are not infected with Epstein-

Barr virus: implications for the biology of the virus. J Virol 1999;73:

1555-64.

42. Laichalk LL, Thorley-Lawson DA. Terminal differentiation into plasma

cells initiates the replicative cycle of Epstein-Barr virus in vivo. J Virol

2005;79:1296-307.

43. Khanna R, Moss DJ, Burrows SR. Vaccine strategies against Epstein-

Barr virus-associated diseases: lessons from studies on cytotoxic T-cell-

mediated immune regulation. Immunol Rev 1999;170:49-64.

44. Khanna R, Slade RW, Poulsen L, Moss DJ, Burrows SR, Nicholls J,

et al. Evolutionary dynamics of genetic variation in Epstein-Barr virus

isolates of diverse geographical origins: evidence for immune pressure-

independent genetic drift. J Virol 1997;71:8340-6.

45. Phillips RE, Rowland-Jones S, Nixon DF, Gotch FM, Edwards JP,

Ogunlesi AO, et al. Human immunodeficiency virus genetic variation

that can escape cytotoxic T cell recognition. Nature 1991;354:453-9.

46. Kurth J, Spieker T, Wustrow J, Strickler GJ, Hansmann LM, Rajewsky

K, et al. EBV-infected B cells in infectious mononucleosis: viral

strategies for spreading in the B cell compartment and establishing

latency. Immunity 2000;13:485-95.

47. Kurth J, Perniok A, Schmitz R, Iking-Konert C, Chiorazzi N, Thompson

KM, et al. Lack of deleterious somatic mutations in the CD95 gene of

plasmablasts from systemic lupus erythematosus patients and autoanti-

body-producing cell lines. Eur J Immunol 2002;32:3785-92.

48. Arpin C, Dechanet J, Van KC, Merville P, Grouard G, Briere F, et al.

Generation of memory B cells and plasma cells in vitro. Science 1995;

268:720-2.

49. Arpin C, Banchereau J, Liu YJ. Memory B cells are biased towards

terminal differentiation: a strategy that may prevent repertoire freezing.

J Exp Med 1997;186:931-40.

50. Thomas JA, Hotchin NA, Allday MJ, Amlot P, Rose M, Yacoub M, et al.

Immunohistology of Epstein-Barr virus-associated antigens in B cell

disorders from immunocompromised individuals. Transplantation 1990;

49:944-53.

51. Kurth J, Hansmann ML, Rajewsky K, Kuppers R. Epstein-Barr virus-

infected B cells expanding in germinal centers of infectious mononucle-

osis patients do not participate in the germinal center reaction. Proc Natl

Acad Sci U S A 2003;100:4730-5.

52. Timms JM, Bell A, Flavell JR, Murray PG, Rickinson AB, Traverse-

Glehen A, et al. Target cells of Epstein-Barr-virus (EBV)-positive post-

transplant lymphoproliferative disease: similarities to EBV-positive

Hodgkin’s lymphoma. Lancet 2003;361:217-23.

53. Glaser SL, Lin RJ, Stewart SL, Ambinder RF, Jarrett RF, Brousset P,

et al. Epstein-Barr virus-associated Hodgkin’s disease: epidemiologic

characteristics in international data. Int J Cancer 1997;70:375-82.

54. Glaser SL, Clarke CA, Gulley ML, Craig FE, DiGiuseppe JA, Dorfman

RF, et al. Population-based patterns of human immunodeficiency virus-

related Hodgkin lymphoma in the Greater San Francisco Bay Area,

1988-1998. Cancer 2003;98:300-9.

55. Oudejans JJ, Dukers DF, Jiwa NM, van den Brule AJ, Grasser FA,

de Bruin PC, et al. Expression of Epstein-Barr virus encoded nuclear

antigen 1 in benign and malignant tissues harbouring EBV. J Clin Pathol

1996;49:897-902.

56. Deacon EM, Pallesen G, Niedobitek G, Crocker J, Brooks L, Rickinson

AB, et al. Epstein-Barr virus and Hodgkin’s disease: transcriptional

analysis of virus latency in the malignant cells. J Exp Med 1993;177:

339-49.

57. Herbst H, Dallenbach F, Hummel M, Niedobitek G, Pileri S, Muller

LN, et al. Epstein-Barr virus latent membrane protein expression in

Hodgkin and Reed-Sternberg cells. Proc Natl Acad Sci U S A 1991;88:

4766-70.

58. Niedobitek G, Kremmer E, Herbst H, Whitehead L, Dawson CW,

Niedobitek E, et al. Immunohistochemical detection of the Epstein-Barr

virus-encoded latent membrane protein 2A in Hodgkin’s disease and

infectious mononucleosis. Blood 1997;90:1664-72.

59. Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C,

Kieff E. The Epstein-Barr virus transforming protein LMP1 engages

signaling proteins for the tumor necrosis factor receptor family. Cell

1995;80:389-99.

60. Gires O, Zimber-Strobl U, Gonnella R, Ueffing M, Marschall G, Zeidler

R, et al. Latent membrane protein 1 of Epstein-Barr virus mimics a

constitutively active receptor molecule. EMBO J 1997;16:6131-40.

61. Caldwell RG, Wilson JB, Anderson SJ, Longnecker R. Epstein-Barr

virus LMP2A drives B cell development and survival in the absence of

normal B cell receptor signals. Immunity 1998;9:405-11.

62. Wang D, Liebowitz D, Kieff E. An EBV membrane protein expressed in

immortalized lymphocytes transforms established rodent cells. Cell 1985;

43:831-40.

63. Thorley-Lawson DA. Epstein-Barr virus. In: Austen KF, Frank MM,

Atkinson JP, Cantor H, editors. Sampter’s immunologic diseases. 6th ed.

New York: Williams and Wilkins; 2001. p. 970-85.

64. Kuppers R, Rajewsky K. The origin of Hodgkin and Reed/Sternberg

cells in Hodgkin’s disease. Annu Rev Immunol 1998;16:471-93.

65. Epstein MA, Achong BG, Barr YM. Virus particles in cultured

lymphoblasts from Burkitt’s lymphoma. Lancet 1964;1:702-3.

66. de-The G. Epstein-Barr virus and Burkitt’s lymphoma worldwide: the

causal relationship revisited. In: Lenoir GM, O’Conor GT, Olweny

CLM, editors. Burkitt’s lymphoma: a human cancer model. New York:

Oxford University Press; 1985. p. 165-76.

67. Gulley ML, Raphael M, Lutz CT, Ross DW, Raab-Traub N. Epstein-Barr

virus integration in human lymphomas and lymphoid cell lines. Cancer

1992;70:185-91.

68. Gregory CD, Rowe M, Rickinson AB. Different Epstein-Barr virus-B

cell interactions in phenotypically distinct clones of a Burkitt’s lym-

phoma cell line. J Gen Virol 1990;71:1481-95.

69. Wilson JB, Bell JL, Levine AJ. Expression of Epstein-Barr virus nuclear

antigen-1 induces B cell neoplasia in transgenic mice. EMBO J 1996;15:

3117-26.

70. Takada K, Nanbo A. The role of EBERs in oncogenesis. Semin Cancer

Biol 2001;11:461-7.

71. Lenoir GM, Bornkamm GW. Burkitt lymphoma a human cancer model

for the study of the multistep development of cancer: a new scenario. In:

Klein G, editor. Advances in viral oncology. New York: Raven Press;

1987. p. 173-206.

72. Klein G. In defense of the ‘‘old’’ Burkitt lymphoma scenario. In: Klein

G, editor. Advances in viral oncology. New York: Raven Press; 1987. p.

207-11.


AUGUST 2005

260 Thorley-Lawson

Review

sand

featu

rearticle

s

73. Kelly G, Bell A, Rickinson A. Epstein-Barr virus-associated Burkitt

lymphomagenesis selects for downregulation of the nuclear antigen

EBNA2. Nat Med 2002;8:1098-104.

74. Niller HH, Salamon D, Banati F, Schwarzmann F, Wolf H, Minarovits J.

The LCR of EBV makes Burkitt’s lymphoma endemic. Trends Microbiol

2004;12:495-9.

75. de The G. The etiology of Burkitt’s lymphoma and the history of the

shaken dogmas. Blood Cells 1993;19:667-75.

76. Burkitt D. A lymphoma syndrome in African children. Ann R Coll Surg

Engl 1962;30:211-9.

77. Klein U, Klein G, Ehlin-Henriksson B, Rajewsky K, Kuppers R.

Burkitt’s lymphoma is a malignancy of mature B cells expressing

somatically mutated V region genes. Mol Med 1995;1:495-505.

78. Gregory CD, Tursz T, Edwards CF, Tetaud C, Talbot M, Caillou B, et al.

Identification of a subset of normal B cells with a Burkitt’s lymphoma

(BL)-like phenotype. J Immunol 1987;139:313-8.

79. Yu MC, Yuan JM. Epidemiology of nasopharyngeal carcinoma. Semin

Cancer Biol 2002;12:421-9.

80. Raab-Traub N, Flynn K. The structure of the termini of the Epstein-Barr

virus as a marker of clonal cellular proliferation. Cell 1986;47:883-9.

81. Brooks L, Yao QY, Rickinson AB, Young LS. Epstein-Barr virus latent

gene transcription in nasopharyngeal carcinoma cells: coexpression of

EBNA1, LMP1, and LMP2 transcripts. J Virol 1992;66:2689-97.

82. Fahraeus R, Fu HL, Ernberg I, Finke J, Rowe M, Klein G, et al.

Expression of Epstein-Barr virus-encoded proteins in nasopharyngeal

carcinoma. Int J Cancer 1988;42:329-38.

83. Young LS, Dawson CW, Clark D, Rupani H, Busson P, Tursz T, et al.

Epstein-Barr virus gene expression in nasopharyngeal carcinoma. J Gen

Virol 1988;69:1051-65.

84. Pathmanathan R, Prasad U, Sadler R, Flynn K, Raab-Traub N. Clonal

proliferations of cells infected with Epstein-Barr virus in preinvasive

lesions related to nasopharyngeal carcinoma [see comments]. N Engl J

Med 1995;333:693-8.

85. Gottschalk S, Heslop HE, Rooney CM. Adoptive immunother-

apy for EBV-associated malignancies. Leuk Lymphoma 2005;46:

1-10.



Thorley-Lawson 261

Review

sand

featu

rearticles

Continuing Medical Education examination

EBV the prototypical human tumor virus—justhow bad is it?

Instructions for category 1 Continuing Medical Education credit

The American Academy of Allergy, Asthma and Immunology is accredited as a provider of Continuing MedicalEducation (CME) by the Accreditation Council for Continuing Medical Education.

Test ID no.: mai0064Contact hours: 1.0Expiration date: July 31, 2006

Category 1 credit can be earned by reading the text material and taking this CME examination online. For completeinstructions, visit the Journal’s Web site at www.mosby.com/jaci.

Learning objectives: ‘‘EBV the prototypical human tumor virus—just how bad is it?’’

1. To understand that EBV uses mature B cell biology to establish latency, persist, and replicate.

2. To understand that even though EBV is so widespread and apparently benign, it is potentially life-threatening.

3. To understand that EBV evolved the capacity to make cells grow because it is an essential part of the mechanism for

establishing latency in resting cells that are not pathogenic.

4. To understand that EBV-associated tumors arise from different stages in the life cycle of latently infected B cells and that

disruption of the immune response is an important component in the development of all of the EBV-associated lymphomas.

CME itemsQuestion 1. EBV persists within what fraction of thehealthy adult population?

A. <1%B. ;10%C. ;50%D. >90%

Question 2. To establish persistent infection, EBVprimarily infects —

A. naive B cells.B. memory cells.C. activated B cells.D. germinal center cells.

Question 3. EBV-associated tumors are relatively rarebecause —

A. EBV is not oncogenic.B. EBV replicates and kills the cells before they

can grow into tumors.C. EBVhas evolved tominimize its oncogenic potential.D. EBV cannot always be detected in tumors.

Question 4. How many mutations in host genes arerequired to make EBV a fatal acute infection?

A. 0B. 1C. 5-6D. >10


Review

sand

featu

rearticle

s


Editorial

Infection versus immunity: What’s the balance?

William T. Shearer, MD, PhD Houston, Tex

This issue of the Journal is devoted to the theme ofinfection and immunity and to attempts to describe thebalance of host forces that protect against those infectiousforces that would invade. For most of our lives we remainunaware of the moment-by-moment interplay of hostresistance factors and infectious agents. Yet as describedin the articles herein, a natural sequence of clonal expan-sion of microbes develops in human beings born withfaulty immunity or in those whose immunity is temporar-ily overwhelmed by infection. There are 3 intersectingconcepts drawn out for us in this month’s Journal—immunodeficiency, infection, and cancer—that meet in acommon point in certain individuals, like the zero point of3-dimensional axes. Most infections of humankind areattributed to being the result of chance, exposure, and doseof infectious agent. Compelling arguments are now beingmade, however, that at least in the case of the more severeforms of these infections, invading organisms have takenadvantage of a hidden chink in the armament of presumednormal immunity. With special patients who need immu-nosuppressive treatments, there is often a manifestation ofchronic infection and even the appearance of cancer. Thisissue of the Journal clarifies some of those mysteriousmechanisms of immunity that enable the preservation oflife.

In the lead Current Reviews article, Tosi summarizesthe rapidly expanding field of innate immunity that,though lacking immune memory and clonal expansionof lymphocytes, probably protects against more infectionsthan does adaptive immunity.1 At epithelial skin surfaces,antimicrobial peptides disrupt the cell membranes ofpathogens and prevent numerous skin infections.2

Atopic eczema is a clinical example of deficiency in

b-defensins that leads to repeated staphylococcal in-fections in eczematous patches of affected skin.3

Mononuclear phagocytes and other cells contain Toll-like receptors that react with bacterial products, such asendotoxin and DNA sequences, and become active secre-tors of cytokines that regulate subsequent inflammatoryresponses.4 Children with deficiencies of the Toll-likereceptor signaling pathway involving the IL-1 receptor–associated kinase (IRAK) are subject to infections withpyogenic bacteria.5 Numerous cytokines are involved inimmune reactions, both protecting the host from infectionsand contributing to the complications of infections. In thelatter category, perhaps IL-1 and TNF-a are best knownfor their pathologic role in gram-negative bacterial toxin-induced shock.6 Chemokines and their receptors are ac-tive in numerous immunologic reactions to infections,but none are more visible than the CCR5 and CXCR4chemokine receptors that induce cognate receptor bindingof the HIV-1 glycoprotein 120 and facilitate entry ofthe HIV-1 virion into target cells.7 The value of naturalkiller (NK) cells assumes more importance as we under-stand the primal role that they serve in host protectionagainst viral infection and the development of cancer.8

Responding in an antigen-independent manner, NK cellsbind and lyse virus-infected host and cancer cells byperforin formation or apoptosis induction. Complement9

and neutrophil-invaded immunity10 round out the reper-toire of innate immunity, each contributing to that imme-diate response to infection that is so important prior to theengagement of the slower acquired immune responses.The illustration on the cover of this issue demonstrateshow neutrophils police the vascular endothelium and,within seconds of detecting chemoattractants created byinfections in the tissues, squeeze through intercellularspaces in pursuit of pathogens.11 On arrival at the site ofinfections, neutrophils engulf and kill microbes throughthe formation of superoxide. No prior memory of thesepathogens is necessary for this bacteriocidal function ofneutrophils.

Thorley-Lawson12 writes in the Molecular Mechanismsarticle on how the Epstein-Barr virus (EBV) takes up long-term residence in almost all human beings and occasion-ally produces lymphomas. Virtually all individuals withEBV lymphomas are either immunosuppressed (eg, trans-plant patients) or lack components of immunity on acongenital basis (eg, X-linked immunoproliferative dis-ease).13 EBV remains in a latent state in resting memoryB cells that do not express EBV proteins on their cell

From the Department of Pediatrics, Baylor College of Medicine, and the

Department of Allergy and Immunology, Texas Children’s Hospital.

Supported by the National Institutes of Health grants AI27551, AI36211,

HD41983, RR0188, HD079533, HL72705, HD078522, contract

202PICL05; the Pediatric Research and Education Fund, Baylor College

of Medicine; and the David Fund, Pediatrics AIDS Fund, and Immunology

Research Fund, Texas Children’s Hospital.

Received for publication May 31, 2005; accepted for publication June 1, 2005.


Reprint requests: William T. Shearer, MD, PhD, Department of Pediatrics,

Section of Allergy and Immunology, Baylor College of Medicine, 6621

Fannin St (MC-FC330.01), Houston, TX 77030. E-mail: wtsheare@

TexasChildrensHospital.org.

J Allergy Clin Immunol 2005;116:263-6.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.06.001

263

Review

sand

featu

rearticles



surface.14 Thus, these EBV-containing B cells remaininvisible to cytotoxic T cells. EBV is thought to play a rolenot only in lymphomas but also in Hodgkin’s disease,Burkitt’s lymphoma, and nasopharyngeal carcinoma.Thorley-Lawson acknowledges that these cells may justcarry EBV rather than being the result of EBV-initatedtransformation events. He also questions why more EBV-driven tumors are not seen in human populations, eventhose immunosuppressed individuals. The answer seemsto reside in the propensity of EBV-infected B cells to stopreplicating under the influence of 2 viral genes—latentmembrane protein (LMP) 1 and LMP2—and enter thelatent resting memory cell condition.15 Circumstances thatdisrupt the immune system change this resting memorycell condition and favor tumor development. Thus, whenhuman beings are given immunosuppressive drugs or havereceived therapeutic irradiation, immune forces are dis-rupted and the EBV B cell enters the replicating cell cycle.Unless killed by cytotoxic T cells, which respond to thenewly expressed EBV cell surface antigens, these acti-vated EBV cells could form oligoclonal tumor cells. Theseobservations hold importance for immunologically nor-mal human beings exposed to occupational or environ-mental conditions that cause immunosuppression, such asradiation in spaceflight.16

Mehandru and colleagues17 contribute a Perspectives/Update article to this issue that details the rapidlyunfolding discoveries of the crucial role of gastrointestinallymphatic tissue in acute HIV-1 infection. In the face ofrelatively stable peripheral blood CD41 T (helper) cellconcentrations in acute HIV-1 infection, there is a massivekill-off of tissue CD41 T cells, particularly those of thegastrointestinal tract.18-20 Moreover, these gastrointestinalCD41 T cells are of the memory phenotype and expressthe CCR5 chemokine receptor that attract the monocyto-tropic HIV-1 viral strains, as demonstrated in the simianmodel of HIV-1 infection.21,22 The emerging model ofpathogenesis of acute HIV-1 infection suggests that thelarge pool of memory CD41 T cells in mucosal surfacesbecomes preferentially infected and stimulates repetitiverounds of viral replication and CD41 T cell killing.Mehandru proposes that these discoveries will revampthe way clinicians decide when to intercede with anti-retroviral agents (ie, immediate versus deferred therapy),rekindle the debate of using immunodulators to reduce thewaves of inflammation and viral replication (eg, cyclo-sporine therapy during acute infection), accelerate the useof protective strategies for the gastrointestinal mucosalsurfaces (eg, microbicides and CCR5 blockers), andredirect vaccine strategies to mucosal surfaces.

In the Advances in Asthma, Allergy, and ImmunologySeries 2005: Basic and Clinical Immunology article inthis issue, Chinen and Shearer mention several note-worthy, recent publications dealing with the interplay be-tween immunity and infection.23 The importance of theHLA allele recognition system in viral infection is seenin human beings with certain HLA-B alleles withHIV-1 infection. HIV-1–infected individuals with HLA-B57 and HLA-B*5801 select for variants with a specific

mutation in the Gag epitope of HIV-1, but when thismutant viral strain is transmitted to another individualwith different HLA alleles, this epitope reverts to thewild type.24 In HIV-1 discordant couples, the risk ofHIV-1 transmission is 2-fold higher (independent ofHIV-1 viral load) if the couples share one or bothHLA-B alleles.25 In perinatal HIV-1 transmission,HLA-B*4901 and B*5301, alleles that inhibit mother-to-infant HIV-1 transmission (despite high HIV-1 viralload), differ from otherwise identical HLA-B*5001 andB*3501 alleles by 5 amino acids encoding the ligand forthe killer inhibitory receptor (KIR) 3DL1 for NK cells.26

The molecular basis for these 3 observations suggestsstrongly that recognition molecules on immune cellsgovern subsequent viral mutation and viral eliminationthrough cytotoxic T cells and NK cells. Also summarizedin the Advances article are the discoveries that mast cellsparticipate in host defense via recognition of Toll-likereceptors and viruses27 and secretion of cytokines thatrecruit effector cell.28 Articles in the area of primaryimmunodeficiency and infectious diseases are also cited,perhaps none more important than the identification ofrisks of malignancy when retroviral vectors are used toinsert gene constructs in stem cells bone marrow derivedin severe combined immunodeficiency.29 In this instance,the retroviral vector has the potential to insert into thehuman genome in the promotor region of oncogenes andto trigger the development of T-cell leukemia.30 Related tothese observations in primary immunodeficiency is theImages in Allergy and Immunology article that pictures thecase of a child who developed severe mosquito bitehypersensitivity, ulcerating skin lesions, enlarged anddraining adjacent lymph nodes, and marked hepatosplen-omegaly.31 Studies reveal that this child developed pro-liferation of EBV-containing NK cells similar, if notidentical, to that seen in the few reported cases of chronicactive EBV infection of NK cells that result in NK cellleukemia and lymphoma.32 It is possible that this child isan example of the atypical immunodeficiency that presentswith a more common and less marked clinical phenotypethat ultimately might be resolved by detection of causalgenes, as proposed by Casanova33 and reviewed byBonilla and Geha34 in an editorial in this issue.

In addition to the interaction of viruses with immuno-deficiency, there is strong evidence for a pathogenic rolefor viruses in allergic diseases. For example, rhinovirusescause more than 50% of upper respiratory infections andare thought to be responsible for the induction of acuteexacerbations of asthma in the lower airway. Friedlanderand Busse35 review this evidence in this issue, and findthat respiratory viruses are associated with approximately80% of children and 50% of adults with wheezing epi-sodes and that infection of the upper and lower respiratorymucosal surfaces induces increased airway hyperrespon-siveness. This concept of rhinovirus induction of asthmaincludes (1) the attachment of the intercellular adhesionmolecule 1 (ICAM-1) to the viral capsid molecules36 and(2) the stimulation of proinflammatory cytokines IL-6,IL-8, IL-16, and RANTES chemokine.37 The net result of


AUGUST 2005

264 Shearer

Review

sand

featu

rearticle

s

upregulation of these mediators is an influx of eosinophils,monocytes, T cells, macrophages, and neutrophils intorespiratory tissue.38 As a result of this increased inflam-mation of the airways, angiogenic growth factors mightinduce tissue remodeling of respiratory mucosa and couldcause a permanent change in lower airway architectureand increased difficulties for treatment programs.

All in all, the theme of this month’s Journal seems tohave been substantially illustrated by the contributions oftalented experts in immunity and infection. These inter-related concepts can be considered the 2 sides of a coin.Perhaps a better analogy is that of a teeter-totter: when thesitting board is horizontal, there is a balance betweenimmunity and infection (Fig 1). When immunity is down,infections rise and immunity must be strengthened to gainbalance, with the result that the inflammation of immunityoften overshoots and infection drops. However, evidenceis being gathered to strongly suggest that when thisbalance is upset, as is the case with immunodeficiencydiseases, certain viruses are able to escape strong immuneresponse and hide in a latent condition. When individualswho harbor latent viruses, such as patients receiving im-munosuppressive drugs or therapeutic radiation, encoun-ter an additional force, the latent virus is forced into its lifecycle that yields outgrowths of clones of virus-containingtransformed cells. More understanding of these balancingforces of immunity and infection is necessary so that we,as clinician-investigators, can intervene with the some-times threatening consequences of imbalances in theforces of the immune system and infection.

I thank Carolyn Jackson and Ruth Herrera for assistance with the

preparation of this manuscript.

REFERENCES

1. Tosi M. Innate immune responses to infection. J Allergy Clin Immunol

2005;116:241-9.

2. Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat Rev

Immunol 2003;3:710-20.

3. Ong PY, Ohtake T, Brandt C, Strickland I, Boguniewicz M, Ganz T,

et al. Endogenous antimicrobial peptides and skin infections in atopic

dermatitis. N Engl J Med 2002;347:1151-60.

4. Zarember KA, Godowski PJ. Tissue expression of human Toll-like

receptors and differential regulation of Toll-like receptor mRNAs in

leukocytes in response to microbes, their products, and cytokines.

J Immunol 2002;168:554-61.

5. Picard C, Puel A, Bonnet M, Ku CL, Bustamante J, Yang K, et al.

Pyogenic bacterial infections in humans with IRAK-4 deficiency.

Science 2003;299:2076-9.

6. Oppenheim JJ, Feldman M. Introduction to the role of cytokines

in innate host defense and adaptive immunity. In: Oppenheim JJ,

Feldman M, Durum SK, Hirano T, Vilcek J, Nicola N, editors. The

cytokine reference. San Diego: Academic Press; 2001. p. 3-20.

7. Glass WG, Rosenberg HF, Murphy PM. Chemokine regulation of

inflammation during acute viral infection. J Allergy Clin Immunol

2003;3:467-73.

8. Smyth MJ, Cretney E, Kelly JM, Westwood JA, Street SE, Yagita H,

et al. Activation of NK cell cytotoxicity. Mol Immunol 2005;42:501-10.

9. Berger M, Frank MM. The serum complement system. In: Stiehm ER,

Ochs HD, Winkelstein JA, editors. Immunologic disorders in infants and

children. 5th ed. Philadelphia: Elsevier Saunders; 2004. p. 20-62.

10. Tosi MF. Immunologic and phagocytic responses to infection. In: Feigin

RD, Cherry JD, Demmler GJ, Kaplan S, editors. Textbook of

pediatric infectious diseases. 5th ed. New York: WB Saunders; 2004.

p. 652-84.

11. Seo SM, McIntire LV, Smith CW. Effects of IL-8, Gro-alpha, and

LTB(4) on the adhesive kinetics of LFA-1 and Mac-1 on human

neutrophils. Am J Physiol Cell Physiol 2001;281:C1568-78.

12. Thorley-Lawson DA. EBV the protypical human tumor virus—just how

bad is it? J Allergy Clin Immunol 2005;116:251-61.

13. Shearer WT, Ritz J, Finegold MJ, Guerra IC, Rosenblatt HM, Lewis DE,

et al. Epstein-Barr virus-associated B-cell proliferations of diverse clonal

origins after bone marrow transplantation in a 12-year-old patient with

severe combined immunodeficiency. N Engl J Med 1985;312:1151-9.

14. Hochberg D, Middeldorp JM, Catalina M, Sullivan JL, Luzuriaga K,

Thorley-Lawson DA. Demonstration of the Burkitt’s lymphoma

Epstein-Barr virus phenotype in dividing latently infected memory cells

in vivo. Proc Natl Acad Sci 2004;101:239-44.

15. Hochberg DR, Thorley-Lawson DA. Quantitative detection of viral gene

expression in populations of Epstein-Barr virus-infected cells in vivo.

Methods Mol Biol 2005;292:39-56.

16. Shearer WT, Zhang S, Reuben RM, Lee B, Butel JS. Effects of radiation

and latent virus on immune responses in a space flight model. J Allergy

Clin Immunol 2005;115:1297-303.

17. Mehandru S, Tenner-Racz K, Racz P, Markowitz M. The gastrointestinal

tract is critical to the pathogenesis of acute HIV-1 infection. J Allergy

Clin Immunol 2005;116:419-22.

18. Guadalupe M, Reay E, Sankaran S, Prindiville T, Flamm J, McNeil A,

et al. Severe CD41 T-cell depletion in gut lymphoid tissue during

primary human immunodeficiency virus type 1 infection and substantial

delay in restoration following highly active antiretroviral therapy. J Virol

2003;77:11708-17.

19. Mehandru S, Poles MA, Tenner-Racz K, Horowitz A, Hurley A, Hogan

C, et al. Primary HIV-1 infection is associated with preferential depletion

of CD41 T lymphocytes from effector sites in the gastrointestinal tract.

J Exp Med 2004;200:761-70.

20. Brenchley JM, Schacker TW, Ruff LE, Price DA, Taylor JH, Beilman

GJ, et al. CD41 T cell depletion during all stages of HIV disease

occurs predominantly in the gastrointestinal tract. J Exp Med 2004;200:

749-59.

FIG 1. Balance between immunity and infection. A, Normal immune system. B, Immunodeficiency.



Shearer 265

Review

sand

featu

rearticles

21. Mattapallil JJ, Douek DC, Hill B, Nishimura Y, Martin M, Roederer M.

Massive infection and loss of memory CD41 T cells in multiple tissues

during acute SIV infection. Nature 2005;434:1093-7.

22. Li Q, Duan L, Estes JD, Ma ZM, Rourke T, Wang Y, et al. Peak SIV

replication in resting memory CD41 T cells depletes gut lamina propria

CD41 T cells. Nature 2005;434:1148-52.

23. Chinen J, Shearer WT. Advances in asthma, allergy, and immunology

series 2005: basic and clinical immunology. J Allergy Clin Immunol

2005;116:411-8.

24. Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M,

et al. HIV evolution: CTL escape mutation and reversion after transmis-

sion. Nat Med 2004;10:282-9.

25. Dorak MT, Tang J, Penman-Aguilar A, Westfall AO, Zulu I, Lobashevsky

ES, et al. Transmission of HIV-1 and HLA-B allele-sharing within

serodiscordant heterosexual Zambian couples. Lancet 2004;363:2137-9.

26. Winchester R, Pitt J, Charurat M, Magder LS, Goring HH, Landay A,

et al. Mother-to-child transmission of HIV-1: strong association with

certain maternal HLA-B alleles independent of viral load implicates innate

immune mechanisms. J Acquir Immune Defic Syndr 2004;36:659-70.




28. Marshall JS, Jawdat DM. Mast cells in innate immunity. J Allergy Clin

Immunol 2004;114:21-7.

29. Chinen J, Puck JM. Successes and risks of gene therapy in primary

immunodeficiencies. J Allergy Clin Immunol 2004;113:595-603.

30. Cavazzana-Calvo M, Lagresle C, Hacein-Bey-Abina S, Fischer A. Gene

therapy for severe combined immunodeficiency. Annu Rev Med 2005;

56:585-602.

31. Pacheco SE, Gottschalk SM, Gresik MV, Dishop MK, Okmaura T,

McCormick TG. Chronic active Epstein-Barr virus (CAEBV) infection

of NK cells presenting a severe skin reaction to mosquito bites. J Allergy

Clin Immunol 2005;116:470-2.

32. Tokura Y, Ishihara S, Tagawa S, Seo N, Ohshima K, Takigawa M.

Hypersensitivity to mosquito bites as the primary clinical manifestation

of a juvenile type of Epstein-Barr virus-associated natural killer cell

leukemia/lymphoma. J Am Acad Dermatol 2001;45:569-78.

33. Casanova JL, Fieschi C, Bustamante J, Reichenbach J, Remus N,

von Bermuth H, et al. From ‘‘idiopathic’’ infectious diseases to ‘‘atyp-

ical’’ primary immunodeficiencies. J Allergy Clin Immunol 2005;116:

426-30.

34. Bonilla S, Geha RS. Are you immunodeficient? J Allergy Clin Immunol

2005;116:423-5.

35. Friedlander G, Busse WW. The role of rhinovirus in asthma exacerba-

tions. J Allergy Clin Immunol 2005;116:267-73.

36. Yamaya M, Sasaki H. Rhinovirus and asthma. Viral Immunol 2003;16:

99-109.

37. Papadopoulos NG, Bates PJ, Bardin PG, Papi A, Leir SH, Fraenkel DJ,

et al. Rhinoviruses infect the lower airways. J Infect Dis 2000;181:

1875-84.

38. Gern JE. Rhinovirus respiratory infections and asthma. Am J Med 2002;

112:19S-27S.


AUGUST 2005

266 Shearer

Review

sand

featu

rearticle

s

Asthma diagnosis and treatment

Rostrum

The role of rhinovirus in asthma exacerbations

Samuel L. Friedlander, MD, and William W. Busse, MD Madison, Wis

Rhinoviruses are a major cause of asthma exacerbations in

children and adults. With the use of sensitive RT-PCR methods,

respiratory viruses are found in approximately 80% of

wheezing episodes in children and in approximately one half of

such episodes in adults. Rhinovirus is a member of the family

Picornaviridae, and acute rhinovirus infections occur

predominantly in the upper airway. This virus has also been

identified in the lower airway, and it might cause acute

wheezing through the production of proinflammatory

mediators with a resulting neutrophilic inflammatory response.

Precisely how this process leads to increases in airway

hyperresponsiveness and airway obstruction is not fully

established. However, risk factors for wheezing with colds

include asthma and atopy, extremes in age, and perhaps having

a deficient TH1 response to rhinovirus. With the use of in vitro

models and experimental inoculation studies, significant

advances have led to a better understanding of the mechanisms

by which rhinovirus infections cause asthma exacerbations.

Advances in our understanding of this interaction might

provide knowledge that could ultimately lead to specific

treatment modalities to prevent and/or treat this significant

burden of asthma exacerbations. (J Allergy Clin Immunol

2005;116:267-73.)

Key words: Rhinovirus, asthma exacerbations, virology, cytokine

response profiles, mechanisms of asthma

Viral upper respiratory tract infections (URIs) areknown to cause exacerbations of asthma. A significantand increasing body of evidence demonstrates that in largepart the primary respiratory infection causing these exac-erbations is rhinovirus (RV), the cause of more than 50%of URIs.1 The frequency of URI-provoked asthma makesit especially important to understand the role and mech-anisms whereby RV infections lead to asthma exacerba-tions, the basic virologic features of RV, their ability toinfect the lower airway, the host susceptibility factors, andmechanisms leading to airflow obstruction.

WHAT ROLE DOES RV PLAY IN ASTHMAEXACERBATIONS?

Asthma exacerbations are most commonly precipitatedby viral URIs, particularly with RV,2 and often occurdespite concurrent use of appropriate controller medica-tions. Detecting respiratory viruses—in particular, RV—by culture methodology alone has been insensitive and haspreviously underestimated the role of respiratory viruses inasthma exacerbations, especially in adults. Viral detectionrates in asthma exacerbations have significantly increasedwith the use of sensitive methods and have thus under-scored the overwhelming importance of respiratory vi-ruses in asthma exacerbations. When RT-PCR is used tosupplement conventional culture techniques, viruses havebeen found in approximately 80% of wheezing episodesin school-age children and in approximately one half of theacute wheezing episodes in adults. Of the respiratoryviruses identified in these circumstances, RV is mostcommonly found and is detected 65% of the time.2,3

A pivotal study by Johnston and colleagues2 in childrenaged 9-11 years old with histories of asthma symptomsfound that 80% to 85% of asthma exacerbations that wereassociated with reduced peak expiratory flow rates andwheezing were due to viral URIs. Without the use ofRT-PCR, the authors reported, the viral detection rate inthis study would have been only around 40%.

Similarly, high rates of asthma attacks due to RV werefound in adults. Nicholson et al3 reported on 138 youngadults with asthma recruited from general practice, thehospital, and the community. In this longitudinal study,80% of asthma episodes (223 of 280), described assymptoms of wheeze, chest tightness, or breathlessness,were associated with colds. Objectively, viruses weredetected in 57% of people with symptomatic colds andasthma exacerbations. In more severe asthma exacerba-tions with reductions in peak flow measurements of

Abbreviations usedG-CSF: Granulocyte colony-stimulating factor

ICAM: Intercellular adhesion molecule

IL-1ra: IL-1 receptor antagonist

RV: Rhinovirus

URI: Upper respiratory tract infection

From the Division of Allergy and Immunology, Department of Medicine,

University of Wisconsin, Madison.

Received for publication April 7, 2005; revised June 3, 2005; accepted for

publication June 7, 2005.


Reprint requests: William W. Busse, MD, Department of Medicine, K4/912

CSC-9988, 600 Highland Avenue, Madison, WI 53792. E-mail: wwb@

medicine.wisc.edu.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.06.003

267

Asthmadiagnosisand

treatm

ent



50 L/min, viruses were detected in 44% of episodes. Incomparison with detection rates by cell culture, RT-PCRwas 5 times more sensitive in identifying human RV inadults with respiratory infections.3

Further evidence to support the role of viral infection inasthma exacerbations also includes reports that peaks inhospital admissions for asthma significantly correlate withseasonal patterns of viral URIs.4 In the United States, RVinfection occurs most commonly in the fall and spring.5

Thus, current evidence strongly supports the concept thatRV respiratory infections are the major cause of acuteasthma exacerbations.

WHAT ARE THE VIROLOGIC FEATURESOF RVS?

The genera RV and Enterovirus are classified within thefamily Picornaviridae. There are more than 100 serotypesof RV; this explains, in part, the lack of an effectivevaccine against the major etiologic agent causing thecommon cold. RV is a small, single-stranded RNA viruswhose capsid contains 4 proteins (Fig 1). Three of theseproteins, VP1, VP2, and VP3, are located on the surfaceof the capsid and are responsible for its antigenic diversity;the fourth, VP4, is located inside the virus and anchorsthe RNA core to the viral capsid.1 The majority of RVserotypes bind to intercellular adhesion molecule (ICAM)1, whereas approximately 10% bind to the low-densitylipoprotein receptor.6,7

Typically, RV infects small clusters of cells in theepithelial layer with little cellular cytotoxicity. Although

increased polymorphonuclear neutrophils are seen ininfected nasal epithelium,8 little or no mucosal damageoccurs from the infection; this suggests that RV is likelyto cause asthma exacerbations by mechanisms other thandirect cellular killing.9 Even with large inoculating dosesof virus, less than 10% of cells in primary airway epithe-lium cultures become infected. However, although RV-induced cytotoxicity is difficult to detect in vivo, an in vitrostudy has demonstrated cytopathic effects when high titersof virus are inoculated with sparsely seeded monolayercultures of human bronchial epithelial cells.10 Moreover,the RV serotypemight also be an important determinant ofthis in vitro–detected cytotoxicity.

ARE RV INFECTIONS LIMITED TO THEUPPER AIRWAY?

An infection with RV leads to symptoms of thecommon cold, which is primarily an upper airway illness.Because RV is primarily an infection of the upper airway,early research efforts were directed toward determiningwhether (a) RV infections could infect the lower airwaysdirectly and provoke asthma, (b) their actions on asthmaoccurred via indirect mechanisms due to the upper airwayinfection only, or (c) a combination of the 2 methods isresponsible. Insight into these questions could suggestpotential target areas to act therapeutically to prevent ortreat an asthma exacerbation.

Debate initially focused on whether RV could existand replicate in the lung to directly cause lower airwayinflammation. This was based on limited studies dem-onstrating that RV replication was optimal at 33C, thetemperature of the upper airways. To address this issue,direct thermal mapping of the lower airways was per-formed. While human subjects breathed room air (26C),the temperature in the subjects averaged 32C in the uppertrachea and 35.5C in the subsegmental bronchi. Thesefindings refuted a possible limitation of RV growth due tohigher temperatures in the lower airway.11Moreover, withthe use of multiple RV serotypes, it was possible to detecthigh viral titers in cell cultures at 37C; little significantdifference in replication was found when wild-type RVisolates were used at 33C compared to 37C. In fact,some serotypes grew more effectively at the highertemperature.10 In addition, when primary cultures oflower airway bronchial epithelial cells and upper airwayadenoidal epithelial cells were used, RV appeared to infectboth upper and lower segments of the respiratory tree withsimilar ability (Fig 2).9

Several additional lines of evidence support the abilityof RV to infect the lower airways directly. When bron-choscopy was used to collect samples from subjects withsymptomatic experimental infections, RV was detectedfrom bronchial brush specimens.12 Furthermore, with theuse of RT-PCR and Southern blotting, RV genetic mate-rial was found in higher amounts in cells of bronchialalveolar lavage fluid than in supernatant; this suggests thatthe virus was located intracellularly.13 However, the role

FIG 1. Transverse section through the center of a pentamer

depicting entry of its cellular receptor, ICAM-1, and the location

of the drug-binding pocket just beneath the canyon floor. An ion,

located at each pentamer center in RV-1A, 214, 216 is tentatively

identified as calcium, which is necessary for attachment of some

RVs. Modified with permission from Gwaltney JM. Rhinovirus. In:

Mandell GL, Douglas RG, Bennett JE, Dolin R, editors. Mandell,

Douglas, and Bennett’s principles and practice of infectious dis-

eases. 6th ed. New York; Elsevier/Churchill Livingstone; 2005. p.

2185-94.


AUGUST 2005

268 Friedlander and Busse

Asth

madiagnosis

and

treatm

ent

of contamination from the upper airway could not bedefinitively excluded in these studies.

Another investigation found RV-16 RNA in 50% ofbronchial biopsies in experimentally inoculated humanvolunteers.10 In this study, in situ hybridization was usedto localize viral RNA by hybridizing the sequence ofinterest to the complementary replicative strand of thevirus. This technique likely excludes the possibility ofcontamination of the lower airway with virus from theupper airway. This was further supported by the finding ofviral replication in the lower airway as well as increases inviral RNA and the production of new viral proteins.Furthermore, the frequency of lower airway infection wassimilar to that observed in the upper airway; this indicatesthat infection of the lower airways might be relativelycommon as part of the natural history of RV infection.Finally, a recent study showed that an experimentalRV infection was associated with virus detection in largelower airways biopsy samples by immunohistochemistryor qPCR in 17 of 19 subjects, but less so in the distalairways.14 Thus, RV is able to infect both the upper andlower airways.

It is likely that the lower airways are infected as a resultof self-inoculation from coughing, sneezing, or perhapsbreathing. Whether the concentration of virus in the lowerairways is large enough to produce clinically relevanteffects is still not established. These studies supportthe concept that RV is a lower as well as an upperrespiratory tract pathogen, and infection of the lower

airway directly likely contributes to viral-induced exacer-bations of asthma.

WHAT ARE THE EFFECTS OF RV INFECTIONON THE MECHANISMS OF AIRWAYPHYSIOLOGY IN ASTHMA?

Multiple studies demonstrate the adverse effects of RVon airway physiology in asthma. In school-age children,symptoms of either upper or lower respiratory tract in-fection were shown to last a week, and during these infec-tious episodes, the peak flow rates fell for a medianduration of 2 weeks.2 In another study, asthmatic subjectswere experimentally inoculated with RV-16 and found todemonstrate modest changes in increased airway hyper-responsiveness, airway obstruction, and inflammation.15

Experimental RV-16 infection also has been shown toreduce FEV1 in patients with mild asthma.16 In addition,increases in existing airway inflammation have occurredafter segmental bronchoprovocation in atopic subjects,suggesting that enhanced airway inflammation is a featureof RV-associated asthma exacerbations.17

To support this possibility, subjects with allergic rhini-tis, but not with active asthma, were inoculated with RV;they were found to have significantly increased airwayhyperreactivity as well as a significantly increased inci-dence of late asthmatic reactions, defined as a 15%decrease in FEV1 approximately 6 hours after antigen

FIG 2. Immunohistochemistry of airway tissue from experimentally infected normal human surgical

specimens infected ex vivo. A, Bronchial tissue specimens were inoculated ex vivo and were incubated in

tissue culturemedium for 24 hours. After washing to eliminate extracellular virus, the tissue was embedded in

paraffin, sectioned, and stained for the presence of RV serotype 16. B, Uninfected bronchial tissue specimen

was processed as in panel A. C, Adenoidal tissue specimen infected ex vivo for 6 hours. Bar: 50 mm. Reprinted

with permission from Mosser AG, Brockman-Schneider R, Amineva S, Burchell L, Sedgwick JB, Busse WW,

et al. Similar frequency of rhinovirus-infectible cells in upper and lower airway epithelium. J Infect Dis

2002;185:734-43.



Friedlander and Busse 269

Asthmadiagnosisand

treatm

ent

challenge.18 Before infection, only 1 patient in this studyhad a late asthmatic reaction. During the acute infection,this number increased to 8 of 10 subjects (P = .0085). Theeffect occurred independently of the enhancement inairway reactivity experienced during a cold alone. Thisdemonstrates that in addition to causing airway hyperre-activity, RV also promotes the development of late-phaseresponses, even in nonasthmatic patients.

RV infection also promotes eosinophil recruitment toairway segments after antigen challenges. Calhoun et al17

used segmental bronchoprovocation with antigen afterinoculation with RV-16 in subjects with allergic rhinitis.After infection, bronchial alveolar lavage fluid revealedenhanced histamine release immediately and increasedeosinophil recruitment 48 hours after antigen challenge.Interestingly, the increase in eosinophils persisted for upto 1 month after infection in some subjects. The effect ofRV on airway inflammation appeared to be an augmen-tation of allergen-specific responses. Thus, enhancementof antigen-induced mediator release from pulmonarymast cells and basophils and eosinophilic recruitment,either directly or via cytokines, could provide onemechanism by which late allergic reactions and airwayhyperresponsiveness are enhanced by viral uncoating andmight act to intensify the airway inflammatory responseto allergen.

Conversely, when nasal allergen challenges in atopicpatients were performed before experimental RV inocu-lation, the onset of cold symptoms was delayed and theresponses were less severe in comparison with whatwas seen in patients without allergies.19 Delayed nasalinflammation, with attenuation of the increase in IL-6,IL-8, and neutrophils seen in infection, was also found inthe group primed with nasal antigen challenge. This mighthave been due to cytokine profile changes with increasedexpression of IFN-g and IL-2, local production of nitric

oxide, or antiviral effects of eosinophil products. Thus, thetiming and intensity of antigen exposure play an importantrole in the severity level and subsequent possible compli-cations of a cold.

HOW DOES RV MODULATE INFLAMMATORYMEDIATORS OF EPITHELIAL CELLSCONTRIBUTING TO ASTHMAEXACERBATIONS?

Epithelial cells are the principal targets of RV infec-tions, allow viral replication, and likely initiate immuneresponses (Fig 3).20,21 Papadopoulos et al10 found localinduction of proinflammatory mediators that could pro-vide a mechanism to explain how lower airway infectioncan lead to inflammation and asthma. RV infection re-sulted in an increase in mRNA expression and subsequenttranslation of IL-6, IL-8, and IL-16. This also occurredwith RANTES, a C-C chemokine with chemoattractantactivity for eosinophils, monocytes, and T lymphocytes.IL-6 and IL-8 are proinflammatory cytokines, and IL-8 is aspecifically potent chemoattractant for neutrophils. IL-16is a powerful lymphocyte chemoattractant and activatorof macrophages and eosinophils and appears to be animportant mediator in the pathogenesis of asthma andlower airway inflammation due to RV.10 The inflamma-tory actions of RV appear to center on its ability togenerate a variety of phlogistic mediators.

Generation of these cytokines correlates with theworsening of respiratory physiology. For example, IL-1enhances airway smoothmuscle contraction in response tobronchospastic agents and attenuates smooth muscledilation responses to bronchodilators.22,23 Differences inimmune response, such as the modulation of costimula-tory molecules and the induction of antigen presentation,

FIG 3. RV induces epithelial cells to produce proinflammatory cytokines leading to airway hyperresponsive-

ness, neurogenic inflammatory responses, mucous secretion, inflammatory cell recruitment and activation,

and plasma leakage. Created with information from Gern JE. Rhinovirus respiratory infections and asthma.

Am JMed 2002;112(Suppl 6A):19S-27S and from YamayaM, Sasaki H. Rhinovirus and asthma. Viral Immunol

2003;16:99-109.


AUGUST 2005


Asth

madiagnosis

and

treatm

ent

might explain how RV infections cause acute exacerba-tions in asthmatic patients.

Virus-induced epithelial damage might cause increasedpermeability of the mucosal layer and thus increaseallergen contact with immune cells to promote neurogenicinflammation. In addition, viruses can enhance vagallymediated reflex bronchoconstriction, possibly by limitingthe function of the M2 muscarinic receptor.24 Viralreplication activates epithelial cells to initiate innate andadaptive immune responses as well as the generation ofoxidative stress.25 Also, double-stranded RNA synthe-sized in virus-infected cells induces the cytokines IL-8 andRANTES, which initiate proinflammatory and antiviralpathways within the cell.24

Upregulation, or activation, of ICAM-1, the principalreceptor for RV, might increase tissue susceptibility to themajor group RV and subsequent infection. The asthmaphenotype, which is associated with increased ICAM-1expression, might therefore be associated with increasedsusceptibility and complications from RV infection.21

Chronic antigen challenge can also increase ICAM-1 ex-pression of the airway epithelium, and RV infection itselfcan increase ICAM-1 expression through production ofIL-1b and a nuclear factor-kb–dependent mechanism.This might lead to the amplification of airway inflamma-tion after RV infection.21,26

In addition, RVmight enhance existing inflammation toa greater degree in asthmatic subjects than in those withoutthe disease. For example, after inoculation with the virus,nasal lavage levels of IL-8 and the proinflammatorymediator IL-1b were increased in asthmatic patients.27

In this study, a small increase in the anti-inflammatorymarker IL-1 receptor antagonist (IL-1ra), a competitiveinhibitor of IL-1, also occurred in asthmatic subjectstreated with budesonide, whereas lower levels were foundin these patients at baseline. These findings suggest thatRV might be able to alter the proinflammatory/anti-inflammatory balance of IL-1b/IL-1ra toward inflamma-tion more markedly in people with existing and activeasthma.

Cytokine response profiles generated by RV mighttranslate into neutrophilic inflammation in both the upperand lower airways. Local RV infection is associated withincreased levels of IL-8, a potent chemoattractant forneutrophils, and also granulocyte colony-stimulating fac-tor (G-CSF) in nasal secretions and later in the circulation.Increased concentrations of circulatory G-CSF could acton the bonemarrow to increase the circulating neutrophils.Thus, a local response in nasal epithelium to RV infectioncan result in a systemic inflammatory reaction.20

Elevated neutrophil counts are also found in the lowerairways with RV infection. Through use of bronchoscopyand bronchial washes, significant increases in airwaylumen neutrophils were found 96 hours after inoculationwith RV-16 in patients with allergic asthma.28 Infectedbronchial epithelium induces the secretion of proinflam-matory cytokines, including IL-1, IL-8, TNF-a, IL-10,and IFN-a, as well. This stimulates the recruitment ofinflammatory cells and neutrophilia. Products of neutro-

phil activation could cause airway obstruction through theproduction of elastase, which also upregulates goblet cellmucus secretion.29

WHAT ARE THE RISK FACTORS FORWHEEZING WITH A COLD?

Various risk factors increase the susceptibility of sub-jects for more severe lower respiratory complicationsfrom an RV infection, such as wheezing, bronchitis, andpneumonia (Table I). These include having low neutral-izing antibody titers to RV, being an infant, being elderly,having chronic lung disease, being a smoker, and being anindividual with existing asthma.22

In addition, subjects who are low producers of IFN-g inresponse to RV and are atopic appear to be more at risk forwheezing or having a severe respiratory infection. Brookset al30 demonstrated that whereas RV induces IFN-g,which is consistent with a strong TH1-like immune re-sponse, those asthmatic patients with diminished, ordeficient, TH1 responses to RV were characterized byincreased airway hyperresponsiveness. Moreover, theratio of RV-16–induced IFN-g:IL-5, a measure ofTH1:TH2 balance, correlated with FEV1. These findingsare similar, in general, to what is known about TH1 andTH2 responses in asthma. In another study, subjects withpersistent and severe asthma displayed a defect in IFN-gproduction, whereas their increased IL-5 responses werefelt to reflect the presence of atopy but were not specif-ically linked to asthma itself.31 Collectively, these findingsdemonstrate that a deficiency of the TH1 response, ratherthan an increased TH2 response, is responsible for RV’sadverse effect on the airways.

The importance of IgE and eosinophilic airway inflam-mation was demonstrated by a study showing synergisticinteractions between RV infection and allergic airwayinflammation.32 In this study, which focused on childrenaged 2-16 years old, the odds ratios for wheezing with RVdetected by RT-PCR in addition to positive radioallergo-sorbent test results, nasal eosinophilia, and elevated nasaleosinophil cationic protein were 17, 21, and 25, respec-tively. The odds ratios for wheezing with any of these4 risk factors alone were much lower, between 3.2 and

TABLE I. Risk factors for severe lower respiratory tract

complications from rhinovirus infection

Asthma

Atopy

Elevated nasal eosinophils or eosinophil cationic protein

Infants and elderly

Low neutralizing antibody titers to rhinovirus

Chronic lung diseases

Smoking

Low IFN-g producers

High IL-5 producers

Modified with permission from Gern JE. Rhinovirus respiratory infections

and asthma. Am J Med 2002;112(Suppl 6A):19S-27S.




Asthmadiagnosisand

treatm

ent

8; this shows the importance of IgE and eosinophil-driveninflammatory responses.

Zambrano et al33 reported that the existence of airwayinflammation prior to virus inoculation predisposed sub-jects to a more deleterious response to RV. Patients wereinoculated with RV-16, and compared with those withoutasthma, those with mild asthma demonstrated increasedairway hyperresponsiveness, decreased FEV1 at baseline,and increased upper and lower respiratory tract symptomscores in response to the infection. Asthmatic patients withelevated IgE profiles also demonstrated higher bloodeosinophil counts, increased eosinophil cationic proteinin nasal washes, and both an increased expired nitricoxide, a marker of inflammation, and decreased solubleICAM-1 in nasal washes at baseline and during coldsymptoms. These findings suggest that patients withasthma, who are highly atopic, might be more likely tohave increased levels of airway inflammation and be atgreater risk for asthma exacerbations in response to RVinfection.

SUMMARY

The importance of RV in asthma exacerbations isestablished in both adults and children. The complexmechanisms by which their interaction provokes asthmaare becoming better understood. RV appears to have adirect and negative impact on the lower airways andcauses an increase in obstructive airway symptoms andphysiology. This effect on airway function is felt to occuras the virus upregulates proinflammatory cytokines andpredisposes the asthmatic patient to more severe respira-tory infections and hence to exacerbations. Defects inTH1-type immune responses appear to be an importantfactor in causing airway inflammation in people withasthma. Further work is needed to better explore themechanisms behind the association between asthmaexacerbations and RV infections. This might ultimatelylead to treatment modalities to prevent and/or treat thesignificant burden of asthma exacerbations caused byRV infection.

REFERENCES

1. Greenberg SB. Respiratory consequences of rhinovirus infection. Arch

Intern Med 2003;163:278-84.

2. Johnston SL, Pattemore PK, Sanderson G, Smith S, Lampe F, Josephs L,

et al. Community study of role of viral infections in exacerbations of

asthma in 9-11 year old children. BMJ 1995;310:1225-9.

3. Nicholson KG, Kent J, Ireland DC. Respiratory viruses and exacerba-

tions of asthma in adults. BMJ 1993;307:982-6.

4. Johnston SL, Pattemore PK, Sanderson G, Smith S, Campbell MJ,

Josephs LK, et al. The relationship between upper respiratory infections

and hospital admissions for asthma: a time-trend analysis. Am J Respir

Crit Care Med 1996;154(3 Pt 1):654-60.

5. Arruda E, Pitkaranta A, Witek TJ Jr, Doyle CA, Hayden FG. Frequency

and natural history of rhinovirus infections in adults during autumn.

J Clin Microbiol 1997;35:2864-8.

6. Casasnovas JM. The dynamics of receptor recognition by human

rhinoviruses. Trends Microbiol 2000;8:251-4.

7. Gwaltney JM. Rhinovirus. In: Mandell GL, Douglas RG, Bennett JE,

Dolin R, editors. Mandell, Douglas, and Bennett’s principles and practice

of infectious diseases. 6th ed. New York: Elsevier/Churchill Livingstone;

2005. p. 2185-94.

8. Winther B, Farr B, Turner RB, Hendley JO, Gwaltney JM Jr, Mygind N.

Histopathologic examination and enumeration of polymorphonuclear

leukocytes in the nasal mucosa during experimental rhinovirus colds.

Acta Otolaryngol Suppl 1984;413:19-24.

9. Mosser AG, Brockman-Schneider R, Amineva S, Burchell L, Sedgwick

JB, Busse WW, et al. Similar frequency of rhinovirus-infectible cells in

upper and lower airway epithelium. J Infect Dis 2002;185:734-43.

10. Papadopoulos NG, Bates PJ, Bardin PG, Papi A, Leir SH, Fraenkel DJ,

et al. Rhinoviruses infect the lower airways. J Infect Dis 2000;181:

1875-84.

11. McFadden ER Jr, Pichurko BM, Bowman HF, Ingenito E, Burns S,

Dowling N, et al. Thermal mapping of the airways in humans. J Appl

Physiol 1985;58:564-70.

12. Halperin SA, Eggleston PA, Hendley JO, Suratt PM, Groschel DH,

Gwaltney JM Jr. Pathogenesis of lower respiratory tract symptoms in

experimental rhinovirus infection. Am Rev Respir Dis 1983;128:806-10.

13. Gern JE, Galagan DM, Jarjour NN, Dick EC, Busse WW. Detection of

rhinovirus RNA in lower airway cells during experimentally induced

infection. Am J Respir Crit Care Med 1997;155:1159-61.

14. Mosser AG, Vrtis R, Burchell L, Lee WM, Dick CR, Weisshaar E, et al.

Quantitative and qualitative analysis of rhinovirus infection in bronchial

tissues. Am J Respir Crit Care Med 2005;171:645-51.

15. Fraenkel DJ, Bardin PG, SandersonG, Lampe F, Johnston SL, Holgate ST.

Lower airways inflammation during rhinovirus colds in normal and in

asthmatic subjects. Am J Respir Crit Care Med 1995;151(3 Pt 1):879-86.

16. Grunberg K, Timmers MC, de Klerk EP, Dick EC, Sterk PJ. Experimental

rhinovirus 16 infection causes variable airway obstruction in subjects with

atopic asthma. Am J Respir Crit Care Med 1999;160:1375-80.

17. Calhoun WJ, Dick EC, Schwartz LB, Busse WW. A common cold virus,

rhinovirus 16, potentiates airway inflammation after segmental antigen

bronchoprovocation in allergic subjects. J Clin Invest 1994;94:2200-8.

18. Lemanske RF Jr, Dick EC, Swenson CA, Vrtis RF, Busse WW.

Rhinovirus upper respiratory infection increases airway hyperreactivity

and late asthmatic reactions. J Clin Invest 1989;83:1-10.

19. Avila PC, Abisheganaden JA, Wong H, Liu J, Yagi S, Schnurr D, et al.

Effects of allergic inflammation of the nasal mucosa on the severity of

rhinovirus 16 cold. J Allergy Clin Immunol 2000;105:923-32.

20. Gern JE. Rhinovirus respiratory infections and asthma. Am J Med 2002;

112(Suppl 6A):19S-27S.

21. Yamaya M, Sasaki H. Rhinovirus and asthma. Viral Immunol 2003;16:

99-109.

22. Gern JE, Busse WW. Association of rhinovirus infections with asthma.

Clin Microbiol Rev 1999;12:9-18.

23. Hakonarson H, Carter C, Maskeri N, Hodinka R, Grunstein MM. Rhino-

virus-mediated changes in airway smooth muscle responsiveness: induced

autocrine role of interleukin-1beta. Am J Physiol 1999;277(1 Pt 1):13-21.

24. Gern JE. Mechanisms of virus-induced asthma. J Pediatr 2003;142:

(2 Suppl):S9-13.

25. Kaul P, Biagioli MC, Singh I, Turner RB. Rhinovirus-induced oxidative

stress and interleukin-8 elaboration involves p47-phox but is independent

of attachment to intercellular adhesion molecule-1 and viral replication.

J Infect Dis 2000;181:1885-90.

26. Terajima M, Yamaya M, Sekizawa K, Okinaga S, Suzuki T, Yamada N,

et al. Rhinovirus infection of primary cultures of human tracheal epithe-

lium: role of ICAM-1 and IL-1beta. AmJPhysiol 1997;273(4 Pt 1):749-59.

27. de Kluijver J, Grunberg K, Pons D, de Klerk EP, Dick CR, Sterk PJ, et al.

Interleukin-1beta and interleukin-1ra levels in nasal lavages during

experimental rhinovirus infection in asthmatic and non-asthmatic sub-

jects. Clin Exp Allergy 2003;33:1415-8.

28. Jarjour NN, Gern JE, Kelly EA, Swenson CA, Dick CR, Busse WW. The

effect of an experimental rhinovirus 16 infection on bronchial lavage

neutrophils. J Allergy Clin Immunol 2000;105(6 Pt 1):1169-77.

29. Cardell LO, Agusti C, Takeyama K, Stjarne P, Nadel JA. LTB(4)-

induced nasal gland serous cell secretion mediated by neutrophil elastase.

Am J Respir Crit Care Med 1999;160:411-4.

30. Brooks GD, Buchta KA, Swenson CA, Gern JE, Busse WW. Rhino-

virus-induced interferon-gamma and airway responsiveness in asthma.



AUGUST 2005


Asth

madiagnosis

and

treatm

ent

31. Smart JM, Horak E, Kemp AS, Robertson CF, Tang ML. Polyclonal and

allergen-induced cytokine responses in adults with asthma: resolution of

asthma is associated with normalization of IFN-gamma responses.


32. Rakes GP, Arruda E, Ingram JM, Hoover GE, Zambrano JC, Hayden FG,

et al. Rhinovirus and respiratory syncytial virus in wheezing children

requiring emergency care. IgE and eosinophil analyses. Am J Respir Crit

Care Med 1999;159:785-90.

33. Zambrano JC, Carper HT, Rakes GP, Patrie J, Murphy DD, Platts-Mills

TA, et al. Experimental rhinovirus challenges in adults with mild asthma:

response to infection in relation to IgE. J Allergy Clin Immunol 2003;

111:1008-16.




Asthmadiagnosisand

treatm

ent

Perspectives in asthmaGuest editor: William W. Busse, MD

Perspectives on the past decade of asthmagenetics

Carole Ober, PhD Chicago, Ill

Although genetic linkage and association studies have identified

more than 25 asthma or allergy susceptibility loci, replication

of significant results remains a problem. Moreover, these

approaches typically ignore the true complexity of these

diseases, such as the role of gene-by-environment and gene-by-

gene interactions. As a result, many important associations

might have been missed. Recent studies demonstrate not only

that such interactions exist but also that the relationship

between genotype and phenotype is more complex than

previously thought. (J Allergy Clin Immunol 2005;116:274-8.)

Key words: Asthma, allergy, genetics, gene-by-environment inter-

actions

EXTENDING THE MENDELIAN PARADIGMTO COMPLEX DISEASES

The search for genes that influence susceptibility tocommon diseases remains one of the greatest challenges inhuman genetics. With the recent completion of the humangenome project,1 the tools are now available to fully meetthis challenge and to redefine medicine in the 21st century.The ultimate goals of molecular medicine are both toidentify genetically susceptible individuals and intervenebefore the onset of disease and to design drugs that areindividualized and genotype specific. Although there havebeen countless successes with respect to defining themolecular basis of Mendelian (monogenic) diseases,2

genetic studies of common diseases with complex causeshave turned out to be considerably more challenging thanoriginally thought. In this perspective I will provide a briefupdate on the status of genetic studies of asthma andallergy and then discuss some of the insights that have

been gained over the past 10 years on the genetic archi-tecture of these traits.

LINKAGE AND ASSOCIATION STUDIESIDENTIFY SUSCEPTIBILITY GENES

Fig 1 shows a common model of susceptibility toasthma and atopy, which implicates many genes andmanyenvironmental factors but implies that the effects of genesand environmental factors individually contribute to risk.However, the truth is much more complex, with genesinteracting both with other genes and with environmentalrisk factors to confer susceptibility. In fact, few genesmight have independent effects, as is typical for Mende-lian diseases. Nonetheless, the approaches that have beenused to find susceptibility genes, either through linkage orassociation studies, have for the most part considered onegene at a time (Fig 2).

Despite this overly simplistic modeling of asthma andatopy genetics, many important discoveries have beenmade (Fig 3). In particular, 5 genes have been identifiedthrough family linkage studies, followed by positionalcloning.3-7 These genes span a wide range of functionsand in all cases were either unknown or would not havebeen considered as candidate asthma genes before theirdiscovery. Among more than 100 genes that have been

Abbreviations usedAD: Atopic dermatitis

ADRB2: b2-Adrenergic receptor

CD14: Monocyte differentiation antigen 14

COAST: Childhood Onset of Asthma Study

FCERB1: FceRb1GSTM1: Glutathione S-transferase M1

GSTP1: Glutathione S-transferase P1

HLAG: Human leukocyte antigen G

IL4RA: IL-4 receptor, a-chain

LTA: Lymphotoxin a

LTC4A: Leukotriene C4 synthase

NOS3: Nitric oxide synthetase 3

TIM1: T-cell immunoglobulin- and mucin

domain–containing molecule 1

TLR4: Toll-like receptor 4

From the Departments of Human Genetics and Obstetrics and Gynecology,

The University of Chicago.

Supported in part by National Institutes of Health grants HL56399, HL66533,

HL70831, and HL72414.

Disclosure of potential conflict of interest: All authors—none disclosed.

Received for publication April 19, 2005; accepted for publication April 25,

2005.

Available online June 30, 2005.

Reprint requests: Carole Ober, PhD, Department of Human Genetics, The

University of Chicago, 920 East 58th St, CLSC 507C, Chicago, IL

60637-1463. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.039

274

Asth

madiagnosis

and

treatm

ent


interrogated through association studies (reviewed inHoffjan et al8), 8 genes have been replicated in morethan 5 studies, and another 13 genes have been replicatedbut in fewer than 5 studies. These 26 genes are likely to betrue susceptibility loci but represent just the tip of theiceberg because additional positionally cloned genes aresoon to be reported, and many other candidate genes willbe identified and replicated. Thus one could conclude thatthe field of asthma genetics has been quite successful andthat many genes have been identified that contribute torisk.

THE PROBLEM OF REPLICATION

Replicating results remains the gold standard forgenetic association studies, but this has proved difficultfor common diseases, such as asthma, irrespective ofwhether the initial association was identified throughcandidate gene association or positional cloning studies.Even among the most replicated genes, including thoseshown in Fig 3, there are many negative studies (forexamples, see Table 1 in Hoffjan et al8). In fact, there areno genes that are associated with asthma, atopy, or arelated phenotype in every study reported.Moreover, evenwhen a gene is replicated, it is often with a differentphenotype (eg, a polymorphism in intron 1 of the LTAgene is associated with asthma in some studies and IgE inothers), with different polymorphisms in the same gene(eg, the 21112C/T promoter polymorphism in the IL13gene is associated with atopic asthma in some studies, butthe Arg130Gln polymorphism in exon 4 of the same geneis associated with asthma and atopic phenotypes in others),and even with different alleles of the same variant (eg, the2159C allele in the promoter region of the CD14 gene isassociated with atopic phenotypes in some populations,whereas the 2159T allele is associated in others). Thislevel of complexity was unexpected and has suggestedthat models of susceptibility that consider one locus at atime, as is the paradigm for Mendelian diseases, are notadequate for discovering and characterizing asthma andallergy susceptibility loci. Rather, models that includeinteractions between genes and between genes and envi-ronmental risk factors might be required to fully elucidatethe genetic architectures of asthma and atopy.

GENE-BY-ENVIRONMENT INTERACTIONSAND ASTHMA

We and others have recently begun to examine interac-tions between individual genotypes and environmentalexposures as a first step in developing more complexmodels of disease susceptibility. These models considerthe possibility that specific genotypes might result in aphenotype only in certain environments or that a specificgenotype might result in different phenotypes, dependingon environmental exposures. Such interactions couldmaskassociations if the study sample is heterogeneous withrespect to the exposure or underlie discrepant resultsbetween samples drawn from populations that differ withrespect to the exposure. A classic example of a gene-by-environment interaction is that of the Mendelian diseasea1-antitrypsin deficiency. The risk for respiratory diseases,such as emphysema and chronic obstructive pulmonarydisease, among homozygotes for the PiZ null allele (ZZgenotype) is nearly 100% in the presence of cigarettesmoke exposure. In this case the exposure is thought of as atrigger of disease in genetically susceptible individuals.Other examples of gene-by-environment interactions onasthma and atopy risk have been recently reported,7,9-16

and these suggest that such effects might be more the rulethan the exception. These studies are summarized in TableI and in all cases provide examples in which genotype-specific effects are modified by environmental exposures.Although not all of these interaction effects have beenreplicated, they provide the basis for future studies and forcharacterizing the range of effects of important environ-mental exposures as modifiers of disease risk.

FIG 1. Common model of the genetics of complex diseases.

Several related disease and quantitative phenotypes result from

the effects of many loci and many environmental factors. BHR,

Bronchial hyperresponsiveness.

FIG 2. Strategies for identifying disease genes. A, With the candi-

date gene approach, typically used in association studies, a gene is

selected on the basis of its known function (ie, functional candidate

gene). Variation in that gene is then examined for associations in

patients and control subjects, cohorts of individuals, or families.

Ultimately, the mechanism of the association is revealed through

genotype-phenotype studies and functional studies of the associ-

ated variant. B, In positional cloning studies initially only informa-

tion on the chromosomal location is known, usually from family

linkage studies. All genes in the linked region become positional

candidates, and association studies are performed as described

above to identify the associated gene and variation that contributes

to disease risk. Once the variation is identified by means of

association studies, the mechanism of the association is studied

as described above.



Ober 275

Asthmadiagnosisand

treatm

ent

THE IMPORTANCE OF EARLY-LIFEEXPOSURES

Epidemiologic studies have identified many environ-mental factors that influence risk for asthma and allergicdisease, such as maternal asthma, birth order, and sibshipsize and early-life exposure to viral infections, endotoxin(LPS), day care, pets, and allergens. Yet few studies todate have examined how exposure to environmental riskfactors during development modifies genotype-specificrisks for asthma and allergic disease. The ChildhoodOnset of Asthma (COAST) Study is a prospective birthcohort study of high-risk children designed to evaluatethe role of genes and environment on the development ofimmune responsiveness and allergic phenotypes.17 Thefirst studies in this cohort to examine gene-by-environ-

ment interactions were recently reported, providing someintriguing examples of interactions. A study of the

effects of dog ownership on the development of immune

responsiveness and atopy in infancy revealed a protec-

tive effect of having a dog in the house at the time the

child was born: only 30% of infants had atopic derma-

titis (AD) if a pet was present in the home compared

with 51% of infants in homes without a dog (P <.0001).14 However, this difference was even more strik-

ing among children with the 2159TT genotype at the

locus encoding the receptor for LPS, CD14: only 5% of

TT children exposed to a dog had AD compared with

43% of unexposed TT children (P =.04). In this exam-

ple, even though both the polymorphism (CD14 2159C/

T) and the environmental exposure (dog) independently

influenced risk for AD, the interaction between the 2 was

FIG 3. Approximate locations of asthma and atopy genes on human chromosomes. Five genes identified

through linkage followed by positional cloning studies are shown in red. Twenty-one genes that were

identified through association studies and replicated in subsequent studies are also shown (summarized from

Hoffjan et al8). Eight genes that have been replicated in more than 5 studies are shown in blue, and 13 genes

that have been replicated but in fewer than 5 studies are shown in black.


AUGUST 2005

276 Ober

Asth

madiagnosis

and

treatm

ent

significant (P = .0071), indicating that the risk associatedwith the TT genotype differs in different exposuregroups. Interactions between CD14 genotype and levelsof endotoxin exposure have been suggested as anexplanation for the discrepant results of associationstudies with this polymorphism,18 as discussed earlier,and these data support that hypothesis.

In a second study in this cohort, the effects of day-careattendance in the first 6 months of life on cytokineresponse profiles and allergic phenotypes were exam-ined.16 Seventy-two polymorphisms in 35 genes wereselected because of their putative role in immune re-

sponses or asthma and genotyped in 99 COAST childrenwho attended day care and 109 COAST children who didnot. Interestingly, neither day-care attendance nor geno-type at these loci by themselves significantly influencedany of the first-year phenotypes examined. However,highly significant interaction effects (P < .001) weredemonstrated with genotypes at 3 loci: NOS3, FCERB1,and IL4RA. In each case the effects of a particulargenotype on the phenotype were opposite depending onwhether the child attended day care (ie, the same genotypewas associated with the highest cytokine responses orprotection from disease among children attending day care

TABLE I. Examples of gene-by-environment interaction effects on asthma and atopic disease

Gene

Environmental

Exposure Phenotype Comment Reference

LTC4S Aspirin exposure Asthma 2444C allele is increased among individuals

with aspirin-induced asthma compared

with individuals with aspirin-tolerant

asthma

Sanak et al9

ADRB2 Cigarette smoke Asthma Increased risk of asthma among smokers

with Arg16 genotype but not among

nonsmokers

Wang et al10

ADRB2 Physical activity Asthma Increased risk of asthma among sedentary

women with Gly16 genotype but not

among active women

Barr et al11

TIM1 HAV Atopy HAV protects against atopy in individuals

with a 6-amino-acid insertion at residue

157 (157insMTTTVP) but not in

individuals without the insertion

McIntire et al12

TLR4 Endotoxin levels Asthma At high levels of endotoxin exposure, carriers

of the Gly299 and Ile399 alleles have

reduced risk for asthma compared with

other genotypes and other exposure groups

Werner et al13

CD14 Dog ownership at birth AD 2159TT genotype is protective against AD

in the first year among children with a dog

in the home at birth

Gern et al14

GSTM1 Diesel exhaust particles IgE and histamine response Enhanced responses among GSTM1-null

individuals but not among individuals with

other genotypes

Gilliland et al15

GSTP1 Diesel exhaust particles IgE and histamine response Enhanced responses among individuals with

the Ile105 allele but not among individuals

without this allele

Gilliland15

NOS3 Day-care exposure in

the first 6 mo of life

Change in TH2 cytokine

(IL-5 and IL-13) response

in first year of life

Asp298 homozygosity associated with

smallest changes in TH2 responses among

children attending day care and largest

changes among children not attending day

care

Hoffjan et al16

FCERB1 Day-care exposure in

the first 6 mo of life

IL-5 response at 1 y of age Gly237 allele associated with decreased IL-5

responsiveness among children attending

day care and increased responsiveness

among children not attending day care

Hoffjan et al16

IL4RA Day-care exposure in the

first 6 mo of life

IFN- g response at 1 y of age Val50 homozygosity associated with lowest

response among children attending day

care and highest response among children

not attending day care

Hoffjan et al16

HLAG Maternal BHR Asthma-BHR in child 2964G allele is associated with asthma in

children of mothers with BHR; 2964A

allele is associated with atopy and asthma

among children of mothers without BHR

Nicolae et al7

HAV, Hepatitis A; BHR, bronchial hyperresponsiveness.



Ober 277

Asthmadiagnosisand

treatm

ent

but the lowest cytokine responses or risk for diseaseamong children not attending day care). That is, thegenotype effects at these loci were modified by theenvironment such that the same genotype was associatedwith protection from or risk for a phenotype depending onthis early-life exposure! In the pooled sample (not strat-ified by day-care attendance) there were no detectabledifferences between genotypes (summarized in Table I).Interestingly, the interaction effects with the FCERB1 andIL4RA genes were likely accounted for by the increasednumber of viral infections among children attending daycare; however, the interaction effects with NOS3 wereindependent of viral infections, suggesting that risk factorsother than viruses but that are correlated with day-careexposure interact with the NOS3 genotype to determinerisk. Complex interactions such as these could underliesome of the association studies in which one allele of apolymorphism is associated in some populations and theother allele of the same polymorphism is associated withthe same phenotype in others.

CONCLUDING REMARKS

The mechanisms underlying these interactions are notyet known. Nonetheless, these studies and others arebeginning to reveal the true complexities of the genetics ofasthma and allergy. The next phase of genetic investiga-tion should continue to unravel the nature and overallimportance of gene-by-environment and gene-by-geneinteractions on the development of asthma and allergicphenotypes on disease progression and severity and onthe response to therapeutic interventions. Thus the next10 years of asthma genetic research will begin to meetthe goals of the new molecular medicine.

I thank Dr Nancy Cox and Dr Robert Lemanske for helpful

discussions.

REFERENCES

1. International Human Genome Sequencing Consortium. Finishing the

euchromatic sequence of the human genome. Nature 2004;431:931-45.

2. Glazier AM, Nadeau JH, Aitman TJ. Finding genes that underlie

complex traits. Science 2002;298:2345-9.

3. Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J,

et al. Association of the ADAM33 gene with asthma and bronchial

hyperresponsiveness. Nature 2002;418:426-30.

4. Zhang Y, Leaves NI, Anderson GG, Ponting CP, Broxholme J, Holt R,

et al. Positional cloning of a quantitative trait locus on chromosome

13q14 that influences immunoglobulin E levels and asthma. Nat Genet

2003;34:181-6.

5. Allen M, Heinzmann A, Noguchi E, Abecasis G, Broxholme J, Ponting

CP, et al. Positional cloning of a novel gene influencing asthma from

chromosome 2q14. Nat Genet 2003;35:258-63.

6. Laitinen T, Polvi A, Rydman P, Vendelin J, Pulkkinen V, Salmikangas

P, et al. Characterization of a common susceptibility locus for asthma-

related traits. Science 2004;304:300-4.

7. Nicolae D, Cox NJ, Lester LA, Schneider D, Tan Z, Billstrand C, et al.

Fine mapping and positional candidate studies identify HLA-G as an

asthma susceptibility gene on chromosome 6p21. Am J Hum Genet

2005;76:349-57.

8. Hoffjan S, Nicolae D, Ober C. Association studies for asthma and atopic

diseases: a comprehensive reviewof the literature.RespirRes2003;4:14-28.

9. Sanak M, Pierzchalska M, Bazan-Socha S, Szczeklik A. Enhanced

expression of the leukotriene C(4) synthase due to overactive transcrip-

tion of an allelic variant associated with aspirin-intolerant asthma. Am

J Respir Cell Mol Biol 2000;23:290-6.

10. Wang Z, Chen C, Niu T, Wu D, Yang J, Wang B, et al. Association of

asthma with beta(2)-adrenergic receptor gene polymorphism and ciga-

rette smoking. Am J Respir Crit Care Med 2001;163:1404-9.

11. Barr RG, Cooper DM, Speizer FE, Drazen JM, Camargo CA Jr. Beta(2)-

adrenoceptor polymorphism and body mass index are associated with

adult-onset asthma in sedentary but not active women. Chest 2001;120:

1474-9.

12. McIntire JJ, Umetsu SE, Macaubas C, Hoyte EG, Cinnioglu C, Cavalli-

Sforza LL, et al. Immunology: hepatitis A virus link to atopic disease.

Nature 2003;425:576.

13. Werner M, Topp R, Wimmer K, Richter K, Bischof W, Wjst M, et al.

TLR4 gene variants modify endotoxin effects on asthma. J Allergy Clin

Immunol 2003;112:323-30.

14. Gern JE, Reardon CL, Hoffjan S, Nicolae D, Li Z, Roberg KA, et al.

Effects of dog ownership and genotype on immune development and

atopy in infancy. J Allergy Clin Immunol 2004;113:307-14.

15. Gilliland FD, Li YF, Saxon A, Diaz-Sanchez D. Effect of glutathione-

S-transferase M1 and P1 genotypes on xenobiotic enhancement of allergic

responses: randomised, placebo-controlled crossover study. Lancet 2004;

363:119-25.

16. Hoffjan S, Nicolae D, Ostrovnaya I, Roberg K, Evans M, Mirel DB, et al.

Gene-environment interaction effects on the development of immune

responses in the 1st year of life. Am J Hum Genet 2005;76:696-704.

17. Lemanske RF. The childhood origins of asthma (COAST) study. Pediatr

Allergy Immunol 2002;15:1-6.

18. Vercelli D. Learning from discrepancies: CD14 polymorphisms, atopy

and the endotoxin switch. Clin Exp Allergy 2003;33:153-5.


AUGUST 2005

278 Ober

Asth

madiagnosis

and

treatm

ent

Original articles

Is it traffic type, volume, or distance? Wheezingin infants living near truck and bus traffic

Patrick H. Ryan, MS,a Grace LeMasters, PhD,a Jocelyn Biagini, MS,a

David Bernstein, MD,b Sergey A. Grinshpun, PhD,a Rakesh Shukla, PhD,a

Kimberly Wilson, MS,a Manuel Villareal, MD,b Jeff Burkle, BS,a and James Lockey, MDa

Cincinnati, Ohio

Background: Previous studies of air pollution have not

examined the association between exposure to varying types,

distance, and amounts of traffic and wheezing in very young

infants.

Objective: We sought to determine the relationship between

types of traffic, traffic volume, and distance and wheezing

among infants less than 1 year of age.

Methods: A geographic information system and a classification

scheme were developed to categorize infants enrolled in the

study as living near moving truck and bus traffic (highway >50

miles per hour, >1000 trucks daily, <400 m), stop-and-go truck

and bus traffic (<50 miles per hour, <100 m), or unexposed and

not residing near either. Symptom data were based on health

questionnaires administered to parents when the infants were

6 months of age and monthly health diaries.

Results: Infants living very near (<100 m) stop-and-go bus

and truck traffic had a significantly increased prevalence of

wheezing (adjusted odds ratio, 2.50; 95% CI, 1.15-5.42) when

compared with unexposed infants. The prevalence of wheezing

among nonwhite infants was at least twice that of white infants,

regardless of exposure. Infants living less than 400 m from a

high volume of moving traffic, however, did not have an

increased prevalence of wheezing.

Conclusion: These results suggest that the distance from and

type of traffic exposures are more significant risk factors than

traffic volume for wheezing in early infancy. (J Allergy Clin

Immunol 2005;116:279-84.)

Key words: Diesel, traffic, truck, bus, wheezing, Geographic Infor-

mation System, infants

Recent articles in this journal have reviewed theepidemiology and biology of air pollution and dieselexhaust and their effects on asthma risk and respiratoryfunction.1,2 Diesel exhaust particles (DEPs) have beenstudied with respect to both the development and exacer-bation of allergic rhinitis and asthma because of theirunique capability to enhance those TH cells that directallergic immune responses (ie, TH2 cells) and productionof IgE.3 Although the mechanism by which DEPs mightinfluence allergy and asthma development and exacerba-tion is still under investigation, the immunologic effects ofDEPs have been recently reviewed.4,5 DEPs are respirableparticles with a large surface area per unit mass thatprovide an excellent medium for absorbing and trans-porting proteins into the peripheral airways.6 Studies havedemonstrated that DEPs are capable of binding with grasspollen allergen (Lol p 1), and this might be similar withother aeroallergens.7,8 In human studies exposure to DEPshas been shown to enhance allergic nasal cytokine andinflammatory responses after direct challenge with aller-gen extracts.9-11

Previous studies of traffic pollutants have focused onroadways with high truck and automobile traffic andminimal bus traffic as the source of air pollution, and thesestudies have been conducted primarily on school-agechildren. The purpose of this study was to determinewhether distance, volume, and/or type of traffic might beassociated with wheezing in infants younger than 1 year.The hypothesis was that infants who reside near majorhighways with heavy truck traffic, as well as infants whoreside near local roads with stop-an-go truck and bustraffic will have a significantly increased risk of wheezing

Abbreviations usedCCAAPS: Cincinnati Childhood Allergy and Air

Pollution Study

DEP: Diesel exhaust particle

GIS: Geographic information system

mph: Miles per hour

OR: Odds ratio

PM: Particulate matter

SPT: Skin prick test

From athe Department of Environmental Health, and bthe Department of

Internal Medicine, Division of Immunology, University of Cincinnati.

Supported by grants ES11170 and ES10957 from the National Institute of

Environmental Health Sciences.

Disclosure of potential conflict of interest: S. A. Grinshpun has received

grants–research support from theNational Institute of Environmental Health

Sciences. M. Villareal has consultant arrangements with Aventis, has stock

or other equity ownership with Pfizer, and is on the speakers’ bureaus for

AstraZeneca, UCB Pharma, Pfizer, Aventis, and GlaxoSmithKline. There

are not other potential conflicts to disclose.

Received for publication March 15, 2005; revised May 9, 2005; accepted for

publication May 10, 2005.


Reprint requests: Patrick H. Ryan, MS, Department of Environmental Health,

University of Cincinnati, Cincinnati, OH 45267-0056. E-mail: ryanph@

email.uc.edu.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.014

279

Asthmadiagnosisand

treatm

ent



when compared with infants residing far from truck andbus traffic.

METHODS

Subject recruitment

The Cincinnati Childhood Allergy and Air Pollution Study

(CCAAPS) is an ongoing birth cohort study. Infants enrolled in

CCAAPS were identified from birth records, and the addresses

obtained from these records were geocoded with EZLocate from

TeleAtlas for the ArcViewGeographic Information System (GIS) 3.2

(Environmental Systems Research Institute, Redlands, Calif). The

distance to the nearest major highway or interstate (defined as >1000trucks daily) was computed for all infants by using the Geoprocessing

extension. Infants whose birth records indicated residency less than

400 m or greater than 1500 m from the nearest major highway or

interstate were eligible.

Parents were recruited when infants were 6 months of age or older

and screened for allergy symptoms.12 Parents with likely atopy were

subsequently tested with skin prick tests (SPTs) with a panel of 15

common indoor and outdoor regional aeroallergens. Infants with at

least one atopic parent were enrolled.

Outcome variables

At the time of the parent SPT, an interviewer administered the

baseline health questionnaire at a physician’s office. This question-

naire gathered demographic information, occupants in the home and

their smoking status, animal ownership, and information on other

possible risk factors. The general health of the infant was queried

from birth until the infants’ age at enrollment. These questions were

based on the well-validated International Study of Allergy and

Asthma in Children questionnaire for children ages 4 to 5 years,

which was adapted for use with infants.13 In addition, monthly diaries

were distributed to the parents of all enrolled infants at the time of the

parental SPT. These diaries recorded parental observation of the

infants’ illnesses and were returned by mail monthly until the child’s

first visit at age 1 year. Wheeze with a cold and wheezing without

a cold were assessed on both questionnaires. Wheezing without

symptoms of a cold was the outcome variable for this study. To

increase the reliability of parental report of wheezing, an infant whose

parent reported wheezing (without a cold) on the parent questionnaire

and returned at least one monthly diary indicating the identical

symptom was designated to have wheezed.

Traffic exposure classification

ArcView shapefiles containing the location of all state roads,

interstates, and traffic counts (both truck and car) were obtained

from the Ohio Department of Transportation and the Kentucky

Transportation Cabinet. Shapefiles containing the location of public

transportation (bus) routes and bus counts for the city of Cincinnati

and Northern Kentucky were obtained from the Cincinnati Area

Geographic Information Systems database, the Northern Kentucky

Area Planning Commission, the Southwest Ohio Regional Transit

Authority, and the Transit Authority of Northern Kentucky. The

distance to the nearest federal interstate, state route, and bus route

from the primary residence of the infant at the time of parent

enrollment was derived by using the Geoprocessing extension for

ArcView GIS 3.2. A traffic exposure classification scheme was

subsequently applied to each infant by using the distance to the 3

types of traffic, the speed limit on the roadway closest to the infant,

and the amount of DEPs producing traffic (trucks or buses) on the

road type nearest each infant.

Classification of exposure by distance was based on the methods

of others, with distances of less than 100 m,14,15 150 m,16,17 200 m,18

and 400 m.19 Fewer than 3% (n = 10) of residences of infants in this

study, however, were within 100 m of an interstate; fewer than 10%

(n = 88) were within 200 m of an interstate, whereas 28% (n = 248)

of the population resided within 400 m of an interstate. Also,

approximately 26% (n = 174) of the infants resided within 100 m of

a state route or a bus route. Hence on the basis of our population’s

geographic distribution and our pilot study, which revealed that the

concentration of ultrafine particles decreased by one half between

50 m and 150 m downwind from a highway and an observable sulfur

concentration gradient up to 400 m from a highway, infants were

classified as exposed to interstate traffic if their residence was within

400 m.20 Exposure to a state or bus route was determined if their

residence was within 100 m from one of these routes. If an infant

resided greater than 400 m from the nearest interstate, greater than

100 m from the nearest state route, and greater than 100 m from the

nearest bus route, the infant was placed in the unexposed category.

Furthermore, exposure tomoving traffic was determined if an infant’s

residence was within 400 m of an interstate or within 100 m of a state

route with a speed limit of greater than or equal to 50 miles per hour

(mph). Fifty miles per hour was used as the cutoff because in Ohio

this is the designation used for classification of an urban (greater

traffic) or rural (less traffic) route. Exposure to stop-and-go traffic was

determined if an infant’s residence was within 100 m of a bus route,

within 100 m of a state route with a speed limit of less than 50 mph,

or both.

Statistical analyses

To determine the presence of an association between traffic

exposure and wheezing, conditional logistic regression was per-

formed with SAS software (version 8.2 for Windows; SAS Institute

Inc, Cary, NC), adjusting for sex, race (white/nonwhite), breast-

feeding (maternal report of breast-feeding<1 week, 1-4 weeks, or>5

weeks), pet ownership, income (<$40,000/$40,000), child care

outside of the home (parent report of infant attending day care or

babysitter), number of siblings, visible mold in the home, maternal

and paternal self-report of asthma, and the number of monthly diaries

returned.

RESULTS

Recruitment for the CCAAPS study was completed onDecember 13, 2003. During year 1 of the study, 633families had returned at least one monthly diary beforeJanuary 1, 2004. Eleven (1.7%) of the 633 eligible familieswere excluded because of inaccuracy in the geocoding oftheir residence at the time of exposure classification. Theaverage age of the infants in this study (at the time of theirenrollment) was 7.5 months (6 2.4 months).

Exposure classification

Table I displays the demographic characteristics of theinfants and their families in the 3 exposure groups; 60.1%(n = 374) of the infants were unexposed, whereas themoving traffic category included 28.3% (n = 176) of theinfants, and the stop-and-go category included 15.9%(n = 99) of the infants. As shown in Table I, the infantsexposed to stop-and-go traffic were more likely to beAfrican American, to have care outside their home, and tohave had a father with asthma, whereas they were lesslikely to have been breast-fed. Because of these differ-ences, these and other possible covariates were adjusted


AUGUST 2005

280 Ryan et al

Asth

madiagnosis

and

treatm

ent

for in the logistic model. In the unexposed category themedian distances to the nearest highway, state route, andbus route were 3287 m, 743 m, and 543 m, respectively.For those infants classified as exposed to moving traffic,the median distances to the nearest highway, state route,and bus route were 252m, 696m, and 341m, respectively.Infants exposed to stop-and-go traffic resided a mediandistance of 2303 m, 439 m, and 43 m from the nearesthighway, state route, and bus route, respectively. Themedian number of trucks on the highway nearest infants(n = 170) in the moving category was 11,820 daily. Forthose infants (n = 6) exposed to moving traffic on a stateroute, the median number of trucks per day was 1050.Infants exposed to buses (n = 71) or trucks (n = 9) only inthe stop-and-go category had a median of 44 buses dailyand 1250 trucks, respectively, on the route nearest theirresidence. Infants exposed to both buses and trucks(n = 19) had a median of 72 and 390, respectively.

Wheeze (without cold)

Of the 622 infants, 50 (8.0%) reported wheezingwithout a cold. In the unexposed category 5.8% of infantsreported wheezing without a cold compared with 7.4%in the moving category and 17.2% in the stop-and-goexposure category (P < .01). The prevalence of wheezingin the infants who were categorized into the 3 exposurecategories was subsequently examined by distance fromthe nearest road and type of traffic (Fig 1). The prevalenceof wheezing was 3 times higher (19%) in the infants whoresided less than 50 m from stop-and-go traffic comparedwith those infants who were unexposed (6%). The prev-alence of wheezing in infants who reside 200 to 300 m

from moving traffic (12%) was more than doubled whencompared with that of infants who were classified asunexposed.

Unadjusted and adjusted odds ratios (ORs) are shown inTable II. Living within 100m of stop-and-go truck and bustraffic was the most important risk factor for early infantwheeze (adjusted OR, 2.50; 95%CI, 1.15-5.42). An infantwith no siblings was at a decreased risk for wheezing(adjusted OR, 0.42; 95% CI, 0.19-0.93), and nonwhiteinfants were at an increased risk for wheezing (adjustedOR, 2.39; 95% CI, 1.20-4.76, respectively). Male sex andpaternal self-report of asthma (although not maternal self-report of asthma) were also significantly associated withwheezing (Table II). Infants classified as exposed tomoving traffic did not have a significant association withwheezing without a cold when compared with infantsclassified as unexposed. A univariate analysis was con-ducted comparing wheezing without a cold and the season(winter [January-March], spring [April-June], summer[July-September], autumn [October-December]) in whichit was first reported to address the possibility of a cold as apossible cause of wheezing. In this analysis there were nodifferences in the prevalence of wheezing among season(P = .50), and the same was true when season was added tothe multivariate model.

DISCUSSION

Infants exposed to stop-and-go bus and truck traffic hada significantly increased risk for wheezing without a coldcompared with infants unexposed to truck or bus traffic orcompared with infants exposed to moving truck trafficwith a larger volume of trucks. Infants with immaturelungs residing in close proximity to stop-and-go truck andbus traffic might be exposed to greater amounts of fine andultrafine particulates.21-24 Sampling for fine particulatematter (PM 2.5 mm) and black carbon inside a bus and acar traveling ahead of the bus showed that the averageDEP levels were approximately 20 mg/m3 and 5 mg/m3,

TABLE I. Prevalence of demographic characteristics by

exposure categorization

Characteristic

Unexposed

(n = 347), %

Moving

(n = 176), %

Stop and go

(n = 99), %

White 83.3 78.7 56.6

Income <$40,000 24.6 35.7 54.2

Male sex 51.3 50.6 61.2

Current smoking mother 11.8 16.5 17.2

Care outside home 28.5 24.3 36.2

Owns dog 33.9 39.0 26.9

Owns cat 27.8 24.4 16.1

Weeks breast-fed

0 26.3 35.2 49.5

1-4 6.9 9.7 14.1

5 66.8 55.1 36.4

Diaries returned

1 27.7 29.6 43.4

2 18.4 11.9 13.1

3 53.9 58.5 43.5

No. of siblings

0 37.5 34.3 36.2

1 33.0 40.7 34.0

2 29.5 25.0 29.8

Has visible mold 65.4 58.0 58.6

Paternal asthma 11.1 13.0 23.6

Maternal asthma 20.6 25.8 17.4

FIG 1. Prevalence of wheeze (without cold) among infants within

exposure categories by distance from the nearest DEP source.

SR, State route; BR, bus route; HW, highway.



Ryan et al 281

Asthmadiagnosisand

treatm

ent

but during stop-and-go traffic, the levels increased to morethan 30mg/m3 and 20mg/m3, respectively.21 Other studieshave also found acceleration, deceleration, and stop-and-go traffic to be associated with higher emissions of organiccarbon, elemental carbon, carbon monoxide, nitric oxide,hydrocarbons, and soot when compared with cruisingtraffic.22-24

To our knowledge, the present study is the first toprospectively examine the effect of living in close prox-imity to roads with stop-and-go bus and truck traffic oninfants’ respiratory health. Although our results are con-sistent with those of other investigations,14,15,18,25-28 thisstudy improves on previous investigations. This prospec-tive design during early infancy minimizes parental recallbias by allowing simultaneous measurements of exposureand outcome, whereas most previous studies have reliedon parental recall of infant illness. The GIS database alsoaccurately geocoded the infant’s residence, as well as thedistance from the nearest traffic source. With the integra-tion of county traffic data, the GIS minimized responsebias by the parents who might inaccurately report thefrequency or proximity of traffic.

Although our a priori hypothesis also expected to findan effect with exposure to moving traffic, no associationwas found. Hence high intermittent exposures to pollu-tants might have a greater detrimental health effect oninfants than exposure to lower continuous exposure.These findings, however, could be related to study limi-tations. Infants living near moving traffic were exposed towide variations in the number of trucks. In addition, themedian distance to stop-and-go traffic was 43 m, whereasthe median distance to moving traffic was 252 m, andalthough state routes might have posted speed limits ofgreater than 50 mph and be classified as a rural route, thepossibility exists that traffic might accelerate and deceler-ate at times of congestion. This scenario is also likely forshort periods on highways where we have designated thetraffic as moving.

PM, a primary constituent of DEPs, has been signifi-cantly associated with emergency department visits forasthma, wheezing bronchitis, lower respiratory tract

symptoms, and physician visits for asthma.29-34 Othershave found that fine and ultrafine particles (PM2.5 andPM1, respectively) have a greater association with respi-ratory symptoms than coarse particles (PM >2.5 mm) andare associated with pulmonary retention of particles.3,35,36

Induction of oxidative stress and mitochondrial damageby ultrafine particulates has been proposed.2,37 Thuswhether the mechanism is total load or oxidative stress,infants who reside less than 100 m away are likelyreceiving a high dose of particulates. It is not possible,however, to separate the contributions of diesel andgasoline engines.

Previous studies have found maternal asthma,38 pater-nal asthma,39 or both to be a significant risk factor for thedevelopment of asthma and wheezing in children.40,41 Inour study only paternal asthma was significant. Althoughstudies have found associations between parental smokingand wheezing in infants, the multivariate model showedno additional smoking effect. Also, multivariate analysesof maternal smoking found no associations with wheezingin the unexposed subpopulation. However, only 14%(n = 87) of the cohort were exposed to maternal smoking(Table I). Of particular interest are the findings regard-ing the high prevalence of wheezing among nonwhiteinfants in all exposure categories. As shown in Fig 2, theprevalence of wheezing in nonwhite infants was nearly 2or more times higher in all groups, suggesting a healthdisparity beginning early in infancy. Thus increasedsusceptibility to wheezing is consistent with nationalstatistics of asthma prevalence, as well as other stud-ies.42-44 Because wheezing in the first year of life isgenerally a poor predictor of later development of child-hood asthma, results must be interpreted cautiously.45

In conclusion, this is the first epidemiologic study toexamine the risk of wheezing in infants younger than1 year who are exposed to varying types and amounts ofurban traffic. It demonstrated that even within an urbanenvironment, the risk of wheezing varies with the type ofand distance from traffic. A health disparity between racesis also evident: nearly one fourth of African Americaninfants exposed to stop-and-go traffic are wheezing before

TABLE II. Unadjusted and adjusted* ORs and 95% CIs

for wheezing without a cold

Unadjusted OR Adjusted OR

OR 95% CI OR 95% CI

Stop-and-go exposure 3.41 1.71-6.81 2.50 1.15-5.42

Nonwhite 2.80 1.54-5.08 2.39 1.20-4.76

Paternal asthma 2.63 1.29-5.37 2.35 1.08-5.13

Male sex 1.93 1.04-3.58 2.49 1.21-5.10

Only child 0.47 0.23-0.96 0.42 0.19-0.93

*Adjusted for maternal smoking, breast-feeding, pet ownership, visible

mold, maternal asthma, care outside the home, and monthly diaries

returned.

Reference category = unexposed.

Reference category = 2 or more siblings.

FIG 2. Prevalence of wheeze (without cold) among infants stratified

by race.


AUGUST 2005

282 Ryan et al

Asth

madiagnosis

and

treatm

ent

age 1 year. This risk will be further investigated as thecohort ages and asthma and atopy can be diagnosed.

We thank Stephanie Maier and Sherry Stanforth for their help with

family interviews and testing.

REFERENCES

1. Peden DB. The epidemiology and genetics of asthma risk associated with

air pollution. J Allergy Clin Immunol 2005;115:213-9.

2. Riedl M, Diaz-Sanchez D. Biology of diesel exhaust effects on respi-

ratory function. J Allergy Clin Immunol 2005;115:221-8.

3. Peden DB. Development of atopy and asthma: candidate environmental

influences and important periods of exposure. Environ Health Perspect

2000;108:475-82.

4. Pandya RJ, Solomon G, Kinner A, Balmes JR. Diesel exhaust and

asthma: hypotheses and molecular mechanisms of action. Environ Health

Perspect 2002;110:103-12.

5. Diaz-Sanchez D, Tsien A, Casillas A, Dotson AR, Saxon A. Enhanced

nasal cytokine production in human beings after in vivo challenge with

diesel exhaust particles. J Allergy Clin Immunol 1996;98:114-23.

6. US Environmental Protection Agency (EPA). Health assessment docu-

ment for diesel engine exhaust. Prepared by the National Center for

Environmental Assessment, Washington, DC, for the Office of Trans-

portation and Air Quality; EPA/600/8–90/057F. 2002. Available at:

http://www.epa.gov/ncea.

7. Knox RB, Suphioglu C, Taylor P, Desai R, Watson HC, Peng JL, et al.

Major grass pollen allergen Lol p 1 binds to diesel exhaust particles:

implications for asthma and air pollution. Clin Exp Allergy 1997;27:246-51.

8. Ormstad H. Suspended particulate matter in indoor air: adjuvants and

allergen carriers. Toxicology 2000;152:53-68.

9. Diaz-Sanchez D, Tsien A, Fleming J, Saxon A. Combined diesel exhaust

particulate and ragweed allergen challenge enhances human in vivo nasal

ragweed-specific IgE and skews cytokine production to a T helper cell 2

type pattern. J Immunol 1997;158:2406-13.

10. Fujieda S, Diaz-Sanchez D, Saxon A. Combined nasal challenge with

diesel exhaust particles and allergen induces in vivo IgE isotype

switching. Am J Respir Cell Mol Biol 1998;19:507-12.

11. Fujimaki H, Nohara O, Ichinose T, Watanabe N, Saito S. IL-4 production

in mediastinal lymph node cells in mice intratracheally instilled with

diesel exhaust particulates and antigen. Toxicology 1994;92:261-8.

12. LeMasters GK, Wilson KA, Levin L, Bernstein DI, Lockey JE, Villareal

M, et al. Validation of a population-based brief allergy symptoms

questionnaire. Abstract presented at: American Thoracic Society Inter-

national Conference; May 16-21, 2003; Seattle (WA).

13. Jenkins MA, Clarke JR, Carlin JB, Robertson CF, Hopper JL, Dalton MF,

et al. Validation of questionnaire and bronchial hyperresponsiveness

against respiratory physician assessment in the diagnosis of asthma. Int

J Epidemiol 1996;25:609-16.

14. Brunekreef B, Janssen NAH, de Hartog J, Harssema H, Knape M

van Vliet P. Air pollution from truck traffic and lung function in chil-

dren living near motorways. Epidemiology 1997;8:298-303.

15. van Vliet P, KnapeM, de Hartog J, Janssen N, Harssema H, Brunekreef B.

Motor vehicle exhaust and chronic respiratory symptoms in children living

near freeways. Environ Res 1997;74:122-32.

16. Venn AJ, Lewis SA, Cooper M, Hubbard R, Britton J. Living near a

main road and the risk of wheezing illness in children. Am J Respir Crit

Care Med 2001;164:2177-80.

17. Wilkinson P, Elliott P, Grundy C, Shaddick G, Thakrar B, Walls P, et al.

Case-control study of hospital admission with asthma in children aged

5-14 years: relation with road traffic in north west London. Thorax 1999;

54:1070-4.

18. Lin S, Munsie JP, Hwang S, Fitzgerald E, Cayo MR. Childhood asthma:

hospitalization and residential exposure to state route traffic. Environ Res

2002;88:73-81.

19. Janssen NAH, Brunekreef B, van Vliet P, Aarts F, Meliefste K, Harssema

H, et al. The relationship between air pollution from heavy traffic and

allergic sensitization, bronchial hyperresponsiveness, and respiratory

symptoms in Dutch schoolchildren. Environ Health Perspect 2003;111:

1512-8.

20. Reponen T, Grinshpun SA, Trakumas S, Martuzevicius D, Wang Z,

LeMasters G, et al. Concentration gradient patterns of aerosol particles

near interstate highways in the Greater Cincinnati airshed. J Environ

Monit 2003;5:557-62.

21. Solomon G, Campbell T, Feuer G, Masters J, Samkian A, Paul K.

No breathing in the aisles: Diesel exhaust inside school buses. Natural

Resources Defense Council, Coalition for Clean Air. 2001. Available at:

http://www.nrdc.org/air/transportation/schoolbus/sbusinx.asp.

22. Shah S, Cocker D, Miller JW, Norbeck JM. Emission rates of particulate

matter and elemental and organic carbon from in-use diesel engines.

Environ Sci Technol 2004;38:2544-50.

23. Tong HY, Hung WT, Cheung CS. On-road vehicle emissions and fuel

consumption in urban driving conditions. J Air Waste Manag Assoc

2000;50:543-54.

24. Unal A, Frey HC, Rouphail NM. Quantification of highway vehicle

emissions hot spots based upon on-board measurements. J Air Waste

Manag Assoc 2004;54:130-40.

25. Hirsch T, Weiland SK, von Mutius E, Safeca AF, Grafe H, Csaplovics E,

et al. Inner city air pollution and respiratory health and atopy in children.

Eur Respir J 1999;14:669-77.

26. Ciccone G, Forastiere F, Agabiti N, Biggeri A, Bisanti L, Chellini E,

et al. Road traffic and adverse respiratory effects in children. SIDRIA

Collaborative Group. Occup Environ Med 1998;55:771-8.

27. Shima M, Nitta Y, Adachi M. Traffic related air pollution and respiratory

symptoms in children living along trunk roads in Chiba Prefecture,

Japan. J Epidemiol 2003;13:108-19.

28. Brauer M, Hoek G, Van Vliet P, Meliefste K, Fischer PH, Wijga A, et al.

Air pollution from traffic and the development of respiratory infections

and asthmatic and allergic symptoms in children. Am J Respir Crit Care

Med 2002;166:1092-8.

29. Norris G, YoungPong S, Koenig J, Larson T, Sheppard L, Stout J. An

association between fine particles and asthma emergency department

visits for children in Seattle. Environ Health Perspect 1999;110:489-93.

30. Pino P, Walter T, Oyarzun M, Villegas R, Romieu I. Fine particulate

matter and wheezing illnesses in the first year of life. Epidemiology

2004;15:702-8.

31. Shima M, Nitta Y, Ando M, Adachi M. Effects of air pollution on the

prevalence and incidence of asthma in children. Arch Environ Health

2002;57:529-35.

32. van der Zee S, Hoek G, Boezen HM, Shouten J, van Wijnen JH,

Brunekreef B. Acute effects of urban air pollution on respiratory health

of children with and without chronic respiratory symptoms. Occup

Environ Med 1999;56:802-13.

33. Jalaludin BB, O’Toole BI, Leeder SR. Acute effects of urban ambient air

pollution on respiratory symptoms, asthma medication use, and doctor

visits for asthma in a cohort of Australian children. Environ Res 2004;95:

32-42.

34. Penttinen P, TimonenKL, Tiittanen P,MirmeA,Ruuskanen J, Pekkanen J.

Ultrafine particles in urban air and respiratory health among adult

asthmatics. Eur Respir J 2001;17:428-35.

35. Schwartz J, Neas LM. Fine particles are more strongly associated than

coarse particles with acute respiratory health effects in schoolchildren.

Epidemiology 2000;11:6-10.

36. Brauer M, Avila-Casado C, Fortoul TI, Vedal S, Stevens B, Churg A.

Air pollution and retained particles in the lung. Environ Health Perspect

2001;109:1039-43.

37. Li N, Sioutas C, Cho A, Schmitz D, Misra C, Sempf J, et al. Ultrafine

particulate pollutants induce oxidative stress and mitochondrial damage.

Environ Health Perspect 2003;111:455-60.

38. Sheriff A, Peters TJ, Henderson J, Strachan D, ALSPAC Study Team.

Risk factor associations with wheezing patterns in children followed

longitudinally from birth to 3(1/2) years. Int J Epidemiol 2001;30:

1473-84.

39. El-Sharif N, Abdeen Z, Barghuthy F, Nemery B. Familial and environ-

mental determinants for wheezing and asthma in a case-control study of

school children in Palestine. Clin Exp Allergy 2003;33:176-86.

40. Arshad SH, Kurukulaaratchy RJ, Fenn M, Matthews S. Early life risk

factors for current wheeze, asthma, and bronchial hyperresponsiveness at

10 years of age. Chest 2005;127:502-8.

41. London SJ, Gauderman J, Avol E, Rappaport EB, Peters JM. Family

history and the risk of early-onset persistent, early onset transient, and

late-onset asthma. Epidemiology 2001;12:577-83.



Ryan et al 283

Asthmadiagnosisand

treatm

ent

http://www.epa.gov/ncea

http://www.nrdc.org/air/transportation/schoolbus/sbusinx.asp

42. Grant EN, Lyttle CS, Weiss KB. The relation of socioeconomic factors

and racial/ethnic differences in the US asthma mortality. Am J Public

Health 2000;90:1923-5.

43. Joseph CLM, Ownby DR, Peterson EL, Johnson CC. Racial differences

in physiologic parameters related to asthma among middle-class children.

Chest 2000;117:1336-44.

44. Simon PA, Zeng Z, Wold CM, Haddock W, Fielding JE. Rate of

childhood asthma and associated morbidity in Los Angeles County:

impacts of race/ethnicity and income. J Asthma 2003;40:535-43.

45. Taussig L, Wright A, Holberg C, Halonen M, Morgan W, Martinez F.

Tucson Children’s Respiratory Study: 1980 to present. J Allergy Clin

Immunol 2003;111:661-75.


AUGUST 2005

284 Ryan et al

Asth

madiagnosis

and

treatm

ent

Effect of low-dose ciclesonide on allergen-induced responses in subjects with mildallergic asthma

Gail M. Gauvreau, PhD,a Louis Philippe Boulet, MD,b Dirkje S. Postma, MD, PhD,c

Tomotaka Kawayama, MD,a Richard M. Watson, BSc,a MyLinh Duong, MD,a Francine

Deschesnes, BSc,b Jan G. R. De Monchy, MD, PhD,c and Paul M. O’Byrne, MDa

Hamilton, Ontario, and Quebec City, Quebec, Canada, and Groningen, The Netherlands

Background: Inhalation of allergens by sensitized patients

with asthma induces reversible airway obstruction, airway

hyperresponsiveness, and eosinophilic airway inflammation.

Attenuation of allergen-induced bronchoconstriction and

inflammation has been used to examine the efficacy of

therapeutic agents such as inhaled corticosteroids in asthma.

Ciclesonide, a nonhalogenated inhaled corticosteroid being

developed for the treatment of persistent asthma, remains

inactive until cleaved by esterases in the lung.

Objective: This study examined the effect of low doses of

inhaled ciclesonide, 40 mg and 80 mg, on allergen-induced

bronchoconstriction, serum eosinophil cationic protein, and

eosinophilic airway inflammation.

Methods: Twenty-one nonsmokers with mild atopic asthma

completed a multicenter, randomized, 3-way crossover study

comparing the effects of 7-day treatment of ciclesonide or

placebo. Allergen-induced responses, including the early and

late fall in FEV1, peripheral blood eosinophils, serum

eosinophil cationic protein levels, and eosinophils in induced

sputum were measured.

Results: Ciclesonide 80 mg attenuated the early and late

asthmatic responses, including the change in FEV1, serum

eosinophil cationic protein, and sputum eosinophils measured

at 24 hours postchallenge (P < .025). Ciclesonide 40 mg

attenuated the late asthmatic responses and sputum eosinophils

measured at 24 hours postchallenge (P < .025), with no effect

on the early allergen-induced bronchoconstriction, 24-hour

FEV1, or serum eosinophil cationic protein levels (P < .025).

Conclusion: With the exception of 24-hour postchallenge

peripheral blood eosinophils, a low dose of ciclesonide, 80 mg,

was effective in blocking all allergen-induced responses

measured. (J Allergy Clin Immunol 2005;116:285-91.)

Key words: Inhaled corticosteroid, allergen inhalation, airwayinflammation

Asthma is characterized by reversible airway obstruc-tion, airway hyperresponsiveness, and eosinophilic bron-chial inflammation. Subjects with allergic asthma developan immediate IgE-mediated early asthmatic response(EAR) after inhalation of an allergen to which they aresensitized. Approximately 50% of these subjects alsodevelop a late asthmatic response (LAR), which begins3 to 4 hours after allergen inhalation.1 The LAR isassociated with elevated levels of airway inflammatorycells including eosinophils, basophils, and mast cells.2,3

This model of allergen-induced bronchoconstriction hasbeen used successfully to assess drug efficacy in subjectswith allergic asthma4-13 and is recommended for evalua-tion of inhaled corticosteroids (ICS).14

Inhaled corticosteroids, having potent anti-inflamma-tory properties, are indicated by the Global Initiativefor Asthma as a primary controller for treatment of mild-persistent to severe asthma.15 Studies of ICS haveconsistently shown a significant attenuation of the allergen-induced LAR,5,16-18 with the proposed mechanism atten-uation of the allergen-induced airway inflammation.However, undesirable side effects of ICS at higher doseshave established a need to evaluate ICS properties at verylow doses.14

Ciclesonide is a nonhalogenated ICS for the treatmentof persistent asthma of all severities. This ICS remainsinactive until cleaved by esterases present in the airway,where its activemetabolite, desisobutyryl-ciclesonide, thenbinds glucocorticoid receptors.

Before developing an optimal solution for metered doseinhaler (MDI) formulation for clinical use, early clinicalstudies were performed with ciclesonide by using a drypowder inhaler (DPI) device. One week of ciclesonide800 mg DPI BID (twice daily; using Cyclohaler device[Pharmachemie, Haarlem, The Netherlands]) has been

From athe Department of Medicine, McMaster University, Hamilton; bInstitut

de cardiologie et de pneumologie de l’Universite Laval, Hopital Laval,

Quebec City; and cthe Department of Pulmonology, University Hospital

Groningen.

Supported by Altana Pharma AG.

Disclosure of potential conflict of interest: P. M. O’Byrne has consultant

arrangements with AstraZeneca, GlaxoSmithKline, Topigen, and Altana,

and has received grants/research support from AstraZeneca, GlaxoSmith-

Kline, Pfizer, Altana, and Dynavax. L.-P. Boulet has been on Advisory

Boards for AstraZeneca, Altana Novartis, GlaxoSmithKline, and Merck

Frost, and received lecture fees from 3M, GlaxoSmithKline, AstraZeneca,

and Merck Frosst. Sponsorship for basic research was received from 3M,

Schering, Genentech, Dynavax, Roche, GlaxoSmithKline, Novartis, As-

traZeneca, Altana, and Merck for participating in multicenter studies of the

pharmacotherapy of asthma. D. Postma is on the Advisory Board of Altana.

Received for publication March 29, 2005; revised May 13, 2005; accepted for



Reprint requests: Paul M. O’Byrne, MD, HSC 3W10, McMaster University,

1200 Main St West, Hamilton, Ontario, Canada L8N 3Z5. E-mail:

[email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.021

285

Asthmadiagnosisand

treatm

ent


Abbreviations usedAE: Adverse event

AUC0-2h: Area under the curve of the early response

AUC3-8h: Area under the curve of the late response

BID: Twice daily

DPI: Dry powder inhaler

EAR: Early asthmatic response; maximum % fall in

FEV1 from 0 to 2 hours after allergen challenge

ECP: Eosinophil cationic protein

ICS: Inhaled corticosteroids

LAR: Late asthmatic response; maximum % fall in

FEV1 from 3 to 8 hours after allergen challenge

MDI: Metered dose inhaler

shown to attenuate significantly the allergen-inducedEAR and LAR,19 confirming that this drug has biologicalactivity in this model of allergic inflammation. Theallergen-induced fall in FEV1 was shown to be a sensitivemarker of dose-response effect in an earlier trial ofmometasone furoate16; therefore, the current trial wasperformed to examine the dose-response of the LAR andto determine the efficacy of low-dose ciclesonide in thereduction of the allergen-induced EAR, LAR, and airwayinflammation. In-house testing has led to the developmentof ciclesonide inMDI formulation with hydrofluoroalkane(HFA) propellent, delivering 40 mg, 80 mg, and 160 mg(ex-actuator) per puff. However, data are limited regardingwhat the lowest effective dose is with theMDI formulationof ciclesonide. This study, therefore, was designed toidentify the lowest effective dose of ciclesonide in MDIformulation in an allergen inhalation model.

METHODS

Subjects

Thirty-five subjects were enrolled in the study. Of these,

13 patients did not meet randomization criteria. Twenty-two subjects

(Groningen, n = 5; Laval, n = 7; McMaster, n = 10), 14 men and

8 women, age 19 to 58 years old (Table I), were randomized to one

of the 6 treatment sequences. One subject discontinued the study

prematurely for nonmedical reasons. Inclusion criteria required

subjects to be nonsmokers with mild atopic asthma, free of other

lung disease, andwithout lower respiratory tract infection for 6 weeks

before entering the study. For randomization, subjects were required

to have stable asthma with FEV1 > 70% of predicted; have baseline

methacholine PC20 < 16 mg/mL; use no regular asthma medication

during the study other than infrequent inhaled b2-agonist, which was

withheld for 8 hours before each visit; and have no exposure to

sensitizing allergens apart from house dust mite. Before entering the

study, subjects could not have used systemic steroids or had an

asthma exacerbation for 6 weeks and could not have used inhaled

steroids for at least 4 weeks. Before morning visits to the lab, subjects

were to refrain from tea or coffee.

Study Design

This trial was a multicenter, double-blind, randomized, placebo-

controlled, 3-period crossover study comparing 7 days of treatment

with ciclesonide at 50 mg and 100 mg exvalve, corresponding to

40 mg and 80 mg ex-actuator, with placebo. The study was approved

by the ethics research board of the respective institutions, and signed

informed consent was given to participate. Screening of subjects was

performed over a period of 2 consecutive days and included a detailed

history, physical examination, allergen skin test, FEV1, methacholine

PC20, blood sampling for serum ECP levels, allergen inhalation

challenge, and sputum induction. Subjects who developed an EAR (at

least 20% fall in FEV1 within 2 hours after allergen inhalation) and

LAR (at least 15% fall in FEV1 between 3 and 8 hours after allergen

inhalation) during 1 screening allergen inhalation challenge were

enrolled in the study. Subjects were screened once only, because we

have shown the LAR to be a reproducible measurement.20,21 Subjects

reported to the laboratory for 3 separate treatment periods separated

by a minimum of 3 weeks (Fig 1). This washout time has been shown

to be adequate in a previous study of ICS using this model.16 Each

treatment period consisted of 4 morning visits. Day 1 consisted of

pretreatment measurements of blood eosinophils, sputum inflamma-

tory cells, and lung function; methacholine PC20 needed to be within

1 doubling dose of that measured during the screening period to

continue with the treatment period. If this criterion was met, subjects

then inhaled the first dose of study medication during the morning

visit to the lab. The subsequent doses of study medication were

inhaled for the next 6 consecutive mornings, immediately after

waking, because ciclesonide has been shown to improve asthma

control irrespective of the time of administration.22 Subjects returned

to the laboratory on day 5 for measurement of preallergen challenge

sputum inflammatory cells and on day 6 for allergen inhalation

challenge. Measurements of FEV1 were taken at regular intervals

until 8 hours after challenge. On day 7, subjects underwent measure-

ments of sputum inflammatory cells, blood eosinophils, and serum

ECP. Postallergenmethacholine PC20was not measured, because this

study was not powered sufficiently for this comparison. All subjects

were considered compliant with study medication according to the

diary cards.

Laboratory procedures

Methacholine inhalation test. Methacholine inhalation challenge

was performed as described by Cockcroft.23 Subjects inhaled normal

saline during 2 minutes of tidal breathing, nebulized at 0.13 mL/

minute from a Wright nebulizer, then doubling concentrations of

methacholine chloride. FEV1 was measured at 30, 90, 180, and 300

seconds after each inhalation. The test was terminated when a fall in

FEV1 of 20% of the baseline value occurred, and the methacholine

PC20 was calculated.

Allergen inhalation challenge. Allergen challenge was performed

as described by O’Byrne et al.1 Subjects were skin tested for allergies

to common aeroallergens. The allergen producing the largest skin

wheal diameter was diluted in 0.9% saline and stored for subsequent

allergen inhalation challenges. The concentration of allergen extract

for inhalation was determined from a formula described by Cockcroft

et al,24 and doubling concentrations of allergen were given until a

<20% early fall in FEV1 at 10 minutes postallergen was reached. The

FEV1 was then measured at regular intervals until 8 hours after

allergen inhalation. The early area under the curve (AUC0-2h) and the

late area under the curve (AUC3-8h) were calculated by using the

trapezoidal rule, were normalized to 1 hour, and were expressed as

liters 3 hour (L3h). Subjects inhaled the same dose of allergen for

the 3 treatment periods.

Sputum analysis. Sputumwas induced and processed by using the

method described by Pizzichini et al.25 Cell plugs were selected and

mixed with 0.1% dithiothreitol (Sputolysin; Calbiochem Corp, San

Diego, Calif) and Dulbecco PBS (Life Technologies Inc, Grand

Island, NY) and filtered through a 48-mm nylon gauze (BNSH

Thompson, Scarborough, Ontario, Canada), and cytospins were


AUGUST 2005

286 Gauvreau et al

Asth

madiagnosis

and

treatm

ent

prepared on glass slides. The total cell count was determined by using

a Neubauer hemocytometer chamber (Hausser Scientific, Blue Bell,

Pa) and expressed as the number of cells per milliliter sputum, and

differential cell counts were obtained from slides stained with Diff

Quik (American Scientific Products, McGaw Park, Ill). All slides

were enumerated at 1 site. Slide preparation and enumeration were

performed before unblinding. The study personnel collecting sputum

samples and the technician enumerating the slides had no knowledge

of the coding of the labels, nor of the airway physiology measured

when these sputum samples were collected. All data were collected

centrally and entered into a master database before the random code

was distributed to participating sites.

Serum eosinophil cationic protein and blood eosinophils. Blood

was collected into 4-mL vacutainer hemogard SST tubes (Becton

Dickinson, Franklin Lakes, NJ) for serum separation. Eosinophil

cationic protein (ECP) was released by allowing blood to clot for

60 to 120 minutes at room temperature, and then samples were

centrifuged at 1000g to 1300g for 10 minutes at room temperature.

Serum was removed and stored at 220C until analysis. All serum

ECP was analyzed at 1 site by using the UniCAP system (Pharmacia,

Uppsala, Sweden). Blood was also collected into 2-mL EDTA

vacutainers, and eosinophil counts were performed by Coulter

Counter (Beckman Coulter, Fullerton, Calif) at the respective

institutions.

Statistical analysis

The target sample size of 18 patients is sufficient to guarantee a

power of 90% in correctly concluding superiority at the .0125 level,

1-sided, if the mean difference accounts for 90% of the SD (based on

paired t test). The 21 subjects who completed the study were included

in the statistical analyses. Summary statistics are expressed as mean

and SEM. Between-treatment differences in FEV1 during the LAR,

FEV1 during the EAR, and sputum eosinophils 24 hours after allergen

were analyzed by using ANOVA. The FEV1 measured 24 hours after

allergen was analyzed with analysis of covariance by using the

baseline value as a covariate to test both within-treatment and

between-treatment differences. Within-treatment differences in spu-

tum variables as well as between-treatment differences in blood

eosinophils, ECP, and sputum variables were analyzed by means of

the nonparametric test for 33 6 crossover design. One-sidedP values

were generated for all variables, and significance is shown at

P < .025.

RESULTS

All subjects inhaled the same dose of allergen for the 3treatment periods. There were no serious adverse events(AEs), and 11 treatment-emergent AEs were reported. Ofthese, 7 AEs were mild and 4 were moderate. AllAEs were assessed as unrelated to the study medication,except 1 case of thrombocytopenia, which was assessedunlikely related to the study medication. No adverse eventled to premature discontinuation or a change in studymedication. Before allergen challenge, the degree ofairway hyperresponsiveness was within 1 doubling dose,and there was no significant difference in FEV1 betweenthe 3 treatment periods. During placebo treatment, allsubjects demonstrated early and late airway responsesafter allergen inhalation challenge; the maximum percentfall in FEV1 was 30.4%6 2.2% during the early response

TABLE I. Subject characteristics at study screening visit

Subject

Age

(y) Sex Predicted FEV1 (%) FEV1/VC ratio

Methacholine

PC20 (mg/mL) Allergen

Final dose,

median

50001 25 F 96 0.86 1.12 Ragweed 1:64

50002 21 M 114 0.86 8.94 HDM 1:16

50004 42 M 106 0.69 4.54 HDM 1:32

50005 20 M 73 0.55 0.31 HDM 1:1024

50008 26 F 89 0.92 1.63 Ragweed 1:16

50011 54 M 107 1.18 0.88 Cat 1:32

50012 33 M 88 0.73 15.51 Ragweed 1:64

50013 54 M 73 0.55 3.39 HDM 1:64

50014 19 F 109 0.83 0.54 HDM 1:128

50015 39 M 80 0.67 5.11 HDM 1:128

50025 26 F 83 0.89 1.07 HDM 1:32

50027 28 F 121 0.87 1.11 Cat 1:64

50028 25 F 96 0.86 1.42 Cat 1:16

50029 23 M 99 0.77 1.47 Cat 1:64

50031 30 M 75 0.70 0.86 Cat 1:128

50032 26 M 93 0.73 2.21 HDM 1:64

50033 38 M 112 0.72 1.56 HDM 1:128

50038 19 M 103 0.84 0.70 HDM 1:1024

50039 58 M 107 0.79 8.80 HDM 1:2048

50040 40 M 84 0.80 0.44 HDM 1:2048

50041 49 F 100 0.86 0.21 HDM 1:2048

50045 33 F 85 0.83 0.33 HDM 1:1024

14 HDM

Mean 6 SEM 33.1 6 2.6 14 M 95.1 6 3.0 0.80 6 0.03 1.46 5 Cat 1:64

8 F (15.51-0.21) 3 RW

HDM, House dust mite; VC, vital capacity.



Gauvreau et al 287

Asthmadiagnosisand

treatm

ent

and 24.0% 6 2.1% during the late response, correspond-ing to AUC0-2h of 233.3 6 4.0 L3h and AUC3-8h of227.5 6 4.3 L3h (Table II).

During treatment with ciclesonide 80 mg, there was asignificant reduction in the maximum percent fall inFEV1 during the EAR to 23.6% 6 2.2% (P = .016), theAUC0-2h (P = .013), the maximum percent fall in FEV1

during the LAR to 10.7% 6 2.1% (P < .001), and theAUC3-8h (P < .001). Treatment with ciclesonide 40 mgsignificantly reduced the maximum percent fall in FEV1

during the late response to 13.3% 6 2.1% (P = .0003)and AUC3-8h (P = .0003), with no significant effect onthe early maximum percent fall in FEV1 or AUC0-2h

(P > .025). There was no significant difference betweenciclesonide 40 mg and 80 mg on the EAR, LAR, AUC0-2h,or AUC3-8h (P > .025; Table II).

FEV1 was measured at pretreatment baseline (day 1),preallergen (day 6), and at 24 hours postallergen inhala-tion challenge (day 7). With placebo treatment, the FEV1

of 3.38 L 6 0.17 L at day 7 was significantly lower thanthe day 1 FEV1 of 3.60 L6 0.15 L (P < .001) and the day6 FEV1 of 3.56 L 6 0.16 L (P < .001). During treatmentwith ciclesonide 40 mg, the FEV1 of 3.50 L 6 0.16 Lat day 7 was significantly lower than that of day 1 or day 6,3.65 L 6 0.17 L and 3.62 L 6 0.15 L, respectively (P <.001). During treatment with ciclesonide 80 mg, the FEV1

of 3.59 L 6 0.18 L at day 7 was significantly lower than3.67 L 6 0.18 L at day 1 (P = .013), but not significantlydifferent from the FEV1 of 3.64 L 6 0.16 L at day 6(P = 0.10). Compared with placebo, ciclesonide 80 mgsignificantly attenuated the day 7 allergen-induced de-crease in FEV1 compared with day 1 (P = .002) and day 6(P = .007; Fig 2). Compared with placebo, ciclesonide40mg did not attenuate day 7 allergen-induced decrease inFEV1 compared with day 1 or day 6 (P > .025). There wasa trend toward a dose-response effect for ciclesonide, withgreater attenuation of the 24-hour postallergen fall inFEV1 by ciclesonide 80 mg compared with ciclesonide40 mg; however, this did not reach statistical significance(1-sided P = .028; Fig 2).

At 24 hours after allergen inhalation challenge, serumECP increased from a baseline of 20.7 mg/L 6 3.0 mg/Lto 31.0 mg/L 6 5.8 mg/L with placebo, 33.0 mg/L 6 5.6mg/L with ciclesonide 40 mg, and 26.0 mg/L 6 5.3 mg/Lwith ciclesonide 80 mg. There was significant attenuationof the allergen-induced increase in serum ECP withciclesonide 80 mg versus placebo treatment (P = .024),but no effect of ciclesonide 40 mg. Peripheral bloodeosinophils were not significantly reduced with cicleso-nide 40 mg or 80 mg (P > .025). There was an allergen-induced increase in the number of peripheral bloodeosinophils at 24 hours after challenge, increasing froma baseline of 0.294 3 106/mL 6 0.037 3 106/mL to0.4043 106/mL6 0.0403 106/mLwith placebo, 0.3853106/mL 6 0.047 3 106/mL with 40 mg ciclesonide,and 0.347 3 106/mL 6 0.052 3 106/mL with 80 mgciclesonide.

There was an allergen-induced increase in the percentsputum eosinophils (Fig 3, A) and the number of eosin-ophils per milliliter sputum (Fig 3, B) on day 7 comparedwith day 5 with placebo, ciclesonide 40 mg, andciclesonide 80 mg treatment (P < .002). However, bothciclesonide 40 mg and 80 mg significantly attenuated theallergen-induced increase in the percentage of sputumeosinophils (P < .001 and P = .006, respectively). Onlyciclesonide 80 mg significantly attenuated the allergen-induced increase in the number of eosinophils per milli-liter sputum, but the trend toward a dose-dependent effect

TABLE II. Effect of ciclesonide on the maximum percent

fall in FEV1 and area under the curve of the early and

late airway responses

Placebo

Ciclesonide

40 mg

Ciclesonide

80 mg

EAR (%) 30.4 6 2.2 28.2 6 2.2 23.6 6 2.2*

AUC0-2h

(L3h)

233.3 6 4.0 229.2 6 4.6 224.9 6 4.1*

LAR (%) 24.0 6 2.1 13.3 6 2.1** 10.7 6 2.1**

AUC3-8h

(L3h)

227.5 6 4.3 211.8 6 4.0** 29.4 6 2.7**

*P < .025 compared with placebo.

**P < .013 compared with placebo.

FIG 1. Study schematic. BL, Pretreatment baseline.


AUGUST 2005

288 Gauvreau et al

Asth

madiagnosis

and

treatm

ent

for ciclesonide did not reach statistical significance(1-sided P = .03).

DISCUSSION

Ciclesonide is a new generation inhaled glucocortico-steroid that remains inactive until cleaved by esterases pre-sent in the lung.Ciclesonide has been shown to be effectiveto reduce allergen-induced early and late responses whenadministered at 800mg twice daily by Cyclohaler (DPI),19

yet it was unknown whether lower doses of 80 mg or40 mg administered by MDI would also be effective. Inthe current study, ciclesonide was administered by MDIonce daily for 7 days at considerably lower doses thanpreviously studied.

This study has demonstrated that once-a-day treatmentwith 40 mg or 80 mg ciclesonide for 7 days is indeedefficacious in this model of allergen-induced bronchocon-striction and airway inflammation. Furthermore, we wereable to determine the lowest effective dose of ciclesonidefor attenuating all allergen-induced responses investigatedin this study. In contrast with the 40 mg dose, ciclesonideat 80 mg per day attenuated the allergen-induced earlyairway response, the sustained fall in FEV1 measured24 hours postchallenge, and the allergen-induced accu-mulation of eosinophils into the airways. Whether 80 mgrepresents a plateau beyond which there is no furtherattenuation of allergen-induced responses is unknown.

Compared with the aforementioned study of 800 mgciclesonide DPI BID resulting in approximately 51%attenuation in the late fall in FEV1,

19 80 mg in the current

study resulted in 58% inhibition and 40 mg resulted in45% inhibition of the fall in FEV1 during the late response.The early response was not attenuated as effectively withthe lower doses of ciclesonide, with 40 mg and 80 mgproviding approximately 8% and 23% inhibition of theearly fall in FEV1, respectively, compared with the 45%inhibition with 800 mg ciclesonide BID previously stud-ied.19 The deposition pattern of HFA-MDI formulations inthe peripheral lung may not inhibit the EAR as effectivelycompared with the DPI formulation, which gets depositedin the large, central airways where the EAR is likely to bemost active. This supports the consistent observation thatmultiple or single doses of inhaled steroids may signifi-cantly attenuate the LAR without significantly attenuatingthe EAR.5,18,26 Moreover, these data suggest that higherlevels of HFA-MDI formulation steroids are necessary toinhibit IgE-mediated early responses to inhaled allergen,such as mast cell degranulation, compared with lowerlevels of steroids that appear to suppress the late responseeffectively, likely through inhibition of proinflammatorycytokine gene expression.27

Ciclesonide HFA-MDI 200 mg 4 times daily adminis-tered in a previous study28 did not affect urine cortisollevels after 4 weeks treatment, which suggests that the80 mg dose in the current study would have a very lowpotential for systemic activity. We did not measuresystemic safety markers, because no effect would havebeen expected. It is unknown, however, whether sys-temic activity in compartments such as the circulationand/or bone marrow is an unidentified yet important siteof action of inhaled steroids. There is certainly evidenceshowing that steroids provide anti-inflammatory effects

FIG 2. A comparison of FEV1 measured at day 1 (before treatment), day 6 (after 6 days of treatment), and day 7

(24 hours after allergen inhalation challenge) with placebo (open bars), ciclesonide 40 mg (hatched bars), and

ciclesonide 80 mg (solid bars) treatments.*P < .025 compared with day 1 pretreatment. P < .025 compared

with day 6 preallergen. P < .025 attenuation of allergen-induced increase compared with placebo.



Gauvreau et al 289

Asthmadiagnosisand

treatm

ent

in these other compartments.17,29 The low potential forsystemic activity with low-dose ciclesonide treatmentsupports the notion that steroid regulation of allergen-induced bone marrow responses is likely through sup-pression of proinflammatory mediators generated inairways that subsequently control the bone marrow,rather than the belief that steroids have a direct effecton cells in the bone marrow.30 Because new generationsteroids are preferred as a result of their low systemiceffects, this question of whether some systemic activity isalso required will become an important issue that needsto be addressed.

There was a numerical, though insignificant (P > .025),dose-response for low doses of ciclesonide during the lateresponse, which is believed to be one of the most sensitivevariablesused todemonstrate adose-response.16Surprisingly,these data demonstrate dose-responses to measurementsof airway obstruction and inflammation at 24 hourspostchallenge. This is an unexpected finding, indicatingthese variables may be worthy of evaluation in subsequenttrials evaluating dose-responses.

Although results from this study suggest that 80 mgciclesonide may be the minimally effective dose forprotection against the EAR and serum ECP levels, it isnoteworthy that there was a significant effect of both80 mg and 40 mg ciclesonide on the LAR and percentageof sputum eosinophils. This implies that the minimallyeffective dose for protection against these parameters islikely to be less than 80mg. However, during the course ofthe study, ciclesonide was always administered in the

morning immediately after waking, and approximately1 hour preallergen challenge. Hence, the drug was presentat the highest possible levels during challenges. Whetherthis degree of efficacy would have been observed had drugadministration and allergen challenge been separated by alonger time interval is unknown, but is important toconsider, because the protective effects of inhaled steroidsagainst allergen-induced early responses, airway eosino-philia, and allergen-induced airway hyperresponsivenessare partially or completely lost as early as 12 hours afterdiscontinuation of therapy.18

Direct comparisons between various ICS were notperformed in this study, largely because of the difficultiesassociated with 4-way crossover studies. Although a directcomparison would need to be performed to indicate therelative potency of ciclesonide, the dose of 80 mg appearsto be similar to proven effective doses of other inhaledsteroids; 50 ug mometasone furoate BID has been shownto attenuate significantly the EAR, LAR, and 24-hourpostallergen sputum eosinophils.16 This study has pro-vided new information regarding the minimally effectivedoses of inhaled ciclesonide for inhibition of allergen-induced airway responses and the apparent local anti-inflammatory effects on the airways. Further evaluation ofciclesonide will be required to address whether these lowdoses are clinically effective.

We thank Dr A. Widmann for logistic support of this study and

Mrs C. Veltman for help in patient recruitment. We also thank Tara

FIG 3. A comparison of percent sputum eosinophils (top panel) and number of sputum eosinophils (bottom

panel)measured on day 1 (before treatment), day 5 (after 5 days treatment), and day 7 (24 hours after allergen

challenge) with placebo (open bars), ciclesonide 40 mg (hatched bars), and ciclesonide 80 mg (solid bars)

treatments.*P < .025 compared with day 5 preallergen. P < .025 attenuation of allergen-induced increase

compared to placebo.


AUGUST 2005

290 Gauvreau et al

Asth

madiagnosis

and

treatm

ent

Strinich, Irene Babirad, and Tracy Rerecich for help in sample

preparation and enumeration.

REFERENCES

1. O’Byrne PM, Dolovich J, Hargreave FE. Late asthmatic responses.

Am Rev Respir Dis 1987;136:740-51.

2. de Monchy JG, Kauffman HF, Venge P, Koeter GH, Jansen HM, Sluiter

HJ, et al. Bronchoalveolar eosinophilia during allergen-induced late

asthmatic reactions. Am Rev Respir Dis 1985;131:373-6.

3. Gauvreau GM, Parameswaran KN, Watson RM, O’Byrne PM. Inhaled

leukotriene E(4), but not leukotriene D(4), increased airway inflamma-

tory cells in subjects with atopic asthma. Am J Respir Crit Care Med

2001;164:1495-500.

4. Fahy JV, Fleming HE, Wong HH, Liu JT, Su JQ, Reimann J, et al. The

effect of an anti-IgE monoclonal antibody on the early- and late-phase

responses to allergen inhalation in asthmatic subjects. Am J Respir Crit

Care Med 1997;155:1828-34.

5. Gauvreau GM, Doctor J, Watson RM, Jordana M, O’Byrne PM. Effects

of inhaled budesonide on allergen-induced airway responses and airway

inflammation. Am J Respir Crit Care Med 1996;154:1267-71.

6. Gauvreau GM, Jordana M, Watson RM, Cockroft DW, O’Byrne PM.

Effect of regular inhaled albuterol on allergen-induced late responses and

sputum eosinophils in asthmatic subjects. Am J Respir Crit Care Med

1997;156:1738-45.

7. Gauvreau GM, Becker AB, Boulet LP, Chakir J, Fick RB, Greene WL,

et al. The effects of an anti-CD11a mAb, efalizumab, on allergen-

induced airway responses and airway inflammation in subjects with

atopic asthma. J Allergy Clin Immunol 2003;112:331-8.

8. Hamilton A, Faiferman I, Stober P, Watson RM, O’Byrne PM.

Pranlukast, a cysteinyl leukotriene receptor antagonist, attenuates aller-

gen-induced early- and late-phase bronchoconstriction and airway

hyperresponsiveness in asthmatic subjects. J Allergy Clin Immunol

1998;102:177-83.

9. Hamilton AL, Watson RM, Wyile G, O’Byrne PM. Attenuation of early

and late phase allergen-induced bronchoconstriction in asthmatic sub-

jects by a 5-lipoxygenase activating protein antagonist, BAYx 1005.

Thorax 1997;52:348-54.

10. Leigh R, Vethanayagam D, Yoshida M, Watson RM, Rerecich T, Inman

MD, et al. Effects of montelukast and budesonide on airway responses

and airway inflammation in asthma. Am J Respir Crit Care Med 2002;

166:1212-7.

11. Parameswaran K, Watson R, Gauvreau GM, Sehmi R, O’Byrne PM. The

effect of pranlukast on allergen-induced bone marrow eosinophilopoiesis

in subjects with asthma. Am J Respir Crit Care Med 2004;169:915-20.

12. Kidney JC, Boulet LP, Hargreave FE, Deschesnes F, Swystun VA,

O’Byrne PM, et al. Evaluation of single-dose inhaled corticosteroid

activity with an allergen challenge model. J Allergy Clin Immunol 1997;

100:65-70.

13. Boulet LP, Chakir J, Milot J, Boutet M, Laviolette M. Effect of

salmeterol on allergen-induced airway inflammation in mild allergic

asthma. Clin Exp Allergy 2001;31:430-7.

14. Boulet LP, Cockcroft DW, Toogood J, Lacasse Y, Baskerville J,

Hargreave FE. Comparative assessment of safety and efficacy of inhaled

corticosteroids: report of a committee of the Canadian Thoracic Society.

Eur Respir J 1998;11:1194-210.

15. Global strategy for asthma management and prevention. NHLBI/WHO

workshop report. Global Initiative for Asthma. Bethesda: National

Institutes of Health, National Heart Lung and Blood Institute; 2002.

NIH publication number 02-3659.

16. Inman MD, Watson RM, Rerecich T, Gauvreau GM, Lutsky BN,

Stryszak P, et al. Dose-dependent effects of inhaled mometasone furoate

on airway function and inflammation after allergen inhalation challenge.


17. Wood LJ, Sehmi R, Gauvreau GM, Watson RM, Foley R, Denburg JA,

et al. An inhaled corticosteroid, budesonide, reduces baseline but not

allergen-induced increases in bone marrow inflammatory cell progenitors

in asthmatic subjects. Am J Respir Crit Care Med 1999;159:1457-63.

18. Subbarao P, Dorman SC, Rerecich T, Watson RM, Gauvreau GM,

O’Byrne PM. Protection by budesonide and fluticasone on allergen-

induced airway responses following discontinuation of therapy. J Allergy

Clin Immunol 2005;15:745-50.

19. Larsen BB, Nielsen LP, Engelstatter R, Steinijans V, Dahl R. Effect of

ciclesonide on allergen challenge in subjects with bronchial asthma.

Allergy 2003;58:207-12.

20. Gauvreau GM, Watson RM, Rerecich TJ, Baswick E, Inman MD,

O’Byrne PM. Repeatability of allergen-induced airway inflammation.


21. Inman MD, Watson R, Cockcroft DW, Wong BJ, Hargreave FE,

O’Byrne PM. Reproducibility of allergen-induced early and late asth-

matic responses. J Allergy Clin Immunol 1995;95:1191-5.

22. Postma DS, Sevette C, Martinat Y, Schlosser N, Aumann J, Kafe H.

Treatment of asthma by the inhaled corticosteroid ciclesonide given

either in the morning or evening. Eur Respir J 2001;17:1083-8.

23. Cockcroft DW. Measure of airway responsiveness to inhaled histamine

or methacholine: method of continuous aerosol generation and tidal

breathing inhalation. In: Hargreave FE, Woolcock AJ, editors. Airway

responsiveness: measurement and interpretation. Mississauga: Astra

Pharmaceuticals Canada Ltd; 1985. p. 22-8.

24. Cockcroft DW, Murdock KY, Kirby J, Hargreave F. Prediction of airway

responsiveness to allergen from skin sensitivity to allergen and airway

responsiveness to histamine. Am Rev Respir Dis 1987;135:264-7.

25. Pizzichini E, Pizzichini MM, Efthimiadis A, Evans S, Morris MM,

Squillace D, et al. Indices of airway inflammation in induced sputum:

reproducibility and validity of cell and fluid-phase measurements. Am J

Respir Crit Care Med 1996;154:308-17.

26. Parameswaran K, Inman MD, Watson RM, Morris MM, Efthimiadis A,

Ventresca PG, et al. Protective effects of fluticasone on allergen-induced

airway responses and sputum inflammatory markers. Can Respir J 2000;

7:313-9.

27. Masuyama K, Jacobson MR, Rak S, Meng Q, Sudderick RM, Kay AB,

et al. Topical glucocorticosteroid (fluticasone propionate) inhibits cells

expressing cytokine mRNA for interleukin-4 in the nasal mucosa in

allergen-induced rhinitis. Immunology 1994;82:192-9.

28. Lee DK, Fardon TC, Bates CE, Haggart K, McFarlane LC, Lipworth BJ.

Airway and systemic effects of hydrofluoroalkane formulations of

high-dose ciclesonide and fluticasone in moderate persistent asthma.

Chest 2005;127:851-60.

29. Gauvreau GM, Wood LJ, Sehmi R, Watson RM, Dorman SC, Schleimer

RP, et al. The effects of inhaled budesonide on circulating eosinophil

progenitors and their expression of cytokines after allergen challenge in

subjects with atopic asthma. Am J Respir Crit Care Med 2000;162:

2139-44.

30. Inman MD, Denburg JA, Ellis R, Dahlback M, O’Byrne PM. Allergen-

induced increase in bone marrow progenitors in airway hyperresponsive

dogs: regulation by a serum hemopoietic factor. Am J Respir Cell Mol

Biol 1996;15:305-11.



Gauvreau et al 291

Asthmadiagnosisand

treatm

ent

Roflumilast, an oral, once-dailyphosphodiesterase 4 inhibitor, attenuatesallergen-induced asthmatic reactions

Emmerentia van Schalkwyk, MBChB,a K. Strydom, MBChB,a Zelda Williams, RN,b

Louis Venter, MSc,c Stefan Leichtl, PhD,d Christine Schmid-Wirlitsch, PhD,d

Dirk Bredenbroker, MD,d and Philip G. Bardin, FRACP, PhDb Cape Town

and Rivonia, South Africa, Melbourne, Australia, and Konstanz, Germany

Background: Asthma is a chronic inflammatory disease with

increasing incidence worldwide. Roflumilast is an oral, once-

daily inhibitor of phosphodiesterase type 4 that prevents the

breakdown of cyclic adenosine monophosphate levels, leading

to inhibition of proinflammatory signaling.

Objective: The objective of this study was to investigate the

effects of repeated doses of 250 or 500 mg of roflumilast on

asthmatic airway responses to allergen.

Methods: Twenty-three patients with mild asthma with an

FEV1 of 70% of predicted value or greater were enrolled in

a randomized, double-blind, placebo-controlled, 3-period

crossover study. Patients participated in 3 treatment periods

(7-10 days) separated by washout periods (2-5 weeks).

Patients received 250 mg of oral roflumilast, 500 mg of

roflumilast, or placebo once daily. Allergen challenge was

performed at the end of each treatment period, followed by

FEV1 measurements over the ensuing 24 hours.

Results: Late asthmatic reactions (LARs) were reduced by 27%

(P = .0110) and 43% (P = .0009) in patients treated with 250

and 500 mg of roflumilast, respectively, versus placebo.

Roflumilast, 250 and 500 mg, also attenuated early asthmatic

reactions by 25% (P = .0038) and 28% (P = .0046), although not

to the same extent as LAR attenuation. Roflumilast was well

tolerated. No serious adverse events or discontinuations caused

by adverse events were reported.

Conclusion: Once-daily oral roflumilast modestly attenuated

early asthmatic reactions and, to a greater extent, LARs to

allergen in patients with mild allergic asthma. Pronounced

suppression of late responses in an allergen challenge model

suggests that roflumilast might have anti-inflammatory activity,

which could provide clinical efficacy in chronic inflammatory

pulmonary diseases, such as asthma. (J Allergy Clin Immunol

2005;116:292-8.)

Key words: Asthma, roflumilast, phosphodiesterase type 4, allergenprovocation, inflammation, late phase

Asthma is a worldwide public health concern that hasbeen increasing in prevalence, particularly in developedcountries.1,2 In the United States approximately 31millionpersons have been given a diagnosis of asthma.3 Further-more, the economic burden of asthma (eg, prescriptiondrugs, hospitalization, and loss of productivity) has alsoincreased over the past 20 years, with the economiccosts associated with asthma estimated to exceed thoseof tuberculosis and HIV-AIDS combined.4

Asthma is characterized by chronic inflammation andairway hyperresponsiveness (AHR), leading to recurrentepisodes of wheezing, breathlessness, chest tightness, andcoughing.5 A primary goal of asthma therapy is to achieveand maintain control of clinical symptoms by improvinglung function and reducing AHR. In addition, reducingthe frequency of asthmatic exacerbations and improvinghealth-related quality of life are important therapeuticgoals. There are several therapeutic options for long-termmaintenance (ie, controllers) and symptom relief (ie,relievers) available to asthmatic patients. Commonly pre-scribed fixed-dose combination therapies based on inhaledcorticosteroids (ICSs) provide effective relief to manypatients with asthma and comprise the current standard ofcare. Unfortunately, this standard of care does not controlasthma in all patients. Long-term use of ICSs can alsopotentially cause serious systemic side effects,6 and poorcompliance is an additional concern.7 Furthermore, a smallnumber of patients are unresponsive to ICS therapy andrequire alternative therapeutic options.8 Novel therapiesthat increase the natural anti-inflammatory response ininflammatory target cells have the potential to meet thecurrent unmet medical need for asthmatic patients.

Inhaled allergen challenge in patients with mild allergicasthma results in an early asthmatic reaction (EAR),followed by a late-phase response (the late asthmaticreaction [LAR]). The EAR is a consequence of theactivation and degranulation of cells expressing allergen-specific IgE. Mediators are released that induce nervestimulation, mucus hypersecretion, vasodilation, andmicrovascular leakage.

From athe Department of Internal Medicine, University of Stellenbosch, Cape

Town; bMonash Centre for Inflammatory Diseases, Monash University and

Medical Centre, Melbourne; cALTANA Madaus (Pty) Ltd, Rivonia; anddALTANA Pharma AG, Konstanz.

Disclosure of potential conflict of interest: Drs Bredenbroker, Schmid-

Wirlitsch, Leichtl, and Venter are employees of ALTANA Pharma, and

Dr Bardin has served as a consultant to ALTANA Pharma and received

research support from GlaxoSmithKline, AstraZeneca, Schering Plough,

and Boehringer-Ingelheim. The other authors have no conflict of interest

to disclose.

Received for publication August 26, 2004; revised April 8, 2005; accepted for

publication April 18, 2005.


Reprint requests: Philip G. Bardin, FRACP, PhD, Monash Medical Centre,

246 Clayton Rd, Clayton 3168, Melbourne, Australia. E-mail: p.bardin@

southernhealth.org.au.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.023

292

Asth

madiagnosis

and

treatm

ent



Abbreviations used

AHR: Airway hyperresponsiveness

AUC: Area under the curve

cAMP: Cyclic adenosine monophosphate

COPD: Chronic obstructive pulmonary disease

EAR: Early asthmatic reaction

ICS: Inhaled corticosteroid

LAR: Late asthmatic reaction

PC20FEV1: Provocative concentration resulting in a

20% decrease in FEV1

PDE4: Phosphodiesterase type 4

The LAR is believed to reflect mechanisms of asthmaticinflammation because in this response activated airwaycells release cytokines and chemokines locally and into thecirculation, thus stimulating the release of inflammatoryleukocytes, particularly eosinophils and their precursors,from the bone marrow into the circulation. Inflammatorycells in the peripheral blood are then recruited into theinflamed airways, where they augment airway inflamma-tion and increase AHR.

Roflumilast (3-cyclo-propylmethoxy-4-difluorome-thoxy-N-[3,5-di-chloropyrid-4-yl]-benzamide) is an oral,once-daily phosphodiesterase type 4 (PDE4) inhibitor inclinical development as long-term maintenance therapyfor chronic obstructive pulmonary disease (COPD) andasthma. Phosphodiesterases hydrolyze the second mes-senger cyclic adenosine monophosphate (cAMP) to5#-adenosine monophosphate, rendering it inactive. ThePDE4 isozyme has localized activity in the lung, andPDE4 inhibitors, such as roflumilast, block cAMP hy-drolysis in the airways. The inflammatory response ishighly sensitive to levels of cAMP, and by preventing thebreakdown of cAMP, PDE4 inhibitors are associated witha natural anti-inflammatory activity. As a second messen-ger, cAMP blocks proliferation and chemotaxis of inflam-matory cells (eg, lymphocytes), inhibits proinflammatorycell activity (eg, phagocytosis and respiratory burst), andsuppresses the release of inflammatory and cytotoxicmediators (eg, TNF-a) in the lungs.9 In previous in vitroand in vivo studies, roflumilast decreased inflammatorycell infiltration into the airways, total protein levels, andTNF-a release.10 Thus by maintaining levels of cAMPin key airway inflammatory cells, roflumilast acts as asteroid-free anti-inflammatory agent that might have util-ity in diseases such as asthma and COPD.11

This proof-of-concept study investigated the effect ofrepeat doses of roflumilast (250 and 500 mg) on allergen-induced asthmatic responses and AHR in patients withmild allergic asthma.

METHODS

Patients

Patients were included in the study if they had a history of

wheezing consistent with mild asthma, had an FEV1 of 70% of

predicted value or greater, were between 18 and 50 years of age, and

were not receiving treatment with asthma medications other than

short-acting bronchodilators for symptom relief. Inclusion criteria

also required a positive allergen skin prick test response and

hyperresponsiveness to methacholine, with a provocative concentra-

tion resulting in a 20% decrease in FEV1 (PC20FEV1) of 16mg/mL or

less. All patients provided informed written consent, and the Human

Research Ethics Committee of the University of Stellenbosch (Cape

Town, South Africa) approved the study.

Study design and measurements

In a double-blind, placebo-controlled, 3-period crossover study,

patients were randomized to 250 mg of roflumilast, 500 mg of

roflumilast, or placebo once daily for 7 to 10 days, with washout

periods of 2 to 5 weeks before each treatment period (Fig 1).

Randomization was performed after the first washout period at the

first treatment visit. Study medication was taken between 7 AM and

10 AM daily.

At the first baseline visit, FEV1 was measured with a mobile

spirometer (Vitalograph, Hamburg, Germany), and AHR to meth-

acholine challenge was assessed for each patient. At the second

baseline visit, allergen challenge was done to determine a provoca-

tive concentration causing an early response to allergen (FEV125%

decrease) and late response (FEV1 15% decrease), as detailed

below. During the baseline visits, patients were assessed for adher-

ence to inclusion and exclusion criteria. In eligible patientsmethacho-

line challenge was performed on the first treatment visit, and allergen

challenge was performed on the second treatment visit (ie, after

taking study medication for 7-10 days). The FEV1 measurements

were done over the ensuing 24 hours at 5, 10, 15, 30, 45, and

60minutes; subsequently at 1-hour intervals for the next 11 hours; and

at 4.5, 5.5, and 24 hours after allergen challenge. At the conclusion of

the 24-hour period after allergen challenge, methacholine challenge

was repeated.

Allergen and methacholine challenges were performed according

to the bronchial provocation technique described by Chai et al,12 as

previously published by Bardin et al.13 The challenge tests were

performed by using a Spira Elektra breath-actuated inhalation

dosimeter (SPIRA OY; Hameenlinna, Finland). The duration of

delivery per actuation was 0.6 seconds, and the flow of compressed

air was 8 L/min. Each inhalation was controlled to result in an

inspiratory flow of 0.6 to 0.8 L/s. Provocation began with 5 breaths of

saline inhalation, each lasting 5 seconds, followed by holding the

breath for 2 seconds (from functional residual capacity to total lung

capacity). Challengeswere performed only if baseline FEV1was 70%

of predicted value or greater, if it was within 12% of the initial value

measured at baseline, and if, after saline inhalation, the decrease in

FEV1 was 10% or less. At initial assessment, a decrease in FEV1 of

25% or greater from the postsaline value within the first 2 hours after

the challenge defined the EAR, whereas the LAR was characterized

by a decrease of 15% or greater in FEV1 from the postsaline value at 2

or more time points after spontaneous reversal of the EAR (>2-12

hours after challenge) and with a typical gradual deterioration in

FEV1.

Each patient was challenged with a single allergen identified on

the basis of reactivity during the skin prick test. The allergens used in

this study were house dust mite, cat hair, South African grass pollen,

and Bermuda grass pollen (dilutions: 1022, 1023, 1024, 1025, and

1026 SQU/mL; BayerMiles, Inc, Cape Town, South Africa). Patients

always started with the lowest allergen concentration, and FEV1 was

recorded 5, 10, and 15 minutes after inhalation. If the decrease in

FEV1 was less than 10% of the postsaline FEV1, a 10-fold higher

allergen concentration was administered; if the decrease was 10% to

15%, a 5-fold higher concentration was administered; and if the



van Schalkwyk et al 293

Asthmadiagnosisand

treatm

ent

decrease in FEV1 was greater than 15% to 25%, the same allergen

concentration was administered a second time. The concentration of

allergen causing a decrease in postsaline FEV1 of greater than 25%

was used in each patient for allergen provocation on subsequent study

days. If needed, patients received single doses of inhaled beclome-

thasone or budesonide (800 mg) for stabilization of their condition

after allergen challenge.

For methacholine challenge, if the decrease in postsaline FEV1

was 10% or less, 5 breaths of the lowest methacholine concentration

(0.03 mg/mL) were inhaled. Doubling concentrations of methacho-

line were then administered until either FEV1 decreased to 20% or

greater of the postsaline value or the highest available methacholine

concentration (16 mg/mL) was reached. The PC20FEV1 was calcu-

lated by means of linear interpolation of the FEV1 dose-response

curve.13

Clinical laboratory evaluations, electrocardiography, vital signs

measurement, and physical examinations were performed at the

beginning and end of the study. Adverse events were assessed

throughout the study.

Statistical methods

The primary efficacy variable of this study was inhibition of the

LAR. Attenuation of the EAR and AHR were secondary end points.

Pairwise tests were used to compare the effects of the 3 treatments on

the area under the curve (AUC) from 2 to 12 hours of the FEV1,

corresponding to the primary variable, LAR, and on the AUC from

0 to 2 hours of the FEV1, corresponding to the EAR. The AUC of

the FEV1 decrease over time was compared by using ANOVA for the

3-period crossover design for both the LAR and EAR. Geometric

means and 95% CIs were calculated for the differences between

population means. Point estimates and 95% CIs were calculated for

the maximum decrease analysis. Percentage reductions in the EAR

and LAR were calculated for the differences between population

least-squares means for the per-protocol population. Other secondary

end points were analyzed in a descriptive manner with geometric

means, and 95% CIs were calculated where appropriate.

RESULTS

Patients

Patient demographics and characteristics at baseline aresummarized in Table I. A total of 23 patients wererandomized in this study; 2 patients terminated the studyprematurely because of nonmedical reasons. The per-protocol population varied among the different efficacyanalyses because of missing or invalid data. Safety dataare reported for the intent-to-treat population that wasactually exposed to each treatment. The numbers ofpatients exposed to 500 mg of roflumilast, 250 mg ofroflumilast, and placebo were 23, 21, and 22, respectively.Three patients were exsmokers, and 1 patient was a currentsmoker. Median age was 28 years. At baseline, the meanFEV1 was 3.17 L (SD, 0.82), 89% of the predicted value.

Lung function and response toallergen challenge

Patients treated with 250 and 500 mg of roflumilastexperienced a significant attenuation of the LAR afterallergen challenge compared with those treated withplacebo. Treatment with 250 mg of roflumilast led to a27% reduction in the LAR by means of AUC analysis(P = .0110) compared with placebo, whereas 500 mg ofroflumilast led to a 43% (P = .0009) reduction comparedwith placebo (Table II). Therefore roflumilast-inducedattenuation of the LAR showed a dose-related trend.Significant attenuation of the LAR compared with placebowas maintained for 12 hours after allergen challenge

FIG 1. Study design. In this randomized, double-blind, 3-period crossover study, each treatment period lasted

7 to 10 days and was preceded by a 2- to 5-week washout period.

TABLE I. Baseline patient demographics and

characteristics

Parameter Patients (n = 23)

Median age, y (range) 28 (20-44)

Mean height 6 SD, cm 168 6 10.7

Mean weight 6 SD, kg 76 6 19.3

Sex distribution, n (%)

Female 12 (52)

Male 11 (48)

Smoking habits, n (%)

Never smoked 19 (83)

Exsmokers 3 (13)

Current smokers 1 (4)

Mean FEV1 6 SD, L 3.17 6 0.82

Mean FEV1 6 SD, % predicted 89 6 10

Mean PC20FEV1 6 SD, mg/mL 4.09 6 3.79


AUGUST 2005

294 van Schalkwyk et al

Asth

madiagnosis

and

treatm

ent

(Fig 2). Treatment with 250 and 500 mg of roflumilastresulted in a modest yet statistically significant inhibitionof the maximum decrease in FEV1 during the LAR of 17%(P = .0321) and 33% (P = .0002), respectively, comparedwith placebo (Table III). Both measures of lung function(AUC and maximum decrease) demonstrated that roflu-milast inhibits bronchoconstriction associated with anti-gen challenge.

On the basis of the analysis of the AUC of FEV1,patients treated with 250 and 500 mg of roflumilast alsoexperienced a statistically significant attenuation in theEAR compared with placebo. Treatment with 250 mg ofroflumilast reduced the EAR by 25% (P = .0038) and500 mg of roflumilast reduced the EAR by 28%(P = .0046) versus placebo, respectively (Fig 2). Therewere no statistically significant differences between the250- and 500-mg roflumilast doses with respect to LARand EAR attenuation (Table II). There was a modestreduction of 14% in themaximumdecrease in FEV1duringthe EAR for both 250 and 500 mg of roflumilast comparedwith placebo (Table III).

AHR was only slightly modified by roflumilast. Duringroflumilast treatment, the PC20FEV1 ratio was 1.0 orgreater, indicating that airway responsiveness did notincrease despite the preceding allergen challenge (1.03for 250 mg of roflumilast and 1.11 for 500 mg ofroflumilast). The PC20FEV1 ratio was less than 1.0(0.87) in patients treated with placebo, which reflects anincrease in AHR (Table IV). There was an apparent trendin doubling doses between roflumilast and placebo treat-ment, but the difference between treatments did not reachstatistical significance.

Safety

Roflumilastwaswell tolerated at both dose levels tested.No serious adverse events or discontinuations because ofadverse events occurred during the study. Most adverseevents were mild to moderate in intensity and were relatedto the digestive tract (eg, diarrhea and gastrointestinaldisorder) or the nervous system (eg, headache). Headachewas the most common adverse event. During baseline,6 (26%) of 23 patients reported headaches. During the

treatment period, headache was reported by 4 (18%) of 22patients, 6 (29%) of 21 patients, and 8 (35%) of 23 patientstreated with placebo, 250 mg of roflumilast, and 500 mg ofroflumilast, respectively. The majority of headachesreported during the treatment period (73%) were consid-ered by the investigator to be unlikely or not related tostudymedication. The 6 adverse events associatedwith thedigestive tract were considered by the investigator to belikely related to study medication, but no adverse eventswere judged definitely related. There were no clinicallyrelevant changes in vital signs, electrocardiographicresults, or clinical laboratory parameters.

DISCUSSION

Asthma is characterized by chronic inflammation of theairways that might be responsive to treatment with PDE4inhibitors. We have assessed the benefits of the PDE4inhibitor roflumilast in patients with mild asthma using anallergen challenge model. Roflumilast had only a modesteffect on the EAR but demonstrated a more pronouncedeffect on the LAR. This suggests a potential anti-inflam-matory role for roflumilast because it is believed thatallergen-induced LARs are linked to an influx of inflam-matory cells and mediators associated with an inflamma-tory response.14-16

In several in vivo and in vitro animal models, roflumi-last has demonstrated multiple anti-inflammatory effects,including inhibition of inflammatory cell infiltration andreduction of TNF-a release in the lungs.11 In a previousstudy in patients with exercise-induced asthma, 500 mg ofroflumilast reduced the decrease in FEV1 after exercisechallenge by 41% and the median TNF-a levels by 21%versus placebo.17 Exercise-induced bronchoconstriction isregarded as an asthmatic airway reaction to nonspecificstimuli.18,19 As a common characteristic of asthma, exer-cise-induced bronchoconstriction is an appropriate modelin which to study the efficacy of roflumilast; however,alternative models are needed to elucidate the mechanismof action of roflumilast in patients with asthma. Theallergen challenge model (particularly the LAR) is a

TABLE II. Treatment differences in the LAR and EAR with AUC analysis

LAR AUC2-12h EAR AUC0-2h

Roflumilast dose Patients,* n

Point estimate

(95% CI) P value Reduction, %

Point estimate


Roflumilast, 250

mg, vs placebo21 20.148

(20.272 to 20.024)

.0110 27 20.146

(20.248 to 20.044)

.0038 25

Roflumilast, 500 mg,vs placebo

19 20.243

(20.382 to 20.104)

.0009 43 20.179

(20.307 to 20.050)

.0046 28

Roflumilast, 500 mg,vs roflumilast, 250 mg

19 20.084

(20.223 to 0.056)

.1113 21 20.015

(20.118 to 0.089)

.3851 3

AUC2-12hr, Area under the curve for FEV1 between 2 and 12 hours; AUC0-2hr, area under the curve for FEV1 between 0 and 2 hours.

*This summarizes the per-protocol population. Data for the comparisons between 500 mg of roflumilast versus placebo or 500 mg of roflumilast versus 250 mg

of roflumilast were only available for 19 patients.

Test.Reference.




Asthmadiagnosisand

treatm

ent

validated technique used to reproduce amplified asthmaticairway inflammation and has previously been used toassess the influence of anti-inflammatory medications onthe EAR and LAR.20 The significant inflammatory com-ponent of the LAR has been attributed to an influx ofactivated eosinophils and the release of cytokines andchemokines that lead to the recruitment of additionalinflammatory cells, such as basophils, lymphocytes, andmonocytes. In a similar challenge model montelukast andbudesonide demonstrated a partial reduction of inflam-mation, as judged by changes in sputum inflammatorycells; however, budesonide did not significantly reduce theEAR.21 As a targeted PDE4 inhibitor with anti-inflamma-

tory actions, roflumilast might confer earlier benefitscompared with corticosteroids and more comprehensiveanti-inflammatory benefits compared with a cysteinylleukotriene receptor antagonist.

The objective of this study was to evaluate the effect ofroflumilast by assessing its effect on the EAR and LAR inpatients with allergen-induced asthma. Roflumilast atdaily doses of 250 mg and 500 mg significantly dimini-shed the LAR, confirming attenuation of allergen-inducedresponses to single-dose administration of roflumilastfound in early studies.22 Roflumilast attenuation of theLAR showed a dose-related trend: compared with pla-cebo, 250 mg of roflumilast resulted in a mean attenuation

TABLE III. Treatment differences in the LAR and EAR with maximum decrease analysis*

LAR EAR

Roflumilast dose

Point estimate


Point estimate



4.712 (0.427 to 8.997) .0321 17 4.015 (0.082 to 7.948) .0457 14


8.901 (4.507 to 13.294) .0002 33 3.909 (20.140 to 7.958) .0580 14

Roflumilast, 500 mg,vs roflumilast 250 mg

4.118 (20.209 to 8.586) .0613 18 20.106 (24.161 to 3.949) .9580 0

*This summarizes the per-protocol population (n = 20, 21, and 21 for patients administered 500 mg of roflumilast, 250 mg of roflumilast, and placebo,

respectively).

Test.Reference.

FIG 2.Mean percentage decrease of FEV1 from postsaline value after allergen challenge. Roflumilast, 250 and

500 mg, significantly attenuated the EAR (0-2 hours) and LAR (2-12 hours) compared with placebo. Data are

presented for the intent-to-treat population.


AUGUST 2005

296 van Schalkwyk et al

Asth

madiagnosis

and

treatm

ent

of 27%, whereas treatment with 500 mg of roflumilastcompared with placebo resulted in a mean attenuation of43%. There was a greater degree of protection againstdecreases in FEV1 during the LAR than the EAR,supported by both AUC and maximum decrease analyses.There was a more modest 14% reduction of the EAR with250 and 500 mg of roflumilast, as determined by means ofmaximum decrease analysis; however, these data supportthe AUC analysis, which is a more comprehensive exam-ination of the entire EAR and LAR. Roflumilast also has amodest effect on the EAR, which is mediated principallyby mast cell degranulation. Corticosteroids do not have asignificant effect on the EAR, which is likely because oftheir lack of effect on mast cell degranulation.21,23 Theearly effect of roflumilast on the EAR might suggest adifferent onset of anti-inflammatory effect than is achievedwith corticosteroids. Although modest, the reduction ofthe EAR by roflumilast might result from mast cellstabilization, mediator release, or both. Roflumilast hadmore pronounced effects on the LAR, and the higher dosereduced responses by almost half. Because of the inflam-matory processes underpinning the LAR, PDE4 inhibitorsmight be particularly suitable to prevent cellular recruit-ment and mediator activity in this phase. In line with theseknown characteristics, roflumilast had a more pronouncedeffect on the LAR than the EAR, possibly because of itsability to reduce inflammation. Roflumilast did not sig-nificantly attenuate AHR, although there was a trendtoward reduction of AHR after allergen.

Roflumilast-induced attenuation of the LAR in thisstudy is similar in magnitude to that achieved with ICStreatment, the primary controller therapy used by patientswith asthma. For example, 200 mg of budesonideattenuates the LAR by approximately 44%,24 and cicle-sonide, a novel ICS under clinical development, attenuatesthe LAR by approximately 47% versus placebo.25 Thus inthis study roflumilast appears to offer anti-inflammatoryeffects comparable with those achieved with ICSs in thismodel. Additionally, long-acting b-agonists do not ex-press anti-inflammatory activity and have been shown toinhibit the LAR after a single dose.26,27 This effect is lostafter multiple doses.28 In this study roflumilast maintainedits inhibitory effect after multiple doses. This addscredence to the hypothesis that the effect on the LARwith roflumilast in this study is more likely caused by anti-inflammatory mechanisms rather than a direct effect on

airway muscle tone. Indeed, previous studies have shownthat roflumilast does not have direct bronchodilatoryeffects.29

Side effects have restricted the use of PDE inhibitors inasthma.20 Because the PDE4 inhibitors are more targeted,they also tend to produce fewer adverse effects, such asnausea and headache. Roflumilast was well tolerated inthis study. Most side effects were mild to moderate inintensity. Headache, diarrhea, and mild nausea were themost common adverse events and appeared to be dosedependent. All were transient and did not result in treat-ment discontinuation. In the current study the frequency ofheadache is likely to be affected by the considerableproportion of patients who reported headache at baseline(26%) and the high incidence of headaches in the placebogroup (18%). The incidence of adverse events in this study,notably headache, was higher than that reported by studiesof the safety and efficacy of roflumilast conducted in largerpopulations.30,31

In conclusion, the current study demonstrates thatroflumilast, a targeted PDE4 inhibitor, inhibits the LAR,which might result from the anti-inflammatory activityof roflumilast. Further clinical studies with surrogatemarkers of inflammation (eg, induced sputum) are neededto confirm the anti-inflammatory effects of roflumilast.Additionally, these data suggest that roflumilast showspromise as an oral, once-daily, steroid-free treatment forasthma. Additional studies to determine the clinicalbenefits of roflumilast in asthma and COPD are needed.

We thank our patients who participated in the study. The nursing

support of Dot Steyn andWendy Lee is acknowledged, and we thank

Marianne Koopman for her expert measurements of pulmonary

function.

REFERENCES

1. Neri M, Spanevello A. Chronic bronchial asthma from challenge to

treatment: epidemiology and social impact. Thorax 2000;55(suppl 2):

S57-8.

2. Adams BK, Cydulka RK. Asthma evaluation and management. Emerg

Med Clin North Am 2003;21:315-30.

3. American Lung Association, Epidemiology and Statistics Unit, Research

and Scientific Affairs. Trends in asthma morbidity and mortality.

Available at: http://www.lungusa.org/atf/cf/7A8D42C2-FCCA-4604-

8ADE-7F5D5E762256/ASTHMA1.PDF. Accessed July 23, 2004.

4. World Health Organization. WHO fact sheet no. 206. Bronchial asthma.

Available at: http://www.who.int/inf-fs/en/fact206.html. Accessed Nov-

ember 14, 2003.

5. National Institutes of Health, National Heart, Lung, and Blood Institute,

Global Initiative for Asthma. Global strategy for asthma management

and prevention. Available at: http://www.ginasthma.com/workshop.pdf.

Accessed December 3, 2004.

6. Dykewicz MS. Newer and alternative non-steroidal treatments for

asthmatic inflammation. Allergy Asthma Proc 2001;22:11-5.

7. Spector S. Noncompliance with asthma therapy—are there solutions?

J Asthma 2000;37:381-8.

8. Adcock IM, Lane SJ. Corticosteroid-insensitive asthma: molecular

mechanisms. J Endocrinol 2003;178:347-55.

9. Essayan DM. Cyclic nucleotide phosphodiesterase (PDE) inhibitors and

immunomodulation. Biochem Pharmacol 1999;57:965-73.

10. Hatzelmann A, Schudt C. Anti-inflammatory and immunomodulatory

potential of the novel PDE4 inhibitor roflumilast in vitro. J Pharmacol

Exp Ther 2001;297:267-79.

TABLE IV. PC20FEV1 ratios* by dose groupy

Placebo

(n = 15)

Roflumilast,

250 mg (n = 18)

Roflumilast,

500 mg (n = 18)

Mean PC20

FEV1 ratio

0.87 1.03 1.11

95% CI 0.55 to 1.36 0.59 to 1.81 0.67 to 1.85

Mean doubling

factor

20.20 0.04 0.15

95% CI 20.85 to 0.45 20.77 to 0.86 20.59 to 0.89

*Ratio of second visit of treatment period versus first visit of treatment period.

This table summarizes PC20FEV1 data for the per-protocol population.




Asthmadiagnosisand

treatm

ent

http://www.lungusa.org/atf/cf/7A8D42C2-FCCA-4604-8ADE-7F5D5E762256/ASTHMA1.PDF

http://www.lungusa.org/atf/cf/7A8D42C2-FCCA-4604-8ADE-7F5D5E762256/ASTHMA1.PDF

http://www.who.int/inf-fs/en/fact206.html

http://www.ginasthma.com/workshop.pdf

11. Bundschuh DS, Eltze M, Barsig J, Wollin L, Hatzelmann A, Beume R. In

vivo efficacy in airway disease models of roflumilast, a novel orally

active PDE4 inhibitor. J Pharmacol Exp Ther 2001;297:280-90.

12. Chai H, Farr RS, Froehlich LA, Mathison DA, McLean JA, Rosenthal

RR, et al. Standardization of bronchial inhalation challenge procedures.


13. Bardin PG, Dorward MA, Lampe FC, Franke B, Holgate ST. Effect of

selective phosphodiesterase 3 inhibition on the early and late asthmatic

responses to inhaled allergen. Br J Clin Pharmacol 1998;45:387-91.

14. Rossi GA, Crimi E, Lantero S, Gianiorio P, Oddera S, Crimi P, et al.

Late-phase asthmatic reaction to inhaled allergen is associated with early

recruitment of eosinophils in the airways. Am Rev Respir Dis 1991;144:

379-83.

15. Niimi A, Amitani R, Yamada K, Tanaka K, Kuze F. Late respiratory

response and associated eosinophilic inflammation induced by repeated

exposure to toluene diisocyanate in guinea pigs. J Allergy Clin Immunol

1996;97:1308-19.

16. Larsen GL, Wilson MC, Clark RA, Behrens BL. The inflammatory

reaction in the airways in an animal model of the late asthmatic response.

Fed Proc 1987;46:105-12.

17. TimmerW, Leclerc V, Birraux G, NeuhauserM, Hatzelmann A, Bethke T,

et al. The new phosphodiesterase 4 inhibitor roflumilast is efficacious

in exercise-induced asthma and leads to suppression of LPS-stimulated

TNF-a ex vivo. J Clin Pharmacol 2002;42:297-303.

18. Hofstra WB, Sont JK, Sterk PJ, Neijens HJ, Kuethe MC, Duiverman EJ.

Sample size estimation in studies monitoring exercise-induced broncho-

constriction in asthmatic children. Thorax 1997;52:739-41.

19. Cockcroft DW. Nonallergic airway responsiveness. J Allergy Clin

Immunol 1988;81:111-9.

20. Torphy TJ. Phosphodiesterase isozymes: molecular targets for novel

antiasthma agents. Am J Respir Crit Care Med 1998;157:351-70.

21. Leigh R, Vethanayagam D, Yoshida M, Watson RM, Rerecich T, Inman

MD, et al. Effects of montelukast and budesonide on airway responses

and airway inflammation in asthma. Am J Respir Crit Care Med 2002;

166:1212-7.

22. Nell H, Louw C, Leichtl S, Rathgeb F, Neuhauser M, Bardin PG. Acute

anti-inflammatory effect of the novel phosphodiesterase 4 inhibitor

roflumilast on allergen challenge in asthmatics after a single dose

[abstract]. Am J Respir Crit Care Med 2000;161(suppl):A200.

23. Palmqvist M, Bruce C, Sjostrand M, Arvidsson P, Lotvall J. Differential

effects of fluticasone and montelukast on allergen-induced asthma.

Allergy 2005;60:65-70.

24. Kidney JC, Boulet LP, Hargreave FE, Deschesnes F, Swystun VA,

O’Byrne PM, et al. Evaluation of single-dose inhaled corticosteroid

activity with an allergen challenge model. J Allergy Clin Immunol 1997;

100:65-70.

25. Larsen BB, Nielsen LP, Engelstatter R, Steinijans V, Dahl R. Effect of

ciclesonide on allergen challenge in subjects with bronchial asthma.

Allergy 2003;58:207-12.

26. Pizzichini MMM, Kidney JC, Wong BJO, Morris MM, Efthimiadis A,

Dolovich J, et al. Effect of salmeterol compared with beclomethasone on

allergen-induced asthmatic and inflammatory responses. Eur Respir J

1996;9:449-55.

27. Weersink EJ, Aalbers R, Koeter GH, Kauffman HF, De Monchy JG,

Postma DS. Partial inhibition of the early and late asthmatic response by a

single dose of salmeterol. Am J Respir Crit Care Med 1994;150:1262-7.

28. Dente FL, Bacci E, Bartoli ML, Cianchetti S, Di Franco A, Giannini D,

et al. One week treatment with salmeterol does not prevent early and late

asthmatic responses and sputum eosinophilia induced by allergen

challenge in asthmatics. Pulm Pharmacol Ther 2004;17:147-53.

29. Engelstatter R, Wingertzahn M, Schmid-Wirlitsch C, Leichtl S,

Bredenbroker D, Wurst W. Roflumilast, an oral, once-daily phosphodi-

esterase 4 (PDE4) inhibitor, does not exhibit bronchodilatory activity.

Presented at: 2004 ACAAI Meeting; November 12-17, 2004; Boston,

Mass.

30. Leichtl S, Schmid-Wirlitsch C, Bredenbroker D, Rathgeb F, Wurst W.

Roflumilast, a new, orally active, selective phosphodiesterase 4 inhibitor,

is effective in the treatment of asthma [abstract]. Eur Respir J 2002;

20(suppl 38):303s.

31. Izquierdo JL, Bateman ED, Villasante C, Schmid-Wirlitsch C,

Bredenbroker D, Wurst W. Long-term efficacy and safety over one

year of once-daily roflumilast, a new, orally active, selective, phospho-

diesterase 4 inhibitor, in asthma [abstract]. Am J Respir Crit Care Med

2003;167:A765.

Correction

With regard to the May 2005 article entitled ‘‘Physical activity and exercise in asthma: Relevance to

etiology and treatment’’ (2005;115:928-34): In the abstract, the seventh sentence should have appeared as

follows:

The allergy community has placed emphasis on medical therapy and allergen avoidance; inaddition, exercise has not been formally incorporated into the National Asthma Education andPrevention Program guidelines.


AUGUST 2005

298 van Schalkwyk

Asth

madiagnosis

and

treatm

ent

Duration of postviral airwayhyperresponsiveness in childrenwith asthma: Effect of atopy

Paraskevi Xepapadaki, MD, PhD,* Nikolaos G. Papadopoulos, MD, PhD,* Apostolos

Bossios, MD, PhD, Emmanuel Manoussakis, MD, Theodoros Manousakas, MD, and

Photini Saxoni-Papageorgiou, MD, PhD Athens, Greece

Background: Respiratory viruses induce asthma

exacerbations and airway hyperresponsiveness (AHR). Atopy

is an important risk factor for asthma persistence.

Objective: We sought to evaluate whether atopy is a risk

factor for prolonged AHR after upper respiratory tract

infections (URIs).

Methods: Twenty-five children (13 atopic and 12 nonatopic

children) with intermittent virus-induced asthma were

studied. Clinical evaluation, skin prick tests, methacholine

bronchoprovocation, questionnaires, and a nasal wash

specimen were obtained at baseline. For 9 months, subjects

completed diary cards with respiratory symptoms. During their

first reported cold, a nasal wash specimen was obtained.

Methacholine provocation was performed 10 days and 5, 7, 9,

and 11 weeks later. In case a new cold developed, the

provocation schedule was followed from the beginning.

Results: Viruses were detected in 17 (68%) of 25 patients

during their first cold, with rhinovirus being most commonly

identified (82%). AHR increased significantly 10 days after the

URI, equally in both groups (P = .67), and remained so up to

the fifth week. Duration of AHR in subjects experiencing a

single URI ranged from 5 to 11 weeks, without a significant

difference between groups. In the duration of the study, atopic

children experienced more colds and asthma exacerbations

than nonatopic children. Thus for duration of AHR,

significant prolongation was noted in the atopic group when

assessed cumulatively.

Conclusion: In asthmatic children the duration of AHR after

a single natural cold is 5 to 11 weeks. However, an increased

rate of symptomatic cold and asthma episodes in atopic children

is associatedwith considerable cumulative prolongation of AHR,

which might help explain the role of atopy as a risk factor for

asthma persistence. (J Allergy Clin Immunol 2005;116:

299-304.)

Key words: Asthma, airway hyperresponsiveness, atopy

Asthma is characterized by abnormalities in lungfunction, variable airway obstruction, and airway hyper-responsiveness (AHR)1; among others, a significant cor-relation exists between the degree of AHR and bothclinical severity and medication needs for asthma.2 Anumber of stimuli, including respiratory viruses, are ableto induce AHR.3 Increased AHR to histamine in healthysubjects after upper respiratory tract infections (URIs)lasting up to 6 weeks was observed more than 20 yearsago.4 Human experimental infections with human rhino-viruses confirmed that increased AHR to nonspecificstimuli is observed for up to 4 weeks after a viral infectionin allergic subjects.5-7 Furthermore, AHR after viralinfections has been documented in animal models ofparamyxovirus, respiratory syncytial virus, and influenzavirus infections.8-10 Most asthma exacerbations in chil-dren are associated with URIs, attributed in their majorityto rhinoviruses.11,12 Virus-induced AHR is increased inatopic individuals compared with in healthy controlsubjects.5,13 Moreover, atopy is one of the strongest riskfactors for the development and persistence of asthma,especially in childhood.14 More than 60% of asthmaticchildren are atopic, whereas the presence of atopy at theage of 12 months increases 3 to 4 times the risk for asthmapersistence during later childhood and adulthood.15,16

There are suggestions that genes responsible for atopy andAHR act together to develop the full asthma phenotype.17

Furthermore, it has been proposed that the defectiveepithelial repair cycle, which is characteristic of asthmaand strongly correlates to AHR, is amplified by exposureto TH2 cytokines.

18 Nevertheless, our current understand-ing of the mechanisms connecting atopy, AHR, andasthma is still incomplete. The classical model of sensi-tization and exposure to particular allergens19 can onlypartly explain the epidemiologic observations. Morerecently, we and others have shown that the response ofatopic asthmatic individuals to rhinovirus infection is

Abbreviations usedAHR: Airway hyperresponsiveness

NW: Nasal wash


URI: Upper respiratory tract infection

From the Allergy Unit, 2nd Pediatric Clinic, University of Athens.

*Drs Xepapadaki and Papadopoulos contributed equally to this work.

Disclosure of potential conflict of interest: N. Papadopoulos has received

grants–research support from GlaxoSmithKline and AstraZeneca.

P. Saxoni-Papageorgiou has received grants–research support from

Novartis and Schering-Plough. All other authors—none disclosed.

Received for publication November 11, 2004; revised March 28, 2005;

accepted for publication April 4, 2005.

Available online May 24, 2005.

Reprint requests: Paraskevi Xepapadaki, MD, PhD, UPC Research

Laboratories, 13, Levadias 11527, Goudi, Greece. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.007

299

Asthmadiagnosisand

treatm

ent


defective, with a suboptimal TH1 response and a shifttoward a TH2 phenotype.

20,21 Such a response might leadto incomplete viral clearance and inflammation persis-tence, potentially perpetuating AHR.22

On the basis of the above, we hypothesized that atopycould be a risk factor for prolongation of AHR afterrespiratory viral infections. The aim of this study was toprospectively evaluate the duration of AHR in childrenwith asthma after naturally occurring respiratory infectionsand assess the role of atopy as a risk factor in this respect.

METHODS

Patients

Twenty-five children (16 boys), 7 to 12 years of age, with

intermittent asthma and a history of disease exacerbations attributable

to URI only, were recruited for the study. Asthma diagnosis was

based on history, 12% or greater improvement in FEV1 after

bronchodilation, and AHR (PC20 16 mg/mL). Classification of

asthma severity was performed according to the Global Initiative for

Asthma.23 Atopy was evaluated by using skin prick tests (SPTs) to a

panel of 18 locally relevant allergens performed outside the pollen

season. Exclusion criteria were a history of seasonal or allergen-

driven symptoms, current or previous use of specific immunotherapy,

use of inhaled steroids within the previous 2 months, recent (<2

months) URI, and chronic conditions potentially affecting airway

responsiveness.24 All parents provided written consent, and the study

was approved by the hospital’s ethics committee.

Study design

A prospective case-control design with 9 months’ follow-up

(September-June) was used. Clinical evaluation, IgE determination,

SPT, and methacholine bronchial provocation were performed at

baseline. Questionnaires and a nasal wash (NW) specimen were also

obtained. Patients were grouped according to the presence of atopy

(with n = 13). Children (or their parents) were asked to prospectively

complete diary cards with upper and lower respiratory tract

symptoms.11 During the first reported URI, judged either subjectively

or by means of increased daily symptom scores (4), an NW was

performed. Methacholine provocation was performed 10 days, and

5, 7, 9, and 11 weeks later. In case a new cold developed within this

time period, the above provocation schedule was followed from

the beginning. Bronchodilators were used as relievers; in case

of persistence or deterioration of asthma symptoms, subjects were

instructed to receive systemic steroids, although these were not

needed on any occasion.

Methacholine provocation

Methacholine provocation was performed with the 2-minute

breathing dosing protocol.25 The aerosols were generated with a

nebulizer output of 0.13 mL/min (Air Dynaval Taema). The proce-

dure was discontinued if FEV1 decreased 20% or greater from

baseline values or when a 16 mg/mL concentration had been

administered.26 Spirometrywas performed according to theAmerican

Thoracic Society guidelines.27

Virus detection

RT-PCR was used for detection of viral RNA in NW samples, as

previously described.28 PCR reactions were performed for rhinovi-

ruses; enteroviruses; respiratory syncytial virus; coronaviruses OC43

and 229E; influenza viruses A and B; parainfluenza viruses 1, 2,

and 3; adenoviruses; Chlamydia pneumoniae; and Mycoplasmapneumoniae.29,30


Comparisons of baseline characteristics were performed with the

Fisher exact test for categoric variables and t tests for continuous

variables. A common cold and an asthma exacerbation were defined

as an increase of the relevant symptom score totals over the personal

median value for 2 days or more. The duration of each episode was

defined as the time from symptom initiation to return to the median

value. Severity of each was defined as the maximum symptom score.

The Wilcoxon signed-rank test and the Mann-Whitney U test were

used to compare independent and paired outcome variables, respec-

tively. The time to PC20 restoration was displayed in Kaplan-Mayer

curves compared with the log-rank test. P values of less than .05 were

considered significant.

RESULTS

Baseline characteristics

Table I shows the demographic, socioeconomic, anddisease characteristics of the 2 groups, as obtained atbaseline. Atopic asthmatic children had statistically sig-nificantly higher levels of total IgE (P = .04). A trendtoward more cases of positive family history of atopy wasalso noted in the atopic group (P = .08). In respect to allremaining characteristics, the groups were homogeneous,with no significant differences when statistically com-pared. Baseline spirometric values did not differ betweenthe 2 groups (Table I).

Detection of respiratory viruses with PCR

All NW specimens obtained at baseline were negativefor the presence of respiratory viruses. PCR revealed thepresence of a virus in 17 (68%) of 25 patients during theirsubsequent first URI. The most commonly identified viruswas rhinovirus (14/17 [82%]). Adenovirus was found in 4(23%) of 17 positive samples, whereas one child had bothrhinovirus and adenovirus. Virus identification rates didnot differ significantly between groups (nonatopic, 58%;atopic, 77%; P = .34).

Duration of AHR

All subjects completed the study. Methacholine re-sponsiveness at baseline was slightly, but not signifi-cantly, higher in atopic individuals (time = 0, atopicPC20 = 5.9 6 1.1 mg/mL, nonatopic PC20 = 8.4 6 1.5mg/mL; P = .18). All subjects experienced at least onenatural cold within the study period. AHR increasedsignificantly 10 days after the first reported URI, equallyin both groups (time = 10th day, atopic PC20 = 2.3 6 1.1mg/mL, nonatopic PC20 = 2.6 6 0.9 mg/mL; P = .002 incomparison with their respective baseline values in bothcases and P = .67 between groups). This increase re-mained statistically significant up to the fifth week afterthe onset of the cold, progressively decreasing, althoughstill without statistically significant differences, betweenthe groups (time = fifth week, atopic PC20 = 3.7 6 1.3mg/mL, nonatopic PC20 = 4.7 6 1.3 mg/mL; P = .005and P = .021, respectively, in comparison with baselineand P = .29 between groups).

The above analysis was performed in subjects withno additional URI during that period (time = 0,


AUGUST 2005

300 Xepapadaki et al

Asth

madiagnosis

and

treatm

ent

n = 25; time = 10 days, n = 24; time = 5 weeks, n = 23).After the fifth week, several children experienced addi-tional colds, and therefore provocations were rescheduledaccording to the study design; comparisons were not doneon fixed time points. Eleven weeks after the initial URI,10 children (3 atopic and 7 nonatopic children) had nofurther cold, and their methacholine responsiveness hadreturned to their respective baseline value. The meanduration of AHR in these subjects was 7.0 6 2.0 weeks(median, 5 weeks; range, 5-11 weeks) for the atopicchildren and 7.3 6 1.0 weeks (median, 7 weeks; range,5-11 weeks) for the nonatopic children. In addition, usinga simple linear regression model based on AHR valuesfrom the 10th day on and including all subjects with asingle URI, the predicted time for AHR to return to itsrespective baseline value was 5.6 to 8.9 weeks for theatopic children and 6.7 to 10.2 weeks for the nonatopicchildren (Fig 1).

However, important differences, as shown in Fig 2,were observed when the cumulative duration of AHR wasassessed prospectively in the complete cohort. All 12nonatopic asthmatic children returned to their baselinePC20 values by day 120, after the first reported URI,

whereas only 9 (67%) of 13 of the atopic children returnedto their baseline PC20 value by day 200 (P = .0068),showing that in assessing the natural history of the disease,increased airway responsiveness, possibly caused byrepetitive colds, is considerably more prolonged in atopicasthmatic children and might in fact last more than6 months.

Common cold and asthma exacerbationcharacteristics

Atopic children had generally more disease episodesthan nonatopic children (Fig 3). The average number ofcolds that each subject experienced during thewhole studyperiod was marginally higher in the atopic group (atopicgroup, 4.6 6 0.4; nonatopic group, 3.5 6 0.4; P = .06).The average duration of each cold did not differ betweengroups (atopic group, 10.4 6 1.8 days; nonatopic group,9.46 1.3 days; P = .81), as was the case for cold severity(atopic group, 2.2 6 0.2; nonatopic group, 2.9 6 0.3;P = .12).

Atopic asthmatic children experienced significantlymore exacerbations during the study period (5.1 6 0.6vs 3.2 6 0.3 of the nonatopic children, P = .01). Therewere no differences in the duration or severity of asthmaexacerbations between the groups (duration: atopic group,

TABLE I. Patient characteristics: Comparison of baseline

characteristics between atopic and nonatopic asthmatic

children included in the study

Baseline

characteristics

Nonatopic

(n = 12)

Atopic

(n = 13) P value

Age (y) 10.4 6 1.8 10.3 6 1.2 NS

Sex (boys) 50% 77% NS

IgE (mg/dL) 78.9 6 14.5 339.4 6 117.1 .04

Residence (urban) 66% 54% NS

Density

(persons per room)

1.5 1.2 NS

Pet ownership 12% 13% NS

Exposure to tobacco

smoke

75% 62% NS

School type (public) 83% 85% NS

Paternal education

level (secondary

education)

73% 69% NS

Maternal education

level (secondary

education)

58% 54% NS

Father or mother with

atopy

60% 85% .08

Colds in the previous

year (n)

3.5 3.6 NS

Asthma exacerbations

in the previous year (n)

3.3 3.3 NS

Emergency visits for

asthma in the previous

year (n)

2.3 2.7 NS

Hospitalizations for

asthma in the previous

year (n)

0.1 0 NS

FVC (% mean) 6 SD 92.9 6 11.3 88.5 6 9.3 NS

FEV1 (% mean) 6 SD 81.4 6 6.7 78.2 6 7.4 NS

FVC, Forced vital capacity.

FIG 1. Linear regression model of AHR changes from the 10th day

to the seventh week after a single URI in atopic (A) and nonatopic

(B) children with intermittent virus-induced asthma. A significant

and time-dependent AHR decrease is observed in both cases. The

predicted time for AHR to return to baseline did not differ between

the groups.



Xepapadaki et al 301

Asthmadiagnosisand

treatm

ent

8.5 6 0.9 days; nonatopic group, 11.4 6 2.3 days;P = .65; severity: atopic group, 4.4 6 0.8; nonatopicgroup, 4.6 6 0.9; P = .60).

No statistically significant differences were found ineither the duration or severity of symptom scores betweenpatients with positive and negative virus identification(data not shown).

All atopic children had positive SPT responses tocommon pollen allergens (with a wheal of 3 mm).31

One atopic child also presented with a positive SPTresponse to mites. To address the question of whetherprolonged AHR in atopic children might be affected byexposure to allergens to which they were sensitized, wecompared the characteristics of asthma exacerbationsduring the pollen season (March-June) and the cold season(October-January). There were no differences in any suchcharacteristics within the nonatopic group (Table II).However, within the atopic group, the number of asthmaexacerbations was significantly higher during the coldseason, suggesting that exposure to allergens could notaccount for additional disease burden in these patients(Table II).

DISCUSSION

This is the first study to prospectively evaluate theduration of AHR in children after naturally occurringcolds for an extended period of time. Two major findingsare reported: (1) the duration of postviral nonspecific AHRis considerably more prolonged than previously thought,and (2) the duration of AHR after a single cold is the samein atopic and nonatopic children; however, an increasednumber of symptomatic colds cumulatively lead theformer to prolonged AHR.

Most previous studies assessing the duration of AHRafter viral infections have used human rhinovirusexperimental infections, which usually result in mild-to-moderate colds and very mild, if any, asthma exacerba-tions.32 Furthermore, fixed time points for up to 8 weeksafter infection have been used for outcome evaluation.7,33

In this respect these studies excluded, to a considerableextent, the possibility of interference from additionalinfections but on the other hand could not evaluate theduration of virus-induced AHR in real-life conditions. Inour study, when the duration of AHR was assessed after asingle infection, either in the subgroup of subjects who didnot have additional infection for up to 11 weeks, or bymeans of a simple regression model, it was 7 weeks onaverage, ranging from 5 to 11 weeks. This time span isstill longer than previously reported in human subjects(10 days to 6 weeks)4,34 and comparable with time spansreported studies prospectively evaluating postviral AHRin experimental animals (2-22 weeks).8,10,35,36 Neverthe-less, when AHR was assessed in the long term aftermultiple colds, a considerable proportion of atopic sub-jects remained hyperresponsive for considerably longer,in some cases more than 6 months. This is consistent withthe notion that additional, naturally occurring colds mightfurther prolong AHR, as recently shown in an experimen-tal animal setting,37 and might prove important in consid-ering the natural history of the disease, as well asmanagement of such patients. Therefore although ourhypothesis that atopy might lead to postviral AHRprolongation was shown not to be directly true becausethere was no difference in AHR duration after a singleinfection, the total duration of AHR was in fact

FIG 2. Kaplan-Meyer diagram of PC20 return to its respective

baseline value after one or more naturally occurring URIs during

a 9-month period in atopic and nonatopic asthmatic children. The

difference is significant (P = .0068, survival analysis).

FIG 3. Upper respiratory tract symptom exacerbations, colds, and

asthma exacerbations experienced during a 9-month period in

atopic (open bars) and nonatopic (filled bars) children with inter-

mittent asthma. *P < .05, #P = .06.

TABLE II. Seasonal distribution of asthma exacerbation

characteristics in atopic and nonatopic asthmatic children

Number Duration (d) Severity (index)

Nonatopic

Winter 1.50 (0.55) 10.17 (15.21) 1.43 (0.55)

Spring 2.00 (1.26) 4.67 (1.63) 1.10 (0.33)

P value .52 .79 .29

Atopic

Winter 2.67 (1.50) 9.25 (4.80) 1.52 (0.53)

Spring 1.42 (0.67) 7.08 (4.89) 1.31 (0.35)

P value .02 .15 .24

Reported values are presented as means (SD).

Winter, October through January; spring, March through June.


AUGUST 2005


Asth

madiagnosis

and

treatm

ent

significantly longer in atopic than nonatopic children andthat was associated with an increased number of colds, aswell as asthma episodes.

Patient characteristics did not differ between the groupsat baseline; it cannot be excluded, however, that subclin-ical inflammation might have been greater in atopicsubjects, which were slightly, but not significantly, morehyperresponsive at baseline.

Several studies have shown that viral infection inducesgreater changes in nonspecific AHR in patients withrespiratory allergy than in healthy control subjects.5,7,38,39

Accordingly, it has been assumed that after a viralinfection, the response to allergen exposure might beexaggerated, a notion elegantly confirmed by usingsegmental bronchial provocations with allergen.40,41

Furthermore, clinical studies have shown synergy be-tween viral infection and allergen exposure in sensitizedindividuals.42,43 Nevertheless, other studies have shownthat allergen exposure does not necessarily augment virus-mediated responses.44,45 This might also be the case in theatopic population of our study, who, even though theywere sensitized to pollen allergens, had more symptomsduring the fall and winter months rather than during thepollen season. This was not completely unexpectedbecause all our subjects were selected for a postinfectiousasthma phenotype during the 2 previous years, apparentlyreporting symptoms only after colds and not after allergenexposure. Certainly, the possibility that unidentifiedallergens might have contributed to some exacerbationscannot be completely excluded. However, the fact that theidentified allergens did not seem to influence the outcome,although the number of colds and subsequent asthmaexacerbations during the study was higher in atopicchildren, suggests that an increased susceptibility to viralinfection might be a more plausible mechanism forprolongation of AHR in that group. Previous evidencesuggests that the immune response to rhinovirus is defec-tive in atopic asthmatic individuals, with reduced IFN-gproduction,20,46 which might be associated with increasedsusceptibility to symptomatic virus infections. In fact,rhinovirus-induced IFN-g production is strongly associ-ated with AHR in patients with asthma.21

There is a longstanding speculation that allergic sub-jects, asthmatic subjects, or both have more respiratoryinfections than the healthy population.47 Most studieshave compared respiratory symptoms after URI in atopicand healthy individuals. In a recent longitudinal cohortstudy, Corne et al48 showed that subjects with atopicasthma are not at greater risk of a rhinovirus infection thanhealthy individuals but have more frequent, severe, andlonger-lasting lower respiratory tract symptoms.48 On theother hand, even within an atopic population, the outcomeof an experimental rhinovirus infection is closely associ-ated with levels of TH1 and TH2 cytokines.

46 It is possiblethat any effects of atopy on the susceptibility to respi-ratory viral infections might differ between adults andchildren, in which cytokine responses, including IFN-gresponses, do not mature before late childhood.49,50

Furthermore, recent evidence suggests that primary epi-

thelial cells derived from atopic subjects are more suscep-tible to ex vivo rhinovirus infection than cells from healthycontrol subjects.51

A weakness of the present study is that PCR viralidentification was performed only for the first reportedcold and not for subsequent colds. Nevertheless, it iswithout doubt that respiratory viral infections are thecause of the common cold, whereas the association ofcolds with subsequent asthma exacerbations is also wellestablished.11,12

The degree of airway responsiveness is indicative ofasthma severity and counts as an indirect marker of airwayinflammation.2,52 In this respect prolongation of virus-induced AHR might well reflect persistent airway inflam-mation after multiple insults. Perpetuation of subclinicalairway inflammation could have a substantial effect on therisk of asthma persistence, relapse later in life, or both,providing a possible mechanism for the well-establishedrole of atopy as a major risk factor for the persistence ofasthma and AHR from childhood to adulthood.53,54

In conclusion, the duration of AHR in children withintermittent virus-induced asthma ranges from 5 to 11weeks after a single infection but might be considerablyprolonged with repeated infections. Atopic subjects pre-sent with more symptomatic colds and thus cumulativelyhave significantly more prolonged AHR. Prolongation ofvirus-induced AHR in atopic subjects might help explainthe well-established role of atopy as a risk factor forasthma persistence.

REFERENCES

1. Lowe L, Custovic MS, Woodcock A. Childhood asthma. Curr Allergy

Asthma Rep 2004;4:159-65.

2. Weiss ST, Van Natta ML, Zeiger RS. Relationship between increased

airway responsiveness and asthma severity in the childhood asthma

management program. Am J Respir Crit Care Med 2000;162:50-6.

3. Louis-Philippe B. Physiopathology of airway hyperresponsiveness.

Curr Allergy Asthma Rep 2003;3:166-71.

4. Empey DW, Laitinen LA, Jacobs L, Gold WM, Nadel JA. Mechanisms

of bronchial hyperreactivity in normal subjects after upper respiratory

tract infection. Am Rev Respir Dis 1976;113:131-9.

5. Gern JE, Calhoun W, Swenson C, Shen G, Busse WW. Rhinovirus

infection preferentially increases lower airway responsiveness in allergic

subjects. Am Respir Crit Care Med 1997;155:1872-6.

6. Grunberg K, Kuijpers EA, de Klerk EP, de Gouw HW, Kroes AC,

Dick EC, et al. Effects of experimental rhinovirus 16 infection on

airway hyperresponsiveness to bradykinin in asthmatic subjects in vivo.


7. Cheung D, Dick EC, Timmers MC, de Klerk EP, Spaan WJ, Sterk PJ.

Rhinovirus inhalation causes long-lasting excessive airway narrowing in

response to methacholine in asthmatic subjects in vivo. Am J Respir Crit

Care Med 1995;152:1490-6.

8. Jafri HS,Chavez-Bueno S,MejiasA,GomezAM,RiosAM,Nassi SS, et al.

Respiratory syncytial virus induces pneumonia, cytokine response,

airway obstruction, and chronic inflammatory infiltrates associated with

long-term airway hyperresponsiveness in mice. J Infect Dis 2004;189:

1856-65.

9. Folkerts G, Busse WW, Nijkamp FP, Sorkness R, Gern JE. Virus

induced airway hyperresponsiveness and asthma. Am J Respir Crit Care

Med 1998;157:1708-20.

10. Walter MJ, Morton JD, Kajiwara N, Agapov E, Holtzman MJ. Viral

induction of a chronic asthma phenotype and genetic segregation from

the acute response. J Clin Invest 2002;110:165-75.



Xepapadaki et al 303

Asthmadiagnosisand

treatm

ent

11. Johnston SL, Pattemore PK, Sanderson G, Smith S, Lampe F, Josephs L,

et al. Community study of role of viral infections in exacerbations of

asthma in 9-11 year old children. BMJ 1995;310:1225-9.

12. Papadopoulos NG, Psarras S, Manoussakis E, Saxoni-Papageorgiou P.

The role of respiratory viruses in the origin and exacerbations of asthma.

Curr Opin Allergy Clin Immunol 2003;3:39-44.

13. Bardin PG, Sanderson G, Robinson BS, Holgate ST, Tyrrell DA.

Experimental rhinovirus infection in volunteers. Eur Respir J 1996;9:

2250-5.

14. Christodoulopoulos P, Cameron L, Nakamura Y, Lemiere C, Muro S,

Dugas M, et al. TH2 cytokine-associated transcription factors in atopic

and nonatopic asthma: evidence for differential signal transducer and

activator of transcription 6 expression. J Allergy Clin Immunol 2001;

107:586-91.

15. Martinez FD. Links between pediatric and adult asthma. J Allergy Clin

Immunol 2001;107(suppl):S449-55.

16. Warner JO. Bronchial hyperresponsiveness, atopy, airway inflammation

and asthma. Pediatr Allergy Immunol 1998;9:55-6.

17. Postma DS, Koppelman GH, Meyers DA. The genetics of atopy and

airway hyperresponsiveness. Am J Respir Crit Care Med 2000;162

(suppl):S118-23.

18. Davies DE, Wicks J, Powell RM, Puddicombe SM, Holgate ST. Airway

remodeling in asthma: new insights. J Allergy Clin Immunol 2003;111:

215-26.

19. Nelson HS, Szefler SJ, Jacobs J, Huss K, Shapiro G, Sternberg AL.

The relationships among environmental allergen sensitization, allergen

exposure, pulmonary function, and bronchial hyperresponsiveness in

the Childhood Asthma Management Program. J Allergy Clin Immunol

1999;104:775-85.

20. Papadopoulos NG, Stanciu LA, Papi A, Holgate ST, Johnston SL.

A defective type 1 response to rhinovirus in atopic asthma. Thorax 2002;

57:328-32.

21. Brooks GD, Buchta KA, Swenson CA, Gern JE, Busse WW. Rhinovirus-

induced interferon-g and airway responsiveness in asthma. Am J Respir

Crit Care Med 2003;168:1091-4.

22. Papadopoulos NG, Stanciu LA, Papi A, Holgate ST, Johnston SL.

Rhinovirus-induced alterations on peripheral blood mononuclear cell

phenotype and costimulatory molecule expression in normal and atopic

asthmatic subjects. Clin Exp Allergy 2002;32:537-42.

23. Global Strategy for Asthma Management and Prevention. Bethesda

(MD): National Institutes of Health, National Heart, Lung and Blood

Institute; 2002. Publication no. 02-3659.

24. Cain H. Bronchoprovocation testing. Clin Chest Med 2001;22:651-9.

25. American Thoracic Society. Guidelines for methacholine and exercise

challenge testing-1999. Am J Respir Crit Care Med 2000;161:309-29.

26. Cockcroft DW. How best to measure airway responsiveness. Am J


27. American Thoracic Society. Standarization of spirometry: 1994 update.


28. Papadopoulos NG, Sanderson G, Hunter J, Johnston SL. Rhinovirus

replicate effectively at lower airway temperatures. J Med Virol 1999;58:

100-4.

29. Papadopoulos NG, Hunter J, Sanderson G, Meyer J, Johnston SL.

Rhinovirus identification by BglI digestion of picornavirus RT-PCR

amplicons. J Virol Methods 1999;80:179-85.

30. Papadopoulos NG, Moustaki M, Tsolia M, Bossios A, Astra E,

Prezerakou A, et al. Association of rhinovirus infection with increased

disease severity in acute bronchiolitis. Am Respir Crit Care Med 2002;

165:1285-9.

31. Sub-Committee on Skin Tests of the European Academy of Allergology

and Clinical Immunology. Skin tests used in type I allergy testing

[position paper]. Allergy 1989;44 Suppl 10:1-59.

32. Skoner DP, Doyle WJ, Seroky J, Fireman P. Lower airway responses to

influenza A virus in healthy allergic and nonallergic subjects. Am J


33. Fleming HE, Little FF, Schnurr D, Avila PC, Wong H, Liu J, et al.

Rhinovirus-16 colds in healthy and in asthmatic subjects: similar

changes in upper and lower airways. Am J Respir Crit Care Med 1999;

160:100-8.

34. Grunberg K, Timmers MC, Smits HH, de Klerk EP, Dick EC, Spaan WJ,

et al. Effect of experimental rhinovirus 16 colds on airway hyper-

responsiveness to histamine and interleukin-8 in nasal lavage in asth-

matic subjects in vivo. Clin Exp Allergy 1997;27:36-45.

35. Schwarze J, Gelfand EW. Respiratory viral infections as promoters of

allergic sensitization and asthma in animal models. Eur Respir J 2002;19:

341-9.

36. Folkerts G, Nijkamp FP. Virus-induced airway hyperresponsiveness.

Role of inflammatory cells and mediators. Am J Respir Crit Care Med

1995;151:1666-74.

37. Matsuse H, Behera AK, Kumar M, Rabb H, Lockey RF, Mohapatra SS.

Recurrent respiratory syncytial virus infections in allergen-sensitized

mice lead to persistent airway inflammation and hyperresponsiveness.

J Immunol 2000;164:6583-92.

38. Trigg CJ, Nicholson KG, Wang JH, Ireland DC, Jordan S, Duddle JM,

et al. Bronchial inflammation and the common cold: a comparison of

atopic and non-atopic individuals. Clin Exp Allergy 1996;26:665-76.

39. Sears MR, Burrows B, Herbison GP, Holdaway MD, Flannery EM.

Atopy in childhood. II. Relationship to airway responsiveness, hay fever

and asthma. Clin Exp Allergy 1993;23:949-56.

40. Lemanske RF, Dick EC, Swenson CA, Vrtis RF, Buse WW. Rhinovirus

upper respiratory tract infection increases airway hyperreactivity and late

asthmatic reactions. J Clin Invest 1989;83:1-10.

41. Calhoun WJ, Dick EC, Schwartz LB, Busse WW. A common cold

virus, rhinovirus 16, potentiates airway inflammation after segmental

antigen bronchoprovocation in allergic subjects. J Clin Invest 1994;94:

2200-8.

42. Green RM, Custovic A, SandersonG, Hunter J, Johnston SL,Woodcock A.

Synergism between allergens and viruses and risk of hospital admission

with asthma: case-control study. BMJ 2002;324:763-8.

43. Papadopoulos NG, Psarros F, Manoussakis E, Hatzipsalti M, Bossios A,

Syrigou E, et al. Association of respiratory viral infections with disease

exacerbations in patients with allergen-driven seasonal asthma and hay

fever [abstract]. Allergy 2002;57:295.

44. Avila PC. Interactions between allergic inflammation and respiratory

viral infections. J Allergy Clin Immunol 2000;106:829-31.

45. de Kluijver J, Evertse CE, Sont JK, Schrumpf JA, van Zeijl-van der

Ham CJ, Dick CR, et al. Are rhinovirus-induced airway responses in

asthma aggravated by chronic allergen exposure? Am J Respir Crit Care

Med 2003;168:1174-80.

46. Parry DE, Busse WW, Sukow KA, Dick CR, Swenson C, Gern JE.

Rhinovirus-induced PBMC responses and outcome of experimental

infection in allergic subjects. J Allergy Clin Immunol 2000;105:692-8.

47. Busse WW, Gern JE. Do allergies protect against the effects of a

rhinovirus cold? J Allergy Clin Immunol 2000;105:889-91.

48. Corne JM, Marshall C, Smith S, Schreiber J, Sanderson G, Holgate ST,

et al. Frequency, severity, and duration of rhinovirus infections in

asthmatic and non-asthmatic individuals: a longitudinal cohort study.

Lancet 2002;359:831-4.

49. Prescott SL. Allergy: when does it begin and where will it end? Allergy

2003;58:864-7.

50. Smart JM, Kemp AS. Ontogeny of T-helper 1 and T-helper 2 cytokine

production in childhood. Pediatr Allergy Immunol 2001;12:181-7.

51. Wark PA, Johnston SL, Bucchieri F, Powell R, Puddicombe S,

Laza-Stanca V, et al. Asthmatic bronchial epithelial cells have a de-

ficient innate immune response to infection with rhinovirus. J Exp

Med 2005;201:937-47.

52. Boulet LP. Asymptomatic airway hyperresponsiveness: a curiosity or an

opportunity to prevent asthma? Am J Respir Crit Care Med 2003;167:

371-8.

53. Arruda LK, Sole D, Baena-Cagnani CE, Naspitz CK. Risk factors for

asthma and atopy. Curr Opin Allergy Clin Immunol 2005;5:153-9.

54. Van den Toorn LM, Overbeek SE, de Jongste JC, Leman K,

Hoogsteden HC, Prins JB. Airway inflammation is present during clinical

remission of atopic asthma. Am JRespir Crit CareMed 2001;164:2107-13.


AUGUST 2005


Asth

madiagnosis

and

treatm

ent

Mechanisms of asthma and allergic inflammation

Dissecting asthma using focused transgenicmodeling and functional genomics

Douglas A. Kuperman, PhD,a,b,c Christina C. Lewis, PhD,b,c Prescott G.Woodruff, MD,b,d,e

Madeleine W. Rodriguez, BS,b,c Yee Hwa Yang, PhD,b,c Gregory M. Dolganov, PhD,b,d,e

John V. Fahy, MD,b,d,e and David J. Erle, MDb,c,d,e,f Chicago, Ill, and San Francisco, Calif

Background: Asthma functional genomics studies are

challenging because it is difficult to relate gene expression

changes to specific disease mechanisms or pathophysiologic

features. Use of simplified model systems might help to address

this problem. One such model is the IL-13/Epi (IL-13–over-

expressing transgenic mice with STAT6 expression limited to

epithelial cells) focused transgenic mouse, which isolates the

effects of a singlemediator, IL-13, on a single cell type, the airway

epithelial cell. These mice develop airway hyperreactivity and

mucus overproduction but not airway inflammation.

Objective: To identify how effects of IL-13 on airway epithelial

cells contribute to gene expression changes in murine asthma

models and determine whether similar changes are seen in

people with asthma.

Methods: We analyzed gene expression in ovalbumin allergic

mice, IL-13–overexpressing mice, and IL-13/Epi mice with

microarrays. We analyzed the expression of human orthologues

of genes identified in the mouse studies in airway epithelial cells

from subjects with asthma and control subjects.

Results: In comparison with the other 2 models, IL-13/Epi mice

had a remarkably small subset of gene expression changes.

Human orthologues of some genes identified as increased in the

mouse models were more highly expressed in airway epithelial

cells from subjects with asthma than in controls. These

included calcium-activated chloride channel 1, 15-lipoxygenase,

trefoil factor 2, and intelectin.

Conclusion: The combination of focused transgenic models,

DNA microarray analyses, and translational studies provides a

powerful approach for analyzing the contributions of specific

mediators and cell types and for focusing attention on a limited

number of genes associated with specific pathophysiologic

aspects of asthma. (J Allergy Clin Immunol 2005;116:

305-11.)

Key words: Asthma, IL-13, calcium-activated chloride channel,

trefoil factor, 15-lipoxygenase, intelectin

Asthma results from a complex interplay betweengenetic and environmental factors.1 Microarray studiesof lung tissue from subjects with asthma and from animalswith experimental allergic asthma typically reveal hun-dreds of genes that are differentially expressed in com-parison with normal lung.2-4 However, it has been difficultto relate gene expression changes to specific mediators,cell types, or pathophysiologic features.

Models that focus on one particular mediator or celltype or on specific pathophysiologic features might helpovercome this obstacle. We developed a transgenic mousemodel to analyze effects of a single cytokine, IL-13, ona single cell type, the nonciliated airway epithelial cell.5

IL-13 expression is increased in the airways of subjectswith asthma,6,7 and IL-13 is necessary and sufficient forthe development of experimental asthma.8-10 IL-13 activatesthe signaling molecule signal transducer and activator oftranscription factor 6 (STAT6) in many cell types.11

Overexpression of IL-13 in the airways of mice withnormal STAT6 expression (tg–IL-13 mice) causes mucusoverproduction, airway hyperreactivity, inflammation,fibrosis, and emphysema.10 To isolate the effects ofIL-13 on airway epithelial cells, we produced mice thatoverexpress IL-13 in the airway and express STAT6 onlyin nonciliated airway epithelial cells. These IL-13/Epimice developed mucus overproduction and airway hyper-reactivity but not inflammation, fibrosis, or emphysema.5

Here we apply a genomics-based approach to identifygene expression changes in the IL-13/Epi focused model,the tg–IL-13 model, and a conventional allergic asthmamodel. Inclusion of the focused model allowed us topinpoint a remarkably small number of gene expressionchanges that were consistently associated with airwayhyperreactivity and mucus production. Furthermore, weused these results to guide translational studies that showthe relevance of some of these gene expression changes inpeople with asthma. These experiments demonstrate thatfocused transgenic models combined with microarrayscan lead to an improved understanding of the pathogenesisof complex diseases such as asthma.

From athe Department of Medicine, Allergy-Immunology Division,

Northwestern University Feinberg School of Medicine, Chicago; and bthe

Department of Medicine, cthe Lung Biology Center, dthe Division of

Pulmonary and Critical Care Medicine, ethe Cardiovascular Research

Institute, and fthe Program in Immunology, University of California San

Francisco School of Medicine.

Supported by National Institute of Health grants HL56835 and HL72301

and by the UCSF Sandler Center for Basic Research in Asthma.

Disclosure of potential conflict of interest: D. A. Kuperman, none disclosed.

C. A. Lewis, none disclosed. P. G. Woodruff, none disclosed. M. W.

Rodriguez, none disclosed. Y. H. Yang, none disclosed. G. M. Dolganov,

none disclosed. J. V. Fahy, none disclosed. D. J. Erle, none disclosed.

Received for publication January 6, 2005; revised February 28, 2005; accepted

for publication March 9, 2005.


Reprint requests: David J. Erle, MD, UCSF Box 2922, San Francisco, CA


0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.024

305

Mech

anismsofasthmaand

allerg

icinflammation


Abbreviations used

GAPD: Glyceraldehyde-3-phosphate dehydrogenase

IL-13/Epi: IL-13–overexpressing transgenic mice with

STAT6 expression limited to epithelial cells

tg–IL-13: IL-13–overexpressing transgenic mice

STAT6: Signal transducer and activator of transcription

factor 6

UCSF: University of California San Francisco

METHODS

Mice

TheUniversity of California San Francisco (UCSF) Committee on

Animal Research approved the use of mice for these experiments.

Care and use of animals compliedwith theUnited States PublicHealth

Service’s Policy on Humane Care and Use of Laboratory Animals by

Awardee Institutions (#3400-01). Three groups of transgenic mice

were used in these experiments: (1) CC10-IL-131, Stat61/2mice (tg–

IL-13mice); (2)CC10–IL-131 Stat62/2mice; and (3)CC10–IL-131,

Stat62/2, CC10-hStat61 mice (IL-13/Epi mice). CC10 refers to the

Clara cell specific promoter used to express IL-13 and human STAT6

(hSTAT6).12 The development and characterization of these trans-

genic mice have been previously described.5 Mice used here were

backcrossed 5 times onto the Balb/c genetic background. tg–IL-13

mice have an intact Stat6 gene and IL-13–driven activation of STAT6

in a wide range of cells in the lung, resulting in airway inflammation,

mucus overproduction, airway hyperreactivity, subepithelial fibrosis,

and emphysema. IL-131 Stat62/2 mice (used as negative controls)

lack STAT6 and did not develop any detectable IL-13–induced lung

pathology. IL-13/Epi mice also lack mouse STAT6 but express

human STAT6 selectively in airway epithelial cells. hSTAT6 is

functional in mice, and epithelial-restricted activation of Stat6 by

IL-13 induced airway hyperreactivity and mucus production without

airway inflammation, subepithelial fibrosis, or emphysema.Wild-type

Balb/c mice 6 to 8 weeks old were used for the ovalbumin challenge

model. There were 5 mice in each experimental and control group.

Antigen sensitization and challenge

Mice were sensitized by intraperitoneal administration of 50 mg

gradeV ovalbuminmixedwith adjuvant (10mg aluminum potassium

sulfate) 3 times at weekly intervals. Control mice received adjuvant

alone. Beginning 1week after the last injection,micewere challenged

3 times by intranasal administration of ovalbumin (1 mg in 50 mL

PBS) at daily intervals. Control mice were challenged with PBS

alone. Tissues were harvested for isolation of RNA 24 hours after the

last challenge.

Isolation and labeling of RNA from mice

Whole-lung RNAs were purified by using Trizol (Invitrogen,

Carlsbad, Calif). Integrity of all RNA samples used in this study was

confirmedwith amodel 2100 bioanalyzer (Agilent Technologies, Inc,

Palo Alto, Calif). Cy3-labeled and Cy5-labeled lung cDNAs were

prepared as described.13 To obtain samples enriched for airway

epithelial cell RNA, the superior portion of the trachea was

cannulated, and the trachea and proximal major bronchi were excised

from the thorax and slowly perfused with 0.35 mL lysis buffer

(RNeasy kit; Qiagen Inc, Valencia, Calif). RNA was isolated from

perfusate according to the manufacturer’s instructions. Because the

amount of RNA obtained from tracheal perfusates was only ~1 mg, a

T7 RNA polymerase-based method13 was used to prepare Cy3-

labeled and Cy5-labeled amplified cRNAs for array hybridizations.

Microarray analysis

Lung gene expression was analyzed by hybridizing Cy5-labeled

cDNA from mouse lungs (5 mice per group, each hybridized

separately) along with Cy3-labeled reference lung cDNA pooled

from wild-type mice. Tracheal perfusate samples were analyzed

similarly, except that amplified cRNA from each mouse was com-

pared with an amplified cRNA reference pool made by using tracheal

perfusate samples from wild-type mice. DNA microarrays used

in these experiments were produced by using the Operon Bio-

technologies (Huntsville, Ala) Mouse Genome Oligo 2.0 set of

70-mer oligonucleotides, supplemented by some additional 70-mers.

A MIAME-compliant description of the array experiments and the

raw array data are available from Gene Expression Omnibus (http://

www.ncbi.nlm.nih.gov/geo, accession number GSE1438).

We used an approach that allowed us to estimate differential gene

expression between the various groups we studied on the basis of

linear models, as previously described.14,15 To determine whether

there were significant differences in gene expression between groups,

we calculated the odds ratio (probability of being differentially

expressed/probability of not being differentially expressed). When

the log2 of the odds ratios (known as the B-statistic) was greater than

0, we classified the gene as differentially expressed.16,17 Hierarchical

clustering, a method for grouping together genes with similar

expression patterns, was performed by using Acuity 4.0 software

(Axon Instruments, Union City, Calif). In addition, genes were

classified on the basis of expression patterns to determine whether

they were increased relative to controls in 1, 2, or all 3 of the

experimental groups (ovalbumin, tg–IL-13, and IL-13/Epi) as fol-

lows. Seven pseudogene vectors were created to represent genes that

were increased only in 1 of the 3models ([0,0,1]; [0,1,0]; and [1,0,0]),

genes increased equally in 2 models ([0,1,1]; [1,0,1]; [1,1,0]), and

genes increased equally in all 3 models ([1,1,1]). Each differentially

expressed gene was assigned a vector in 3-dimensional space

according to the median log fold-change gene expression values

determined for that gene in the 3 experimental groups. All vectors

were scaled to unit length, and each gene was matched to the closest

pseudogene, as determined by Euclidean distance.

Human subjects

These studies were approved by the UCSF Committee on Human

Research and conducted in compliance with the Declaration of

Helsinki principles. Written informed consent was obtained from all

subjects. All subjects were adult nonsmokers (,10 pack-year total

smoking historywith last cigarette.1 year before the study).Medical

histories were obtained, physical examinations were performed,

symptom questionnaires were collected, and spirometry was per-

formed as described.18 Airway reactivity was measured by deter-

mining the PC20.19 The 28 healthy control subjects had no history of

lung disease and were not hyperreactive (PC20 . 16 mg/mL). All 30

subjects with asthma had a previous physician diagnosis of asthma,

were hyperreactive (PC20 , 8 mg/mL), and used only short-acting

inhaled b-adrenergic–agonist medications for therapy. Individuals

with an asthma exacerbation or respiratory infection within the

previous 6 weeks or significant medical problems other than asthma

and those using inhaled or systemic corticosteroids or leukotriene

antagonists were excluded.

Bronchial epithelial brushings

Bronchoscopy was performed, and bronchial brushings were

obtained randomly from right or left lower lobe bronchial segments

by using 4 disposable cytology brushes. The brushes were gently

vortexed in sterile saline. Cells from all brushes were pooled, yielding

a single sample for each subject. An aliquot was removed for

cytocentrifugation, stained with Diff-Quik (Baxter, McGraw Park,


AUGUST 2005

306 Kuperman et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

http://www.ncbi.nlm.nih.gov/geo

http://www.ncbi.nlm.nih.gov/geo

Ill), and examined by light microscopy. On average, the bronchial

brushings contained 97% epithelial cells. Total RNA was extracted

by using the RNeasy kit. Thirty-four of the 58 human epithelial cell

RNA samples analyzed for this study were analyzed separately for

another study (Woodruff et al, unpublished data).

Cultured human bronchial epithelial cells

Air-liquid interface cultures were established by using published

protocols.20 Primary normal human bronchial epithelial cells (lot

3F1191; Cambrex Bio Science, Baltimore, Md) were seeded onto

12-mm–diameter Corning Transwells (Corning, NY) containing

0.4-mm pores (4-5 replicates per condition). Cells were submerged

in bronchial epithelial growth media (Clonetics) until confluent

(3 days), and then the apical media was removed and the basolateral

media was changed to differentiation media (1:1 of Dulbecco

modified Eagle media to bronchial epithelial growth media) for

8 days. After the differentiation period, IL-13 (0-10 ng/mL) was

maintained in the basolateral media over the period of the next 4 days.

At the end of the IL-13 exposure, total RNAwas isolated by using the

RNeasy kit.

Analysis of gene expression by real-time PCR

Primers and a probe formouse genes (see Table E1 in the Journal’s

Online Repository at www.mosby.com/jaci) and human genes (see

Table E2 in the Online Repository at www.mosby.com/jaci) were

designed by using Primer Express software (Perkin Elmer, Boston,

Mass). First-strand cDNA synthesis and PCR were performed by

usingABI Prizm 7700 or 7900 SequenceDetection Systems (Applied

Biosystems, Foster City, Calif). For mouse lung and cultured human

airway epithelial cell studies, cycle thresholds for each gene were

normalized to those of glyceraldehyde-3-phosphate dehydrogenase

(GAPD). For human bronchial brushing samples, a 2-step PCR

approach21 was used, and transcript copy numbers were normalized

on the basis of the geometric mean expression values of 3 house-

keeping genes (GAPD, elongation factor 1a1, and cyclophilin A) as

described.22 To identify genes that were differentially expressed

between subjects with asthma and normal subjects, we performed a

Wilcoxon rank-sum test and calculated the corresponding adjusted

P value by using the Westfall and Young23 maxT algorithm

implemented in Bioconductor’s multtest package (Bioconductor;

open source software for bioinformatics, www.bioconductor.org).24

RESULTS

Microarray analysis of gene expression inmouse asthma models

We used DNA microarrays to analyze gene expressionin 3 murine models of asthma. The first model wascomposed of mice that were sensitized and challengedwith ovalbumin (ovalbumin model). The second modelwas composed of mice with transgenic overexpression ofIL-13 in the lung (tg–IL-13 model); these mice had anintact gene for STAT6, a key signaling molecule requiredfor IL-13 activity. The third model was composed of IL-13transgenic mice with STAT6 expression limited to non-ciliated airway epithelial cells (IL-13/Epi model). Inwhole-lung samples, 805 gene transcripts were differen-tially expressed (increased or decreased) in at least 1model(Fig 1). Of these, 509 genes were increased or decreasedby 2-fold in at least 1 model. There were 583 geneexpression changes in the ovalbumin model and 351changes in the tg–IL-13 model. In contrast, only 18changes were seen in the IL-13/Epi focused transgenicmodel. Comparison of the focused transgenic model to theother models allowed for the identification of a smallsubset of gene expression changes that are attributable to aspecific mechanism and also are relevant to pathogenesisin complex systems.

FIG 1. Lung gene expression changes in 3 mouse asthma models.

Differentially expressed genes were arranged by hierarchical

clustering. Each column represents data from 1 of 5 individual

mice in each group. Colors represent fold-change compared with

the appropriate controls. The arrow indicates a small group of

genes that were increased in all 3 models. Ova, Ovalbumin.

FIG 2. Gene expression patterns. Grouping revealed genes with

increased expression in (A) the ovalbumin (Ova) model only, (B)

the Ova and tg–IL-13 models, (C) the tg–IL-13 model only, and (D)

all 3 models. The number of genes (left) and representative genes

(right) is shown for each group. Phenotypic attributes of each

model are shown at the bottom. AHR, Airway hyperreactivity.



Kuperman et al 307

Mech

anismsofasthmaand

allerg

icinflammation



http://www.sciencedirect.com/\\www.bioconductor.org

To aid in interpretation, we grouped transcripts accord-ing to the magnitude of change in expression across the3 models. Transcripts were grouped by determining whetherthe expression changes were seen in 1, 2, or all 3 ofthe models (see Table E3 in the Online Repository atwww.mosby.com/jaci). There were 17 transcripts with atleast 2-fold increases in the IL-13/Epi model but only 3transcripts with at least 2-fold decreases; therefore, wefocused our subsequent analyses on increased genes. Atotal of 239 transcripts were increased in the ovalbuminmodel but not induced (or induced to a much lesser extent)in the tg–IL-13 or IL-13/Epi models (Fig 2, A). Thesechanges are likely largely attributable to allergen-inducedlymphocyte activation and to production of mediatorsother than IL-13. There were 247 genes increased simi-larly in the ovalbumin and tg–IL-13 models but not in theIL-13/Epi model (Fig 2, B). These allergen-induced genesare apparently increased because of IL-13 effects on cellsother than nonciliated airway epithelial cells, and the large

number of gene transcripts is consistent with the idea thateffects of IL-13 on these cells are important in allergicinflammatory responses.8,9 A total of 73 transcripts wereincreased in the tg–IL-13 model but not increased (orincreased to a much lesser extent) in the other 2 models(Fig 2, C). These represent genes that are induced byprolonged high-level overexpression of IL-13 but not byacute allergen challenge and may be involved in thepathogenesis of subepithelial fibrosis and emphysema,pathologic features present in the tg–IL-13 model but notthe other 2 models.

We were especially interested in identifying genes thatwere increased in all 3 models. Analysis of whole-lungsamples revealed 35 genes that had similar fold increasesin all models (Fig 2, D). Because some epithelial geneexpression changes might have been undetectable inwhole lung samples, we also analyzed gene expressionin tracheal perfusate samples enriched for airway epithe-lial cell RNA. In these samples, we detected 276 geneswith altered expression (increased or decreased) in at least1 model (see Table E4 in the Online Repository atwww.mosby.com/jaci). Of these, 107 genes were changedby 2-fold. There were 43 genes changed in the ovalbu-min model and 239 in the tg–IL-13 model. There were 76genes changed in the IL-13/Epi model, but only 22 ofthese were changed by 2-fold or more. Some genes hadsimilar expression changes in both analyses, but manyother changes were identified only in lung samples or onlyin tracheal samples. In part, this reflects the fact that thetracheal sample was enriched for large airway epithelialcells, whereas the whole-lung sample contained smallerairway epithelial cells as well as many other cell types. In

TABLE I. Genes induced by allergen and by direct effects of IL-13 on airway epithelial cells*

Fold-change by arrays Fold-change by PCR

Symbol Description Ova tg–IL-13 IL-13/Epi Ova tg–IL-13 IL-13/Epi

Clca3 Chloride channel calcium activated 3 78.3 59.8 28.2 820.3 1530.7 1448.2

Slc26a4 Solute carrier family 26, member 4 9.9 29.8 20.0 18.7 40.1 10.3

Retnla Resistin-like a 4.3 23.8 23.7 ND ND ND

Itln Intelectin 7.6 10.8 19.2 9.5 73.8 146.2

Nadsyn NAD synthetase 1 17.2 12.5 6.4 0.7 0.5 0.8

Atp2a1 Ca11 transporting ATPase 15.3 8.0 3.9 0.4 0.1 0.9

AMCase Acidic mammalian chitinase 2.5 12.3 8.6 3.7 7.4 4.4

Muc5ac Mucin 5, subtypes A and C 12.1 6.8 3.8 33.5 23.4 7.8

Agr2 Anterior gradient 2 3.8 7.7 4.3 11.1 21.6 10.2

Tff1 Trefoil factor 1 3.3 6.4 5.7 52.6 592.2 224.7

Reg3g Regenerating islet-derived 3g 6.4 4.1 3.2 ND ND ND

Pigr Polymeric immunoglobulin receptor 2.5 8.4 2.3 1.5 12.3 2.2

Muc5b Mucin 5, subtype B 4.0 5.1 2.8 3.6 7.4 13.2

D630002J15Rik RIKEN cDNA D630002J15 gene 2.8 4.3 2.0 ND ND ND

Scin Scinderin 2.8 2.9 3.0 21.9 21.8 34.5

Alox15 15-lipoxygenase 3.9 1.9 2.0 5.2 1.6 2.8

H2-Q7 Histocompatibility 2, Q region locus 7 2.4 2.1 2.1 ND ND ND

1110001D15Rik RIKEN cDNA 1110001D15 gene 2.1 2.1 2.4 ND ND ND

Tff2 Trefoil factor 2 4.5 4.5 1.8 3.5 8.6 4.9

*Genes with a 2.0-fold or greater increase in expression in both the ovalbumin (Ova) allergic model and the IL-13/Epi model are included. Array fold-change

values represent the larger median fold change from lung or tracheal perfusate compared with the appropriate control group.

ND indicates that we did not use PCR to determine fold change for these mouse genes, which do not have obvious human orthologues.

TABLE II. Characteristics of human subjects*

Control

subjects

Subjects

with asthma P

Number 28 30 —

Age 36 6 8 38 6 13 NS

Sex 16 F/12 M 18 F/12 M NS

PC20 (mg/mL) 60.8 6 1.4 0.9 6 1.3 ,.0001

FEV1/forced vital

capacity (%)

80.6 6 1.4 71.1 6 1.4 ,.0001

FEV1 (% predicted) 106.5 6 13.2 85.2 6 13.9 ,.0001

*Values represent means 6 SDs.


AUGUST 2005

308 Kuperman et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n



addition, we expect that there were differences in thesensitivity of the lung and tracheal sample analysesbecause we did not need to use RNA amplification forlung samples but did need to amplify tracheal samples. Bycombining microarray results from whole-lung and tra-cheal perfusate samples, we identified 18 genes that wereincreased at least 2-fold by allergen and by direct effects ofIL-13 on airway epithelial cells (Table I). We alsoincluded trefoil factor 2 because previous PCR analysis5

suggested that the microarrays underestimated changes inexpression of this gene.

PCR validation of gene expression changesin mouse models

In preparation for translational studies, we generated avalidated list of genes with human orthologues that wereincreased in both ovalbumin and IL-13/Epi mice. Fivetranscripts listed in Table I did not have obvious humanorthologues. Eleven of the remaining 14 transcripts wereincreased by at least 2-fold in both ovalbumin allergic andIL-13/Epi mice by PCR (Table I). Our mouse modelingapproach therefore resulted in the selection of 11 genesthat could be studied in people with asthma.

Airway epithelial gene expression in humansubjects with and without asthma

We used real-time PCR to analyze gene expression inairway epithelial cells from 30 subjects with mild tomoderate asthma and 28 controls (Table II). There weresubstantial and highly significant increases in the expres-sion of the calcium activated chloride channel Clca1(orthologue of murine Clca3) and intelectin in subjectswith asthma (Table III and Fig 3, A). There were smallerincreases in expression of 15-lipoxygenase and trefoilfactor 2 in subjects with asthma. One transcript, mucin 5b,was less abundant in subjects with asthma, and othertranscripts examined were not significantly different.We were unable to consistently detect acidic mammalianchitinase transcripts in either group by using 2 differentsets of PCR primers and probes.

IL-13 induces intelectin expression in culturedhuman airway epithelial cells

One of the transcripts substantially increased in airwayepithelial cells from subjects with asthma was intelectin, apattern recognition molecule that has not been previouslyimplicated in asthma. IL-13 treatment of cultured primaryhuman bronchial epithelial cells led to a large increase inintelectin transcripts (Fig 3, B), demonstrating that IL-13directly increases intelectin expression and that other celltypes are not required.

DISCUSSION

We combined focused transgenic modeling, functionalgenomics, and translational studies in human subjects tohelp understand important aspects of the complex patho-genesis of asthma. As expected, microarray analysis ofan allergic model of asthma revealed hundreds of geneexpression changes. Analysis of the tg–IL-13 model,where IL-13 acts on many different cell types to produceextensive lung pathology, also showed hundreds ofchanges, many of which were similar to those seen inthe allergic model. By using the data from these 2 models,

FIG 3. Intelectin (Itln) expression in human airway epithelial cells.

A, Intelectin gene expression by airway epithelial cells from control

subjects and subjects with asthma. Medians and interquartile

ranges are shown. B, IL-13 treatment of cultured human airway

epithelial cells induced increased expression of intelectin tran-

scripts.

TABLE III. Gene expression in airway epithelial cells from control subjects and subjects with asthma

Fold

Difference*

Median gene copy number

Symbol Description Control Asthmatic Adjusted P

Clca1 Chloride channel calcium activated 1 216.5 0.06 3 104 8.21 3 104 .0001

Itln Intelectin 3.4 1.73 3 105 6.12 3 105 .0002

Alox15 15-lipoxygenase 1.3 2.28 3 106 3.27 3 106 .06

Tff2 Trefoil factor 2 2.7 0.55 3 104 1.00 3 104 .06

Muc5ac Mucin 5, subtypes A and C 1.5 1.67 3 107 2.07 3 107 NS

Tff1 Trefoil factor 1 1.4 1.47 3 105 2.17 3 105 NS

Agr2 Anterior gradient 2 1.1 0.90 3 107 1.04 3 107 NS

Slc26a4 Solute carrier family 26, member 4 0.8 3.72 3 104 3.35 3 104 NS

Scin Scinderin 1.0 4.70 3 105 4.99 3 105 NS

Muc5b Mucin 5, subtype B 25.0 1.84 3 106 0.42 3 106 .0001

AMCase Acidic mammalian chitinase — ND ND —

*Fold difference between subjects with asthma and control subjects, adjusted for age and sex.

Human Clca1 is orthologous to mouse Clca3.

ND indicates that this transcript was not detected in most samples from both groups.



Kuperman et al 309

Mech

anismsofasthmaand

allerg

icinflammation

it would have been extremely challenging to determinehow particular gene expression changes relate to specificdisease mechanisms or to select a reasonable number ofcandidate genes for further study. By analyzing a thirdmodel (IL-13/Epi) that focused on the effects of a singlemediator, IL-13, on a single cell type, the airway epithelialcell, we were able to generate a dramatically shorter list ofdifferentially expressed genes. These expression changeswere associated with airway hyperreactivity and mucusoverproduction, which are features of all 3 models, but notwith inflammation, fibrosis, or emphysema, which areabsent in the IL-13/Epi model.

Our approach was designed to reduce the complexityinherent in microarray studies of disease pathogenesis.Other microarray studies have reduced complexity byanalyzing homogeneous populations of cultured cells orcells isolated from tissues by laser capture microscopy orflow-cytometric sorting. Although those approaches canbe useful, an important advantage of the focused trans-genic model approach that we describe here is that itisolates the effects that a single mediator exerts on a singlecell type in vivo.

We used translational studies to determine whether thegene expression changes first identified by using mousemodels were also seen in human disease. Four of theorthologues identified in the models (the chloride channelClca1, intelectin, 15-lipoxygenase, and trefoil factor 2)were increased in airway epithelial cells from people withasthma. In another study involving a subset of the humansubjects used for this study, we found that expression ofClca1 transcripts and protein was increased in airwayepithelial cells from subjects with asthma (Woodruff et al,unpublished data). Previous reports suggest that inductionof the mClca3/hClca1 chloride channel in airway epithe-lial cells is important for mucus production.25-27 15-Lipoxygenase produces 15S-hydroxyeicosatetraenoic acid,which has been reported to triggermucus secretion in dogs,28

promote contraction of human airway smooth muscle,29

and potentiate increases in allergen-induced earlyasthmatic responses in human beings.30 Trefoil factor2 was shown to be increased in an allergic mouse modelof asthma,31 and we now show that trefoil factor 2 isincreased in airway epithelial cells from people withasthma. Trefoil factor 2 promotes migration of culturedbronchial epithelial cells32 and contributes to repair ofinjured intestinal epithelium.33 Intelectin is a recentlydescribed pattern recognition molecule not previouslyimplicated in asthma. There are 2 related intelectins, Itln1and Itln2. The probes and primers that we used weredesigned to recognize Itln1, but they might not distinguishbetween these 2 very closely related sequences. Themolec-ular patterns recognized by intelectin include furanosidessuch as galactofuranose.34 Galactofuranosyl residues arepresent in bacterial and fungal cell walls and in protozoanparasites but not in mammalian cells. The possiblecontributions of intelectin to asthma pathogenesis requirefurther exploration, but intelectin in the airway might alterthe response of subjects with asthma to infection orcolonization with bacterial or fungal pathogens.

We did not find significant increases in expression ofthe other human homologues that we examined. To someextent, this likely reflects difficulties inherent in detectinggene expression changes in limited samples (epithelialbrushings) taken from human subjects with stable mildto moderate disease, as opposed to whole-lung analysisof mice with homogeneous genetic backgrounds housedin a controlled environment and subjected to a potentdisease-inducing stimulus. One transcript, Muc5b, wassignificantly increased in each of the mouse models, butexpression of both the transcript and the protein is de-creased in people with asthma (Woodruff et al, unpub-lished data). Differences inMuc5b expressionmight relateto the fact that human Muc5b is expressed predominantlyin submucosal glands.35 These glands are abundant inhuman airways but are absent in the mouse, with theexception of a single gland in the trachea. This is anexample of the limits of mouse modeling. Our PCR assaywas unable consistently to detect the transcript for anotherorthologue, AMCase, but a recent report demonstratedthat airway epithelial production of AMCase is alsoincreased in subjects with asthma.36 Including transla-tional studies helped to clarify the relevance of the animalmodels to human disease and allowed us to draw novelinferences about the activity of a specific mechanism inhuman disease.

We conclude that the combination of focused trans-genic models, DNAmicroarray analyses, and translationalstudies provides a powerful approach for analyzing thecontributions of specific mediators and cell types and forfocusing attention on a limited number of genes associatedwith specific pathophysiologic aspects of complex dis-eases like asthma.

We thank Dean Sheppard and Andrea Barczak for their advice and

Xiaozhu Huang, Louis Nguyenvu, Michael Salazar, and the staffs of

the Sandler Center Animal Physiology and Microscopy Core and the

UCSF National Heart, Lung, and Blood Institute Shared Microarray

Facility for technical assistance.

REFERENCES

1. Bousquet J, Jeffery PK, Busse WW, Johnson M, Vignola AM. Asthma:

from bronchoconstriction to airways inflammation and remodeling. Am J


2. Laprise C, Sladek R, Ponton A, Bernier MC, Hudson TJ, Laviolette M.

Functional classes of bronchial mucosa genes that are differentially

expressed in asthma. BMC Genomics 2004;5:21.

3. Zimmermann N, King NE, Laporte J, Yang M, Mishra A, Pope SM,

et al. Dissection of experimental asthma with DNA microarray analysis

identifies arginase in asthma pathogenesis. J Clin Invest 2003;111:

1863-74.

4. Zou J, Young S, Zhu F, Gheyas F, Skeans S, Wan Y, et al. Microarray

profile of differentially expressed genes in a monkey model of allergic

asthma. Genome Biol 2002;3:research0020.1-0020.13.

5. Kuperman DA, Huang X, Koth LL, Chang GH, Dolganov GM, Zhu Z,

et al. Direct effects of interleukin-13 on epithelial cells cause airway

hyperreactivity and mucus overproduction in asthma. Nat Med 2002;8:

885-9.

6. Huang SK, Xiao HQ, Kleine-Tebbe J, Paciotti G, Marsh DG,

Lichtenstein LM, et al. IL-13 expression at the sites of allergen challenge

in patients with asthma. J Immunol 1995;155:2688-94.


AUGUST 2005

310 Kuperman et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

7. Humbert M, Durham SR, Kimmitt P, Powell N, Assoufi B, Pfister R,

et al. Elevated expression of messenger ribonucleic acid encoding IL-13

in the bronchial mucosa of atopic and nonatopic subjects with asthma.


8. Grunig G, Warnock M, Wakil AE, Venkayya R, Brombacher F, Rennick

DM, et al. Requirement for IL-13 independently of IL-4 in experimental

asthma. Science 1998;282:2261-3.

9. Wills-Karp M, Luyimbazi J, Xu X, Schofield B, Neben TY, Karp CL,

et al. Interleukin-13: central mediator of allergic asthma. Science 1998;

282:2258-61.

10. Zhu Z, Homer RJ, Wang Z, Chen Q, Geba GP, Wang J, et al. Pulmonary

expression of interleukin-13 causes inflammation, mucus hypersecretion,

subepithelial fibrosis, physiologic abnormalities, and eotaxin production.

J Clin Invest 1999;103:779-88.

11. Nelms K, Keegan AD, Zamorano J, Ryan JJ, Paul WE. The IL-4

receptor: signaling mechanisms and biologic functions. Annu Rev

Immunol 1999;17:701-38.

12. Stripp BR, Sawaya PL, Luse DS, Wikenheiser KA, Wert SE, Huffman

JA, et al. cis-acting elements that confer lung epithelial cell expression of

the CC10 gene. J Biol Chem 1992;267:14703-12.

13. Barczak A, Rodriguez MW, Hanspers K, Koth LL, Tai YC, Bolstad BM,

et al. Spotted long oligonucleotide arrays for human gene expression

analysis. Genome Res 2003;13:1775-85.

14. Diaz E, Yang YH, Ferreira T, Loh KC, Okazaki Y, Hayashizaki Y, et al.

Analysis of gene expression in the developing mouse retina. Proc Natl

Acad Sci U S A 2003;100:5491-6.

15. JinW,RileyRM,WolfingerRD,WhiteKP, Passador-Gurgel G,GibsonG.

The contributions of sex, genotype and age to transcriptional variance in

Drosophila melanogaster. Nat Genet 2001;29:389-95.

16. Lonnstedt I, Speed TP. Replicated microarray data. Statistica Sinica

2002;12:31-46.

17. Smyth GK. Linear models and empirical Bayes methods for assessing

differential expression in microarray experiments. Stat Appl Genet

Mol Biol 2004;3:3. Available at: http://www.bepress.com/sagmb/vol3/

iss1/art3/.

18. Claman DM, Boushey HA, Liu J, Wong H, Fahy JV. Analysis of

induced sputum to examine the effects of prednisone on airway

inflammation in asthmatic subjects. J Allergy Clin Immunol 1994;94:

861-9.

19. Chai H, Farr RS, Froehlich LA, Mathison DA, McLean JA, Rosenthal

RR, et al. Standardization of bronchial inhalation challenge procedures.


20. Atherton HC, Jones G, Danahay H. IL-13-induced changes in the goblet

cell density of human bronchial epithelial cell cultures: MAP kinase and

phosphatidylinositol 3-kinase regulation. Am J Physiol Lung Cell Mol

Physiol 2003;285:L730-9.

21. Dolganov GM, Woodruff PG, Novikov AA, Zhang Y, Ferrando RE,

Szubin R, et al. A novel method of gene transcript profiling in airway

biopsy homogenates reveals increased expression of a Na1-K1-Cl-

cotransporter (NKCC1) in asthmatic subjects. Genome Res 2001;11:

1473-83.

22. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A,

et al. Accurate normalization of real-time quantitative RT-PCR data

by geometric averaging of multiple internal control genes. Genome Biol

2002;3:research0034.1-0034.11.

23. Westfall PH, Young SS. Resampling-based multiple testing: examples and

methods for p-value adjustment. New York: John Wiley & Sons; 1993.

24. Ge Y, Dudoit S, Speed TP. Resampling-based multiple testing for

microarray data hypothesis. Test 2003;12:1-44.

25. Hoshino M, Morita S, Iwashita H, Sagiya Y, Nagi T, Nakanishi A, et al.

Increased expression of the human Ca21-activated Cl- channel

1 (CaCC1) gene in the asthmatic airway. Am J Respir Crit Care Med

2002;165:1132-6.

26. Nakanishi A, Morita S, Iwashita H, Sagiya Y, Ashida Y, Shirafuji H,

et al. Role of gob-5 in mucus overproduction and airway hyper-

responsiveness in asthma. Proc Natl Acad Sci U S A 2001;98:5175-80.

27. Zhou Y, Shapiro M, Dong Q, Louahed J, Weiss C, Wan S, et al. A

calcium-activated chloride channel blocker inhibits goblet cell metaplasia

and mucus overproduction. Novartis Found Symp 2002;248:150-65;

discussion 65-70, 277-82.

28. Johnson HG, McNee ML, Sun FF. 15-Hydroxyeicosatetraenoic acid is a

potent inflammatory mediator and agonist of canine tracheal mucus

secretion. Am Rev Respir Dis 1985;131:917-22.

29. Copas JL, Borgeat P, Gardiner PJ. The actions of 5-, 12-, and 15-HETE

on tracheobronchial smooth muscle. Prostaglandins Leukot Med 1982;8:

105-14.

30. Lai CK, Polosa R, Holgate ST. Effect of 15-(s)-hydroxyeicosatetraenoic

acid on allergen-induced asthmatic responses. Am Rev Respir Dis 1990;

141:1423-7.

31. Nikolaidis NM, Zimmermann N, King NE, Mishra A, Pope SM,

Finkelman FD, et al. Trefoil factor-2 is an allergen-induced gene

regulated by Th2 cytokines and STAT6 in the lung. Am J Respir Cell

Mol Biol 2003;29:458-64.

32. Oertel M, Graness A, Thim L, Buhling F, Kalbacher H, Hoffmann W.

Trefoil factor family-peptides promote migration of human bronchial

epithelial cells: synergistic effect with epidermal growth factor. Am J

Respir Cell Mol Biol 2001;25:418-24.

33. Farrell JJ, Taupin D, Koh TJ, Chen D, Zhao CM, Podolsky DK, et al.

TFF2/SP-deficient mice show decreased gastric proliferation, increased

acid secretion, and increased susceptibility to NSAID injury. J Clin

Invest 2002;109:193-204.

34. Tsuji S, Uehori J, Matsumoto M, Suzuki Y, Matsuhisa A, Toyoshima K,

et al. Human intelectin is a novel soluble lectin that recognizes

galactofuranose in carbohydrate chains of bacterial cell wall. J Biol

Chem 2001;276:23456-63.

35. Sharma P, Dudus L, Nielsen PA, Clausen H, Yankaskas JR, Hollings-

worth MA, et al. MUC5B and MUC7 are differentially expressed in

mucous and serous cells of submucosal glands in human bronchial

airways. Am J Respir Cell Mol Biol 1998;19:30-7.

36. Zhu Z, Zheng T, Homer RJ, Kim YK, Chen NY, Cohn L, et al. Acidic

mammalian chitinase in asthmatic Th2 inflammation and IL-13 pathway

activation. Science 2004;304:1678-82.



Kuperman et al 311

Mech

anismsofasthmaand

allerg

icinflammation

http://www.bepress.com/sagmb/vol3/iss1/art3/

http://www.bepress.com/sagmb/vol3/iss1/art3/

Endobronchial adenosine monophosphatechallenge causes tachykinin release in thehuman airway

Fionnuala Crummy, MD, MRCP,a,c Mark Livingston, PhD,a,b Joy E. S. Ardill, PhD,

FCRPath,c Catherine Adamson, MSc,a,b Madeleine Ennis, PhD,a,b and

Liam G. Heaney, MD, MRCPa,c Belfast, Northern Ireland, United Kingdom

Background: Adenosine 5 monophosphate (AMP) has been

shown to cause bronchoconstriction and a sensation of chest

tightness when inhaled by asthmatic subjects. This response is

attenuated after repeated inhalation of bradykinin, suggesting

that AMP may act in part by the release of neuropeptides.

Objective: This study examined neuropeptide release in the

human airway after endobronchial AMP challenge.

Methods: Endobronchial AMP challenge was performed in

20 subjects and tachykinin levels were measured after

endobronchial AMP challenge and after placebo endobronchial

challenge with saline.

Results: All subjects coughed immediately after adenosine

challenge. There was a significant increase in neurokinin A and

substance P levels (P < .01, P< .01 respectively) when post-saline

and post-AMP levels were compared. There was, however, no

significant change in calcitonin gene related peptide levels

(P=.37).

Conclusion: This study demonstrates that endobronchial AMP

challenge causes tachykinin release in the human airway

in vivo. (J Allergy Clin Immunol 2005;116:312-7.)

Key words: Tachykinin, neuropeptide, adenosine, endobronchial

challenge

Adenosine is a naturally occurring purine nucleosidethat functions as a constituent of nucleic acid, as anintracellular and autocoid mediator.1 Elevated levels havebeen found in bronchoalveolar lavage fluid of asthmaticsubjects as compared to normal subjects, suggesting thatadenosine may be a mediator in asthma.2 Inhalation ofadenosine monophosphate (AMP), which is rapidlydephosphorylated to adenosine in vivo,3 causes broncho-constriction in atopic asthmatic and non-asthmatic sub-jects but not in non-atopic, non-asthmatic subjects.4 Therelated nucleoside guanosine has no effect, suggesting thatthis is a specific receptor mediated effect.5

Atopic asthmatic and non-asthmatic subjects cough andbronchoconstrict in response to stimuli such as inhaledAMP and sulphur dioxide; however, asthmatics tend torespond to lower concentrations.6 Cough and chest tight-ness are both common symptoms in asthma and are relatedto stimulation of sensory nerves.6

Evidence exists that while AMP acts mainly via primedmast cells, the agent also stimulates vagal nerves. Pre-treatment with ipratropium (an anti-cholinergic agent) hasa bronchoprotective effect on the response to AMP,suggesting activation of cholinergic nerves in the responseto AMP.7,8 Inhalation of AMP and bradykinin cause agreater sensation of chest tightness than does inhalation ofmethacholine, for the same degree of bronchoconstriction,suggesting that the former acts on sensory pathways.9

Repeated inhalation of bradykinin attenuates the responseto inhaled AMP suggesting that both of these agents act inpart via liberation of neuropeptides from sensory nerves.10

Hong et al11 have shown that pulmonary C fibers (thenerve fibers containing neuropeptides) in the rat areactivated after right atrial injection of adenosine, impli-cating these nerves in the response to AMP in this model.The purpose of this study was to examine neuropeptiderelease in vivo in the human airway after endobronchialAMP challenge.

METHODS

Subjects

Ethical approval was granted by the Research Ethics Committee

of the Queen’s University of Belfast. All subjects gave written

informed consent. All subjects were non-smokers and had not

received any anti-histamines or inhaled or oral steroids in the preceding

six months. Asthmatics were recruited if they (1) had a prior clinical

Abbreviations usedAMP: Adenosine 5# monophosphate

CGRP: Calcitonin gene related peptide

NEP: Neutral endopeptidase

NKA: Neurokinin A

NKB: Neurokinin B

NPK: Neuropeptide K

PC20 AMP: Provocative concentration of AMP causing a

20% fall in FEV1

SP: Substance P

From aRespiratory Research Group, School of Medicine, Queen’s University

of Belfast and Departments of bClinical Biochemistry and Metabolic

Medicine and cMedicine, Queens University, Belfast.

Funding: Northern Ireland Chest Heart and Stroke Association.


accepted for publication March 28, 2005.


Reprint requests: Liam G. Heaney, MD, MRCP, Regional Respiratory Centre,

Level 8, Belfast City Hospital, Lisburn Road, Belfast, Northern Ireland, UK.

BT9 7AB. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.034

312

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n


diagnosis of asthma and a history of intermittent shortness of breath

or wheeze, (2) were atopic (reacted to at least one allergen on skin

prick testing), and (3) had FEV1 greater than 60% predicted. All other

subjects had no symptoms suggestive of asthma; atopic non-asthmatic

subjects had at least one positive skin prick test as aforementioned.

All subjects attended on two occasions. At the screening visit,

informed consent was obtained and clinical assessment, skin prick

testing, and AMP inhalation challenge were performed. At the

subsequent visit (at least 72 hours after screening visit), bronchos-

copy and endobronchial AMP challenge were performed.

Skin prick testing

Skin prick testing was performed using a standardized puncture

technique,12 using allergen preparations of house dust mite, cat, and

dog protein and grass pollen (Dome-Hollister-Stier, Epernon Cedex,

France). A positive reaction was taken as a wheal size of 3 mm or

more.

Inhalational challenging

Spirometry was performed according to American Thoracic

Society Guidelines13 using a Vitalograph spirometer (Buckingham,

UK). AMP (Sigma-Aldrich Ltd., Poole, UK) was freshly prepared in

0.9% saline in doubling concentrations ranging from 0.391mg/mL to

400mg/mL. TheAMP provocation test was performed using the two-

minute tidal breathing method of Cockcroft et al14 using a Medix

Turbonebuliser (Leicestershire, UK) with an output of 0.13 mL/min.

PC20 AMP was calculated by linear interpolation.

Bronchoscopy

AMPwas freshly made up on the morning of the bronchoscopy in

0.9% saline from a stock solution of 400 mg/mL. At bronchoscopy,

subjects were given intravenous Midazolam (up to 14 mg) to achieve

mild sedation and the hypopharynx was anaesthetised using 4%

lignocaine spray. Vocal cord and tracheal anaesthesia was achieved

using 4 mL of 4% lignocaine introduced trans-cricoidally. Oxygen

was routinely applied at 2 L/min via nasal cannulae. Heart rate,

ECG, and oxygen saturations were monitored throughout the

procedure. The bronchoscope (240 IT Olympus Optical Co. Ltd.

Tokyo, Japan) was introduced orally and 2-mL aliquots of 2%

lignocaine were used as necessary to anesthetize the airways below

the carina to suppress coughing.

The site of the subsequent endobronchial challenge was rando-

mized prior to bronchoscopy. Subjects were randomly assigned a

number, which determined the site of the active challenge to either

the right middle or upper lobes, and randomization was constrained

to achieve balance. The placebo challenge was automatically

assigned to the opposite site from the active challenge.

The bronchoscope was initially wedged in a segmental orifice of

the site randomized for the placebo challenge and a baseline bronchial

wash with 20 mL of saline was performed and aspirated back after

minimum dwell time. A placebo challenge of 5 mL of saline was

administered to the same segment and the segment closely observed

for any visible reaction. After 3 minutes, a second bronchial wash

using 20 mL of saline was performed and aspirated back after

minimal dwell time.

The active (AMP) challenge was then performed in the other site.

Again a baseline bronchial wash was performed using 20 mL saline

and immediately aspirated back under gentle suction. Then the active

challenge with 5 mL AMP was performed. The initial AMP con-

centration administered was one tenth that which caused a 20% fall in

FEV1 on the prior inhalational challenge or if the subject had been

unresponsive to adenosine one tenth of the maximum concentration

during the inhalation challenge (400 mg/mL). Up to two subsequent

AMP doses were given at quadrupling concentrations, the maximum

administered endobronchial dose being 400 mg/mL. There was

a time lapse of 3 minutes after each concentration given to observe

for any visual reaction using the analogue outlined in Table I. The

endobronchial challenge was terminated either when there was a

visible reaction to AMP, when the maximum concentration of

AMP had been administered or when it was necessary to terminate

the challenge for reasons of patient comfort. Three minutes after the

final concentration of adenosine had been administered, a further

bronchial wash of 20 mL of saline was performed and aspirated

after a minimum dwell time. Subjects remained under observation for

a period of at least two hours after the procedure.

Processing of samples

A total cell count was measured using a modified Neubauer

hemocytometer and was expressed as the number of cells 3105/mL

of BAL. Cell viability was assessed by Trypan blue exclusion

staining. Viable cells are expressed as a percentage of total cell

numbers. Samples were centrifuged at 2003 g for 10 minutes at 4Cto separate any debris and added to a protease inhibitor cocktail

(see Appendix) and stored at 270C for subsequent analysis.

Neuropeptide measurement

NKA was measured using radioimmunoassay, utilizing a

N-terminal specific anti-serum that was raised in guinea pigs to

synthetic human NKA (Amersham Bioscience UK Ltd product

number IM168, Buckinghamshire, UK). It cross-reacts fully with

NKB and NPK but less than 0.1%with SP. The detection limit for the

assay is 2 ng/L.

CGRP immunoreactivity was measured using a commercial

CGRP human radioimmunoassay (RIA) kit (catalogue number

RIK009, Peninsula Laboratories, San Carlos, Calif). This antibody

is a rabbit anti-human CGRP peptide (II) antibody. The label was125I-Tyr0-CGRP (catalog number Y6011). The limit of detection for

this assay is 2 ng/L and the antibody cross-reacts 100% with human

CGRP (II), human CGRP, and rat CGRP. It cross-reacts <0.001%

with rat calcitonin C-terminal adjacent peptide and less than 0.02%

with insulin, glucagon, somatostatin, SP, vasoactive intestinal pep-

tide, and gastrin releasing peptide.

Substance P (SP) was measured using a commercially available

ELISA (catolog number DE1400, R&D Systems, Minneapolis,

Minn). It shows no significant cross-reactivity with NKA, neurokinin

B (NKB), and neuropeptide K (NPK). The limit of detection of this

assay is 8 pg/mL.

For radioimmunoassays, lavage fluid was extracted using a

previously validated technique.16 In brief, cleared plasma and

traysolol were added to equal volume of lavage fluid followed by

precipitation of large molecular weight proteins in 60% alcohol and

the sample centrifuged (30 min,3 1500 g). Thiomersal was added to

the supernatant. This was then decanted, the extract was evaporated to

TABLE I. Visual analogue reaction for grading of response

after endobronchial challenge

Analogue

score Reaction observed

0 No reaction

1 Subject coughs after instillation of AMP

(no coughing after saline challenge)

2 Subject coughs/immediate bronchial pallor

then hyperemia/increased mucus secretion

after instillation of AMP

3 Bronchoconstriction observed after instillation

of AMP

Adapted from Polosa et al.1



Crummy et al 313

Mech

anismsofasthmaand

allerg

icinflammation

dryness and the sample assayed. For SP ELISA, samples were Sep-

pakked (C18 Sep-pak; Waters, Milford, Mass) and eluted using a

60:40 solution of acetontirile and 1% trifluoro-acetic acid, dried down

and reconstituted in buffer prior to assay. Using these extraction and

assay methods, peptide recovery was >90%. We have previously

characterised NKA, SP, and CGRP immunoreactivity in bronchoal-

veolar lavage fluid using HPLC, confirming it to be target peptide.17

Measurement of osmolality ofAMP solutions

AMPwas freshlymade up in 0.9%normal saline in concentrations

ranging from 0.39 mg/mL to 400 mg/mL and osmolality measured

(Advanced Micro-osmometer 3300; Advanced Instruments Inc,

Pomona, Calif).


All statistical analysis was performed using SPSS version 11.0

(Statistical Package for Social Sciences, Chicago, Ill). All data were

tested for normality of distribution using Shapiro-Wilk W tests.

Where the data were skewed values are quoted as median and inter-

quartile range (IQR), unless otherwise stated. For non-parametric

data, comparisons were made between paired samples using the

Wilcoxon analysis and for unpaired samples using the Mann-

Whitney U test. Comparison between more than two groups were

performed using Kruskall-Wallis analysis. Post hoc multiple com-

parisons were then performed to demonstrate the underlying statis-

tical differences (Stats Direct, Cambridge, UK). Non–parametric

correlations were calculated using Spearman’s rank correlations.

Subjects who did not achieve at least a 20% drop in FEV1 after

inhaling 400 mg/mL of AMP were given an assigned PC20 value of

640 mg/mL for statistical analysis. P values <.05 were regarded as

statistically significant.

RESULTS

A total of 24 subjects were recruited. Three subjects(1 normal, 1 atopic, and 1 asthmatic) did not completethe endobronchial challenge. Twenty-one subjects com-pleted the endobronchial challenge protocol (1 samplewas inadequate for processing). The demographic detailsfor the remaining 20 subjects (7 normals, 6 atopicnon-asthmatics, and 7 atopic asthmatics) are shown inTable II. There was no difference between the groups interms of age or baseline spirometry.

Measurements of osmolality of saline and AMP solu-tions showed the following: 0.9% saline, 281 mOsm;0.39 mg/mL, 282 mOsm; 3.125 mg/mL, 297 mOsm;

12.5 mg/mL, 326 mOsm; 40 mg/mL, 403 mOsm; 100mg/mL, 567 mOsm; 400 mg/mL, 1292 mOsm.

Endobronchial responses to AMP challenge are shownin Figure 1. Instillation of endobronchial AMP led toimmediate coughing in all subjects after administration ofadenosine, which was not seen after instillation of thesaline placebo challenge. However, there was no evidenceof generalised lung involvement (eg, wheeze, hypoxia)in response to the endobronchial AMP challenge in anysubject.

As immediate coughing was observed in all groups,for the purposes of analysis all subjects were consideredtogether. There was a significant increase in median (IQR)NKA and substance P levels [10.0 (5.0–15.0) vs. 20.0(10.6–25.0) pg/mL, P < .01 and 227. 8 (176.9–278.6) vs.318.1 (190.6–422.9) pg/mL, P < .01] respectively whenpost-saline and post-AMP levels were compared (Fig 2).There was no significant change in CGRP levels (P=.37,Fig 2). There was no significant difference betweengroups in NKA, CGRP, or substance P levels at baselineor post-challenge.

There was no correlation between the change (post-AMP–post-saline) in NKA and substance P after AMPchallenging (r = 0.27, P=.25) or between the changein CGRP and substance P (r = 0.25, P=.34). There wasno significant correlation between the change in NKA orsubstance P and the PC20 AMP for the group as a whole(r = 0.09, P=.71, r = 0.26, P=.31 respectively). There wasno difference between the changes in NKA observedbetween those who bronchoconstricted after endobron-chial challenge (visual reaction 3) and those who did not(P=.30).

DISCUSSION

This study provides the first evidence of in vivotachykinin release after a chemical airway challenge inhumans. During the study it was observed that many of thesubjects who were sedated and who did not cough duringthe placebo saline challenge or baseline lavage at theactive challenge site, started to cough immediately afterthe instillation of endobronchial AMP. In three cases thiscoughing induced by AMP was believed to be distressingenough to necessitate termination of the procedure.

TABLE II. Demographic details of subjects

Normals (N) Atopics (AT) Asthmatics (AS)

Number 7 6 7

Age (y), Median (IQR) 21 (21-24) 21 (21-24) 26 (21-30)

Sex (n = male) 4 3 2

FEV1% predicted, Mean (SD) 114.0 (8.4) 99.8 (14.2) 90.0 (14.8)

FVC% predicted, Mean (SD) 97.7 (11.0) 99.2 (15.4) 95.6 (16.4)

PC20 AMP (mg/mL), Median (IQR) 640.0 413.9 (144.8-640.0) 3.6 (2.3-7.9)

Maximum dose of endobronchial AMP (mg/mL), Median (range) 160.0 (40.0-400.0) 40.0 (10.0-160.0) 1.3 (0.15-10.0)

Number of aliquots of endobronchial AMP, Median (range) 2 (1.0-3.0) 11 (1.0-2.0) 11 (1.0-3.0)


AUGUST 2005

314 Crummy et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

The immediacy of the cough after AMP challengesuggests that the cough is mediated by afferent sensorynerves. It has previously been shown that AMP acts onvagal nerves.7,8,9 The co-existent increase in NKA andsubstance P suggest the mechanism may involve stimu-lation of sensory C fibres with antidromic release oftachykinins.

Stimulation of sensory C fibers and release of neuro-peptides have previously been implicated in cough. It isknown that inhalation of capsaicin acting through theVR-1 receptor causes cough.18,19 Capsaicin also liberatesneuropeptides from sensory nerves and after repeatedinhalation the cough diminishes, possibly due to depletionof neuropeptides and suggesting propagation of coughby neuropeptides in this model. There is evidence, inanimal models, that AMP can act directly on pulmonaryC fibers11; however, the exact receptor involved seems tovary between species. In rats it was found that adenosinepotentiated the response of pulmonary C fibers to chem-ical stimuli and that this response was attenuated by pre-treatment with an A1 receptor antagonist20; while inguinea pig lung A2 agonists decreased the release of SPfrom pulmonary C fibers.21 Thus, it is possible in thisstudy that AMP was acting directly on nerve endings tocause cough via a central mechanism rather than directlyby neuropeptide release, although this potential mecha-nism has not been well studied in the human airway.

The coughing after endobronchial AMP challenge andsubsequent release of tachykinins was observed across allgroups of subjects. While it has been suggested that thenerves of asthmatics may be more sensitive to stimuli than

those of non-asthmatics, the asthmatics in this study weremild both in terms of airway hyperresponsiveness andtreatment requirements and may not reflect differencesthat may occur with more significant asthma.

There was no correlation between the changes in NKA,CGRP, or substance P. This is perhaps not surprising inthe case of NKA, as this neuropeptide may be locatedindependently of the other two peptides.22 However, it issurprising that there was no correlation between thechanges seen in substance P and CGRP because they areoften co-localized in the same neurones. There are anumber of possible explanations for this discrepancy.While neuropeptides are co-localized the exact amountspresent may vary depending on the location of the tissue.23

There may also be differential paths of degradation.Substance P and NKA are primarily degraded in theairways by the enzyme neutral endopeptidase (NEP).However, the exact degradation pathway of CGRP hasnot yet been elucidated. It has been shown that CGRP issubject to degradation by tryptase while the tachykininsNKA and substance P are not.24 Most of the subjects inthis study had tryptase release after endobronchial AMPchallenging,15 which may explain why there was nodifference in CGRP levels after AMP challenge and alsowhy there is no relationship between CGRP and SP levels.In addition, it may be that differential release accounts forthe patterns of neuropeptide release observed. The relativeamount of neuropeptide released may be dependent onthe frequency with which the nerve is stimulated as well asthe quantity of each peptide in the nerve ending.25 How-ever, the precise mechanisms remain to be established.

FIG 1. The frequency of endobronchial response to AMP challenge in normals, atopics, and asthmatics.



Crummy et al 315

Mech

anismsofasthmaand

allerg

icinflammation

Another possibility is that while AMP mediates thecough through afferent sensory nerves, the tachykininrelease is from a non-neuronal source. Sensory nervescontaining neuropeptides account for around 1% of all thenerve fibers found in human airways.22 However, muchhigher levels of neuropeptides have been found in inducedsputum 26 and in BAL17 than would be expected from thissource alone. It has been shown that tachykinins may beproduced by eosinophils, macrophages, lymphocytes,neutrophils, and epithelial cells,27,28,29,30 and it may be

that these non-neuronal sources are responsible for theneuropeptide release induced by AMP in this study. Thus,the neuropeptide release and the cough may be twoconcomitant but separate events in the airway after AMPchallenge.

Nasal challenge with hypertonic saline induces neuro-peptide release31 and mannitol challenge in asthmaticscauses cough that is independent of bronchoconstric-tion.32 Osmolality of the adenosine concentrations dem-onstrated that the challenges used in normal subjects wererelatively hyperosmolar compared to normal saline/plasma; however, this was not the case for the concen-trations used in asthmatic subjects (Table II). Thus, whilewe cannot completely exclude an osmotic effect in normalsubjects, given this mechanism is not applicable in asth-matic subjects, we believe it is unlikely that two entirelyseparate mechanisms are producing cough and causingtachykinin release.

There was no relationship between PC20 AMP and thechange in NKA, SP, or CGRP levels. Bronchoconstrictionafter inhalation of AMP is mediated by multiple mecha-nisms including the release of mast cell mediators, whichwere present in many of these subjects.15 Thus a simplecorrelation between changes in airway physiology andindividual endobronchial tachykinin or other mediatorrelease would seem unlikely.

This study has provided evidence that endobron-chial AMP challenge causes immediate cough and sig-nificant NKA and substance P release in non-atopicnon-asthmatic, atopic non-asthmatic, and atopic asthmaticsubjects, which occurs in the absence of significantCGRP release. This is the first demonstration of in vivotachykinin release, after chemical stimulation, in thehuman airway. It therefore seems likely that AMP canact through a number of mechanisms, in addition to mastcell mediator release, to generate responses in the humanairway.

REFERENCES

1. Polosa R, Ng WH, Crimi N, Vancheri C, Holgate ST, Church MK, et al.

Release of mast-cell-derived mediators after endobronchial adenosine

challenge in asthma. Am J Respir Crit Care Med 1995;151:624-9.

2. Driver AG, Kukoly CA, Ali S, Mustafa SJ. Adenosine in bronchoalve-

olar lavage fluid in asthma. Am Rev Respir Dis 1993;148:91-7.

3. Mann JS, Holgate ST, Renwick AG, Cushley MJ. Airway effects of

purine nucleosides and nucleotides and release with bronchial provoca-

tion in asthma. J Appl Physiol 1986;61:1667-76.

4. Cushley MJ, Tattersfield AE, Holgate ST. Inhaled adenosine and

guanosine on airway resistance in normal and asthmatic subjects. Br J

Clin Pharmacol 1983;15:161-5.

5. Holgate ST, Cushley MJ, Mann JS, Hughes P, Church MK. The action

of purines on human airways. Arch Int Pharmacodyn Ther 1986;280(2

Suppl):240-52.

6. Spina D. Airway sensory nerves: a burning issue in asthma? Thorax

1996;51:335-7.

7. Crimi N, Palermo F, Oliveri R, Polosa R, Settinieri I, Mistretta A.

Protective effects of inhaled ipratropium bromide on bronchoconstriction

induced by adenosine and methacholine in asthma. Eur Respir J 1992;5:

560-5.

8. Polosa R, Phillips GD, Rajakulasingam K, Holgate ST. The effect of

inhaled ipratropium bromide alone and in combination with oral terfen-

adine on bronchoconstriction provoked by adenosine 5#- monophosphate

and histamine in asthma. J Allergy Clin Immunol 1991;87:939-47.

FIG 2. The change in post-saline and post-AMP levels of (A) NKA,

(B) Substance P, and (C) CGRP.


AUGUST 2005

316 Crummy et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

9. Marks GB, Yates DH, Sist M, Ceyhan B, De Campos M, Scott DM,

et al. Respiratory sensation during bronchial challenge testing with

methacholine, sodium metabisulphite, and adenosine monophosphate.

Thorax 1996;51:793-8.

10. Polosa R, Rajakulasingam K, Church MK, Holgate ST. Repeated inhala-

tion of bradykinin attenuates adenosine 5#-monophosphate (AMP) in-

duced bronchoconstriction in asthmatic airways. Eur Respir J 1992;5:

700-6.

11. Hong JL, Ho CY, Kwong K, Lee LY. Activation of pulmonary C fibres

by adenosine in anaesthetized rats: role of adenosine A1 receptors.

J Physiol 1998;508:109-18.

12. Skin tests used in type I allergy testing Position paper. Sub-Committee

on Skin Tests of the European Academy of Allergology and Clinical

Immunology. Allergy 1989;44(Suppl 10):1-59.

13. Standardization of spirometry–1987 update. Statement of the American

Thoracic Society. Am Rev Respir Dis 1987;136:1285-98.

14. Cockcroft DW, Killian DN, Mellon JJ, Hargreave FE. Bronchial

reactivity to inhaled histamine: a method and clinical survey. Clin

Allergy 1977;7:235-43.

15. Crummy F, Livingston M, Ennis M, Heaney LG. Mast cell mediator

release in non-asthmatic subjects after endobronchial adenosine chal-

lenge. J Allergy Clin Immunol 2004;114:34-9.

16. Heding LG. Radioimmunological determination of pancreatic and gut

glucagon in plasma. Diabetologia 1971;7:10-9.

17. Heaney LG, Cross LJ, McGarvey LP, Buchanan KD, Ennis M, Shaw C.

Neurokinin A is the predominant tachykinin in human bronchoalveolar

lavage fluid in normal and asthmatic subjects. Thorax 1998;53:357-62.

18. Bevan S, Szolcsanyi J. Sensory neuron -specific actions of capsiacin:

mechanisms and applications. Trends Pharmacol Sci 1990;11:330-3.

19. Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD,

Julius D. The capsaicin receptor: a heat activated ion channel in the pain

pathway. Nature 1997;389:816-24.

20. Gu Q, Ruan T, Hong JL, Burki N, Lee LY. Hypersensitivity of

pulmonary C fibres induced by adenosine in anaesthetized rats. J Appl

Physiol 2003;95:1315-24.

21. Morimoto H, Yamashita M, Imazumi K, Matsuda A, Ochi T, Seki N, et al.

Effects of adensoine A2 receptor agonists on the excitation of capsaicin-

sensitive afferent sensory nerves in airway tissues. Eur J Pharmacol 1993;

240:121-6.

22. Joos GF, Germonpre PR, Pauwels RA. Role of tachykinins in asthma.

Allergy 2000;55:321-37.

23. Bakhle YS, Bell C. Neurokinin A and substance P vary independently in

different regions of rat sensory neurons. Neuropeptides 1995;28:237-41.

24. Sommerhoff CP. Mast cell tryptases and airway remodeling. Am J Respir

Crit Care Med 2001;164:S52-8.

25. Widdicombe JG. Autonomic regulation. i-NANC/e-NANC. Am J Respir

Crit Care Med 1998;158:S171-5.

26. Tomaki M, Ichinose M, Miura M, Hirayama Y, Yamauchi H, Nakajima

N, et al. Elevated substance P content in induced sputum from patients

with asthma and patients with chronic bronchitis. Am J Respir Crit Care

Med 1995;151:613-7.

27. Maggi CA. The effects of tachykinins on inflammatory and immune

cells. Regul Pept 1997;70:75-90.

28. Joos GF, Germonpre PR, Pauwels RA. Neural mechanisms in asthma.

Clin Exp Allergy 2000;30(Suppl 1):60-5.

29. Joos GF, De Swert KO, Pauwels RA. Airway inflammation and

tachykinins: prospects for the development of tachykinin receptor

antagonists. Eur J Pharmacol 2001;429:239-50.

30. Reynolds PN, Scicchitano R, Holmes MD. Pre-protachykinin-A mRNA

is increased in the airway epithelium of smokers with chronic bronchitis.

Respirology 2001;6:187-97.

31. Baraniuk JN, Mushtaq A, Atsushi Y, Sheen-Yie F, Naranch K. Hyper-

tonic saline nasal provocation stimulates nocioceptive nerves, substance

P release and glandular mucus exocytosis in normal humans. Am J


32. Koskela HO, Hyvarinen L, Brannan JD, Chan HK, Anderson SD.

Coughing during mannitol challenge is associated with asthma. Chest

2004;125:1985-92.

APPENDIX

Protease inhibitor cocktail contained the followingreagents at recommended concentrations:

Soybean Trypsin Inhibitor - BDH Laboratory Supplies,Poole, UK

Aprotinin - Sigma-Aldrich Ltd., Poole, UKAlpha-1-antitrypsin - Sigma-Aldrich Ltd., Poole, UKPepstatin A - BDH Laboratory Supplies, Poole, UKPhenanthroline - Acros Organics, Geel, BelgiumEDTA acid - Acros Organics, Geel, BelgiumBenzamidine - Sigma-Aldrich Ltd., Poole, UK



Crummy et al 317

Mech

anismsofasthmaand

allerg

icinflammation

Rat tracheal epithelial responses to wateravoidance stress

Hiroshi Akiyama, PhD,a Hiroo Amano, MD, PhD,b and John Bienenstock, MDc Tokyo

and Maebashi, Japan, and Hamilton, Ontario, Canada

Background: Psychologic stress has major effects on many

organs and cellular systems. The hypothalamic-pituitary-

adrenal axis, corticotropin-releasing factor (CRF), mast cells,

and nerves have all been shown to be involved in intestinal

epithelial responses to stress. There has been little information

in the literature on stress and the lung.

Objective: To investigate Wistar rat tracheal epithelial

responses to acute water avoidance stress (1 hour).

Methods: Tracheal tissue was examined in Ussing chambers.

Results: Increases in short-circuit current, but not in

conductance, occurred after stress and were inhibited by

previous injection of the CRF 1 and 2 receptor antagonist,

a-helical CRF–(9-41). Electron microscopic morphologic

evidence for tracheal mast cell activation and degranulation

was found after stress. Stress and CRF injection both enhanced

responses to substance P, but these effects were not inhibited by

a-helical CRF.

Conclusion: The data suggest that acute stress affects tracheal

epithelium and sensitizes it to enhanced responses to substance P,

partly through mast cell activation. Many but not all of these

effects are mediated by CRF. These results offer the possibility

that stress may be involved in inflammatory diseases of the lung

such as asthma. (J Allergy Clin Immunol 2005;116:318-24.)

Key words: Mast cell, hypothalamic-pituitary-adrenal axis, short-

circuit current, corticotropin-releasing factor, substance P

The role of stress in asthma is a controversial subject,which is reflected by the literature on this subject.1 Arecent investigation of college students clearly showedthat stress associated with examinations enhanced severaloutcome measures such as sputum eosinophils, eosino-phil-derived neurotoxin levels, and IL-5. The authorsconcluded that stress could act as a cofactor with inhaledantigen to enhance inflammation and asthma severity.2

Asthma exacerbations in children increase significantlywithin 1 to 2 days of a major stressful event. This effect

was statistically evident for as long as 5 to 7 weeks.3

Indeed, it has become increasingly evident in recent yearsthat the effects of acute and chronic stress on physiologicfunction can be important in the initiation, development,and/or perpetuation of many human diseases.4 The intes-tine has long been thought of as a target for stress. The roleof stress as a determinant in intestinal disease is wellestablished both experimentally and clinically5-7 and maywell be relevant to the study of asthma.8 Irritable bowelsyndrome, as in asthma, is characterized by smoothmuscle hyperreactivity.9 The major prospective determi-nant for irritable bowel syndrome after acute infectiousdiarrhea is accompanying poor psychosocial circumstan-ces.10 Activation of mast cells colocalized with entericnerves11,12 is a significant hallmark of disease. Finally,experimental acute and chronic stress causes intestinalhypersecretion, permeability, mucin release, and smoothmuscle hyperreactivity.5-7,13-16 These effects seem to bemediated peripherally by neuropeptides such as substanceP and neurotensin, as well as by the autonomic nervoussystem, and centrally by corticotrophin-releasing factor(CRF). Despite this evidence for stress effects on theintestine, little attention has been paid experimentally tothe possible effects of stress on lung function,17 althoughthe effect of stress on the development and expression ofatopy has begun to receive attention.18 This reportrepresents one of the first such attempts. In it we showthat acute stress, mediated in part through the action ofCRF, has a stimulatory effect on ion secretion by rattracheal epithelium, and is accompanied by evidence formast cell activation and degranulation. At the same time,both stress and CRF prepare (sensitize) tissues for en-hanced subsequent responses to substance P (SP). Thesestudies may have important implications for the role ofstress in lung diseases such as asthma.

METHODS

Details of the methods can be found in the Journal’s Online

Repository (www.mosby.com/jaci).

Abbreviations useda-Helical CRF: a-Helical CRF–(9-41)

CRF: Corticotropin-releasing factor

Isc: Short-circuit current

PD: Potential difference

SP: Substance P

WAS: Water avoidance stress

From aDivision of Foods, National Institute of Health Sciences, Tokyo; bthe

Department of Dermatology, Gunma University Graduate School of

Medicine, Maebashi; and cthe Departments of Medicine and Pathology

and Molecular Medicine, McMaster University and the Brain-Body

Institute, St Joseph’s Healthcare, Hamilton.

Dr Akiyama and Dr Amano contributed equally to this work.

Supported by the Medical Research Council of Canada and St Joseph’s

Hospital Foundation, Hamilton, Ontario, Canada.


accepted for publication March 28, 2005.


Reprint requests: Hiroshi Akiyama, PhD, Division of Foods, National Institute

of Health Sciences, 1-18-1, Kamiyoga, Setagaya-ku, Tokyo 158-8501

Japan. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.040

318

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n


Animals

Specific pathogen-freemaleWistar rats (300-420 g; CharlesRiver,

St Constant, Quebec, Canada) were maintained on a 12-hour:

12-hour light-dark cycle. Rats were provided with food and water

ad libitum and handled daily for 1 week before the study to minimize

effects of stress occasioned by human handling. All experiments

were approved by the Animal Care Committee at McMaster

University.

Stress model

Water avoidance stress (WAS) was chosen as the model of stress.

This model appears to have similar effects and modes of action

insofar as intestinal responses and the hypothalamic-pituitary-adrenal

axis are concerned.5,19 Stress sessions were performed between 1000

and 1300 hours to minimize diurnal variations in response.

Drugs

Corticotrophin-releasing factor and a-helical CRF–(9-41)

(a-helical CRF), a CRF 1 and 2 antagonist (Peninsula Lab, Belmont,

Calif), were dissolved in PBS according to the manufacturer’s

instructions, aliquoted, and kept frozen at 270C until used. The

neutral endopeptidase inhibitor, phosphoramidon, was obtained from

SigmaChemicalCo (StLouis,Mo) anddissolved inPBS.SP(Peninsula

Lab) was dissolved in PBS, aliquoted, and stored at 0C before use. All

drugs were used fresh. Immediately before the experiments, the drugs

were diluted in PBS to the appropriate concentrations.

Ussing chamber experiments

Trachea. The trachea was removed, immersed in 37C oxygen-

ated Krebs buffer, opened along the anterior margin, and immediately

mounted in Ussing chambers. Ussing chambers measure current

across living tissue. In the case of tracheal tissue, the current

measured reflects chloride ion secretion by the epithelium. Tissue

was mounted in an Ussing chamber and the spontaneous potential

difference (PD) clamped at 0 volts (WP Instruments automated volt-

age clamp; WP Instruments, Nacro Scientific, Mississauga, Ontario,

Canada). The injected current, the short-circuit current (Isc in mA/

cm2), required to maintain 0 PD, was continuously measured and

indicates net active ion transport. At intervals, the PD across the tissue

was measured to allow calculation of tissue ion conduction (indicates

net passive ion flux across the preparation) using the Ohm law and the

PD and Isc values. Baseline values for Isc, an indicator of ion

secretion, and G, an indicator of ion permeability, were measured 15

minutes after mounting the tissues, and then every 5 minutes for 40

minutes. The tissue was considered equilibrated after 15 minutes.

Abnormal baseline values of G > 50 millisiemens/cm2 were consid-

ered damaged and were excluded.

Colon. Tissue was treated as described elsewhere.5

Tracheal responses to SP in vitro

Substance P in amounts of 10 mL to a final concentration of

1026 mol/L was added to the luminal or the serosal buffer of the

Ussing chambers 50 minutes after mounting. After adding SP to the

luminal buffer, immediate, sharp increases in Isc were observed.

The Isc was calculated as the difference between the baseline Isc

value and the maximum value within 10 seconds of addition of SP. In

all experiments, phosphoramidon (final concentration: 1025 mol/L)

was added to the chamber buffer 10 minutes before adding SP.

Electron microscopy

The tissue was fixed in 2% glutaraldehyde in 0.1 mol/L sodium

cacodylate buffer (pH 7.4) for 2 hours at room temperature, washed,

left overnight at 4C in 0.2 mol/L sodium cacodylate buffer, pH 7.4,

and then washed 3 times in the same buffer. Samples were postfixed

in 2% osmium tetroxide for 1 hour, dehydrated in graded ethanol,

and embedded in Spurr resin. The sections were photographed at a

magnification of 30003 in a transmission electron microscope

(JEOL 1200 EX, Tokyo, Japan). At least 20 mast cells from each

rat were analyzed by 2 investigators who were blind to the

experiments. Interobserver error did not vary by more than 3%.

Activation of mast cells was defined by the presence of granules with

altered or absent electron-dense content.

Fecal pellet output

The number of fecal pellets expelled by each rat per hour was

used as an indirect measure of colonic motility, as previously

described.13,20

Experimental design and pharmacologicaltreatments

A pilot study showed that colonic epithelial abnormalities were

maximal at 1 hour after initiation of WAS. The rats were injected

intraperitoneally with CRF (50 mg/kg) 1 hour before mounting the

tissue. Others had previously determined that this dose was maximal

in intestinal Isc responses.5 The dose (250 mg/kg) of a-helical CRF

has been shown to block intestinal responses toWAS effectively5 and

was injected intraperitoneally 30 minutes before the WAS protocol.

Measurement of serum corticosterone

Levels were determined by theHPLCmethod ofWong et al21 with

some minor modifications.

Statistics

The data were expressed as means 6 SEMs. A P value of less

than .05was considered significant.Differences between the valueswere

tested by using the Student t test or the Scheffe method after ANOVA.22

RESULTS

Corticosterone levels

For corticosterone levels, see Fig E1 in the OnlineRepository (www.mosby.com/jaci). The results from theHPLC standards were linear over a dose range of 10 to1000 ng/mL carbon-treated serum. Control values innormal rats were 102.6 6 19.5. It was evident that thesham animals were undergoing mild stress, because theirlevels were 2706 56.1, which differed significantly from

FIG 1. Effect of WAS on tracheal baseline Isc. Rats were subjected

to WAS or sham stress for 1 hour or injected intraperitoneally with

CRF (50 mg/kg) or a-helical CRF (250 mg/kg) 30 minutes before

stress or CRF protocol. Bars represent means 6 SEMs; n = 4 to 12

rats/group. **P < .01 vs control; P < .05 vs stress;CP < .05 vs CRF.



Akiyama, Amano, and Bienenstock 319

Mech

anismsofasthmaand

allerg

icinflammation

the controls (P < .05). The levels in stressed animals(785.1 6 121.8) differed significantly both from sham(P< .01) and control (P< .01) groups. The responses to anintraperitoneal injection of CRF (50 mg/kg) in PBS 1 hourbefore mounting the tissues in Ussing chambers were912.6 6 201.8, which again differed significantly fromcontrol (P < .01). The injection of a-helical CRF beforethe stress or CRF protocol reduced the levels to approx-imately the same as those of the sham (234.26 29.4). Thislevel, however, differed significantly from the controlvalues (P < .05).

Defecation during stress and responseto a-helical CRF

See Fig E2 in theOnline Repository (www.mosby.com/jaci). Measurement of the number of fecal pellets per hourcorresponded to values expected in controls (0.5 6 0.3),sham (1.1 6 1.0), or stress (6.4 6 0.7) and those animalspretreated with a-helical CRF (3.8 6 1.0). The numberseen in the stress group differed significantly from bothcontrol and sham groups (P < .01), whereas the animalspretreated with a-helical CRF had a reduced but notsignificantly reduced number of pellets.

Effects of stress, CRF, and a-helical CRFon colonic epithelial physiology

See Fig E3 in theOnline Repository (www.mosby.com/jaci). Baseline Isc responses 15 minutes after mountingwere approximately the same in control (30.7 6 6.3) andsham groups (37.7 6 6.1). WAS significantly increasedIsc relative to sham (P< .05) and control groups (P< .01).Intraperitoneal injection of CRF (25 mg/kg in saline)confirmed that this produced significant increases in Iscresponses. a-Helical CRF had no effect on colonic Isc5

and reduced the effects of stress to baseline.5

Effects of stress, CRF, and a-helical CRFon tracheal epithelial physiology

Stress caused increases in Isc compared with both sham(P < .05) and control groups (P < .01; Fig 1). The effect ofintraperitoneal injection of CRF (50 mg/kg) 1 hour beforethe tissue was mounted in the Ussing chamber wassignificant compared with sham (P < .01) and control(P < .01) groups. This concentration has been shownby others5 to have maximal effects on colon epithelial

physiology. The time point of 1 hour was chosen toexamine the effect of CRF on the trachea because thestress exposure lasted for 1 hour. We saw no differenceat 2 or 4 hours from that seen at 1 hour. There wasno significant difference between any of the groupsin conductance values (see Table E1 in the OnlineRepository at www.mosby.com/jaci). Pretreatment witha-helical CRF reduced the Isc seen in stress to the levelsof control and sham (P < 0.05). The effects of a-helicalCRF (42.8 6 6.4) alone did not differ significantlybetween control (43.5 6 1.9) and sham (41.8 6 3.4)groups. No changes in conductance were seen as a resultof treatment with a-helical CRF in any of the groupsexamined.

Effects of stress and CRF on SP-inducedtracheal epithelial Isc

We wished to establish whether stress sensitizedtissues for responses to other agonists such as SP. SPeffects on the generation of tracheal epithelial Isc havebeen shown by others to be effective only if delivered onthe luminal side.23-25 We confirmed these results (Fig 2)and established that the effect of SP at a concentrationof 1026 mol/L was optimal on tissue equilibrated for15 minutes. No differences were seen with or without theaddition of phosphoramidon. As seen in Table I, bothstress (P < .01) and sham stress (P < .05) significantlyincreased the response to standardized amounts of SPcompared with controls. Similarly, CRF injection (50 mg/kg) intraperitoneally 1 hour before the experiment sig-nificantly increased responsiveness of the tissue to SP(P < .01). No changes in conductance were seen at anytime in these experiments.

Effects of a-helical CRF on stressand CRF effects

Results of experiments with a-helical CRF are shownin Fig 3. a-Helical CRF, which in earlier experiments atthis concentration (250 mg/kg) was effective in reducingstress effects, had no effect on the sensitization inducedby stress for subsequent SP responses. Surprisingly, thecontrol experiment of injection of a-helical CRF by itselfsensitized the trachea equally as well as stress or CRF sothat there were no significant differences compared withcontrol between the stress, CRF, a-helical CRF plus

FIG 2. Effect of SP (1026 mol/L) added to the luminal or serosal sides of trachea on Isc generation in Ussing

chamber. SP was added 50 minutes after mounting. These tracings are representative of 3 similar

experiments.


AUGUST 2005

320 Akiyama, Amano, and Bienenstock

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

stress, and a-helical CRF groups. Pretreatment witha-helical CRF before injection with CRF did not reducethe enhanced response to SP. In fact, the response of thisgroup showed a significant enhancement over CRF itself(P < .05), but not over the a-helical CRF alone.

Effects of stress on mast cell morphology

Acute stress caused mast cell activation (Table II) andin some cases degranulation (Fig 4). There were signifi-cant differences between control and stressed animals(P < .05) in the trachea as well as in the colon (P < .01).Nevertheless, some loss of density of granule contentswas also found in both colon and trachea of controlanimals. If this criterion is used to establish activation,an average of 22% of total mast cell granules in variousnormal tissues, such as skin, ileum, heart, and cranialdura mater,26-30 show signs of activation under normalphysiologic conditions.

DISCUSSION

To the best of our knowledge, this is the first report ofthe effects of psychological stress on lung epithelialphysiology. We have shown that in Wistar rats, acutewater avoidance stress caused increased tracheal epithelialIsc activity, and that this was accompanied by morpho-logic evidence of mast cell activation and degranulation,was mimicked by systemic CRF injection, and could belargely reversed by a CRF receptor competitive inhibitor,a-helical CRF.

There are many reports now that have focused onthe effects of different forms of acute or chronic psy-chological stress on various intestinal physiologic param-eters.6,7,14,15,20,31,32 These include experimental results,obtained mostly16 (but not exclusively) with animal tis-sues, and include effects on fluid or ion secretion andabsorption, permeability,5,15,23,33,34 motility,6,13,20 andpain perception.32 Tissue responses have been shown tovary according to the experimental animal, strain, andtissue source, ie, stomach,35 jejunum,15,31,33 ileum,36 andcolon.5,6,13,32

Thus, a significant body of literature now exists thatsuggests acute stress effects in the intestine are largelymediated by CRF,5,6,37,38 because many of the physio-logical consequences can be reproduced equally eitherby intracerebral injection of CRF6,7,32,38 or through its

peripheral systemic administration.5-7,37,38 These effectsare mostly mediated through mast cell activation, becausepharmacological inhibition of mast cell degranulationprevents many of the consequences of acute stress,5,6,32

and mast cell deficient rats fail to exhibit these find-ings.14,15 Mast cell–nerve interactions and communica-tion11,39,40 have been shown to be involved in thegeneration of a variety of intestinal and lung physiologicresponses to antigen.16,25,39-41 These interactions35 appearto be crucial for the generation of many of the functionalintestinal effects of stress, because SP antagonists6 blockthese effects. They are also inhibited by cholinergic,adrenergic, and autonomic ganglion blockade throughthe use of a variety of well defined pharmacologicagents.5

At no time did we see any increase in conductance, areflection of the integrity of the tracheal epithelium,whereas this is invariably seen as a stress response in theintestine.5,34 This probably reflects a general differencebetween trachea and intestine, because antigen exposureof tracheal tissue from sensitized rats, despite causinglarge increases in Isc25 and mast cell degranulation,21

never showed increased conductance.25,30

Stress effects are complex end results of a rangeof interacting signals between a variety of systems andcells of different types. Indeed, the results obtained witha-helical CRF inhibition of stress on corticosterone levelsare instructive, because the levels of corticosterone werereduced only to those found in sham animals, and stilldiffered significantly from those of controls. This suggeststhat factors in addition to CRF and its pathway areoperative in the production of corticosterone and notinhibited by this antagonist of CRF receptors 1 and 2.42,43

Indeed, behavioral changes apparently mediated by CRF1 receptors can still be seen in transgenic CRF knockoutmice.44 Furthermore, stress-enhanced inflammation in ahapten sensitization model of colitis was shown not to becaused by CRF.45

Corticotropin-releasing factor is known to have bothproinflammatory29,46 and anti-inflammatory47-50 effects.

TABLE I. Effect of stress and intraperitoneal CRF

injection on tracheal responses to SPy

DIsc (mA/cm2)

Control 47.9 6 4.4

Sham 105.1 6 21.7*

Stress 107.1 6 6.4**

CRF 129.5 6 4.1**

Values are means 6 SEMs. The DIsc was calculated as the difference

between the baseline Isc value and the maximum value within 10 seconds

of addition of SP.

**P < .01; *P < .05 vs control.FIG 3. Effect ofWAS,CRF,ora-helicalCRF injectiononSP responses

of the trachea in Wistar rats. Rats were subjected to WAS or sham

stress for 1 hour or injected intraperitoneally with CRF (50 mg/kg).

One hour after the initiation of stress, 90 minutes after a-helical

CRF, or 1 hour after CRF injection alone, tracheas were mounted

in Ussing chambers. SP (1026 mol/L) was added to the luminal

side of the chamber at 50 minutes after mounting. Bars represent

means 6 SEMs; **P < .01, *P < .05 vs control; P < .05 vs CRF.




Mech

anismsofasthmaand

allerg

icinflammation

It has been shown to cause mast cell degranulation wheninjected into the skin.29 It also appears to be involved instress-induced cardiac mast cell degranulation,51 degran-ulation of cranial dura mater mast cells,52 and those of theurinary bladder.53 However, CRF has not previously beenshown to be involved in tracheal or lung mast cellactivation. Because we have shown that the tracheal Iscresponses to stress are inhibited by the a-helical CRF, andbecause stress increased tracheal mast cell degranulation(Table II; Fig 4), it appears likely that directly or indirectly,CRF causes tracheal mast cell degranulation. Mast celldegranulation is one of the first effects of allergen inha-lation. Stress mediated lung mast cell degranulation isanother mechanism for enhancement of the consequenceof allergen inhalation in the sensitized host and wouldfurther promote airways inflammation. Although rat tra-cheal and lung mast cells are a mixture of mucosal andconnective tissue types,54 both seem to respond to stressand CRF, as evidenced by the fact that injection of CRFinto rat skin caused degranulation of primarily connectivetissue–type mast cells,29 and stress caused elevations ofrat mast cell protease II,6 a specific marker of mucosalmast cells.

Our results also showed that both acute stress and theperipheral systemic injection of CRF enhanced the actionof SP on epithelial cell secretion. Joachim et al55 recentlyshowed that stressed mice exhibited increased bronchialhyperreactivity and that this and the increase in airwaysinflammation was caused by SP. Our experimental resultswere anticipated in view of the data of McAlexander andUndem,56 who showed that CRF caused enhancement oftachykinin-induced contractions of guinea pig tracheo-bronchial smooth muscle. However, our results could notbe predicted, because there are now several reportsshowing that CRF can have a protective effect.47-50

These results range from experiments involving antigen-induced plasma extravasation and that induced by SPor vagal stimulation, all the way to stress-induced worsen-ing of a hapten model of colitis48 both in hypo Lewis(LEW/N) and hyper (Fischer, 344/N) CRF responders tostress.57,58 Interpretation of the role of CRF in themodulation of stress effects is complicated by our ownresults, which clearly indicated not only that effects onstress-induced Isc increases were inhibited totally by the

a-helical CRF inhibitor but also that the stress-inducedenhancement of the effect of SP was not inhibited at all bythe a-helical CRF inhibitor. Further complexity to inter-pretation was added by the observation that the a-helicalCRF inhibitor alone enhanced the SP effect but notablyhad no effect itself on Isc generation either in the colon ortrachea. Last, a-helical CRF did not reduce the activity ofCRF in this model but actually enhanced it, confirming itsactivity independent of CRF. We can conclude thatalthough stress effects on Isc may be mediated largelyby CRF, its effects on enhancement of SP may be throughadditional and different mechanisms.

Several possible explanations may be entertained toexplain these unexpected findings. First, a-helical CRFhas weak intrinsic activity37,42 in several different sys-tems.43,59 It might have influenced SP receptor number oraffinity or have agonist activity on new, as yet undiscov-ered CRF receptors expressed in target tissue. That thestress effect itself is, however, not wholly dependent onCRF is further evidenced by our observations that corti-costerone levels after stress in a-helical CRF pretreatedanimals were not reduced to levels seen in controls, butrather only to those seen in sham animals.

The biological significance and relevance of theseexperiments to asthma lie in the primary observationthat acute stress can cause moderate effects on trachealfunction. It seems reasonable to argue teleologically that itwould be detrimental to the host if hypersecretion by the

TABLE II. Mast cell activation during stressy

Tissue Condition

Total

granules

Mast cell

activation

granular change

Total low densityz% Total

activation§

Trachea Control 2631 600 22.5 6 2.0

Stress 3234 1163 35.3 6 2.9*

Colon Control 2728 619 22.0 6 2.0

Stress 2352 940 41.0 6 1.7**

At least 20 mast cells from each rat were evaluated (n = 4 in each group).

Loss of granule electron density content.

§Indicates the percentage of changed granules (means 6 SEMs).

**P < .01; *P < .05 vs control.

FIG 4. Representative electron micrographs of mast cells in rat

tracheal epithelium from control (A) and stress (B) rats. Tracheas

were fixed 15 minutes after mounting in Ussing chambers. Mast

cells from control groups showed some signs of activation. Mast

cells from stress groups were more activated, as indicated by loss

of granule density (arrows point to such granules in B). Magnifi-

cation: 30003.


AUGUST 2005


Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

tracheal epithelium occurred in response to acute stress,whereas intestinal hypersecretionwould not be deleteriousto the host’s survival. Therefore, it may be expected thatthese tracheal responses would indeed be moderate andlikely to be subject to a significant array of physiologicmechanisms of inhibition and regulation. However, stressactivation of epithelium15 may cause upregulation ofcytokines such as IL-8 and possibly eotaxin to accountfor the findings of Liu et al2 that stress enhances sputumeosinophil numbers and products.

Our observation of profound enhancement of the effectof SP as a result of stress may have great significance interms of understanding the mechanisms of lung diseasessuch as asthma. SP and other neuropeptides may havemajor effect in the generation of bronchial smooth musclecontraction experimentally and in asthma.60 Extensivereduction of thresholds for this activity would likely haveprofound deleterious effects in asthma. It is becomingincreasingly recognized that several signals may have tooccur simultaneously, or in prescribed succession, andonly when these occur in genetically vulnerable individ-uals may this constellation of stimuli lead to disease.61,62

In this scenario, acute stress occurring in a particularcontext of infection and genetic susceptibility could havesignificant detrimental consequences to the host.63

We thank Linda Builder, Laurie Nielsen, and Todd Prior for

technical assistance and advice and Professor Yuzo Endo, Dr Ping-

Chang Yang, and the staff of the Electron Microscopy Unit,

McMaster University, for expert assistance. The help of Professor

Jack Rosenfeld with development of the corticosterone assay is

gratefully acknowledged.

REFERENCES

1. Busse WW, Kiecolt-Glaser JK, Coe C, Martin RJ, Weiss ST, Parker SR.

NHLBI Workshop summary: stress and asthma. Am J Respir Crit Care

Med 1995;151:249-52.

2. Liu LY, Coe CL, Swenson CA, Kelly EA, Kita H, Busse WW. School

examinations enhance airway inflammation to antigen challenge. Am J


3. Sandberg S, Jarvenpaa S, Penttinen A, Paton JY, McCann DC. Asthma

exacerbations in children immediately following stressful life events:

a Cox’s hierarchical regression. Thorax 2004;59:1046-51.

4. Csermely P. Stress of life: from molecules to man: proceedings of a

conference: Budapest, Hungary, July 1-5, 1997. Ann N Y Acad Sci

1998;851:1-47.

5. Santos J, Saunders PR, Hanssen NP, Yang PC, Yates D, Groot GA, et al.

Corticotropin-releasing hormone mimics stress-induced colonic epithe-

lial pathophysiology in the rat. Am J Physiol 1999;277:G391-9.

6. Castagliuolo I, LaMont JT, Qiu B, Fleming SM, Bhaskar KR,

Nikulasson ST, et al. Acute stress causes mucin release from rat colon:

role of corticotropin releasing factor and mast cells. Am J Physiol 1996;

271:G884-92.

7. Pothoulakis C, Castagliuolo I, Leeman SE. Neuroimmune mechanisms

of intestinal responses to stress: role of corticotropin-releasing factor and

neurotensin. Ann N Y Acad Sci 1998;840:635-48.

8. Bienenstock J. Stress and asthma: the plot thickens. Am J Respir Crit

Care Med 2002;165:1034-5.

9. Verdu EF, Collins SM. Microbial-gut interactions in health and disease:

irritable bowel syndrome. Best Pract Res Clin Gastroenterol 2004;18:

315-21.

10. Bennett EJ, Piesse C, Palmer K, Badcock CA, Tennant CC, Kellow JE.

Functional gastrointestinal disorders: psychological, social, and somatic

features. Gut 1998;43:256-61.

11. Stead RH, TomiokaM, QuinonezG, Simon GT, Felten SY, Bienenstock J.

Intestinal mucosal mast cells in normal and nematode-infected rat

intestines are in intimate contact with peptidergic nerves. Proc Natl Acad

Sci U S A 1987;84:2975-9.

12. Barbara G, Stanghellini V, De Giorgio R, Cremon C, Cottrell GS, Santini

D, et al. Activated mast cells in proximity to colonic nerves correlate with

abdominal pain in irritable bowel syndrome. Gastroenterology 2004;126:

693-702.

13. Monnikes H, Schmidt BG, Tache Y. Psychological stress-induced

accelerated colonic transit in rats involves hypothalamic corticotropin-

releasing factor. Gastroenterology 1993;104:716-23.

14. Castagliuolo I, Wershil BK, Karalis K, Pasha A, Nikulasson ST,

Pothoulakis C. Colonic mucin release in response to immobilization

stress is mast cell dependent. Am J Physiol 1998;274:G1094-100.

15. Santos J, Benjamin M, Yang PC, Prior T, Perdue MH. Chronic stress

impairs rat growth and jejunal epithelial barrier function: role of mast

cells. Am J Physiol Gastrointest Liver Physiol 2000;278:G847-54.

16. Santos J, Saperas E, Nogueiras C, Mourelle M, Antolin M, Cadahia A,

et al. Release of mast cell mediators into the jejunum by cold pain stress

in humans. Gastroenterology 1998;114:640-8.

17. Joachim RA, Sagach V, Quarcoo D, Dinh QT, Arck PC, Klapp BF.

Neurokinin-1 receptor mediates stress-exacerbated allergic airway in-

flammation and airway hyperresponsiveness in mice. Psychosom Med

2004;66:564-71.

18. Wright RJ, Cohen RT, Cohen S. The impact of stress on the develop-

ment and expression of atopy. Curr Opin Allergy Clin Immunol 2005;5:

23-9.

19. Bonaz B, Tache Y. Water-avoidance stress-induced c-fos expression in

the rat brain and stimulation of fecal output: role of corticotropin-

releasing factor. Brain Res 1994;641:21-8.

20. Barone FC, Deegan JF, Price WJ, Fowler PJ, Fondacaro JD, Ormsbee

HSI. Cold-restraint stress increases rat fecal pellet output and colonic

transit. Am J Physiol 1990;258:G329-37.

21. Wong YN, Chien BM, D’mello AP. Analysis of corticosterone in rat

plasma by high-performance liquid chromatography. J Chromatogr

B Biomed Appl 1994;661:211-8.

22. Hopkins KD, Glass GV, Hopkins BR. Basic statistics for the behavioral

sciences. 2nd ed. Englewood Cliffs (NJ): Prentice-Hall; 1987.

23. Al-Bazzaz FJ, Kelsey G, Kaage WD. Substance P stimulation of chloride

secretion by canine tracheal mucosa. Am Rev Respir Dis 1985;131:86-9.

24. Rangachari PK, McWade D. Effects of tachykinins on the electrical

activity of isolated canine tracheal epithelium: an exploratory study.

Regul Pept 1985;12:9-19.

25. Sestini P, Bienenstock J, Crowe SE, Marshall JS, Stead RH, Kakuta Y,

et al. Ion transport in rat tracheal epithelium in vitro: role of capsaicin-

sensitive nerves in allergic reactions. Am Rev Respir Dis 1990;141:393-7.

26. Rozniecki JJ, Dimitriadou V, Lambracht-Hall M, Pang X, Theoharides

TC. Morphological and functional demonstration of rat dura mater mast

cell-neuron interactions in vitro and in vivo. Brain Res 1999;849:1-15.

27. Singh LK, Pang X, Alexacos N, Letourneau R, Theoharides TC. Acute

immobilization stress triggers skin mast cell degranulation via cortico-

tropin releasing hormone, neurotensin, and substance P: a link to

neurogenic skin disorders. Brain Behav Immun 1999;13:225-39.

28. Spanos C, Pang X, Ligris K, Letourneau R, Alferes L, Alexacos N, et al.

Stress-induced bladder mast cell activation: implications for interstitial

cystitis. J Urol 1997;157:669-72.

29. Theoharides TC, Singh LK, Boucher W, Pang X, Letourneau R, Webster

E, et al. Corticotropin-releasing hormone induces skin mast cell degran-

ulation and increased vascular permeability, a possible explanation for its

proinflammatory effects. Endocrinology 1998;139:403-13.

30. Yang PC, Berin MC, Perdue MH. Enhanced antigen transport across rat

tracheal epithelium induced by sensitization and mast cell activation.

J Immunol 1999;163:2769-76.

31. Yates DA, Santos J, Soderholm JD, Perdue MH. Adaptation of stress-

induced mucosal pathophysiology in rat colon involves opioid pathways.

Am J Physiol Gastrointest Liver Physiol 2001;281:124-8.

32. Gue M, Del Rio-Lacheze C, Eutamene H, Theodorou V, Fioramonti J,

Bueno L. Stress-induced visceral hypersensitivity to rectal distension in

rats: role of CRF and mast cells. Neurogastroenterol Motil 1997;9:271-9.

33. Kiliaan AJ, Saunders PR, Bijlsma PB, Berin MC, Taminiau JA, Groot

JA, et al. Stress stimulates transepithelial macromolecular uptake in rat

jejunum. Am J Physiol 1998;275:G1037-44.




Mech

anismsofasthmaand

allerg

icinflammation

34. Saunders PR, Kosecka U, McKay DM, Perdue MH. Acute stressors

stimulate ion secretion and increase epithelial permeability in rat

intestine. Am J Physiol 1994;267:G794-9.

35. Pare WP. Stress ulcer susceptibility and depression in Wistar Kyoto

(WKY) rats. Physiol Behav 1989;46:993-8.

36. Theoharides TC, Letourneau R, Patra P, Hesse L, Pang X, Boucher W,

et al. Stress-induced rat intestinal mast cell intragranular activation and

inhibitory effect of sulfated proteoglycans. Dig Dis Sci 1999;44:87S-93S.

37. Tache Y, Martinez V, Million M, Rivier J. Corticotropin-releasing factor

and the brain-gut motor response to stress. Can J Gastroenterol 1999;

13(suppl A):18A-25A.

38. Williams CL, Peterson JM, Villar RG, Burks TF. Corticotropin-releasing

factor directly mediates colonic responses to stress. Am J Physiol 1987;

253:G582-6.

39. McKay DM, Bienenstock J. The interaction between mast cells and

nerves in the gastrointestinal tract. Immunol Today 1994;15:533-8.

40. Marshall JS, Bienenstock J. The role of mast cells in inflammatory

reactions of the airways, skin and intestine. Curr Opin Immunol 1994;6:

853-9.

41. Sestini P, Dolovich M, Vancheri C, Stead RH, Marshall JS, Perdue M,

et al. Antigen-induced lung solute clearance in rats is dependent on

capsaicin-sensitive nerves. Am Rev Respir Dis 1989;139:401-6.

42. Fisher L, Rivier C, Rivier J, Brown M. Differential antagonist activity

of alpha-helical corticotropin-releasing factor 9-41 in three bioassay

systems. Endocrinology 1991;129:1312-6.

43. Gulyas J, Rivier C, Perrin M, Koerber SC, Sutton S, Corrigan A, et al.

Potent, structurally constrained agonists and competitive antagonists

of corticotropin-releasing factor. Proc Natl Acad Sci U S A 1995;92:

10575-9.

44. Weninger SC, Dunn AJ, Muglia LJ, Dikkes P, Miczek KA, Swiergiel

AH, et al. Stress-induced behaviors require the corticotropin-releasing

hormone (CRH) receptor, but not CRH. Proc Natl Acad Sci U S A 1999;

96:8283-8.

45. Gue M, Bonbonne C, Fioramonti J, More J, Del Rio-Lacheze C, Comera

C, et al. Stress-induced enhancement of colitis in rats: CRF and arginine

vasopressin are not involved. Am J Physiol 1997;272:G84-91.

46. Karalis K, Sano H, Redwine J, Listwak S, Wilder RL, Chrousos GP.

Autocrine or paracrine inflammatory actions of corticotropin-releasing

hormone in vivo. Science 1991;254:421-3.

47. Gao GC, Dashwood MR, Wei ET. Corticotropin-releasing factor inhibi-

tion of substance P-induced vascular leakage in rats: possible sites of

action. Peptides 1991;12:639-44.

48. Million M, Tache Y, Anton P. Susceptibility of Lewis and Fischer rats to

stress-induced worsening of TNB-colitis: protective role of brain CRF.

Am J Physiol 1999;276:G1027-36.

49. Wei ET, Kiang JG. Inhibition of protein exudation from the trachea by

corticotropin-releasing factor. Eur J Pharmacol 1987;140:63-7.

50. Yoshihara S, Ricciardolo FL, Geppetti P, Linden A, Hara M, Chan B,

et al. Corticotropin-releasing factor inhibits antigen-induced plasma

extravasation in airways. Eur J Pharmacol 1995;280:113-8.

51. Pang X, Alexacos N, Letourneau R, Seretakis D, Gao W, Boucher W,

et al. A neurotensin receptor antagonist inhibits acute immobilization

stress-induced cardiac mast cell degranulation, a corticotropin-releasing

hormone-dependent process. J Pharmacol Exp Ther 1998;287:307-14.

52. Theoharides TC, Spanos C, Pang X, Alferes L, Ligris K, Letourneau R,

et al. Stress-induced intracranial mast cell degranulation: a corticotropin-

releasing hormone-mediated effect. Endocrinology 1995;136:5745-50.

53. Alexacos N, Pang X, Boucher W, Cochrane DE, Sant GR, Theoharides

TC. Neurotensin mediates rat bladder mast cell degranulation triggered

by acute psychological stress. Urology 1999;53:1035-40.

54. Bienenstock J, Befus D, Denburg J, Goto T, Lee T, Otsuka H, et al.

Comparative aspects of mast cell heterogeneity in different species and

sites. Int Arch Allergy Appl Immunol 1985;77:126-9.

55. Joachim RA, Quarcoo D, Arck PC, Herz U, Renz H, Klapp BF. Stress

enhances airway reactivity and airway inflammation in an animal model

of allergic bronchial asthma. Psychosom Med 2003;65:811-5.

56. McAlexander MA, Undem BJ. Enhancement of tachykinin-induced

contractions of guinea pig isolated bronchus by corticotropin-releasing

factor. Neuropeptides 1997;31:293-9.

57. Sternberg EM, Glowa JR, Smith MA, Calogero AE, Listwak SJ,

Aksentijevich S, et al. Corticotropin releasing hormone related behav-

ioral and neuroendocrine responses to stress in Lewis and Fischer rats.

Brain Res 1992;570:54-60.

58. Sternberg EM, Zareie M, Young WS III, Alogero AE, Chrousos GP,

Gold PW, et al. A central nervous system defect in biosynthesis of

corticotropin-releasing hormone is associated with susceptibility to

streptococcal cell wall-induced arthritis in Lewis rats. Proc Natl Acad

Sci U S A 1989;86:4771-5.

59. Menzaghi F, Howard RL, Heinrichs SC, Vale W, Rivier J, Koob GF.

Characterization of a novel and potent corticotropin-releasing factor

antagonist in rats. J Pharmacol Exp Ther 1994;269:564-72.

60. van der Kleij H, Kraneveld AD, Redegeld FA, Gerard NP, Morteau O,

Nijkamp FP. The tachykinin NK1 receptor is crucial for the development

of non-atopic airway inflammation and hyperresponsiveness. Eur J

Pharmacol 2003;476:249-55.

61. Holgate ST. The epidemic of allergy and asthma. Nature 1999;402:B2-4.

62. Holgate ST. Science, medicine, and the future: allergic disorders.

BMJ 2000;320:231-4.

63. Christiansen SC. Day care, siblings, and asthma: please, sneeze on my

child. N Engl J Med 2000;343:574-5.


AUGUST 2005


Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

Allergen-induced substance P synthesis inlarge-diameter sensory neurons innervatingthe lungs

Benjamas Chuaychoo, MD, Dawn D. Hunter, PhD, Allen C. Myers, PhD,

Marian Kollarik, MD, PhD, and Bradley J. Undem, PhD Baltimore, Md

Background: Tachykinins such as substance P are localized in

unmyelinated slow-conducting C fibers that can be activated by

noxious stimuli and tissue inflammation. Substance P is seldom

expressed in fast-conducting large-diameter (A-fiber) vagal

sensory neurons. We have previously found that allergic

inflammation causes a phenotypic change in tachykinergic

innervation of the trachea such that the production of

substance P is induced in large-diameter sensory neurons

projecting mechanosensitive A fibers to the trachea.

Objective: To evaluate whether allergic inflammation also

induces substance P synthesis in large-diameter sensory

stretch-receptor neurons innervating guinea pig lungs, and to

investigate potential mechanisms by which this may occur.

Methods: Sensitized guinea pigs were exposed to allergen

(ovalbumin) aerosol. One day later, immunohistochemical

analysis was performed on vagal sensory neurons that had been

retrogradely labeled from the lungs.

Results: Ovalbumin inhalation caused a significant increase

in substance P expression in large-diameter neurofilament-

positive nodose ganglion neurons that innervate the lungs

(P < .05). This effect was decreased by ipsilateral vagotomy.

Exposing isolated nodose ganglia to the sensitizing antigen,

ovalbumin, also significantly increased substance P expression

compared with control.

Conclusion: Allergic inflammation induces substance P

synthesis in large-diameter (A-fiber) nodose ganglion neurons

innervating guinea pig lungs. This could contribute to the

hyperreflexia seen in allergic airway disease. The full

expression of this phenotypic switch in vagus nodose

ganglion neurons requires intact vagus nerve, but if allergen

reached the systemic circulation in sufficient quantities, it

could also affect substance P synthesis by local activation of

vagal ganglionic mast cells. (J Allergy Clin Immunol

2005;116:325-31.)

Key words: Allergic inflammation, phenotypic switch, neurogenic

inflammation, substance P, nodose ganglion, sensory nerve, vagus

nerve

In both the somatosensory and visceral-sensory system,sensory neuropeptides, exemplified by the tachykininsubstance P (SP), are commonly found stored in theperipheral and central terminals of unmyelinated sensoryC fibers.1-3 Inflammatory reactions in the rat paw havebeen found to evoke a phenotypic switch in the nature oftachykinergic innervation.4,5 After a local inflammatoryreaction in the paw, the preprotachykinin gene is inducedde novo in large-diameter, low-threshold, touch-sensitiveAb fibers.4 Simply touching the paw can then lead toneuropeptide release in the spinal cord that, in theory,could lead to an increase in neurotransmission to theextent that this innocuous touch stimulus is interpretedas a noxious pain-producing stimulus (ie, an allodynia).

A similar inflammation-induced phenotypic switch canoccur in vagal sensory innervation of the airways. In therespiratory tract of most mammals, including guinea pigsand human beings, SP-containing fibers are located mainlyin the upper airways, trachea, and main bronchi.1,6 Thereare relatively few SP containing fibers in the lungs. Inhealthy guinea pig trachea and large bronchi, tachykininsare localized nearly exclusively in nociceptive C fibersderived from small-diameter cell bodies in jugular sensoryganglia.2,3 One day after allergic inflammation in the lungs,SP production is increased in the lungs and in vagal sensoryneurons.7 On further analysis, it was determined that part ofthe increase in SP production occurred de novo in large-diameter neurons that project mechanosensitive Ad fibersto the trachea.8 This same type of phenotypic switch occurs2 to 3 days after respiratory tract viral infection.9

The touch-sensitive Ad fibers in the guinea pig trachearepresent a unique subtype of vagal afferent mechanosen-sors apparently designed to evoke cough on activation.10

They differ from classically described intrapulmonarystretch-sensitive fibers in neurophysiology, activationprofile, and distribution in the lungs.10 The intrapulmonarystretch-sensitive Ab fibers are further subclassified aseither rapidly adapting receptors (RARs) or slowly adapt-ing receptors (SARs). Whether the stretch-sensitive fibersin the lungs can be induced by inflammation to produceneuropeptides in a fashion similar the tracheal touch-sensitive fibers is unknown.

There is also little known about the mechanism bywhich airway inflammation leads to phenotypic changes

From the Johns Hopkins University School of Medicine and Bloomberg

School of Public Health.

Supported by the Heart, Lung and Blood Institute of the National Institutes of

Health (Bethesda, Md) and scholarship funding to Dr Chuaychoo from the

Faculty of Medicine, Siriraj Hospital, Mahidol University (Bangkok,

Thailand).

Received for publication February 13, 2005; revised March 30, 2005; accepted

for publication April 4, 2005.


Reprint requests: Bradley J. Undem, PhD, Johns Hopkins Asthma and Allergy

Center, 5501 Hopkins Bayview Circle, Baltimore, MD 21224. E-mail:

[email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.005

325

Mech

anismsofasthmaand

allerg

icinflammation


Abbreviations usedNGF: Nerve growth factor

RAR: Rapidly adapting receptor

SAR: Slowly adapting receptor

SP: Substance P

Triton-BSA-PBS: PBS containing 0.3 % TritonX-100 and

1% BSA

Trk: Tyrosine kinase receptor

of the neurons located in the remote vagal sensory ganglia(nodose and jugular ganglia situated near the base of thebrain). A logical scenario is that the allergic reaction leadsto the release of certain neurotrophic factors such as nervegrowth factor (NGF) that interact with tyrosine kinasereceptor (Trk) A on the terminals. Activation of neuro-trophin Trk receptors at nerve terminals can signal eventsto the cell body via internalization and transport of theactivated receptors along the nerve axon.11 However,there are alternative mechanisms to consider. The vagalsensory ganglia that contain the cell bodies of vagalsensory neurons are enriched with mast cells. In sensitizedanimals, antigen can activate these ganglionic mast cells,leading to local mediator release and change in neuronalexcitability.12 It is possible, therefore, that the inhaledallergen is not triggering tachykinin synthesis from withinthe lungs, but rather, allergen is systemically absorbed andinteracts with mast cells locally within the vagal sensoryganglia.

In the current study, experiments were performed toaddress 2 hypotheses. First, allergic inflammation canlead to induction of tachykinin synthesis not only intracheal Ad neurons but also in neurons that project Abstretch-sensitive fibers to the lungs. Second, the mecha-nism for allergen-induced SP production in bronchopul-monary A-fiber nodose neurons requires intact vagalnerve fibers.

METHODS

All experiments were approved by the Johns Hopkins Animal

Care and Use Committee. For active sensitization, male Hartley

guinea pigs (Hilltop, Scottsdale, Pa) weighing 100 to 300 g were

immunized by intraperitoneal injection (10 mg/kg) of ovalbumin

(10 mg/mL) dissolved in saline on day 1, day 3, and day 5. On day 21

after the final sensitization, the actively sensitized animals were killed

by CO2 asphyxiation and exsanguinated. The blood was collected,

and serum containing ovalbumin-specific IgG1 was isolated. For

passive sensitization to ovalbumin, male Hartley guinea pigs

weighing 100 to 300 g were sensitized by intraperitoneal injection

of serum (2 mL/kg) containing IgG1 that was collected from guinea

pigs actively sensitized to ovalbumin13 and challenged 1 day after

sensitization. Control guinea pigs were not injected with serum but

challenged in a similar fashion.13 The studies evaluating lung-labeled

afferent neurons, and the ex vivo studies were performed by using

actively sensitized animals. The passive sensitization protocol was

used in the vagotomy studies as a means to decrease the intra-animal

variation in the allergen response. In the in vivo studies, sensitized

(active or passive) and unsensitized (control) animals were exposed

to aerosolized antigen in a Plexiglas chamber (volume, 8 L) with

consecutive increasing concentrations of ovalbumin (0.01%, 0.03%,

0.1% ovalbumin diluted in saline), with 10 minutes of exposure for

each concentration to themaximumdose unless the animal developed

signs of allergic response (gasping; rapid, shallow breathing; or

coughing), whereupon it was removed to breathe ambient air. After

24 hours, the animals were killed with CO2 inhalation and

exsanguinated. We found no difference in the overresponse of the

animal to ovalbumin between actively and passively sensitized

animals. Nonsensitized animals did not respond to the ovalbumin

treatment. The ovalbumin treatment resulted in an eosinophilic

bronchitis within 24 hours of exposure (data not shown).

Identification of nonretrogradely labeledlarge-diameter nodose neurons

We previously reported that allergen challenge induced a pheno-

typic switch to SP production in large-diameter nodose ganglion

neurons (>25 mm diameter) that project nerve fibers specifically to

the trachea.8 To evaluate whether allergic inflammation induces a

similar phenotypic switch in large-diameter neurons innervating

beyond the trachea, we first evaluated all nodose ganglion neurons

with diameters larger than 25 mm for SP immunoreactivity in control

animals and after antigen challenge. Nodose ganglia were removed

and fixed in 4% formaldehyde (fixative) for 2 hours at 4C, rinsed3 times in 0.1 M PBS (pH 7.4) and then cryoprotected with 18%

sucrose in PBS for 18 to 24 hours. Serial cryostat sections of the

nodose ganglia (12 mm thickness) were mounted on 4 consecutive

slides, such that the first slide had sections 1, 5, 9., the second 2, 6,

10, and so forth, and the alternate slides were used for the analysis.

The sections were dried at room temperature for 30 minutes and

rinsed with water and PBS and incubated with blocking solution

containing 1% BSA, 10% goat serum, and 0.5% Tween 20 in PBS at

room temperature for 1 hour. The slides were then processed for

double immunofluorescence staining with a mixture of rabbit poly-

clonal antibody to SP (1 mg/mL; Peninsula Laboratories Inc, San

Carlos, Calif) and mouse mAb to 160 kd neurofilament protein (13

mg/mL; Chemicon, Temecula, Calif) diluted in PBS containing 0.3%

TritonX-100 and 1%BSA (Triton-BSA-PBS; SigmaChemical Co, St

Louis, Mo) for 24 hours at 4C. The sections were washed 3 times

with Triton-BSA-PBS and covered with a mixture of goat antirabbit

fluorescein (20 mg/mL diluted in Triton-BSA-PBS; Vector

Laboratories, Burlingame, Calif) and goat antimouse Texas red (30

mg/mL diluted in Triton-BSA-PBS; Vector Lab) for 2 hours at room

temperature. The sections were then rinsed twice with PBS and once

with higher pH PBS (pH 8.6), then coverslipped with antifade

glycerol (Fluoromount, Molecular Probes, Eugene, Ore). Slides were

examined under epifluorescence (Olympus DX60 microscope;

Olympus Corp, Melville, NY) by using appropriate filter combina-

tions for fluorescein (excitation filter 450-480 nm; barrier filter 500-

515 nm) and Texas red (excitation filter 510-550 nm; barrier filter

570-590 nm). The 160-kd neurofilament protein has been reported to

be a selective marker for myelinated neurons,2,3 most of which are

large-diameter neurons (>25 mm and mean diameter ;40 mm3,8);

only large-diameter, neurofilament-positive, nodose ganglion neu-

rons were evaluated for the expression of SP. Negative controls

included slides treated with only secondary antibodies and slides in

which available commercially obtained isotype controls were used as

the primary antibodies for rabbit and mouse (Vector Lab). These

resulted in categorically negative staining.

Identification of lung-projectinglarge-diameter neurons

Nodose ganglion neurons were retrogradely labeled from the lung

with True Blue (Chemicon) a week before sensitization and allergen

challenge. True Blue is fluorescent dye that is taken up by the nerve


AUGUST 2005

326 Chuaychoo et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

terminals and travels inmembranous vesicles back to the cell soma by

retrograde axoplasmic flow. Accordingly, it is extensively used in

neurobiology to retrogradely backfill neurons. Guinea pigs were

anesthetized with intramuscular injection of ketamine (50mg/kg) and

xylazine (2.5 mg/kg) and received two 1-mL (3%) transdermal

injections of True Blue dye (in PBS containing 1% dimethyl

sulfoxide; Molecular Probes) into the right lung using a 5-mL

Hamilton syringe. One week after the retrograde labeling, the animals

were exposed to aerosolized antigen as described.

Twenty-four hours after antigen challenge, the animals were killed

by intraperitoneal injection of an overdose of sodium pentobarbital

(150 mg/kg) and perfused via the ascending aorta with a rinsing

solution containing procaine (100mg/mL) and heparin (10,000 IU/L)

in PBS followed by fixative. Intrathoracic organs (lung, trachea,

esophagus, heart, and surrounding soft tissues) and nodose ganglia

were removed and put into fixative for 2 hours at 4C, then rinsed

3 times in PBS. Lungs were cut in sections of 1 to 2mm to identify the

dye distribution in parenchyma, blood vessels, and airways. Trachea

and esophaguswere examined inside and outside the lumens. Ganglia

from animals were used only if the dye was delimited primarily to the

lung parenchyma and distal airways, with no dye diffusion into the

large airways or other extrapulmonary sites.

These nodose ganglia from the allergen challenged animals were

prepared and sectioned as described. The alternate slides were

examined, and all True Blue–labeled neurons were photographed

to identify labeled cells in case some labeled neurons faded after

immunostaining. The sections were double stained as described with

the exception of using a rat mAb to SP (5 mg/mL diluted in Triton-

BSA-PBS; Chemicon) with the mouse mAb to 160-kd neurofilament

(13 mg/mL diluted in Triton-BSA-PBS; Chemicon) for 24 hours at

4C as primary antibodies, followed by a mixture of goat antirat

antibody labeled with Alexa Fluor 594 (20 mg/ mL diluted in Triton-

BSA-PBS; Molecular Probes) and goat antimouse antibody labeled

with Alexa Fluor 488 (10 mg/mL diluted in Triton-BSA-PBS;

Molecular Probes) for 2 hours at room temperature as secondary

antibodies. The slides were examined as described, except the True

Blue was visualized with an UV filter (excitation filter 330-385 nm;

barrier filter 400-420 nm). Only True Blue–labeled neurons with

neurofilament-positive immunofluorescence were evaluated to de-

termine whether allergic inflammation induces SP production. We

tested both SP antibodies in the same ganglion sections and found

there was no difference in staining (not shown).

Isolated ganglia studies

The animals were actively sensitized with intraperitoneal injec-

tions of ovalbumin as described. Three weeks after the sensitization,

the animalswere killed byCO2 inhalation and exsanguinated. Nodose

ganglia from both sides were isolated, cut nearly in half to expose the

neurons to the antigen ex vivo, and placed immediately into Krebs

bicarbonate solution (composed of NaCl, 118 mM; KCl, 5.4 mM;

NaH2PO4, 1.0 mM;MgSO4, 1.2 mM;CaCl2, 1.9 mM;NaHCO3, 25.0

mM; dextrose, 11.1 mM; and gassed with 95%O2, 5% CO2, pH 7.4).

Each individual nodose ganglion was incubated (37 C) in 2 mL

Krebs bicarbonate solution containing antibiotics (penicillin/strepto-

mycin, 100 U/mL) and continuously oxygenated (95%O2, 5% CO2).

The solution was changed every 15 minutes 4 times to wash and

equilibrate the tissues. The nodose ganglia from the same animalwere

divided into a control and an antigen challenge group. In the antigen

challenge group, ovalbumin (10 mg/mL) was added to the solution.

The ganglia were incubated for various periods (0, 4, 8, 12, 18 hours)

at 37 C and then removed from the test tubes and fixed and processed

for double immunofluorescence staining for SP and neurofilament as

described in the nonlabeled large-diameter nodose neuron study. We

evaluated all large-diameter, neurofilament-positive nodose ganglion

neurons to determine whether antigen induced the expression of

SP ex vivo.

Vagotomy studies

The animals were anesthetized as described, and 25 mL 5% Fast

Blue dye (Sigma Chemical Co, St Louis, Mo) was instilled into the

trachea as describe previously.14 A week later, the animals were

divided into unsensitized (control) and sensitized groups. In the

sensitized group, the animals were passively sensitized with intra-

peritoneal injection of ovalbumin-sensitized serum as described. A

day after the sensitization, under general anesthesia as described, the

vagus nerve was unilaterally ligated caudal to the nodose ganglion

with suture in 2 places approximately 1 cm apart, and the section of

vagus nerve between sutures was removed. On the following day, the

animals were challenged with ovalbumin inhalation as described.

Unsensitized (control) animals had a similar unilateral vagotomy and

were challengedwith ovalbumin. Twenty-four hours after the antigen

challenge, the animals were killed with CO2 inhalation and

exsanguinated. The nodose ganglia were removed bilaterally and

fixed for double immunofluorescence staining with antibodies to

neurofilament (160 kd) and SP as described in the nonlabeled large-

diameter nodose neuron study. Only Fast Blue–labeled neurons that

were neurofilament-positive were evaluated. For vagotomy, Fast

Blue was used; no differences in labeling were noted between Fast

Blue and True Blue, and the latter was used for the lung labeling

procedure described.


The data were analyzed by using the Student paired or nonpaired

t tests. Datawere expressed asmeans6 SEMs.P< .05 was considered

significant.

RESULTS

Nonretrogradely labeled large-diameternodose neuron study

Wehave previouslymade exhaustive studies of cell diam-eters and neurofilament staining in vagal sensory neuronsprojecting to the respiratory tract.3,9 The cell diametersof neurofilament-negative neurons form a unimodal pop-ulation with a mean diameter of about 20 mm. The celldiameters of neurofilament-positive neurons also form aunimodal distribution with an average diameter of 436 8(n = 713) in one study and 40 6 3 (n = 307) in anotherstudy. In the current study, we evaluated 781 nodoseganglion neurons that were <25 mm in diameter andfound that >99% were neurofilament-negative, whereasamong the 792 neurons with diameters >30 mM, all wereneurofilament-positive. On the basis of these studies, andin accordance with conclusions drawn from more directstudies in the somatosensory system,15 we conclude thatthe large diameter neurofilament-positive population ofneurons project myelinated A fibers to peripheral tissues.

In unsensitized (control) animals, we found that only 78of 2034 large neurons were SP-immunoreactive (averageof 3.8%6 0.6%; n = 4 animals). In ovalbumin-sensitizedanimals, the number of large-diameter neurons positive forSP was substantially higher, 712 of 1939 (average of36.7% 6 3.1%; n = 4 animals; P < .01; compared withunsensitized animals, Fig 1, C; part of these data was



Chuaychoo et al 327

Mech

anismsofasthmaand

allerg

icinflammation

obtained from our previous study8 and is included here forcomparison purposes).

Lung-labeling experiment

To address more specifically the hypothesis that SPsynthesis could be induced in intrapulmonary A-fiberneurons, we used a retrograde labeling strategy. Theretrograde tracer True Blue was injected into right lung.One week after the injection, both actively sensitized andunsensitized (control) groups were challenged with in-haled ovalbumin. The animals were killed 1 day afterantigen challenge, and the distribution of the dye wasexamined in the thoracic tissues. Thirty-two animals wereused for this experiment. In 17 animals, we confirmed thatthe dye was delimited primarily to the lung parenchymaand distal airways, with no dye diffusion into the largeairways or extrapulmonary sites. Fifteen animals wereexcluded from further analysis as a result of diffusion ofthe dye to the large airways and/or cardiac tissue.

Among the successfully labeled animals, 9 animalswere unsensitized (controls) and 8 were sensitized. Allanimals were challenged with ovalbumin, and 1 day later,the neurons were evaluated for SP immunoreactivity.In control animals, an average of 18% 6 4% of thelarge-diameter neurofilament-positive neurons were SP-immunoreactive. In sensitized animals challenged withovalbumin, 30% 6 6% of large-diameter neurofilament-positive neurons were SP-positive (P< .05 compared withcontrol animals; Fig 1, C).

Isolated ganglia studies

We have previously noted that mast cells located withinthe nodose ganglia isolated from ovalbumin-sensitized

animals degranulate and release inflammatory mediatorson antigen challenge.12 We therefore next addressed thequestion whether a local allergic reaction within thenodose ganglion is sufficient to induce SP expression inlarge-diameter neurofilament-positive neurons.

The nodose ganglia were isolated from actively sensi-tized animals and exposed to 10 mg/mL ovalbumin orvehicle (Krebs bicarbonate solution) ex vivo. Ovalbumineffectively induced SP synthesis in neurofilament-positiveneurons beginning at between 8 and 12 hours of exposure.After 18 hours of ovalbumin treatment, approximately16 % of neurofilament-positive neurons were SP-positive(n = 8), compared with;2% in the vehicle treated ganglia(n = 6; P < .01; Fig 2).

Vagotomy studies

The data from the isolated ganglia studies suggest that ifinhaled ovalbumin can reach the vagal sensory ganglia insensitized animals, it may be capable of acting locally toinduce SP synthesis in neurofilament-positive neurons. Toaddress this question further, in vivo experiments wereconducted in passively sensitized (or control) animals withunilateral vagotomy performed before ovalbumin inhala-tion. The induction of SP immunoreactivity in the nodoseganglion neurons was compared between the vagotomyside and contralateral side with intact vagus. If theallergen-inducing signal for SP synthesis travels to thecell body via the nerve fiber, then the induction will benoted only in the ganglia associated with an intact vagusnerve. In these studies, either the right or left vagus nervewas severed in a given animal. The data obtained were thesame whether the right or left ganglia was studied, andtherefore, the results were pooled.

Fast Blue dye (25mL) was instilled into cervical trachea1 week before the antigen challenge. This procedurelabeled both extrapulmonary and intrapulmonary compart-ments. All animals were then challenged with aerosolizedovalbumin. Twenty-four hours after the antigen challenge,the animals were killed, and labeled neurofilament-positiveneurons were evaluated for the SP immunoreactivity.

FIG 1. SP immunoreactivity (SP-IR) in nodose ganglion neurons

1 day after the antigen challenge with ovalbumin (OVA) inhalation

in (A) unsensitized (control) and (B) sensitized guinea pigs. Repre-

sentative photomicrographs of lung neurons labeled with True

Blue (TB) dye with double immunofluorescence staining of SP

and neurofilament (NF; 160-kd) antibodies (scale bar = 50 mm).

C, Histogram showing the percentage of SP immunoreactivity in

large-diameter NF-positive neurons in nodose ganglia after OVA

challenge in unlabeled (n = 4), tracheal-labeled (n = 3), and lung-

labeled (n = 8, unsensitized; n = 9, sensitized). Note, tracheal la-

beled neuron data taken from our previous study.8 Data are

presented as the means 6 SEMs. *P < .01 between sensitized

and unsensitized animals).

FIG 2. Nodose ganglia (ex vivo) were treated ovalbumin (OVA;

10 mg/mL) or saline for the times indicated (n5 1-2 for 0, 4, 8,

12 h and n 5 8 for 18 h). At 18 hours after OVA treatment, the

percentage of SP immunoreactivity (SP-IR) is significantly in-

creased in large-diameter (>25 mm) neurofilament (NF)–positive

neurons (16% 6 2%; n = 8) compared with the control (2% 6 0.8%;

n = 6). Data are presented as means 6 SEMs. *P < .01.


AUGUST 2005

328 Chuaychoo et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

Vagus nerve intact side. In unsensitized animals,3% 6 1% (n = 4) of neurofilament-positive neurons ex-pressed SP in nodose ganglia of an intact vagus nerve. Asexpected, in sensitized animals, the percentage of neuro-filament-positive neurons expressing SP in nodose gangliawith intact vagus averaged 29% 6 3 % (n = 4; Fig 3).

Vagus nerve severed side. In unsensitized animals,9%6 1% of the neurofilament-positive neurons were SP-positive in nodose ganglia on the side in which the vagusnerve was cut. Comparing these results with those ob-served in ganglia obtained from the side with an intactvagus (3%6 1%) suggests that severing the vagus per sesignals the induction of tachykinin synthesis in a subset ofneurofilament-positive nodose neurons. In sensitized andchallenged animals, there was a significant (P < .05) butslight increase in the percentage of the neurofilament-positive neurons that were SP-positive in nodose gangliaon the side of the vagotomy (Fig 3). However, thisapparent ovalbumin effect in ganglia ipsilateral to thevagotomy (1.8-fold increase) was significantly less thanthat observed on the vagus intact side (9.7-fold increase).

DISCUSSION

The results support the hypothesis that allergic inflam-mation in guinea pig lungs can lead to a switch in sensorySP innervation such that, in addition to sensory C fibers,SP is found in cough-causing A fibers in the trachea8 andin stretch-sensitive A fibers in the lungs (current study).In addition, the results are consistent with the hypothesisthat this phenotypic switch in tachykinergic innervationrequires signals that, at least in part, reach the cell bodies inthe distant sensory ganglia via the vagal nerve fibers.

In healthy mammals, most neuronal tachykinins (SPand neurokinin A) in the mammalian respiratory tract arefound in sensory C fibers.1,2 The vagal C-fiber neurons inthe guinea pig respiratory tract are derived from cellbodies situated in the jugular and nodose ganglia.2,3,16 Thejugular C fibers, found in extrapulmonary airways andintrapulmonary bronchi, are more likely to contain SP thanthe nodose C fibers located within the lungs.16 Regardlessof the type of C fiber, these nerves are typically not thoughtto be activated in healthy animals under normal circum-stances. Rather, they are recruited to action by noxiousstimuli or pathological conditions such as tissue inflam-mation.17 On activation of a tachykinin-containing nerve,the tachykinins are released into synapses in the centralnervous system, where they serve to evoke pain in thesomatosensory system4,18 and augment cough19,20 andparasympathetic reflexes in the respiratory system.21-24

When released from the peripheral nerve terminals viaaxon reflexes, the tachykinins may cause or contribute tolocal inflammatory reactions.25-27

Substance P is seldom expressed in large-diameter(A-fiber) sensory neurons innervating the trachea orlungs.2,3,28 The large-diameter neurons innervating theguinea pig respiratory system are located mainly in thenodose ganglia and project touch-sensitive and stretch-

sensitive A fibers to the trachea and lungs, respec-tively.3,10,16 Fischer et al7 were first to publish that allergicinflammation in guinea pig airways is associated with anincrease in preprotachykinin gene transcription and SPsynthesis in nodose ganglion neurons. Subsequently, weshowed that this could occur in large-diameter neurofila-ment-positive neurons that projected low-threshold touch-sensitive Ad fibers to the trachea.8 Further investigationindicated that tracheal nodose Ad fibers in the guinea pigrepresent a unique type of sensory nerve that evokes coughon activation.10

In the current study, we designed experiments todetermine the effects of allergic inflammation in the lungon SP content of neurons that project A fibers to intra-pulmonary structures (ie, RAR/SAR population). Weinjected small volumes of dye directly into the lungparenchyma and subsequently evaluated retrogradelylabeled neurofilament-positive neurons. In the allergen-challenged animals, about 30% of the lung-specificneurofilament-positive neurons expressed SP immunore-activity. This value is similar to that observed in tracheal-labeled neurons. We have previously reported that undercontrol conditions, very few (1% to 3%) of neurofilament-positive neurons in the nodose ganglia express SP immu-noreactivity.8 Likewise, Kummer et al2 have reported thatnodose neurons labeled from healthy guinea pig lungs areessentially all SP-negative. We were therefore surprisedby our finding that 18% of the neurofilament-positiveneurons labeled from the lung were SP-positive in thecontrol animals. We suspect that this may be a result ofsome inflammation caused by our intraparenchymalinjections of the dye itself. In any event, the resultssupport the hypothesis that allergic inflammation can leadto induction of tachykinin synthesis in stretch-sensitive Afibers within the lungs.

In our previous electrophysiological analysis of nodoseA fibers projecting to the guinea pig lungs, we found thatthey represented a rather homogenous population of fast-conducting stretch-sensitive nerves.16 All of the fibers hadconduction velocities in the range of 9 to 25 m/s and couldbe segregated on the basis of their adaptation to prolongedlung inflation into either RAR or SAR phenotypes. Thefinding that 30% of nodose A-fiber neurons express SPafter allergen challenge indicates that SP production can

FIG 3. Histogram demonstrating the percentage of SP immunore-

activity (SP-IR) in Fast Blue–labeled neurofilament (NF)–positive

nodose neurons in intact versus severed vagus nerves 1 day after

the ovalbumin inhalation. Data are presented as means 6 SEMs;

n = 4 per group. *P .05.



Chuaychoo et al 329

Mech

anismsofasthmaand

allerg

icinflammation

be induced in some vagal stretch-sensitive nerves. Thisraises the possibility that SP could be released from theircentral terminals in the brain stem during breathing,independently of noxious stimuli. This would be analo-gous to the phenotypic switch described in touch-sensitiveAb fibers of the somatosensory system, in which suchan effect has been suggested to underlie sensations ofallodynia.4

How allergic inflammation causes increased SP immu-noreactivity in nodose neurons is not known. By using asimilar model system, Fischer et al7 provided directevidence that allergen-induced increase in tachykininsynthesis occurs at the level of preprotachykinin genetranscription in the cell nucleus. We considered 2 path-ways by which inhaled allergen can influence genetranscription in the cell bodies located at a distant vagalsensory ganglion. First, the allergen could reach thesystemic circulation and activate mast cells within theganglion, causing a local effect on the neighboringneurons. Second, the allergic inflammation within theairway wall could influence the nerve terminals in amanner that sends long-distance signals to the cell nucleusvia the afferent nerve fiber.

When the nodose ganglia isolated from sensitizedguinea pig are subsequently exposed to the sensitizingantigen, ganglionic mast cells degranulate, mast cell–associated mediators are released, and the electricalexcitability of resident neurons is altered.12 In the currentstudy, we found that antigen administered to the nodoseganglia ex vivo caused a;10-fold increase in the percent-age of SP-expressing neurofilament-positive neurons.This effect is presumably secondary to the activation ofresident mast cells, although this hypothesis was notfurther addressed. On the basis of these findings, it can besurmised that if inhaled ovalbumin reached the systemiccirculation in sufficient quantities, it could affect SPsynthesis in nodose neurons via a local activation ofganglionic mast cells.

The experiment in which the vagus nerve was unilat-erally severed 1 day before the allergen inhalation moredirectly addresses this issue. Regrettably, severing thenerve itself appeared to increase the SP production inneurofilament-positive neurons to nearly 10% (comparedwith 3% with the vagus nerve intact). This is consistentwith the findings of Noguchi et al,29,30 who reported thatsevering the peripheral fibers in vivo in the somatosensorysystem leads to the induction of neuropeptides productionin dorsal root ganglion neurons. This vagotomy-inducedincrease in SP immunoreactivity in neurons clouds theinterpretation of the data. Nevertheless, allergen inhala-tion caused the expected robust increase (9.7-fold) in thepercentage of neurofilament-positive, SP-positive nodoseneurons only in the side with the vagus nerve intact. On theside in which the vagus was severed, ovalbumin inhalationwas associated with only a marginal increase (1.8-fold) inthe percentage of neurofilament-positive neurons express-ing SP. This is the expected result if the site of action of theallergen is within the lung, but it would not be expected ifthe allergen were acting at the level of the sensory ganglia.

Also supporting the hypothesis that the site of initiation ofthe phenotypic switch is within the lung is the finding thatthe effect of allergic inflammation on SP induction inlarge-diameter neurons can be mimicked with respiratorytract viral infection.9 The virus infection in this model isthought not to extend beyond the airway epithelium. Thesmall ovalbumin-induced effect observed in nodoseneurons ipsilateral to the vagotomy may have been aresult of the fact that the vagus was cut caudal to the pointwhere the superior laryngeal nerve meets the vagus. Thus,the pathway remained intact for that population of A fibersthat project to the larynx and pharynx via the superiorlaryngeal nerve.

The question remains as to the nature of the chemicalmediators capable of being released after allergic inflam-mation that can lead to long-distance nuclear signaling viathe vagal axons. On the basis of the literature, it seemslikely that a neurotrophin molecule such as NGF may becausally involved.31 Neurotrophins and their Trk recep-tors are, in fact, designed to signal gene transcriptionalevents from interactions occurring at distant terminals.11

In addition, NGF is associated with allergic reactions,32

and we have previously reported that microinjection ofNGF into the tracheal wall leads to SP production in large-diameter neurofilament-positive nodose neurons innervat-ing the trachea.33 Similar results have also been observedin the mouse.34

We thank Ms Holly K. Rohde for her technical assistance.

REFERENCES

1. Lundberg JM, Hokfelt T, Martling CR, Saria A, Cuello C. Substance P-

immunoreactive sensory nerves in the lower respiratory tract of various

mammals including man. Cell Tissue Res 1984;235:251-61.

2. Kummer W, Fischer A, Kurkowski R, Heym C. The sensory and

sympathetic innervation of guinea-pig lung and trachea as studied by

retrograde neuronal tracing and double-labelling immunohistochemistry.

Neuroscience 1992;49:715-37.

3. Riccio MM, Kummer W, Biglari B, Myers AC, Undem BJ. Intergan-

glionic segregation of distinct vagal afferent fibre phenotypes in guinea-

pig airways. J Physiol 1996;496:521-30.

4. Neumann S, Doubell TP, Leslie T, Woolf CJ. Inflammatory pain

hypersensitivity mediated by phenotypic switch in myelinated primary

sensory neurons. Nature 1996;384:360-4.

5. Ma QP, Woolf CJ. Progressive tactile hypersensitivity: an inflammation-

induced incremental increase in the excitability of the spinal cord. Pain

1996;67:97-106.

6. Lilly CM, Bai TR, Shore SA, Hall AE, Drazen JM. Neuropeptide content

of lungs from asthmatic and nonasthmatic patients. Am J Respir Crit

Care Med 1995;151:548-53.

7. Fischer A, McGregor GP, Saria A, Philippin B, Kummer W. Induction of

tachykinin gene and peptide expression in guinea pig nodose primary

afferent neurons by allergic airway inflammation. J Clin Invest 1996;98:

2284-91.

8. Myers AC, Kajekar R, Undem BJ. Allergic inflammation-induced

neuropeptide production in rapidly adapting afferent nerves in guinea

pig airways. Am J Physiol Lung Cell Mol Physiol 2002;282:L775-81.

9. Carr MJ, Hunter DD, Jacoby DB, Undem BJ. Expression of tachykinins

in nonnociceptive vagal afferent neurons during respiratory viral infec-

tion in guinea pigs. Am J Respir Crit Care Med 2002;165:1071-5.

10. Canning BJ, Mazzone SB, Meeker SN, Mori N, Reynolds SM, Undem

BJ. Identification of the tracheal and laryngeal afferent neurones medi-

ating cough in anaesthetized guinea-pigs. J Physiol 2004;557:543-58.


AUGUST 2005

330 Chuaychoo et al

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

11. Neet KE, Campenot RB. Receptor binding, internalization, and retrograde

transport of neurotrophic factors. Cell Mol Life Sci 2001;58:1021-35.

12. Undem BJ, Hubbard W, Weinreich D. Immunologically induced neuro-

modulation of guinea pig nodose ganglion neurons. J Auton Nerv Syst

1993;44:35-44.

13. Undem BJ, Buckner CK, Harley P, Graziano FM. Smooth muscle

contraction and release of histamine and slow-reacting substance of

anaphylaxis in pulmonary tissues isolated from guinea pigs passively

sensitized with IgG1 or IgE antibodies. Am Rev Respir Dis 1985;131:

260-6.

14. Hunter DD, Undem BJ. Identification and substance P content of vagal

afferent neurons innervating the epithelium of the guinea pig trachea.


15. Lawson SN, Perry MJ, Prabhakar E, McCarthy PW. Primary sensory

neurones: neurofilament, neuropeptides, and conduction velocity. Brain

Res Bull 1993;30:239-43.

16. Undem BJ, Chuaychoo B, Lee MG, Weinreich D, Myers AC, Kollarik

M. Subtypes of vagal afferent C-fibres in guinea-pig lungs. J Physiol

2004;556:905-17.

17. Coleridge JC, Coleridge HM. Afferent vagal C fibre innervation of the

lungs and airways and its functional significance. Rev Physiol Biochem

Pharmacol 1984;99:1-110.

18. Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain.

Science 2000;288:1765-9.

19. Mazzone SB. Sensory regulation of the cough reflex. Pulm Pharmacol

Ther 2004;17:361-8.

20. Bonham AC, Sekizawa SI, Joad JP. Plasticity of central mechanisms for

cough. Pulm Pharmacol Ther 2004;17:453-7.

21. Carr MJ, Undem BJ. Inflammation-induced plasticity of the afferent inner-

vation of the airways. Environ Health Perspect 2001;109(suppl 4):567-71.

22. Myers AC, Undem BJ. Functional interactions between capsaicin-

sensitive and cholinergic nerves in the guinea pig bronchus. J Pharmacol

Exp Ther 1991;259:104-9.

23. Rumsey WL, Aharony D, Bialecki RA, Abbott BM, Barthlow HG,

Caccese R, et al. Pharmacological characterization of ZD6021: a novel,

orally active antagonist of the tachykinin receptors. J Pharmacol Exp

Ther 2001;298:307-15.

24. Canning BJ, Fischer A. Neural regulation of airway smooth muscle tone.

Respir Physiol 2001;125:113-27.

25. Groneberg DA, Quarcoo D, Frossard N, Fischer A. Neurogenic

mechanisms in bronchial inflammatory diseases. Allergy 2004;59:

1139-52.

26. Barnes PJ. Neurogenic inflammation in the airways. Respir Physiol 2001;

125:145-54.

27. Lundberg JM. Tachykinins, sensory nerves, and asthma: an overview.

Can J Physiol Pharmacol 1995;73:908-14.

28. Springall DR, Cadieux A, Oliveira H, Su H, Royston D, Polak JM.

Retrograde tracing shows that CGRP-immunoreactive nerves of rat

trachea and lung originate from vagal and dorsal root ganglia. J Auton

Nerv Syst 1987;20:155-66.

29. Noguchi K, Dubner R, De Leon M, Senba E, Ruda MA. Axotomy

induces preprotachykinin gene expression in a subpopulation of dorsal

root ganglion neurons. J Neurosci Res 1994;37:596-603.

30. Noguchi K, Kawai Y, Fukuoka T, Senba E, Miki K. Substance P

induced by peripheral nerve injury in primary afferent sensory neurons

and its effect on dorsal column nucleus neurons. J Neurosci 1995;15:

7633-43.

31. Wilfong ER, Dey RD. Nerve growth factor and substance P regulation in

nasal sensory neurons after toluene diisocyanate exposure. Am J Respir

Cell Mol Biol 2004;30:793-800.

32. Bonini S, Lambiase A, Levi-Schaffer F, Aloe L. Nerve growth factor:

an important molecule in allergic inflammation and tissue remodelling.

Int Arch Allergy Immunol 1999;118:159-62.

33. Hunter DD, Myers AC, Undem BJ. Nerve growth factor-induced

phenotypic switch in guinea pig airway sensory neurons. Am J Respir

Crit Care Med 2000;161:1985-90.

34. Dinh QT, Groneberg DA, Peiser C, Springer J, Joachim RA, Arck PC,

et al. Nerve growth factor-induced substance P in capsaicin-insensitive

vagal neurons innervating the lower mouse airway. Clin Exp Allergy

2004;34:1474-9.



Chuaychoo et al 331

Mech

anismsofasthmaand

allerg

icinflammation

Differential effects of (S)- and (R)-enantiomersof albuterol in a mouse asthma model

William R. Henderson, Jr, MD, Ena Ray Banerjee, PhD, and Emil Y. Chi, PhD

Seattle, Wash

Background: (R)- and (S)-Enantiomers of albuterol likely

exert differential effects in patients with asthma. The

(R)-enantiomer binds to the b2-adrenergic receptor with

greater affinity than the (S)-enantiomer and is responsible

for albuterol’s bronchodilating activity. (S)-Albuterol

augments bronchospasm and has proinflammatory

actions.

Objective: The study aim was to determine whether the

(S)-enantiomer, in contrast to the (R)-enantiomer, has

adverse effects on allergic airway inflammation and

hyperresponsiveness in a mouse asthma model.

Methods: Mice sensitized to ovalbumin (OVA) intraperitoneally

on days 0 and 14 were challenged with OVA intranasally on

days 14, 25, and 35. On day 36, 24 hours after the final allergen

challenge, the effect of the (R)- and (S)-enantiomers of albuterol

(1 mg kg21 d21 administered by means of a miniosmotic

pump from days 13-36) on airway inflammation and

hyperreactivity was determined.

Results: In OVA-sensitized/OVA-challenged mice, (R)-albuterol

significantly reduced the influx of eosinophils into the

bronchoalveolar lavage fluid and airway tissue. (R)-Albuterol

also significantly decreased airway goblet cell hyperplasia and

mucus occlusion and levels of IL-4 in bronchoalveolar lavage

fluid and OVA-specific IgE in plasma. Although (S)-albuterol

significantly reduced airway eosinophil infiltration, goblet

cell hyperplasia, and mucus occlusion, it increased airway

edema and responsiveness to methacholine in OVA-sensitized/

OVA-challenged mice. Allergen-induced airway edema

and pulmonary mechanics were unaffected by

(R)-albuterol.

Conclusion: Both (R)- and (S)-enantiomers of albuterol reduce

airway eosinophil trafficking and mucus hypersecretion in a

mouse model of asthma. However, (S)-albuterol increases

allergen-induced airway edema and hyperresponsiveness.

These adverse effects of the (S)-enantiomer on lung function

might limit the clinical efficacy of racemic albuterol. (J Allergy

Clin Immunol 2005;116:332-40.)

Key words: b2-adrenergic agonist enantiomers, airways, mucus,

edema, inflammation, hyperresponsiveness

Adrenergic receptors are composed of a- and b-receptors that bind endogenous catecholamines, such asepinephrine. Although 3 subtypes of b-adrenergic recep-tors exist, smooth muscle relaxation producing vasodila-tion and bronchodilation is mediated by the b2-receptor.Short-acting b2-adrenergic receptor agonists rapidlyinduce bronchodilation in patients with asthma and areused for relief of acute symptoms, prevention of exercise-induced asthma, and management of acute severe asthma.

Racemic albuterol contains equal concentrations(50:50) of the (R)- and (S)-enantiomers (ie, enantiomersthat are nonsuperimposable mirror images).1 The (R)-enantiomer of albuterol binds to b2-adrenergic receptorswith nearly 100-fold greater affinity than the (S)-enanti-omer, suggesting that the (S)-enantiomer does not actthrough b-adrenergic receptor activation.1 Whereas the(R)-enantiomer of albuterol (levalbuterol) exerts thebronchodilating properties of albuterol, the (S)-enantio-mer has adverse effects, including augmentation of bron-chospasm and proinflammatory activities.2-5 In murinemast cells (S)-albuterol increases IgE-induced histamineand IL-4 production, whereas the (R)-enantiomer lacksthese effects.2 Anti-inflammatory effects of (R)-albuterol,such as inhibition of T-cell proliferation, might be negatedby the presence of the (S)-enantiomer.3

Differential effects of the enantiomers might resultfrom differences in pharmacokinetics.1 The initial step inthe metabolism of the (S)- and (R)-enantiomers is sulfateconjugation, a stereospecific process in human airwayepithelial cells and other cells and tissues.6 The greaterrate of sulfate conjugation of (R)-albuterol might leadto lower plasma levels of (R)- than (S)-albuterol inhuman subjects.7 Potential adverse effects of (S)-albuterolon asthma control might also be augmented by increasedbinding to lung tissue.

In this study we characterized the effects of the(R)- and (S)-enantiomers of albuterol on allergic airwayinflammation and hyperresponsiveness in a mouse asthmamodel that mimics key features of human asthma.8

Although prior studies in guinea pigs and human subjectshave demonstrated that the (S)-enantiomer of albuterol caninduce airway hyperreactivity, there are no prior studiesexamining the effect of (S)-albuterol versus (R)-albuterolon both airway hyperresponsiveness and the TH2phenotype (ie, allergen-induced airway eosinophil traf-ficking, mucus metaplasia, edema, and TH2 cytokinerelease) in an in vivo asthma model. We report that both

From the Departments of Medicine and Pathology, University of Washington.

Supported by National Institutes of Health grants AI04989 and HL073722

and by a grant from Sepracor Inc.

Disclosure of potential conflict of interest: E. Chi and E. R. Banerjee—none

disclosed. W. R. Henderson, Jr, receives grants–research support from

Sepracor, Inc.

Received for publication August 4, 2004; revised April 1, 2005; accepted for



Reprint requests: William R. Henderson, Jr, MD, Department of Medicine,

Center for Allergy and Inflammation, Box 358050, University of

Washington, 815 Mercer Street, Seattle, WA 98109. E-mail: joangb@

u.washington.edu.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.013

332

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n



Abbreviations usedAP-1: Activator protein 1

BAL: Bronchoalveolar lavage

OVA: Ovalbumin

Penh: Enhanced pause

(R)- and (S)-enantiomers reduce allergen-induced airwayeosinophil and mucus gland hyperplasia. However, only(S)-albuterol increases airway edema and responsivenessto methacholine, effects that would decrease the thera-peutic efficacy of racemic albuterol.

METHODS

Study protocol

All animal use procedures were approved by the University of

Washington Animal Care Committee. Female BALB/c mice (6-8

weeks of age; The Jackson Laboratory, Bar Harbor, Me) received an

intraperitoneal injection of 100 mg of ovalbumin (OVA; 0.2 mL of

500 mg/mL) complexed with alum on days 0 and 14 (Fig 1). Mice

were anesthetized with 0.2 mL of ketamine (6.5 mg/mL)/xylazine

(0.44 mg/mL) in normal saline administered intraperitoneally before

receiving an intranasal dose of 50mg of OVA (50mL of 1 mg/mL) on

days 14, 25, and 35 (Fig 1). The control group received 0.2 mL of

normal saline with alum administered intraperitoneally on days 0 and

14 and 0.4 mL of saline without alum administered intranasally on

days 14, 25, and 35. In both the saline- and OVA-treated groups,

miniosmotic pumps (200 mL, Alzet Model 2004; Durect Corp,

Cupertino, Calif) containing either (R)- or (S)-albuterol (1 mg kg21 d21, 6 mL/d delivery administration) were inserted subcutaneously

on the back slightly posterior to the scapulae on day 13 and remained

in place until study conclusion on day 36 (Fig 1). Absorption of the

compounds by local capillaries results in systemic administration.

Each study group consisted of 4 to 6 animals. The 1 mg kg21 d21

dose of albuterol enantiomer infusion was selected on the basis

of prior work by Sartori et al,9 demonstrating that continuous release

of racemic albuterol (2 mg kg21 d21) subcutaneously by means of

miniosmotic pump produced steady-state, high-plasma levels of

albuterol (1025 M) in mice.

Pulmonary function testing

In vivo airway responsiveness to methacholine was determined

on day 36 in conscious, freely moving, spontaneously breathing mice

by using whole-body plethysmography (Model PLY 3211; Buxco

Electronics Inc, Sharon, Conn), as described by Hamelmann et al.10

Mice were challenged with aerosolized saline or increasing doses

(2 and 10 mg/mL) of methacholine generated by an ultrasonic

nebulizer (DeVilbiss Health Care, Inc, Somerset, Pa) for 2 minutes.

The degree of bronchoconstriction was expressed as enhanced pause

(Penh), a calculated dimensionless value that correlates with mea-

surement of airway resistance, impedance, and intrapleural pres-

sure.9-11 Penh readings were taken and averaged for 4 minutes

after each nebulization challenge. Penh is calculated as follows:

Penh ¼ ½ðTe=Tr21Þ3ðPEF=PIFÞ, where Te is expiration time, Tr is

relaxation time, PEF is peak expiratory flow, and PIF is peak

inspiratory flow 3 0.67 coefficient. The time for the box pressure

to change from a maximum to a user-defined percentage of the

maximum represents the relaxation time. The Tr measurement begins

at the maximum box pressure and ends at 40%. Because Penh is the

ratio of measurements obtained during the same breath, it is mainly

independent of functional residual capacity, tidal volume, and

respiratory rate.

Light microscopy-morphometry

After pulmonary function testing, bronchoalveolar lavage (BAL)

was performed on the right lung, with total BAL fluid cells counted

and eosinophils identified by means of eosin staining.12 Left lung

tissue was obtained for histopathology, and plasma was obtained

for OVA-specific IgE levels. Ten lung sections per animal were

randomly selected and examined in a blinded manner. Sections were

stained with hematoxylin and eosin, the total inflammatory cell

infiltrate was assessed on a semiquantitative scale (0-41), the number

of eosinophils per unit of airway area (2200 mm2) was determined by

using a point-counting system (Image-Pro Plus point-counting sys-

tem software, Version 1.2 for Windows; Media Cybernetics, Silver

Spring, Md),12 and interstitial and perivascular airway edema were

assessed.13,14 Airway goblet cells (as a percentage of total airway

cells) were identified by means of Alcian blue staining,12 and the

degree of mucus plugging of the airways (0.5 mm to 0.8 mm in

diameter) with the percentage occlusion of the airway diameter was

classified on a 0 to 41 scale on the basis of the following criteria:

0, no mucus; 1, approximately 10% occlusion; 2, approximately 30%

occlusion; 3, approximately 50% occlusion; 4, greater than approx-

imately 80% occlusion.12

Cytokine assays

IL-4, IL-5, IL-10, GM-CSF, TNF-a, IL-2, and IFN-g were

assayed in BAL fluid with Bio-Plex Mouse Cytokine assays (Bio-

Rad Laboratories, Hercules, Calif) that are bead-based multiplex

sandwich immunoassays with a limit of detection of less than 10 pg/

mL. IL-13 was assayed in BAL fluid with a mouse IL-13 immuno-

assay (Quantikine M; R&D Systems, Minneapolis, Minn), with a

limit of detection of less than 1.5 pg/mL.

OVA-specific IgE assay

OVA-specific IgE was assayed by modification of the method

of Iio et al.15 Nunc 96-well flat-bottom plates (Nalge Nunc

International, Rochester, NY) were coated with 50 mg/mL OVA in

13 PBS overnight at room temperature, washed 3 times with

13 PBS plus 0.05% Tween-20 (wash buffer), blocked with 3% BSA

in 13 PBS for 1 hour at room temperature, and washed 4 times with

wash buffer. Fifty-microliter plasma samples (1:1 in 13 PBS) were

added per well and incubated for 90 minutes at 37C, then washed

4 times with wash buffer, and blotted dry by inverting over paper

towels. One hundred microliters (1:100 in 13 PBS) of biotin-

conjugated rat anti-mouse IgE mAb (clone R35-72; BD Biosciences,

San Diego, Calif) was added to each well and incubated overnight at

4C and then washed 4 times with wash buffer and blotted dry. Then

100 mL per well (1:1000 in 13 PBS) streptavidin-horseradish

peroxidase–conjugated secondary antibody (BD Biosciences) was

added, and samples were incubated at 37C for 90 minutes, then

washed 4 times with wash buffer, and blotted dry. One hundred

microliters of substrate solution (ie, 1 tablet of 2,2#-azinobis [3-

ethylbenzthiazoline-sulfonic acid, ABTS; Sigma Chemical Co, St

Louis, Mo] dissolved in 100 mL of 0.05 M phosphate-citrate buffer,

pH 5.0, and 25 mL of 30% H2O2) was added per well, and color was

developed for 30 minutes at room temperature. OD 405 nm was

measured by using an AD 340C absorbance detector (Beckman

Coulter Inc, Fullerton, Calif). For the IgE standard curve, a sandwich

ELISA was used in which in separate assay plates biotin-conjugated

rat anti-mouse IgE mAb (clone R35-72, BD Biosciences) was used

to coat the wells, and instead of plasma samples, known concen-

trations of purified anti-mouse IgE (clone C38-2, BD Biosciences)



Henderson, Banerjee, and Chi 333

Mech

anismsofasthmaand

allerg

icinflammation

were incubated; assays were run as described above. The standard

curve was constructed by using a linear regression analysis of the

absorbances against serial dilutions of known concentrations of

mouse IgE. Pooled mouse plasma from OVA-sensitized/OVA-

challenged mice was used as a positive control.


The data are reported as the means 6 SE of the combined

experiments. Differences were analyzed for significance (P < .05) by

means of ANOVA with the protected least-significant-difference

method (Statview II; Abacus Concepts, Berkeley, Calif).

RESULTS

Effect of (R)- and (S)-enantiomers of albuterolon allergen-induced airway inflammation

Airway infiltration by eosinophils. A marked infiltra-tion of inflammatory cells that were predominantlyeosinophils around the airways and pulmonary bloodvessels was observed in the lung interstitium of OVA-treated mice (Fig 2, B) compared with that seen in saline-treated control mice (Fig 2,A) on day 36, 24 hours after thelast intranasal OVA or saline challenge. By means ofmorphometric analysis (Fig 3, A and B) of the histologicsections (Fig 2, C and D vs B), administration of (R)-and (S)-albuterol by means of miniosmotic pumps(1 mg kg21 d21 dose from days 13-36) significantlydecreased the influx of total inflammatory cells (P =.0035, R-albuterol/OVA vs OVA; P = .0226, S-albuterol/OVA vs OVA; Fig 3, A) and eosinophils (P = .021, R-albuterol/OVA vs OVA; P = .008; S-albuterol/OVA vsOVA; Fig 3, B) into the lung interstitium. Compared withthe saline group (Fig 3, C), OVA-sensitized/OVA-chal-lenged mice exhibited a marked increase in BAL fluideosinophils to 2.5 6 0.5 3 105 eosinophils/mL (P <.0001, OVA vs saline; Fig 3,C), which represented 41.8%of total BAL fluid cells. In OVA-sensitized/OVA-chal-lenged mice, treatment with (R)-albuterol significantlyinhibited the influx of eosinophils into BAL fluid by40.6% (P = .0043, R-albuterol/OVA vs OVA; Fig 3, C).In contrast, (S)-albuterol had no significant effect oneosinophil influx into the BAL fluid of OVA-treated mice(Fig 3, C).Airway mucus hypersecretion. Hyperplasia of airway

goblet cells and hypersecretion of mucus were observed inOVA-treated mice (Fig 2, B) compared with in control

mice (Fig 2, A). Airway goblet cells increased to 38.0% oftotal airway cells in OVA-treated mice compared with0.4% in saline control mice (P = .0001, OVA vs saline;Fig 4, A). The mucus occlusion of the airway diametermorphometric score increased 15-fold in the OVA-treatedmice compared with control mice (P < .0001, OVA vssaline; Fig 4, B). Allergen-induced goblet cell hyperplasiaand mucus occlusion of airway diameter were inhibited byboth the (R)-enantiomer (Fig 2, C) and (S)-enantiomer(Fig 2, D) of albuterol. By means of morphometricanalysis, (R)-albuterol decreased goblet cell hyperplasiaby 48.9% (P = .0066, R-albuterol/OVA vs OVA; Fig 4,A)and airway mucus occlusion by 41.4% (P = .0042,R-albuterol/OVA vs OVA; Fig 4, B). (S)-Albuterolreduced goblet cell hyperplasia by 44.8% (P = .0095,S-albuterol/OVA vs OVA; Fig 4, A) and mucus occlusionof the airways by 35.7% (P = .0088, S-albuterol/OVAvs OVA; Fig 4, B).Airway edema. Airway edema was observed in the

lungs of OVA-treated mice (Fig 2, B) compared withthat seen in saline-treated control mice (Fig 2, A).(S)-Albuterol markedly increased airway edema in OVA-treated mice (Fig 2, D vs B). In contrast, (R)-albuterol hadno effect on allergen-induced edema in the airways ofOVA-treated mice (Fig 2, C).Cytokine release. Significant levels (P < .05) of IL-4,

IL-5, IL-13, and GM-CSF were found in the BAL fluid ofOVA-treated mice compared with levels in the salinecontrol group (Fig 5). The increased levels of IL-4 inOVA-treated mice were reduced 70.5% and 52.2% by(R)-albuterol and (S)-albuterol, respectively; the reductionwas statistically significant only for the (R)-enantiomer(P = .043, R-albuterol/OVA vs OVA; Fig 5). There wasno significant effect of either the (R)- or (S)-enantiomer ofalbuterol on the increased BAL fluid levels of IL-5, IL-13,and GM-CSF in the OVA-treated animals. The levels ofIL-10, TNF-a, IL-2, and IFN-g were not significantlyincreased in the BAL fluid of OVA-treatedmice comparedwith levels seen in saline-treated control mice; neitherenantiomer affected the levels of these cytokines in OVA-treated animals (Fig 5).OVA-specific IgE. Plasma OVA-specific IgE was

absent in saline-treated control mice and present inOVA-treated control mice (Fig 6). (R)-Albuterol and(S)-albuterol reduced OVA-specific IgE levels in OVA-treated mice; the reduction was statistically significant

FIG 1. Study protocol. i.p., Intraperitoneal; i.n., intranasal


AUGUST 2005

334 Henderson, Banerjee, and Chi

Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

only for the (R)-enantiomer (P = .0120, R-albuterol/OVAvs OVA; Fig 6).

Effect of (S)- and (R)-enantiomers ofalbuterol on allergen-induced airwayhyperresponsiveness

Pulmonary mechanics were assessed in response toaerosolized methacholine by means of noninvasive in vivoplethysmography on day 36, 24 hours after the lastintranasal OVA challenge. The OVA-sensitized mice

had been challenged with 3 intranasal doses of OVA, aprotocol that we have previously shown to induce airwayinflammation and mucus hypersecretion but that is sub-optimal for inducing airway hyperresponsiveness.16,17

This protocol was used because an augmenting effectof (S)-albuterol on airway hyperresponsiveness couldhave been masked in our mouse asthma model protocol,in which bronchial hyperresponsiveness is achievedthrough administration of 4 intranasal doses of OVAin mice sensitized by 2 intraperitoneal OVA doses.16,17

FIG 2. Effect of (R)- and (S)-albuterol enantiomers on airway histopathology in a mouse asthma model. Lung

tissue was obtained on day 36 from saline-treated control animals (A), OVA-treated control animals (B), OVA-

treatedmice administered (R)-albuterol (C), and OVA-treatedmice administered (S)-albuterol (D), and sections

were stained with hematoxylin and eosin. Arrows indicate eosinophils and other inflammatory cells,

arrowheads indicate mucus, and asterisks indicate edema. AW, Airway; BV, blood vessel. Bars = 100 mm.




Mech

anismsofasthmaand

allerg

icinflammation

In OVA-treated mice (S)-albuterol significantly increasedbronchial responsiveness to methacholine (Fig 7). Incontrast, (R)-albuterol did not alter airway responsivenessto methacholine in OVA-sensitized/OVA-challengedmice (Fig 7).

Effect of (R)- and (S)-enantiomers of albuterolin non–OVA-sensitized/OVA-challenged mice

Miniosmotic pumps containing either (R)- or (S)-albuterol were placed subcutaneously in saline controlanimals for a 24-day treatment period before pulmonaryfunction testing and assessment of lung histopathology(Fig 8) to examine the effect of the albuterol enantiomersindependently of a modification of the allergic inflamma-tory response.Lung morphology. The (S)-enantiomer of albuterol did

not induce airway edema independently of an allergicresponse. No airway edema or inflammation was seenin saline-treated mice administered either (S)-albuterol(Fig 8, B) or (R)-albuterol (Fig 8, C).Pulmonary mechanics. The (S)-enantiomer of albuterol

did not affect airway reactivity in non–OVA-sensitized/OVA-challenged mice. At 0, 2, and 10 mg/mL methacho-line challenge doses, Penh (percentage of air) valueswere, respectively, 105.1%, 92.3%, and 100.0% of thevalues of saline control animals. Similarly, (R)-albuterol

had no effect on Penh (percentage of air) values ofthe saline control group. At 0, 2, and 10 mg/mL metha-choline doses, non–OVA-sensitized/OVA-challengedmice administered (R)-albuterol had 106.4%, 95.7%,and 101.1% Penh (percentage of air) values of the salinecontrol animals.

DISCUSSION

In this mouse model of asthma, we found both over-lapping and distinct actions of the (S)- and (R)-enan-tiomers of albuterol on key features of allergen-inducedairway inflammation and responsiveness to methacholine.(R)-Albuterol significantly reduced the following featuresof allergen-induced airway inflammation: BAL fluidlevels of IL-4 and eosinophils, airway eosinophil infiltra-tion, goblet cell hyperplasia, and mucus occlusion andcirculating levels of OVA-specific IgE. Although (S)-albuterol also decreased airway tissue eosinophilia, gobletcell hyperplasia, and mucus plugging of the airways, nosignificant effect on the influx of eosinophils into the BALfluid was observed. (S)-Albuterol, but not (R)-albuterol,had adverse effects on airway function by increasingairway edema and hyperresponsiveness in OVA-treatedmice.

FIG 3. Effect of (R)- and (S)-albuterol enantiomers on allergen-induced airway inflammatory cell infiltration.

The total inflammatory cell infiltration of the airways (A), the number of eosinophils per unit area (2200 mm2)

of lung tissue (B), and eosinophils per milliliter of BAL fluid (C) were determined. *P < .05 versus OVA.


AUGUST 2005


Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

We found in thismouse asthmamodel that both (S)- and(R)-albuterol inhibited infiltration of eosinophils intoairway tissue, but only (R)-albuterol significantly reducedeosinophil influx into BAL fluid. Although a correlationbetween BAL fluid levels of IL-5 and eosinophilia istypically seen, there was no reduction in the increased IL-5levels of OVA-treated mice administered either albuterolenantiomer. Short- and long-actingb2-adrenergic agonistsmight facilitate eosinophil apoptosis, thereby reducingairway eosinophilia.18 Albuterol (0.1-10mM) in vitro dosedependently decreases colony numbers and increasesapoptosis of eosinophil progenitor cells from the bloodof patients with asthma.19 In contrast, racemic albuterol,(R)-albuterol, and (S)-albuterol do not affect apoptosis ofantigen-specific human T-cell lines.3

An unexpected finding of this study was the reductionin airway goblet cell hyperplasia and mucus hypersecre-tion by both the (R)- and (S)-enantiomers of albuterol.Limited data exist regarding the effect of (R)- and(S)-enantiomers of albuterol on airway mucus gland

function. Sympathetic (ie, adrenergic), parasympathetic(ie, cholinergic), and sensory-efferent (ie, tachykinin-mediated) pathways regulate mucus secretion from airwayepithelial goblet cells and submucosal glands.20 In humanairways the cholinergic response is predominant and ismediated by muscarinic M3-receptors on the mucus secre-tory cells.20 In ovine tracheal epithelial cells (R)-albuterolincreased ciliary beat frequency, whereas (S)-albuterolhad no significant effect.21 Increased mucociliary clear-ance rates by b2-adrenergic agonists have been reported insome patients with asthma.22 The TH2 cytokines IL-4 andIL-13 have potent effects on mucus secretion.23,24

Administration of each cytokine independently stimulatesairway mucus accumulation in mice.24 MUC5AC geneexpression and BAL fluid mucus protein release areincreased in IL-4 transgenic mice.23 In addition, inhibitionof IL-4 by administration of soluble IL-4 receptor reducesairway mucus hypersecretion and inflammatory cell traf-ficking to the lungs in OVA-treated mice.25 We found thatthe increased BAL fluid levels of IL-4, but not IL-13,

FIG 5. Effect of (R)- and (S)-albuterol enantiomers on BAL fluid cytokine levels in OVA-treated mice. BAL fluid

was assayed for TH1 and TH2 cytokines. *P < .05 versus OVA.

FIG 4. Effect of (R)- and (S)-albuterol enantiomers on allergen-induced airway mucus hypersecretion. The

number of goblet cells (A) andmucus occlusion of airway diameter (B)were determined. *P < .05 versus OVA.




Mech

anismsofasthmaand

allerg

icinflammation

in OVA-treated mice were significantly reduced by(R)-albuterol, with a trend toward IL-4 reduction by(S)-albuterol. Thus the albuterol enantiomers might mod-ulate allergen-induced airway inflammation and mucushypersecretion through IL-4, rather than IL-5 or IL-13,signaling. We have recently demonstrated a similar dis-cordance between IL-4 and IL-5/IL-13 in mediation ofallergen-induced airway eosinophilia and mucus hyper-secretion.14 In a mouse asthma model the selectiveredox effector factor 1 inhibitor PNRI-299, which inhibitsthe transcription factor activator protein-1 (AP-1), sig-nificantly decreased airway eosinophil infiltration andmucus occlusion and lung gene expression of IL-4 but notIL-5 or IL-13.14 During T-cell activation, a complexinteraction between nuclear factor of activated T cells andAP-1 is necessary for inducible expression of IL-4.26 Incontrast, transcriptional regulation of IL-5 and IL-13might be independent of AP-1 binding.27 The inductionof goblet cell hyperplasia and eosinophilia by IL-4 in tripleIL-5/IL-9/IL-13–knockout mice further demonstrates thekey role IL-4 exerts in the development of the TH2phenotype.28 In our studies the reduction in IL-4 inOVA-treated mice administered (R)-albuterol correlatedwith a decrease in circulating OVA-specific IgE.

We found that (S)-albuterol, but not (R)-albuterol,augmented the interstitial edema observed in OVA-treated

mice, suggesting a proinflammatory effect unique to(S)-albuterol. This effect of (S)-albuterol was not observedin nonsensitized/nonchallenged mice. In 2 sheep modelsof altered lung fluid balance, the effect of aerosolizedracemic albuterol and its (R)- and (S)-enantiomers on lungepithelial permeability has been examined.29 Pretreatmentwith (S)-albuterol increased the level of albumin in theepithelial lining fluid in sheep receiving histamine toincrease lung permeability. This effect was not observedafter pretreatment with either (R)-albuterol or its racemate.In sheep receiving an increase in left atrial pressure toincrease hydrostatic forces, only (S)-albuterol increasedlung water volume. Thus in the presence of alteredlung fluid balance, (S)-albuterol, but not (R)-albuterol,might increase lung epithelial permeability. Our findingthat (S)-albuterol increases allergen-induced interstitialedema indicates a potential adverse effect of this enanti-omer in patients with asthma.

(S)-Albuterol significantly increased bronchial re-sponsiveness to methacholine challenge in OVA-sensi-tized/OVA-challenged mice, an adverse effect on airwayfunction not shared by (R)-albuterol. In nonsensitized/nonchallenged mice (S)-albuterol did not independentlyaffect the airway response to methacholine. We usedwhole-body plethysmography to assess airway hyperre-activity to methacholine in the study groups. Although

FIG 6. (R)-Albuterol decreases OVA-specific IgE levels in OVA-treated mice. Plasma OVA-specific IgE levels

were determined. *P < .05 versus OVA.

FIG 7. (S)-Albuterol increases allergen-induced airway hyperresponsiveness. The degree of bronchoconstric-

tion to aerosolized methacholine (0, 2, and 10 mg/mL) was expressed as Penh (percentage of air as control).

*P < .05 versus OVA.


AUGUST 2005


Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

there has been recent controversy regarding the use ofPenh as an indirect measure of pulmonary mechanics,30-32

Penh values correlate well with airway resistance mea-sured directly in anesthetized, tracheotomized, andmechanically ventilated mice.10,11 A strong correlationalso exists between Penh values and the intensity of theallergen-induced airway eosinophil infiltration in themouse asthma model.33

(S)-Albuterol might mediate its augmenting effect onbronchial hyperresponsiveness through several modes ofaction, including parasympathetic and sympathetic path-ways. In OVA-sensitized guinea pigs continuous expo-sure to (R,S)- and (S)-albuterol, but not (R)-albuterol,

for a 10-day period increased bronchial hyperresponsive-ness to both histamine and OVA.5 Chronic capsaicintreatment prevented the (R,S)- and (S)-albuterol–inducedbronchial hyperresponsiveness in this model to indicatethe importance of capsaicin-sensitive sensory nerves in(S)-albuterol-mediated development of airway hyperre-sponsiveness.5 Recent studies by Agrawal et al4 indicatethat (S)-albuterol activates proconstrictor and proinflam-matory pathways in human bronchial smooth musclecells. (S)-Albuterol significantly increased the expressionand activity of Gia-1 protein and reduced Gs protein inthese cells.4 (S)-Albuterol also increased the intracellularfree calcium concentration in the bronchial smooth

FIG 8. Effect of (R)- and (S)-albuterol enantiomers on airway histology in non–OVA-sensitized/non–OVA-

challenged mice. Lung tissue was obtained on day 36 from saline control animals (A), saline-treated mice

administered (S)-albuterol (B), and saline-treated mice administered (R)-albuterol (C), and sections were

stained with hematoxylin and eosin. AW, Airway. Bars = 100 mm.




Mech

anismsofasthmaand

allerg

icinflammation

muscle cells after methacholine stimulation.4 Theseproconstrictor effects of (S)-albuterol were accompaniedby stimulation of phosphatidylinositol 3#-OH-kinaseand nuclear factor kB proinflammatory pathways in thesmooth muscle cells. (R)-Albuterol induced the oppositeeffects on Gia-1, Gs, and intracellular free calcium con-centration in the bronchial smooth muscle cells, whichindicates separate mechanisms of action of the enan-tiomers.4

In summary, the actions of the (R)- and (S)-enantiomersof albuterol in the lungs of allergen-sensitized/allergen-challenged mice are complex. Although both enantiomersreduce mucus hypersecretion and trafficking of eosino-phils to the lungs after allergen challenge, only the(S)-enantiomer induces airway edema and hyperrespon-siveness to methacholine. Additional studies are neededto delineate the specific effects of the (R)- and (S)-enantiomers of albuterol on airway inflammation andhyperresponsiveness in patients with asthma.

We thank Gertrude Chiang, Falaah Jones, and Ying-tzang Tien

for excellent technical assistance and Rachel Norris for typing this

manuscript.

REFERENCES

1. Slattery D, Wong SW, Colin AA. Levalbuterol hydrochloride. Pediatr

Pulmonol 2002;33:151-7.

2. Cho SH, Hartleroad JY, Oh CK. (S)-Albuterol increases the production

of histamine and IL-4 in mast cells. Int Arch Allergy Immunol 2001;124:

478-84.

3. Baramki D, Koester J, Anderson AJ, Borish L. Modulation of T-cell

function by (R)- and (S)-isomers of albuterol: anti-inflammatory

influences of (R)-isomers are negated in the presence of the (S)-isomer.


4. Agrawal DK, Ariyarathna K, Kelbe PW. (S)-Albuterol activates pro-

constrictory and pro-inflammatory pathways in human bronchial smooth

muscle cells. J Allergy Clin Immunol 2004;113:503-10.

5. Keir S, Page C, Spina D. Bronchial hyperresponsiveness induced by

chronic treatment with albuterol: role of sensory nerves. J Allergy Clin

Immunol 2002;110:388-94.

6. Eaton EA, Walle UK, Wilson HM, Aberg G, Walle T. Stereoselective

sulphate conjugation of salbutamol by human lung and bronchial

epithelial cells. Br J Clin Pharmacol 1996;41:201-6.

7. Boulton DW, Fawcett JP. Enantioselective disposition of salbutamol in

man following oral and intravenous administration. Br J Clin Pharmacol

1996;41:35-40.

8. Henderson WR Jr, Lodewick MJ. Animal models of asthma. In:

Adkinson NF Jr, Yuninger JW, Busse WW, Bochner BS, Holgate ST,

Simons FER, editors. Middleton’s allergy: principles and practice.

6th ed. St Louis: Mosby; 2003. p. 465-81.

9. Sartori C, Fang X, McGraw DW, Koch P, Snider ME, Folkesson HG,

et al. Selected contribution: long-term effects of b2-adrenergic receptor

stimulation on alveolar fluid clearance in mice. J Appl Physiol 2002;93:

1875-80.

10. Hamelmann E, Schwarze J, Takeda K, Oshiba A, Larsen GL, Irvin CG,

et al. Noninvasive measurement of airway responsiveness in allergic

mice using barometric plethysmography. Am J Respir Crit Care Med

1997;156:766-75.

11. Justice JP, Shibata Y, Sur S, Mustafa J, Fan M, Van Scott MR. IL-10

gene knockout attenuates allergen-induced airway hyperresponsiveness

in C57BL/6 mice. Am J Physiol Lung Cell Mol Physiol 2001;280:

L363-8.

12. Henderson WR Jr, Tang L-O, Chu S-J, Tsao S-M, Chiang GKS, Jones F,

et al. A role for cysteinyl leukotrienes in airway remodeling in a mouse

asthma model. Am J Respir Crit Care Med 2002;165:108-16.

13. Oh SW, Chong IP, Dong Keun L, Jones F, Chiang GKS, Kim HO, et al.

Tryptase inhibition blocks airway inflammation in a mouse asthma

model. J Immunol 2002;168:1992-2000.

14. Nguyen C, Teo J-L, Matsuda A, Eguchi M, Chi E, Henderson WR Jr,

et al. Chemogenomic identification of Ref-1/AP-1 as a novel therapeutic

target for asthma. Proc Natl Acad Sci U S A 2003;100:1169-73.

15. Iio J, Katamura K, Takeda H, Ohmura K, Yasumi T, Meguro TA, et al.

Lipid A analogue, ONO-4007, inhibits IgE response and antigen-induced

eosinophilic recruitment into airways in BALB/c mice. Int Arch Allergy

Immunol 2002;127:217-25.

16. Henderson WR Jr, Lewis DB, Albert RK, Zhang Y, Lamm WJE, Chiang

GKS, et al. The importance of leukotrienes in airway inflammation in a

mouse model of asthma. J Exp Med 1996;184:1483-94.

17. Zhang Y, Lamm WJE, Albert RK, Chi EY, Henderson WR Jr, Lewis

DB. Influence of the route of allergen administration and genetic

background on the murine allergic pulmonary response. Am J Respir

Crit Care Med 1997;155:661-9.

18. Johnson M. Effects of b2-agonists on resident and infiltrating inflam-

matory cells. J Allergy Clin Immunol 2002;110(suppl):S282-90.

19. Wang CH, Lin HC, Lin CH, Yu CT, Liu SL, Huang KH, et al. Effect of

theophylline and specific phosphodiesterase IV inhibition on prolifera-

tion and apoptosis of progenitor cells in bronchial asthma. Br J

Pharmacol 2003;138:1147-55.

20. Rogers DF. Pharmacological regulation of the neuronal control of airway

mucus secretion. Curr Opin Pharmacol 2002;2:249-55.

21. Frohock JI, Wijkstrom-Frei C, Salathe M. Effects of albuterol enan-

tiomers on ciliary beat frequency in ovine tracheal epithelial cells. J Appl

Physiol 2002;92:2396-402.

22. Bennett WD. Effect of b-adrenergic agonists on mucociliary clearance.

J Allergy Clin Immunol 2002;110(suppl):S291-7.

23. Temann U-A, Prasad B, Gallup MW, Basbaum C, Ho SB, Flavell RA,

et al. A novel role for murine IL-4 in vivo: induction of MUC5AC gene

expression and mucin hypersecretion. Am J Respir Cell Mol Biol 1997;

16:471-8.

24. Grunig G, Warnock M, Wakil AE, Venkayya R, Brombacher F,

Rennick DM, et al. Requirement for IL-13 independently of IL-4 in

experimental asthma. Science 1998;282:2261-3.

25. Henderson WR Jr, Chi EY, Maliszewski CR. Soluble IL-4 receptor

inhibits airway inflammation following allergen challenge in a mouse

model of asthma. J Immunol 2000;164:1086-95.

26. Burke TF, Casolaro V, Georas SN. Characterization of P5, a novel

NFAT/AP-1 site in the human IL-4 promoter. Biochem Biophys Res

Commun 2000;270:1016-23.

27. Macian F, Garcia-Rodriguez C, Rao A. Gene expression elicited by

NFAT in the presence or absence of cooperative recruitment of Fos and

Jun. EMBO J 2000;19:4783-5.

28. Fallon PG, Jolin HE, Smith P, Emson CL, Townsend MJ, Fallon R, et al.

IL-4 induces characteristic Th2 responses even in the combined absence

of IL-5, IL-9, and IL-13. Immunity 2002;17:7-17.

29. Peterson BT, Miller EJ. Effects of enantiomers of albuterol on lung

epithelial permeability [abstract]. Am J Respir Crit Care Med 2000;161:

A416.

30. Petak F, Habre W, Donati YR, Hantos Z, Barazzone-Argiroffo C.

Hyperoxia-induced changes in mouse lung mechanics: forced oscilla-

tions vs. barometric plethysmography. J Appl Physiol 2001;90:2221-30.

31. Lundblad LK, Irvin CG, Adler A, Bates JH. A reevaluation of the

validity of unrestrained plethysmography in mice. J Appl Physiol 2002;

93:1198-207.

32. Mitzner W, Tankersley C. Interpreting Penh in mice. J Appl Physiol

2003;94:828-31.

33. Dohi M, Tsukamoto S, Nagahori T, Shinagawa K, Saitoh K, Tanaka Y,

et al. Noninvasive system for evaluating the allergen-specific airway

response in a murine model of asthma. Lab Invest 1999;79:1559-71.


AUGUST 2005


Mech

anism

sofasth

maand

alle

rgic

inflammatio

n

Rhinitis, sinusitis, and ocular diseases

Comparison of test devices for skin pricktesting

Warner W. Carr, MD,a Bryan Martin, DO,a Robin S. Howard, MA,b Linda Cox, MD,c

Larry Borish, MD,d and the Immunotherapy Committee of the American Academy

of Allergy, Asthma and Immunology Silver Spring, Md, Fort Lauderdale, Fla,

and Charlottesville, Va

Background: Allergy skin testing guides developing avoidance

plans and writing an immunotherapy prescription. The goal

for the allergist is to apply allergen skin testing to the

appropriate patient population by using a device that

minimizes both false-negative and false-positive findings while

minimizing patient discomfort. New skin testing devices

continue to be developed with a trend toward production of

multiheaded devices. Data on the performance of these devices

in a head-to-head prospective fashion are limited.

Objective: Our goal was to study 8 commonly used devices to

compare their performance in a head-to-head fashion.

Methods: In a prospective, double-blind fashion, the

performance of 8 skin test devices was evaluated. Devices

were tested with histamine and saline on both the arms and

back of each subject. Devices were rotated over 4 testing

sessions, at least a week apart, so each device was tested in

each anatomic testing location. Performance elements

examined included wheal, flare, pain, sensitivity, specificity,

and intradevice variability.

Results: We found significant differences in all areas of

device performance among all devices examined. Multiheaded

devices also demonstrated significant intradevice variability

and were more painful than single devices. Furthermore,

multiheaded devices had larger reactions on the back, whereas

single devices had larger reactions on the arms.

Conclusion: Statistically significant differences exist among

all devices tested. Providers should consider this data when

choosing a device that suits their practice setting and ensure

that technicians are sufficiently trained on the correct use of

that device. (J Allergy Clin Immunol 2005;116:341-6.)

Key words: Skin testing, device, performance, variability, pain

The US Joint Council of Allergy, Asthma and Immu-nology1 and the European Academy of Allergology andClinical Immunology2 recommend percutaneous testingas the primary test for diagnosis of IgE mediated allergicdisease. Skin testing is also the preferred method forselecting allergens to be included in immunotherapy.3

Given this, the findings on the initial skin test panel arevery important clinical data. If a particular device is toosensitive (resulting in false-positive findings), the patientmay receive an antigen that is not required to achieveclinical benefit. On the other hand, a high false-negativerate for a particular device will result in a patient notreceiving a needed antigen while undergoing immuno-therapy. The goal for the allergist is to perform allergenskin testing in an appropriate patient population by usinga device that minimizes both false-negative and false-positive findings. In addition, it is desirable to use a devicethat results inminimal patient discomfort. Previous studiescomparing devices for skin prick (ie, prick and puncture)testing have revealed significant differences in the sizeof wheal and flare reactions. These differences have beenseen at both positive (allergen extract or histamine) andnegative (saline) sites.4-7 In these studies, the differenceappeared to result from the degree of trauma impartedto the skin by the device, an interpretation that wasreinforced by the fact that those producing larger whealsalso caused more patient discomfort.6

New skin devices continue to be developed, with acurrent trend toward devices that allow for applicationof several antigens simultaneously, referred to as multi-headed. This may limit technician time and increaseefficiency. In addition, multiheaded devices have increas-ing popularity in children, in whom the acceptance of afew multiple test devices tends to be better than manyindividually applied devices. In a recent letter to the editor,Nelson et al7 compared 3 new multidevices with previ-ously reviewed devices. In this communication, signifi-cant differences were noted with the smallpox needle onthe back and the Greer Track (Greer Labs, Lenoir, NC) onthe arm. Given this, we reviewed 4 devices that allow

From athe Department of Allergy and Immunology and bthe Department of

Clinical Investigations, Walter Reed Army Medical Center, Silver Spring;cprivate practice, Fort Lauderdale; and dthe Asthma and Allergic Disease

Center, University of Virginia Health System.

Supported by the Walter Reed Army Medical Center, Department of Clinical

Investigations.

Disclosure of potential conflict of interest: L. Borish has consultant arrange-

ments with PDL, Syngenta, and Sepracor. No other relevant conflicts of

interest to disclose.

The opinions or assertions contained herein are the private views of the authors

and are not to be construed as official or as reflecting the views of the

Department of the Army or the Department of Defense.

Received for publication September 3, 2004; revisedMarch 28, 2005; accepted



Reprint requests: Warner W. Carr, MD, Division of Experimental

Therapeutics, Walter Reed Army Institute of Research, 503 Robert Grant

Ave, Silver Spring, MD 20910. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.035

Rhinitis,sinusitis,

and

ocu

lardisease

s

341


Abbreviation usedCV: Coefficient of variation

application of multiple antigens at once (multiheaded) and4 devices that allow application of only 1 antigen at a time(single devices).

Our goal was to study a cohort of devices to comparetheir performance in a head-to-head fashion. We specif-ically intended to determine sensitivity, specificity, var-iability, and pain. With these results, providers will beable to determine which device is best suited for theirpractices.

METHODS

Study design

The study was a prospective, double-blind clinical trial and was

reviewed and approved by the Walter Reed Army Medical Center

Clinical Investigation Committee and theHumanUseCommittee. All

subjects enrolled into the study voluntarily agreed to participate and

gave written informed consent. Each subject underwent testing in 4

sessions, with each at least 1 week apart. Each device was tested both

on the arm and the back, with histamine (10 mg/mL; Hollister-Stier,

Spokane, Wash) and glycerol-saline (Hollister-Stier) during each

session. During the course of 4 sessions, the locations on the arm and

back were rotated to ensure each device was tested on the upper and

lower arm and upper and lower back. Therefore, each session yielded

4 test sites per device: 2 histamine tests (1 back and 1 arm) and 2

glycerol-saline tests (1 back and 1 arm). Fig 1 illustrates the back and

left arm test regions for one test session. Over the course of the study,

sites were rotated to give an equal number of tests in all areas to offset

any differences in reactivity.6 At the end of the study, each device had

been tested on the upper and lower back and the upper and lower

arms, with an even distribution between left and right. All heads of a

multiheaded device were tested with histamine at the histamine

site, and all heads were tested with saline at the saline site. At the

conclusion of the fourth session, a mean result was determined for

each head of the multiheaded devices, and from this, intradevice

variability was determined. Single device test sites were spaced at

least 30 mm apart, and multiheaded spacing was fixed at 20 mm to

30 mm on the basis of the design of the device. With the devices

examined, this resulted in 132 individual pricks per subject per

session. The total number of individual pricks over the course of

the study for each subject was 528. With 13 subjects completing

the study, this yielded 6864 individual prick sites for examination.

Before each session, antihistamines were withheld for at least 1 week,

and H2 antagonists and leukotriene antagonists were withheld for

72 hours.

To maintain objectivity, the technician who performed all of the

tests was blind to the contents of the test solution, either histamine

or saline. A second technician who was not in the room during

application of each device recorded the results. This technician was

blind to the device used as well as to the solution used. Before the

study was initiated, a representative of the manufacturer trained the

technicianwho performed the skin tests on each device. This step was

taken to achieve the best possible results by using the manufacturer’s

recommended skin testing technique.

Pain assessment was performed by using theWong-Baker FACES

pain rating scale8 immediately after application of each skin test

device (measured on a scale from 0-10). On the basis of this scale, a

level of 1 to 2 is considered minimal pain. The greatest reported pain

was recorded for that particular test site and session. Pain was

recorded within seconds of application of the skin test device to

minimize the influence of histamine on pain perception. Results were

recorded for pain sensation on the arm and on the back.

Subjects

Male or female subjects age 18 to 70 years, with or without

allergies, were eligible for the study. Subjects were excluded if they

had dermatographism, severe atopic dermatitis, or asthma, or were

taking antidepressants. Antihistamines were withheld for 1 week

before testing. Leukotriene antagonists and H2 antagonists were

withheld for 72 hours before testing.

Devices

Four single-headed devices and 4 multiheaded devices were

tested. Single headed devices included the Greer Pick (Greer Labs),

Accuset (ALK-Abello, Inc, Round Rock, Tex), Sharptest (Panatrex,

Inc, Placentia, Calif), and Quintip (Hollister-Stier). Multiheaded

devices tested were the Quintest (Hollister-Stier), Quantitest

(Panatrex, Inc), Greer Track, and Multi-Test II (Lincoln Diag-

nostics, Inc, Decatur, Ill; Fig 2).

Skin testing

All testing was performed first on the arms, and once results were

obtained and recorded, testing proceeded on the back. The wheal and

flare results were recorded at 15 minutes by obtaining the longest

orthogonal diameters. Mean diameters were used for statistical

analyses. Pain was recorded immediately after application of each

skin test device. Positive test solution consisted of 10 mg/mL

histamine (Hollister-Stier), with standard glycerol saline (Hollister-

Stier) used as a negative solution.


Results were analyzed by using repeated-measures ANOVA, with

the within-subject factors body site (upper arm, lower arm, upper

back, lower back) and device. Thirteen subjects were needed to power

the study adequately to detect a minimum difference of 2 mm

between each device. When calculating sensitivity and specificity, a

true-positive result was considered a histamine wheal of 3 mm or

greater, and a true-negative result was a glycerol-saline wheal less

than 3 mm. A result was considered false-negative if a histamine

FIG 1. The left image illustrates the 4 test zones of the back, and the

image on the right represents the 2 left arm test zones. Right arm

test zones are not shown but mirror those of the left arm. LLA, Left

lower arm; LLB, left lower back; LUA, left upper arm; LUB, left

upper back; RLB, right lower back; RUB, right upper back.


AUGUST 2005

342 Carr et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

wheal was less than 3 mm, and a result was considered false-positive

if the glycerol-saline site was 3 mm or greater. Results are presented

as the means6 SDs, and for multiheaded devices, the average of all

heads was used in the calculation of sensitivity and specificity.

Sensitivity and specificity of each device are presented as proportions

with 95% CIs, and devices were compared by using the Fisher exact

test (2-tailed). Sensitivity was calculated by dividing true-positive

results by the sum of true-positive and false-negative results.

Specificity was calculated by dividing true-negative results by the

sum of true-negative and false-positive results.

When multiheaded devices were analyzed, intradevice variability

was described by using the coefficients of variation (CVs; presented

as medians with the interquartile range) for each device. For each

multiheaded device, the wheal produced by each head was compared

by using repeated-measures ANOVA.

Pain scores were compared among devices by using theWilcoxon

signed-rank test: median pain scores were presented as well as the

proportion of pain scores above a value of 2 (representing mild pain

on the Wong-Baker FACES pain rating scale). For interdevice

comparisons of pain, wheal, and flare size within the single-headed

or multiheaded groups, there are 6 possible pairwise analyses. Using

a Bonferroni correction of the overall experimental P value of .05,

a P value of .008 (.05/6) or less is considered significant.

RESULTS

Twenty subjects were recruited for the study, and7 subjects did not complete because of pregnancy (1)and military operational requirements (6). Eight menand 5 women completed the study. The mean age was

32.2 years (range, 22-57), and 7 subjects had a history ofatopy.

Interdevice comparisons

Histamine and saline reactions are presented in Table I.Controlling for site (arm vs back), there was a significantdifference in histamine wheal size among all devices ineach device group (P < .008 for all comparisons), exceptfor no significant difference between the Accuset and theQuintip (P = .28) and the Multi-Test II and Quantitest(P = .27). The largest reactions to histamine base werefound with 2 single devices, Sharptest and Greer Pick.There were no significant differences in saline whealreactions. In addition, all mean histamine flares weregreater than 10 mm, and mean saline flares were below5 mm. Table II gives the number of results that exceededthe limits for positive and negative reactions set forhistamine and saline. For histamine wheal reactions, theGreer Pick gave the lowest number of false-negativeresults (2/208 or 0.96%); the range for single devices was0.96% to 3.8% (Accuset). The range for multiheadeddevices was 57/1664 (3.4%) with the Multi-Test II and366/1664 (22%) with Greer Track.

Single devices demonstrated a high degree of repro-ducibility, with CVs ranging from 0.22 to 0.37 (Table I).The CV reported in Table I for the multiheaded devicesrepresents a CV of the mean of all heads. For intradevicevariability, or differences between each head of a multi-headed device, see Table III.

FIG 2. Skin test devices investigated. Multiheaded devices from top left to right followed by midleft to right:

Quintest, Greer Track, Multi-Test II, and Quantitest. Single devices from bottom left to right: Accuset, Quintip,

Sharptest, Greer Pick.



Carr et al 343

Rhinitis,sinusitis,

and

ocu

lardisease

s

Sensitivity and specificity

The results of device sensitivity and specificity arelisted in Table I. All single devices and the Multi-Test IIhad sensitivities >90%, and there was no significantdifference in sensitivity among the single devices andthe Multi-Test II. The Multi-Test II was more sensitivecompared with all other multiheaded devices (P < .002).The Quintest was less sensitive than the Greer Pick,Sharptest, and Multi-Test II. The Greer Track was lesssensitive than all other devices (P < .0005).

Arm versus back comparisons

There was a significant difference in histamine whealsizes between the arms and backs for all devices (P <.0005; Fig 3). Histamine wheals for all single devices weresignificantly larger on the arms (P < .05 for all compar-isons), and wheals for all multiheaded devices (except theQuintest) were larger on the back (P < .0013). TheQuintest device was larger on the back, but this differencedid not reach statistical significance (P = .17). There wasno significant difference between upper and lower arm.There also was no significant difference between upperand lower back.

Multidevices: intradevice variabilityIntradevice variability reactions are presented in Table

III. Analyzing the multiheaded devices for intradevicevariability, there were significant differences in the whealsizes between the various heads for each device (P < .009for all devices). Fig 4 illustrates the intradevice variabilityfor the Greer Track. With the 8-headed devices (GreerTrack, Multi-Test II, and Quantitest), the greatest degreeof variability was found comparing the interior heads (S2,S3, S6, S7) with the corner heads (S1, S4, S5, S8) for all ofthem.

TABLE I. Outcome measures for 8 devices*

Histamine wheal,

mean 6 SD

Histamine flare,

mean 6 SD CV

Saline wheal,

mean 6 SD

Saline flare,

mean 6 SD

Sensitivity %

(95% CI)

Specificity %

(95% CI)

Single devices

Sharptest 7.1 6 1.7 31.6 6 8.4 0.22 0.03 6 0.3 3.2 6 2.8 97 (91-991) 99 (94-991)

Greer Pick 6.6 6 1.8 33.3 6 9.5 0.37 0.0 6 0.0 2.7 6 2.4 98 (93-991) 100 (97-100)

Accuset 5.1 6 1.9 24.3 6 10.7 0.34 0.1 6 0.5 1.5 6 2.4 92 (85-97) 98 (93-991)

Quintip 4.8 6 1.7 22.6 6 9.3 0.36 0.0 6 0.0 1.1 6 2.6 95 (89-99) 100 (97-100)

Multidevices

Multi-Test II 5.9 6 1.3 26.0 6 5.7 0.23 0.02 6 0.2 3.3 6 1.5 93 (91-95) 99 (98-991)

Quantitest 5.7 6 1.6 25.6 6 7.3 0.34 0.01 6 1.6 3.2 6 1.8 89 (86-91) 99 (98-991)

Quintest 4.3 6 1.4 19.9 6 7.8 0.25 0.0 6 0.0 0.8 6 1.5 86 (82-89) 100 (99-100)

Greer Track 3.2 6 1.3 16.5 6 6.4 0.42 0.012 6 0.1 3.4 6 1.4 56 (52-60) 99 (98-991)

*Values for wheal and flare expressed in millimeters.

TABLE II. Number of tests that exceed 3 mm for saline wheal and 10 mm for saline flare, and number of

tests that are below 3 mm for histamine wheal and 10 mm for histamine flare

Histamine wheal, mm Histamine flare, mm Saline wheal, mm Saline flare, mm

Total test <3 Range <10 Range >3 Range >10 Range

Single devices

Sharptest 208 3 0-10 1 9-60 0 0-3 1 0-15

Greer Pick 208 2 0-12 2 7-75 0 0 0 0-10

Accuset 208 8 0-8 11 0-50 1 0-4 1 0-12

Quintip 208 5 0-10 9 0-40 0 0 2 0-15

Multidevices

Multi-Test II 1664 57 0-12 42 0-50 4 0-4 4 0-20

Quantitest 1664 94 0-11 85 0-62 1 0-5 3 0-32

Quintest 1040 73 0-11 93 0-50 0 0 1 0-15

Greer Track 1664 366 0-11 361 0-47 2 0-5 1 0-22

TABLE III. Intradevice variability for multiheaded

devices expressed as CV

Multidevices

Histamine wheal,*

mean 6 SD

CVy (interquartile

range)

Multi-Test II 5.9 6 1.3 0.20 (0.14-0.44)

Quantitest 5.7 6 1.6 0.23 (0.14-0.50)

Quintest 4.3 6 1.4 0.25 (0.18-0.59)

GreerTrack 3.2 6 1.3 0.93 (0.70-1.39)

*Values expressed in millimeters.

CV median (interquartile range), intradevice variability.


AUGUST 2005

344 Carr et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

The greatest degree of intradevice variability was foundwithin the Greer Track.

Pain

Median pain scores for all of the devices was 1.0, exceptfor the Greer Track, with a median of 2.0 (Table IV).Reports of pain were considered minor, with only 1 painrating reported above 6 on a scale from 0 to 10 on theWong-Baker FACES pain rating scale.8 The highest painrating was for the Greer Track (34% of pain scores above2), and the minimum pain reported was for the Greer Pick(5% of pain scores above 2). All single devices weresignificantly less painful than the multiheaded devices(P < .0005). Comparing the single devices, Sharptest painscores were significantly higher than Greer Pick (P <.0005) and Accuset (P = .001). For the multidevices,Greer Track scores were significantly higher than all otherdevices (P< .0005 for all comparisons), and theQuantitestwas more painful than the Quintest (P = .001). In addition,pain was negatively associated with sensitivity(r = 20.77; P = .027), because the devices with greatersensitivity had lower pain scores. For the Greer Trackmultidevice with 56% sensitivity, 34% of pain scoreswere above 2. For the Greer Pick single device with98% sensitivity, only 5% of pain scores were above 2.

DISCUSSION

We have concluded a head-to-head prospective com-parative study of 8 skin test devices and found that there

are statistically significant differences among virtually alldevices tested. One device that stands out with the lowestperformance in all areas is the Greer Track. This devicewas the most painful, had the smallest mean histaminewheals and flares, was the least sensitive, and had thegreatest degree of intradevice variability. These statisticalfindings of performance may very well equate to clinicallysignificant differences in performance. Excluding theGreer Track, it is unknown whether the statistical differ-ences among the remaining 7 devices will equate toclinically significant differences in performance. Of theremaining 7 devices, all had mean histamine whealsgreater than 3 mm and mean histamine flares greaterthan 10 mm, with sensitivities from 86% to 97%. Inaddition, all of the remaining 7 devices had specificitiesof 98% or greater. Therefore, each individual providershould determine which device is best for that provider’spractice. Keep in mind that these studies were performedunder the best of circumstances, with all tests conductedby 1 technician who was certified by a representative ofthe manufacturer on the proper use of each device. Wewould recommend that technicians within a given practiceundergo this same type of training before using a givendevice. In addition, these findings may not be directlyapplicable to allergen skin testing, because we looked onlyat histamine and glycerol-saline responses. A separatestudymay be required to compare devices for this purpose.

FIG 4. Mean intradevice variability of the Greer Track. Sites are

labeled S1 through S8, with sites S1, S4, S5, and S8 representing

the corners and S2, S3, S6, and S7 representing the interior heads.

FIG 3. Mean histamine wheal size in millimeters for all devices.

TABLE IV. Pain outcome measures for 8 devices

Mean pain Median pain

Pain %*

(95% CI)

Single devices

Sharptest 1.17 1 13% (7-22)

Greer Pick 0.88 1 5% (1-11)

Accuset 0.94 1 9% (4-16)

Quintip 1.0 1 7% (2-14)

Multidevices

Multi-Test II 1.62 1 26% (17-36)

Quantitest 1.74 1 26% (17-36)

Quintest 1.45 1 17% (10-26)

Greer Track 2.04 2 34% (23-43)

*Percentage of values above 2 on the Wong-Baker FACES pain rating

scale (percentage of values interpreted as greater than mild pain).



Carr et al 345

Rhinitis,sinusitis,

and

ocu

lardisease

s

When choosing a skin test device, a few points areworth consideration. First, in our study, single devices hadlarger reactions on the arm, and multidevices had largerreactions on the back. Historically, it has been thoughtthat the back has been more reactive than the arms, andalthough our multiheaded device data concur with this,our single-headed device data do not. There are severalpossibilities for this observation, including intraoperatorvariability, inadvertent operator bias, or a true difference.With regard to the difference between single and multi-headed devices, we think that this difference is related tothe back being a flatter surface; therefore, better contact ismade with all of the test sites on amultiheaded device. Thearms are technically more challenging when placing amultiheaded device, given natural curvatures. To com-pensate for the curvatures and to ensure contact with allof the device heads, manufacturers have recommended arocking motion. With this motion, contact between allheads on a multiheaded device and the skin is achieved.However, we think that this rocking motion is responsiblefor the differences noted between individual test siteswithin a given multiheaded device, or intradevice varia-bility (Fig 4). With this rocking motion, more pressure isexerted on the skin from the corner test sites. It is importantto note that single devices also have differences in therecommended technique of application. The Quintip andSharptest use a simple downward perpendicular pressure,and both of these devices have a depth control feature.Manufacturer-recommended techniques for theGreer Pickand Accuset are slightly more complicated. The skinsurface is penetrated at an angle, and then a flick, prick-not-puncture technique is used. Neither of these last 2devices has a depth control feature. Given the tech-nique and lack of depth control, the Greer Pick andAccuset may result in greater intertechnician variabilityif care is not taken to control for these features.However, with correct technique, and by using 1 tech-nician, all single devices had sensitivities greater than90% while maintaining specificities of 98% or greater.

Another observation from our study is that skin testingis not a painful procedure on average. The mean painscores for all devices ranged from 0.88 to 2.04. Using theWong-Baker FACES pain rating scale,8 this is consideredmild pain. The degree of pain was significantly associatedwith the type of device used, with the multiheaded devicesmore painful than the single devices. However, whenmaking this comparison, it is noteworthy that with aminimal increase in pain, as many as 8 times more tests areapplied. Therefore, with a pain score of 0.88, the GreerPick applied 1 skin test, and with a pain score of 1.62, theMulti-Test II applied 8 skin tests. It is unclear whetherthese observed statistical differences in pain would equateto significant clinical differences, because all pain wasconsidered mild.

In contrast with previous studies, we did not find a clearrelationship between pain and wheal size.6 In fact, thedevice that resulted in the greatest degree of pain had thesmallest mean histamine wheal size. Again, we think it isup to the individual practitioner to consider these differ-ences when using a given device in practice.

Finally, when comparing sensitivity and specificity,there are few differences among devices, with 2 excep-tions. The Greer Track was less sensitive than all otherdevices, and the Quintest was less sensitive than the GreerPick, Sharptest, and Multi-Test II. With the 6 remainingdevices, there was no significant difference amongsensitivities. In addition, we found no significant differ-ences in specificity among devices. Overall, we found avery low false-positive rate in all devices when using themanufacturer’s recommended skin testing technique.

CONCLUSION

We have completed a prospective, head-to-head com-parison of the performance of 8 skin test devices. Thisstudy was performed under the best of clinical circum-stances, with 1 technician, trained by a representative ofthe manufacturer, who performed all skin testing, andanother technician who read all of the results. We havefound significant differences among all devices tested.Whether this equates to clinical differences is yet to bedetermined. Overall, skin testing is associated with min-imal pain, and individual providers should choose a skintest device on the basis of their own practice setting.As new devices are being produced, this study suggeststhe need for continued evaluation of these devices in aprospective, nonbiased fashion.

REFERENCES

1. Li JT, Lockey RF, Bernstein IL, Portnoy JM, Nicklas RA. Allergen

immunotherapy: a practice parameter. Ann Allergy Asthma Immunol

2003;90(suppl):1-40.

2. Position paper: immunotherapy. (EAACI) The European Academy of

Allergology and Clinical Immunology. Allergy 1993;48(suppl 14):7-35.

3. Spector SL, Nicklas RA. Practice parameters for the diagnosis and

treatment of asthma. J Allergy Clin Immunol 1995;96:707-870.

4. Adinoff AD, Rosloniec DM, McCall LI, Nelson HS. A comparison of six

epicutaneous devices in the performance of immediate hypersensitivity

skin testing. J Allergy Clin Immunol 1989;84:168-74.

5. Nelson HS, Rosloniec DM, McCall LI, Ikle D. Comparative performance

of five commercial prick skin test devices. J Allergy Clin Immunol 1993;

92:750-6.

6. Nelson HS, Lahr J, Buchmeier A, McCormick D. Evaluation of devices

for skin prick testing. J Allergy Clin Immunol 1998;101:153-6.

7. Nelson HS, Kolehmainen C, Lahr J, Murphy J, Buchmeier A. A com-

parison of multiheaded devices for allergy skin testing. J Allergy Clin

Immunol 2004;113:1218-9.

8. Wong-Baker FACES Pain Rating Scale. In: Wong DL, Hockenberry-

Eaton M, Wilson D, Winkelstein ML, Schwartz P. Wong’s essentials of

pediatric nursing. 6th ed. St Louis: Elsevier; 2001. p. 1301-2.


AUGUST 2005

346 Carr et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

Allergen-specific nasal IgG antibodies inducedby vaccination with genetically modifiedallergens are associated with reduced nasalallergen sensitivity

Jurgen Reisinger, MSc,a Friedrich Horak, MD,a Gabrielle Pauli, MD,e

Marianne van Hage, MD,d Oliver Cromwell, PhD,f Franz Konig,c

Rudolf Valenta, MD,b and Verena Niederberger, MDa Vienna, Austria,

Stockholm, Sweden, Strasbourg, France, and Reinbek, Germany

Background: We have performed a double-blind, placebo-

controlled injection immunotherapy study with genetically

modified derivatives of the major birch pollen allergen,

Bet v 1 (Bet v 1–trimer, Bet v 1–fragments).

Objective: To investigate whether vaccination with genetically

modified allergens induces allergen-specific antibodies in nasal

secretions and to study whether these antibodies affect nasal

allergen sensitivity.

Methods: A randomly picked subgroup of patients (n = 23;

placebo, n = 10; trimer, n = 10; fragments, n = 3) was subjected

to an extensive analysis of serum samples and nasal lavage

fluids and to nasal provocation testing. Bet v 1–specific IgG1-4

and IgA antibodies were determined in serum samples

obtained before and after vaccination, after the birch pollen

season, and 1 year after start of vaccination as well as in nasal

lavage fluids obtained after the birch pollen season and 1 year

after start of vaccination by ELISA. Nasal sensitivity to

natural, birch pollen–derived Bet v 1 was determined by active

anterior rhinomanometry after the birch pollen season and

1 year after start of vaccination.

Results: Vaccination with genetically modified Bet v 1

derivatives, but not with placebo, induced Bet v 1–specific

IgG1, IgG2, and IgG4, and low IgA antibodies in serum, which

also appeared in nasal secretions, but no IgG3 antibodies. The

levels of therapy-induced Bet v 1–specific IgG4 antibodies in

nasal secretions were significantly (P < .05) associated with

reduced nasal sensitivity to natural, birch pollen–derived

Bet v 1 as objectively determined by controlled nasal

provocation experiments.

Conclusion: Our data demonstrate that vaccination with

genetically modified allergens induces IgG antibody responses

against the corresponding natural allergen not only in serum

but also in mucosal fluids, where they may protect against

allergen-induced inflammation. (J Allergy Clin Immunol

2005;116:347-54.)

Key words: Allergen-specific immunotherapy, IgG antibodies,genetically modified allergens, nasal provocation

In the last 15 years, considerable progress has beenmade in the field of molecular allergen characterization.The repertoire of the disease-eliciting allergens has beenrecreated in the form of recombinant allergens for the mostcommon allergen sources and can be used for diagnostictests that allow dissection of the reactivity profiles ofpatients down to amolecular level.1,2 In attempts to reduceIgE-mediated side effects during allergen-specific immu-notherapy, several research groups have used geneticengineering and peptide chemistry to develop allergenderivatives with reduced allergenic activity.3-7 By usinggenetically engineered derivatives of the major birchpollen allergen, Bet v 1, we have conducted a first double-blind, placebo-controlled immunotherapy study in pa-tients allergic to birch pollen.8 One group of patients wastreated with a mixture of 2 recombinant Bet v 1 fragmentsthat exhibited strongly reduced IgE reactivity and aller-genic activity; a second group received a recombinantBet v 1 trimer, also characterized by reduced allergenicactivity; and the third group was treated with aluminiumhydroxide alone, which was used as adjuvant for thesubcutaneous injections.8-12 The analysis of the immuno-logical mechanisms showed that vaccination with thegenetically modified Bet v 1 derivatives induced de novoserum IgG antibodies that recognized the Bet v 1–wild-type allergen and inhibited Bet v 1–induced histaminerelease from basophils.8 Furthermore, we obtained evi-dence that boosts of the IgE memory responses caused byseasonal allergen exposure were reduced in activelyvaccinated patients.8

In this sub-study, we asked whether vaccination withthe genetically engineered allergens also leads to aninduction of allergen-specific antibodies in the nasalmucosa, the main site of allergic inflammation, and ifso, whether such antibodies are associated with reduced

From athe Department of Otorhinolaryngology, and bthe Department of

Pathophysiology, Center for Physiology and Pathophysiology, Vienna

General Hospital, and cthe Department of Medical Statistics, Medical

University of Vienna; dthe Department of Medicine, Clinical Immunology

and Allergy Unit, Karolinska Institutet and Hospital, Stockholm; eService

de Pneumologie, Hopitaux Universitaires de Strasbourg; and fAllergopharma

Joachim-Ganzer KG, Reinbek.

Supported by grants F01811 and F01815 of the Austrian Science Fund and by

a research grant from Biomay, Vienna.




Reprint requests: Verena Niederberger, MD, Department of Otorhinolaryn-

gology, Vienna General Hospital, AKH, Medical University of Vienna,

Waehringer Guertel 18-20, A-1090 Vienna, Austria. E-mail: Verena.

[email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.003

347

Rhinitis,sinusitis,

and

ocu

lardisease

s



Abbreviation used

r: Recombinant

allergen sensitivity in the nose as evaluated objectively bynasal provocation. For this purpose, we performed an in-depth analysis of serum and nasal lavage fluids regardingallergen-specific antibodies and extensive nasal provoca-tion in a subgroup of 23 subjects recruited at random frompatients participating in the immunotherapy trial.

METHODS

Recombinant allergens and vaccineformulation

Recombinant major birch pollen allergen, Bet v 1,13 mimicking

natural birch pollen–derived Bet v 1 was obtained from Biomay

(Vienna, Austria). Recombinant Bet v 1 fragments and trimer were

expressed in Escherichia coli and purified as described.9,10 Alumin-

ium hydroxide adsorbates containing 100 mg protein/mL adsorbate

trimer or aluminium hydroxide alone (placebo) were formulated as

described14 following Good Manufacturing Practice guidelines.

Vaccination of patients allergic to birch pollenwith Bet v 1 derivatives or placebo

The immunotherapy studywas conducted as a placebo-controlled,

double-blind, randomized vaccination trial over a period of 12

months with 1 preseasonal treatment course in 3 study centers

(Stockholm, Strasbourg, Vienna). Patients were included if they had a

positive result with skin prick test and ImmunoCAP (3.5 kU/L;

Pharmacia, Uppsala, Sweden) for Bet v 1 and natural birch pollen

extract and had a history of moderate to severe seasonal allergic

rhinoconjunctivitis attributable to birch pollen. Subcutaneous injec-

tions of aluminium hydroxide–adsorbed allergen derivatives con-

taining increasing doses (1-80mg) of recombinant Bet v 1 derivatives

(Bet v 1–trimer, Bet v 1–fragments) or placebo were administered

at intervals of 1 to 2 weeks as a preseasonal treatment. After reaching

the maximal dose, the treatment was continued at 4-week intervals

until before the flowering season. Clinical outcome was monitored

by questionnaires, skin prick tests, and nasal provocation tests. The

study was approved by the local ethical committees, and written

informed consent was obtained from each patient.

In a substudy, nasal lavage fluids were collected in 23 randomly

picked patients from a total of 71 at the Vienna center. Serum samples

TABLE I. Demographic and clinical data of the 23 patients (trimer, n = 10; fragments, n = 3; placebo, n = 10) included in this

study. The numbering of the immunotherapy trial was preserved and indicates that the patients were randomly picked

when the study was still blind.

Pat.-Nr. Age Sex Sensitized to Symptoms Treatment Nr. of injections

Cumulative injected

dose (mg)

2 33 m b, g r, c trimer 10 265

8 39 m b r, c, a trimer 9 185

15 42 m b r, c, a trimer 8 165

18 36 f b r, c, a trimer 10 169

26 27 f b, g, w r, a trimer 8 165

33 51 f b, a r, c, a trimer 8 163

34 28 m b, g, w, a r, c, a trimer 8 165

41 45 m b r, c, a trimer 9 245

44 37 m b r, c trimer 8 69

49 33 m b r, c trimer 8 165

36 45 f b, w, a r, c, a, d fragments 9 103

42 51 m b, g, w, a r, c, a fragments 7 85

64 36 m b, g r, c fragments 9 245

3 28 f b, g, w r, c placebo 8 0

5 38 m b, w r, c, a placebo 8 0

6 23 m b, g, a r, c placebo 8 0

28 26 f b r, c placebo 9 0

37 30 m b, g, w, a, m r, c placebo 9 0

51 55 f b, w, a r, c placebo 9 0

56 36 m b r, c placebo 8 0

57 40 f b r, c, a placebo 8 0

58 43 m b, g, w, a r, c, a placebo 9 0

61 35 m b, g r, c, a placebo 9 0

Abbreviations used: m: male, f: female; b: birch pollen, g: grass pollen, w: weed pollen, a: animal dander, m: mites; symptoms: r: rhinitis, c: conjunctivitis,

a: asthma, d: dermatitis

FIG 1. Experimental protocol. Patients received a single course of

subcutaneous injections (therapy) before the birch pollen season

and were monitored for almost 1 year. Times of blood sampling,

nasal washings, and provocations are indicated.


AUGUST 2005

348 Reisinger et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

and nasal lavage fluids as well the results from the nasal provocation

tests were analyzed after deblinding of the immunotherapy study

(Table I).

Experimental protocol

The sequence of serum sampling, nasal washings, and nasal

provocation tests is illustrated in Fig 1.

Collection of nasal lavage fluidsand serum samples

Nasal lavage fluids were obtained from 23 patients (treated with

Bet v 1 fragments, n = 3; Bet v 1 trimer, n = 10; placebo, n = 10)

after the birch pollen season (May 2001) and 1 year after beginning of

the study (autumn/winter 2001).

To obtain nasal secretions, patients sat with their head bent

forward, and 1 nostril was sealed with a foam plastic plug encircling

the end piece of a narrow plastic tube. The respective side of the nose

was slowly filled and emptied 5 times with 5 mL prewarmed 0.9%

saline from a syringe attached to the other end of the plastic tube.

Subsequently, lavages were obtained from the other nostril.

Serum samples were obtained before treatment (autumn/winter

2000), after treatment (February/March 2001), after the birch pollen

season (May 2001), and in autumn/winter 2001. Nasal lavage fluids

and serum samples were stored at 220C until analysis.

Nasal provocation tests

Nasal provocation tests were performed with natural birch pollen

extract containing defined Bet v 1 concentrations before treatment

(November/December 2000), after the birch pollen season (May

2001), and after 1 year (October 2001). Allergen solutions were

freshly prepared from the standardized lyophilized birch pollen

extract, were kept at 4C between tests, and were discarded after

48 hours.

FIG 2. Bet v 1–specific antibody levels in nasal lavage fluids (diluted 1:2) and serum (diluted 1:50). IgG1 (A),

IgG2 (B), IgG4 (C), and IgA (D) levels (y-axis, OD values) were measured in nasal secretions (circles, left panel)

and sera (rhombuses, right panel) from patients treated with trimer (black circles or rhombuses), fragments

(grey circles or rhombuses), or placebo (white circles or rhombuses) at different times (x-axis). Bars indicate

mean values. Statistically significant differences for nasal lavage fluid data are indicated with P values.



Reisinger et al 349

Rhinitis,sinusitis,

and

ocu

lardisease

s

The baseline levels of nasal parameters (nasal flow and resistance)

were established by active anterior rhinomanometry (Allergopharma

Rhinomanometer; Allergopharma, Reinbek, Germany) without any

interaction and 10 minutes after application of 0.9% sodium chloride.

Thereafter, the patient received increasing doses of birch pollen

extract solution containing 0.0064 mg/mL, 0.064 mg/mL, 0.64 mg/

mL, 6.4 mg/mL, and 64 mg/mL Bet v 1, or placebo. Approximately

0.05 mL (1 pump action) of allergen solutions was administered into

both nostrils by using ametered dose pump.During application of test

solutions, patients had to hold their breath in full inspiration to avoid

bronchial provocation. Changes in nasal parameters were determined

15 and 20 minutes after allergen application. The nasal flow and

resistance values with saline were used as a reference for calculations

of changes of flow and resistance with birch pollen allergen. Local

and systemic symptoms were recorded 15 minutes after provocation

(local symptoms—secretion: mild = 1 score, intensive = 2 scores;

sneezing: 3-53 = 1 score, >53 = 2 scores; systemic symptoms—

tear flow and/or itching of throat and/or ears = 1 score, conjunctivitis

and/or chemosis and/or urticaria and/or cough and/or dyspnea = 2

scores). The test was regarded as positive if a 40% or more reduction

of nasal air flow was obtained or if a symptom score of 3 was

exceeded. If the test was negative after 15 and 20 minutes, the testing

procedure was repeated with a 10-fold dose increase. When the test

became positive, the maximum allergen concentration tolerated by

the patient was recorded, and the provocation test was concluded.

Reduction of nasal sensitivity to birch pollen extract in nasal

provocation experiments was expressed as change in tolerated

concentration steps.

Measurement of antibodies specific forwild-type Bet v 1, Bet v 1–fragments,and Bet v 1–trimer by ELISA

Serum IgG1, IgG2, IgG3, and IgG4 subclass antibodies (serum

dilution, 1:50) specific for recombinant (r) Bet v 1, Bet v 1–fragments,

and Bet v 1–trimer were measured as described.15 Bet v 1–specific

serum IgA antibodies (serum dilution, 1:50) were detected by using

an alkaline phosphatase–conjugated mouse monoclonal antihuman

IgA antibody (G20-359; dilution, 1:1000; BD Pharmingen, San

Diego, Calif) that recognizes IgA1, IgA2, and secretory IgA. The

alkaline phosphatase–labeled detection antibody was developed as

described.16

Nasal lavage fluids were diluted 1:2 in PBS, 0.05% vol/vol Tween

20, 0.5 % wt/vol BSA (for IgG1, IgG2, IgG3, and IgG4 determi-

nations); or in TRIS-buffered saline, 0.05% vol/vol Tween 20, 0.5 %

wt/vol BSA (for IgA determinations). ELISA measurements were

performed as described for serum samples. All determinations were

performed in duplicate with less than 5% deviation, and results are

displayed as means.


Spearman rank correlation coefficient r was used to assess

correlations between parameters. Wilcoxon 2-sample tests were

used to check for differences in IgG1, IgG2, IgG4, and IgA levels

between actively treated and placebo-treated patients. P values <.05were considered statistically significant.

RESULTS

Demographic, clinical, and immunologicalcharacterization of patients allergicto birch pollen

Thirteen men and 10 women with a mean age of38 years were analyzed. Nine of these patients were aller-gic exclusively to birch pollen, whereas the other 14 pa-tients also had allergic symptoms to other allergen sources.The patients were sensitized exclusively to the majorbirch pollen allergen, Bet v 1, within birch pollen. Allpatients had allergic rhinitis, all but 1 also had allergicconjunctivitis, and 13 had mild seasonal allergic asthma.On an average, patients had received 8.52 injections(placebo = 8.50, range = 8-9; treatment = 8.54, range = 7-10).Actively treated patients had received a cumulativeinjected dose of 164.4 mg recombinant protein on average(range = 69-265 mg).

Vaccination with recombinant Bet v 1derivatives induces Bet v 1–specificIgG1, IgG2, and IgG4 but not IgG3 andIgA antibodies in nasal secretions

Levels of IgG1-4 subclass and IgA antibodies specificfor Bet v 1 were measured in nasal secretions obtained inMay (after the birch pollen season) and in October 2001(1 year after the beginning of immunotherapy). In May2001, IgG1 levels to Bet v 1 were significantly higher innasal lavage fluids from actively treated patients than innasal lavage fluids from patients who had receivedplacebo (P < .05; Fig 2, A, left panel). We also foundhigher Bet v 1–specific IgG2 and IgG4 levels in nasalsecretions from actively treated patients compared withthe placebo group, but these differences did not reachsignificance (IgG2, Fig 2, B, left panel; IgG4, Fig 2, C, leftpanel). In October 2001, 1 year after treatment, antibodyconcentrations declined in the nasal lavage samples,indicating that vaccination-induced elevations of Bet v 1–specific IgG levels in nasal secretions show a kineticsimilar to that of serum antibody levels (Fig 2, A-C, left

FIG 3. Association of vaccination-induced nasal Bet v 1–specific

IgG1 antibody concentrations with the cumulative injected dose

(mg) of either Bet v 1 trimer or fragments. The cumulative injected

dose (mg) received during treatment is shown on the x-axis. IgG1

levels specific for Bet v 1 are indicated as OD values on the y-axis.

Black spots, treatment with Bet v 1–trimer; grey spots, treatment

with Bet v 1–fragments; white spots, placebo treatment.


AUGUST 2005

350 Reisinger et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

panels).8 Active treatment did not lead to increased nasalIgG3 (data not shown) and IgA (Fig 2,D, left panel) levels.Furthermore, we noted no relevant difference betweenMay 2001 and October 2001 measurements of nasal IgG3

and IgA levels.The levels of Bet v 1–specific IgG1 in nasal secretions

were dependent on the dose of genetically modifiedallergens injected during immunotherapy. A statisticallysignificant correlation (r = 0.446; P < .05) was found

between the cumulative injected dose of the derivativesand IgG1 (May 2001) in nasal lavage fluids (Fig 3).

Bet v 1–specific nasal antibodies mirrorserum antibody responses

Sera had been collected at several time points (Fig 1;before therapy, autumn 2000; after therapy, spring 2001;after the birch pollen season, May 2001; and 1 year afterthe beginning of the study, autumn 2001).

FIG 4. Vaccination with Bet v 1 derivatives (trimer, fragments) induces IgG1 (A) and IgG4 (B) antibodies to wild-

type Bet v 1 and to the derivatives. Levels of IgG1 and IgG4 to wild-type Bet v 1, Bet v 1–fragment 1, Bet v 1–

fragment 2, and Bet v 1–trimer (x-axis) were determined in the serum of the 13 patients who had received

active treatment at 4 different time points. IgG1 (A) and IgG4 (B) levels are indicated as OD values on the y-axis.

FIG 5. Immunotherapy-induced nasal Bet v 1–specific IgG1 (A), IgG2 (B), and IgG4 (C) antibody levels in

May 2001 and Bet v 1–specific nasal IgG4 levels in October 2001 (D) are associated with the respective

serum antibody concentrations. IgG1,2,4 antibody levels in nasal lavage fluids are shown as OD values on the

x-axis. The serum IgG1,2,4 concentrations are indicated as OD values on the y-axis. Black spots, treatment

with Bet v 1–trimer; grey spots, treatment with Bet v 1–fragments; white spots, placebo treatment.



Reisinger et al 351

Rhinitis,sinusitis,

and

ocu

lardisease

s

Actively treated patients exhibited significantly higherBet v 1–specific serum IgG1 (Fig 2, A, right panel), IgG2

(Fig 2, B, right panel), and IgG4 (Fig 2, C, right panel)antibody levels than placebo-treated patients in February2001 and May 2001 (all P values <.05). No relevantalterations in IgG3 (data not shown) and IgA (Fig 2, D,right panel) levels were observed after active treatment.Active treatment induced IgG1 and IgG4 antibodiesspecific for intact Bet v 1 as well as for Bet v 1 derivatives(Fig 4).

Significant correlations were found between serumand nasal lavage fluid IgG1 (r = 0.572; P < .01; Fig 5, A),IgG2 (r = 0.498; P < .05; Fig 5, B), and IgG4 (r = 0.623;P < .01; Fig 5, C) antibody levels in May and betweenIgG4 levels in October 2001 (r = 0.461; P < .05; Fig 5, D).No such correlations were found concerning IgA levels(data not shown).

Vaccination-induced Bet v 1–specific IgGantibody levels are associated with reducednasal sensitivity to Bet v 1

All 23 patients completed the first (November/December 2000) and third (October 2001) provocationtest, but 2 trimer-treated patients and 1 placebo-treatedpatient did not undergo the second (May 2001) provoca-tion. Therapy-induced Bet v 1–specific IgG4 antibodylevels (May 2001) in nasal secretions were significantlycorrelated with a reduced nasal sensitivity to birch pollen–derived Bet v 1 allergen (r = 0.590; P < .01; Fig 6, B).Furthermore, there is a weak correlation between nasalIgG1 levels and reduced nasal sensitivity (r = 0.425), butthis did not achieve statistical significance (P = .062; Fig6, A). By contrast, there were no associations between

either the nasal IgG2, IgG3, and IgA levels and the resultsfrom the May nasal provocation tests, or the October nasalantibody levels and the outcomes of the October nasalprovocation tests. None of the patients reacted to saline innasal provocation tests.

DISCUSSION

We have recently described that injection immunother-apy with genetically modified derivatives (ie, rBet v 1fragments, rBet v 1 trimer) of the major birch pollenallergen, Bet v 1, induces serum IgG antibody responsesrecognizing the wild-type allergen, which inhibit allergeninduced histamine release from basophils in vitro.8 In thissubstudy, we demonstrate that IgG antibodies appear alsoin nasal secretions and that their presence is associatedwith reduced allergen-specific nasal sensitivity.

Treatment of patients with genetically modified ver-sions of Bet v 1 induced the generation of antibodies, notonly against the Bet v 1–fragments and Bet v 1–trimer, butalso against wild-type Bet v 1 allergen in serum and nasallavage fluids. This finding can be explained by the fact thatboth types of genetically engineered Bet v 1 derivativeswere found to induce IgG antibodies in animals. TheseIgG antibodies recognized natural, pollen-derived Bet v 1and inhibited allergic patients’ IgE binding to Bet v 1.9,17

Because the therapy-induced responses to wild-typeBet v 1 were as strong as (ie, trimer) and sometimeseven stronger than to the derivatives (ie, fragments), itappears that most of the IgG response induced by vacci-nation with the derivatives is directed against the wild-type allergen. The composition of the Bet v 1–specific

FIG 6. Association of vaccination-induced IgG1 and IgG4 antibody levels with changes in nasal sensitivity to

birch pollen–derived Bet v 1. Differences between the concentration of birch pollen extract tolerated in nasal

provocation experiments before (November 2000) and after (May 2001) immunotherapy are shown on the

x-axis as concentration steps. A value of 11 indicates that a 10-fold higher concentration of Bet v 1 was

tolerated in May 2001 compared with November 2000. IgG1 (A) and IgG4 (B) levels specific for Bet v 1 are

displayed as OD values on the y-axis. Two trimer-treated patients and 1 placebo-treated patient did

not undergo the second (May 2001) provocation. Black spots, treatment with Bet v 1–trimer; grey spots,

treatment with Bet v 1–fragments; white spots, placebo treatment.


AUGUST 2005

352 Reisinger et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

antibodies in nasal lavage fluids mirrored that of the serumantibodies regarding subclasses and kinetics. In serum aswell as in nasal lavage fluids, Bet v 1–specific IgG1, IgG4,and IgG2, low IgA, and no IgG3 responses were inducedby vaccination, and the kinetics of antibody levels weresimilar in serum and nasal lavage fluids. Bet v 1–specificIgG levels were high after vaccination (May) and declinedalmost to baseline by October of the same year. The latterobservations and previous studies18,19 on allergen-specificantibody reactivities in mucosal fluids support the as-sumption that the nasal allergen-specific IgG may resultfrom transudation.

The importance of the vaccine-induced Bet v 1–specificIgG antibodies is indicated by the fact that their concen-trations were associated with a reduction of nasal allergensensitivity and by the observation that this protectiveeffect disappeared after their decline in October. We andothers have found that the vaccine-induced IgG antibodiesinhibited allergen-induced degranulation of basophils20-23

in vitro and hence assume also that the nasal Bet v 1–specific IgG may act as blocking antibodies that mask IgEepitopes of Bet v 1 and thus prevent mast cell degranu-lation in the submucosal tissues. In addition, it is possiblethat the Bet v 1–specific IgG acts as protective shieldpreventing the intrusion of allergen into the submucosa. Infact, several recent studies point to the importance ofallergen-specific IgG antibodies for the success of immu-notherapy.20,24 For example, it has been demonstrated thatallergen-specific IgG can inhibit IgE-mediated allergenpresentation to T cells.25

The nasal mucosa is one of the most important sites forallergic inflammation, and there is also increasing evi-dence that local IgE production in the nasal mucosa playsan important role in systemic IgE responses.26 In thiscontext, it should be noted that patients vaccinated withthe genetically modified Bet v 1 derivatives had exhibitedlower boosts of systemic IgE production after seasonalbirch pollen exposure than the placebo-treated group.8 It ishence tempting to speculate that vaccine-induced nasalIgG antibodies may inhibit the boosting of IgE memorycells in the nasal mucosa by allergen contact in vaccinatedpatients.

Should further studies provide evidence for the impor-tance of allergen-specific nasal IgG antibody responsesfor the success of immunotherapy, it will be possible todevelop more effective protocols for allergen-specificimmunotherapy.

REFERENCES

1. Valenta R, Kraft D. From allergen structure to new forms of allergen-

specific immunotherapy. Curr Opin Immunol 2002;14:718-27.

2. Hiller R, Laffer S, Harwanegg C, Huber M, Schmidt WM, Twardosz A,

et al. Microarrayed allergen molecules: diagnostic gatekeepers for allergy

treatment. FASEB J 2002;16:414-6.

3. Valenta R. The future of antigen-specific immunotherapy of allergy. Nat

Rev Immunol 2002;2:446-53.

4. Singh MB, de Weerd N, Bhalla PL. Genetically engineered plant

allergens with reduced anaphylactic activity. Int Arch Allergy Immunol

1999;119:75-85.

5. Ferreira F, Wallner M, Breiteneder H, Hartl A, Thalhammer J, Ebner C.

Genetic engineering of allergens: future therapeutic products. Int Arch


6. Vrtala S, Focke-Tejkl M, Swoboda I, Kraft D, Valenta R. Strategies for

converting allergens into hypoallergenic vaccine candidates. Methods

2004;32:313-20.

7. Drew AC, Eusebius NP, Kenins L, de Silva HD, Suphioglu C, Rollan

JM, et al. Hypoallergenic variants of the major latex allergen Hev b 6

retaining human T lymphocyte reactivity. J Immunol 2004;173:5872-9.

8. Niederberger V, Horak F, Vrtala S, Spitzauer S, Krauth MT, Valent P,

et al. Development of a novel allergy vaccine: genetically engineered

allergens prevent progression of allergic disease. Proc Nat Acad Sci

U S A 2004;101(suppl 2):14677-82.

9. Vrtala S, Hirtenlehner K, Vangelista L, Pastore A, Eichler HG, Sperr

WR, et al. Conversion of the major birch pollen allergen, Bet v 1 into

two nonanaphylactic T cell epitope-containing fragments: candidates for

a novel form of specific immunotherapy. J Clin Invest 1997;99:1673-81.

10. Vrtala S, Hirtenlehner K, Susani M, Akdis M, Kussebi F, Akdis CA,

et al. Genetic engineering of a hypoallergenic trimer of the major birch

pollen allergen Bet v 1. FASEB J 2001;15:2045-7. (Express article

Available at: www.fasebj.org/cgi/reprint/00-0767fjev1).

11. Pauli G, Purohit A, Oster JP, De Blay F, Vrtala S, Niederberger V, et al.

Comparison of genetically engineered hypoallergenic rBet v 1 derivatives

with rBet v 1 wild-type by skin prick and intradermal testing: results

obtained in a French population. Clin Exp Allergy 2000;30:1076-84.

12. van Hage-Hamsten M, Kronqvist M, Zetterstrom O, Johansson E,

Niederberger V, Vrtala S, et al. Skin test evaluation of genetically

engineered hypoallergenic derivatives of the major birch pollen allergen,

Bet v 1: results obtained with a mix of two recombinant Bet v 1 fragments

and recombinant Bet v 1 trimer in a Swedish population before the birch

pollen season. J Allergy Clin Immunol 1999;104:969-77.

13. Ferreira FD, Hoffmann-Sommergruber K, Breiteneder H, Pettenburger

K, Ebner C, Sommergruber W. Purification and characterization of

recombinant Bet v 1, the major birch pollen allergen: immunological

equivalence to natural Bet v 1. J Biol Chem 1993;268:19574-80.

14. Mahler V, Vrtala S, Kuss O, Diepgen TL, Suck R, Cromwell O, et al.

Vaccines for birch pollen allergy based on genetically-engineered

hypoallergenic derivatives of the major birch pollen allergen, Bet v 1.

Clin Exp Allergy 2004;34:510-2.

15. Vrtala S, Susani M, Sperr WR, Valent P, Laffer S, Dolecek C, et al.

Immunologic characterization of purified recombinant timothy grass

pollen (Phleum pratense) allergens (Phl p 1, Phl p 2, Phl p 5). J Allergy

Clin Immunol 1996;97:781-7.

16. Denepoux S, Eibensteiner PB, Steinberger P, Vrtala S, Visco V,

Weyer A, et al. Molecular characterization of human IgG monoclonal

antibodies specific for the major birch pollen allergen Bet v 1: anti-

allergen IgG can enhance the anaphylactic reaction. FEBS Lett 2000;

465:39-46.

17. Vrtala S, Akdis CA, Budak F, Akdis M, Blaser K, Kraft D, et al. T cell

epitope-containing hypoallergenic recombinant fragments of the major

birch pollen allergen, Bet v 1, induce blocking antibodies. J Immunol

2000;165:6653-9.

18. Hoffmann-Sommergruber K, Ferreira ED, Ebner C, Barisani T,

Korninger L, Kraft D, et al. Detection of allergen-specific IgE in tears

of grass pollen-allergic patients with allergic rhinoconjunctivitis. Clin

Exp Allergy 1996;26:79-87.

19. Aghayan-Ugurluoglu R, Ball T, Vrtala S, Schweiger C, Kraft D,

Valenta R. Dissociation of allergen-specific IgE and IgA responses in

sera and tears of pollen-allergic patients: a study performed with

purified recombinant pollen allergens. J Allergy Clin Immunol 2000;

105:803-13.

20. Flicker S, Valenta R. Renaissance of the blocking antibody concept in

type I allergy. Int Arch Allergy Immunol 2003;132:13-24.

21. Mothes N, Heinzkill M, Drachenberg KJ, Sperr WR, Krauth MT, Majlesi

Y, et al. Allergen-specific immunotherapy with a monophosphoryl lipid

A-adjuvanted vaccine: reduced seasonally boosted IgE production and

inhibition of basophil histamine release by therapy-induced blocking

antibodies. Clin Exp Allergy 2003;33:1-11.

22. Ball T, Sperr WR, Valent P, Lidholm J, Spitzauer S, Ebner C, et al.

Induction of antibody responses to new B cell epitopes indicates

vaccination character of allergen immunotherapy. Eur J Immunol 1999;

29:2026-36.



Reisinger et al 353

Rhinitis,sinusitis,

and

ocu

lardisease

s

23. Clinton PM,KemenyDM,YoultenLJ, LessofMH.Histamine release from

peripheral blood leukocytes with purified bee venom allergens: effect of

hyperimmune beekeeper plasma. Int ArchAllergy Immunol 1989;89:43-8.

24. Wachholz PA, Durham SR. Mechanisms of immunotherapy: IgG

revisited. Curr Opin Allergy Clin Immunol 2004;4:313-8.

25. Wachholz PA, Kristensen Soni N, Till SJ, Durham SR. Inhibition of

allergen-IgE binding to B cells by IgG antibodies after grass pollen

immunotherapy. J Allergy Clin Immunol 2003;112:915-22.

26. Durham SR, Gould HJ, Hamid QA. Local IgE production in nasal

allergy. Int Arch Allergy Immunol 1997;113:128-30.


AUGUST 2005

354 Reisinger et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

Levocetirizine: Pharmacokinetics andpharmacodynamics in children age6 to 11 years

F. Estelle R. Simons, MD, FRCPC,a and Keith J. Simons, PhDa,b Winnipeg,

Manitoba, Canada

Background: The pharmacokinetics and pharmacodynamics

of medications may differ between children and adults,

necessitating different dose regimens for different age groups.

Levocetirizine, the active enantiomer of cetirizine, is used in

the treatment of allergic rhinitis and chronic urticaria in

Europe. Its pharmacokinetics and pharmacodynamics have

not yet been studied prospectively in school-age children.

Objectives: This study was performed to investigate

levocetirizine pharmacokinetic disposition and pharma-

codynamics in relation to skin reactivity to histamine

in children aged 6 to 11 years.

Methods: Blood samples were obtained at predose baseline

and at defined intervals up to and including 28 hours after

a 5-mg levocetirizine dose. Concurrently, epicutaneous tests

with histamine phosphate, 1 mg/mL, were performed.

Wheals and flares were traced at 10 minutes, and the areas

were measured with a computerized digitizing system.

Results: In children aged 8.6 6 0.4 years (6 SEM), the peak

levocetirizine concentration was 450 6 37 ng/mL, and the time

at which peak concentrations occurred was 1.2 6 0.2 hours.

The terminal elimination half-life was 5.7 6 0.2 hours, the

oral clearance was 0.82 6 0.05 mL/min/kg, and the volume of

distribution was 0.4 6 0.02 L/kg. Compared with predose

areas, the wheals and flares produced by histamine

phosphate were significantly decreased from 1 to 28 hours,

inclusive (P < .05). Mean maximum inhibition of wheals

and flares occurred from 2 to 10 hours (97% 6 1%) and

from 2 to 24 hours (93% 6 1%), respectively.

Conclusions: Levocetirizine had an onset of action within

1 hour and provided significant peripheral antihistaminic

activity for 28 hours after a single dose. Once-daily dosing

may be optimal in children aged 6 to 11 years, as it is in

adults. (J Allergy Clin Immunol 2005;116:355-61.)

Key words: H1-antihistamine, levocetirizine, pharmacokinetics,

pharmacodynamics, wheal, flare, allergic rhinitis, urticaria, children

The pharmacokinetics (absorption, distribution, metab-olism, and excretion) and pharmacodynamics of manymedications, including those used in the treatment ofallergic diseases, have not been optimally investigated inthe pediatric population.1 In the absence of such clinicalpharmacology data, drug doses and dose intervals have tobe extrapolated from those recommended for adults, andthe dose and dose interval selected may not be optimallyefficacious or safe in children. Indeed, many drug regu-latory agencies now mandate clinical pharmacology stud-ies in the pediatric population.2

More than 40 H1-antihistamines are used in the treat-ment of allergic rhinitis, urticaria, and other diseases.3

Most of the orally administered H1-antihistamines areavailable in dosage formulations suitable for administra-tion to children and even to infants; however, only 11 ofthe 40 H1-antihistamines have been studied prospectivelyin children with regard to their pharmacokinetics andpharmacodynamics.4-23 These studies have generally beenconducted after administration of a single dose,5-10,12-20

but 3 studies have been performed at steady state,11,12,20

and in a few studies, a population pharmacokineticdesign21-23 has been used. The clinical pharmacology ofa few of the first-generation H1-antihistamines, such aschlorpheniramine, brompheniramine, diphenhydramine,and hydroxyzine, was investigated after they had beenused in children for several decades. In contrast, thepharmacokinetics and pharmacodynamics of the second-generation H1-antihistamines cetirizine, fexofenadine,ebastine, loratadine, levocetirizine, and mizolastine havebeen investigated in the pediatric population relativelyearly in drug development.

In the present study our objective was to characterizethe pharmacokinetics and pharmacodynamics of thenew H1-antihistamine levocetirizine in children aged 6to 11 years. Levocetirizine24-29 (Fig 1) is the active R-enantiomer of the racemate cetirizine. It is highly selectivefor the human histamine H1-receptor, at which it hastwice the binding affinity of cetirizine. Levocetirizinehas conformational stability and is not converted todextrocetirizine, the S-enantiomer, which has 30-foldless binding affinity than cetirizine at the H1-receptor.Levocetirizine is minimally metabolized; during theweek after administration of a single oral 14C-labeled dose

From athe Department of Pediatrics and Child Health, Department of

Immunology, Canadian Institutes of Health Research National Training

Program in Allergy and Asthma, Faculty of Medicine, and bthe Faculty of

Pharmacy, Department of Pediatrics and Child Health, Faculty of Medicine,

The University of Manitoba.

Supported by an Institutional Grant from UCB Pharma, Inc (Belgium) to the

University of Manitoba and the Health Sciences Centre.

Disclosure of potential conflict of interest: Grants and research support from

UCB Pharma.

Received for publication December 22, 2004; revised April 4, 2005; accepted



Reprint requests: F. Estelle R. Simons, MD, FRCPC, 820 Sherbrook St,

Winnipeg, Manitoba, Canada R3A 1R9. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.010

355

Rhinitis,sinusitis,

and

ocu

lardisease

s


Abbreviations usedEC50: Plasma concentration producing 50% of Emax

Emax: Maximum effect attributable to medication

to adults, 85.4% and 12.9% of the drug can be recoveredunchanged in urine and feces, respectively.26 Like enan-tiomers of other medications, levocetirizine is consideredto be a new chemical entity, and as such, its pharmaco-kinetics, pharmacodynamics, efficacy, and safety needto be defined in individuals in various age groups. Wehypothesized that in children aged 6 to 11 years, as inadults, it would have prompt onset of action and wouldalso have peripheral H1-antihistaminic activity lasting atleast 24 hours after a single dose.

METHODS

To test the hypothesis stated above, we performed a prospective,

open-label, single-dose study of levocetirizine involving objective

pharmacokinetic and pharmacodynamic measurements. Approval

for levocetirizine administration was obtained through a New Drug

Submission to Health Canada. The study protocol was approved

by the University of Manitoba Research Ethics Board on the Use of

Human Subjects in Research. Before study entry, written assent was

obtained from each child, and written informed consent was obtained

from the parent or parents of each child.

Selection of participants

Children were eligible to participate if they were 6 to 11 years of

age, weighed 20 to 40 kg, and had mild allergic rhinitis. They were

excluded if they had any recent acute illness or any other health

problem except for mild intermittent or persistent asthma or if they

required any oral medication, including any oral H1-antihistamines,

in the week before study entry or during the study. The only medi-

cations permitted before and during the study were as follows: low-

dose (100 mg) intranasal glucocorticoids for rhinitis and low-dose

(250 mg) inhaled glucocorticoids and as-needed inhaled albuterol

for asthma.

Study outline

During a preliminary visit to theManitoba Institute of ChildHealth

Pediatric Allergy Laboratory, the children were assessed for their

ability to meet the inclusion criteria of the study. Medical history was

obtained, and physical examination, complete blood count, urinalysis,

and assessment of hepatic and renal function were performed. The

children were given the opportunity to become familiar with the test

procedures.

In addition to the medication restrictions noted previously, before

the levocetirizine dose and for 28 hours afterward, study participants

refrained from ingesting methylxanthine-containing substances (eg,

cola, chocolate, or cocoa). After an 8- to 10-hour overnight fast, at

8 AM, a single dose of levocetirizine was administered as a 5-mg

tablet, followed by 150 mL of water. For the first 1.5 to 2 hours after

dosing, only clear juice or water was permitted.

EMLA local anesthetic cream (Astra, Mississauga, ON, Canada)

was applied to potential venipuncture sites. An indwelling intrave-

nous catheter (Critikon, Tampa, FL) was inserted, and 2.5-mL blood

samples were obtained before dosing and at 0.5, 1, 2, 3, 4, 6, 8, 10, 24,

26, and 28 hours afterward. The first 1 mL of blood was discarded.

After each sample was obtained, the catheter was rinsed with 1.5 mL

of 0.9% saline. Blood samples were centrifuged at room temperature

at 3700 rpm for 10 minutes. The plasma was transferred to

polypropylene tubes, which were sealed and frozen at 220C until

measurement of levocetirizine concentrations was performed.22

After each blood sample was collected at the times stated above,

peripheral H1-antihistaminic activity was evaluated by one investi-

gator who performed epicutaneous tests with histamine phosphate,

1 mg/mL, on the volar surfaces of the forearms by using sterilized

disposable straight needles (Coates&Clark, Greer, SC) and the prick-

through drop technique. All skin tests were performed in duplicate.

A different site on the volar surfaces of the forearmswas used for each

skin test. The sequence of test sites was identical in all children.

Analytic methods

Levocetirizine concentrations were determined in plasma samples

by using chiral HPLCwith tandemmass spectrometric detection after

online processing through the column-switching method.22 Quality

control samples at 7.5, 150, and 750 ng/mLwere assayed in duplicate

with each batch of clinical samples. Between-run accuracy and

precision were better than 10% throughout the range. The lower limit

of quantification for the assay was 12 ng/mL.

Wheal-and-flare circumferences were traced with a pen at

10 minutes and transferred to paper by using transparent tape. The

tracings were scanned, and the areas were calculated with Sigma-

Scan (Jandel Scientific, San Rafael, Calif). By using this system, with

wheal-and-flare sizes ranging from 0.05 to 5.0 cm2 and a sample size

of 14 children, differences of 20% could be detected with a 95% level

of confidence.

Data analysis

Pharmacokinetics. The pharmacokinetic parameters were calcu-

lated by using the noncompartmental analysis approach. The elimina-

tion rate constant (Ke) was calculated from the plasma levocetirizine

concentration (C) versus time (t) data measured after Cmax had

occurred, within 0.5 to 2 hours after dosing, by using equation 1:

C ¼ Ce-Ket

where C is the plasma concentration extrapolated to zero time after

application of equation 1 by using WIN-NONLIN (Scientific

Consulting, Apex, NC). The elimination half-life (t1/2) was calcu-

lated by using equation 2:

t1=2 ¼ ln 2=Ke

Although levocetirizine appears to be well absorbed, there is no

intravenous formulation, and therefore the absolute bioavailability

(F) is unknown. Total body clearance (Cl) and apparent volume of

distribution (Vd) were calculated as Cl/F and Vd/F, as shown in

equation 3:

Cl=F ¼ AUC=Dose

and equation 4:

Vd=F ¼ Cl=F

Ke

where AUC is the area under the plasma levocetirizine concentration

versus time curve from time zero to 28 hours.

Pharmacodynamics. The pharmacodynamic parameters maxi-

mum effect attributable to medication (Emax) and plasma concen-

tration producing 50% of Emax (EC50) were calculated by using

WIN-NONLIN (Scientific Consulting) and equation 5:


AUGUST 2005

356 Simons and Simons

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

E ¼ Emax C

EC50 1C

where E is the clinical effect, percentage suppression of histamine-

induced wheal or flare, and C is the levocetirizine plasma concen-

tration at which E occurs.30

Statistical analysis. Absolute wheal-and-flare areas over time

were analyzed by using 1-way ANOVA, with subject and time as

variates, analysis of covariance with predose wheal-or-flare areas as

the covariates, and the Tukey and Bonferroni multiple-range tests.

Differences were considered to be significant at P values of less than

or equal to .05.

RESULTS

The 14 Caucasian children (9 boys) with mild allergicrhinitis enrolled in the study had a mean (6 SEM) age of8.6 6 0.4 years, a mean weight of 30.4 6 2.2 kg, amean height of 132.5 6 3.3 cm, and a mean body massindex of 16.9 6 0.6. They received a single 5-mglevocetirizine dose, which was equivalent to a mean doseof 0.18 6 0.01 mg/kg (Table I). Complete pharmacoki-netic data were available on only 13 of the 14 childrenbecause of missing blood samples in 1 child. The meanplasma levocetirizine concentration versus time plot isshown in Fig 2. The pharmacokinetic parameters, in-cluding elimination rate constants, area under the plasmaconcentration versus time curve, oral clearance, andapparent volume of distribution, were calculated byusing noncompartmental analysis.

The mean maximum plasma levocetirizine concen-tration of 450 6 37 ng/mL occurred at a mean time of1.26 0.2 hours (Table II). The mean terminal eliminationhalf-life was 5.7 6 0.2 hours, the mean area under theplasma levocetirizine concentration versus time plot was3549 6 342 ng/mL/h, the mean oral clearance rate was

0.826 0.05mL/min/kg, and the mean apparent volume ofdistribution was 0.4 6 0.02 L/kg. The mean residencetime was 6.8 6 0.3 hours.

Pharmacodynamic data were available on all 14 chil-dren. Wheal-and-flare areas after testing with histaminephosphate, 1 mg/mL, are shown in Fig 3. Comparedwith predose values, the wheals were significantly sup-pressed (P < .05) from 1 to 28 hours, inclusive, withthe mean maximum suppression of 97% 6 1% occurringfrom 2 to 10 hours. Compared with predose values, theflares were significantly suppressed (P < .05) from 1 to 28hours, inclusive, with the mean maximum suppression of93%6 1%occurring from 2 to 24 hours. The relationshipsbetween the plasma levocetirizine concentrations andpercentage suppression of wheal-and-flare responsescompared with predose values versus time are shown inFig 4.

Pharmacodynamic analysis resulted in calculation of anEC50 estimate of 16.1 6 2.2 ng/mL and an Emax estimateof 104.3%6 2.6% for wheal suppression, and an EC50 of1.4 6 0.1 ng/mL and an Emax of 94.5% 6 0.4% for flaresuppression.

FIG 1. Structure of levocetirizine. The asterisk indicates the position of the chiral center. The molecular weight

is 461.8 g/mole. The formula is C21H25CIN2O3.2HCl.

TABLE I. Demographics

N = 14* (9 boys)

Age: 8.6 6 0.4 y

Weight: 30.4 6 2.2 kg

Height: 132.5 6 3.3 cm

Body mass index: 16.9 6 0.6

Levocetirizine dose: 0.18 6 0.01 mg/kg

All values are presented as means 6 SEM.

*At the time of the study, 4 children were using an intranasal

glucocorticoid for mild persistent allergic rhinitis, and 2 were using an

orally inhaled glucocorticoid for mild persistent asthma.



Simons and Simons 357

Rhinitis,sinusitis,

and

ocu

lardisease

s

There were no serious adverse effects. Two childrenexperienced some sneezing, nasal congestion, and dis-charge, and 1 child had intermittent coughing. The nasalsymptoms were attributed to allergic rhinitis, and thecough was attributed to asthma; that is, the respiratorysymptoms were considered to be due to underlyingallergic diseases and to be unrelated to administration ofthe study drug. Two hours after dosing, one child hadnausea that was relieved by eating, and 23 hours afterdosing, another child had a sore stomach that was relievedby eating. These gastrointestinal symptoms were attrib-uted either to overnight fasting or to anxiety about testprocedures and were considered to be probably unrelatedto the study drug. One child was more tired than usual 6hours after the dose of the study drug, and 2 children weremore tired than usual 12 hours after dosing. Although thisfatigue might have been due to the intensity of theprocedures during the first 12 to 13 hours of the study, itwas considered to be possibly related to the study drug.

DISCUSSION

In this study levocetirizine appeared to be wellabsorbed, with peak plasma concentrations occurring at

about 1 hour. In the absence of an intravenous levocetir-izine formulation, true bioavailability cannot be deter-mined. On the basis of the mean levocetirizine terminalelimination half-life of 5.7 hours that was found, dosingevery 24 hours on a regular basis would be expected tolead to minimal or no levocetirizine accumulation inplasma. On the basis of significant wheal-and-flare sup-pression from 1 to 28 hours after dosing, levocetirizinewould be expected to have significant H1-antihistaminicactivity throughout the dosing interval.

In pharmacokinetic and pharmacodynamic studies ofH1-antihistamines, although outcome measures such asblood tests and skin tests are highly objective, they areinherently invasive, and the studies therefore presentunique challenges in children.2,4 Study designs do notusually involve a placebo control,7-20 not only because ofethical constraints and parental concerns about the use ofplacebo, but also because a potent H1-antihistaminesuppresses wheals and flares by up to 100%,5,6,12,13 thusmaking it difficult to maintain double-masked observa-tions and measurements.

The objective, standardized, histamine-induced wheal-and-flare bioassay is useful for studying the onset, amount,and duration of activity of H1-antihistamines. Skin testswith histamine relate to the suppression of wheals, aprimary symptom and sign in urticaria, and flares, whichare caused by an axon reflex and are thus related tohistamine indirectly rather than directly. Whether skin testsuppression correlates with events in the airways remainscontroversial31; however, it is noteworthy that in allergypractice worldwide, skin tests with allergen are performedin lieu of nasal and bronchial allergen challenges toascertain the relevance of allergens to allergic rhinitisand asthma symptoms, and in the clinical setting cutane-ous responses are assumed to reflect airways responses.

The pharmacokinetics and pharmacodynamics of lev-ocetirizine, reported here in children aged 6 to 11 years,differ slightly from those reported previously in adultswith a mean age of 35 6 2 years and a mean weight of

FIG 2. Plasma levocetirizine concentration (mean 1 SEM) versus time plot after ingestion of levocetirizine,

5 mg.

TABLE II. Levocetirizine pharmacokinetics

Cmax (ng/mL) 450 6 37

tmax (h) 1.2 6 0.2

t1/2 (h) 5.7 6 0.2

AUC (ng/mL/h) 3549 6 342

Cl/F (mL/min/kg) 0.82 6 0.05

Vd/F (L/kg) 0.4 6 0.02

Values are presented as means 6 SEM.

Cmax, Maximum plasma concentration; tmax, time of maximum plasma

concentration; t1/2, terminal elimination half-life; AUC, area under the

plasma concentration versus time plot; Cl, oral clearance; F, oral

bioavailability; Vd, apparent volume of distribution.


AUGUST 2005


Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

FIG 3. The effect of levocetirizine, 5 mg, on the wheals and flares produced by epicutaneous tests with

histamine phosphate, 1 mg/mL. A, The wheals (1 SEM) were suppressed from 1 to 28 hours, inclusive

(P .05). B, The flares (1 SEM) were suppressed from 1 to 28 hours, inclusive (P .05).

FIG 4. Mean plasma levocetirizine concentrations and mean wheal-and-flare percentage suppression

compared with predose values, plotted versus time.




Rhinitis,sinusitis,

and

ocu

lardisease

s

67.3 6 2.3 kg, in whom the time of maximum plasmaconcentration is 0.736 0.7 hours, the terminal eliminationhalf-life value is 7.8 6 0.3 hours, the clearance rate is0.62 6 0.02 mL/min/kg, and the apparent volume ofdistribution is 0.416 0.02 L/kg.24,25 As noted previously,after administration of radioactively labeled levocetirizineto adults, 85.4% of the drug is eliminated unchanged in theurine, and 12.9% is eliminated unchanged in the feceswithin 1 week.26 The duration of action of a singlelevocetirizine dose is greater than 24 hours in adults.27

The pharmacokinetics and pharmacodynamics of lev-ocetirizine reported here in children aged 6 to 11 yearsalso differ from those reported previously in very youngchildren. In a prospective study in children with a meanage of 20.7 6 3.7 months and a mean weight of 11.6 6

1.8 kg, the time of maximum plasma concentration was1 hour, the terminal elimination half-life was 4.1 6 0.67hours, the clearance was 1.056 0.10 mL/min/kg, and theapparent volume of distribution was 0.37 6 0.06 L/kg.20

Rapid elimination of levocetirizine was also found in apopulation pharmacokinetic study in which cetirizine wasgiven to 343 children aged 14 to 46 months, and timedsparse blood samples were obtained at steady statefor measurement of plasma levocetirizine (the activeenantiomer or eutomer) and dextrocetirizine (the inactiveenantiomer or distomer) values.22,23 The population phar-macokinetic model used predicted that with increasingbody weight, levocetirizine oral clearance would increaseby 0.044 L/h/kg, and levocetirizine volume of distributionwould increase by 0.639 L/kg. Taken together, the resultsof these 2 studies indicate that in very young children,compared with older children and adults, higher levoce-tirizine doses may be needed on a milligram per kilogrambasis, and twice-daily dosing may be required.

Development of organ function and many of thematurational changes affecting pharmacokinetic disposi-tion of drugs is ongoing throughout infancy and child-hood, but elimination through the renal route is largelycompleted by age 4 to 5 years.1 This study provides arationale for administration of levocetirizine, 5 mg, oncedaily in children aged 6 to 11 years, as in adults, with theexpectation of prompt onset of action and significant long-lasting antihistaminic activity at the H1-receptor.

We thank Dr Marc de Longueville and Dr Eugene Baltes for their

collaboration.We also thankDrNestorCisneros;GailBonin, RN; and

especially Sandra S. Goritz, RN, for their contributions to this study.

REFERENCES

1. Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL, Leeder

JS, Kauffman RE. Developmental pharmacology—drug disposition,

action and therapy in infants and children. N Engl J Med 2003;349:

1157-67.

2. Steinbrook R. Testing medications in children. N Engl J Med 2002;347:

1462-70.

3. Simons FER. Advances in H1-antihistamines. N Engl J Med 2004;351:

2203-17.

4. Simons FER. H1-antihistamines in children. In: Simons FER, editor.

Histamine and H1-antihistamines in allergic disease. 2nd Ed. New York:

Marcel Dekker, Inc; 2002. p. 437-64.

5. Simons FER, Johnston L, Simons KJ. Clinical pharmacology of the

H1-receptor antagonists cetirizine and loratadine in children. Pediatr


6. Simons FER, Semus MJ, Goritz SS, Simons KJ. H1-antihistaminic

activity of cetirizine and fexofenadine in allergic children. Pediatr


7. Simons FER, Luciuk GH, Simons KJ. Pharmacokinetics and efficacy of

chlorpheniramine in children. J Allergy Clin Immunol 1982;69:376-81.

8. Simons FER, Simons KJ, Becker AB, Haydey RP. Pharmacokinetics and

antipruritic effects of hydroxyzine in children with atopic dermatitis.

J Pediatr 1984;104:123-7.

9. Simons KJ, Watson WTA, Martin TJ, Chen XY, Simons FER. Diphen-

hydramine: pharmacokinetics and pharmacodynamics in elderly adults,

young adults, and children. J Clin Pharmacol 1990;30:665-71.

10. Simons FER, Roberts JR, Gu X, Kapur S, Simons KJ. The clinical

pharmacology of brompheniramine in children. J Allergy Clin Immunol

1999;103:223-6.

11. Schmidt-Redemann B, Brenneisen P, Schmidt-Redemann W, Gonda S.

The determination of pharmacokinetic parameters of ketotifen in steady

state in young children. Int J Clin Pharmacol Ther Toxicol 1986;24:496-8.

12. Watson WTA, Simons KJ, Chen XY, Simons FER. Cetirizine: a phar-

macokinetic and pharmacodynamic evaluation in children with seasonal

allergic rhinitis. J Allergy Clin Immunol 1989;84:457-64.

13. Simons FER, Watson WTA, Simons KJ. Pharmacokinetics and pharma-

codynamics of ebastine in children. J Pediatr 1993;122:641-6.

14. Simons FER, Bergman JN, Watson WTA, Simons KJ. The clinical

pharmacology of fexofenadine in children. J Allergy Clin Immunol 1996;

98:1062-4.

15. Lin CC, Radwanski E, Affrime MB, Cayen MN. Pharmacokinetics of

loratadine in pediatric subjects. Am J Ther 1995;2:504-8.

16. Salmun LM, Herron JM, Banfield C, Padhi D, Lorber R, Affrime MB.

The pharmacokinetics, electrocardiographic effects, and tolerability of

loratadine syrup in children aged 2 to 5 years. Clin Ther 2000;22:613-21.

17. Desager JP, Dab I, Horsmans Y, Harvengt C. A pharmacokinetic

evaluation of the second-generation H1-receptor antagonist cetirizine in

very young children. Clin Pharmacol Ther 1993;53:431-5.

18. Pariente-Khayat A, Rey E, Dubois MC, Vauzelle-Kervroedan F, Pons G,

D’Athis P, et al. Pharmacokinetics of cetirizine in 2- to 6-year-old

children. Int J Clin Pharmacol Ther 1995;33:340-4.

19. Spicak V, Dab I, Hulhoven R, Desager J-P, Klanova M, de Longueville

M, et al. Pharmacokinetics and pharmacodynamics of cetirizine in infants

and toddlers. Clin Pharmacol Ther 1997;61:325-30.

20. Cranswick NE, Fuchs M, Turzikova J. Efficacy, safety and pharmaco-

kinetics of levocetirizine in allergic children age 1-2 years. Clin

Pharmacol 2004;75:P46.

21. Mentre F, Dubruc C, Thenot J-P. Population pharmacokinetic analysis

and optimization of the experimental design for mizolastine solution in

children. J Pharmacokinet Pharmacodyn 2001;28:299-319.

22. Hussein Z, Pitsiu M, Majid O, Aarons L, de Longueville M, Stockis A,

et al. Retrospective population pharmacokinetics of levocetirizine in

atopic children receiving cetirizine: the ETAC study. Br J Clin

Pharmacol 2005;59:28-37.

23. Simons FER, on behalf of the ETAC Study Group. Population pharma-

cokinetics of levocetirizine in very young children: the pediatricians’

perspective. Pediatr Allergy Immunol 2005;16:97-103.

24. Baltes E, Coupez R, Giezek H, Voss G, Meyerhoff C, Strolin Benedetti M.

Absorption and disposition of levocetirizine, the eutomer of cetirizine,

administered alone or as cetirizine to healthy volunteers. Fundam Clin

Pharmacol 2001;15:269-77.

25. Tillement JP, Testa B, Bree F. Compared pharmacological character-

istics in humans of racemic cetirizine and levocetirizine, two histamine

H1-receptor antagonists. Biochem Pharmacol 2003;66:1123-6.

26. Benedetti MS, Plisnier M, Kaise J, Maier L, Baltes E, Arendt C, et al.

Absorption, distribution, metabolism and excretion of [14C]levocetiri-

zine, the R enantiomer of cetirizine, in healthy volunteers. Eur J Clin

Pharmacol 2001;57:571-82.

27. Devalia JL, De Vos C, Hanotte F, Baltes E. A randomized, double-blind,

crossover comparison among cetirizine, levocetirizine, and ucb 28557 on

histamine-induced cutaneous responses in healthy adult volunteers.

Allergy 2001;56:50-7.

28. Bachert C, Bousquet J, Canonica GW, Durham SR, Klimek L, Mullol J,

et al. Levocetirizine improves quality of life and reduces costs in


AUGUST 2005


Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

long-term management of persistent allergic rhinitis. J Allergy Clin

Immunol 2004;114:838-44.

29. Potter PC, for the Pediatric Levocetirizine Study Group. Efficacy and

safety of levocetirizine on symptoms and health-related quality of life of

children with perennial allergic rhinitis. Ann Allergy Asthma Immunol.

In press 2005.

30. Holford NHG, Sheiner LB. Understanding the dose-effect relationship:

clinical application of pharmacokinetic-pharmacodynamic models. Clin

Pharmacokinet 1981;6:429-53.

31. Simons FER. Antihistamines. In: Adkinson NF Jr, Yunginger JW, Busse

WW, Bochner BS, Holgate ST, Simons FER, editors. Middleton’s allergy:

principles and practice. 6th ed. St Louis: Mosby, Inc; 2003. p. 834-69.




Rhinitis,sinusitis,

and

ocu

lardisease

s

Striking deposition of toxic eosinophilmajor basic protein in mucus: Implicationsfor chronic rhinosinusitis

Jens U. Ponikau,MD,a David A. Sherris, MD,b Gail M. Kephart, BS,c Eugene B. Kern, MD,b

David J. Congdon, MD,a Cheryl R. Adolphson, MS,c Margaret J. Springett, BS,d

Gerald J. Gleich, MD,e and Hirohito Kita, MDc Rochester, Minn, Buffalo, NY,

and Salt Lake City, Utah

Background: The mechanisms by which eosinophilic

inflammation damages the epithelium and contributes to

recurrent acute exacerbations in chronic rhinosinusitis (CRS)

have not been fully elucidated.

Objective: We tested the hypotheses that eosinophils deposit

toxic major basic protein (MBP) in the mucus and that MBP

reaches concentrations able to damage the sinonasal

epithelium.

Methods: Tissue specimens with mucus attached to the tissue

were carefully collected from 22 patients with CRS and

examined by using immunofluorescence staining for MBP. This

immunofluorescence was digitally analyzed to determine the

area covered by MBP and the intensity of the staining

(estimating MBP concentration). Levels of MBP in extracts

from nasal mucus were quantitated by means of RIA.

Results: Heterogeneous eosinophilia was evident within tissue

and mucus specimens. All tissue specimens showed intact

eosinophils, but diffuse extracellular MBP deposition, as a

marker of eosinophil degranulation, was rare. In contrast, all

mucus specimens showed diffuse MBP throughout and

abundant diffuse extracellular MBP deposition within clusters

of eosinophils. Digitized analyses of MBP immunofluorescence

revealed increased area coverage (P < .0001) in mucus

compared with that seen in tissue. Estimated concentrations

of MBP within the clusters suggested toxic levels. MBP

concentrations in mucus extract reached 11.7 mg/mL; MBP

was not detectable in healthy control subjects.

Conclusion: In patients with CRS, eosinophils form clusters

in the mucus where they release MBP, which is diffusely

deposited on the epithelium, a process not observed in the

tissue. Estimated MBP levels far exceed those needed to

damage epithelium from the luminal side and could predispose

patients with CRS to secondary bacterial infections. (J Allergy

Clin Immunol 2005;116:362-9.)

Key words: Eosinophils, chronic rhinosinusitis, mucus, degranula-tion, major basic protein

A recent survey by the National Center for HealthStatistics reported that 14.2% (29.2 million patients) ofthe US adult population recalled a health professional’sdiagnosis of sinusitis.1 Rhinosinusitis is now preferred tothe previous term sinusitis ‘‘. because sinusitis is almostalways accompanied by concurrent nasal airway inflam-mation. .’’2 The economic effect of chronic rhinosinus-itis (CRS) is huge; in the US the direct cost was estimatedin 1996 at $5.6 billion per year, and the indirect cost wasestimated as more than 70 million lost activity days peryear.3 Patients with CRS have long-term nasal congestion,thick mucus production, loss of sense of smell, andintermittent acute exacerbations secondary to bacterialinfections; they also experience severe quality-of-lifeimpairment.2,4 As an additional burden, CRS lacks aplausible cause. To date, the US Food and DrugAdministration has not approved any drug or treatmentfor CRS; no medical intervention has ever been efficaciousin a controlled clinical trial.

CRS is an inflammatory disease of the nasal andparanasal mucosa with persistent symptoms for longerthan 3 months; its ultimate end stage is inflammatorymucosal thickening and, in a subset of patients, polypoidchanges.2,5 The histologic hallmark of CRS is persistentunderlying eosinophilic inflammation.5-7 Eosinophil gran-ules contain several cytotoxic proteins,8 and eosinophilgranule major basic protein (MBP) is directly toxic toextracellular microorganisms as well as host tissue,including respiratory mucosa.9 CRS specimens showepithelial damage that is colocalized with MBP deposi-tion.6,7,10 In vitro, MBP directly damages respiratoryand sinus epithelium in a time- and dose-dependentmanner.11,12

Recent histologic analyses of CRS specimens sug-gested that intact eosinophils migrate from the tissue intothe mucus to form distinct and characteristic clusters.13,14

Therefore we tested whether eosinophils release MBP inthe mucus, but not in the tissue, and whether MBP reachesconcentrations capable of damaging the sinonasal epithe-lium in patients with CRS.

From athe Department of Otorhinolaryngology–Head and Neck Surgery, cthe

Department of Internal Medicine, Division of Allergic Diseases, and dthe

Department of Biochemistry and Molecular Biology, Mayo Clinic

Rochester; bthe Department of Otorhinolaryngology, University at

Buffalo, The State University of New York; and ethe Departments of

Dermatology and Medicine, University of Utah, Salt Lake City.

Supported by grants from the National Institutes of Health (AI 49235,

AI 09728) and from the Mayo Foundation.

Received for publication May 7, 2004; revised March 4, 2005; accepted for

publication March 31, 2005.


Reprint requests: Jens Uwe Ponikau, MD, Department of Otorhinolaryngol-

ogy, Mayo Clinic Rochester, 200 First St SW, Rochester, MN 55905.


0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.049

362

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s


Abbreviations used

CRS: Chronic rhinosinusitis

MBP: Major basic protein

METHODS

Patient selection

The diagnostic guidelines and criteria for CRS were consistent

with those adopted at the recent Rhinosinusitis Consensus

Conference.2 All patients had symptoms consistent with CRS for

longer than 3 months, inflamed mucosa on endoscopy, and a coronal

computed tomographic scan demonstrating mucosal thickening of

greater than 5 mm in more than 2 sinuses. Retention cysts and cystic

fibrosis are differential diagnoses to CRS, and if these diseases were

diagnosed, those patients were excluded from the study. Because

complete immunologic evaluations were not performed in all pa-

tients, we have not excluded patients (if any) with immunodeficien-

cies. With regard to noneosinophilic inflammatory sinusitis, we found

eosinophilia in tissues from all patients of our otherwise unselected

patient population. Patients were not preselected for having eosino-

philic CRS.

Histologic analyses of specimens

For the histologic analyses, specimens were collected from 22

consecutive patients with CRS undergoing endoscopic sinus surgery.

During surgical intervention, we used Blakesley surgical forceps to

carefully and gently collect the maximum amounts of tissue and

mucus, and we ensured that the mucus remained attached to the tissue,

which was immediately fixed in formalin. Four specimens from the

ethmoid sinuses of healthy individuals (nonallergic and no asthma)

undergoing septoplasty procedures served as negative controls. The

Institutional Review Board of the Mayo Clinic approved the study.

Paraffin-embedded tissue blocks with attached mucus were cut in

5-mm-thick serial sections, mounted on positively charged slides, and

stained with the following: (1) hematoxylin and eosin; (2) antibody to

eosinophil MBP using rabbit antihuman MBP15-17; (3) antibody to

neutrophil elastase10 using rabbit antihuman elastase (IgG fraction;

Cortex Biochem, San Leandro, Calif); and (4) negative control for

MBP and elastase (normal rabbit IgG). All specimens were incubated

in 10% normal goat serum to block nonspecific binding by the

second-stage antibody and in 1% chromotrope 2R to block non-

specific binding of fluorescein dye to the eosinophils.6 Fluorescein

isothiocyanate–conjugated goat anti-rabbit IgG was used as the

secondary antibody.15-17 MBP was chosen to assess eosinophil

infiltration and degranulation because it is the predominant eosino-

phil granule protein. It accounts for roughly 50% of the total protein

mass in the eosinophil granule,18 and the release of MBP is highly

correlated with the release of the other granule proteins.19 Elastase,

the predominant neutrophil granule protein, was studied to assess

neutrophil infiltration and degranulation.

Semiquantitative analysis of MBP andelastase immunofluorescence in tissueand mucus

Three examiners independently examined the entire specimen. To

exclude artifacts from trauma caused through the removal of the

specimen during surgery, only areas of untouched mucosa covered

with mucus were evaluated. In contrast, areas with obvious tears or

influx of red blood cells, indicating trauma with bleeding in the areas

touched by the forceps, were excluded.

The tissue and the attached mucus were each evaluated with the

following scoring system, which was previously used to describe

eosinophil7 and neutrophil10 infiltration and degranulation. Because

of the heterogeneity of the eosinophilic and neutrophilic inflamma-

tion, only those areas of tissue and attached mucus with the most

prominent cellular infiltrate for these leukocytes were scored in this

semiquantitative scheme. For each specimen, we calculated the

means of the examiners’ scores (0, not present; 1, few present/

scattered; 2, many present/abundant) for tissue and for attached

mucus using the criteria listed below:

d intact eosinophils or neutrophils;

d punctate staining (MBP or elastase within intact extracellular

granules); and

d diffuse staining (extracellular MBP or elastase not in granules).

In addition, eosinophils forming clusters within the mucus

(eosinophilic mucin) were noted as present (1) or absent (2), and

diffuse MBP staining within the clusters was noted as present (1) or

absent (2). Neutrophil clusters and elastase deposition were evalu-

ated similarly.

Digital analyses of MBP and elastaseimmunofluorescence in tissue and mucus

Computer analysis of sections stained for MBP or elastase by

means of immunofluorescence was performed to evaluate objectively

the overall areas covered by MBP or elastase, as well as the intensities

of the staining (as indirect markers for MBP or elastase concen-

trations).6 Briefly, we used a confocal microscope (LSM510

Confocal Microscope; Carl Zeiss, Inc, Oberkochen, Germany) to

survey and select both the least and most intense areas of MBP or

elastase staining in the tissue and the mucus. First, digital images

(512 3 512 pixels, 4003 magnification, 488-nm excitation wave-

length) of areas that showed maximal fluorescence staining (most

intense accumulation of either eosinophils or neutrophils or diffuse

extracellular MBP or elastase deposition) were obtained. Second,

digital images of the corresponding areas on the serial section, which

was stained with normal rabbit IgG, as the negative control, were

recorded. Third, by using image-analysis software (KS400 Image

Analysis System, Carl Zeiss, Inc), the threshold for each negative

control image was calibrated to a baseline value that showed no

positive pixels. Fourth, this background threshold was used to analyze

the corresponding area on the MBP or elastase immunofluorescence-

stained specimen. Any pixels recorded were quantitated as a percent-

age of an area (512 3 512 pixels) positive for MBP or elastase. The

image-analysis software then compared the different areas within the

specimen and determined the area with the highest percentage of

positive pixels; this percentage indirectly indicated the area of

maximal inflammatory eosinophilic or neutrophilic infiltrate (tissue)

or maximal MBP or elastase deposition (mucus). A similar survey was

made to find the area in the MBP or elastase immunofluorescence

specimen with the least fluorescence in the tissue and in the mucus;

once located, these areas were compared with the respective

corresponding areas in the negative control serial section and analyzed

as above.

Electron microscopy and immunogoldlabeling for MBP

We also used electron microscopy to investigate the morphology

of eosinophils and eosinophil granules and to localize MBP in tissues

from 3 patients with CRS, as described earlier.20 After the primary

antibody, affinity-purified antibody to MBP, we used a secondary

antibody conjugated to 15 nm colloidal gold particles.17



Ponikau et al 363

Rhinitis,sinusitis,

and

ocu

lardisease

s

Quantitative analysis of MBP in mucus

Nasal mucus was collected by means of direct trap suctioning

from the nasal cavity (middle meatus region) and from one maxillary

sinus under endoscopic guidance with a sinus secretion collector

(Xomed Surgical Products, Jacksonville, Fla) from 12 additional

patients with CRS who met the same diagnostic criteria as described

above and from 9 healthy control subjects. The mass of each mucus

specimen was obtained, and a 3-fold excess of normal saline (0.15 M

NaCl) was added. After vigorous vortexing for 10 seconds (33), the

mucus suspension was centrifuged, and the resulting supernatant

fluid was frozen at 270C. (This vortexing step probably did not

release the intracellular intragranular MBP, which requires more

severe conditions. Specifically, to extract MBP, eosinophils need to

be incubated for 30 minutes at room temperature in 0.5% NP-40

[Sigma-Aldrich, St Louis, Mo] and 0.01 M HCl8). Finally, MBP

levels in the mucus supernatant fluid were determined by means of

RIA, essentially as described earlier.21


The groups were compared with a 2-sided Student t test or the

Wilcoxon matched-pairs signed-rank test, and a P value of less than

.05 was considered significant. Data are presented as means (6 SD)

or medians (ranges).

RESULTS

Patient demographics

Patient demographics have been previously described.6

Briefly, the mean age of the 22 patients with CRS was47 years (range, 16-86 years); 11 were women; the meannumber of sinus operations was 1.8 (range, 0-7); the meanduration of disease was 8.6 years (range, 2-27 years); theincidence of aspirin idiosyncrasy was 41%; 11 hadincreased serum levels of total IgE (>128 U/mL, 2 SDsabove the mean value of healthy adult control subjects);and 10 were considered allergic, as defined by a positiveskin prick test response to at least one allergen from apanel of 16 common aeroallergens. The incidence ofphysician-diagnosed asthma was 68% (15/22); the other7 patients underwent a methacholine challenge, and 5 hadpositive responses.

Eosinophil and neutrophil infiltration anddegranulation in tissue and mucus

Although intact eosinophils were abundant in numer-ous areas in the tissue (Fig 1, A and B), the most strikingobservation was the abundance of diffuse MBP staining(not in cells and not in granules) in the mucus comparedwith its absence in the tissue (Fig 1, A-D). Eosinophilsformed cell clusters in the mucus, and diffuse MBPstaining was observed within and around these clusters(Fig 1, A-D and F). Punctate staining, indicating intactextracellular eosinophil granules, was frequently observedin both tissue and mucus (Fig 1, E and F). Compared withMBP staining, only isolated areas in tissue and mucusstained for neutrophil elastase (results not shown).

As summarized in Fig 2 (top), tissue in all specimensshowed abundant intact eosinophils (22/22) comparedwith mucus (11/22). The numbers of patient specimens

showing various amounts of punctate staining were sim-ilar between tissue and mucus. Diffuse MBP deposition inthe mucus was abundant in all (22/22) CRS specimens butwas not observed in the tissue, except in one small areafrom 1 specimen (1/22). Intact neutrophils were evident inboth tissue (10/22) and mucus (14/22) specimens. Incontrast to the numerous areas showing diffuse MBPdeposition in the mucus of all 22 specimens (Fig 2, A),isolated areas of diffuse elastase deposition were noted inthe mucus of only 9 of 22 patients. The majority ofspecimens showed a virtual absence of diffuse elastase inthe mucus (Fig 2, B).

Eosinophils forming clusters within the mucus (eosin-ophilic mucin) were present in 22 of 22 specimens, anddiffuse deposition of MBP in and around these clusterswas seen in 22 of 22 specimens. In contrast, neutrophilcluster formation was observed in only 5 of 22 specimens,and elastase deposition in these clusters was observed inonly 4 of 22 specimens. None of the specimens fromhealthy control subjects (0/4) were positive for intacteosinophils or neutrophils or punctate or diffuse stainingfor MBP or elastase (results not shown).

Digital analyses of MBP and elastaseimmunofluorescence

In the tissue the maximum area positive for MBPimmunofluorescence staining had a median of 18.35%(range, 0.64% to 56.63%); by contrast, the maximum areapositive for elastase immunofluorescence was signifi-cantly less (median, 3.11%; range, 0.05% to 46.2%;P < .002; Fig 3). In the mucus the maximum area positivefor MBP immunofluorescence staining had a median of93.28% (range, 3.62% to 100%); by contrast, the maxi-mum area positive for elastase immunofluorescence wasalso significantly less (median, 29.30%; range, 0.10% to85.42%; P< .001; Fig 3). Furthermore, the maximum areapositive for MBP immunofluorescence in the mucus(median, 93.28%) was significantly increased comparedwith the maximum mean area in tissue (median, 18.35%;P < .0001).

Electron microscopy and immunogoldlabeling for MBP

In tissue from a patient with CRS, both an intacteosinophil containing the characteristic electron-densegranule cores with MBP and extracellular granules arevisible (Fig 4, A). However, the immunogold MBP stainshows MBP confined within the intact granules (Fig 4, B),corresponding with the punctate staining in MBP immu-nofluorescence (see Fig 1, E). Diffuse MBP immunogoldstaining is not seen in the tissue (Fig 4, B), which isconsistent with the lack of diffuse MBP immunofluores-cence staining seen in the tissue (see Fig 1, E).

Measurement of MBP in the mucus

As shown in Fig 5, the mean concentration of detectableMBP in mucus extracts from the maxillary sinuses inpatients with CRS was 4.2 mg/mL (6 3.1 mg/mL; range,


AUGUST 2005

364 Ponikau et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

FIG 1. Photomicrographs of CRS specimens stained for MBP by means of immunofluorescence or stained

with hematoxylin and eosin. Panel a demonstrates eosinophilic inflammation in tissue, eosinophil clusters

(black arrows) in mucus, subepithelial basement membrane thickening, and damaged epithelium (yellow

arrows) (hematoxylin and eosin counterstain of Panel b; original magnification, 1603). Panel b shows MBP

in tissue is contained within the cells or in intact granules (punctate staining) outside the cells. In mucus,

diffuse MBP staining is in eosinophil clusters (white arrows) and outside of clusters (anti-MBP; original

magnification, 1603). Panel c shows minimal tissue eosinophilia, massive eosinophilia in mucus, subepi-

thelial basement membrane thickening, and the damaged epithelium (yellow arrows) (hematoxylin and eosin;

original magnification, 4003). Panel d (serial section of Panel c) shows few intact eosinophils in tissue, intense

diffuse MBP deposition within the mucus, and MBP adjacent to the epithelial surface (anti-MBP; original

magnification, 4003). Panels e (tissue) and f (mucus) show intact eosinophils (white arrows) and free granules

(punctate staining, blue arrows); diffuse extracellular MBP staining (orange arrows) appears unique to mucus

(anti-MBP; original magnification, 14003). Serial sections stained with normal rabbit IgG were negative

(results not shown).



Ponikau et al 365

Rhinitis,sinusitis,

and

ocu

lardisease

s

0.3-11.7 mg/mL; n = 12). In 9 of 12 patients with CRS, wewere also able to harvest sufficient mucus from the nasalcavity to detect a mean MBP concentration of 4.1 mg/mL(6 2.6 mg/mL; range, 0-8.0 mg/mL), which did not differsignificantly from the mean concentration in the maxillarysinuses. In contrast, in mucus specimens from the nasalcavities of the healthy control subjects, no MBP could bedetected above the sensitivity of the assay (0.010 mg/mL).Thus even the lowest MBP concentration detected inmucus from the maxillary sinus of a patient with CRS(0.31 mg/mL) was at least 30-fold greater than that ofhealthy control subjects.

DISCUSSION

The underlying eosinophilic inflammation is in-creasingly recognized to play an important role in thepathogenesis of CRS, and its association with epithelialdamage has been suspected.6,7 How the eosinophilactually mediates the pathophysiology, such as damagingthe epithelium, remains unclear. Earlier studies in patientswith CRS used tissue biopsy specimens without mucus7,10

and showed MBP deposition within damaged sinus epi-

thelium; in contrast, our specimens were carefully col-lected to ensure that the mucus remained attached to theharvested tissue. Thus we could document the extent,localization, and degranulation pattern of the heteroge-neous eosinophilia and neutrophilia in both the tissue andthe mucus. Eosinophil, rather than neutrophil, inflamma-tion was predominant in both tissue and mucus. Strikingextracellular deposition of diffuse MBP (especially inclusters) was unique to the mucus in all 22 patients withCRS and was not found in the tissue (with the exception ofone small area in 1 patient). In contrast, extracellulardeposition of diffuse elastase was found in less than half ofthe patients (9/22) and only in isolated areas.

The pattern of eosinophil degranulation described inthis study could explain the sinonasal epithelial erosionobserved in patients with CRS.6 As seen in Fig 1, A and C,the outer layers of the epithelium are eroded away, but alayer of basal epithelial cells still remains. This observa-tion, as well as the marked deposition of toxic eosinophilgranule MBP in the mucus, suggests that the damage tothe epithelium occurs from the outside (luminal side). Thedamaged epithelium might provide an entry port for colo-nizing bacteria that are present in the sinuses of patientswith CRS, as well as healthy control subjects.22,23 Thus

FIG 2. Comparisons of eosinophilic and neutrophilic inflammation in tissue versus mucus. The graphs show

the mean MBP and elastase immunofluorescence scores for intact eosinophils and neutrophils, punctate

staining (MBP and elastase within intact extracellular granules), and diffuse staining (extracellular MBP and

elastase not in granules) in tissue and mucus from 22 patients with CRS. Each dot represents the mean score

of 3 independent examiners for each patient (scoring on vertical axis follows the grading system presented in

theMethods section). Panel a shows a CRS specimen stainedwith anti-MBP; note the abundant diffuseMBP in

mucus that is absent in tissue. Panel b shows a CRS specimen stained with anti-elastase; note the virtual

absence of diffuse elastase in both mucus and tissue (original magnification, 4003).


AUGUST 2005

366 Ponikau et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

the release of toxic eosinophil granule proteins, such asMBP, in the mucus could be a crucial predisposing factorfor the secondary bacterial infections that likely mediatethe acute exacerbations of CRS. In healthy controlsubjects, the absence of MBP in the mucus might alsoexplain the lack of epithelial damage and, consequently,the lack of bacterial infections, despite the presence ofbacteria.22,23 Taken together, these findings might explainthe complex clinical course of CRS with its occasionalbacterial exacerbations; indeed, the numbers of infiltrat-ing tissue neutrophils have been directly correlated to thenumbers of bacteria present.24 Overall, these exacerba-tions are superimposed on the persistent and underlyingeosinophilic inflammation uniformly seen in all patients.

Although we used an RIA for MBP in extracts frommucus specimens, this procedure likely underestimatedthe actual local concentrations for the following reasons.First, we attempted to extract MBP from the thick mucusthrough vortexing in saline, and thus only the MBP thatactually dissolved in the saline could be measured. Theremaining mucus and probably a large amount of undis-solved MBP had to be discarded. Second, the concentra-tion of MBP was based on the entire volume of the mucusspecimen; because portions of the mucus do not containMBP (Fig 3), our results underestimate the local maximalMBP concentrations.

To address this potential underestimation, we calcu-lated the approximate MBP concentration in an eosinophilto be 33 mg/mL or 2.1 3 1023 M (8.98 3 1029 mg [massMBP/eosinophil]/2.68 3 10210 mL [volume/eosino-phil]).8 Although the immunogenic epitopes might notbe equally available for MBP in mucus compared withMBP in the granules of intact cells, we used digital

analyses to compare the brightness of anti-MBP stainingin confocal microscopic images. Overall, 17 of 22 speci-mens showed brighter immunofluorescence staining forMBP in mucus compared with that seen in tissue; nobrightness differences were noted for intact cells in tissue,blood vessels, or mucus. Thus we estimate that areas of dif-fuse MBP in the mucus might exceed 33 mg of MBP/mL.Because MBP is toxic to and causes erosion of theepithelium at concentrations less than 10 mg/mL,12 ourMBP immunofluorescence results suggest that MBPreaches local concentrations in the mucus of patientswith CRS far exceeding those necessary to mediateepithelial damage. In addition, inspissation (dessication)of mucus at mucosal surfaces in vivomight lead to a furtherincrease in the local MBP concentration. We observedstriking MBP deposition directly adjacent to the damagedepithelium (Fig 1, A-D).

In the tissue, extracellular MBP is confined to intacteosinophil granules (ie, punctate staining), as shown bymeans of MBP immunofluorescence staining (Fig 1, B,D, and E) and by means of electron microscopy andimmunogold labeling for MBP (Fig 4). The apparent lackof subepithelial tissue damage, as assessed with hematox-ylin-and-eosin staining of specimens from patients withCRS, suggests that intact granules containing MBP mightnot have physiologic or damaging actions in the tissue.The accumulation of diffuse MBP immunofluorescence inthe mucus (and not in the tissue) suggests that the cellsfound in the tissue are in transit toward their final target inthe mucus, with ongoing deposition of MBP (ie, thecluster of eosinophils surrounded by the diffuse cloud ofMBP). This pattern of cluster-forming eosinophils appearsstrikingly similar to the eosinophils’ role in their immune

FIG 3. Comparison of digitized areas ofminimal andmaximalMBP or elastase staining in the tissue andmucus

of patients with CRS. To demonstrate the heterogenicity within the 22 specimens, these data points represent

the minimal and maximal percentages of area positive for MBP or elastase immunofluorescence. The

horizontal lines show the median values in each group.



Ponikau et al 367

Rhinitis,sinusitis,

and

ocu

lardisease

s

defense against parasites when they cluster around theorganisms, subsequently releasing granular proteins (in-cluding MBP) that destroy the parasite.25

It is generally assumed that eosinophil cytolysis andpiecemeal degranulation are distinct mechanisms bywhich granules and, subsequently, granule proteins arereleased in diseased airway tissue.26 Instead, we found that

eosinophils release cytotoxic MBP in the mucus, but not inthe tissue, at concentrations likely exceeding those neededto damage the epithelium in patients with CRS. Thereforeone might need to take not only the tissue but also themucus into account when trying to understand the patho-physiology of CRS and probably other airway diseases. Inaddition, this new understanding suggests a beneficial

FIG 4. Transmission electron micrographs of sinus tissue from a patient with CRS. Panel a shows the

characteristic electron-dense secondary granules within an intact cell (white arrows) and intact extracellular

granules in the tissue (black arrows; original magnification, 10,0003). Panel b shows immunogold labeling

(black dots) for MBP and demonstrates that MBP is localized within the intact granules; note the lack of MBP

labeling in the surrounding tissue (original magnification, 33,0003).

FIG 5. MBP concentrations in mucus specimens from patients with CRS and healthy control subjects. Mucus

specimens were extracted with 0.15 M NaCl, and MBP was measured in the supernatants by means of RIA.

MBP was detected in the maxillary sinus mucus and in the nasal cavity mucus of patients with CRS but not in

mucus from the healthy control subjects. Horizontal bars indicate mean values for each group.


AUGUST 2005

368 Ponikau et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

effect in therapies that target primarily the underlying andpresumably damage-inflicting eosinophilic inflammationinstead of the secondary bacterial infection.

REFERENCES

1. Lethbridge-Cejku M, Schiller JS, Bernadel L. Summary health statistics

for U.S. Adults; National Health Interview Survey, 2002. National

Center for Health Statistics. Vital Health Stat 2004;10:23.

2. Meltzer EO, Hamilos DL, Hadley JA, Lanza DC, Marple BF, Nicklas

RA, et al. Rhinosinusitis: establishing definitions for clinical research

and patient care. J Allergy Clin Immunol 2004;114(suppl):S155-212.

3. Ray NF, Baraniuk JN, Thamer M, Rhinehart CS, Gergen PJ, Kaliner M,

et al. Healthcare expenditures for sinusitis in 1996: contributions of

asthma, rhinitis, and other airway disorders. J Allergy Clin Immunol

1999;103:408-14.

4. Gliklich RE, Metson R. The health impact of chronic sinusitis in patients

seeking otolaryngologic care. Otolaryngol Head Neck Surg 1995;113:

104-9.

5. Kaliner MA, Osguthorpe JD, Fireman P, Anon J, Georgitis J, Davis JL.

Sinusitis: bench to bedside. Current findings, future directions. Otolar-

yngol Head Neck Surg 1997;116(suppl):S1-20.

6. Ponikau JU, Sherris DA, Kephart GM, Kern EB, Gaffey TA, Tarara JE,

et al. Features of airway remodeling and eosinophilic inflammation in

chronic rhinosinusitis: is the histopathology similar to asthma? J Allergy

Clin Immunol 2003;112:877-82.

7. Harlin SL, Ansel DG, Lane SR, Myers J, Kephart GM, Gleich GJ.

A clinical and pathologic study of chronic sinusitis: the role of the

eosinophil. J Allergy Clin Immunol 1988;31:867-75.

8. Abu-Ghazaleh RI, Dunnette SL, Loegering DA, Checkel JL, Kita H,

Thomas LL, et al. Eosinophil granule proteins in peripheral blood

granulocytes. J Leukoc Biol 1992;52:611-8.

9. Gleich GJ, Adolphson CR, Leiferman KM. The biology of the eosin-

ophilic leukocyte. Annu Rev Med 1993;44:85-101.

10. Fujisawa T, Kephart GM, Gray BH, Gleich GJ. The neutrophil and

chronic allergic inflammation: immunochemical localization of neutro-

phil elastase. Am Rev Respir Dis 1990;141:689-97.

11. Frigas E, Loegering DA, Gleich GJ. Cytotoxic effects of the guinea pig

eosinophil major basic protein on tracheal epithelium. Lab Invest 1980;

42:35-43.

12. Hisamatsu K, Ganbo T, Nakazawa T, Murakami Y, Gleich GJ,

Makiyama K, et al. Cytotoxicity of human eosinophil granule major basic

protein to human nasal sinus mucosa in vitro. J Allergy Clin Immunol

1990;86:52-63.

13. Ponikau JU, Sherris DA, Kern EB, Homburger HA, Frigas E, Gaffey TA,

et al. The diagnosis and incidence of allergic fungal sinusitis. Mayo Clin

Proc 1999;74:877-84.

14. Braun H, Busina W, Freudenschuss K, Beham A, Stammberger H.

‘‘Eosinophilic fungal rhinosinusitis’’: a common disorder in Europe?

Laryngoscope 2003;113:264-9.

15. Filley WV, Ackerman SJ, Gleich GJ. An immunofluorescent method for

specific staining of eosinophil granule major basic protein. J Immunol

Methods 1981;47:227-38.

16. Filley WV, Holley KE, Kephart GM, Gleich GJ. Identification by

immunofluorescence of eosinophil granule major basic protein in lung

tissues of patients with bronchial asthma. Lancet 1982;2:11-6.

17. Peters MS, Schroeter AL, Kephart GM, Gleich GJ. Localization of

eosinophil granule major basic protein in chronic urticaria. J Invest

Dermatol 1983;81:39-43.

18. Kita H, Adolphson CR, Gleich GJ. Biology of eosinophils. In:

Adkinson NF Jr, Bochner BS, Yunginger JW, Holgate ST, Busse WW,

Simons FE, editors. Middleton’s allergy: principles and practice. 6th ed.

Philadelphia: Mosby; 2003. p. 305-32.

19. Ott NL, Gleich GJ, Peterson EA, Fujisawa T, Sur S, Leiferman KM.

Assessment of eosinophil and neutrophil participation in atopic derma-

titis: comparison with the IgE-mediated late-phase reaction. J Allergy


20. Popken-Harris P, Checkel J, Loegering D, Madden B, Springett M,

Kephart G, et al. Regulation and processing of a precursor form of

eosinophil granule major basic protein (proMBP) in differentiating

eosinophils. Blood 1998;92:623-31.

21. Wagner JM, Bartemes K, Vernof KK, Dunnette S, Offord KP, Checkel

JL. Analysis of pregnancy-associated major basic protein levels through-

out gestation. Placenta 1993;14:671-81.

22. Kalcioglu MT, Durmaz B, Aktas E, Ozturano O, Durmaz R. Bacteriology

of chronic maxillary sinusitis and normal maxillary sinuses: using culture

and multiplex polymerase chain reaction. Am J Rhinol 2003;17:143-7.

23. Nadel DM, Lanza DC, Kennedy DW. Endoscopically guided cultures

in chronic sinusitis. Am J Rhinol 1998;12:233-41.

24. Dunnette SL, Hall MM, Washington JA 2nd, Kern EB, McDonald TJ,

Facer GW, et al. Microbiologic analyses of nasal polyp tissue. J Allergy


25. Kephart GM, Gleich GJ, Connor DH, Gibson DW, Ackerman SJ.

Deposition of eosinophil granule major basic protein onto microfilariae

of Onchocerca volvulus in the skin of patients treated with diethylcar-

bamazine. Lab Invest 1984;50:51-61.

26. Erjefalt JS, Persson CG. New aspects of degranulation and fates of

airway mucosal eosinophils. Am J Respir Crit Care Med 2000;161:

2074-8.



Ponikau et al 369

Rhinitis,sinusitis,

and

ocu

lardisease

s

Intranasal tolerance induction withpolypeptides derived from 3 noncross-reactive major aeroallergens preventsallergic polysensitization in mice

Karin Hufnagl, PhD,a Birgit Winkler, MD,a Margit Focke, PhD,a Rudolf Valenta, MD,a

Otto Scheiner, PhD,a Harald Renz, MD,b and Ursula Wiedermann, MD, PhDa Vienna,

Austria, and Marburg, Germany

Background: Specific immunotherapy is less effective in

patients with multiple allergic sensitizations compared with

monosensitized patients.

Objective: We therefore established a mouse model of

polysensitization to the major birch and timothy grass pollen

allergens to test whether allergic polysensitization can be

prevented by multiple allergen application via the mucosal

route.

Methods: Female BALB/c mice were immunized

intraperitoneally with recombinant (r) Bet v 1, rPhl p 1, and

rPhl p 5. For intranasal tolerance induction, a mixture of the

complete allergens was compared with allergen-derived

immunodominant peptides applied either as a mixture or as

a synthetic hybrid peptide composed of the T-cell epitopes

of the 3 allergens.

Results: Intranasal application of the mixture of the complete

allergen molecules did not prevent polysensitization to the same

allergens. In contrast, pretreatment with a mixture of the

immunodominant peptides or the hybrid peptide led to

significantly reduced allergen-specific IgE responses in sera,

IL-4 production in vitro, and suppressed airway inflammation.

TGF-b mRNA levels did not change, and IL-10 production was

significantly suppressed after the pretreatment. The fact that

the reduction of IL-10 was not abrogated after IL-10 receptor

neutralization and that tolerance was not transferable with

splenocytes indicates that the suppression of TH2 responses

in polysensitized mice might not be mediated by

immunosuppressive cytokines.

Conclusion: Our study demonstrates that it is possible to

suppress allergic immune responses simultaneously to several

clinical important allergens. Thus, mucosal coapplication of

selected peptides/hybrid peptides could be the basis of a

mucosal polyvalent vaccine to prevent multiple sensitivities in

atopic patients. (J Allergy Clin Immunol 2005;116:370-6.)

Key words: Animal model, type I allergy, polysensitization, mucosaltolerance, polypeptides, hybrid, IL-10, TGF-b

About 25% of the population in industrialized countrieshas type I allergy, a genetically determined immunolog-ical disorder. In Europe, the most common seasonalairborne allergens are derived from white birch (Betulaverrucosa) and timothy grass (Phleum pratense).1 Morethan 90% of the patients allergic to birch pollen (BP) reactto the major allergen Bet v 1, and in patients allergicto grass pollen, Phl p 1 and Phl p 5 represent majorallergens.2-4

There is substantial clinical evidence that many patientswith allergy are cosensitized to several unrelated airborneallergens. In this context, recent studies demonstrated thatBP allergy is frequently associated with sensitization toother inhalant allergens, including grass pollen, and thatmultiallergies often develop with increasing age.5,6

One of the most effective treatments for type I allergy,especially in young and monosensitized patients, is spe-cific immunotherapy, performed by repeated subcutane-ous injections of increasing amounts of allergen extracts.7

In patients with multiple sensitivities, specific immuno-therapy is known to be of low efficacy and is associatedwith an increased risk of anaphylactic side reactions.7,8

Thus, there is a clear necessity to improve or develop newtreatment strategies particularly for polysensitized indi-viduals.

Such improvements could be achieved by the use ofrecombinant allergens or allergen-derived peptides ac-cording to the patient’s sensitization profile,9-11 insteadof natural allergen extracts containing a variety ofallergenic molecules. Recently the possibility of usingprophylactic allergy vaccines has been discussed.12

Following the concept of vaccination against manyinfectious diseases, prophylactic treatment to most com-mon allergens could be a possible strategy for earlyallergy prevention. In addition, the change to a lessinvasive route of administration, such as application ofallergens via the mucosal surfaces, might even enhance

From athe Department of Specific Prophylaxis and Tropical Medicine, Center

for Physiology and Pathophysiology, Medical University of Vienna; andbthe Department of Clinical Chemistry andMolecular Diagnostics, Hospital

of the Philipps University Marburg.

Supported by grants from the Austrian Science Fund (F01814, P14634-PAT,

Y0784GEN, FWF T163).

Received for publication July 9, 2004; revised March 10, 2005; accepted for



Reprint requests: Ursula Wiedermann, MD, PhD, Department of Specific

Prophylaxis and Tropical Medicine, Center for Physiology and Patho-

physiology, Medical University of Vienna, Kinderspitalgasse 15, A-1095

Vienna, Austria. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.002

370

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s


Abbreviations usedaa: Amino acid

BP: Birch pollen

r: Recombinant

RBL: Rat basophil leukemia

the patient’s compliance to treatment. In several experi-mental animal studies, it has been shown that mucosaltolerance is highly effective in preventing allergic dis-ease.13-17 A recent study in mice demonstrated that theintranasal route of desensitization can be even moreeffective than the intradermal route.18 We previouslydemonstrated in a murine model of allergic sensitizationto BP that mucosal administration of recombinant (r) Bet v 1suppressed allergic sensitization and airway inflamma-tion in naive and in sensitized mice.16,17,19 In the currentstudy, we established a murine model of polysensitizationto the noncross-reactive major birch and grass pollenallergens Bet v 1, Phl p 1, and Phl p 5. By using this model,we tested whether it is possible to induce tolerancesimultaneously against these different pollen allergensby intranasal application of either the allergen proteins orthe allergen-derived peptides composed of the immuno-dominant epitopes thereof. Our results show that intra-nasal application of polypeptides successfully preventedallergic polysensitization.

METHODS

Animals

Female inbred 7-week-old BALB/c mice (n = 5 per group) were

obtained from Charles River (Sulzfeld, Germany). All experiments

were approved by the Animal Experimentation Committee of the

University of Vienna and the Federal Ministry of Education, Science

and Culture.

Recombinant allergens and naturalallergen extracts

rBet v 1, rPhl p 1, and rPhl p5 were obtained from Biomay AG

(Vienna, Austria). Birch pollen (B verrucosa) and timothy grass

pollen (P pratense) were purchased from Allergon (Valinge,

Sweden), and extracts were prepared as previously described.16

Epitope mapping studies

For T-cell epitope mapping, a panel of 50 peptides of the Bet v 1

molecule (Cambridge Research Biochemicals Limited, Cambridge,

United Kingdom), 77 peptides of Phl p 1 (Cambridge Research

Biochemicals Limited), and 92 peptides of Phl p 5 (provided by

Dr Helmut Fiebig, Reinbek, Germany) were used. Spleen cell sus-

pensions from Bet v 1, Phl p 1, and Phl p 5 immunized mice were

incubated with each of the dodecapeptides (2 mg/well), spanning the

whole amino acid (aa) sequence of the respective antigens. The

dodecapeptides overlapped by 9 residues. Proliferative responses

were measured according to a previous description.16

Synthesis, purification, and characterizationof peptides for intranasal pretreatment

Subsequent to the identification of the immunodominant regions

of Bet v 1, Phl p 1, and Phl p 5 by epitope mapping, the respective

peptides (see Results and Tables E1-E3 in the Journal’s Online

Repository at www.mosby.com/jaci) for intranasal pretreatment were

synthesized by using a 9-fluorenylmethoxycarbonyl strategy with

2-(1H-benzotriazol-1-yl)1,1,3,3 tetramethyluronium hexafluoro-

phosphat activation (0.1-mmol small-scale cycles) on the Applied

Biosystems (Foster City, Calif) peptide synthesizer Model 433A.20

The identity to the peptides was checked by mass spectrometry, and

the peptides were purified to >90% purity by preparative HPLC

(Pichem, Graz, Austria).

Dose-finding and kinetic studies forintranasal pretreatment

Dose-finding experiments and kinetic studies with Bet v 1, Phl p 1,

and Phl p 5 proteins were performed by using 0.1 to 100 mg/antigen

applied 3 times at 7-day intervals or every third day for 3 weeks. In

line with previous studies in monosensitized mice, a treatment

regimen with 10 mg applied 3 times every 7 days was chosen.16,17

Dose-finding experiments with the immunodominant peptides (1-100

mg/peptide) revealed that low-dose application (ie, 5mg/peptide) was

optimal for intranasal tolerance induction.

Polysensitization and intranasal pretreatment

Polysensitization was performed by 3 intraperitoneal injections

(days 22, 36, and 50) of a mixture of 5mg rBet v 1, 5 mg rPhl p 1, and

5 mg rPhl p 5 adsorbed to aluminium hydroxide (Al[OH]3; Serva,

Heidelberg, Germany) at 14-day intervals (group 1). For intranasal

pretreatment, a mixture of the 3 allergens, Bet v 1, Phl p 1, and Phl p 5

(10 mg each), was applied intranasally in 30mL 0.9%NaCl 3 times at

7-day intervals (days 0, 7, and 14) before polysensitization (group 2).

Peptide pretreatment was performed by applying a mixture of 5 mg

Bet v 1 peptide, 5 mg Phl p 1 peptide, and 5 mg Phl p 5 peptide 2 in a

volume of 30 mL (group 3). Intranasal pretreatment with the hybrid

peptide was performed by using a concentration of 20mg construct in

30mLper application (group 4). Controlmicewere intranasally sham-

treated with 30 mL 0.9% NaCl before polysensitization (group 1).

One week after the last intraperitoneal immunization, an aerosol

challenge with 1% wt/vol BP and Phleum extract was performed on

2 consecutive days, as previously described.17

Sampling

Blood samples were taken before treatment and 2 days after the

aerosol challenge (day 60) by tail bleeding. After the bleeding, the

mice were sacrificed, spleen cell suspensions were prepared, and

bronchoalveolar lavages were collected, as previously described.16,17

Airway eosinophilia

Bronchoalveolar lavage samples were spun onto microscope

slides and stained with hematoxylin and eosin (Hemacolor; Merck,

Darmstadt, Germany). The number of eosinophils was expressed

as percentage of the total counted cell number, as previously

described.19

Detection of allergen-specific antibodylevels in serum

Microtiter plates (Nunc, Roskilde, Denmark) were coated with

each of the recombinant allergens (5 mg/mL) before incubation with

sera. Rat antimouse IgG1, IgG2a, and IgE antibodies (1/500;

Pharmingen, San Diego, Calif) were used, followed by peroxidase-

conjugated mouse antirat IgG antibodies (1/2000; Jackson Immuno



Hufnagl et al 371

Rhinitis,sinusitis,

and

ocu

lardisease

s


Lab,West Grove, Pa).16 Results show the OD values after subtraction

of baseline levels (0.058 6 0.025) from preimmune sera.

Rat basophil leukemia cell mediatorrelease assay

Rat basophil leukemia (RBL)-2H3 cells were incubated with sera

obtained from pretreated and polysensitized mice at dilutions of 1/30

to 1/300. Degranulation of RBL cells was induced by adding 0.03mg

of each allergen diluted in 100mLTyrode’s buffer. Supernatantswere

analyzed for b-hexosaminidase activity as previously described.21

Results are reported as percentages of total b-hexosaminidase

released after addition of 1% Triton X-100 and are shown after

subtraction of baseline release levels (1.13 6 0.81) obtained with

preimmune sera.

Lymphocyte proliferation andcytokine production

Proliferation of splenocytes (2 3 105 cells/well) after stimulation

with recombinant allergens (2 mg/well) was measured as previously

described.16

IFN-g, IL-4, IL-5, and IL-10 production was measured in spleen

cell suspensions incubated for 40 hours with BP (25 mg/well) or

Phleum extract (25 mg/well) as described.16 IL-5 was measured in

bronchoalveolar lavage fluids as previously described.17 For neutral-

ization of the IL-10 receptor in vitro pooled splenocytes were cultured

in the presence of rat antimouse IL-10 receptor mAb (60 mg/mL;

BD Biosciences, Heidelberg, Germany) or control rat IgG1,

k isotype antibody (BD Biosciences).22 Cytokine levels are shown

in pg/mL after subtraction of baseline levels (IFN-g, 147.23 6 61.6

pg/mL; IL-4/5, 4.8 6 4.1 pg/mL; IL-10, 27.75 6 23.3 pg/mL) of

unstimulated cultures.

Quantification of TGF-b mRNA expressionby real-time RT-PCR

Total RNA was isolated from pooled spleen cell suspensions on

the day of sacrifice by using RNeasy Minikit (Quiagen, Valencia,

Calif), treated with DNase (Quiagen), and then reverse-transcribed

into cDNA by using random hexamers (GeneAmp RT-PCR kit;

Perkin Elmer, Boston, Mass). Gene expression was determined by

quantitative real-time PCR by using predesigned TaqMan Gene

Expression Assays according to the manufacturer’s protocol on an

ABI 7700 sequence detection system (Applied Biosystems, Foster

City, Calif). Amplification of the endogenous control 18S rRNA

(Applied Biosystems) was performed to standardize the amount of

sample cDNA added, and relative quantitation was performed by

using the standard curve method (Applied Biosystems). The thermal

cycle conditions were 50C for 2 minutes and 95C for 10 minutes,

followed by 40 cycles of amplification at 95C for 15 seconds and

60C for 1 minute.

Adoptive cell transfer

Pretreatment with the peptide mixture or the hybrid peptide was

performed as described. Eight days after the last intranasal applica-

tion, donor spleen cells (1 3 107) were intravenously injected into

naive recipients.17 Four hours after the cell transfer, the recipients

were polysensitized as described, and immune responses were

evaluated 7 days after the last immunization.

Statistics

Data are expressed as means 6 SEMs from 3 independent

experiments. Differences between groups were tested by Kruskal-

Wallis tests. Post hoc comparisons for all pairs of groups were

performed applying Tukey-Kramer tests. P values below .05 were

considered significant. Pairwise comparison of sensitized versus

pretreated groups was performed by using the Mann-Whitney U test.

RESULTS

Polysensitized mice displayed comparableimmune responses to each of the3 allergens

Immunization with rBet v 1, rPhl p 1, and rPhl p 5induced comparably high IgG1, IgE, and IgG2a levelsand strong lymphocyte proliferative responses to each ofthe allergens (Table I). In addition, similar levels of IL-4,IL-5, and IFN-g were detected after stimulation of spleencell suspensions from polysensitized mice with BP andPhleum extract (Table I). Naive splenocytes stimulatedwith rBet v 1, rPhl p 1, rPhl p 5, BP, or Phleum extract didnot differ in their proliferative responses or in cytokinelevels from medium control levels (data not shown).

Epitope mapping studies for characterizationof the immunodominant peptides of Bet v 1,Phl p 1, and Phl p 5

In Bet v 1 immunized mice, 1 immunodominant T-cellepitope, MGETLLRAVESY, was located at the C-termi-nus corresponding to the aa sequence position 139-150.23

One immunodominant region of Phl p 1, AGELELQF-RRVKCKY, corresponding to the aa sequence position127-141, was identified, also described as a T-cell–reactive region in human beings.24 Concerning Phl p 5,2 immunodominant regions, KVDAAFKVAATAANAcorresponding to the aa sequence 166-180 (peptide 1) andTVATAPEVKYTVFETALK corresponding to the aa

TABLE I. Allergen-specific antibody levels, T-cell proliferation, and cytokine production in polysensitized mice

rBet v 1 rPhl p 1 rPhl p 5 BP Phleum

IgG1* 1.98 6 0.28 1.92 6 0.05 1.61 6 0.27 IFN-g 1501.9 6 621.1 1307.9 6 372.7

IgE* 0.85 6 0.21 0.48 6 0.05 0.53 6 0.02 IL-4 28.6 6 14.6 65.8 6 17.3

IgG2a* 0.51 6 0.35 0.31 6 0.12 0.75 6 0.37 IL-5 51.7 6 42.52 246.7 6 105.1

Stimulation index 2.97 6 0.51 2.63 6 1.02 4.41 6 2.38

*Serum antibody levels (OD) were measured by ELISA.

Spleen cells from polysensitized mice were cultured with the respective antigens for 4 days (stimulation index, background medium values

5820.2 6 1519.3 cpm).

Cytokine levels (pg/mL) were measured by ELISA. All results are mean values (6SDs) from 3 independent experiments with 5 animals per experiment.


AUGUST 2005

372 Hufnagl et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

sequence 226-243 (peptide 2), were identified. Compar-ison of the tolerogenicity of these peptides revealed thatpeptide 2 led to the highest immunosuppression inPhl p 5–sensitized mice (data not shown). Peptide 2 wastherefore selected for tolerance induction in polysen-sitized mice.

Synthesis of peptides for intranasaltolerance induction

According to the epitope mapping studies, wesynthesized single peptides from Bet v 1 (aa: SKEM-GETLLRAVESYLLAHSDE), Phl p 1 (aa: LRSAGEL-ELQFRRVKCKYPEG), and Phl p 5 (aa sequence:YAATVATAPEVKYTVFETALKKAI) for intranasalapplication as peptide mixture, or a hybrid peptidecomposed of the immunodominant regions of these 3allergens (aa: MGETLLRAVESYAGELELQFRRVK-CKYTVATAPEVKYTVFETALK; see Tables E1-E3 inthe Online Repository at www.mosby.com/jaci).

Intranasal tolerance induction with thepeptide mixture and the hybrid peptide,but not with the protein mixture,suppressed TH2 immune responses andairway inflammation in polysensitized mice

Sera from peptide mixture as well as from hybridpeptide–pretreated mice, but not from mice pretreatedwith the complete proteins, induced significantly lowerIgE-induced basophil degranulation in vitro than sera frompolysensitized mice (Fig 1, A). In accordance with thisreduction, IL-4 levels were significantly suppressed inspleen cell cultures from mice pretreated with the peptidemixture or the hybrid peptide. In contrast, IL-4 levels wereenhanced or remained unchanged in the protein-pretreatedgroup (Fig 1, B). Unlike the allergen-specific TH2responses, IgG2a antibody production was not signifi-cantly changed by either of the pretreatments (Table II).Similarly, IFN-g levels in vitro remained unchangedexcept for a significant enhancement in BP-stimulatedsplenocytes from the hybrid peptide–pretreated mice(Table II).

Airway inflammation, characterized by eosinophils andIL-5 production in bronchoalveolar lavages, was signifi-cantly reduced after pretreatment with the peptide mixtureor the hybrid peptide (Fig 2).

Polytolerance induction is not regulatedby TGF-b or IL-10

No significant changes in TGF-b mRNA expressionwere detected after pretreatment with the peptide mixtureor the hybrid peptide compared with polysensitizedcontrols (Fig 3, A). IL-10 levels were significantly sup-pressed in peptide mixture as well as hybrid peptide–pretreatedmice (Fig 3,B). Neutralization of IL-10 receptorin vitro did not abrogate the suppression of IL-10 levels inthe pretreated groups (Table III). Moreover, adoptive celltransfer experiments showed that tolerance was not trans-ferable by spleen cells, because IgE-induced basophil

degranulation remained unchanged in polysensitizedrecipients (Fig 3, C).

DISCUSSION

In the current study, we demonstrate that it is possible toprevent the development of multiple sensitivities to birchand grass pollen allergens by intranasal antigen applica-tion. Successful tolerance induction was achieved byintranasal administration of the immunodominant pepti-des of Bet v 1, Phl p 1, and Phl p 5, but not by the mixtureof the allergen proteins.

It was recently shown that the dose of coadministeredantigens is most important for the establishment of abalanced TH2 immune response.25 Accordingly, in thecurrent study, we demonstrated that immunization withthe optimal doses of Bet v 1, Phl p 1, and Phl p 5 (ie, 5 mgeach) led to comparable humoral and cellular immuneresponses to all 3 allergens/antigens (Table I). Moreover,the immunodominant epitopes recognized by T cells frompolysensitized mice were identical to some of the T-cellepitopes in patients allergic to birch and grass pol-len.24,26,27 Thus, our murine model of multiple allergensensitivity showed similar immunological characteristicsto those of human pollinosis.

From previous studies we know that rBet v 1 acts as apotent mucosal tolerogen in monosensitized mice.16,17,19

However, when rBet v 1 was intranasally applied beforepolysensitization with the 3 allergens, the tolerizingeffects of Bet v 1 toward itself were considerably im-paired.28 In addition, in polysensitized mice, mucosalpretreatment with Bet v 1 alone did not alter the immuneresponse toward the coapplied grass pollen allergens,indicating that no bystander or linked suppression can be

FIG 1. IgE-dependent allergen-specific basophil degranulation (RBL

assay) (A), and IL-4 production (B) in vitro from protein mixture

(hatched bars), peptide mixture (gray bars), and hybrid peptide–

pretreatedmice (black bars) comparedwith polysensitized controls

(white bars). *P < .05, **P < .01, pretreated vs polysensitized.



Hufnagl et al 373

Rhinitis,sinusitis,

and

ocu

lardisease

s


achieved in mice with multiple sensitivities (data notshown).22,29 Therefore, a mixture of all 3 allergen proteinswas used for intranasal pretreatment. However, it was notpossible to reduce IgE-dependent basophil degranulation,allergen-specific antibody responses, or cytokine produc-tion with the mixture of the proteins in polysensitizedmice. These results are in line with other studies showingthat the tolerogenic efficacy was impaired when morethan 1 protein antigen was ingested via the nasal or oralroute.30,31

Several studies have demonstrated that allergen-derived immunodominant peptides applied via mucosalsurfaces have a high tolerogenicity in naive and in primedanimals.15,32,33 It was further shown that treatment of micewith a peptide containing a single immunodominantepitope led to inhibition of immune responses directedto the other epitopes on the natural allergen, a phenom-enon called intramolecular suppression.33 Therefore, toenhance the efficacy of tolerance induction in the poly-sensitized mice, we applied a mixture of the immunodo-minant peptides or a synthetic hybrid composed of themajor T-cell epitopes of Bet v 1, Phl p 1, and Phl p 5.

Indeed, intranasal administration of either of the peptidemixture or the hybrid peptide led to a highly significantreduction of allergen-specific TH2 responses (Fig 1) andairway inflammation (Fig 2). Other studies showing thatapplication of a panel of peptides suppressed TH2 immuneresponses were directed only against single allergens,32,34

whereas this is the first report showing that peptide-induced tolerance can be directed against 3—and probablymore—different antigens/allergens.

Our observation that the mixture of protein allergenswas unable to induce tolerance, whereas the polypeptides

TABLE II. Allergen-specific IgG2a antibody levels and IFN-g production in vitro in polysensitized and pretreated mice

IgG2a (OD)* IFN-g (pg/mL)y

rBet v 1 rPhl p 1 rPhl p 5 BP Phleum

Polysensitized 0.51 6 0.35 0.31 6 0.12 0.75 6 0.37 1501.9 6 621.1 1307.9 6 372.7

Protein-tolerized 0.29 6 0.19 0.29 6 0.26 0.71 6 0.51 2160.7 6 752.1 1885.6 6 553.3

Peptide-tolerized 0.93 6 0.71 1.02 6 0.77 1.56 6 0.46 2305.7 6 798.5 1753.5 6 657.1

Hybrid-tolerized 1.25 6 1.03 1.07 6 0.82 1.38 6 0.61 3486.9 6 184.1** 1869.7 6 204.1

*IgG2a antibody levels were measured by ELISA.

IFN-g levels were measured in spleen cell cultures after stimulation with BP or Phleum extract.

**P < .01 hybrid-tolerized vs polysensitized as determined by Mann-Whitney U test.

FIG 2. Number of eosinophils (A) and IL-5 levels (B) in bronchoal-

veolar lavage fluids from polysensitized mice (white bars), from

peptide mixture–pretreated mice (gray bars), and from mice pre-

treated with the hybrid peptide (black bars). *P < .05, **P < .01,

pretreated vs polysensitized. BALF, Bronchoalveolar lavage fluid.

FIG 3. A, TGF-b mRNA expression from peptide mixture (gray bar)

and hybrid peptide–pretreated mice (black bar) shown as relative

values in comparison with polysensitized controls (white bar). B,

IL-10 production in vitro from polysensitized (white bars), peptide

mixture (gray bars), and hybrid peptide–pretreated (black bars)

mice. *P < .05, **P < .01, pretreated vs polysensitized. C, RBL assay

after adoptive transfer of spleen cells from peptide mixture (trans-

fer, gray bars) or hybrid peptide–pretreated mice (transfer, black

bars) compared with polysensitized controls (white bars).


AUGUST 2005

374 Hufnagl et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

were able to do so, leads to the speculation that because ofthe different conformation of the allergens, distinct antigen-presenting cells were targeted.35 It might further bespeculated that the reduced/nontolerogenic capacity ofthe proteinmixture is a result of an intermolecular competi-tion among a higher number of epitopes for the accessi-bility of MHC molecules than might be the case with only3 (or few) immunodominant peptides.36 In studies onsecondary prevention, the efficacy of the peptide treatmentdepending on their length was evaluated: whereas shortpeptides (20 aa or less) have the advantage that theycannot cross-link IgE on mast cells, they may requirecharacterization of major HLA-restricted T-cell epitopeson a patient basis.37 Long peptides (50 aa or more) mightovercome this problem, but at the cost of safety.32 Theobvious advantage of primary prevention, as performed inthis study, lies in the risk-free application of selectedpeptides, but limitations might still be based on thepolymorphisms of the HLA molecules. However, it isalso known that peptides capable of bindingmultiple HLAtypes and of being immunogenic in context with differentHLA molecules do exist.38 This has been also shown forseveral allergens.27,39 The fact that the immunodominantpeptides of the 3 allergens—representing major T-cellepitopes in a high number of patients allergic to birch andgrass pollen24,26,27—were also tolerogenic in anothermouse strain, ie, C57Bl/6J (data not shown), may indicatethat tolerance induction was not restricted to a particularMHC haplotype.

Concerning the underlying mechanisms, we previouslydemonstrated in Bet v 1–monosensitized mice that toler-ance induction depends on the conformation of the anti-gen: whereas tolerance induction with the whole Bet v 1molecule led to enhanced TGF-b and IL-10 mRNA levelsand tolerance was transferable with spleen cells, toleranceinduction with a fragment of Bet v 1—including theimmunodominant T-cell epitope—was not mediated byregulatory mechanisms.16,17 Similarly, in this study ofpolypeptide-induced tolerance, we found significantlyreduced IL-10 levels along with unchanged TGF-bmRNA levels. Neutralization of the IL-10 receptorin vitro did not abrogate the reduction of IL-10 (Table III).Moreover, because tolerance was not transferable withsplenocytes, it seems unlikely that regulatory cytokinesmediated tolerance in polysensitized mice. Thus, mecha-

nisms such as clonal anergy, described in a model ofpeptide-induced tolerance,32 or cytokine-independentmechanisms, such as cell-cell contact interaction via theNotch signalling pathway40 or via cytotoxic T lympho-cyte-associated antigen 4 expression on antigen-specificregulatory T cells,41 might rather play a role in our modelof polytolerance induction.

In conclusion, this is the first report showing thatmucosal application of polypeptide constructs inhibitedpolysensitization to several antigens/allergens. Thus,these data could be the basis for the development of amucosal polyvalent allergy vaccine for primary preven-tion of multiple sensitivities in atopic individuals. Itremains to be investigated whether polytolerance induc-tion can be used for treatment of already establishedmultisensitivities.

We thank Karin Baier for technical assistance, Dr Michael Kundi

for statistical analysis, and Dr Helmut Fiebig for providing the Phl p 5

peptides.

REFERENCES

1. D’Amato G, Spieksma FT, Liccardi G, Jager S, Russo M, Kontou-Fili K,

et al. Pollen-related allergy in Europe. Allergy 1998;53:567-78.

2. Ferreira FD, Hoffmann-Sommergruber K, Breiteneder H, Pettenburger

K, Ebner C, Sommergruber W, et al. Purification and characterization of

recombinant Bet v 1, the major birch pollen allergen: immunological

equivalence to natural Bet v 1. J Biol Chem 1993;268:19574-80.

3. Laffer S, Valenta R, Vrtala S, Susani M, van Ree R, Kraft D, et al.

Complementary DNA cloning of the major allergen Phl p I from timothy

grass (Phleum pratense); recombinant Phl p I inhibits IgE binding to

group I allergenes from eight different grass species. J Allergy Clin

Immunol 1994;94:689-98.

4. Vrtala S, Sperr WR, Reimitzer I, van Ree R, Laffer S, Muller WD, et al.

cDNA cloning of a major allergen from timothy grass (Phleum pratense)

pollen; characterization of the recombinant Phl p V Allergen. J Immunol

1993;151:4773-81.

5. Eriksson NE, Holmen A. Skin prick tests with standardized extracts of

inhalant allergens in 7099 adult patients with asthma or rhinitis: cross-

sensitizations and relationships to age, sex, month of birth and year of

testing. J Investig Allergol Clin Immunol 1996;6:36-46.

6. Silvestri M, Rossi GA, Cozzani S, Pulvirenti G, Fasce L. Age-dependent

tendency to become sensitized to other classes of aeroallergens in atopic

asthmatic children. Ann Allergy Asthma Immunol 1999;83:335-40.

7. Bousquet J, Lockey R, Malling HJ. Allergen immunotherapy: therapeu-

tic vaccines for allergic diseases: a WHO position paper. J Allergy Clin

Immunol 1998;102:558-62.

8. Bousquet J, Van Cauwenberge P, Khaltaev N. Aria Workshop Group,

World Health Organization. Allergic rhinitis and its impact on asthma.

J Allergy Clin Immunol 2001;108:S147-334.

9. Scheiner O, Kraft D. Basic and practical aspects of recombinant

allergens. Allergy 1995;50:384-91.

10. Valenta R, Lidholm J, Niederberger V, Hayek B, Kraft D, Gronlund H.

The recombinant allergen-based concept of component-resolved diag-

nostics and immunotherapy (CRD and CRIT). Clin Exp Allergy 1999;29:

896-904.

11. Alexander C, Kay AB, Larche M. Peptide-based vaccines in the

treatment of specific allergy. Curr Drug Targets Inflamm Allergy 2002;

1:353-61.

12. Valenta R. The future of antigen-specific immunotherapy of allergy.

Nat Rev Immunol 2002;2:446-53.

13. Holt PG, Patty JE, Turner KG. Inhibition of specific IgE responses in

mice by pre-exposure to inhaled antigen. Immunology 1981;42:409-17.

14. Tsitoura DC, DeKruyff RH, Lamb JR, Umetsu DT. Intranasal exposure

to protein antigen induces immunological tolerance mediated by func-

tionally disabled CD41 T cells. J Immunol 1999;163:2592-600.

TABLE III. IL-10 levels after neutralization of IL-10

receptor in vitro*

Control Ab aIL-10 receptor Ab

BP Phleum BP Phleum

Poly-sens 241.82 940.66 283.94 1095.78

Peptide-tol 138.63 472.41 158.91 661.14

Hybrid-tol 165.84 630.72 283.74 790.85

*Pooled splenocytes from polysensitized (poly-sens), peptide mixture

(peptide-tol), or hybrid peptide (hybrid-tol) pretreated mice were stimulated

with BP or Phleum extract in the presence of a neutralizing anti IL-10

receptor antibody or an isotype control. IL-10 levels (pg/mL) in

supernatants were measured by ELISA.



Hufnagl et al 375

Rhinitis,sinusitis,

and

ocu

lardisease

s

15. Jarnicki AG, Tsuji T, Thomas WR. Inhibition of mucosal and systemic

T(h)2-type immune responses by intranasal peptides containing a dom-

inant T cell epitope of the allergen Der p 1. Int Immunol 2001;13:

1223-31.

16. Wiedermann U, Jahn-Schmid B, Bohle B, Repa A, Renz H, Kraft D,

et al. Suppression of antigen-specific T-and B-cell responses by intrana-

sal or oral administration of recombinant Bet v 1, the major birch pollen

allergen, in a murine model of type I allergy. J Allergy Clin Immunol

1999;103:1202-10.

17. Wiedermann U, Herz U, Baier K, Vrtala S, Neuhaus-Steinmetz U,

Bohle B, et al. Intranasal treatment with a recombinant hypoallergenic

derivative of the major birch pollen allergen Bet v 1 prevents allergic

sensitization and airway inflammation in mice. Int Arch Allergy

Immunol 2001;126:68-77.

18. Takabayashi K, Libet L, Chisholm D, Zubeldia J, Horner AA. Intranasal

immunotherapy is more effective than intradermal immunotherapy for

the induction of airway allergen tolerance in Th2-sensitized mice.

J Immunol 2003;170:3898-905.

19. Winkler B, Baier K, Wagner S, Repa A, Scheiner O, Kraft D, et al.

Mucosal tolerance as therapy of type I allergy: intranasal application of

recombinant Bet v 1, the major birch pollen allergen, leads to the

suppression of allergic immune responses and airway inflammation in

sensitized mice. Clin Exp Allergy 2002;32:30-6.

20. Focke M, Mahler V, Ball T, Sperr WR, Majlesi Y, Valent P, et al.

Nonanaphylactic synthetic peptides derived from B cell epitopes of the

major grass pollen allergen, Phl p 1, for allergy vaccination. FASEB J

2001;15:2042-4.

21. Hufnagl K, Wagner B, Winkler B, Baier K, Hochreiter R, Thalhamer J,

et al. Induction of mucosal tolerance with recombinant Hev b 1 and

recombinant Hev b 3 for prevention of latex allergy in BALB/c mice.

Clin Exp Immunol 2003;133:170-6.

22. Winkler B, Bolwig C, Seppala U, Spangfort MD, Ebner C, Wiedermann

U. Allergen-specific immunosuppression by mucosal treatment with

recombinant Ves v 5, a major allergen of Vespula vulgaris venom, in a

murine model of wasp venom allergy. Immunology 2003;110:376-85.

23. Bauer L, Bohle B, Jahn-Schmid B, Wiedermann U, Daser A, Renz H,

et al. Modulation of the allergic immune response in BALB/c mice by

subcutaneous injection of high doses of the dominant T cell epitope from

the major birch pollen allergen Bet v 1. Clin Exp Immunol 1997;107:

536-41.

24. Schenk S, Breiteneder H, Susani M, Najafian N, Laffer S, Duchene M,

et al. T-cell epitopes of Phl p 1, major pollen allergen of timothy grass

(Phleum pratense): evidence for crossreacting and non-crossreacting

T-cell epitopes within grass group I allergens. J Allergy Clin Immunol

1995;96:986-96.

25. Von Garnier C, Astori M, Kettner A, Dufour N, Corradin G, Spertini F.

In vivo kinetics of the immunoglobulin E response to allergen: bystander

effect of coimmunization and relationship with anaphylaxis. Clin Exp

Allergy 2002;32:401-10.

26. Muller WD, Karamfilov T, Kahlert H, Stuwe HT, Fahlbusch B,

Cromwell O, et al. Mapping of T-cell epitopes of Phl p 5: evidence for

crossreacting and non-crossreacting T-cell epitopes within Phl p 5

isoallergens. Clin Exp Allergy 1998;28:1538-48.

27. Ebner C, Schenk S, Najafian N, Siemann U, Steiner R, Fischer GW, et al.

Nonallergic individuals recognize the same T cell epitopes of Bet v 1, the

major birch pollen allergen, as atopic patients. J Immunol 1995;154:

1932-40.

28. Hufnagl K, Winkler B, Baier K, Valenta R, Kraft D, Wiedermann U.

Induction ofmucosal tolerance by coapplication of recombinant Bet v 1, the

major birch pollen allergen, recombinant Phl p 1 and Phl p 5, major grass

pollen allergens, in polysensitized mice. Allergy 2002;57(suppl 73):49.

29. Miller A, Lider O, Weiner HL. Antigen-driven bystander suppression

after oral administration of antigens. J Exp Med 1991;174:791-8.

30. Mowat AM. The regulation of immune responses to dietary protein

antigens. Immunol Today 1987;8:93-8.

31. Shi FD, Bai XF, Xiao BG, van der Meide PH, Link H. Nasal

administration of multiple antigens suppresses experimental autoimmune

myasthenia gravis, encephalomyelitis and neuritis. J Neurol Sci 1998;

155:1-12.

32. Astori M, von Garnier C, Kettner A, Dufour N, Corradin G, Spertini F.

Inducing tolerance by intranasal administration of long peptides in naive

and primed CBA/J mice. J Immunol 2000;165:3497-505.

33. Hoyne GF, Jarnicki AG, Thomas WR, Lamb JR. Characterization of the

specificity and duration of T cell tolerance to intranasally administered

peptides in mice: a role for intramolecular epitope suppression. Int

Immunol 1997;9:1165-73.

34. Hirahara K, Tatsuta T, Takatori T, Ohtsuki M, Kirinaka H, Kawaguchi J,

et al. Preclinical evaluation of an immunotherapeutic peptide comprising

7 T-cell determinants of Cry j 1 and Cry j 2, the major Japanese cedar

pollen allergens. J Allergy Clin Immunol 2001;108:94-100.

35. Akdis CA, Blesken T, Wymann D, Akdis M, Blaser K. Differential

regulation of human T cell cytokine patterns and IgE and IgG4 responses

by conformational antigen variants. Eur J Immunol 1998;28:914-25.

36. Lo-Man R, Leclerc C. Parameters affecting the immunogenicity of

recombinant T cell epitopes inserted into hybrid proteins. Hum Immunol

1997;54:180-8.

37. Smith TR, Larche M. Investigating T cell activation and tolerance in

vivo: peptide challenge in allergic asthmatics. Cytokine 2004;28:49-54.

38. Southwood S, Sidney J, Kondo A, del Guercio MF, Appella E, Hoffman

S, et al. Several common HLA-DR types share largely overlapping

peptide binding repertoires. J Immunol 1998;160:3363-73.

39. Bohle B, Radakovics A, Jahn-Schmid B, Hoffmann-Sommergruber K,

Fischer GF, Ebner C. Bet v 1, the major birch pollen allergen, initiates

sensitization to Api g 1, the major allergen in celery: evidence at the

T cell level. Eur J Immunol 2003;33:3303-10.

40. Hoyne GF, Le Roux I, Corsin-Jimenez M, Tan K, Dunne J, Forsyth LM,

et al. Serrate1-induced notch signalling regulates the decision between

immunity and tolerance made by peripheral CD4(1) T cells. Int

Immunol 2000;12:177-85.

41. Fowler S, Powrie F. CTLA-4 expression on antigen-specific cells but not

IL-10 secretion is required for oral tolerance. Eur J Immunol 2002;32:

2997-3006.


AUGUST 2005

376 Hufnagl et al

Rhinitis,

sinusitis,

and

ocu

lardise

ase

s

Environmental and occupational respiratory disorders

Prevalences of positive skin test responses to10 common allergens in the US population:Results from the Third National Healthand Nutrition Examination Survey

Samuel J. Arbes, Jr, DDS, MPH, PhD,a Peter J. Gergen, MD, MPH,b

Leslie Elliott, MPH, PhD,a and Darryl C. Zeldin, MDa Research Triangle Park,

NC, and Bethesda, Md

Background: Allergy skin tests were administered in the second

and third National Health and Nutrition Examination Surveys

(NHANES II and III) conducted in the United States from 1976

through 1980 and 1988 through 1994, respectively.

Objectives: This study estimated positive skin test response

rates in NHANES III and identified predictors of one or more

positive test responses. Comparisons with NHANES II were

also made.

Methods: In NHANES III, 10 allergens and 2 controls were

tested in all subjects aged 6 to 19 years and a random half-

sample of subjects aged 20 to 59 years. Awheal-based definition

of a positive test response was used.

Results: In NHANES III, 54.3% of the population had positive

test responses to 1 or more allergens. Prevalences were 27.5%

for dust mite, 26.9% for perennial rye, 26.2% for short

ragweed, 26.1% for German cockroach, 18.1% for Bermuda

grass, 17.0% for cat, 15.2% for Russian thistle, 13.2% for white

oak, 12.9% for Alternaria alternata, and 8.6% for peanut.

Among those with positive test responses, the median number

of positive responses was 3.0. Adjusted odds of a positive test

response were higher for the following variables: age of 20 to 29

years, male sex, minority race, western region, old homes, and

lower serum cotinine levels. For the 6 allergens common to

NHANES II and III, prevalences were 2.1 to 5.5 times higher in

NHANES III.

Conclusions: The majority of the US population represented in

NHANES III was sensitized to 1 or more allergens. Whether

the higher prevalences observed in NHANES III reflect true

changes in prevalence or methodological differences between

the surveys cannot be determined with certainty. (J Allergy

Clin Immunol 2005;116:377-83.)

Key words: Allergens, allergic sensitization, allergy skin test,epidemiology, NHANES II, NHANES III, survey

Over the last 2 or more decades, rates of asthma haveincreased in the United States and worldwide, althoughthere is some evidence that asthma rates might havepeaked.1-3 One of the most important risk factors forasthma is sensitization to one or more allergens. TheNational Center for Health Statistics included allergyskin testing in the second and third National Health andNutrition Examination Surveys (NHANES II and III),which were conducted from 1976 through 1980 and 1988through 1994, respectively, to estimate and monitor theprevalence of allergic sensitization in the United States.

Although skin test results from NHANES II have beenpublished,4 a comprehensive summary of skin test resultsfrom NHANES III has not been published, nor has acomparison between NHANES II and III data beenpublished. The primary objectives were to estimate ratesof positive skin test responses in NHANES III and toidentify predictors of a positive test response to 1 or moreallergens. A secondary objective was to compare positiveskin test response rates between NHANES II and III;however, methodological differences between the 2surveys, which this article describes in detail, providechallenges for comparing and interpreting rate differencesbetween the 2 surveys.

METHODS

NHANES II and III

NHANES II and III were two in a series of population-based

surveys conducted by the National Center for Health Statistics to

determine the health and nutritional status of the US population. Both

surveys used a complex design to sample the civilian, noninstitu-

tionalized population. In NHANES II, questionnaires and medical

examinations were administered to 20,322 individuals aged 6 months

to 74 years, whereas in NHANES III, 31,311 individuals aged 2

months to 90 years were interviewed and examined.

Allergy skin testing in NHANES II and III

Prick-puncture allergy skin testing was performed in NHANES II

and III; however, there were important differences in age eligibility,

From athe Laboratory of Respiratory Biology, Division of Intramural Research,

National Institute of Environmental Health Sciences, National Institutes of

Health, Research Triangle Park, and bthe Division of Allergy, Immunology,

and Transplantation, National Institute of Allergy and Infectious Diseases,

National Institutes of Health, Bethesda.

This analysis of National Health and Nutrition Examination Survey (NHANES

III) data was funded by the National Institute of Environmental Health

Sciences and the National Institute of Allergy and Infectious Diseases.

Received for publication March 1, 2005; revised April 25, 2005; accepted for



Reprint requests: Darryl C. Zeldin, MD, NIEHS/NIH, PO Box 12233, MD

D2-01, Research Triangle Park, NC 27709. E-mail: [email protected].

0091-6749

doi:10.1016/j.jaci.2005.05.017

377

Environmenta

land

occupationalr

esp

irato

rydisord

ers


Abbreviations used

NHANES II: The second National Health and Nutrition

Examination Survey

NHANES III: The third National Health and Nutrition

Examination Survey

medical exclusion criteria, number and types of allergens tested,

standardization of allergen extracts, and reading times for the

reactions. Overviews of the allergy skin test protocols for both

surveys are presented here; however, details of the protocols can be

found elsewhere.5,6

In NHANES II, prick-puncture allergy skin tests to 8 allergens

(house dust, cat, dog, Alternaria alternata, mixed giant-short rag-

weed, oak, perennial ryegrass, and Bermuda grass) and 2 controls

(positive and negative) were administered to all subjects aged 6 to 74

years. Each of the allergens was commercially available and US Food

and Drug Administration licensed, but none was standardized. A

standardized extract is one for which a reference standard for potency

exists. The positive control was histamine phosphate, and the nega-

tive control was 50% glycerol saline. Subjects in 48 of the 64 primary

sampling units were tested with a histamine base concentration of

0.1 mg/mL, a less than optimal concentration, whereas the rest were

tested with the optimal concentration of 1.0 mg/mL.7 Lengths and

widths of wheals (raised area in the middle of the reaction) and flares

(reddish area around the wheal) were measured at 10 and 20 minutes.

Subjects with a history of allergy to cats, dogs, or ragweed were not

initially tested for those allergens. At the 10-minute reading, if the

subject reacted to fewer than 3 of the remaining allergens, then dog,

cat, and ragweed were tested on the other arm. If 3 or more responses

were positive, then only ragweed was tested on the other arm.

In NHANES III, prick-puncture allergy skin tests to 10 allergens

(Table I) and 2 controls (positive and negative) were administered to

all subjects aged 6 to 19 years and a random half-sample of subjects

aged 20 to 59 years. The positive control was histamine phosphate

(concentration is not published), and the negative control was 50%

glycerol saline.8 Only house dust mite, cat, and short ragweed

allergens were standardized (personal communication with Paul

Turkeltaub, MD, December 2, 2004). Lengths and widths of wheals

and flares were measured after 15 minutes (6 5 minutes). Subjects

weremedically excluded from skin testing if they usually did not have

trouble breathing in their chest or lungs but were having trouble

breathing at the time of the examination, although not from a cold; if

they usually had trouble breathing in their chest or lungs and had

more trouble breathing at the time of the examination; if they had a

severe response to allergen skin testing previously; or if they had

severe eczema or infection on both arms.

For comparisons between NHANES II and III, prevalences of

positive skin test reactions in NHANES II were estimated for the

6 allergens and ages (6-59 years) common to both surveys. The

6 allergens were cat, ragweed (mixed giant and short in NHANES II

and short in NHANES III), perennial rye, oak (oak inNHANES II and

white oak in NHANES III), Bermuda grass, and A alternata.

Definition of a positive skin test response

For our analyses of NHANES II and III skin test data, we

considered an allergen-specific skin test response positive if the skin

test panel was valid and the difference between the mean of the

wheal’s length and width for the allergen-specific test and the

negative control was at least 3 mm. A skin test panel was considered

valid if the difference between the mean wheal diameters of the

positive and negative controls was at least 1 mm. For NHANES II

results, measurements from the 20-minute reading were used.

In NHANES II, 11,769 of the 16,204 subjects who were age

eligible for skin testing were aged 6 to 59 years, and of those, 11,062

had a wheal-based result for the 6 allergens common to both surveys.

Of the 11,062 subjects, 7230 had a valid skin test panel, 3024 had an

invalid panel, and 808 were missing a positive control result. The

NHANES II analysis was limited to the 7230 subjects; however, a

secondary analysis was conducted without regard to the valid panel

criterion (n = 11,062).

In NHANES III, there were 12,106 age-eligible subjects, and of

those, 10,863 participated in skin testing, 174 were excluded for

medical reasons, and 1069 refused or were unavailable for testing. Of

the 10,863 subjects, 10,841 had a result for all 10 allergens, and of

those, 10,508 had a valid skin test panel, 332 had an invalid panel, and

1 subject was missing a positive control result. The NHANES III

analysis was limited to the 10,508 subjects.


Percentages (with SEs) of the population with positive skin test

responses were estimated among the populations aged 6 to 59 years

represented by the surveys. Sociodemographic or medical examina-

tion variables were assessed as potential predictors of one or more

positive skin test responses in NHANES III. The complete list can be

viewed in Table E1 in the Online Repository in the online version of

this article at www.mosby.com/jaci. Potential predictors were eval-

uated first with x2 statistics and then with multivariable logistic

regression by using a backward selection process. The process began

with all potential predictors in the model and endedwith variables at a

P value of .050 or less. Education, rather than poverty/income ratio,

was modeled as an indicator of socioeconomic status because the

latter had a large number of missing values, and serum cotinine level

was modeled in place of smoker in the home because serum cotinine

level is a biomarker for tobacco smoke exposure. Two-way interac-

tions between sex, age, and race-ethnicity were evaluated and

adjusted for the other predictors in the model. Only interactions

significant at the .050 level were reported.

Statistical analyses were conducted with SAS Version 9.1 (SAS

Institute, Cary, NC) or SUDAAN Release 9.0 (RTI International,

Research Triangle Park, NC) software. All percentages and odds

TABLE I. Prevalences of positive skin test responses

among the US population aged 6 to 59 years

represented in NHANES III

Allergen tested Percentage (SE)

Indoor allergens

Dust mite 27.5 (1.02)

German cockroach 26.1 (0.82)

Cat 17.0 (1.00)

At least one indoor allergen 43.0 (1.12)

Outdoor allergens

Perennial rye 26.9 (0.88)

Short ragweed 26.2 (1.03)

Bermuda grass 18.1 (0.81)

Russian thistle 15.2 (0.92)

White oak 13.2 (0.78)

Alternaria alternata 12.9 (0.69)

At least one outdoor allergen 40.0 (1.22)

Food allergen: peanut 8.6 (0.51)

At least one indoor or outdoor allergen 53.9 (1.02)

At least one of any type 54.3 (1.00)


AUGUST 2005

378 Arbes et al

Enviro

nmenta

land

occu

patio

nalre

spira

tory

diso

rders

ratios reported in this article were weighted to represent population

estimates, and all SEs were adjusted for the complex survey design.

Numbers of subjects reported in this article are unweighted.

RESULTS

NHANES III: Prevalences of positiveskin test responses

Table I shows the prevalences of positive skin testresponses among the US population aged 6 to 59 years.More than half of the population had positive testresponses to one or more allergens. The highest preva-lences were for dust mite, rye, ragweed, and cockroach,and the lowest prevalence was for peanut. A positive testresponse to at least 1 indoor allergen was slightly morecommon than a positive test response to at least 1 outdoorallergen (43.0% vs 40.0%), even though twice as manyoutdoor allergens were tested.

The percentage of the population with a positive testresponse decreased as the number of positive test re-sponses increased from 1 to 10 (Fig 1). A solitary positiveskin test response was seen in 15.5% (SE = 0.48) of thetotal population and 28.7% (SE = 0.95) of the populationwith positive test responses. The 2 most common solitaryreactions were to cockroach (4.3% [SE = 0.42] of the totalpopulation) and dust mite (4.2%, SE = 0.24). The preva-lences of a solitary reaction to the other allergens rangedfrom 0.10% to 1.70% of the total population. The meanand median numbers of positive test responses amongthose with positive test responses were 3.5 (SE = 0.06)and 3.0, respectively.

Table II shows how positive skin test responses—classified as indoor, outdoor, and peanut—were distrib-uted among the total US population and among those withpositive test responses. Among those with positive test re-sponses, 41% reacted to a combination of indoor and out-door allergens (but had negative test responses to peanut).A positive test response to peanut alone was quite rare(0.6%), as were positive test responses to indoor allergensand peanut (0.3%) and outdoor allergens and peanut(2.3%).

NHANES III: Predictors of 1 or morepositive test responses

The independent predictors of 1 or more positive testresponses were sex, age, race-ethnicity, census region,home construction year, and serum cotinine level. Thedistributions of these predictors in the US population andtheir adjusted odds ratios are shown in Table III. Thedistributions of all tested predictors and their bivariateassociations with each of the 10 allergen skin tests can befound in Table E1 in the Online Repository in the onlineversion of this article at www.mosby.com/jaci.

Age was bivariately associated with each allergen test(Table E1). The prevalence of 1 or more positive testresponses, as well as the adjusted odds ratio, increasedfrom the first decade of age to the second, peaked in thethird decade, and then decreased through the sixth decade(Table III).

For each allergen tested, male subjects were more likelythan female subjects to have positive test responses (TableE1). The adjusted odds of having 1 or more positive testresponses were 1.6 times greater in male subjects (TableIII). The odds ratio for sex did not differ by age (P valuefor sex-age interaction term = .518); however, it diddiffer by race-ethnicity (P value for sex-race interactionterm = .027). With the sex-race interaction term in themodel (model not shown), the adjusted odds ratios com-paring male subjects with female subjects were 1.6 (95%CI, 1.3-2.0) for non-Hispanic whites, 1.4 (95% CI, 1.1-1.8) for non-Hispanic blacks, 1.1 (95% CI, 0.9-1.4) forMexican Americans, and 2.0 (95% CI, 1.1-3.8) for others.

Race-ethnicity was bivariately associated with a posi-tive test response to 7 of the 10 allergens (Table E1).Compared with non-Hispanic whites, the adjusted oddsof having 1 or more positive test responses were greaterfor each of the other 3 race-ethnicity categories (TableIII). However, as mentioned in the previous paragraph,

FIG 1. Percentage of the US population aged 6 to 59 years

(NHANES III) by numbers of positive skin test responses.

TABLE II. Distribution of positive skin test responses by

allergen classification among the US population aged

6 to 59 years represented in NHANES III

Percentage (SE)

Allergen type

Among

the total

population

Among the

population

with positive

test responses

Indoor only* 13.7 (0.55) 25.3 (1.13)

Outdoor only 9.7 (0.68) 17.9 (1.26)

Peanut only 0.3 (0.11) 0.6 (0.20)

Indoor and outdoor only 22.2 (1.10) 41.0 (1.57)

Indoor and peanut only 0.2 (0.06) 0.3 (0.11)

Outdoor and peanut only 1.2 (0.18) 2.3 (0.32)

Indoor, outdoor, and peanut 6.9 (0.50) 12.6 (0.88)

None 45.7 (1.00) –

Total 100.0 (0.00) 100.0 (0.00)

*House dust mite, cat, or German cockroach.

Short ragweed, perennial rye, Alternaria alternata, Bermuda grass,

Russian thistle, or white oak.



Arbes et al 379

Environmenta

land

occupationalr

esp

irato

rydisord

ers

there was a significant interaction between sex andrace-ethnicity. Among female subjects, the adjustedodds ratios for race-ethnicity (with non-Hispanic whitesas the referent) were 1.8 (95% CI, 1.4-2.2) for non-Hispanic blacks, 1.4 (95% CI, 1.2-2.7) for MexicanAmericans, and 1.4 (95% CI, 0.9-2.1) for others. Amongmale subjects, those adjusted odds ratios were 1.5 (95%CI, 1.2-1.8), 1.0 (95% CI, 0.8-1.2), and 1.7 (95% CI, 1.2-2.5), respectively.

Census region was bivariately associated with tests tothe outdoor allergens ragweed, rye, grass, and thistle(Table E1). For 1 or more positive test responses, theadjusted odds ratio was lowest for the south and highestfor the west (Table III).

Home construction year was bivariately associated witha positive test response to dust mite, cockroach, ragweed,and peanut (Table E1). The prevalence of one or morepositive test responses, as well as the adjusted odds ratio,was greatest in the oldest homes (Table III).

Serum cotinine levels were bivariately associated with4 of the 6 outdoor allergens (Table E1); however, for thoseallergens, the lowest cotinine level was associated with thehighest prevalence of a positive test response. That same

pattern remained in the adjusted model for 1 or morepositive test responses (Table III).

Comparisons between NHANES II and III

The prevalences in NHANES II for positive testresponses to the 6 allergens and ages (6-59 years) commonto both surveys were 12.5% (SE = 0.74) for ragweed,11.9% (SE = 0.62) for rye, 5.8% (SE = 0.54) for oak,5.2% (SE = 0.49) for Bermuda grass, 4.5% (SE = 0.29)for A alternata, and 3.1% (SE = 0.32) for cat. TheNHANES III prevalences for those 6 allergens were 2.1to 5.5 times higher (Table I), and the NHANES IIIpopulation was much more likely to react to at least 1 ofthe 6 allergens (41.9% [SE = 1.23] vs 21.8% [SE = 0.94]).As shown in Fig 2, rates of positive test responses wereconsistently higher in NHANES III than in NHANES II ateach age group.

For both surveys, the prevalences without the valid-panel criterion were systematically less, although onlyslightly less. For example, the rate for a positive testresponse to 1 or more of the 6 allergens decreased from41.9% to 41.4% in NHANES III and 21.8% to 19.6% inNHANES II.

DISCUSSION

The main finding of this study was that 54.3% of thepopulation represented by NHANES III had 1 or morepositive skin test responses to 10 common allergens. Withthe limited number of allergens tested, this might be anunderestimation of the prevalence of allergic sensitizationin the US population. On average, an individual with apositive test response reacted to 3 to 4 allergens, and mostwith positive test responses reacted to a combination ofindoor and outdoor allergens as opposed to indoor,outdoor, or peanut allergens alone. For each of the 6allergens tested in both NHANES II and III, the preva-lence of a positive test response was higher in NHANESIII at each decade of age.

Even though we analyzed positive skin test responserates for the allergens and ages common to both surveys,there were differences between the surveys that could notbe controlled, such as differences in medical exclusioncriteria, in reading times of the reactions, in the histamineconcentrations for the positive controls, and in the qualityof the allergen extracts. It would seem doubtful thatdifferences in medical exclusion criteria, reading times,and histamine concentrations would have contributedsignificantly to the differences in rates because only asmall percentage of subjects were excluded for medicalreasons in each survey, reading times overlapped some-what, and results remained essentially the same irrespec-tive of whether the histamine control was used in thedefinition of a positive test response. One methodologicaldifference that could potentially explain the differences inpositive skin test response rates is the potency of theallergens used. Only the cat and ragweed allergens testedin NHANES III were standardized, and without standard-

TABLE III. Prevalences and odds ratios for the indepen-

dent predictors of 1 or more positive skin test responses

among the US population aged 6 to 59 years represented

in NHANES III

Predictor

Percentage

(SE)

Adjusted*

odds ratio

(95% CI)

Wald F

test,

P value

Age (y)

6-9 45.6 (2.19) 1.0 (reference)

10-19 55.5 (1.33) 1.7 (1.3-2.1)

20-29 60.0 (1.79) 2.1 (1.6-2.8)

30-39 56.5 (1.96) 1.8 (1.4-2.4)

40-49 50.5 (2.71) 1.5 (1.1-2.0)

50-59 49.1 (2.70) 1.4 (1.0-1.8) <.001

Sex

Female 49.2 (1.23) 1.0 (reference)

Male 59.4 (1.21) 1.6 (1.4-1.8) <.001

Race-ethnicity

Non-Hispanic white 51.3 (1.17) 1.0 (reference)

Non-Hispanic black 62.0 (1.26) 1.6 (1.4-1.9)

Mexican American 57.1 (1.28) 1.2 (1.0-1.4)

Other 64.0 (2.85) 1.5 (1.2-2.0) <.001

Census region

South 50.8 (1.40) 1.0 (reference)

West 58.0 (1.51) 1.3 (1.1-1.6)

Northeast 57.9 (2.78) 1.2 (0.9-1.8)

Midwest 52.8 (2.57) 1.1 (0.9-1.5) .042

Year home constructed

1974 to present 53.1 (1.40) 1.0 (reference)

1946-1973 52.1 (1.60) 0.9 (0.8-1.1)

Before 1946 59.4 (1.69) 1.3 (1.1-1.6) .002

Cotinine (ng/mL)

0.035-0.100 56.9 (2.17) 1.0 (reference)

0.100-10.00 55.4 (1.63) 0.9 (0.7-1.1)

10.00-1080.00 51.0 (1.48) 0.7 (0.5-0.9) .012

*Adjusted for each variable in the table.


AUGUST 2005

380 Arbes et al

Enviro

nmenta

land

occu

patio

nalre

spira

tory

diso

rders

ization, it cannot be assumed that the potencies ofthe allergens were the same between surveys. In fact,unstandardized allergen extracts can vary greatly in theirpotencies.9

The relative potencies of the allergens used in these 2surveys are unknown, and because of this, it cannot bestated with any certainty that the increases in positive skintest response rates observed between NHANES II and IIIwere due to true increases in reactivity in the US popu-lation. However, we would like to present 2 argumentsthat support true increases.

First, potency between unstandardized allergens couldbe greater, less, or the same, and it would seem unlikelythat the potency would have been systematically greaterfor all 6 of the NHANES III allergens. An example of thevariability one might expect between allergen prepara-tions can be found within NHANES II itself. WithinNHANES II, complete panels of allergens were purchasedfrom 2 different manufacturers, and subjects were testedwith one panel or the other.7 In a comparison of positiveskin test response rates between these 2 panels of aller-gens, Gergen and Turkeltaub7 found that one panel gavehigher rates for 3 allergens, lower rates for 1 allergen, andsimilar rates for 4 allergens; however, none of the absolutedifferences was greater than 4.7%. Second, the increasesseen between NHANES II and III are consistent withreports from other countries, such as Japan,10 the UnitedKingdom,11 and Denmark.12

In NHANES III, sex, race-ethnicity, age, census region,home construction date, and serum cotinine level wereindependent predictors of 1 or more positive skin testresponses. Age was the strongest independent predictor of

1 or more positive skin test responses, with rates peakingat age 20 to 29 years. In cross-sectional studies it is oftendifficult to determine whether age effects are real or aredue to a cohort effect (ie, the effect of capturing a high-riskcohort at a point in time). However, the age-specificcomparisons between NHANES II and III (Fig 2) providestrong evidence that the prevalence of allergic sensitiza-tion truly peaks in the third decade of life. If the NHANESIII finding had been due to a cohort effect, then NHANESII rates would have peaked at a younger age.

The prevalence of 1 or more positive test responses washigher among male than female subjects, and the preva-lence was higher for male subjects at each decade of life. Inthe general population, male subjects have higher levels ofserum IgE than female subjects at any given age,13 butwhether sex influences sensitization primarily through agenetic or an environmental pathway is not known. Thehigher prevalence of allergic sensitization among malesubjects at any age is in contrast to the pattern seen withasthma. For asthma, the prevalence is greater in malesubjects during childhood but greater in female subjectsduring the teenage and adult years.14 This contrastsuggests that factors other than allergic sensitization areresponsible for the sex-related shift in asthma prevalenceobserved at or near puberty.

Compared with non-Hispanic whites, the odds of hav-ing 1 or more positive skin test responses were increasedfor the other 3 race-ethnicity categories. For NHANES II,Gergen et al4 reported that the age-adjusted prevalence of1 or more positive test responses was higher in blacks thanwhites; however, the difference was not statistically sig-nificant. In a study of allergic sensitization among children

FIG 2. Age-specific comparisons of positive skin test response rates for the 6 allergens tested in NHANES II

(dashed lines) and NHANES III (solid lines).



Arbes et al 381

Environmenta

land

occupationalr

esp

irato

rydisord

ers

in NHANES III, Stevenson et al15 argued that race orethnicity differences in sensitization likely reflect differ-ences in environmental exposures rather than genetics. Infinding race-ethnicity a strong predictor of positive testresponses to dust mite, cockroach, and A alternata,those authors reasoned that the association was mostlikely to be due to differences in housing and communityenvironments, which would lead to differences in allergenexposures.

For census region and home construction date, theallergen-specific results suggest that these predictorsaffect sensitization primarily through an environmentalpathway. Census region was bivariately associated withpositive test responses to outdoor allergens only, whichlikely reflects geographic differences in exposures to thoseallergens. Consistent with the NHANES III results,Gergen et al4 reported that positive test response rates inNHANES II were lowest in the south. Older homes werebivariately associated with positive test responses to theindoor allergens dust mite and cockroach. In the NationalSurvey of Lead and Allergens in Housing, a representativesurvey of US housing, it was shown that older homes hadhigher levels of dust mite, cockroach, and mouse allergensthan newer homes.16,17

Higher serum cotinine levels predicted lower preva-lences of 1 or more positive test responses. Activesmoking has been associated with increased serum levelsof total IgE; however, the published literature on therelationship between either active or passive smoking andskin test response positivity is inconclusive.18,19 Chronictobacco smoke exposure can suppress the immune systemand impair host defenses,20 which could potentially lead tolower sensitization rates; however, smoke avoidanceamong persons with allergies and asthma would alsolead to lower rates.

Two potential predictors worth discussing that did notremain in the final prediction model were the presenceof an indoor cat and the presence of an indoor dog. Therole of pet exposure in the cause of allergic sensitizationand disease is controversial. One limitation to this cross-sectional analysis was the inability to assess the timing ofexposures and the development of allergic sensitization,which could be an explanation for the lack of associationwith indoor cat and dog. Interestingly, the presence of anindoor cat was not associated with a positive skin testresponse to cat allergen. One potential explanation for thisnull result could be the pervasiveness of cat allergen in UShomes. In the National Survey of Lead and Allergens inHousing, 99% of homes with an indoor cat and 56% ofhomes without an indoor cat had cat allergen levels thatexceeded the proposed threshold for allergic sensitiza-tion.21 In epidemiologic studies the more widespread anexposure is within a population, the more difficult itbecomes to demonstrate its effects.22 Another potentialexplanation could be cat avoidance among persons whoare sensitized to cats.

In conclusion, the majority of the US populationrepresented in NHANES III was sensitized to 1 or moreallergens. Although it cannot be definitively concluded

that the increases in positive skin test response ratesobserved between NHANES II and III represent anincrease in the reactivity of the US population, such anincrease would be consistent with studies from othercountries. In NHANES 2005-2006, total and allergen-specific IgE levels are being measured in all subjects,along with levels of indoor allergens in their homes.

We thank Drs Stephanie London and Donna Baird (Epidemiology

Branch, National Institute of Environmental Health Sciences) for

providing helpful comments during the preparation of this

manuscript.

REFERENCES

1. Mannino DM, Homa DM, Akinbami LJ, Moorman JE, Gwynn C, Redd

SC. Surveillance for asthma—United States, 1980-1999. MMWR

Surveill Summ 2002;51:1-13.

2. Robertson CF, Roberts MF, Kappers JH. Asthma prevalence in

Melbourne schoolchildren: have we reached the peak? Med J Aust

2004;180:273-6.

3. Verlato G, Corsico A, Villani S, Cerveri I, Migliore E, Accordini S, et al.

Is the prevalence of adult asthma and allergic rhinitis still increasing?

Results of an Italian study. J Allergy Clin Immunol 2003;111:1232-8.

4. Gergen PJ, Turkeltaub PC, Kovar MG. The prevalence of allergic skin

test reactivity to eight common aeroallergens in the U.S. population:

results from the second National Health and Nutrition Examination

Survey. J Allergy Clin Immunol 1987;80:669-79.

5. National Health and Nutrition Examination Survey III. Training manual

for allergy component. Available at: http://www.cdc.gov/nchs/data/

nhanes/nhanes3/cdrom/nchs/manuals/train.pdf. Accessed November 9,

2004.

6. National Center for Health Statistics (U.S.). Public use data tape

documentation: allergy skin testing: tape number 5309: National Health

and Nutrition Examination Survey, 1976-80. Hyattsville (MD): US

Department of Health and Human Services, Public Health Service,

National Center for Health Statistics; 1986.

7. Gergen PJ, Turkeltaub PC. National Center for Health Statistics (US).

Percutaneous immediate hypersensitivity to eight allergens, United

States, 1976-80. Washington (DC): US Department of Health and

Human Services Public Health Service, National Center for Health

Statistics; 1986.

8. Plan and operation of the Third National Health and Nutrition Exam-

ination Survey, 1988-94. Series 1: programs and collection procedures.

Vital Health Stat 1 1994:(32)1-407.

9. Gleich GJ, Leiferman KM, Jones RT, Hooton ML, Baer H. Analysis of

the potency of extracts of June grass pollen by their inhibitory capacities

in the radioallergosorbent test. J Allergy Clin Immunol 1976;58:31-8.

10. Nakagomi T, Itaya H, Tominaga T, Yamaki M, Hisamatsu S, Nakagomi

O. Is atopy increasing? Lancet 1994;343:121-2.

11. Sibbald B, Rink E, D’Souza M. Is the prevalence of atopy increasing?

Br J Gen Pract 1990;40:338-40.

12. Linneberg A, Nielsen NH, Madsen F, Frolund L, Dirksen A, Jorgensen

T. Increasing prevalence of specific IgE to aeroallergens in an adult

population: two cross-sectional surveys 8 years apart: the Copenhagen

Allergy Study. J Allergy Clin Immunol 2000;106:247-52.

13. Barbee RA, Halonen M, Lebowitz M, Burrows B. Distribution of IgE in

a community population sample: correlations with age, sex, and allergen

skin test reactivity. J Allergy Clin Immunol 1981;68:106-11.

14. Asthma Prevalence, Health Care Use and Mortality, 2000-2001. Avail-

able at: http://www.cdc.gov/nchs/products/pubs/pubd/hestats/asthma/

asthma.htm. Accessed April 29, 2003.

15. Stevenson LA, Gergen PJ, Hoover DR, Rosenstreich D, Mannino DM,

Matte TD. Sociodemographic correlates of indoor allergen sensitivity

among United States children. J Allergy Clin Immunol 2001;108:747-52.

16. Arbes SJ Jr, Cohn RD, Yin M, Muilenberg ML, Burge HA, Friedman W,

et al. House dust mite allergen in US beds: results from the First National

Survey of Lead and Allergens in Housing. J Allergy Clin Immunol 2003;

111:408-14.


AUGUST 2005

382 Arbes et al

Enviro

nmenta

land

occu

patio

nalre

spira

tory

diso

rders

http://www.cdc.gov/nchs/data/nhanes/nhanes3/cdrom/nchs/manuals/train.pdf

http://www.cdc.gov/nchs/data/nhanes/nhanes3/cdrom/nchs/manuals/train.pdf

http://www.cdc.gov/nchs/products/pubs/pubd/hestats/asthma/asthma.htm

http://www.cdc.gov/nchs/products/pubs/pubd/hestats/asthma/asthma.htm

17. Cohn RD, Arbes SJ Jr, Yin M, Jaramillo R, Zeldin DC. National

prevalence and exposure risk for mouse allergen in US households.


18. Burrows B, Halonen M, Lebowitz MD, Knudson RJ, Barbee RA. The

relationship of serum immunoglobulin E, allergy skin tests, and smoking

to respiratory disorders. J Allergy Clin Immunol 1982;70:199-204.

19. Strachan DP, Cook DG. Health effects of passive smoking. 5. Parental

smoking and allergic sensitisation in children. Thorax 1998;53:117-23.

20. Sopori ML, Kozak W. Immunomodulatory effects of cigarette smoke.

J Neuroimmunol 1998;83:148-56.

21. Arbes SJ Jr, Cohn RD, Yin M, Muilenberg ML, Friedman W, Zeldin DC.

Dog allergen (Can f 1) and cat allergen (Fel d 1) in US homes: results

from the National Survey of Lead and Allergens in Housing. J Allergy

Clin Immunol 2004;114:111-7.

22. Rose G. Sick individuals and sick populations 1985. Bull World Health

Organ 2001;79:990-6.



Arbes et al 383

Environmenta

land

occupationalr

esp

irato

rydisord

ers

Airborne endotoxin in homes with domesticanimals: Implications for cat-specific tolerance

James A. Platts-Mills, BA, Natalie J. Custis, BA, Judith A. Woodfolk, MD, PhD, and

Thomas A. E. Platts-Mills, MD, PhD Charlottesville, Va

Background: Although endotoxin is known to increase

symptoms in allergic individuals, early exposure might

decrease sensitization. Similarly, the presence of an animal in

the home has been associated with decreased sensitization to

animal allergens. It has been suggested that the effect of

animals could be explained by increased endotoxin exposure.

Objective: We sought to investigate the effects of domestic

animals on airborne endotoxin.

Methods: By using a silent particle collector, air was sampled

over 24 hours in homes with or without animals. The total

volume sampled was approximately 1000 m3, which provides

quantities of allergen and endotoxin that can easily be

measured with standard assays.

Results: The quantity of endotoxin ranged from less than 0.5 to

more than 500 pg/m3, whereas cat and dog allergen ranged

from less than 0.002 to more than 5 ng/m3. Overall, the quantity

of airborne endotoxin was not higher in homes with at least one

animal. However, airborne endotoxin levels were significantly

lower in homes with a cat compared with homes with a dog

(P < .001). In keeping with this, there was a significant

correlation between airborne Can f 1 and airborne endotoxin

(r = 0.50, P < .01) but not between endotoxin and Fel d 1

(r = 0.17, P = .27).

Conclusions: The results demonstrate that endotoxin is present

in the air of almost all homes. Although higher levels were seen

in homes with a dog, similar levels might be present in homes

with no animals. The results argue that the effects of cat

ownership cannot be explained by increased exposure to

endotoxin. (J Allergy Clin Immunol 2005;116:384-9.)

Key words: Airborne endotoxin, cats, dogs

Respiratory symptoms related to domestic animals are asignificant health issue. However, in many studies theprevalence of IgE specific for cats or dogs is lower than forother major allergens, such as pollens or dust mites.1-3

This is not due to inadequate exposure because the majorallergens Fel d 1 and Can f 1 are found in schools, public

places, and houses without a cat, as well as in those with ananimal.4-7 Despite (or because of) this high allergenexposure, children raised in a house with an animal areless likely to become sensitized to animal allergens.8-12

One possible explanation for this paradoxical finding isthat allergic families choose to avoid owning animalsbecause of the perceived risk of sensitization.10,13 Wehave previously reported that high exposure to the catallergen Fel d 1 induces a form of immune tolerance that isallergen specific.9,14,15 Alternatively, it has been arguedthat cats and other animals might increase agents such asbacterial LPS (endotoxin) in the home.12 Endotoxin isknown to favor a shift away from TH2 responses in miceand might have the same effect on children.16,17 Inkeeping with this, children raised in close contact withfarm animals (ie, with high endotoxin exposure) have lessallergic disease.18,19 However, published reports are notconsistent about the effects of pets on either floor orairborne endotoxin levels.15,20,21

Measuring airborne allergen or endotoxin requires bothsensitive assays and a technique for collecting airborneparticles.22,23 The quantity measured is a function of theairborne concentration and the volume of air sampled. Ifthe concentration in the air is low, low-volume samplerswill require very sensitive assays and might still provideinadequate samples.22,24 On the other hand, a high-volume collector may sample the air repeatedly and thusunderestimate the airborne concentration, whereas collec-tors with a fan are at risk of artificially increasing the fluxof particles into the air.23,25 The ion-charging device(ICD) used here is silent but has a moderately high flowrate. The device has 3 stainless-steel collection plates fromwhich the particles can be removed and analyzed.We haveused this technique to measure airborne allergen levels inhomes and airborne endotoxin levels in animal facili-ties.26,27 Those studies included validating the assaytechniques and the collection efficiency of the device.

METHODS

Airborne sampling

The machines used for airborne sampling (Ionic Breeze Quadras

from The Sharper Image, San Francisco, Calif) cycle between 2

Abbreviations usedEU: Endotoxin units

ICD: Ion-charging device

From the Asthma and Allergic Diseases Center, University of Virginia.

Supported by National Institutes of Health grants AI-20565 and AID/EHS

grant P01-AI-50989. In addition, J.P.M received an unrestricted educational

grant from The Sharper Image.

Disclosure of potential conflict of interest: None disclosed.

Received for publication January 10, 2005; revised April 29, 2005; accepted

for publication May 9, 2005.


Reprint requests: Thomas A. E. Platts-Mills, MD, PhD, University of Virginia

Health Systems, Asthma and Allergic Diseases Center, PO Box 801355,

Charlottesville, VA 22908-1355. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.012

384

Enviro

nmenta

land

occu

patio

nalre

spira

tory

diso

rders


distinct flow rates. For each device, we timed the periods P1 and P2 of

each rate. By using a vanometer, the wind speed was measured at the

center of 114 squares, each with an area of 9 cm2, on a 2-dimensional

grid placed orthogonal to and directly in front of the machine. These

speeds were averaged and multiplied by the measurement area to

determine the flow rate at each speed, V1 and V2. The flow rate of

each device was calculated as follows: ðP1V11P2V2Þ=ðP11P2Þ incubic meters per minute. The average flow rate of the 14 devices used

in this study was 1.48 6 0.09 m3/min.26,27

A 20 L/min air-sampling pump and an ICD were run in parallel in

a room with artificially disturbed dust (by using a vacuum cleaner

without a filter) for 10-minute periods (n = 8) to determine the

collection efficiency of the devices. The 20 L/min pump collected

airborne particles by using the same prefilter as was used to clean the

collection plates of the ICDs. All samples were extracted overnight in

2 mL of PBS and assayed for endotoxin and Fel d 1. The flow rates of

the 2 devices and the amount of endotoxin and cat allergen measured

were used to determine the collection efficiency for ICDs. The mean

collection efficiency was 40.6%6 9.0% for endotoxin and 51.7%6

12.0% for Fel d 1, which were not significantly different. For the

purposes of comparison with other studies, an estimated sampling

rate of 0.67 m3/min was used, the product of the flow rate and a

mean particle collection efficiency of 45%. Using this estimated

sampling rate, we can convert values for the total quantities of

airborne endotoxin or allergen collected to airborne concentrations

(Figs 1-3).

All sampling used 2 ICDs in parallel running for 24 hours and

placed at least 6 feet apart and at least 4 feet from the wall. In each

case, the stainless-steel plates of the 2 ICDs were removed and

cleaned with a series of 3 filters (Millipore prefilters, AP20, 35 mm;

Millipore Corporation, Bedford, Mass) dampened with sterile water.

Each filter was placed in a 3-mL syringe and extracted overnight at

4C. The 3 filters from the first ICD were extracted in 2 mL of 1%

BSA in PBS-Tween for measuring Can f 1 and Fel d 1, whereas those

from the latter were extracted in 2 mL of endotoxin-free PBS for

measuring endotoxin. In preliminary experiments 2 ICDs were run in

parallel and both were assayed for either endotoxin or Fel d 1, and

there was a close correlation between samples for both endotoxin

(n = 56, r = 0.91) and Fel d 1 (n = 44, r = 0.93).

Domestic sampling

A total of 71 homes in Central Virginia were studied between

November 2003 and May 2004 to collect dust and carry out air

sampling for 24 hours. Because seasonal variation has been reported

for endotoxin, 43 homes were studied both in November-December

and April-May. A floor dust sample was also collected with a Hoover

handheld vacuum.

Animal room sampling

We sampled 20 mouse rooms from 4 different animal facilities

(vivariums) at the University of Virginia. Each room was sampled at

least twice. A variety of cage types are used, but for the purposes of

this study, we have distinguished only between open cages, which

have only a metal grill to prevent the animals from exiting the cage,

and filter-topped cages of any configuration. The detailed methods

have been published elsewhere.27 All animals were used for research

studies that had been approved by the University of Virginia

Institutional Animal Care and Use Committee.

Assays

Samples were assayed for Fel d 1 and Can f 1 by using 2-site mAb-

based ELISAs (Indoor Biotechnologies, Inc, Charlottesville, Va),

which are sensitive to 1 ng/mL, and for endotoxin with the Limulus

Amoebocyte Lysate test QCL 1000, which is sensitive to 0.3

endotoxin units (EU)/mL (equivalent to 30 pg of endotoxin/mL;

Bio-Whittaker/Cambrex). Because extract freeze-thaw cycles are

associated with a significant decrease in endotoxin concentration,

samples were assayed for endotoxin immediately on extraction. The

buffer used for extractionwas used in each assay as a negative control

and was less than the level of detection in all cases.


Because the exposure data had a log-normal distribution, all

values were reported as geometric means with 95%CIs, and statistics

FIG 1. Airborne endotoxin concentrations in picograms per cubic meter calculated from quantities collected

on the basis of 1 EU = 100 pg and a sampling rate of 0.67 m3/min (see the ‘‘Methods’’ section). Geometric

means are indicated. Any subset of homes with dogs had significantly more airborne endotoxin than any

subset without dogs.



Platts-Mills et al 385

Environmenta

land

occupationalr

esp

irato

rydisord

ers

were performed on log-transformed data. Means were compared

by using independent-sample t tests, whereas a least squares linear

regression was used to assess the seasonal variation in airborne

endotoxin, the correlation between airborne allergens and endotoxin,

and the correlation between duplicate measurements for endotoxin

and cat allergen. AP value of less than .05was considered significant.

All statistical analyses were performed with SPSS 11 (SPSS Inc,

Chicago, Ill).

RESULTS

Easily measurable concentrations of endotoxin andanimal allergens were present in the floor dust samples.There was a wide range of values, from 0.82 EU/mg to 660EU/mg, and overall, there was no significant effect ofanimal ownership (Table I). There were large differencesin the concentration of cat and dog allergens, which was inkeeping with the presence of animals.

The results for airborne endotoxin also showed nooverall difference between homes with animals and homeswithout animals (Table II). The same data are presented asairborne concentrations (Fig 1). When the results wereanalyzed by the species of animal present, there werehighly significant differences. Homes with a cat or catshad significantly lower airborne endotoxin levels thanhomes with a dog or with both species (Fig 1). In 43houses airborne sampling for endotoxin was carried outtwice, first in November-December and again in April-May (Fig 2). Overall, there was a good correlationbetween the 2 measurements (r = 0.57, P < .001). Inparticular, the values in homes with a cat were consistentlylower (ie, <30 pg/m3). Comparing airborne dog allergenlevels with airborne endotoxin levels showed a significantpositive correlation (r = 0.56, P < .001; Fig 3, A). Theassociation was not significant between endotoxin and catallergen (r = 0.13, P = .33; Fig 3, B).

Presenting the results as endotoxin collected over 24hours allows comparison between different homes andwith animal facilities sampled in the sameway. The resultsshow that endotoxin exposure in homes is lower than inanimal rooms where rats or mice are kept without a filtertop on the cage. However, in animal rooms where cagetops are present, which is the case for the majority ofanimal facilities, the mean endotoxin airborne concentra-tion was lower than that for homes (Table III).

DISCUSSION

In many studies the prevalence of IgE antibodiesspecific for cat or dog allergen is lower than for othermajor allergens, such as pollens or dust mites.1-3 This isnot due to inadequate exposure because (1) the phenom-enon is more marked among children living in a housewith a cat,8-12 (2) the allergens are present and airbornecontinuously, and (3) themajor allergens Fel d 1 andCan f 1are found in schools, public places, and houses withoutanimals, as well as those with an animal. Our datashow that the presence of a cat in the home does notincrease airborne endotoxin levels. Thus the allergen-specific tolerance to cat allergens cannot be attributedto increased endotoxin exposure in homes with a cat.

Inevitably, any sampling technique that collects aquantity of allergen or endotoxin that can be confidentlymeasured will alter the particles present in the air. Inaddition, it is well established that the airflow created by ahigh-volume air filter can increase airborne allergenlevels.23 Thus all methods of measuring airborne concen-trations of allergen or endotoxin involve a compromise.The particle collector used here has several disadvantagesand some major advantages. The disadvantages are (1)those associated with high-volume sampling and (2) that

FIG 2. Airborne endotoxin was measured in 43 homes on 2 separate occasions 4 months apart. Although there

was considerable variation between the 2 samples, the correlation was: r = 0.57, P < .001. For homes with cats,

shown with open circles, each sample was less than 30 pg/m3.


AUGUST 2005

386 Platts-Mills et al

Enviro

nmenta

land

occu

patio

nalre

spira

tory

diso

rders

the interpretation of the quantity collected requires anestimate of the efficiency with which particles are col-lected. The advantages of the device are that (1) it is silentand therefore well accepted for use in any room, includingbedrooms; (2) sampling of particles off the stainless-steelplates is simple and very consistent (this is not possiblewith most devices designed to clean the air, including allhigh-efficiency particulate air filters); and (3) it collectsparticles from large volumes of air, but because of thewide aperture, the velocity of air coming out is relativelylow.

In preliminary experiments sampling air at 18 L/min for2 to 6 hours, we were not able to detect significantendotoxin levels in most houses. Other groups have

successfully measured airborne endotoxin levels by usingvery low-volume sampling (ie, 2 L/min).22,28 However,that required extrasensitive assays and extensive precau-tions to avoid endotoxin contamination. The collectorused here has no electrical safety concerns. Most pumpsused for collecting airborne allergen are not approved asdomestic appliances and therefore should not be left in ahomewithout the presence of an investigator. The range ofresults observed (ie, from <5 to >5000 EU/24 hours or<0.5 to >500 pg/m3) is such that small differences in theestimated collection efficiency (ie, between 41% and52%) would not affect the interpretation of our results.

In an additional experiment (data not shown), airborneendotoxin and Fel d 1 levels were measured before and

FIG 3. A, Airborne dog allergen concentration in nanograms per cubic meter compared with airborne

endotoxin concentration in picograms per cubic meter (r = 0.56, P < .001). B, Airborne cat allergen

concentration in nanograms per cubic meter compared with airborne endotoxin concentration in nanograms

per cubic meter (r = 0.17, P = .27).




Environmenta

land

occupationalr

esp

irato

rydisord

ers

after disturbance of dust in an experimental room by usinga domestic vacuum cleaner without a filter. The resultssuggested that 90% of both airborne endotoxin and catallergen levels fell within 10 minutes. The rapid fallingrates of both endotoxin and cat allergen suggest that thelevel of disturbance is as important as the concentration inreservoirs of dust. The observation that endotoxin levels inthe floor dust of houses with cats or dogs are similar, yetthe airborne levels are significantly different, might sug-gest that dogs were a greater cause of disturbance.

Differences in the quantity of endotoxin either airborneor in the floor dust of houses with a cat cannot explain whythe presence of a cat in the house is associated with a formof immune tolerance. However, this is not to say thatairborne endotoxin is irrelevant to the response. In micethe immune response to inhaled ovalbumin can besuppressed or changed by high exposure to endotoxin,but the IgE antibody response is enhanced by smallquantities of endotoxin.16 We have found airborne endo-toxin in almost all homes, and this might be sufficient toact as an adjuvant for responses to inhaled allergens.29 At

present, it remains unclear whether inhaled endotoxin inhomes contributes either to sensitization or symptoms.The concentrations reported here in homes are higher thanthe concentrations reported to give rise to symptomsamong animal handlers.28 However, those studies usedlow-volume sampling and reported a much smaller rangeof results than we have found here or in animal facilities.27

Furthermore, recent short-term challenge studies foundthat doses of less than 10,000 EU produced very littleimmediate change in the lungs.30

Some authors have implied that both the effects of cowownership in Europe and the paradoxical effects of catownership are in keeping with the hygiene hypothesis.Our data argue in favor of a completely different inter-pretation of the effects of cat ownership. In our studies on4 different cohorts, the effect of cat ownership has been catspecific. In particular, cat ownership has no effect on theIgE antibody response to dust mite allergens.9,15 Inaddition, the immune response to Fel d 1 includes theIL-4–dependent isotype IgG4.9,10 Thus the nonallergicresponse to cat has the features of a modified TH2 responseand not the TH1 response that would be predicted if thetolerant response was related to increased endotoxinexposure.16-18

It is important to recognize that there are major differ-ences in the ways that cats and dogs are kept, and ourresults might be relevant to specific housing conditions. InNew Zealand the floor dust levels of endotoxin in homeswith or without cats (13.7 vs 17.4 EU/mg, not significant)were lower than the levels seen here.15 Thus we could belooking at 2 different phenomena overlapping in differentstudies. The first effect is tolerance induced specifically bycat allergen exposure, whereas the second, a nonspecificeffect of animal ownership, including dog ownership,

TABLE I. Floor dust concentration of endotoxin and allergens*

N Endotoxin (EU/mg)y Fel d 1 (mg/g) Can f 1 (mg/g)

No animals 23 49.8 (29.9-82.9) 2.5 (1.3-4.7) 2.3 (1.4-4.0)

Animals 39 63.9 (47.1-86.8) 25.1 (7.9-79.9) 49.4 (17.2-142)

Cat(s) only 13 82.8 (45.3-151) 583 (201-1690) 0.65 (0.24-1.7)

Dog(s) only 15 49.9 (31.3-79.5) 0.52 (0.24-1.1) 405 (226-728)

Both species 11 66.1 (39.5-110.7) 121 (36.1-405) 315 (169-586)

*All values are presented as geometric means (95% CIs).

Endotoxin levels were not significantly different between cats and dogs (P = .189) or cats and no animals (P = .221).

TABLE II. Airborne quantity of endotoxin and allergens collected in 24 hours*

n Endotoxin (EU/24 h)y Fel d 1 (ng/24 h) Can f 1 (ng/24 h)

No animals 28 115 (61-217) 6.6 (4.9-8.8) 3.3 (2.7-4)

Animals 43 240 (155-372) 102 (64-162) 138 (87-219)

Cat(s) only 16 68 (44-105) 488 (361-660) 3.4 (2.3-4.9)

Dog(s) only 15 447 (250-800) 4.7 (3.6-6.1) 777 (566-1067)

Both species 12 588 (282-1230) 1076 (545-2126) 1030 (790-1342)

*All values are presented as geometric means (95% CIs).

Homes with dogs had higher airborne endotoxin than homes with cats (P < .001) or homes with no animals (P = .003), as did homes with both animals

(vs homes with cats, P < .001; vs homes with no animals, P = .005).

TABLE III. Airborne endotoxin in homes and animal rooms

n EU/24 h (GM)

Homes without animals 28 115 (61-217)

Homes with animals 43 240 (155-372)

Mouse rooms (open cages) 8 1930 (752–4940)*

Mouse rooms (filter tops on cages) 20 47.8 (31.9-71.6)

GM, Geometric mean.

*Significantly higher than homes with or without animals (P < .001).

Significantly lower than homes with animals (P < .001) or without

animals (P = .024).


AUGUST 2005

388 Platts-Mills et al

Enviro

nmenta

land

occu

patio

nalre

spira

tory

diso

rders

could reflect increased exposure to endotoxin or otherbacterial products.

REFERENCES

1. Roost HP, Kunzli N, Schindler C, Jarvis D, Chinn S, Perruchoud AP.

Role of current and childhood exposure to cat and atopic sensitization.

European Community Respiratory Health Survey. J Allergy Clin

Immunol 1999;104:941-7.

2. Sears MR, Herbison GP, Holdaway MD, Hewitt CJ, Flannery EM,

Silva PA. The relative risks of sensitivity to grass pollen, house dust

mite and cat dander in the development of childhood asthma. Clin Exp

Allergy 1989;19:419-24.

3. Lewis SA, Weiss ST, Platts-Mills TA, Syring M, Gold DR. Association

of specific allergen sensitization with socioeconomic factors and allergic

disease in a population of Boston women. J Allergy Clin Immunol 2001;

107:615-22.

4. Sporik R, Ingram JM, Price W, Sussman JH, Honsinger RW, Platts-Mills

TA. Association of asthma with serum IgE and skin test reactivity to

allergens among children living at high altitude. Tickling the dragons

breath. Am J Respir Crit Care Med 1995;151:1388-92.

5. Almquist C, Larsson PH, Egmar AC, Hedren M, Malmberg P, Wickman

M. School as a risk environment for children allergic to cats and a site for

transfer of cat allergen to homes. J Allergy Clin Immunol 1999;103:

1012-7.

6. Custovic A, Green R, Taggart SC, Smith A, Pickering CA, ChapmanMD,

et al. Domestic allergens in public places. II: Dog (Can f1) and cockroach

(Bla g 2) allergens in dust and mite, cat, dog and cockroach allergens in the

air in public buildings. Clin Exp Allergy 1996;26:1246-52.

7. Perzanowski MS, Ronmark E, Nold B, Lundback B, Platts-Mills TA.

Relevance of allergens from cats and dogs to asthma in the northernmost

province of Sweden: schools as a major site of exposure. J Allergy Clin

Immunol 1999;103:1018-24.

8. Hesselmar B, Aberg N, Aberg B, Eriksson B, Bjorksten B. Does early

exposure to cat or dog protect against later allergy development? Clin

Exp Allergy 1999;29:611-7.

9. Platts-Mills T, Vaughan J, Squillace S, Woodfolk J, Sporik R. Sensiti-

zation, asthma, and a modified Th2 response in children exposed to cat

allergen: a population-based cross-sectional study. Lancet 2001;357:

752-6.

10. Perzanowski MS, Ronmark E, Platts-Mills TA, Lundback B. Effect of

cat and dog ownership on sensitization and development of asthma

among preteenage children. Am J Respir Crit Care Med 2002;166:

696-702.

11. Custovic A, Hallam CL, Simpson BM, Craven M, Simpson A, Wood-

cock A. Decreased prevalence of sensitization to cats with high exposure

to cat allergen. J Allergy Clin Immunol 2001;108:537-9.

12. Ownby DR, Johnson CC, Peterson EL. Exposure to dogs and cats in the

first year of life and risk of allergic sensitization at 6 to 7 years of age.

JAMA 2002;288:963-72.

13. Anyo G, Brunekreef B, de Meer G, Aarts F, Janssen NA, van Vliet P.

Early, current and past pet ownership: associations with sensitization,

bronchial responsiveness and allergic symptoms in school children. Clin

Exp Allergy 2002;32:361-6.

14. Reefer AJ, Carneiro RM, Custis NJ, Platts-Mills TA, Sung SS, Hammer

J, et al. A role for IL-10-mediated HLA-DR7-restricted T cell-dependent

events in development of the modified Th2 response to cat allergen.

J Immunol 2004;172:2763-72.

15. Erwin EA, Wickens K, Custis NJ, Siebers R, Woodfolk JA, Barry D,

et al. Cat and dust mite sensitivity and tolerance in relation to wheezing

among children with high exposure to allergens. J Allergy Clin Immunol

2005;115:74-9.

16. Eisenbarth SC, Piggott DA, Huleatt JW, Visintin I, Herrick CA,

Bottomly K. Lipopolysaccharide-enhanced, toll-like receptor 4-depen-

dent T helper cell type 2 responses to inhaled antigen. J Exp Med 2002;

196:1645-51.

17. Gereda JE, Leung DY, Thatayatikom A, Strieb JE, Price MR, Klinnert

MD, et al. Relation between house-dust endotoxin exposure, type

1 T-cell development, and allergen sensitization in infants at high risk

of asthma. Lancet 2000;355:1680-3.

18. Braun-Fahrlander C, Riedler J, Herz U, Eder W, Waser M, Grize L, et al.

Environmental exposure to endotoxin and its relation to asthma in

school-age children. N Engl J Med 2002;347:869-77.

19. Gehring U, Bischof W, Fahlbusch B, Wichmann H-E, Heinrich J. House

dust endotoxin and allergic sensitization in children. Am J Respir Crit

Care Med 2002;166:939-44.

20. Heinrich J, Gehring U, Douwes J, Koch A, Fahlbusch B, Bischof W,

et al. Pets and vermin are associated with high endotoxin levels in house

dust. Clin Exp Allergy 2001;31:1839-45.

21. El Sharif N, Douwes J, Hoet PHM, Doekes G, Nemery B. Concentrations

of domestic mite and pet allergens and endotoxin in Palestine. Allergy

2004;59:623-31.

22. Park JH, Spiegelman DL, Gold DR, Burge HA, Milton DK. Predictors of

airborne endotoxin in the home. Environ Health Perspect 2001;109:

859-64.

23. Luczynska CM, Li Y, Chapman MD, Platts-Mills TA. Airborne

concentrations and particle size distribution of allergen derived from

domestic cats (Felis domesticus). Measurements using cascade impactor,

liquid impinger, and a two-site monoclonal antibody assay for Fel d 1.

Am Rev Respir Dis 1990;141:361-7.

24. Schweitzer IB, Smith E, Harrison DJ, Myers DD, Eggleston PA,

Stockwell JD, et al. Reducing exposure to laboratory animal allergens.

Comp Med 2003;53:486-92.

25. Swanson MC, Agarwal MK, Reed CE. An immunological approach to

indoor aeroallergen quantitation with a new volumetric air sampler:

studies with mite, roach, cat, mouse, and guinea pig antigens. J Allergy


26. Custis NJ, Woodfolk JA, Vaughan JW, Platts-Mills TAE. Quantitive

measurement of airborne allergens from dust mites, dogs, and cats using

an ion charging device. Clin Exp Allergy 2003;33:986-91.

27. Platts-Mills J, Custis N, Kenney A, Tsay A, Chapman M, Feldman S,

et al. The effects of cage design on airborne allergens and endotoxin in

animal rooms: high-volume measurements with an ion-charging device.

Contemp Top Lab Anim Sci 2005;44:12-6.

28. Pacheco KA, McCammon C, Liu AH, Thorne PS, O’Neill M, Martyny J,

et al. Airborne endotoxin predicts symptoms in non-mouse-sensitized

technicians and research scientists exposed to laboratory mice. Am J


29. Liu AH. Endotoxin exposure in allergy and asthma: reconciling a

paradox. J Allergy Clin Immunol 2002;109:379-92.

30. Alexis NE, Lay JC, Almond M, Peden DB. Inhalation of low-dose

endotoxin favors local T(H)2 response and primes airway phagocytes

in vivo. J Allergy Clin Immunol 2004;114:1325-31.




Environmenta

land

occupationalr

esp

irato

rydisord

ers

Food allergy, dermatologic diseases, and anaphylaxis

COX-2 inhibition enhances the TH2 immuneresponse to epicutaneous sensitization

Dhafer Laouini, PhD, Abdala ElKhal, PhD, Ali Yalcindag, MD, Seiji Kawamoto, MD, PhD,

Hans Oettgen, MD, PhD, and Raif S. Geha, MD Boston, Mass

Background: Mechanical injury to the skin by scratching is an

important feature of atopic dermatitis (AD).

Objective: To investigate the role of COX-2 in allergic skin

inflammation elicited by epicutaneous (EC) sensitization via

introduction of ovalbumin through shaved tape-stripped skin.

Methods: COX-2 mRNA was measured by quantitative PCR,

and COX-2 protein was measured by Western blotting. We

investigated the effect of administration of the COX-2 selective

inhibitor NS-398 during EC sensitization with ovalbumin in a

mouse model of AD characterized by eosinophil skin

infiltration, elevated total and antigen specific IgE, and a

systemic TH2 response to antigen. We further examined the

response of COX-2–deficient mice to EC immunization with

ovalbumin.

Results: Tape stripping caused a transient increase in skin

COX-2 mRNA. In contrast, COX-2 mRNA was not increased

after ovalbumin sensitization. Infiltration by eosinophils and

expression of IL-4 mRNA in ovalbumin-sensitized skin sites,

ovalbumin specific IgE and IgG1 antibody responses, and IL-4

secretion by splenocytes after ovalbumin stimulation were all

significantly increased in EC mice that received NS-398. In

contrast, ovalbumin specific IgG2a antibody response and

IFN-g secretion by splenocytes after ovalbumin stimulation

were significantly decreased in these mice. COX-2–deficient

mice also exhibited an enhanced systemic TH2 response to

EC sensitization.

Conclusion: These results demonstrate that COX-2 limits the

TH2 response to EC sensitization and suggest that COX

inhibitors may worsen allergic skin inflammation in patients

with AD. (J Allergy Clin Immunol 2005;116:390-6.)

Key words: Atopic dermatitis, allergic skin Inflammation, NS-398,

COX-2, TH1, TH2

Prostaglandins are formed by the oxidative cyclizationof the central carbons within 20 carbon polyunsaturatedfatty acids. 5-COX is the key enzyme involved in the

conversion of arachidonic acid to prostaglandin (PG) G2

and PGH2. PGH2 is subsequently converted to a variety ofeicosanoids that include PGE2, PGD2, PGF2a, PGI2, andthromboxane A2.

1 The spectrum of prostaglandins pro-duced depends on the downstream enzymatic machineryexpressed in a particular cell type. Prostaglandins haveboth autocrine and paracrine effects. These are mediatedby an array of receptors, which are differentially expressedby various cell types. Two classes of prostaglandinreceptors exist: the membrane G-coupled receptor class,ie, E-prostanoid 1-4 receptors for PGE2; and the nuclearperoxisome proliferator-activated receptor (PPAR) class,ie, PPARa, PPARg, and PPARd, which acts as a tran-scription factor on ligand binding.2 Nonsteroidal anti-inflammatory drugs inhibit COX, leading to a markeddecrease in prostaglandin synthesis and inflammation.3

Two COX isoforms, COX-1 and COX-2, have beenidentified and are encoded by distinct genes.4 COX-1 isexpressed in nearly all tissues under basal conditions,suggesting that its major function is to generate prosta-glandin precursors for homeostatic regulation.5 COX-2 ismainly an inducible enzyme. Inflammatory cytokines,which include IL-1 and TNF-a, and growth factors, whichinclude TGF-a, platelet-derived growth factor, epidermalgrowth factor, and fibroblast growth factor, all have beenshown to induce COX-2 expression.6-9

Prostaglandins have profound effects on the immuneresponse. A large body of data suggests that addition ofPGE2 in vitro inhibits IL-12 production and promotesIL-10 production by antigen-presenting cells, inhibits theproduction of TH1 cytokines, and promotes TH2 celldifferentiation.10,11 Furthermore, PGE2 was shown toenhance IL-4–driven isotype switching to IgE.12 Topicalapplication of PGE2 suppresses the cutaneous immune

Abbreviations usedAD: Atopic dermatitis

BAL: Bronchoalveolar lavage

CysLT: Cysteinyl leukotriene

DC: Dendritic cell

EC: Epicutaneous

HPF: High-power field

PG: Prostaglandin

PPAR: Peroxisome proliferator-activated receptor

WT: Wild type

From the Division of Immunology, Children’s Hospital; and the Department

of Pediatrics, Harvard Medical School.

Dr Laouini and Dr ElKhal contributed equally to the article.

Supported by National Institutes of Health grant AI-31541.

Received for publication July 27, 2004; revised March 31, 2005; accepted for



Reprint requests: Raif S. Geha, MD, Enders 8, Division of Immunology,

Children’s Hospital, 300 Longwood Avenue, Boston, MA 02115. E-mail:

[email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.042

390

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis


response.13 The PPARg agonist 15-deoxy-d(12,14)-pros-taglandin J(2) also inhibits IL-12 production by macro-phages14 and ameliorates experimental autoimmuneencephalomyelitis.15 In a model of allergic airway in-flammation, COX inhibition by nonsteroidal anti-in-flammatory drug increased IL-5 and IL-13 productionin bronchoalveolar lavage (BAL) fluid and airwayhyperresponsivenessAHR.16 Furthermore, lung inflam-matory indices, which include BAL cells, proteins, andIgE as well as lung inflammation as determined byhistopathology, were significantly increased in the ab-sence of either COX-1 or COX-2.17

COX-1 is basally expressed at low levels in skin.18 Skininjury by UV light has been shown to induce COX-2expression,19 whereas mechanical injury increases PGE2

in the skin.20 This may be mediated by IL-1 and TNF-areleased fromkeratinocytes, fibroblasts, andmast cells.21-23

Atopic dermatitis (AD) is an inflammatory skin diseasethat frequently occurs in subjects with personal or familyhistory of atopic disease.24 Mechanical injury to the skinby scratching is an important feature of AD. We havedeveloped a mouse model of allergic skin inflammationelicited by epicutaneous (EC) sensitization with ovalbu-min. This model displays many of the features of humanAD, including a dermatitis characterized by dermal infil-tration of T cells and eosinophils and increased localexpression of TH2 cytokines and by a systemic allergenspecific TH2 response characterized by IgG1 and IgEantibodies and IL-4 secretion by splenocytes after in vitrostimulation with ovalbumin.25 We used this model toassess the role of COX-2 in allergic skin inflammation.

METHODS

Mice

BALB/c mice were obtained from the Jackson Laboratory (Bar

Harbor, Me). COX-2+/2 mice on C57BL6x129/SvlmJ background

and genetically matched controls were obtained from Taconic

(Germantown, NY). Homozygous COX-2–deficient mice were

obtained by genotyping the offspring of COX-2+/2 parents, and, as

previously described, were not fertile.26 All mice were kept in a

pathogen-free environment. All procedures performed on the mice

were in accordance with the Animal Care and Use Committee of the

Children’s Hospital.

EC sensitization

EC sensitization of female mice 4 to 6weeks old was performed as

described previously.25 Briefly, the skin of anesthetized mice was

shaved and tape-stripped 6 times. Ovalbumin (grade V; Sigma

Chemical Co, St Louis, Mo) 100 mg in 100 mL normal saline, or

placebo (100 mL normal saline), was placed on a patch of sterile

gauze (1 cm3 1 cm), whichwas secured to the skinwith a transparent

bio-occlusive dressing (Tegaderm, Owens & Minor Inc, Franklin,

Mass). Each mouse had a total of three 1-week exposures to the patch

separated by 2-week intervals. On day 49, the mice were killed and

their tissues examined.

Treatment with COX-2 inhibitor

Micewere given 1mg/kg of the selective COX-2 inhibitor NS-398

(Biomol Research Laboratories, Inc, Plymouth Meeting, Pa) intra-

peritoneally daily for the duration of the sensitization period. NS-398

(25 mg/mL in dimethyl sulfoxide) was diluted in a 5% NaHCO3

solution before injection.

Histological analysis

Specimens were fixed in 10% buffered formalin and embedded in

paraffin. Multiple 4-mm sections were stained with hematoxylin and

eosin. Individual cell types were counted blinded in 15 to 20 high-

power fields (HPFs) at 10003.

Quantitative RT-PCR for COX enzymemRNA expression

Five hundred milligrams of skin was homogenized by using a

Polytron RT-3000 (Kinematica AG, Brinkmann Instruments Inc) in

lysis buffer solution provided in the RNAqueous extraction kit

(Ambion Inc, Austin, Tex). RT was performed by using transcriptor

first-strand cDNA synthesis kit (Roche Diagnostic, Foster City,

Calif). PCR reactions were run on an ABI Prism 7700 (Applied

Biosystems, Foster City, Calif) sequence detection system platform.

Taqman primers with 6-carboxyfluorescein-labeled probe were

obtained from Applied Biosystems. The housekeeping gene b2-

microglobulin was used as a control. The relative gene expression

among the different samples was determined by using the method

described by Pfaffl.27

Determination of COX-2 protein expressionand PGE2 levels in skin

Five hundred milligrams of skin was homogenized in 1 mL 0.1

mol/L PBS solution (pH = 7.4) containing 1 mmol/L EDTA, 0.1

mmol/L indomethacin, and a cocktail of protease inhibitors. Fifteen

microliters of this solution was used for Western blotting for COX-2,

with a rabbit anti–COX-2 antiserum (Abcam Inc, Cambridge, Mass)

followed by horseradish peroxidase–conjugated donkey antirabbit

antibody (Amersham Bioscience, Temecula, Calif). The blots were

reprobed with mAb to actin (Chemicon International, Piscataway,

NJ) followed by horseradish peroxidase–conjugated sheep antimouse

antibody (Amersham Bioscience) for loading control. The rest of the

material was extracted as described,28 and the extract used to

determine PGE2 concentration by ELISA (Cayman Chemicals, Ann

Arbor, Mich).

Competitive RT-PCR evaluation ofcytokine mRNA in skin

Competitive RT-PCR evaluation of cytokine mRNA in skin was

performed as described previously.29 Skin biopsies were immediately

frozen in dry ice. The samples were homogenized in Trizol (GIBCO

BRL, Carlsbad, Calif) by using a Polytron RT-3000. RNA extraction

was performed following the manufacturer’s instructions. cDNAwas

synthesized from 10 mg total RNA in a 40-mL reaction mix by using

Superscript II (GIBCO BRL). The primers used to amplify cDNA for

b2-microglobulin, IL-4, and IFN-g and DNA amplification were as

described previously.29 To quantify cytokine mRNA, a fixed amount

of reverse-transcribed cellular mRNA was coamplified in the pres-

ence of serial dilutions of a multispecific internal plasmid control

(pMUS3), which contains nucleotide sequences of multiple cyto-

kines.30 Results were expressed as a ratio of cytokine cDNA to

b2-microglobulin cDNA. We have recently found that the results of

competitive RT-PCR for determination of cytokine mRNA in skin

compare favorably with those of quantitative RT-PCR. In 2 experi-

ments in BALB/c mice, each using 6 mice with EC with ovalbumin

and 6 mice with EC with saline, we found that the mean increase in

skin IL-4 mRNA expression after ovalbumin sensitization was 4.9-

fold using competitive PCR compared with 4.3-fold using quantita-

tive RT-PCR.



Laouini et al 391

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

Serum antibody determinations

IgG1, IgG2a, and IgE antiovalbumin antibodies were determined

by ELISA following the procedures we previously described.25

IL-4 and IFN-g synthesis by spleen cells

Single cell suspensions of spleen cells were prepared and cultured

at 2 3 106/mL in 24-well plates in the presence of ovalbumin (50

mg/mL) as previously described.31 Supernatants were collected after

96 hours. IL-4 and IFN-g were determined by ELISA (Pharmingen,

San Diego, Calif).


The nonparametric Mann-Whitney test was used to compare the

different mice groups.

RESULTS

Tape stripping induces expression ofCOX-2 mRNA in normal mouse skin

We used quantitative RT-PCR to examine the effect oftape stripping on COX-2 mRNA expression in mouseskin. Low levels of COX-2 mRNA were detectable inuninjured skin. After tape stripping 6 times, COX-2mRNA expression increased, with peak levels 8 hourspoststripping, and returning to normal 48 hours later (Fig1, A). In contrast, there was no detectable increase in thelevels of COX-1 mRNA levels in the skin after tapestripping. COX-2 and COX-1 mRNA levels in ovalbu-min-sensitized skin sites did not significantly differ fromthose in saline-sensitized or unmanipulated skin sites (Fig1, B). Western blotting analysis demonstrated that COX-2protein expression in the skin increased 8 hours after tapestripping (Fig 1, C). In contrast, there was no detectableincrease in COX-2 protein expression in ovalbumin-

sensitized skin sites compared with saline-sensitizedskin sites or with unmanipulated skin (Fig 1, C). Theincreased COX-2 mRNA expression observed 8 hoursafter stripping was associated with significantly increasedlevel of the COX metabolite PGE2 (Fig 1, D). There wasno increase in PGE2 levels in either ovalbumin-sensitizedor saline-sensitized skin sites.

Eosinophil infiltration is increased inovalbumin-sensitized skin sites of micetreated with COX-2 inhibitor

There was no difference in the numbers of eosinophilsin saline sensitized sites of untreatedmice andmice treatedwith NS-398 (Fig 2, A). Ovalbumin sensitization caused asignificant increase in the number of eosinophils in theskin of control untreated BALB/c mice, consistent withprevious observations.25 There were significantly moreeosinophils in ovalbumin-sensitized skin of mice treatedwith NS-398 than in ovalbumin-sensitized skin of un-treated controls (Fig 2,A). Ovalbumin sensitization causedan increase in skin mononuclear cells that was modestlybut significantly higher in mice treated with NS-398compared with untreated controls (Fig 2, B).

IL-4 expression is increased in EC skin sitesof mice treated with COX-2 inhibitor

We used competitive PCR to measure cytokine mRNAin skin. Low and comparable levels of IL-4 and IFN-gmRNA were detected in saline sensitized skin fromuntreated mice and mice treated with NS-398 (Fig 3).Consistent with previous results,25 expression of IL-4mRNA, but not IFN-g mRNA, markedly increased inovalbumin-sensitized skin sites of untreated controlBALB/c mice. IL-4 mRNA was significantly increased

FIG 1. COX-2 and COX-1 mRNA expression in the skin of BALB/C mice after tape stripping (n = 2 per group;

A) and after EC sensitization with ovalbumin (OVA) and saline (SAL) sensitization (n = 5 per group; B), expressed

as fold induction over levels in unmanipulated skin. C, COX-2 protein expression in skin after tape stripping

(left panel) and EC sensitization (right panel). Results are representative of 3 experiments. D, PGE2 levels in

skin (n = 4 per group). Columns and bars represent means and SEMs. *P < .05.


AUGUST 2005

392 Laouini et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

in ovalbumin-sensitized skin of mice treated with NS-398compared with untreated ovalbumin-sensitized controls.These results suggest that COX-2 products normallydownregulate the TH2 cytokine profile of infiltratingT cells.

Treatment with COX-2 inhibitor enhancesantigen specific IgE and IgG1 antibodyresponses to EC sensitization with ovalbumin

The TH2 cytokine IL-4 plays an important role inisotype switching to IgE and IgG1, whereas the TH1cytokine IFN-g plays an important role in isotype switch-ing to IgG2a.

32 To investigate whether COX-2 inhibitionenhanced the systemic TH2 response to EC sensitizationwith ovalbumin, we measured total and ovalbumin spe-cific IgE and IgG1 in serum. Fig 4, A, shows thatovalbumin specific IgG1 and IgE levels were significantlyhigher in mice treated with NS-398. Treatment withNS-398 had no effect on the IgG2a antibody response.These results suggest that COX products normally down-regulate the systemic IgE and IgG1 antibody response toEC-introduced antigen.

COX-2 inhibition causes increased systemicTH2 response and decreased systemic TH1response to EC sensitization

We have previously shown that splenocytes fromBALB/c mice with EC with ovalbumin secrete IL-4, and

IFN-g after ovalbumin stimulation in vitro.31 Fig 4, B,shows that splenocytes from ECmice treated with NS-398secreted significantly higher amounts of IL-4, and signif-icantly less IFN-g, than splenocytes of unsensitized,untreated controls. These results suggest that COXproducts normally limit the systemic TH2 response toEC-introduced antigen and promote the systemic TH1response.

Increased systemic TH2 response anddecreased systemic TH1 response inCOX-2–deficient mice

NS-398 may have effects other than COX-2 enzymeinhibition. To ascertain that the effect of NS-398 on theTH response to EC sensitization was a result of COX-2inhibition, we examined the response of COX-22/2

mice to EC immunization. Fig 5 shows that ovalbuminspecific IgE levels were significantly higher and ovalbu-min specific IgG2a levels were significantly lower in

FIG 2. Effect of the COX-2 inhibitor NS-398 on the number of

infiltrating eosinophils (A) and mononuclear cells (B) in ovalbumin

(OVA)–sensitized and saline-sensitized skin sites of untreated mice

and in mice treated with NS-398. The columns and error bars

represent means 6 SEMs/HPF of cells calculated by examining

15 to 20 HPFs per mouse (n = 6). *P < .05. **P < .01.

FIG 3. Effect of the COX-2 inhibitor NS-398 on IL-4 (A) and IFN-g (B)

mRNA expression in saline and ovalbumin (OVA)–sensitized skin.

Levels were normalized to b2-microglobulin. Pooled results of

experiments using 6 mice per group. Bars represent means 6

SEMs. *P < .05. **P < .01.

FIG 4. Effect of the COX-2 inhibitor NS-398 on (A) serum levels

of ovalbumin (OVA) specific IgG, IgE, and IgG2a and (B) cytokine

production by spleen cells from EC mice (n = 6 for each group).

Columns and error bars represent means 6 SEMs. *P < .05.

**P < .01. SAL., Saline; Sens., sensitization.

FIG 5. Serum levels of ovalbumin (OVA) specific IgG1 (A), IgE (B),

and IgG2a (C) in ECmice, COX-22/2mice, andWT controls (n = 6 for

each group). Columns and error bars represent means 6 SEMs.

*P < .05.



Laouini et al 393

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

COX-2–deficient mice than in wild-type (WT) controls.COX-2 deficiency had no detectable effect on the IgG1

antibody response. Fig 6 shows that splenocytes from EC-immunized COX-22/2 mice secreted significantly moreIL-4 and significantly less IFN-g than splenocytes ofgenetically matched WT controls.

DISCUSSION

The results of this study suggest that COX-2 productslimit the systemic TH2 response and enhance the systemicTH1 response to epicutaneously introduced antigen andpromote skin infiltration with eosinophils in a mousemodel of allergic dermatitis.

Both COX-1 and COX-2 were expressed in unmanip-ulated mouse skin (Fig 1, A), consistent with previousreports on mouse and human skin.18,33,34 Mechanicalinjury by tape stripping transiently upregulated COX-2mRNA but not COX-1 mRNA expression in mouse skin.This finding is consistent with the observation that PGE2

levels increase in human skin after tape stripping,20 whichwe confirmed in this mouse study (Fig 1, C).Keratinocytes, fibroblasts, mast cells, endothelial cells,and tissue macrophages have all been reported to expressCOX-2 after activation.21-23 IL-1b and TNF-a are knowninducers of COX-2 mRNA expression.6,7 A correlationhas been observed between levels of PGE2 and IL-1a intape-stripped skin.20

The COX-2 selective inhibitor NS-398 enhanced der-mal infiltration with eosinophils in our model (Fig 2, A).Eosinophil infiltration of the skin in our model is depen-dent on their expression of the eotaxin receptor CCR3,31

and that eotaxin expression is dependent on IL-4.29

Expression of mRNA for the TH2 cytokine IL-4, but notfor the TH1 cytokine IFN-g, in ovalbumin-sensitized skinsites was significantly enhanced in NS-398–treated mice(Fig 3). This may have contributed to increased eosinophilskin infiltration. The TH2 cytokine IL-5 is also importantfor eosinophil infiltration of the skin in our model.Although we did not measure IL-5 expression, it is likelythat it was also enhanced in NS-398–treated mice andcontributed to their exaggerated skin eosinophilia, be-cause IL-5 and IL-4 expression by TH2 cells is usuallyconcordant.

The increased infiltration by mononuclear cells ob-served in ovalbumin-sensitized skin of NS-398–treatedmice (Fig 2, B) may reflect the stronger TH2 responseexhibited by these mice. Tape stripping induces theexpression of the TH2 selective chemokines thymus andactivation-regulated chemokine and cutaneous T cell–attracting chemokine, which attract TH2 cells in the skin(unpublished data). Secretion of the TH2 cytokines IL-4and IL-13 by infiltrating TH2 cells further upregulates theexpression of TH2 selective chemokines by skin cells35

and enhances TH2 cell infiltration. Increased secretion ofTH2 cytokines in the skin of NS-398–treated mice maycontribute to the enhanced infiltration of T cells in the skinof these mice.

Treatment of mice with the COX-2 inhibitor NS-398promoted the TH2 systemic response to EC sensitization.This was evidenced by enhanced serum IgG1 and IgEantibody levels to ovalbuminmice (Fig 4,A) and increasedIL-4 secretion by splenocytes in response to stimulationwith ovalbumin (Fig 4, B). In contrast, IFN-g secretionwas significantly decreased (Fig 4, B). Studies with COX-22/2 mice strongly supported the conclusion that theenhancing effect of NS-398 on the TH2 response to ECimmunization was a result of inhibition of COX-2 activity.After EC immunization, these mice mounted a signifi-cantly higher ovalbumin specific IgE antibody responseand their splenocytes secreted significantly more IL-4 andsignificantly less IFN-g than splenocytes from WTcontrols (Figs 5 and 6).

Taken together, our results suggest that COX-2 prod-ucts normally limit the development of TH2 cytokines andpromote the development of TH1 cytokines in response toEC sensitization with antigen. The fact that there was nodetectable increase in COX-2 or COX-1 mRNA expres-sion, COX-2 protein expression, or PGE2 levels in oval-bumin-sensitized skin compared with either salinesensitized or unmanipulated skin (Fig 1, B-D) suggeststhat the effects of COX-2 inhibition may be exerted earlyin the sensitization phase of our model, which is depen-dent on tape stripping, because we are unable to sensitizethe mice without it. However, we cannot rule out an effecton later events via the inhibition of baseline COX-2activity in the skin.

Our findings of increased IL-4 response to EC sensiti-zation with inhibition or lack of COX-2 is in agreementwith previous findings that administration of the COXinhibitor indomethacin 2 days before and during intraper-itoneal immunization enhances mRNA expression andsecretion of the TH2 cytokines IL-5 and IL-13 in the lungafter allergen challenge.16 Similarly, allergic lung inflam-mation as evidenced by eosinophils in BAL fluid and lunghistopathology and TH2 cytokine secretion are enhancedin intraperitoneally immunized mice deficient in COX-1or COX-2.17,36 In these studies, COX inhibition did notresult in a significant increase in the IgE antibody responseto intraperitoneal immunization, and TH cytokine secre-tion by splenic T cells was not examined.

There is a plethora of evidence that the COX-2 productPGE2, which is present in skin of patients with AD,37

FIG 6. Cytokine production by spleen cells from ECmice, COX-22/2

mice, and WT controls (n = 6 for each group). Columns and error

bars represent means 6 SEMs. *P < .05. SAL., Saline.


AUGUST 2005

394 Laouini et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

inhibits the development of TH1 cells and promotes thedevelopment of TH2 cells in vitro,

10,38 although discrepantresults have also been reported.39,40 This inhibitory effectis exerted in a large part at the level of the dendritic cells,because PGE2 is a potent inhibitor of IL-12 production bythese cells10 and an inhibitor of IL-12b1 receptor andIL-12b2 receptor expression.41 On the basis of thesein vitro results, one would expect that decreased PGE2

generation in the skin subsequent to inhibition or lackof COX-2 may promote the TH1 response and inhibit theTH2 response to EC sensitization. In fact, the reverse wasobserved. However, PGE2 may not be the most abundantor most relevant prostaglandin generated in injured skin.Langerhans cells generate high amounts of PGD2 butvery little amounts of other prostaglandins.42 The prosta-noid PGI2 limits lung allergic inflammation,43 and micedeficient in the PGI2 receptor mount an exaggerated TH2response.43,44 Further studies are needed to examine thenature of prostaglandins that accumulate in the skin aftermechanical injury and the roles of individual prostaglan-dins in modulating the TH response to EC sensitization.

Inhibition or lack of COX-2 activity may result inenhanced leukotriene synthesis because of both increasedavailability of arachidonic acid substrate and release fromthe inhibitory effect of prostaglandins on 5-lipooxyge-nase–activating protein expression.45 Increased amountsof leukotriene C4 in the skin may promote CC chemokineligand 19-dependent mobilization of antigen bearingdendritic cells (DC) to lymph nodes,46 resulting in theexaggeration of what is already a TH2-skewed response toEC sensitization. Cysteinyl leukotrienes (cysLTs) maypromote the induction of TH2 responses by DCs, assuggested by the observation that intranasal administra-tion of DCs pulsed with antigen and cysLTs enhanceallergic inflammation compared with DCs pulsed withantigen alone.47 CysLTs also promote eosinophil loco-motion and hence infiltration at skin sites of allergicsensitization.48

EC sensitization of mice is relevant to human sensiti-zation because it mimics allergen sensitization via abradedskin in patients with AD. Although there are no data onCOX-2 expression in human AD, COX products areincreased in the skin of patients with AD,37 and COX-2gene expression is induced by IL-13, which is expressed inAD lesions.49 Our results clearly show that COX-2inhibition may exacerbate AD by promoting the systemicand cutaneous TH2 response and are best avoided in thisdisease.

REFERENCES

1. Smith W, Marnett L, DeWitt D. Prostaglandin and thromboxane

biosynthesis. Pharmacol Ther 1991;49:153-79.

2. Forman BM, Chen J, Evans RM. The peroxisome proliferator-activated

receptors: ligands and activators. Ann N Y Acad Sci 1996;804:266-75.

3. Simon LS. Actions and toxicity of nonsteroidal anti-inflammatory drugs.

Curr Opin Rheumatol 1996;8:169-75.

4. Smith WL, Garavito RM, DeWitt DL. Prostaglandin endoperoxide H

synthases (cyclooxygenases)-1 and -2. J Biol Chem 1996;271:33157-60.

5. Crofford LJ. COX-1 and COX-2 tissue expression: implications and

predictions. J Rheumatol 1997;24(suppl 49):15-9.

6. Raz A, Wyche A, Siegel N, Needleman P. Regulation of fibroblast

cyclooxygenase synthesis by interleukin-1. J Biol Chem 1988;263:

3022-8.

7. Diaz A, Chepenik KP, Korn JH, Reginato AM, Jimenez SA. Differential

regulation of cyclooxygenases 1 and 2 by interleukin-1 beta, tumor

necrosis factor-alpha, and transforming growth factor-beta 1 in human

lung fibroblasts. Exp Cell Res 1998;241:222-9.

8. Saha D, Datta PK, Sheng H, Morrow JD, Wada M, Moses HL, et al.

Synergistic induction of cyclooxygenase-2 by transforming growth

factor-beta1 and epidermal growth factor inhibits apoptosis in epithelial

cells. Neoplasia 1999;1:508-17.

9. Goppelt-Struebe M, Rehm M, Schaefers HJ. Induction of cyclooxygen-

ase-2 by platelet-derived growth factor (PDGF) and its inhibition by

dexamethasone are independent of NF-kappaB/IkappaB transcription

factors. Naunyn Schmiedebergs Arch Pharmacol 2000;361:636-45.

10. van der Pouw Kraan TC, Boeije LC, Smeenk RJ, Wijdenes J, Aarden

LA. Prostaglandin-E2 is a potent inhibitor of human interleukin 12

production. J Exp Med 1995;181:775-9.

11. Kalinski P, Hilkens CM, Snijders A, Snijdewint FG, Kapsenberg ML.

IL-12-deficient dendritic cells, generated in the presence of prostaglandin

E2, promote type 2 cytokine production in maturing human naive T

helper cells. J Immunol 1997;159:28-35.

12. Roper RL, Brown DM, Phipps RP. Prostaglandin E2 promotes B

lymphocyte Ig isotype switching to IgE. J Immunol 1995;154:162-70.

13. Rheins LA, Barnes L, Amornsiripanitch S, Collins CE, Nordlund JJ.

Suppression of the cutaneous immune response following topical appli-

cation of the prostaglandin PGE2. Cell Immunol 1987;106:33-42.

14. Azuma Y, Shinohara M, Wang PL, Ohura K. 15-Deoxy-delta(12,14)-

prostaglandin J(2) inhibits IL-10 and IL-12 production by macrophages.

Biochem Biophys Res Commun 2001;283:344-6.

15. Diab A, Deng C, Smith JD, Hussain RZ, Phanavanh B, Lovett-Racke

AE, et al. Peroxisome proliferator-activated receptor-gamma agonist

15-deoxy-delta(12,14)-prostaglandin J(2) ameliorates experimental auto-

immune encephalomyelitis. J Immunol 2002;168:2508-15.

16. Peebles RS Jr, Dworski R, Collins RD, Jarzecka K, Mitchell DB,

Graham BS, et al. Cyclooxygenase inhibition increases interleukin 5 and

interleukin 13 production and airway hyperresponsiveness in allergic

mice. Am J Respir Crit Care Med 2000;162:676-81.

17. Gavett SH, Madison SL, Chulada PC, Scarborough PE, Qu W, Boyle JE,

et al. Allergic lung responses are increased in prostaglandin H synthase-

deficient mice. J Clin Invest 1999;104:721-32.

18. Abd-El-Aleem SA, Ferguson MW, Appleton I, Bhowmick A, McCollum

CN, Ireland GW. Expression of cyclooxygenase isoforms in normal

human skin and chronic venous ulcers. J Pathol 2001;195:616-23.

19. Athar M, An KP, Morel KD, Kim AL, Aszterbaum M, Longley J, et al.

Ultraviolet B(UVB)-induced cox-2 expression in murine skin: an

immunohistochemical study. Biochem Biophys Res Commun 2001;

280:1042-7.

20. Reilly DM, Green MR. Eicosanoid and cytokine levels in acute skin

irritation in response to tape stripping and capsaicin. Acta Derm Venereol

1999;79:187-90.

21. Scholz K, Furstenberger G, Muller-Decker K, Marks F. Differential

expression of prostaglandin-H synthase isoenzymes in normal and

activated keratinocytes in vivo and in vitro. Biochem J 1995;309:263-9.

22. Warnock LJ, Hunninghake GW. Multiple second messenger pathways

regulate IL-1 beta-induced expression of PGHS-2 mRNA in normal

human skin fibroblasts. J Cell Physiol 1995;163:172-8.

23. Leong J, Hughes-Fulford M, Rakhlin N, Habib A, Maclouf J, Goldyne

ME. Cyclooxygenases in human and mouse skin and cultured human

keratinocytes: association of COX-2 expression with human keratinocyte

differentiation. Exp Cell Res 1996;224:79-87.

24. Leung DY. Pathogenesis of atopic dermatitis. J Allergy Clin Immunol

1999;104:S99-108.

25. Spergel J, Mizoguchi E, Brewer J, Martin T, Bhan A, Geha R.

Epicutaneous sensitization with protein antigen induces localized allergic

dermatitis and hyperresponsiveness to methacholine after single exposure

to aerosolized antigen in mice. J Clin Invest 1998;101:1614-22.

26. Morham SG, Langenbach R, Loftin CD, Tiano HF, Vouloumanos N,

Jennette JC, et al. Prostaglandin synthase 2 gene disruption causes severe

renal pathology in the mouse. Cell 1995;83:473-82.

27. Pfaffl MW. A new mathematical model for relative quantification in

real-time RT-PCR. Nucleic Acids Res 2001;29:e45.



Laouini et al 395

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

28. Takemura N, Takahashi K, Tanaka H, Ihara Y, Ikemoto A, Fujii Y, et al.

Dietary, but not topical, alpha-linolenic acid suppresses UVB-induced

skin injury in hairless mice when compared with linoleic acids. Photo-

chem Photobiol 2002;76:657-63.

29. Spergel JM, Mizoguchi E, Oettgen H, Bhan AK, Geha RS. Roles of TH1

and TH2 cytokines in a murine model of allergic dermatitis. J Clin Invest

1999;103:1103-11.

30. Shire D, Legoux P. Gene expression analysis using competitive reverse

transcription polymerase chain reaction and multispecific internal con-

trol. Totowa (NJ): Humana Press, Inc; 1995.

31. Ma W, Bryce PJ, Humbles AA, Laouini D, Yalcindag A, Alenius H,

et al. CCR3 is essential for skin eosinophilia and airway hyper-

responsiveness in a murine model of allergic skin inflammation. J Clin

Invest 2002;109:621-8.

32. Snapper CM, Paul WE. Interferon-gamma and B cell stimulatory

factor-1 reciprocally regulate Ig isotype production. Science 1987;236:

944-7.

33. Muller-Decker K, Reinerth G, Krieg P, Zimmermann R, Heise H, Bayerl

C, et al. Prostaglandin-H-synthase isozyme expression in normal and

neoplastic human skin. Int J Cancer 1999;82:648-56.

34. Muller-Decker K, Scholz K, Neufang G, Marks F, Furstenberger G.

Localization of prostaglandin-H synthase-1 and -2 in mouse skin:

implications for cutaneous function. Exp Cell Res 1998;242:84-91.

35. Romagnani S. Cytokines and chemoattractants in allergic inflammation.

Mol Immunol 2002;38:881-5.

36. Carey MA, Germolec DR, Bradbury JA, Gooch RA, Moorman MP,

Flake GP, et al. Accentuated T helper type 2 airway response after

allergen challenge in cyclooxygenase-1-/- but not cyclooxygenase-2-/-

mice. Am J Respir Crit Care Med 2003;167:1509-15.

37. Fogh K, Herlin T, Kragballe K. Eicosanoids in skin of patients with

atopic dermatitis: prostaglandin E2 and leukotriene B4 are present in

biologically active concentrations. J Allergy Clin Immunol 1989;83:

450-5.

38. Katamura K, Shintaku N, Yamauchi Y, Fukui T, Ohshima Y, Mayumi

M, et al. Prostaglandin E2 at priming of naive CD4+ T cells inhibits

acquisition of ability to produce IFN-gamma and IL-2, but not IL-4 and

IL-5. J Immunol 1995;155:4604-12.

39. Rieser C, Bock G, Klocker H, Bartsch G, Thurnher M. Prostaglandin E2

and tumor necrosis factor alpha cooperate to activate human dendritic

cells: synergistic activation of interleukin 12 production. J Exp Med

1997;186:1603-8.

40. Parker CW, Huber MG, Godt SM. Modulation of IL-4 production in

murine spleen cells by prostaglandins. Cell Immunol 1995;160:278-85.

41. Wu CY, Wang K, McDyer JF, Seder RA. Prostaglandin E2 and

dexamethasone inhibit IL-12 receptor expression and IL-12 responsive-

ness. J Immunol 1998;161:2723-30.

42. Ruzicka T, Aubock J. Arachidonic acid metabolism in guinea pig

Langerhans cells: studies on cyclooxygenase and lipoxygenase path-

ways. J Immunol 1987;138:539-43.

43. Jaffar Z, Wan KS, Roberts K. A key role for prostaglandin I2 in limiting

lung mucosal Th2, but not Th1, responses to inhaled allergen. J Immunol

2002;169:5997-6004.

44. Takahashi Y, Tokuoka S, Masuda T, Hirano Y, Nagao M, Tanaka H,

et al. Augmentation of allergic inflammation in prostanoid IP receptor

deficient mice. Br J Pharmacol 2002;137:315-22.

45. Harizi H, Juzan M, Moreau JF, Gualde N. Prostaglandins inhibit 5-

lipoxygenase-activating protein expression and leukotriene B4 produc-

tion from dendritic cells via an IL-10-dependent mechanism. J Immunol

2003;170:139-46.

46. Robbiani DF, Finch RA, Jager D, Muller WA, Sartorelli AC, Randolph

GJ. The leukotriene C(4) transporter MRP1 regulates CCL19 (MIP-

3beta, ELC)-dependent mobilization of dendritic cells to lymph nodes.

Cell 2000;103:757-68.

47. Machida I, Matsuse H, Kondo Y, Kawano T, Saeki S, Tomari S, et al.

Cysteinyl leukotrienes regulate dendritic cell functions in a murine model

of asthma. J Immunol 2004;172:1833-8.

48. Fregonese L, Silvestri M, Sabatini F, Rossi GA. Cysteinyl leukotrienes

induce human eosinophil locomotion and adhesion molecule expression

via a CysLT1 receptor-mediated mechanism. Clin Exp Allergy 2002;32:

745-50.

49. Yu CL, Huang MH, Kung YY, Tsai CY, Tsai YY, Tsai ST, et al.

Interleukin-13 increases prostaglandin E2 (PGE2) production by normal

human polymorphonuclear neutrophils by enhancing cyclooxygenase 2

(COX-2) gene expression. Inflamm Res 1998;47:167-73.


AUGUST 2005

396 Laouini et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

Responsiveness to autologous sweat andserum in cholinergic urticaria classifiesits clinical subtypes

Atsushi Fukunaga, MD, Toshinori Bito, MD, Kenta Tsuru, MD, Akiko Oohashi, MD,

Xijun Yu, MD, Masamitsu Ichihashi, MD, Chikako Nishigori, MD, and

Tatsuya Horikawa, MD Kobe, Japan

Background: It has been reported that patients with cholinergic

urticaria have a type 1 allergy to autologous sweat; however,

the pathogenesis of that disorder has not been fully elucidated.

Objective: We investigated the responsiveness to autologous

sweat and serum in patients with cholinergic urticaria in

relation to their clinical characteristics. We further classified

the clinical subtypes that are clearly characterized by

responsiveness to in vivo and in vitro tests as well as their

clinical features.

Methods: Intradermal tests with autologous sweat and serum

were performed in 18 patients with cholinergic urticaria.

Histamine release from peripheral blood basophils induced

by autologous sweat was measured.

Results: Eleven of 17 patients with cholinergic urticaria showed

positive reactions in skin tests with their own diluted sweat.

Substantial amounts of sweat-induced histamine release from

autologous basophils were observed in 10 of 17 patients. Eight

of 15 patients with cholinergic urticaria showed positive

reactions in the autologous serum skin tests. All 6 patients who

developed satellite wheals after the acetylcholine test showed

hypersensitivity to sweat. Further, patients whose eruptions

were coincident with hair follicles showed positive responses to

the skin test with autologous serum, whereas patients whose

eruptions were not coincident with hair follicles did not.

Conclusion: On the basis of these findings, we propose that

cholinergic urticaria should be classified into 2 distinct

subtypes. The first (nonfollicular) subtype shows strong positive

reactions to autologous sweat and negative reactions to

autologous serum. The second (follicular) subtype shows

weak reactions to autologous sweat and positive reactions to

autologous serum. (J Allergy Clin Immunol 2005;116:

397-402.)

Key words: Cholinergic urticaria, sweat, autologous serum, skin

test, histamine release test, acetylcholine test

Cholinergic urticaria (CU), which was first describedby Duke1 in 1924, is characterized by unique clinicalfeatures: pinpoint sized, highly pruritic wheals withsurrounding erythema that occur after sweating duringphysical exercise, taking a bath, raising the body temper-ature, and emotional stress. In typical cases, this disorderusually occurs in young adults. Occasionally, this disorderis accompanied with angioedema and anaphylactic reac-tions.2,3

The pathogenesis of CU has not yet been well clarifieddespite the fact that numerous investigators have de-scribed its clinical characteristics and possible pathogen-esis. In patients with CU, injection of acetylcholine(mecholyl) into normal-appearing skin produces a whealand flare reaction, often surrounded by smaller satellitelesions that are similar to the skin symptoms of CU.4

Acetylcholine is thus believed to play a significant role inthe development of the symptoms of CU. Another aspectof the pathogenesis of CU has focused on sweat itself onthe basis of evidence that this unique eruption occurs aftersweating. Adachi et al5 found that 20 patients with CUshowed immediate-type reactions after an intradermalskin test with autologous sweat. Kobayashi et al6 pre-sumed that the leakage of sweat into the dermis because ofductal occlusion at the superficial acrosyringium causesCU. Kaplan et al7 and Sigler et al8 demonstrated plasmahistamine elevations after exercise challenge of patientswith CU.

We performed this study to clarify further the possi-ble involvement of sweat-mediated and autoimmune-mediated mechanisms in CU and in its clinical features.Skin responsiveness was evaluated after the intracutane-ous injection of autologous sweat and serum. We assessedthe correlation between the degrees of skin reactions andamounts of in vitro histamine released from basophils afterstimulation with autologous sweat. We further analyzedthe relationship between the clinical symptoms of CU andthe characteristics of these tests.

Abbreviations usedCU: Cholinergic urticaria

ASST: Autologous serum skin test

From the Division of Dermatology, Department of Clinical Molecular

Medicine, Kobe University Graduate School of Medicine.

Disclosure of potential conflict of interest: A. Fukunaga, none disclosed;

T. Bito, none disclosed; K. Tsuru, none disclosed; A. Oohashi, none

disclosed; X.Yu, none disclosed;M. Ichihashi, none disclosed; C.Nishigori,

none disclosed; T. Horikawa, none disclosed.

Received for publication November 11, 2004; revised May 13, 2005; accepted

for publication May 17, 2005.


Reprint requests: Tatsuya Horikawa, MD, Division of Dermatology,

Department of Clinical Molecular Medicine, Kobe University Graduate

School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe, 650-0017,

Japan. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.024

397

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis


METHODS

Subjects

Eighteen patients with CU were enrolled at the Dermatological

Institute of Kobe University Hospital. CU was confirmed by the

development of numerous small wheals after exercise until sweat-

ing.9 All of the patients had no aquagenic urticaria. The character-

istics of patients with CU are described in Table I. Their age ranged

between 15 and 31 years (mean, 21.8 years). Nine patients had a past

history of atopic diseases (6 atopic dermatitis and 3 allergic rhinitis).

Five patients were accompanied with cold urticaria. Healthy control

subjects were enrolled from the staff at Kobe University Hospital. All

subjects provided oral consent for this study after oral and written

explanations. Relevant drugs such as histamine H1-receptor antag-

onists were withdrawn for at least 24 hours before the examination.

All of the patients had never had systemic corticosteroids for at least

3 months before the examination.

Materials

Venous bloodwas taken into sterile glass tubes and allowed to clot

at room temperature for 30 minutes. Serum was separated by cen-

trifugation at 500g for 20 minutes and passed through a 0.45-mm

MILLEX HV membrane (Millipore, Molsheim, France). Sweat was

collected from each patient’s forearm after exercise. The sweat was

sterilized after collection by using a 0.45-mm MILLEX HV mem-

brane, and it was preserved at280C before use. Sweat samples were

diluted with saline (1/100 dilution) before the skin test. Histamine

contents of sweat samples (1/100 dilution) from healthy control

subjects and patients with CU were less than 10 nmol/L.

Skin test technique

Samples of autologous diluted sweat (0.02mL), autologous serum

(0.05 mL), and 0.9% sterile saline (0.02 or 0.05 mL) were separately

injected intradermally into the volar aspect of the forearm of each

subjects when they were quiet and with no wheal. The diameters of

wheals and erythema were measured after 15 minutes. Reactions

were assessed as positive if the diameter of the wheal induced by

sweat and serum was equal to or larger than 6 mm. The sterile saline-

induced wheals of all subjects were below 4 mm and 2 mm when the

amounts of 0.05 and 0.02 mL were injected, respectively.

Local provocation test

Responses to acetylcholine chloride (Ovisot; Daiichi, Tokyo,

Japan) were evaluated. Acetylcholine (0.1 mL) was intradermally

injected at a concentration of 100 mg/mL diluted with saline. The

development of satellite wheals around the injection site was con-

sidered as positive. Simultaneously, we checked sweating around

the injection site by the iodine-starch technique, and all patients

tested showed sweating by this test.10 All of the normal controls

tested showed a significant number of tiny sweating points by this

method.

Basophil histamine release test

A histamine release test was performed in vitro by using

HRT (Shionogi, Osaka, Japan) as previously described.11 Venous

peripheral blood samples from patients with CU and normal healthy

controls, 20 mL, and antibasophil antibodies conjugated to magnetic

beads were added to each well of a 96-well plate and incubated for

10 minutes at room temperature on a plate mixer. Antibody-binding

basophils in each well were then trapped with a chandelier-shaped

magnet and transferred to another microplate, where the basophils

were stimulated at 37C for 1 hour with autologous sweat, anti-IgE

antibody, and digitonin, respectively. Histamine released into the

medium was measured by an ELISA with a characteristic detection

profile.12


The statistical significance of differences was determined by using

the Student t test. Some data were analyzed by regression analysis

by using the statistical package StatView J (Abacus Concepts Inc,

Palo Alto, CA). A difference was considered statistically significant

at P < .05.

RESULTS

Skin tests for autologous sweat

Of 17 patients with CU, except for 1 patient whoshowed mechanical urticaria in skin test for autologoussweat, 11 (64.7%) showed positive reactions to their own1/100 diluted sweat by measuring the diameter of wheals(Table II). In contrast, all 10 healthy controls showednegative reactions to their own 1/100 diluted sweat,whereas a few healthy controls had positive reactions totheir own 1/10 diluted sweat (data not shown). Sweat-induced wheals in the skin tests were significantly greaterfor patients with CU than for healthy control subjects(Fig 1, A).

Sweat-induced histamine releasefrom basophils

We investigated the histamine release from basophils of17 patients with CU and of 10 healthy controls afterincubation with autologous sweat. Of 17 CU patients’basophils, 10 (58.8%) showed positive responses (morethan 5% histamine release) after incubation with 1/100diluted autologous CU sweat, whereas none of the 10healthy controls’ basophils did with 1/100 diluted autol-ogous normal sweat. Four of the CU patients’ basophilsshowed positive responses to 1/1000 diluted CU sweat.The overall values of percent histamine release frombasophils of patients with CU were significantly largerthan those from healthy control subjects (Fig 1, B).

Correlation of skin tests for autologous sweatwith sweat-induced histamine release frombasophils

We examined whether sweat-induced histamine releasefrom CU basophils correlates with skin tests for autolo-gous CU sweat in 16 patients. As shown in Fig 1, C,% histamine release correlated positively with the area inwheal using skin tests on 1/100 diluted sweat. Theseresults indicate that the degree of percent histamine releasefor autologous sweat represents the responsiveness of skintests for autologous sweat.

Autologous serum skin tests

Of 15 patients with CU, 8 (53.3%) showed a positiveresponse in the autologous serum skin test (ASST; TableII). In contrast, all 6 healthy controls showed a negativeresponse for ASST. Most patients with CU who had anegative response for ASST tended to show a positive


AUGUST 2005

398 Fukunaga et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

response for skin tests and for the histamine release testwith autologous CU sweat. In contrast, a few patients witha positive response for ASST tended to show hypersen-sitivity for sweat (Table II).

Intradermal acetylcholine test

After intradermal injections of relatively high concen-trations of cholinergic agents, the typical satellite pinpoint

wheals around the central large wheal were seen only inpatients with CU.4 However, these satellite wheals seemedto develop only in a few patients with CU.13 We thereforeexamined whether the patients with CU showed satellitewheals by acetylcholine injection. In this study, 6 (50%) ofthe 12 patients with CU tested showed a positive responsefor acetylcholine (Table II). Almost all of the patients whowere checked for sweating by iodine-starch method

TABLE I. Clinical characteristics of patients with cholinergic urticaria

Patient Age (y) Sex History Accompanying symptoms Total IgE (IU/mL) IgE RAST

1 26 Female Atopic dermatitis None 919 Mite, Candida

2 25 Female Atopic dermatitis Asthma 5518 Cedar, orchard grass

3 22 Male None Cold urticaria 432 Wheat

4 20 Male None None 85 Not done

5 21 Male Atopic dermatitis None 4080 Mite, Candida, wheat


7 22 Female Allergic rhinitis Cold urticaria 1842 Mite, Candida

8 19 Male None None Not done Not done


10 26 Male None None Not done Not done

11 31 Male Allergic rhinitis None 194 Not done

12 19 Female Atopic dermatitis Cold urticaria, angioedema 342 Mite

13 24 Female Atopic dermatitis None 907 Mite

14 15 Male None Cold urticaria 1302 Mite, Candida, orchard grass

15 26 Female None None 144 Negative

16 18 Female Atopic dermatitis None 118 Mite

17 19 Male Urticaria Cold urticaria Not done Not done

18 17 Male Allergic rhinitis None 423 Not done

TABLE II. Details of results for skin tests, histamine release test, acetylcholine test, and clinical chracterization

Autologous sweat skin test

% Histamine release

by sweatAutologous

serum skin test§

Acetylcholine

testkCharacteristics

of eruptionPatient (Erythema*) (Whealy) 1/100z 1/1000z

1 31 3 16 20 3 14 100 36.6 Negative Not done Nonfollicular

2 25 3 20 10 3 10 89.6 23.6 Negative Positive Nonfollicular

3 11 3 10 9 3 8 0.5 0 Negative Positive Nonfollicular

4 23 3 20 8 3 8 7.3 0.4 Negative Positive Nonfollicular

5 11 3 10 11 3 10 21.4 6.5 Negative Positive Undetermined

6 10 3 10 7 3 7 Not done Not done Negative Positive Nonfollicular

7 24 3 23 10 3 8 40 3.9 Not done Positive Nonfollicular

8 25 3 22 9 3 9 48.8 0.2 Not done Not done Nonfollicular

9 20 3 15 7 3 6 54.8 6.7 Positive Not done Undetermined

10 15 3 13 6 3 5 63.2 4.6 Positive Not done Undetermined

11 0 3 0 0 3 0 10.6 1.3 Positive Not done Follicular

12 10 3 8 10 3 8 2.1 0.2 Positive Negative Follicular

13 9 3 9 0 3 0 1 0 Positive Negative Follicular

14 Mechanical urticaria 0.2 0 Not done Not done Follicular

15 0 3 0 0 3 0 0.4 0 Positive Negative Follicular

16 7 3 6 0 3 0 0 0 Positive Negative Nonfollicular

17 0 3 0 0 3 0 0.4 0.4 Positive Negative Follicular

18 7 3 7 4 3 4 1.1 0 Negative Negative Nonfollicular

*Long axis and short axis of oval area are presented. 1/100 diluted sweat is used in autologous sweat skin test.

Dilution of sweat.

§Autologous serum was injected intradermally into the volar aspect of the forearm.

kAcetylcholine was intradermally injected 0.1 mL in concentration of 100 mg/mL diluted with saline.



Fukunaga et al 399

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

showed sweating after acetylcholine injection in ourseries, indicating that the absence of the satellite whealsby the agent is not attributed to the dyshidrosis. Comparedwith patients who showed a negative response for theacetylcholine test, the areas of CU sweat-induced whealsin the skin tests were significantly greater in patients whoshowed satellite wheals for the acetylcholine test (Fig 2,A). The values of CU sweat-induced histamine releasedfrom basophils of patients who showed positive responsesfor the acetylcholine test were significantly greater than

those from patients who showed negative responses forthe acetylcholine test (Fig 2, B). In addition, certainsatellite wheals after the acetylcholine test were recogniz-able as coincident with perspiration points when sweat-ing points were detected by the starch-iodine method(Fig 2, C).

Characterization of the clinical phenotype

When patients with CU developed wheals after exer-cise, we observed that the wheals sometimes coincidedwith hair follicles. Of 16 patients with CU, 6 (37.5%) hadwheals coincident with follicles (the follicular type), and 8(50%) had wheals that were not coincident with follicles(the nonfollicular type). Compared with the folliculartype, the areas of CU sweat-induced wheals in the skin testwere significantly greater in the nonfollicular type (Fig 3,A). The values of CU sweat-induced histamine releasedfrom basophils of the nonfollicular type tended to begreater than those released from basophils of the folliculartype (Fig 3, B). A representative clinical picture of whealsconsisting of follicles is shown (Fig 3, C).

DISCUSSION

We demonstrated that the majority of patients with CUare highly sensitive to autologous sweat. The heteroge-neous responses in skin tests with autologous sweatsuggest that patients with CU have various degrees ofhypersensitivity to sweat. We further observed that var-ious amounts of histamine were detected in the mediumwhen basophils were incubated in the presence of autol-ogous sweat, which suggests that autologous sweat itselfcontains factors that can induce histamine release. Theamounts of histamine released from CU basophils corre-lated relatively well with the degree of response in the skintests to autologous CU sweat. In contrast, normal healthycontrols did not respond to intracutaneous challenge withautologous sweat and did not show histamine release frombasophils by stimulation with autologous sweat. Theseresults indicate that patients with CU have various degreesof hypersensitivity to sweat and that in vitro histaminerelease tests using autologous sweat correlate with theintracutaneous test. Previously, Adachi et al5 reported thatall patients with CU examined showed immediate-typeskin reactions to intradermal tests with sweat at variousdilutions (20-29). We observed that a few healthy controlshad positive reactions after intradermal injection of 1/10autologous sweat, whereas no healthy controls showed apositive reaction to their own 1/100 diluted sweat. Certainpatients who showed a negative response at 1/100 dilutionmight have shown a positive response at higher concen-trations (1/10 and higher dilutions). In other words, thosepatients with CU who showed positive skin responses inthis study might represent the presence of strong hyper-sensitivity to sweat.

Interestingly, Hide et al14 recently reported that patientswith atopic dermatitis show hypersensitivity to autologoussweat antigens. They speculate that this phenomenon is

FIG 1. Differences in responses to sweat in patients with CU and

normal controls. A, The areas of wheals induced by intradermal

injection with autologous sweat. B, Values of the histamine release

from basophils stimulated with CU sweat or normal sweat. C, The

relationship of responsiveness of skin tests with CU sweat and CU

sweat-induced histamine release from CU basophils.


AUGUST 2005

400 Fukunaga et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

an IgE-mediated response, because histamine release wasimpaired by removal of IgE from patients’ basophils andmyeloma IgE blocked the sensitization of basophils withthe patient’s serum. It is of interest to note that patientswith CU frequently have atopic dermatitis.5,15 Adachiet al5 have shown that leukocytes from a normal healthydonor did release histamine on sweat challenging afterbeing sensitized with a patient’s serum. Therefore, itmight be possible that, similar to atopic dermatitis, hyper-sensitivity to sweat in patients with CU could be anIgE-mediated response.

An attractive hypothesis for the pathomechanisms ofCU is that sweat leaks from the sweat duct into thedermis.6,16 Several cases of CU have been described thatare associated with hypohidrosis/anhidrosis.6,16 Occlu-sion of the superficial acrosyringium might result in sweatleakage into the dermis in patients with CU and anhidro-sis.6 If those patients with CU are hypersensitive to sweat,the leaking sweat possibly induces urticarial symptomsaround the sweat ducts, resulting in small pinpoint wheals.Commens and Greaves4 examined 12 patients with CU byintradermal testing with methacholine and found thatsatellite wheals were induced in only 6 of them. It is not yetclear why only some patients with CU develop satellitewheals after injection of cholinergic agents. We showed

here that patients developing satellite wheals in theacetylcholine test had significantly enhanced responsesto sweat in the skin tests and in histamine release tests (Fig2). This means that those with hypersensitivity to sweattend to develop satellite wheals after stimulation withacetylcholine, a sweat inducer. Moreover, we observedthat satellite wheals after the acetylcholine test werecoincident with perspiration points by the iodine-starchmethod (Fig 2, C). These results are compatible with theidea that sweat leakage from sweat ducts induces smallwheals in certain patients with CU.

Circulating functional histamine-releasing autoanti-bodies reactive against either the a-subunit of the high-affinity IgE receptor (FceRIa) or IgE have been identifiedin more than one third of patients with chronic idiopathicurticaria.17-22 The ASST is now recognized as a suitablescreening test for such autoantibodies in such patients.4

However, it is still unclear whether the wheal-inducingfactors in the patient’s sera in this study are theseautoantibodies, and this issue should be further clarifiedin the future. Sabroe et al23 have reported that only 1 of 9patients with CU had a positive ASST. In contrast, weshowed here that 8 of 15 patients had a positive ASST. Thediscrepancy of the ratio of responsiveness in ASST be-tween their findings and ours might be attributed to thepatient population; that is, the majority of the patients intheir study might have had the nonfollicular type. So far, itis unclear whether we might enroll more follicular-type

FIG 2. The relationship between responses to sweat and acetyl-

choline tests in patients with CU.A, The areas of wheals induced by

intradermal injection with CU sweat in patients with CUs with or

without satellite wheals. B, The CU sweat-induced histamine

release from basophils in patients with CU with or without satellite

wheals. C, A representative clinical picture that satellite wheals are

coincident with perspiration points.

FIG 3. The difference of response to sweat in the relationship

between follicles and eruption in patients with CU. A, The areas of

wheals induced by intradermal injection with CU sweat in patients

with CU with nonfollicular or follicular wheals. B, The CU sweat-

induced histamine release from basophils in CU patients with

nonfollicular or follicular wheals. C, A representative picture of

wheals consisting of follicles.



Fukunaga et al 401

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

patients than the usual population in CU. Therefore, thisissue should be further studied in the future.

We observed that certain patients with CU developwheals in association with hair follicles, whereas the otherpatients do not. This phenomenon is similar to that seenin aquagenic urticaria, in which follicular wheals developafter contact with water. We found that patients with non-follicular-type CU tend not only to show satellite whealsafter the acetylcholine test but also to have hypersensitiv-ity to sweat as determined by skin tests and by histaminerelease tests (Table II). On the other hand, most of thepatients with follicular-type CU showed a positive reac-tion to ASST and no satellite wheals by acetylcholine orhypersensitivity to sweat (Table II). On the basis of thesefindings, we strongly believe that CU should be classifiedinto 2 subtypes from the clinical and pathological aspects.The relationship between CU and hair follicles should beexamined further.

In summary, 2 subtypes were identified in patients withCU. The first subtype shows nonfollicular wheals, ahypersensitivity to autologous sweat, satellite wheals inthe acetylcholine test, and negative reactions to autolo-gous serum. The second subtype shows follicular wheals,a very weak hypersensitivity to sweat, no satellite whealsin the acetylcholine test, and positive reactions to autol-ogous serum. Thus, we suggest that the pathogenesis ofCU involves hypersensitivity or autoimmunity to sweat. Inconsidering the pathogenesis of CU, the classificationpresented here may be useful to this unique disorder.

REFERENCES

1. Duke WW. Urticaria caused specifically by the action of physical agents.

JAMA 1924;83:3-9.

2. Kaplan AP, Natbony SF, Tawil AP, Fruchter L, Foster M. Exercise-

induced anaphylaxis as a manifestation of cholinergic urticaria. J Allergy

Clin Immunol 1981;68:319-24.

3. Lawrence CM, Jorizzo JL, Kobza-Black A, Coutts A, Greaves MW.

Cholinergic urticaria with associated angio-oedema. Br J Dermatol 1981;

105:543-50.

4. Commens CA, Greaves MW. Tests to establish the diagnosis in

cholinergic urticaria. Br J Dermatol 1978;98:47-51.

5. Adachi J, Aoki T, Yamatodani A. Demonstration of sweat allergy in

cholinergic urticaria. J Dermatol Sci 1994;7:142-9.

6. Kobayashi H, Aiba S, Yamagishi T, Tanita M, Hara M, Saito H, et al.

Cholinergic urticaria, a new pathogenic concept: hypohidrosis due to

interference with the delivery of sweat to the skin surface. Dermatology

2001;204:173-8.

7. Kaplan AP, Gray L, Shaff RE, Horakova Z, Beaven MA. In vivo studies

of mediator release in cold urticaria and cholinergic urticaria. J Allergy

Clin Immunol 1975;55:394-402.

8. Sigler RW, Levinson AL, Evans R III, Horakova Z, Kaplan AP.

Evaluation of patients with cold and cholinergic urticaria. J Allergy


9. Kobza Black A, Lawlor F, Greaves MW. Consensus meeting on the

definition of physical urticarias and urticarial vasculitis. Clin Exp

Dermatol 1996;21:424-6.

10. Wada M, Takagaki T. A simple and accurate method for detecting the

secretion of sweat. Tohoku J Exp Med 1948;49:284.

11. Adachi A, Fukunaga A, Hayashi K, Kunisada M, Horikawa T. Anaphy-

laxis to polyvinylpyrrolidone after vaginal application of povidone-

iodine. Contact Dermatitis 2003;48:133-6.

12. Nishi H, Nishimura S, Higashiura M, Ikeya N, Ohta H, Tsuji T, et al.

A new method for histamine release from purified peripheral blood

basophils using monoclonal antibody-coated magnetic beads. J Immunol

Methods 2000;240:39-46.

13. Commens CA, Greaves MW. Tests to establish the diagnosis in

cholinergic urticaria. Br J Dermatol 1978;98:47-51.

14. Hide M, Tanaka T, Yamamura Y, Koro O, Yamamoto S. IgE-mediated

hypersensitivity against human sweat antigen in patients with atopic

dermatitis. Acta Derm Venereol 2002;82:335-40.

15. Freedberg IM, Eisen AZ, Wolff K, Austen KF, Goldsmith LA, Katz SI.

Fitzpatrick’s dermatology in general medicine. 6th ed. New York:

McGraw-Hill; 2003. p. 1132-3.

16. Itakura E, Urabe K, Yasumoto S, Nakayama J, Furue M. Cholinergic

urticaria associated with acquired generalized hypohidrosis: report of a

case and review of literature. Br J Dermatol 2000;143:1064-6.

17. Hide M, Francis DM, Grattran CE, Hikimi J, Kochan JP, Greaves MW.

Autoantibodies against the high-affinity IgE receptor as a cause of

histamine release in chronic urticaria. N Engl J Med 1993;328:

1599-604.

18. Niimi N, Francis DM, Kermani F, O’Donnell BF, Hide M, Kobza Black

A, et al. Dermal mast cell activation by autoantibodies against the high

affinity IgE receptor in chronic urticaria. J Invest Dermatol 1996;106:

1001-6.

19. Fiebiger E, Maurer D, Holub H, Reininger B, Hartmann G, Woisetschl-

ager M, et al. Serum IgG autoantibodies directed against the alpha

chain of Fc epsilon RI: a selective marker and pathogenetic factor for a

distinct subset of chronic urticaria patients? J Clin Invest 1995;96:

2606-12.

20. Tong LJ, Balakrishnan G, Kochan JP, Kinet JP, Kaplan AP. Assessment

of autoimmunity in patients with chronic urticaria. J Allergy Clin

Immunol 1997;99:461-5.

21. Zweiman B, Valenzazo M, Atkins PC, Tanus T, Getsy JA. Character-

istics of histamine-releasing activity in the sera of patients with chronic

idiopathic urticaria. J Allergy Clin Immunol 1996;98:89-98.

22. Ferrer M, Kinet JP, Kaplan AP. Comparative studies of functional and

binding assays for IgG anti-Fc(epsilon)RIalpha (alpha-subunit) in

chronic urticaria. J Allergy Clin Immunol 1998;101:672-6.

23. Sabroe RA, Grattan CEH, Francis DM, Barr RM, Black AK, Graves MW.

The autologous serum skin test: a screening test for autoantibodies in

chronic idiopathic urticaria. Br J Dermatol 1999;140:446-52.


AUGUST 2005

402 Fukunaga et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

Lack of detectable allergenicity of transgenicmaize and soya samples

Rita Batista, BSc,a,b Baltazar Nunes, MSc,aManuela Carmo,a Carlos Cardoso, PharmD,c

Helena Sao Jose,c Antonio Bugalho de Almeida, MD, PhD,d Alda Manique, MD,d

Leonor Bento, MD, PhD,e Candido Pinto Ricardo, PhD,b,f and Maria Margarida

Oliveira, PhDb,g Lisboa, Oeiras, and Alges, Portugal

Background: The safety issues regarding foods derived from

genetically modified (GM) plants are central to their

acceptance into the food supply. The potential allergenicity

of proteins newly introduced in GM foods is a major safety

concern.

Objective: We sought to monitor, in potentially sensitive

human populations, the allergenicity effects of 5 GM

materials obtained from sources with no allergenic potential

and already under commercialization in the European Union.

Methods: We have performed skin prick tests with protein

extracts prepared from transgenic maize (MON810, Bt11, T25,

Bt176) and soya (Roundup Ready) samples and from

nontransgenic control samples in 2 sensitive groups: children

with food and inhalant allergy and individuals with asthma-

rhinitis. We have also tested IgE immunoblot reactivity of sera

from patients with food allergy to soya (Roundup Ready) and

maize (MON810, Bt11, Bt176) samples, as well as to the pure

transgenic proteins (CryIA[b] andCP4 5-enolpyruvylshikimate-

3-phosphate synthase).

Results: None of the individuals undergoing tests reacted

differentially to the transgenic and nontransgenic samples

under study. None of the volunteers tested presented detectable

IgE antibodies against pure transgenic proteins.

Conclusion: The transgenic products under testing seem to be

safe in terms of allergenic potential. We propose postmarket

testing as an important screening strategy for putative allergic

sensitization to proteins introduced in transgenic plants.

(J Allergy Clin Immunol 2005;116:403-10.)

Key words: Transgenic food, allergenicity, immune response,

public health, food safety, recombinant DNA technology

Recombinant DNA technology or genetic engineeringallows the transfer of single genes from one organism toanother, even if distantly related, a feat impossible throughconventional plant breeding. As a result, a geneticallymodified organism (GMO) will contain a modified oradditional trait encoded by the introduced gene or genes,which generally results in additional proteins.

Potential benefits for world agriculture derived fromGMOs could be enormous, including the possibility ofproducing higher yields of more nutritious food in moresustainable regimens.1-5

With the development of the new modification tech-niques, there is the increasing concern of emergence ofnew food allergies. An example of such a situation is theBrazil nut allergen (2S protein), which when overex-pressed in soybean was found to retain its allergenicityand was therefore never commercialized.6

Food allergy is a term that should be used to describeadverse reactions to certain foods because of immunologicmechanisms.7Themajority of individualswithdocumentedimmunologic reactions to foods exhibit IgE-mediatedhypersensitivity reactions that can be sudden, severe, andlife-threatening.8 The best estimates are that IgE-mediatedfood allergies affect approximately 1% to 2% of the adultpopulation9,10; in children this value is estimated to be2%to8%.11,12

Before market introduction, genetically modified (GM)food products are subjected to extensive assessment ofpotential effects to human health, including toxicity andpotential allergenicity. When the gene source is an aller-genic food, in vitro and clinical tests are available to assessthe allergenicity of the transferred protein or proteins.However, most genes transferred through genetic engi-neering are obtained from organisms with no allergenichistory. In such cases the assessment of allergenicitybecomes more difficult to obtain because of the absence ofvalid methods and models.13-16

Abbreviations usedBt: Bacillus thuringiensis

EPSPS: 5-Enolpyruvylshikimate-3-phosphate synthase

GM: Genetically modified

GMO: Genetically modified organism

PAT: Phosphinotricine acetyl transferase

RUR: Roundup Ready


From aInstituto Nacional de Saude Dr Ricardo Jorge, Lisboa; bInstituto de

Tecnologia Quımica e Biologica/Instituto de Biologia Experimental e

Tecnologica, Oeiras; cClınica Medica e de Diagnostico Dr Joaquim

Chaves, Alges; dClınica Universitaria de Pneumologia do Hospital de

Santa Maria, Lisboa; eDepartamento de Clınica Pediatrica do Hospital de

Santa Maria, Lisboa; fInstituto Superior de Agronomia, Tapada da Ajuda,

Lisboa; and gDepartamento Biologia Vegetal, Faculdade de Ciencias de

Lisboa, Lisboa.

Supported by Fundacxao Calouste Gulbenkian, research project

SDH.SP.I.01.11 and by Comissao de Fomento da Investigacxao em

Cuidados de Saude, research project no. 186/01.

Received for publication January 11, 2005; revised March 22, 2005; accepted



Reprint requests: Rita Batista, BSc, Instituto Nacional de Saude Dr Ricardo

Jorge, Av Padre Cruz, 1649-016 Lisboa, Portugal; E-mail: rbatista@

itqb.unl.pt.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.014

403

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis



In this study we have monitored the IgE response ofallergy-sensitive populations to GM maize and soyaproducts (Table I). The transgenes in maize and soyawere obtained from sources with no allergenic history andapproved for human consumption in the European Union.The IgE response of the same individuals to nonmodifiedproducts was also analyzed for comparison.

METHODS

This study was evaluated and approved by the Research Ethic

Committees of the Hospital of Santa Maria and the National Institute

of Health, Lisbon, Portugal. All individuals participating in this study

or their parents also provided informed consent.

Food inquiry

Because of the fact that IgE-mediated allergic reactions require

prior exposure, resulting in sensitization, we have performed a food

inquiry to evaluate the consumption of soya and maize food-derived

products. Bearing in mind that since 1998 all the GM products

under testing were approved for commercialization in the European

Union (Table I), we assumed that consumption of maize and soya

food-derived products implied a consumption of GM soya and

maize.

The food inquiry was performed on 106 healthy volunteers to find

out which maize- and soya-derived products (from a list of 205 dif-

ferent products) they had already consumed. The population studied

included individuals with ages from 1 to 41 years, with an average of

12.4 years (48 male and 58 subjects).

Transgenic quality of the noncertifiedflour samples

In addition to the 3 noncertified transgenic products listed in

Table I, nontransgenic analogues were also tested as controls. For

the noncertified material (Table I), we have first confirmed the

transformation event and the absence of cross-contamination among

them.

For these analyses, DNAwas isolated by using the cetyltrimethyl-

ammonium bromide method,17 with 3 replicas per sample. DNA

quality and concentration were analyzed by means of agarose gel

electrophoresis, and maize-specific amplifiable DNA was detected

by using PCR amplification of a 226-bp sequence from the maize

invertase gene.18

The presence or absence of the 35S Cauliflower Mosaic Virus

promoter in the transgenic (Table I) and control samples was checked

TABLE I. Transgenic flour products tested in SPTs and in IgE Immunoblot reactivity assays

Material Characteristics

Date of

commercialization

Responsible

company

Origin and

certification of

the material

Method of testing

and human

population studied

2% GM

Bt11 maize

Insect resistance (CryIA[b]

gene) and ammonium

glufosinate tolerance

(PAT gene); 35S pro;

NOS 3# t

1998 Syngenta Institute of Reference

Materials and

Measurements

certified

SPTs in a population

of allergic children

(27 individuals);

IgE immunoblot

reactivity assay with sera

from patients with food

allergy (57 individuals)

100% GM

Bt176 maize


gene) and ammonium

glufosinate tolerance

(PAT gene); 35S pro; 35S t

1997 Syngenta

National Service of

Plant Protection

(DGPC); not certified

SPTs in a population

of allergic children

(27 individuals);

IgE immunoblot

reactivity assay with

sera from patients

with food allergy

(57 individuals)

100% GM

T25 maize

Ammonium glufosinate

tolerance (PAT gene);

35S pro; 35S t

1998 Bayer Crop

Sciences

SPTs in a population of

patients with asthma-

rhinitis (50 individuals)

100% GM

MON810

maize


gene); 35S pro; NOS3# t1998 Monsanto SPTs in a population

with asthma-rhinitis

(50 individuals); IgE

immunoblot reactivity

assay with sera from

patients with food allergy

(24 of the 57 individuals)

5% GM

RUR soya

Gliphosate resistance

(CP4EPSPS gene);

35S pro; NOS 3# t

1996 Monsanto Institute of Reference

Materials and

Measurements

certified

SPTs in a population of

allergic children (27

individuals); IgE

immunoblot reactivity

assay with sera from

patients with food allergy

(57 individuals)

35S pro, 35S Cauliflower Mosaic Virus promoter; 35S t, 35S Cauliflower Mosaic Virus terminator; NOS 3# t, Agrobacterium tumefaciens nopaline synthase

terminator; DGPC, Direccxao Geral de Proteccxao de Culturas.


AUGUST 2005

404 Batista et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

by using standard protocols for the amplification of a 195-bp DNA

sequence.18

Transformation event–specific PCR reactions were performed to

verify the presence of MON810, T25, and Bt176 transgenic events.18

Different internal controls were always used to detect putative

contaminations. In each case whole or digested (HaeIII or Hinf I)PCR product size was compared with expected values.18

Preparation of protein extracts forhuman skin prick testing andIgE immunoblot reactivity

Maize and soya protein extracts were made by Laboratorios Leti,

SL (Madrid, Spain) according to approved pharmaceutical prepara-

tive and safety procedures for the production of diagnostic skin prick

test (SPT) materials. About 10 g of each of the maize and soya flour

samples was extracted for 16 hours in 1:20 (wt/vol) PBS (pH 7.4).

After centrifugation, the pellet was discarded, and the supernatant

was extensively dialyzed against bidistilled water. The extracts were

centrifuged, filter sterilized, and freeze-dried. For human SPTs, the

extracts were resuspended to 10 mg/mL maize or soya freeze-dried

material (approximately 2mg of total protein/mL forMON810, Bt11,

Bt176, and control samples; approximately 3 mg of total protein/mL

for T25 and control samples; and approximately 3.5 mg of total

protein/mL for Roundup Ready [RUR] and control samples).

For the IgE immunoblot reactivity assay conducted with sera from

patients with food allergy, we used an extract prepared with food to

which the person undergoing the test was allergic as a positive control

extract. Four grams of food material was homogenized in liquid

nitrogen and precipitated with 20 mL of 10% trichloroacetic acid

(wt/vol) in cold acetone containing 20 mM dithiothreitol for 1 hour at

220C. The precipitate was collected by means of centrifugation

(15 minutes at 14,000g at 4C), washed twice with 20 mM dithio-

threitol in cold acetone, and allowed to dry completely.

Quality of transgenic proteins in maizeand soya extracts

ELISA GMO Check Bt maize test kit (SDI Europe, London,

United Kingdom) was used to evaluate the presence-absence of

Bt CryIA(b) protein in the lyophilized extracts prepared by

Laboratorios Leti. Ten milligrams of dry extract was resuspended

in 200 mL of the kit extraction buffer provided, and all nonsoluble

material was removed by means of centrifugation (10 minutes at

11,000g). The manufacturer’s instructions were followed thereafter,

using approximately 200 mg of total protein.

To evaluate the presence or absence of CP4 5-enolpyruvylshiki-

mate-3-phosphate synthase (CP4EPSPS) protein in RUR soya and

nontransgenic analogues, the lyophilized materials were tested with

an ELISA GMO Check RUR Soya Grain test kit (Strategic Diag-

nostics Inc). Five milligrams of dry extract was diluted in 200 mL of

kit extraction buffer. The nonsoluble material was removed bymeans

of centrifugation (10 minutes at 11,000g). The manufacturer’s

instructions were followed thereafter, using approximately 2.5 mg

of total protein.

Thirty micrograms of each sample was also run by means of

SDS-PAGE and immunobloted with rabbit anti-Bt CryIA(b) poly-

clonal antibodies (RDI, Flanders, NJ) or goat anti-CP4EPSPS serum

(Monsanto Co, St Louis, Mo; see description below).

It was impossible to obtain commercial anti-phosphinotricine

acetyl transferase (anti-PAT) antibodies, and there is no commercially

available ELISAkit for PAT.We therefore decided to use the Trait LL

corn grain test kit (Strategic Diagnostics Inc) to evaluate the presence

or absence of PAT inBt176, Bt11, T25, and nontransgenic analogues.

This kit uses PAT-specific antibodies coupled to a color reagent and

incorporated into strips, allowing the detection of PAT in an extract

through color development. Ten milligrams of lyophilized samples

was diluted in the kit buffer provided, and 100 mL of each sample

(approximately 200 mg of total protein) was eluted along the strip.

For protein quantification, the Bio-Rad protein assay (Bio-Rad

laboratories) was used, with turkey albumin (Merck) as a standard.

Skin testing of the 2 populations

Skin tests were performed in 2 human populations with positive

histories of food allergy, inhalant allergy, or both, as well as a

positive SPT response for related allergens; one group was composed

of 27 children with food and inhalant allergy from the Paediatrics

Allergy Department of the Hospital of SantaMaria, and the other was

composed of 50 patients with asthma-rhinitis from the University

Clinic of Pneumology from the Hospital of Santa Maria (see Tables

E1 and E2 in the Online Repository in the online version of this

article at www.mosby.com/jaci). The children were tested with the

extracts of Bt176, Bt11, RUR, and nontransgenic analogues; for the

asthma-rhinitis population, we used the extracts of MON810, T25,

and nontransgenic analogues for testing (Table I).

Skin tests were performed by using the prick procedure,19 and

results were read after 20 minutes. The results were classified as

positive when the larger diameter of the wheal exceeded 3 mm.

Histamine hydrochloride, 10 mg/mL (Leti), was used as a positive

control, and Phenolate saline serum with glycerine (Leti) was used as

a negative control.

All the protein extracts were first tested on a control population of

20 nonallergic healthy individuals.

Sera for the IgE immunoblot reactivity assay

Patient sera were provided by the JoaquimChaves Clinic andwere

obtained from 57 individuals who had a positive history of docu-

mented food allergy, as well as a positive value equal to or higher than

class 3 on specific UniCAP test (Pharmacia Diagnostics, Seixal,

Portugal; see Table E3 in the Online Repository in the online version

of this article at www.mosby.com/jaci). All 57 sera were first assayed

for reactivity against nontransgenic maize and soya by means of

specific IgE UniCAP testing. The sera were then tested for IgE

immunoblot reactivity against Bt11, Bt176 maize, and RUR soya, as

well as against nontransgenic analogues (Table I). MON810 maize

and its nontransgenic analogue, as well as pure CryIA(b) (Research

Diagnostics, Inc) and CP4EPSPS (Monsanto Co), were used to test

the IgE immunoblot reactivity of sera of the 24 more sensitive

patients (Table I).

SDS-PAGE and protein transfer tonitrocellulose membranes

Samples were diluted 1:2 in sample buffer (0.125M Tris-HCl [pH

6.8], 4% SDS, 20% vol/vol glycerol, 0.2 M dithiothreitol, and 0.02%

bromophenol blue) and boiled for 5 minutes before electrophoresis

in a 0.75-mm-thick 10% acrylamide gel with 4% stacking gel.20 After

electrophoresis, the proteins were blotted onto hybond ECL nitro-

cellulose membranes (Amersham Biosciences, Carnaxide, Portugal)

by means of wet transfer in 25 mMTris, 192 mM glycine, 0.1% SDS,

and 20% methanol for 1 hour at 75V at room temperature.

IgE immunoblot reactivity assay of serafrom patients with food allergy

The detection of patient sera IgE reactivity was carried out after

electrophoresis of 30 mg (60 mg/cm gel width) of MON810, Bt11,

Bt176, and RUR transgenic samples and nontransgenic analogues

and 25 ng (50 ng/cm gel width) of pure CryIA(b) and CP4EPSPS and

transfer to nitrocellulose membrane.

Blots were blocked overnight at 4C with PBS-T (58 mM

Na2HPO4, 17 mM NaH2PO4.H2O, 68 mM NaCl, and 0.2% Tween



Batista et al 405

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

20) and 5% skimmedmilk powder (or 3%BSA for patients with milk

allergy) and washed with PBS-T before incubation in serum diluted

1:10 in blocking solution for 1 hour and 30 minutes at room

temperature. After washing with PBS-T, the membranes were

incubated for 1 hour at room temperature in alkaline phosphatase–

conjugated monoclonal anti-human IgE (Southern Biotechnology

Associates, Birmingham, Ala) diluted 1:2000 in blocking solution,

washed with PBS-T and assay buffer, and incubated for 5 minutes

with CDP-Star solution with Nitro-Block II enhancer (Tropix

Western-Star Immunodetection System).

Blots were observed after exposure (5 seconds-30 minutes) to a

high-performance chemiluminescence Hyperfilm ECL (Amersham

Biosciences).

Immunoblot detection of Bt CryIA(b)and CP4EPSPS

The procedure was identical to the one described for IgE

immunobloting of patient’s sera, with the following differences. For

Bt CryIA(b), the first antibody incubation was performed in rabbit

anti-Bt CryIA(b) polyclonal (Research Diagnostics, Inc) diluted

1:1400 in blocking solution, and the second antibody incubation

was performed in goat anti-rabbit IgG-AP conjugate (Tropix-Applied

Biosystems, Porto, Portugal) diluted 1:2800 in blocking solution. For

CP4EPSPS, the first antibody incubation was performed in goat anti-

CP4EPSPS serum (Monsanto Co) diluted 1:5000 in blocking solu-

tion, and the second antibody incubation was performed in anti-goat

IgG-alkaline phosphatase conjugate (Sigma, Sintra, Portugal) diluted

1:2500 in blocking solution.


To estimate the probability of one individual from the Portuguese

population having once been in contact with transgenic proteins

present in maize or soya foods, we used (1) the results from the food

inquiry and (2) the percentage data of maize and soya products with

detectable transgenic proteins provided by Instituto de Biologia

Experimental e Tecnologica Good Laboratory Practices Microbiol-

ogy laboratory. This laboratory is one of the 2 national laboratories

responsible for food GMO detection.

Assuming that the number of products with maize or soya

consumed by the population is a Poisson random variable with the

expected value l and that the probability of an individual having

consumed a product with transgenic proteins (provided he or she had

consumed n products with maize or soya) is modeled by using

binomial distributions21 (n = number of experiences, p = probability

of one product withmaize or soya having transgenic proteins), we can

then calculate the probability of one individual having been in contact

with transgenic proteins, which is 12e2lp.

To estimate this probability, we used as l the mean number of

consumed products with maize or soya obtained in the survey, and as

p the proportion of maize and soya products detected with transgenic

proteins calculated by using the Instituto de Biologia Experimental e

Tecnologica Good Laboratory Practices Microbiology laboratory

data during the last 2 years.

RESULTS

Food inquiry

All 106 individuals participating in this inquiry con-sumed some of the 205 products presented. The extremecases, with lower and higher numbers of consumedproducts, were relative to a 1-year-old and 9-year-oldgirl with 4 and 129 consumed products, respectively.The mean of consumed products with maize and soyawas 39.3 (95% CI, 35.2-43.4), and the probability ofan individual having eaten GM food was near 100%(Table II).

Transgenic quality of the noncertifiedflour samples

All 6 tested samples (Bt176, T25, MON810, and thenontransgenic analogues) showed the expected bandswhen checking for amplifiable maize DNA (data notshown), and only the 3 transgenic samples showed theexpected amplicon of the 35S promoter (data not shown).

The final confirmation that all the samples tested werecorrectly labeled and that there was no cross-contamina-tion among them was obtained from construct-specificPCR (Fig 1). As expected, the digestion of the obtainedamplicons confirmed the accuracy of the specific PCR(data not shown).

Quality of transgenic proteins in maizeand soya extracts

As described in the Methods section, LaboratoriosLeti protein extracts were tested for the presence of the

TABLE II. Results of the food inquiry regarding the probability of an individual having consumed a transgenic maize

or soya sample

Mean number of consumed

products with maize or soya

(l estimates)

Probability of consumption of products

with transgenic protein

n 95% CI P = .235 95% CI

Total 106 39.3 35.2-43.4 0.999902 0.99974-0.99999

Sex

Male 48 34.8 29.2-40.4 0.999959 0.99895-0.99993

Female 58 43.0 37.1-48.2 0.999719 0.99983-0.99999

Age group (y)

<5 20 29.5 22.6-36.3 0.999024 0.99506-0.99980

5-10 56 41.1 35.2-46.9 0.999936 0.99974-0.99998

10-25 11 48.8 35.6-62.0 0.999990 0.99976-1.00000

25 19 38.9 27.5-50.3 0.999893 0.99844-0.99999

n, Number of valid responses; P, probability of one product with maize or soya having transgenic proteins (Instituto de Biologia Experimental e

Tecnologica Good Laboratory Practices Microbiology laboratory data).


AUGUST 2005

406 Batista et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

transgenic proteins under testing. CryIA(b) was detectedby using ELISA (data not shown) and Western blotting(Fig 2) in MON810, Bt11, and Bt176 extracts andwas absent from the nontransgenic control analogues.CP4EPSPS was also detected by means of ELISA (datanot shown) and Western blotting (Fig 3) in RUR ex-tract and was absent from the nontransgenic analogue.Both pure CryIA(b) and CP4EPSPS proteins were de-tected with the respective specific antibodies (data notshown).

In T25, Bt11, and Bt176 samples PAT protein wasdetected in 200 mg of total protein solutions by using the

Trait LL corn grain test kit. With this system, we have alsoconfirmed the absence of PAT in nontransgenic analogues(data not shown).

Allergenicity tests

Skin testing of the 2 populations. Only individualswith maize sensitivity, soybean sensitivity, or both hadpositive results against the protein extracts under testing;however, none of the volunteers reacted differentially toGM versus non-GM samples (Table III and Tables E1 andE2 in the Online Repository in the online version of thisarticle at www.mosby.com/jaci).

FIG 1. Construct-specific PCR for the detection of modified DNA sequences from T25, Bt176, and MON810

maize. M, 100-bp DNA ladder; MM, Mastermix; Bl, DNA extraction blank; T252, Bt1762, MON8102, non-GM

controls; T251, Bt1761, MON8101, 100% GM T25, Bt176, and MON810 maize, respectively.

FIG 2.Western blot for the detection of CryIA(b) protein in Laboratorios Leti protein extracts. I, 10%Acrylamide

SDS-PAGE; II, immunoblot with rabbit anti-Bt CryIA(b) polyclonal. M, Molecular weight marker; Bt112,

MON8102, Bt1762, non-GM controls; Bt111, MON8101, Bt1761, GM material 2% Bt11, 100% MON810, and

100% Bt176, respectively.



Batista et al 407

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

All the patients had wheals larger than 3 mm forhistamine, and none of them reacted against the negativecontrol.

IgE Immunoblot reactivity assay of sera frompatients with food allergy. Two types of Westernblotting results (Figs 4 and 5) were observed. In Fig 4serum from an individual with octopus allergy of class 4on specific UniCAP testing reacted only against positivecontrols. In Fig 5, concerning an individual with peanutallergy of class 6 on a specific UniCAP test, positivesignals were observed against the positive control but alsoagainst other maize and soya protein extracts.

None of the volunteers tested presented differentialsignals against nontransgenic versus transgenic proteinextracts (Table III). All 24 individuals tested against puretransgenic proteins (CP4EPSPS and CryIAb) presented nodetectable reactions against these controls.

DISCUSSION

Although absolute certainties regarding GM food risksto health and the environment will hardly be obtained,reports regarding potential problems have raised publicconcern. Some of the concerning issues include theputative toxicity-allergenicity of crops expressing foreignproteins,22-25 although these fears have not been con-firmed in some later studies,26,27 and the adequacy of the

methods of testing have been questioned.28 Consideringthat the past few decades have witnessed a significantincrease in IgE-mediated allergic diseases, the allergenicpotential of these novel foods is a major concern in publichealth.

The food inquiry performed in this study indicated thatthe probability of an individual having eaten GM food wasnear 100% (Table II). This value is probably underesti-mated because each individual probably consumed eachproduct several times, which was not considered in statis-tical calculations.

Also, it is possible that the first sensitization occurredduring breast-feeding in the individuals submitted to SPTsand Western blot analyses who were younger than 6 years(the time between the first commercialization in 1998 and2004).29 It therefore seems reasonable to assume that allthe individuals participating in this study had already beenin contact with the products tested.

The DNA and protein quality analysis performed in thisstudy confirmed the quality of flour samples and maizeand soya protein extracts (Figs 1-3). In the Western assayfor the detection of CryIA(b) in Bt11, Bt176, MON810,and nontransgenic control analogues (Fig 2), the multiplebands approximately equal to the CryIA(b) trypsin resis-tant core observed are likely the products of endogenousgrain proteases.30 Some of the protein is degraded furtherto produce lower-molecular-weight bands, including a30-kd product previously reported.30

As already mentioned, we have performed this study onsensitive populations. The population submitted to SPTsand immunoblot analyses was composed of individualswith food allergy and inhalant allergy, many of themchildren. Children are more susceptible to food allergiesthan adults. This higher susceptibility is probably theresult of immunologic immaturity and, to some extent,immaturity of the gut.31,32 In addition, children who havepreexisting food allergies are more likely to experienceallergic reactions to other foods introduced in their diets.

The absence of detectable differences in IgE reactivitybetween GMmaize and soya samples and the correspond-ing wild-type samples obtained in this study is in accor-dance with some previously published results.33,34

The appearance of nondifferential bands on somechemiluminescence films for maize and soya protein

FIG 3. Western blot for the detection of CP4EPSPS protein in Laboratorios Leti protein extracts. I, 10%

Acrylamide SDS-PAGE; II, immunoblot with anti-CP4EPSPS goat serum (IgG). M, Molecular weight marker;

Bt112, Bt1762, RUR2, non-GM controls; Bt111, Bt1761, RUR1, GM material 2% Bt11, 100% Bt176, and

5% RUR, respectively.

TABLE III. Results obtained with the allergenicity

tests performed by using SPTs and IgE immunoblot

reactivity assays

SPTs

IgE immunoblot

reactivity assay

No. of

individuals

tested

Positive

responses

(%)

No. of

individuals

tested

Positive

responses

(%)

GM protein

PAT 77 0 NT 2

CRY1A(b) 77 0 57 0

CP4EPSPS 27 0 57 0

NT, Not tested.


AUGUST 2005

408 Batista et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

extract lanes (Fig 5) might be due to the phenomenon ofcross-reactivity among various plant and animal pro-teins.35,36 In the example presented in Fig 5, althoughthe patient tested had only documented peanut allergy(class 6 on UniCAP test), it was shown that he also hadIgE binding to other foods, such as almond (class 2),hazelnut (class 3), walnut (class 2), cashew (class 4),soybean (class 3), and maize (class 3). This fact justifiesthe appearance of the nondifferential bands on maize andsoya lanes.

Although IgE detection (either SPT or specific IgE)serves as a good indicator of sensitization but not neces-sarily of disease, in the clinical setting the absence ofdetectable IgE was found to have excellent negative pre-dictive accuracy indices and therefore might be veryuseful in excluding the presence of immediate food hyper-sensitivity.37

In this study we did not obtain any differential positiveresults, which allows us to conclude that the transgenicproducts under testing seem to be safe regarding theirallergenic potential. Although we succeeded in integratinga private clinic and a hospital in this study, it would bedesirable to increase the size of the analyzed populationand eventually extend this work to other countries.

We also propose the development and use of clinicaltesting with specific IgE in the postmarketing surveillanceof foods produced through biotechnology. Positive testresults should be followed by double-blind, placebo-controlled food challenges under appropriate clinicalobservation to identify true clinical reactions.38

We gratefully acknowledge the National Service of Plant

Protection (DGPC) for providing BT176, T25, MON810, and

nontransgenic analogue maize samples; Laboratorios Leti for the

FIG 4. IgE antibody reactivity assay from an octopus-sensitive patient. I, 10% Acrylamide SDS-PAGE;

II, immunoblot. M, Molecular weight marker; Bt112, Bt1762, RUR2, MON8102, non-GM controls; Bt111,

Bt1761, RUR1, MON8101, GM material 2% Bt11, 100% Bt176, 5% RUR, and 100% MON810, respectively;

Otp, Octopus protein extract; Cry, CryIA(b); CP4, CP4EPSPS.

FIG 5. IgE antibody reactivity assay from a peanut-sensitive patient. I, 10% Acrylamide SDS-PAGE;

II, immunoblot. M, Molecular weight marker; Bt112, Bt1762, RUR2, non-GM controls; Bt111, Bt1761,

RUR1, GM material 2% Bt11, 100% Bt176, and 5% RUR, respectively; Pnt, Peanut protein extract.



Batista et al 409

Foodallerg

y,derm

ato

logic

disease

s,andanaphylaxis

preparation of maize and soya protein extracts; and Monsanto,

especially Dr Richard Goodman, for providing the CP4EPSPS

protein and the corresponding antiserum. Fernanda Spınola and

Catia Peres are gratefully acknowledged for their advice regarding

GMO detection. We also thank Margarida Santos, Helena Raquel,

Madalena Martins, and Sara Silva for help in the preparation of the

food inquiry. Finally, we thank Phil Jackson for the final revision of

the manuscript and Fundacxao Calouste Gulbenkian and Comissao de

Fomento da Investigacxao em Cuidados de Saude for funding.

REFERENCES

1. Alam MF, Datta K, Abrigo E, Vasquez A, Senadhira D, Datta SK.

Production of transgenic deepwater indica rice plants expressing a

synthetic Bacillus thuringiensis cryIA(b) gene with enhanced resistance

to yellow stem borer. Plant Sci 1998;135:25-30.

2. Shintani D, Della Penna D. Elevating the vitamin E content of plants

through metabolic engineering. Science 1998;282:2098-100.

3. De la Fuente JM, Ramirez Rodriguez V, Cabrera Ponce JL, Herrera

Estrella L. Aluminium tolerance in transgenic plants by alteration of

citrate synthesis. Science 1997;276:1566-8.

4. Ye X, Al Babili S, Kloeti A, Zhang J, Lucca P, Beyer P, et al.

Engineering the provitamin A (beta-carotene) biosynthetic pathway into

(carotenoid-free) rice endosperm. Science 2000;287:303-5.

5. Goto F, Yoshihara T, Shigemoto N, Toki S, Takaiwa F. Iron fortification

of rice seed by soybean ferritin gene. Nat Biotechnol 1999;17:282-6.

6. Nordlee JA, Taylor SL, Townsend JA, Thomas LA, Bush RK. Identi-

fication of a Brazil-nut allergen in transgenic soybeans. N Engl J Med

1996;334:688-92.

7. Matsuda T, Nakamura R. Molecular structure and immunological

properties of food allergens. Trends Food Sci Technol 1993;4:289-93.

8. Taylor SL, Leher SB. Principles and characteristics of food allergens.

Crit Rev Food Sci Nutr 1996;36(suppl):S91-118.

9. Sampson HA. Food Allergy. JAMA 1997;278:1888-94.

10. Anderson JA. Allergic reactions to foods. Crit Rev Food Sci Nutr 1996;

36(suppl):S19-38.

11. Bock SA. Prospective appraisal of complaints of adverse reactions to

foods in children during the first three years of life. Paediatrics 1987;79:

683-8.

12. Helm RM, Burks AW. Mechanisms of food allergy. Curr Opin Immunol

2000;12:647-53.

13. Metcalfe DD, Astwood JD, Townsend R, Sampson HA, Taylor SL,

Fuchs RL. Assessment of the allergenic potential of foods derived from

genetically engineered crop plants. Crit Rev Food Sci Nutr 1996;

36(suppl):S165-86.

14. Mendieta NLR, Nagy AM, Lints FA. The potential allergenicity of novel

foods. J Sci Food Agric 1997;75:405-11.

15. Taylor SL. Assessment of the allergenicity of genetically modified foods.

Nutr Abstracts Rev (Series A) 1997;67:1163-8.

16. Lack G, Chapman M, Kalsheker N, King V, Robinson C, Venables K.

Report on the potential allergenicity of genetically modified organisms

and their products. Clin Exp Allergy 2002;32:1131-43.

17. Draft International Standard ISO/DIS 21571. Foodstuffs—methods of

analysis for the detection of genetically modified organisms and derived

products—nucleic acid extraction. Nov 2002. p. 25-8.

18. Draft International Standard ISO/DIS 21569. Foodstuffs—methods

of analysis for the detection of genetically modified organisms and

derived products—qualitative nucleic acid based methods. Nov 2002.

p. 23-66.

19. Sub-Committee on Skin Tests of the European Academy of Allergology

and Clinical Immunology. Skin tests used in type I allergy testing.

Allergy 1989;44(suppl 10):1-59.

20. Laemmli UK. Cleavage of structural proteins during the assembly of the

head of bacteriophage T4. Nature 1970;227:680-5.

21. Bliss CI. Statistics in biology. Vol 1. New York: MacGraw-Hill; 1967.

p. 558.

22. Losey JE, Rayon LS, Carter ME. Transgenic pollen harms monarch

larvae. Nature 1999;395:214.

23. Ewen SWB, Pusztai A. Effect of diets containing genetically modified

potatoes expressing Galanthus nivalis lectin on rat small intestine. Lancet

1999;354:1353-4.

24. TacoBell taco shells sold in grocery stores contain banned corn [transcript].

CNN. September 18, 2000. Available at: http://europe.cnn.com/2000/

FOOD/news/09/18/food.corn.reut/. Accessed January 10, 2005.

25. EPA Assessment of Scientific Information concerning Starlink Corn

Cry9C Bt Corn Plant-pesticide. Federal Register 65 (October 31, 2000).

Environmental ProtectionAgency publication no. 65246-65251.Available

at: http://www.access.gpo.gov/su_docs/fedreg/frcont00.html. Accessed

January 10, 2005.

26. Sears MK, Hellmich RL, Stanley-Horn DE, Oberhauser KS, Pleasants

JM, Matilla HR, et al. Impact of Bt corn pollen on monarch butterfly

populations: a risk assessment. Proc Natl Acad Sci U S A 2001;21:

11937-42.

27. Sutton SA, Assa’ad AH, Steinmetz C, Rothenberg ME. A negative,

double-blind, placebo-controlled challenge to genetically modified corn.


28. Kuiper HA, Noteborn HPJM, Peijnenburg ACM. Adequacy of methods

for testing the safety of genetically modified foods. Lancet 1999;354:

1315-6.

29. Vadas P, Wai Y, Burks W, Perelman B. Detection of peanut allergens in

breast milk of lactating women. JAMA 2001;285:1746-8.

30. Miranda R, Zamudio FZ, Bravo A. Processing of Cry1Ab d-endotoxin

from Bacillus thuringiensis by Manduca sexta and Spodoptera frugi-

perda midgut proteases: role in protoxin activation and toxin inactiva-

tion. Insect Biochem Mol Biol 2001;31:1155-63.

31. Sampson HA, Metcalfe DD. Food allergies. JAMA 1992;268:2840-4.

32. Sampson HA. Food allergy. Part 2: diagnosis and management. J Allergy

Clin Immunol 1999;103:981-9.

33. Sten E, Skov PS, Andersen SB, Torp AM, Olesen A, Bindslev-Jensen U,

et al. A comparative study of the allergenic potency of wild-type and

glyphosate-tolerant gene-modified soybean cultivars. APMIS 2004;112:

21-8.

34. Burks AW, Fuchs RL. Assessment of the endogenous allergens in

glyphosate-tolerant and commercial soybean varieties. J Allergy Clin

Immunol 1995;96:1008-10.

35. Sicherer SH. Clinical implications of cross-reactive food allergens.


36. Vieths S, Scheurer S, Ballmer-Weber B. Current understanding of cross-

reactivity of food allergens and pollen. Ann N Y Acad Sci 2002;964:

47-68.

37. Sampson HA, Albergo R. Comparison of results of skin tests, RAST and

double-blind, placebo-controlled food challenges in children with atopic

dermatitis. J Allergy Clin Immunol 1984;74:26-33.

38. Bindslev-Jensen C, Poulsen LK. Accuracy of in vivo and in vitro tests.

Allergy 1998;53:72-4.


AUGUST 2005

410 Batista et al

Foodalle

rgy,derm

ato

logic

dise

ase

s,andanaphylaxis

http://europe.cnn.com/2000/FOOD/news/09/18/food.corn.reut/

http://europe.cnn.com/2000/FOOD/news/09/18/food.corn.reut/

http://www.access.gpo.gov/su_docs/fedreg/frcont00.html


Advances in Asthma, Allergy, and Immunology Series 2005


Javier Chinen, MD, PhD,a and William T. Shearer, MD, PhDb Bethesda, Md,

and Houston, Tex

The authors selected articles published in the literature from

January 2004 through December 2004 that were relevant to

the areas of basic and clinical immunology. Several articles

explored the development of TH1 or TH2 response and the role

of the monocyte–T cell interaction. Others were articles

describing the action of drugs commonly used in asthma to

inhibit cytokine responses and the anti-inflammatory role of

nonimmune pulmonary cells present in the lung. Several

reports show how dendritic cells are being developed as

vehicles for DNA vaccines aimed at stimulating cellular

responses, an advance of great importance for HIV researchers

working on vaccines, who are concerned about the different

ways HIV evades the immune response. Other publications

described Toll-like receptors in diverse cells, including mast

cells and CD41 T cells, for the recognition of viruses and

bacteria. In the area of clinical immunology, an updated

classification for primary immunodeficiencies with more than

100 identified genes responsible for these diseases and the

report on the second clinical trial of gene therapy for X-linked

severe combined immunodeficiency syndrome were published.

Significant advances included the clinical prognosis in common

variable immunodeficiency for patients presenting with lung

pathology, the safety of live vaccines in partial DiGeorge

syndrome, the report of patients with complete DiGeorge

syndrome with the presence of peripheral blood T cells, the

clinical spectrum of patients with NF-kB essential modifier

(NEMO) gene deficiency, the publication of a consensus

algorithm for the management of hereditary angioedema, and

the report of immune restoration syndrome in pediatric HIV

infection. (J Allergy Clin Immunol 2005;116:411-8.)

Key words: Immunoregulation, HIV, immunodeficiency, innateimmunity, complement

The areas of basic and clinical immunology continue todevelop at a fast pace, with numerous reports exploringrelatively new and old areas of these fields, such as thebiology of Toll-like receptors (TLRs) and the descriptionof primary immunodeficiencies (PIDs), respectively. Thegoal of this article is to review some of the significantprogress in basic and clinical immunology published in2004, with focus on articles that the authors consideredof interest to the readers of The Journal of Allergy andClinical Immunology.

BASIC IMMUNOLOGY

Some key advances in basic immunology are listedin Table I.

Regulation of the T-cell response:TH1 versus TH2

Wittmann et al1 investigated the cytokine secretionprofile resulting from the interaction of monocytes derivedfrom peripheral blood and autologous CD41 T cellsisolated from inflammatory skin lesions induced by anallergen patch test. These activated T cells induced IL-12secretion by monocytes that were stimulated with IFN-g.However, when the T cells were incubated with restingmonocytes, IL-12 secretion was not induced. In contrast,resting T cells did not inhibit IL-12 secretion in restingmonocytes. The authors further determined that this effecton IL-12 secretion was cell-contact specific and dependent

Abbreviations usedAPC: Antigen-presenting cell

CTL: Cytotoxic T cell

CVID: Common variable immunodeficiency

DGS: DiGeorge syndrome

DSS: Dextran sulfate sodium

HIGM: Hyper-IgM syndrome

ICOS: Inducible costimulatory molecule

IRD: Immune restoration disease

NEMO: NF-kB essential modifier

NK: Natural killer

PID: Primary immunodeficiency

TCR: T-cell receptor

TLR: Toll-like receptor

From aGenetics and Molecular Biology Branch, National Human Genome

Research Institute, National Institutes of Health, Bethesda, and bthe

Department of Allergy and Immunology, Texas Children’s Hospital, and

the Departments of Pediatrics and Immunology, Baylor College of

Medicine, Houston.

The opinions expressed in this article do not necessarily represent the views of

the National Human Genome Research Institute or the National Institutes of

Health.


Received for publication May 4, 2005; accepted for publication May 6, 2005.


Reprint requests: Javier Chinen,MD, PhD, National Human Genome Research

Institute, 10 Center Drive, MSC 1611, Building 10/CRC Room 6-3340,

Bethesda, MD 20892. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.010

411

Basicandclinicalimmunology


on the induction of T-bet expression in monocytes. T-betis a signal transduction factor that is essential for thedevelopment of the TH1 response. On the basis of thesefindings, the authors suggest that initial events in skinatopic disease might involve the infiltration of activatedT cells into the skin and interaction with resting antigen-presenting cells (APCs), resulting in absence of IL-12 anddevelopment of a TH2 environment. The role of T-bet asthe key protein for the induction of the TH1 response wassupported by the work of Lametschwandtner et al,2 whoinduced T-bet expression in TH2 cells obtained from skinbiopsy specimens of atopic individuals. TH2 cells ex-pressing T-bet were able to secrete and express high levelsof IFN-g, TNF-a, IL-2, and IL-12, with a decrease of theexpression levels of IL-4 and IL-5. In addition, theyshowed a reversion of chemokine expression profile fromTH2 to TH1. The immunologic events early in infancy thatcan lead to atopic disease were investigated by Uphamet al,3 who described that the HLA-DR expression inmonocytes obtained from cord blood stimulated withIFN-g correlated with IL-12 secretion induced by endo-toxin and had an inverted association with IL-13 secretioninduced by ovalbumin or dust mite. HLA-DR expressionin unstimulated monocytes was inversely associated withallergic disease at the 2-year follow up. These findingssuggest that early activation of APCs might decrease theTH2 response and the risk of development of atopicdisease.

Two drugs commonly used in asthma, fluticasone andsalmeterol, were shown to synergistically inhibit cytokinesecretion and to influence the TH2 to TH1 balance.4 Thedrug combination inhibited the secretion of the TH1cytokines TNF-a and IFN-g by mitogen-stimulatedPBMCs from normal and asthmatic patients. In contrast,the secretion of the TH2 cytokines IL-5 and IL-13 wasinhibited only in PBMC cultures from control subjects butnot from asthmatic patients. The addition of a phospho-diesterase inhibitor to the combination suppressed IL-13secretion in PBMCs from asthmatic patients, an effect thatcan be explained by the maintenance of high cyclicadenosine monophosphate levels. Pace et al5 were alsointerested in the mechanism of action of the combinationof fluticasone and salmeterol in T cells. These investiga-tors found that the induction of apoptosis in peripheral

blood T cells by fluticasone increases synergistically withthe addition of salmeterol. Increased apoptosis was asso-ciated with a more efficient caspase processing, increasedtranslocation of the glucocorticoid receptor, and reductionof the expression of phosphorylated IkBa. Salmeterolalone did not produce apoptosis in the T cells.

A factor that affects the immune regulation of theallergic response is the 10-kd protein secreted bythe pulmonary Clara cells.6 This protein decreased theexpression of the TH2 cytokines IL-4, IL-5, and IL-13of ovalbumin-sensitized mouse splenocytes and of CD41

T cells that had been polarized into TH2 cells. This potenteffect was dose dependent and associated with a reductionof intracellular GATA-3, the transcription factor thatmediates TH2 response. This result underscores thedynamic interaction between immune cells and highlyspecialized lung cells in lung inflammation and asthmapathogenesis.

DNA vaccines

DNA vaccination is being actively explored as analternative for the treatment of allergic diseases. DNA-based immunomodulation has been shown to switch theimmune response from a TH2- to a TH1-dominant re-sponse in several mouse models.7 Klostermann et al8 usedhuman dendritic cells transduced with an adenovirusvector carrying the expression cassette for the grass pollenPhl p 1 protein. When these cells were cocultured withT cells, they induced a TH1-like response, with prolifer-ation of the CD81 T-cell subtype, increased IFN-gsecretion, and less IL-4 and IL-5 secretion than whennontransduced dendritic cells were pulsed with the Phl p 1protein. A different delivery strategy for DNA vaccinationin allergy was shown by Ludwig-Portugall et al9 using agene gun–mediated delivery in vivo. Although thismethod requires 100- to 1000-fold more DNA, it does notinvolve a viral vector. The authors used the gene gun toimmunize mice percutaneously with a b-galactosidase–encoding plasmid and then sensitized the animal tob-galactosidase protein. IgG2a was 10-fold increased,and specific IgE was not detectable. In contrast, mice thatwere immunized intraperitoneally with b-galactosidasehad high specific IgE levels. To increase the immunoge-nicity of DNA vaccines, Jilek et al10 examined the use ofbiodegradable microspheres as DNA carriers for prophy-laxis against anaphylaxis. They used microspheres madeof polylactidecoglycolide, a biodegradable material thatis readily phagocytosed by dendritic cells.11 When micereceived subcutaneous polylactidecoglycolide micro-spheres with DNA encoding phospholipase A, the beevenom major allergenic protein, and then were sensitizedand challenged with a lethal dose of the phospholipase Aprotein, anaphylaxis was prevented in 50 of 54 experi-mental mice. Interestingly, the preventive effect was alsoachieved in animals that received microspheres withnonspecific DNA. The authors also demonstrated similarproduction of IgG2a, IL-4, IFN-g, and IL-10, suggestingthat polylactidecoglycolide microspheres drive APCstoward the TH1 phenotype and could be considered for

TABLE I. Key advances in basic immunology

1. Monocyte activation is necessary for induction of a TH1

response.

2. Salmeterol and fluticasone act synergistically to inhibit

secretion of inflammatory mediators in asthma.

3. A 10-kd protein from pulmonary Clara cells and IL-17F from

bronchial epithelial cells are potent anti-inflammatory cytokines.

4. HLA alleles influence the generation of HIV CTL

escape mutants and the risk of HIV transmission.

5. Mast cells recognize viruses and are activated through

Toll receptors.

6. CD4 T cells can be directly activated through Toll receptors.

7. Alternative splicing might be responsible for breaking immune

tolerance and the development of autoimmune disorders.


AUGUST 2005

412 Chinen and Shearer

Basic

andclin

icalim

munology

therapeutic use in atopic patients. Another moleculestudied was the CpG oligonucleotide, which was shownto drive TH1 cytokine expression in plasmacytoid den-dritic dells obtained from patients with allergic rhinitis.12

When a CpG oligonucleotide was introduced to coculturesof dendritic cells and CD41 T cells, the cytokine secretionprofile changed from being predominantly composed ofIL-4 and IL-5 to being mostly IFN-a and TNF-a. Thisstudy suggests that mucosal dendritic cells can be con-sidered as targets for DNA-based immunomodulationof the T-cell response.

Advances in cytokine research

The production of IL-13 by human B lymphocytes andits role in IgE synthesis was investigated by Hajoui et al13

using B cells isolated from the tonsils of healthy volun-teers. They reported a 10-fold increase of IL-13 secretionwhen these cells were stimulated with anti-CD40 antibodyand IL-4. When they added neutralizing anti-IL-13 anti-bodies, IgE levels decreased by 80%, and IgE transcriptsdecreased by 50%, suggesting that B-cell secretion of IgEis regulated in part by IL-13 produced by the same B-cellpopulation. Two articles published in the Journal focusedon the IL-17 family of inflammation proteins and the roleof nonimmune cells in lung inflammation. Kawaguchiet al14 reported that a function of a newly identified IL-17Fprotein in primary bronchial epithelial cells was to induceGM-CSF secretion through activation of Raf-1/MEK-ERK1/2, and therefore IL-17F participates in the patho-physiology of allergic inflammation. A second articleexamined the regulation of IL-17A in human airwaysmooth cells obtained from patients undergoing lungsurgery. IL-17A induced secretion of IL-6 after stimula-tion with TNF-a but not after stimulation with IL-1b. Ofnote, there was no induction of other inflammationmarkers, such as intercellular adhesion molecule expres-sion or GM-CSF secretion.15 The expression of IL-10 andFoxP3, which phenotypically define regulatory T cells,was compared in CD41 T cells obtained from patientswith moderate and severe asthma and from healthy controlsubjects.16 This study found that FoxP3 mRNA expres-sion correlated with IL-10 mRNA expression, and it was2-fold higher in asthmatic patients receiving steroids thanin healthy control subjects or patients with mild asthma.In addition, it was shown that CD41CD251 T cellsexpressed 11-fold more IL-10 and FoxP3 than totalCD41 T cells after being exposed to corticosteroidsin vitro. These results suggest that the anti-inflammatoryeffect of corticosteroids include the development ofregulatory T cells secreting IL-10.

HIV immunopathogenesis

The mechanisms by which HIV evades the immunesystem are far from being completely elucidated. Leslieet al17 studied the association of a specific cytotoxic T-cell(CTL) epitope in the HIV1 Gag protein and HLA allelesin an HIV-infected population. They found that HIV-infected individuals expressing the HLA alleles B57 andB*5801 had selected for variants with a specific mutation

in this Gag epitope. However, when this HIV strain wastransmitted to an individual with different HLA alleles,this epitope reverted to the wild type. Supporting theirfinding, they demonstrated that a second mutation in thesame epitope did not revert. The reasons why someepitope mutations persist and others revert to the wildtype are not clearly related to virus fitness, but certainlythis is of concern for the design of vaccines targeting theanti-HIV CTL response. A related article by Dorak et al18

examined 125 couples who were initially HIV discordantafter which the spouse converted and 104 persistent HIV-discordant couples. They found that the risk of HIVtransmission to the spouse was 2-fold higher, independentof viral load, if the couples shared one or both HLA-Balleles than if they had different HLA-B alleles. Thissuggests that HIV CTL escape mutants are transmittedmore efficiently in a homogenous population and that theywill vary from population to population (Fig 1). Addingone more strategy for HIV immune evasion, Draenertet al19 reported mutations in the Gag protein outside aparticular epitope that altered a target site for proteinprocessing in APCs and therefore cannot be presented,making CTLs unable to lyse infected cells.

Winchester et al20 investigated innate immunity inmother-to-child transmission of HIV, mediated by mater-nal natural killer (NK) cells. The expression of HLA-B*4901 and B*5301 alleles inhibited mother-to-infantHIV transmission despite high maternal viral loads. Theyalso bound the KIR30L1 NK receptor. The HLA-B*5001and B*3501 alleles, which differ from B*4901 andB*5301 by only 5 amino acids, did not bind theKIR30L1 receptor and were associated with enhancedvertical transmission. The authors proposed that themolecular basis of this observation involved maternalNK cell recognition by engagement of NK cell receptorswith polymorphic ligands encoded by maternal HLA-Balleles. Moreover, they believe that the placenta is thesite where protection against vertical HIV transmissionoccurs, mediated by interrelating adoptive and innateimmune recognition mechanisms.

In the B-cell compartment, Moir et al21 found 42 genesupregulated in B cells from HIV-viremic patients com-pared with B cells from healthy control subjects. Mostof these genes were associated with the activation of theIFN-g pathway or with terminal differentiation of B cells.In addition, they showed that CD95 expression in B cellscorrelated with HIV viremia. This report is valuable forthe identification of genes involved in the mechanismsof B-cell dysfunction in HIV infection.

Innate immunity

The role of TLRs in innate immunity continues toexpand and involve many different immune cells andprocesses. Kulka et al22 added mast cells to the list ofeffector cells participating in the recognition of virusesthrough TLRs. Mast cells had already been shown torespond toLPS and peptidoglycan throughTLR-1, TLR-2,TLR-4, and TLR-6. The importance of mast cells onshaping the innate immunity response to infection was



Chinen and Shearer 413


reviewed by Marshall and Jawdat.23 Mast cells arenot only activated by TLRs but also by complementcomponents, leading to the secretion of cytokines impor-tant for selective recruitment of effector cells. The expres-sion of TLR-3 in mast cells derived from peripheral bloodand 2 mast cell lines was newly reported, and theirproduction of IFN-a and IFN-b in response to exposureto dsDNA and to PolyI:C was also reported. A similarresponse was obtained when mast cells were exposedto UV light–inactivated influenza virus and to type 1reovirus.

Flo et al24 demonstrated that TLR-4 stimulation withLPS inmacrophages helped to control bacterial replicationby increasing levels of lipocalin 2 expression. This proteininhibits iron uptake by Escherichia coli. Lipocalin 2knockout mice became highly bacteremic after infectionwith E coli but not after infection with other bacteria lessdependent on iron. In the gut the existence of commensalbacteria has prompted the question of how the gut controlsinflammation, and it has been thought that inflammatorybowel disease could be caused by inappropriate recogni-tion of antigens. Rakoff-Nahoum et al25 showed that micedeficient in TLRs have increased mortality than wild-typemice after receiving dextran sulfate sodium (DSS) as amodel for inflammation. These mice were deficient onMYD88, a signal transduction factor essential for sev-eral TLR-mediated responses. The mice presented withepithelial injury and severe colonic bleeding but notleukocyte infiltration or bacterial overload. Previousadministration of antibiotics did not modify the pathologicchanges. The TLR-deficient mice had increased prolifer-

ation of colonic cells and were more susceptible to injurycaused by DSS or radiation. Similar mortality occurredwhen wild-type mice were deprived of commensal bacte-ria and then were treated with DSS. When these animalswere given LPS, the animals were protected, suggestingthat TLR stimulation might mediate epithelial barrierrepair. To provide a direct link between innate andadaptive immunity, Gelman et al26 reported that mouseCD41 T cells expressed TLR-3 and TLR-9 and respondedto CpG and polyI:C stimulation with increased survivaland nuclear factor kB (NF-kB) activation. This observa-tion might represent the response of the immune systemfor infectious organisms that impair APCs by directlyactivating the adaptive immune cells.

Immune mechanisms of drug allergy

Depta et al27 challenged the classical notion thathaptens stimulate specific T cells only when they arecovalently bound to proteins. The investigators trans-fected a mouse T-cell hybridoma to express a plasmidencoding a T-cell receptor (TCR) specific to sulfametox-azole. When these cells were exposed to the drug in thepresence of fixed EBV-transformed B cells, they prolif-erated with a reactivity that was dependent on the level ofthe specific TCR expression. Increased TCR expressionalso correlated with cross-reactivity with other drugs thatshare the sulfanilamide core structure but not with othersulfonamides, like furosemide or celecoxib. Because fixedB cells were used as APCs, these results showed thatdrugs can directly interact with TCRs and that there is noneed for antigen processing to obtain T-cell reactivity to

FIG 1. HIV escape mutants (blue particles) appear over time in HIV-infected individuals with effective specific

CTL responses because of survival selective pressure and depending on viral fitness of the newmutants. They

persist after transmission between individuals sharing the same HLA alleles but might revert to the wild type

(red particles) in the newly infected individual with different HLA alleles. There is a higher transmission rate

when HIV-infected couples share the same HLA alleles.


AUGUST 2005


Basic

andclin

icalim

munology

sulfonamides. A study by Nassif et al28 examined thephenotypes of lymphocytes obtained from 6 patients withtoxic epidermal necrolysis caused by hypersensitivity toa single drug, either cotrimoxazole and carbamazepine,tetrazepam, or piroxicam. These lymphocytes were 70%to 90% CD81 CTLs and were specific for the offendingdrug in the cases of cotrimoxazole and carbamazepine butnot for tetrazepam or piroxicam. The authors believe thatthe lack of specificity for the last 2 drugs can be explainedby the uncertainty of the offending drug, but overall, thedata support the hypothesis that specific CTLs are respon-sible for toxic epidermal necrolysis pathology.

Applications of the human genome sequencein immunology

Ng et al29 took advantage of the completion of themapping and sequencing of the human genome to supporttheir hypothesis that alternative splicing of self-antigensmight play a role for the generation of autoantigens inautoimmune disorders. Alternative splicing might disturbperipheral tolerance that has already been attained for thenormal spliced protein. The authors randomly chose 45self-proteins that have been implicated in autoimmunityand compared their alternative splicing frequency with9554 random proteins of the human genome. Forty-twopercent of the random proteins present alternative splicingin contrast to 100% of the 45 proteins implicated inautoimmunity. Eighty percent of these proteins mightundergo noncanonical alternative splicing, which wasmuch higher than the 1% of randomly selected proteins.More experimental data are needed to confirm this originalhypothesis, which would provide insight into the patho-genesis of autoimmunity disorders and novel therapydevelopment. In an example of high-yield gene searchstudies, Nakajima et al30 used a gene chip contain-ing about 22,000 gene probes to compare transcriptsexpressed in CD41 cells, CD81 cells, basophils, eosino-phils, neutrophils, CD141 cells, and CD191 cells, focus-ing on the expression of granulocyte-selective genes forion channels. The authors found 17 novel transcripts from51 with 5-fold greater expression than other leukocytelineages. Six of these 17 were eosinophil and basophilspecific. The authors reported the list of genes withspecific expression and stressed their importance fordrug targets in allergic and inflammatory processes andtheir significance for drug development in allergic diseaseand inflammation. Genetic variations influencing allergywere described by Hoffjan et al31 by screening 200children for 61 polymorphisms in 35 immunoregulatorygenes. The polymorphisms were analyzed in regard tocytokine production and allergic sensitization, as well asinteraction between the polymorphisms. The authorsfound 5 associations that involved a reduced IL-13secretion, including polymorphisms in the genes forIL-13, TGF-b, IgE receptor, and nitric oxide. None ofthe genes were associated with atopic dermatitis. Theauthors concluded that variations in immunoregulatorygenes might be risk factors for the development of allergicdisease and childhood asthma.

CLINICAL IMMUNOLOGY

Some key advances in clinical immunology are listedin Table II.

Asthma and the immune response

Hanania et al32 asked whether the corticosteroid ther-apy in patients with asthma affects the immune response toinfluenza vaccine. Asthmatic subjects (n = 294) who wererandomized to receive either placebo or inactivated influ-enza vaccine were divided in 2 groups, one that receivedmedium- or high-dose inhaled corticosteroids and anotherthat received none or only low corticosteroid doses. Theserologic response to influenza serotype A was notimpaired with the use of corticosteroids, but the responseto serotype B was slightly decreased, with a 2.1-foldincrease of the titers compared with an increase of 2.5-foldin the group receiving no steroids or only low-dosesteroids. Although actual protection against influenzainfection was not measured, the study places a word ofcaution on the possible decreased immune response inasthmatic patients taking steroids.

PIDs

A must-read article is the update in PIDs written byNotarangelo et al33 representing the International Union ofImmunological Societies Primary ImmunodeficiencyDiseases Classification Committee. The authors reviewedand classified PIDs reported up to their last meeting in2003. More than 100 PIDs have been defined and char-acterized. Although rare, PIDs are diagnosed with morefrequency and in more diverse ethnic groups. However,more efforts are needed for PID awareness in minoritygroups, as noted by Cunningham-Rundles et al,34 wholooked in a database of over 120,000 inpatients of ageneral hospital for conditions suggestive of immuno-deficiency. Fifty-nine patients were identified, and 17 ofthem had an undiagnosed PID. Eighty-six percent of thesepreviously undiagnosed patients with PIDs were AfricanAmerican or Hispanic.

It is common to think that the de novo genetic defects ofPIDs should occur during egg fertilization and embryoformation. This idea was challenged by Holzelova et al.35

They investigated patients with the autoimmune lympho-proliferative syndrome but without identified mutationsin the causative genes Fas, Fas L, Casp8, and Casp10.The authors cleverly explored the double-negative T cellsthat accumulate in these patients. They identified Fasmutations in these cells and subsequently in monocytesand CD341 cells, but the mutations were not present inmucosal cells or B cells or when total T cells were tested.The authors were able to demonstrate these somaticmutations in 2 of 6 patients studied and showed that thesurvival advantage conferred to a subset of lymphocyteswas enough to produce autoimmune lymphoproliferativesyndrome.

Chinen and Puck36 reviewed the progress and currenthurdles of gene therapy for PIDs. The French clinical trialfor X-linked severe combined immunodeficiency has now





treated 11 infants, with successful T-cell restoration in 9of them; the other 2 had only partial reconstitution andsubsequently received a conventional bone marrow trans-plantation.37 In addition, Gaspar et al38 published theirexperience in 4 infants with X-linked severe combinedimmunodeficiency in England who received gene therapyand achieved normal T-cell numbers. However, thesesuccesses were tempered with the occurrence of T-cellleukemia in 3 of the children from the French trial.37 Thesemalignancies were proved to be caused, at least in part, byinsertion of the retroviral vector in oncogenes. New genevector designs are being investigated to reduce the risk ofcancer caused by insertional mutagenesis.

Gardulf et al39 reported their experience with thesubcutaneous self-administration of immunoglobulinsfor patients with PID done at home, suggesting that thequality of life of these patients might increase with thismodality. Chinen and Shearer40 reviewed the pros andcons of subcutaneous administration of immunoglobulinsfor immunodeficiency, which is being established as aviable alternative to traditional intravenous infusions.

Regarding specific PIDs, several advances have beenmade in common variable immunodeficiency (CVID),hyper-IgM syndrome (HIGM), and DiGeorge syndrome(DGS). Bates et al41 investigated the clinical features ofnoninfectious pulmonary disease in patients with CVID.They found that 29 of 69 patients with CVID presentedwith none of these abnormalities, 23 had respiratorysymptoms but were radiologically normal, and 18 hadrespiratory symptoms and radiologic diffuse abnormali-ties. Within this last group, those who had granulomatouslung disease, follicular bronchiolitis, lymphoid hyperpla-sia, and lymphoid interstitial pneumonia (13/18 patients)had worse prognosis and survival than the other groups,with a survival of 13.7 years since the time of diagnosiscompared with 28.8 years in the other groups. In addition,these patients also were at higher risk of lymphoprolifer-ative disease. Salzer et al42 investigated the role of

inducible costimulatory molecule (ICOS) in their cohortof patients with CVID and found mutations in the ICOSgene in 2 of 9 families with autosomal-recessive CVIDand no mutations in the ICOS ligand gene. No polymor-phic variants found in the ICOS gene sequence weremore common in patients with CVID than in the generalpopulation. An additional 181 patients with sporadicCVID were examined, and no mutations were found.This report confirms that an ICOS gene defect might causeCVID, although it is responsible for only a minority ofcases.

A comprehensive review of DGS or chromosome22q11.2 deletion syndrome by Sullivan43 emphasizesthe spectrum of severity of each of the clinical findingsof this condition, including T-cell deficiency. A few casesmight remain undiagnosed until the patient reachedchildhood and receives live vaccines before immunologicstatus is assessed. Moylett et al44 studied a cohort of 53such patients and found that 25 of them had received a livevaccine. However, no serious adverse effects related tolive vaccines occurred. Although reassuring, it is impor-tant to note that this cohort of patients had only a mild-to-moderate decrease of T cells. The recommendation onthe use of live vaccines in this group of patients is stillcontroversial, and a cautious approach is advised againstlive vaccines administration until more data are avail-able. Markert et al45 reported an unusual presentation of5 patients with DGS who had heart, parathyroid, andimmune defects. In addition, they presented with rashand lymphadenopathy. Although they had T cells andsome had mitogen proliferative responses, these T cellswere oligoclonal and there was no evidence of thymusactivity, as measured by the absence and low output ofnaive cells. The authors concluded that the presence ofT cells did not necessarily mean the presence of thymusactivity and recommended that patients with DGS shouldundergo an evaluation of thymus activity in addition to theassessment of T-cell numbers.

Two review articles by Etzioni and Ochs46 and Luoet al47 described and summarized the recent develop-ments on HIGM and the biology of the activation-inducedcytosine deaminase, one of the proteins that, when miss-ing, results in HIGM. Orange et al48 described 7 patientswith anhydrotic ectodermal dysplasia with immuno-deficiency caused by mutations in the NF-kB essentialmodifier (NEMO) gene. They demonstrated a particularsusceptibility to pyogenic and mycobacterial infections.NEMO deficiency was initially described as a form ofHIGM but is currently classified as an innate immunitydefect.33 In a follow-up article, the authors described a16-year-old patient who had a mutation in the Ikb kinaseportion of the NEMO gene, and although presenting withimmunologic defects, the patient did not have the ecto-dermal defect components of this condition.49 Niehueset al50 reported a similar patient, although with a mutationin exon 2 inducing a premature stop codon. These 2 casesunderscore the variety of presentations of these raredisorders and the need for continuing awareness fordiagnosis.

TABLE II. Key advances in clinical immunology

1. The immune response to influenza vaccine in patients with

moderate-to-severe asthma taking inhaled or oral corticosteroids

is slightly decreased.

2. Somatic mutations in T-cell progenitors might cause ALPS.

3. Four infants with XSCID were successfully treated with gene

therapy in England.

4. Patients with CVID with granulomatous lung disease, lymphoid

hyperplasia, and lymphoid interstitial pneumonia have poor

survival prognosis.

5. Some patients with complete DGS might have detectable

numbers of T cells in peripheral blood, although oligoclonal in

nature and with poor function.

6. NEMO-deficient patients have increased susceptibility to pyo-

genic and mycobacterial infections. Some of these patients might

not have the ectodermal component of this syndrome.

7. IRD occurs in pediatric HIV infection.

ALPS, Autoimmune lymphoproliferative syndrome; XSCID, X-linked severe

combined immunodeficiency.


AUGUST 2005


Basic

andclin

icalim

munology

Complement deficiency

Three highly recommended reference articles for themanagement of complement deficiencies were publishedin 2004. One is a comprehensive review for the clinicalevaluation of complement deficiencies addressed for theclinician and containing useful algorithms for the evalu-ation of patients with suspected complement deficiency.51

This article explains the genetics of these deficiencies,clinical manifestations, and current therapeutic advances.The second article is a review of the history, genetics,clinical presentation, and current management of heredi-tary angioedema caused by deficiency of C1 inhibitor,written by Frank.52 In an effort to gain consensus in themanagement of hereditary angioedema, an internationalconference was held in Ontario, Canada, in 2003 with theparticipation of European and American researchers.A summarizing consensus algorithm was drafted andpublished.53 Diagnostics and managements availablewere reviewed, including appropriate use of tranexamicacid, androgens, and C1 inhibitor concentrate.

Immunorestoration syndrome in HIV infection

Seven cases of immune restoration disease (IRD) weredescribed in a cohort of 69 perinatally acquired HIV-infected children.54 IRD presents as a severe inflammatoryreaction to opportunistic infections in HIV-infected pa-tients who received highly active antiretroviral therapyand have a good response with recovery of normal T-cellcounts. This is the first report of IRD in children.Interestingly, all 7 cases were caused by herpes zosterand occurred in those with the most severe immunodefi-ciency at baseline.

CONCLUSIONS

In 2004, several exciting developments were reportedin the areas of basic and clinical immunology. The role ofmonocyte activation in the development of TH1 responseswas characterized, as well as the action of drugs com-monly used in asthma to inhibit cytokine responses andinduce regulatory T-cell differentiation. The regulatoryeffect of nonimmune cells present in the lung in thedevelopment of a TH1 or TH2 response has been estab-lished, with reports of pulmonary Clara cells inhibitingIL-4 and IL-13 expression and bronchial epithelial cellssecreting the newly described IL-17F, a cytokine thatparticipates in controlling allergic inflammation. The useof dendritic cells genetically engineered to favor theexpression of TH1 cytokines on specific allergen stimula-tion has proved to be effective in in vitro models. In thearea of HIV immunology, several investigators havedescribed several methods used by HIV to escape CTLsurveillance and control. This work is of importance toresearchers developing vaccines on the basis of cellularresponse because it suggests that multiple epitopes spe-cific for target populations need to be included. Researchon Toll receptors continues to reveal more mechanisms of

innate immunity. They are present in mast cells and CD41

T cells for the recognition of viruses and bacteria, and theyare involved in the control of iron metabolism that isessential for some bacteria species.

Clinical research in immunologic diseases continues toshow remarkable progress. An updated classification forPIDs is now available, with more than 100 genes identifiedas responsible for these diseases. ICOS is known now tobe responsible for a minority of patients with CVID. Withthe advances in the field of genetics, a second successfultrial of gene therapy has been published, although 3 casesof leukemia have occurred in patients from the first trial.Other advances were the definition of poor prognosis inCVID for patients presenting with specific inflammatorylung pathology, the estimation of safety for administrationof live vaccines in patients with partial DGS, the unusualpresentation of complete DGS with the presence of hostT cells in peripheral blood, the presentation of the clinicalspectrum of patients with NEMO deficiency, the elabora-tion of a consensus algorithm for the management ofhereditary angioedema, and the description of immunerestoration syndrome in HIV-infected children with pre-dominance of herpes zoster infections.

REFERENCES

1. Wittmann M, Alter M, Stunkel T, Kapp A, Werfel T. Cell-to-cell contact

between activated CD41 T lymphocytes and unprimed monocytes

interferes with a TH1 response. J Allergy Clin Immunol 2004;114:

965-73.

2. Lametschwandtner G, Biedermann T, Schwarzler C, Gunther C, Kund J,

Fassl S, et al. Sustained T-bet expression confers polarized human TH2

cells with TH1-like cytokine production and migratory capacities.


3. Upham JW, Holt PG, Taylor A, Thornton CA, Prescott SL. HLA-DR

expression on neonatal monocytes is associated with allergen-specific

immune responses. J Allergy Clin Immunol 2004;114:1202-8.

4. Goleva E, Dunlap A, Leung DY. Differential control of TH1 versus

TH2 cell responses by the combination of low-dose steroids with

beta2-adrenergic agonists. J Allergy Clin Immunol 2004;114:183-91.

5. Pace E, Gagliardo R, Melis M, La Grutta S, Ferraro M, Siena L, et al.

Synergistic effects of fluticasone propionate and salmeterol on in vitro

T-cell activation and apoptosis in asthma. J Allergy Clin Immunol

2004;114:1216-23.

6. Hung CH, Chen LC, Zhang Z, Chowdhury B, Lee WL, Plunkett B, et al.

Regulation of TH2 responses by the pulmonary Clara cell secretory

10-kd protein. J Allergy Clin Immunol 2004;114:664-70.

7. Spiegelberg HL, Raz E. DNA-based approaches to the treatment of

allergies. Curr Opin Mol Ther 2002;4:64-71.

8. Klostermann B, Bellinghausen I, Bottcher I, Petersen A, Becker WM,

Knop J, et al. Modification of the human allergic immune response by

allergen-DNA-transfected dendritic cells in vitro. J Allergy Clin

Immunol 2004;113:327-33.

9. Ludwig-Portugall I, Montermann E, Kremer A, Reske-Kunz AB,

Sudowe S. Prevention of long-term IgE antibody production by gene

gun-mediated DNA vaccination. J Allergy Clin Immunol 2004;114:

951-7.

10. Jilek S, Walter E, Merkle HP, Corthesy B. Modulation of allergic

responses in mice by using biodegradable poly(lactide-co-glycolide)

microspheres. J Allergy Clin Immunol 2004;114:943-50.

11. O’Hagan DT, Singh M, Ulmer JB. Microparticles for the delivery of

DNA vaccines. Immunol Rev 2004;199:191-200.

12. Farkas L, Kvale EO, Johansen FE, Jahnsen FL, Lund-Johansen F.

Plasmacytoid dendritic cells activate allergen-specific TH2 memory cells:

modulation by CpG oligodeoxynucleotides. J Allergy Clin Immunol

2004;114:436-43.





13. Hajoui O, Janani R, Tulic M, Joubert P, Ronis T, Hamid Q, et al.

Synthesis of IL-13 by human B lymphocytes: regulation and role in IgE

production. J Allergy Clin Immunol 2004;114:657-63.

14. Kawaguchi M, Kokubu F, Odaka M, Watanabe S, Suzuki S, Ieki K, et al.

Induction of granulocyte-macrophage colony-stimulating factor by a new

cytokine, ML-1 (IL-17F), via Raf I-MEK-ERK pathway. J Allergy Clin

Immunol 2004;114:444-50.

15. Henness S, Johnson CK, Ge Q, Armour CL, Hughes JM, Ammit AJ.

IL-17A augments TNF-alpha-induced IL-6 expression in airway smooth

muscle by enhancing mRNA stability. J Allergy Clin Immunol 2004;

114:958-64.

16. Karagiannidis C, Akdis M, Holopainen P, Woolley NJ, Hense G,

Ruckert B, et al. Glucocorticoids upregulate FOXP3 expression and

regulatory T cells in asthma. J Allergy Clin Immunol 2004;114:1425-33.

17. Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M,

et al. HIV evolution: CTL escape mutation and reversion after transmis-

sion. Nat Med 2004;10:282-9.

18. Dorak MT, Tang J, Penman-Aguilar A, Westfall AO, Zulu I,

Lobashevsky ES, et al. Transmission of HIV-1 and HLA-B allele-

sharing within serodiscordant heterosexual Zambian couples. Lancet

2004;363:2137-9.

19. Draenert R, Le Gall S, Pfafferott KJ, Leslie AJ, Chetty P, Brander C,

et al. Immune selection for altered antigen processing leads to cytotoxic

T lymphocyte escape in chronic HIV-1 infection. J Exp Med 2004;199:

905-15.

20. Winchester R, Pitt J, Charurat M, Magder LS, Goring HH, Landay A,

Read JS, et al. Mother-to-child transmission of HIV-1: strong association

with certain maternal HLA-B alleles independent of viral load implicates

innate immune mechanisms. J Acquir Immune Defic Syndr 2004;36:

659-70.

21. Moir S, Malaspina A, Pickeral OK, Donoghue ET, Vasquez J, Miller NJ,

et al. Decreased survival of B cells of HIV-viremic patients mediated by

altered expression of receptors of the TNF superfamily. J Exp Med 2004;

200:587-92.




23. Marshall JS, Jawdat DM. Mast cells in innate immunity. J Allergy Clin

Immunol 2004;114:21-7.

24. Flo TH, Smith KD, Sato S, Rodriguez DJ, Holmes MA, Strong RK, et al.

Lipocalin 2 mediates an innate immune response to bacterial infection by

sequestrating iron. Nature 2004;432:917-21.

25. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, Edberg S, Medzhitov

R. Recognition of commensal microflora by toll-like receptors is required

for intestinal homeostasis. Cell 2004;118:229-41.

26. Gelman AE, Zhang J, Choi Y, Turka LA. Toll-like receptor ligands

directly promote activated CD41 T cell survival. J Immunol 2004;172:

6065-73.

27. Depta JP, Altznauer F, Gamerdinger K, Burkhart C, Weltzien HU,

Pichler WJ. Drug interaction with T-cell receptors: T-cell receptor

density determines degree of cross-reactivity. J Allergy Clin Immunol

2004;113:519-27.

28. Nassif A, Bensussan A, Boumsell L, Deniaud A, Moslehi H,

Wolkenstein P, et al. Toxic epidermal necrolysis: effector cells are

drug-specific cytotoxic T cells. J Allergy Clin Immunol 2004;114:

1209-12.

29. Ng B, Yang F, Huston DP, Yan Y, Yang Y, Xiong Z, Peterson LE, et al.

Increased noncanonical splicing of autoantigen transcripts provides the

structural basis for expression of untolerized epitopes. J Allergy Clin

Immunol 2004;114:1463-70.

30. Nakajima T, Iikura M, Okayama Y, Matsumoto K, Uchiyama C,

Shirakawa T, et al. Identification of granulocyte subtype-selective

receptors and ion channels by using a high-density oligonucleotide

probe array. J Allergy Clin Immunol 2004;113:528-35.

31. Hoffjan S, Ostrovnaja I, Nicolae D, Newman DL, Nicolae R, Gangnon R,

et al. Genetic variation in immunoregulatory pathways and atopic

phenotypes in infancy. J Allergy Clin Immunol 2004;113:511-8.

32. Hanania NA, Sockrider M, Castro M, Holbrook JT, Tonascia J, Wise R,

et al. Immune response to influenza vaccination in children and adults

with asthma: effect of corticosteroid therapy. J Allergy Clin Immunol

2004;113:717-24.

33. Notarangelo L, Casanova JL, Fischer A, Puck J, Rosen F, Seger R, et al.

Primary immunodeficiency diseases: an update. J Allergy Clin Immunol

2004;114:677-87.

34. Cunningham-Rundles C, Sidi P, Estrella L, Doucette J. Identifying

undiagnosed primary immunodeficiency diseases in minority subjects by

using computer sorting of diagnosis codes. J Allergy Clin Immunol 2004;

113:747-55.

35. Holzelova E, Vonarbourg C, Stolzenberg MC, Arkwright PD, Selz F,

Prieur AM, et al. Autoimmune lymphoproliferative syndrome with

somatic Fas mutations. N Engl J Med 2004;351:1409-18.

36. Chinen J, Puck JM. Successes and risks of gene therapy in primary


37. Cavazzana-Calvo M, Lagresle C, Hacein-Bey-Abina S, Fischer A. Gene

therapy for severe combined immunodeficiency. Annu Rev Med 2005;

56:585-602.

38. Gaspar HB, Parsley KL, Howe S, King D, Gilmour KC, Sinclair J, et al.

Gene therapy of X-linked severe combined immunodeficiency by use of

a pseudotyped gammaretroviral vector. Lancet 2004;364:2181-7.

39. Gardulf A, Nicolay U, Math D, Asensio O, Bernatowska E, Bock A,

et al. Children and adults with primary antibody deficiencies gain quality

of life by subcutaneous IgG self-infusions at home. J Allergy Clin

Immunol 2004;114:936-42.

40. Chinen J, Shearer WT. Subcutaneous immunoglobulins: alternative for

the hypogammaglobulinemic patient? J Allergy Clin Immunol 2004;114:

934-5.

41. Bates CA, Ellison MC, Lynch DA, Cool CD, Brown KK, Routes

JM. Granulomatous-lymphocytic lung disease shortens survival in

common variable immunodeficiency. J Allergy Clin Immunol 2004;

114:415-6.

42. Salzer U, Maul-Pavicic A, Cunningham-Rundles C, Urschel S,

Belohradsky BH, Litzman J, et al. ICOS deficiency in patients with

common variable immunodeficiency. Clin Immunol 2004;113:234-40.

43. Sullivan KE. The clinical, immunological, and molecular spectrum of

chromosome 22q11.2 deletion syndrome and DiGeorge syndrome.


44. Moylett EH, Wasan AN, Noroski LM, Shearer WT. Live viral vaccines

in patients with partial DiGeorge syndrome: clinical experience and

cellular immunity. Clin Immunol 2004;112:106-24.

45. Markert ML, Alexieff MJ, Li J, Sarzotti M, Ozaki DA, Devlin BH, et al.

Complete DiGeorge syndrome: development of rash, lymphadenopathy,

and oligoclonal T cells in 5 cases. J Allergy Clin Immunol 2004;113:

734-41.

46. Etzioni A, Ochs HD. The Hyper IgM syndrome—an evolving story.

Pediatr Res 2004;56:519-25.

47. Luo Z, Ronai D, Scharff MD. The role of activation-induced cytidine

deaminase in antibody diversification, immunodeficiency, and B-cell

malignancies. J Allergy Clin Immunol 2004;114:726-35.

48. Orange JS, Jain A, Ballas ZK, Schneider LC, Geha RS, Bonilla FA. The

presentation and natural history of immunodeficiency caused by nuclear

factor kappaB essential modulator mutation. J Allergy Clin Immunol

2004;113:725-33.

49. Orange JS, Levy O, Brodeur SR, Krzewski K, Roy RM, Niemela JE,

et al. Human nuclear factor kappa B essential modulator mutation can

result in immunodeficiency without ectodermal dysplasia. J Allergy Clin

Immunol 2004;114:650-6.

50. Niehues T, Reichenbach J, Neubert J, Gudowius S, Puel A, Horneff G,

et al. Nuclear factor kappaB essential modulator-deficient child with

immunodeficiency yet without anhidrotic ectodermal dysplasia. J Allergy

Clin Immunol 2004;114:1456-62.

51. Wen L, Atkinson JP, Giclas PC. Clinical and laboratory evaluation of

complement deficiency. J Allergy Clin Immunol 2004;113:585-93.

52. Frank MM. Hereditary angioedema: a half century of progress. J Allergy

Clin Immunol 2004;114:626-8.

53. Bowen T, Cicardi M, Farkas H, Bork K, Kreuz W, Zingale L, et al.

Canadian 2003 international consensus algorithm for the diagnosis,

therapy, and management of hereditary angioedema. J Allergy Clin

Immunol 2004;114:629-37.

54. Tangsinmankong N, Kamchaisatian W, Lujan-Zilbermann J, Brown CL,

Sleasman JW, Emmanuel PJ. Varicella zoster as a manifestation of

immune restoration disease in HIV-infected children. J Allergy Clin

Immunol 2004;113:742-63.


AUGUST 2005


Basic

andclin

icalim

munology

Current perspectives

The gastrointestinal tract is critical to thepathogenesis of acute HIV-1 infection

Saurabh Mehandru, MD,a Klara Tenner-Racz, MD,b Paul Racz, MD, PhD,b and

Martin Markowitz, MDa New York, NY, and Hamburg, Germany

It has become evident that the gastrointestinal tract is

preferentially and profoundly depleted of CD41 T cells during

acute HIV-1 infection. The enhanced susceptibility of

gastrointestinal lymphoid tissue to HIV-1 is in part due to the

large complement of CCR51 memory CD41 T cells resident at

this site. Here we summarize the recent findings demonstrating

that the gastrointestinal tract plays a critical role in the

pathogenesis of acute HIV-1 and simian immunodeficiency

virus infections. Ongoing work in this field is likely to have

a significant effect on HIV research in the near future.


Key words: Acute HIV-1, gastrointestinal tract, CD41 T cells

Acute infection is a critical time in the course of HIV-1infection. During this phase, the virus gains access to itstarget cells, infects, replicates, disseminates, and simulta-neously establishes a pool of latently infected cells. Asevidenced by the explosive growth of the epidemic in thelast 2 decades, it is clear that HIV-1 is extremely adeptat accomplishing these tasks. Once established, untreatedHIV-1 infection culminates in profound immunodefi-ciency and death in the majority of individuals. Recentevidence derived from human subjects with acute HIV-1infection and macaques with acute simian immunode-ficiency virus (SIV) infection suggests that the course ofthese infections might be determined during the acutephase of infection. Because the virus entering a new hostmust negotiate a series of obstacles between entry,amplification, and dissemination, it is plausible that in-terventions made during acute HIV-1 infection have thepotential to change the natural history of this disease. Todate, much work has focused on understanding theseevents and have described changes exclusively within theperipheral blood. Until recently, mucosal sites, such asthe gastrointestinal (GI) tract, have been relatively under-

emphasized in the study of acute HIV-1 infection. Here wewill summarize the recent findings suggesting the criticalrole of the lymphoid system of the GI tract during acuteHIV and SIV infection.

It has been well established that the GI tract harbors themajority of the body’s complement of immune cells.1 GItract lymphocytes, placed in close proximity to theexternal environment, are phenotypically distinct fromperipheral blood lymphocytes; the majority of the intes-tinal lymphocytes (>90%) exhibit a memory phenotype.2

In addition, because of constant exposure to a myriad offood and microbial antigens, GI tract lymphocytes aresignificantly more activated than peripheral blood lym-phocytes.2 Furthermore, up to 70% of GI tract lympho-cytes express CCR5, a chemokine receptor that serves asan essential coreceptor for the entry of CCR5-tropic HIV-1into CD41 T cells.3 (In contrast, approximately 20% ofperipheral blood lymphocytes express CCR5.) Thus thevast population of activated memory CD41 T cells withabundant expression of chemokine receptors providesHIV-1 with an ideal environment to establish infection.

Unfortunately, the study of the human GI tract duringacute HIV-1 infection is challenging. It is fraught withdifficulties in identifying individuals during acute infec-tion and the added complexity of obtaining GI tract biopsyspecimens in the face of psychological and physicalcomplications associated with acute HIV-1 infection.Consequently, initial work in this area emerged from theSIV macaque model. A striking depletion of intestinalCD41 T cells was noted in macaques within days of SIVinfection, at a time when little or no CD41T-cell depletionwas evident in the peripheral blood.4,5 Furthermore,intestinal CD41 T-cell depletion occurred regardless ofwhether viral inoculum was delivered intrarectally orintravenously.5 These studies were subsequently extendedto demonstrate that GI tract lymphocyte depletion occursin all stages of SIV infection, including acute infection.4

Studies conducted during the 1990s indicated that intes-tinal CD41 T-cell depletion might be an early feature ofHIV-1 infection as well6 and that intestinal CD41 T-cell

Abbreviations used

GI: Gastrointestinal

SIV: Simian immunodeficiency virus

From aAaron Diamond AIDS Research Center and Rockefeller University,

New York, and bBernhard-Nocht Institut fur Tropenmedizin, Hamburg.

Disclosure of potential conflict of interest: All authors—none disclosed.


2005.


Reprint requests: Saurabh Mehandru, MD, Aaron Diamond AIDS Research

Center, The Rockefeller University, 455 First Ave, 7th Floor, New York,

NY 10016. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.040

419



depletion is more significant than depletion of CD41 Tcells in the peripheral blood.7 However, such studies werenot conducted in patients identified during acute infection.

Along with 2 other groups, we have recently demon-strated that during acute HIV-1 infection, a preferentialand profound depletion of CD41 T cells occurs within theGI tract.8-10 In the first of these studies, Guadalupe et al8

described 2 individuals with acute HIV-1 infection inwhich significant CD41 T-cell depletion occurred withinapproximately 4 to 6 weeks of infection. Brenchley et al10

studied one individual with acute HIV-1 infection (in-fected for <1 month) and 4 individuals with early HIV-1infection (duration of infection was 4-9 months). In all5 subjects, a preferential CD41 T-cell depletion wasnoted in the intestines. In fact, significant GI tract CD41

T-cell depletion was characteristic of all stages of HIV-1infection.9,10

Our initial focus was to describe changes within the GItract of individuals with acute and early HIV-1 infection.To this end, we studied 13 individuals identified duringacute (n = 7) and early (n = 6) infection. Mean CD41 T-cell percentage in the GI tract was 15.7% 6 3.6%compared with a mean of 42.3% 6 14.7% in the blood(P < .001). Thus in all of our subjects, profound CD41

T-cell depletion was observed in the GI tract and wassignificantly greater than depletion in the peripheral blood(Fig 1). Since then, we have extended our studies to26 subjects with acute and early HIV-1 infection andhave observed far greater CD41 T-cell depletion in theintestines compared with in the peripheral blood (unpub-lished data). Having demonstrated a striking CD41 T-celldepletion within the GI tract, we have examined thesubsets of CD41 T cells in which this depletion was mostprominent. Because a majority of intestinal CD41 T cellsexpress CCR53 and given that viruses during the earlystages of HIV-1 infection are predominantly CCR5-tropic,11 it was not surprising that Brenchley et al10 andwe9 both observed that GI tract lymphocyte depletion wasmost significant among the CCR51 subsets of CD41 Tcells. In addition, we have observed marked CD41 T-celldepletion in the effector sites (lamina propria) of the GItract, with relative sparing of the inductive sites (organizedlymphoid tissue). In contrast, however, HIV-1 RNA waslocalized to the inductive sites. We believe that thisdiscrepancy is best explained by loss of target cells inthe effector compartment.We hypothesize that if we couldexamine a subject within the first 7 to 10 days of infection,viral RNA would be evident in the effector compartment

FIG 1. Effector sites (lamina propria) of the GI tract show pronounced CD41 T-cell depletion in acute and early

HIV-1 infection. At magnifications of 253 (A) and 803 (B), CD41 T cells (stained red) are seen in a

representative biopsy specimen from an HIV-uninfected individual. In comparison, profound CD41 T-cell

depletion is noted in low-power (C, 253) and high-power (D, 803) views in a specimen from a subject with

acute HIV-1 infection.


AUGUST 2005

420 Mehandru et al

Basic

andclin

icalim

munology

as well before the annihilation of the resident CD41 cells.In a small cross-sectional cohort, we have observed that incontrast to the peripheral blood, restoration of CD41 Tcells in the GI compartment with highly active antiretro-viral therapy is incomplete. We are currently examiningpatients longitudinally to determine whether the timing ofonset of antiretroviral therapy is associated with better GItract CD41 T-cell reconstitution, as has been suggestedby Guadalupe et al8 in a limited cohort comprised of 2subjects.

Recent data obtained from the SIV macaque modelprovide startling insights into the degree of GI tractCD41 T-cell infection and depletion as well. Studiesby Mattapallil et al12 and Li et al13 demonstrate rapidinfection and destruction of GI tract memory CD41T cellswithin days of infectionwith SIV. The study byMattapallilet al12 suggests that at peak viremia (day 10), as many as60% of GI tract memory CD41 T cells might be infectedwith SIV and that these cells are lost within 14 days ofinfection. This results in profound immunodeficiencythat begins within days of infection, not months to yearsas was previously thought. The authors put forth the notionthat memory CD41 T cells are killed by means of direct,virus-mediated destruction rather than bystander effects orsuppression of CD41 T-cell production.

Li et al13 showed, somewhat surprisingly, that GI tractcells infected initially with SIV have a nonactivated(CD692CD252Ki672) phenotype. Peak infection ofmemory CD41 T cells within the GI tract correspondedto peak viremia, and depletion of GI tract CD41 T cellscoincided with a decrease in the peripheral viral load. Liet al therefore suggested the concept of ‘‘substrate de-pletion,’’ resulting in viral load reduction in the host. Incontrast to Mattapallil et al,12 Li et al13 propose that CD41

T-cell depletion is caused by virus-triggered, Fas–Fasligand–mediated apoptosis. It is likely that CD41 T-celldestruction is multifactorial, caused by virus-inducedcytolysis, apoptosis, and the host’s own cytotoxic T-lymphocyte, natural killer cell responses. Further workneeds to be done to resolve this issue.

Concurrent studies in our laboratory have focused ondetermining virologic and immunologic correlates of GItract CD41 T-cell depletion. Our data (unpublished)suggest that acute HIV-1 infection results in immunologicactivation and proliferation of GI tract CD41 T cells,creating a local niche of viral replication. We haveobserved a highly significant difference between the levelof CD41 T-cell infection in the GI tract and peripheralblood. In addition, we have observed a striking differencein HIV-1 viral RNA production within CD41 T cellsderived from the GI tract compared with from the periph-eral blood. Thus we hypothesize that during acute infec-tion, HIV-1 encounters a vast population of susceptiblecells within the GI tract and preferentially infects them.Viral infection results in immune activation and CD41

T-cell proliferation, both of which augment viral produc-tion, setting up the next round of infection. The end re-sult is profound CD41 T-cell depletion within the GItract. Combined, the results from recent studies in human

subjects and SIV-infected macaques suggest an emergingmodel for the pathogenesis of HIV-1 infection. The largecomplement of resident memory cells in mucosal surfacesget preferentially infected and fuel subsequent roundsof replication and infection. Therefore in acute stagesmucosal sites appear to propel the infection forward.

What are the implications of these findings? A numberof significant issues emerge from these studies:

1. By demonstrating that significant viral replication andimmune depletion occurs at mucosal sites duringacute infection, these findings provide compellingevidence to the argument that mucosal sites shouldbe the focus of further examination and should beconsidered in the monitoring of patients on therapy.

2. These findings shatter the dogma that during acuteHIV-1 infection, there is little CD41 T-cell depletionin the body.

3. These findings reinvigorate the debate regarding thetiming of therapy in HIV-1 infection. Guidelinesrecommending that treatment need not be initiateduntil CD41 T-cell count decreases to less than 350cells/mm3 or until plasma viral load is more than100,000 copies/mL14 should take into account theserecent data demonstrating severe mucosal CD41

T-cell depletion in acute and early HIV-1 infection.This said, however, it must be mentioned that theclinical significance of mucosal immune depletionremains uncertain. Redundancy in our immune systemmight prevent long-term consequences, yet the pos-sibility of early immune senescence exists, and thelong-term consequences are not clear.

4. These findings suggest a potential for the use of im-munomodulators, such as cyclosporine, during acuteHIV-1 infection with the goal of curtailing successiverounds of viral infection and CD41 T-cell depletionwithin the GI tract.

5. With regard to preventive efforts directed againstHIV-1, recent data underscore the need to developstrategies to protect mucosal surfaces from infection.The use of microbicides and CCR5 blockers wouldrepresent such interventions.

6. Finally and perhaps most importantly, these findingsreemphasize the fact that mucosal immune responsesmust be targeted in the development of effectiveHIV-1 vaccines.

The weight of current evidence places mucosal lym-phoid tissue as pivotal in HIV-1 pathogenesis. Thesefindings are likely to have a significant effect on HIVresearch in the near future.

REFERENCES

1. Mowat AM. Anatomical basis of tolerance and immunity to intestinal

antigens. Nat Rev Immunol 2003;3:331-41.

2. Schieferdecker HL, Ullrich R, Hirseland H, Zeitz M. T cell differenti-

ation antigens on lymphocytes in the human intestinal lamina propria.

J Immunol 1992;149:2816-22.

3. Anton PA, Elliott J, Poles MA, McGowan IM, Matud J, Hultin LE, et al.

Enhanced levels of functional HIV-1 co-receptors on human mucosal



Mehandru et al 421


T cells demonstrated using intestinal biopsy tissue. AIDS 2000;14:

1761-5.

4. Smit-McBride Z, Mattapallil JJ, McChesney M, Ferrick D, Dandekar S.

Gastrointestinal T lymphocytes retain high potential for cytokine

responses but have severe CD4(1) T-cell depletion at all stages of

simian immunodeficiency virus infection compared to peripheral lym-

phocytes. J Virol 1998;72:6646-56.

5. Veazey RS, DeMaria M, Chalifoux LV, Shvetz DE, Pauley DR, Knight

HL, et al. Gastrointestinal tract as a major site of CD41 T cell depletion

and viral replication in SIV infection. Science 1998;280:427-31.

6. Lim SG, Condez A, Lee CA, Johnson MA, Elia C, Poulter LW. Loss

of mucosal CD4 lymphocytes is an early feature of HIV infection. Clin

Exp Immunol 1993;92:448-54.

7. Schneider T, Jahn HU, Schmidt W, Riecken EO, Zeitz M, Ullrich R.

Loss of CD4 T lymphocytes in patients infected with human immuno-

deficiency virus type 1 is more pronounced in the duodenal mucosa than

in the peripheral blood. Berlin Diarrhea/Wasting Syndrome Study

Group. Gut 1995;37:524-9.

8. Guadalupe M, Reay E, Sankaran S, Prindiville T, Flamm J, McNeil A,

et al. Severe CD41 T-cell depletion in gut lymphoid tissue during

primary human immunodeficiency virus type 1 infection and substantial

delay in restoration following highly active antiretroviral therapy. J Virol

2003;77:11708-17.

9. Mehandru S, Poles MA, Tenner-Racz K, Horowitz A, Hurley A, Hogan

C, et al. Primary HIV-1 infection is associated with preferential depletion

of CD41 T lymphocytes from effector sites in the gastrointestinal tract.

J Exp Med 2004;200:761-70.

10. Brenchley JM, Schacker TW, Ruff LE, Price DA, Taylor JH, Beilman

GJ, et al. CD41 T cell depletion during all stages of HIV disease occurs

predominantly in the gastrointestinal tract. J Exp Med 2004;200:749-59.

11. Connor RI, Sheridan KE, Ceradini D, Choe S, Landau NR. Change

in coreceptor use coreceptor use correlates with disease progression in

HIV-1–infected individuals. J Exp Med 1997;185:621-8.

12. Mattapallil JJ, Douek DC, Hill B, Nishimura Y, Martin M, Roederer M.

Massive infection and loss of memory CD41 T cells in multiple tissues

during acute SIV infection. Nature 2005;434:1093-7.

13. Li Q, Duan L, Estes JD, Ma ZM, Rourke T, Wang Y, et al. Peak SIV

replication in resting memory CD41 T cells depletes gut lamina propria

CD41 T cells. Nature 2005;434:1148-52.

14. Guidelines for the use of antiretroviral agents in HIV-infected adults

and adolescents. Washington (DC): Department of health and Human

Services; 2005.


AUGUST 2005

422 Mehandru et al

Basic

andclin

icalim

munology

Editorial

Are you immunodeficient?

Francisco A. Bonilla, MD, PhD, and Raif S. Geha, MD Boston, Mass

Key words: Clinical immunology, primary immunodeficiency, in-

fectious diseases, genetics

In their Rostrum monograph, Casanova et al1 considerthe problems of definition and classification of primaryimmunodeficiencies (PIs). They begin with a standard andstraightforward premise: ‘‘immunodeficiency is a failureto achieve immune function to provide efficient, self-limited host defense against the biotic and abioticenvironment while preserving tolerance to self.’’ Thechallenge to practitioners is to translate this axiom intoprinciples that answer specific clinical and academicneeds. One essential message of Casanova et al is toconsider the susceptibility to infection limited to one or afew pathogens and having Mendelian inheritance to bewithin the spectrum of PI, regardless of the immunologicphenotype. They further argue that in light of this,academic and clinical needs will best be met by aclassification system on the basis of clinical phenotype,in contrast to the classic, or perhaps traditional, system onthe basis of immunologic phenotype.2,3 According to theirscheme, Casanova et al1 distinguish conventional andunconventional immunodeficiencies as those that do ordo not have clearly defined immunologic phenotypes,respectively. The authors consider these to be end pointsof a spectrum rather than dichotomous. They further statethere has ‘‘never been a fully satisfactory classification ofPID.’’ However, it is worth discussing whether any singlesystem on the basis of clinical phenotype or immunologicphenotype could ever be so.

Regardless of how we might classify their diseases,immunodeficient patients come to clinical attention pre-dominantly as a result of a predisposition to infection. Thispredisposition is manifested in one or more of the clinicaldimensions of infection: the inherent virulence of the

organism, the site of infection (localized vs disseminated),the infection’s severity (degree of tissue or organ damage),the infection’s persistence or resistance to therapy, and thefrequency of relapse or reinfection. In spite of the com-monality of these considerations early in the approach tothe potentially immunodeficient patient, there are very fewdata or agreement on where to best draw the dividing linebetween normal and abnormal along any of these dimen-sions. It might also be important to consider that this linecan be drawn differently under circumstances that differwith respect to, for example, the level of public hygiene,the prevalence of particular pathogens, or the availabilityof vaccinations. (Casanova et al,1 in fact, consider thesefactors as masking the true prevalence of PI.) In addition,Casanova et al consider the importance of Mendelian(single gene) inheritance in a clinical definition. Perhapsthis element could be generalized to any definable geneticcomponent to include interactions among mutations,polymorphisms, or both that might determine a pheno-type, as has been observed in some PI diseases.4 Whetherone is prepared to alter one’s conceptualization of immunedeficiency to include unconventional forms, the matter ofdefinition requires further study from all sides (epidemi-ology, immunology, and genetics) to provide a more solidframework for further discussion. Ultimately, the issue ofwhere to draw the line between normal and abnormal iscritical if the search for ‘‘currently unknown Mendelianprimary immunodeficiencies’’1 is to set out with hope formeaningful discovery.

The criteria of normalcy in a system based on immu-nologic phenotype relate to population distributions ofscreening laboratory studies of immune function (eg,serum immunoglobulin levels and specific antibody titers,peripheral blood lymphocyte subpopulations, in vivo orin vitro measures of T-cell function, assays of phagocyteoxidative burst or adhesion, and complement functionor serum component levels).2 This system affords con-venient statistical labels for normal and abnormal.Unfortunately, the biologic and clinical correlates arealso largely lacking here, as in the discussion of predis-position to infection above. This sometimes leads to animportant point raised by Casanova et al1 that certainlybears repeating: ‘‘patients with specific clinical infectious

Abbreviation used

PI: Primary immunodeficiency

From the Division of Immunology, Children’s Hospital, and the Department

of Pediatrics, Harvard Medical School, Boston, Mass.

Disclosure of potential conflict of interest: F. A. Bonilla has consultant

arrangements with Talecris Biotherapeutics and is on the speakers’ bureau

for Accredo Therapeutics. R. Geha has no conflicts of interest to disclose.


2005.


Reprint requests: Francisco A. Bonilla, MD, PhD, Children’s Hospital,

Immunology, Enders 809, 300 Longwood Ave, Boston, MA 02115.



0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.026

423



diseases but no overt immunological phenotype [are]being largely neglected.’’

However, this is not precisely a failure of a system orclassification based on immunologic phenotype, as muchas it reflects 3 distinct elements. First, one might choosesimply not to categorize a limited infection predispositionas a PI, one of the main points of Casanova et al1 (ie, if it isdefinable genetically, then we should). Second, we mustconsider further what is meant by immunologic pheno-type. Casanova et al use the adjective ‘‘overt’’ to denote anabnormality detectable by the battery of screening testslisted above. However, at some level, all genetically de-finable PIs must have some immunologic phenotype; it isthe clinical phenotype that dictates the specific eval-uation that will ultimately identify it (therefore we canagree on the value of the clinical classification here). Andfinally, it is clear that we have incomplete knowledgeregarding how the organism as a whole (including theimmune system) protects itself from infection. Only inrecent years have the importance of defects of naturalkiller cell function5 and toll-like receptor signaling6,7

become the foci of attention in PI. It is inevitable that theclinical screening immunologic evaluation of patients willcontinue to develop along with expanding knowledge ofimmune mechanisms.

In the final analysis, the matter of classification mighttruly be secondary. Consider Table I, which shows a verysimple scheme outlining the biologic and clinical elementsof PI and the classification systems that might be proposedin response to or motivated by these different aspects ofthe interaction of a pathogen with a susceptible host. Nosystem is superior in all cases, and some might even befrankly cumbersome in some situations, but there is nosingle system that optimally serves all needs.

We all work together toward the determination ofcomplete sets or elements of a PI knowledge base, suchas the complete set of gene alterations that lead tosusceptibility to infection, the complete set of clinicalphenotypes of PIs, and the complete set of immunologicphenotypes. These will not be conveniently ordered along1 or 2 dimensions that will be useful in every situation,as outlined above. What might serve best is a multiseg-mented database in which each segment orders the infor-mation according to a distinct classification system andevery entry is linked to its entries in the other segments.This type of organization is legion on the Internet (see,for example, the National Center for BiotechnologyInformation of the National Institutes of Health, Bethesda,Maryland, at http://www.ncbi.nih.gov/, and the Institutefor Medical Technology Bioinformatics Group of theUniversity of Tampere, Tampere, Finland, at http://bioinf.uta.fi/).

The adaptive immune system might be dispensable forhuman development and survival in a germ-free environ-ment.8 On the other hand, elements of innate immunityinteract with commensal flora and are required for normalfunction of some systems.9 In light of the interrelations ofimmunity with other organ systems, the boundaries of theimmune system, as a whole, become less distinct.Whether

it is truly ‘‘the least efficient physiological system at theindividual level’’1 and whether we might discover that alarge fraction of human subjects are immunodeficientdepends on one’s perspective, as we have discussed, not-withstanding the benefits of public hygiene, immuniza-tion, and antibiotics. This is apparent if one shifts focusfrom the anthropocentric view and remembers that humanpathogens and human beings evolve together.

At the bedside, the clinical immunologist’s interestis piqued by the constellation of history, symptoms, andfindings. Some version of the clinical classification ofwhich Casanova et al1 speak is foremost in our mindswhile we determine the most efficient path towarddefining the immunologic phenotype, making a definitivediagnosis, or both. Experience informs us that our abilityto define the immunologic phenotype with precisiondepends utterly on the sophistication of the laboratorymethods available for study of the individual patient. Wenote in passing that recent advances in molecular methodscan, in some instances, divorce the processes of definingthe immunologic phenotype from making a diagnosis inthe case of a PI that has already been defined at themolecular level. For example, a 15-month-old boy pre-sents with severe recurrent respiratory tract infections withencapsulated bacteria.We sequence hisBTK gene and finda mutation or deletion consistent with X-linked agamma-globulinemia. We have established a diagnosis withoutknowing whether he is agammaglobulinemic or B lym-phopenic.We do not advocate such an approach, however,because it perpetuates or even creates critical gaps in ourknowledge base.

As we stated, the example applies only where themolecular defect is known. The situation is differentfor the patient with a less well-understood form of PI.The level of laboratory sophistication required for thedefinition of new forms of PI is an order of magnitude

TABLE I. Clinical-academic considerations or questions

in PI diseases and potential classification schemes to

address them

Biomedical aspects

of immunodeficiency

Basis for

classification of PI

Infecting microbe(s) Microbial taxonomy

Host with genetic

lesion(s)-polymorphism(s)

Genetic catalog, mode

of inheritance

Altered immunopathogenesis

of infection

Biochemical and cell

biologic mechanisms

Immune dysregulation

(atopy, autoimmunity,

lymphoproliferation, malignancy)

Biochemical and cell

biologic mechanisms

Clinical syndrome of

immunodeficiency

All clinical features

of the disease

Diagnostic evaluation Immunologic phenotype

Therapy Anti-infective, immune

reconstitution, response

to therapy

Outcome Natural history, prognosis


AUGUST 2005

424 Bonilla and Geha

Basic

andclin

icalim

munology

http://www.ncbi.nih.gov/

http://bioinf.uta.fi/

http://bioinf.uta.fi/

beyond what suffices for diagnosis of known entities.Technologic advances might soon provide us with theability to automate functional studies of pathogen-specificimmune responses and to link these to genomics-derivedstrategies to identify loci for study.10 Wherever they exist,these resources must be made available in some way to thelarger community of clinical immunologists. The impor-tance of this cannot be overstated. We agree withCasanova et al1 that the full spectrum of human suscep-tibility to infection is largely waiting to be discovered. Forthose with PI and those who study it, hope derives from achance encounter with ‘‘a prepared mind,’’ timely recog-nition, and the technology to repair it.11,12

REFERENCES

1. Casanova J-L, Fieschi C, Bustamante J, Reichenbach J, Remus N,

von Bernuth H, et al. From idiopathic infectious diseases to novel primary


2. Bonilla FA, Bernstein IL, KhanDA,Ballas ZK, Chinen J, FrankMM, et al.

Practice Parameter for the diagnosis and management of primary immu-

nodeficiency. Ann Allergy Asthma Immunol 2005;94(suppl):S1-63.



2004;114:677-87.

4. Foster CB, Lehrnbecher T, Mol F, Steinberg SM, Venzon DJ, Walsh TJ,

et al. Host defense molecule polymorphisms influence the risk for

immune-mediated complications in chronic granulomatous disease.

J Clin Invest 1998;102:2146-55.

5. Orange JS. Human natural killer cell deficiencies and susceptibility to

infection. Microbes Infect 2002;4:1545-58.

6. Orange JS, Levy O, Brodeur SR, Krzewski K, Roy RM, Niemela JE,

et al. Human nuclear factor kappa B essential modulator mutation can

result in immunodeficiency without ectodermal dysplasia. J Allergy Clin

Immunol 2004;114:650-6.



Science 2003;299:2076-9.

8. Guerra IC, Shearer WT. Environmental control in management of

immunodeficient patients: experience with ‘‘David’’. Clin Immunol

Immunopathol 1986;40:128-35.

9. Rakoff-NahoumS, Paglino J, Eslami-Varzaneh F, Edberg S,Medzhitov R.

Recognition of commensal microflora by toll-like receptors is required for

intestinal homeostasis. Cell 2004;118:229-41.

10. Tian Q, Stepaniants SB, Mao M, Weng L, Feetham MC, Doyle MJ, et al.

Integrated genomic and proteomic analyses of gene expression in

Mammalian cells. Mol Cell Proteomics 2004;3:960-9.

11. Aiuti A, Slavin S, Aker M, Ficara F, Deola S, Mortellaro A, et al.

Correction of ADA-SCID by stem cell gene therapy combined with

nonmyeloablative conditioning. Science 2002;296:2410-3.

12. Cavazzana-Calvo M, Hacein-Bey S, de Saint Basile G, Gross F, Yvon E,

Nusbaum P, et al. Gene therapy of human severe combined immuno-

deficiency (SCID)-X1 disease. Science 2000;288:669-72.



Bonilla and Geha 425


Rostrum

From idiopathic infectious diseases tonovel primary immunodeficiencies

Jean-Laurent Casanova, MD, PhD,a,b Claire Fieschi, MD, PhD,a,c Jacinta Bustamante, MD,a

Janine Reichenbach, MD,a,d Natasha Remus, MD,a,e Horst von Bernuth, MD,a and

Capucine Picard, MD, PhDa,b Paris and Creteil, France, and Frankfurt, Germany

Primary immunodeficiencies are typically seen as rare

monogenic conditions associated with detectable immunologic

abnormalities, resulting in a broad susceptibility to multiple

and recurrent infections caused by weakly pathogenic and

more virulent microorganisms. By opposition to these

conventional primary immunodeficiencies, we describe

nonconventional primary immunodeficiencies as Mendelian

conditions manifesting in otherwise healthy patients as a

narrow susceptibility to infections, recurrent or otherwise,

caused by weakly pathogenic or more virulent microbes.

Conventional primary immunodeficiencies are suspected on the

basis of a rare, striking, clinical phenotype and are defined on

the basis of an overt immunologic phenotype, often leading to

identification of the disease-causing gene. Nonconventional

primary immunodeficiencies are defined on the basis of a more

common and less marked clinical phenotype, which remains

isolated until molecular cloning of the causal gene reveals a

hitherto undetected immunologic phenotype. Similar concepts

can be applied to primary immunodeficiencies presenting other

clinical features, such as allergy and autoimmunity.

Nonconventional primary immunodeficiencies thus expand

the clinical boundaries of this group of inherited disorders

considerably, suggesting that Mendelian primary

immunodeficiencies are more common in the general

population than previously thought and might affect children

with a single infectious, allergic, or autoimmune disease.


Key words: Primary immunodeficiency, infectious diseases, idio-pathic infections, inborn errors, Mendelian disorders, predisposi-

tion to infection

An immunologic definition and classification of pri-mary immunodeficiencies currently prevails and is ex-pected to do so for the foreseeable future.1 Unfortunately,this has resulted in studies of otherwise healthy patientswith specific infectious clinical diseases but no overtimmunologic phenotype being largely neglected. Theattention of most investigators and clinicians has remainedfocused on the tip of the iceberg: those rare patients with anoisy clinical phenotype (multiple, recurrent, and severeinfections) and a visible immunologic phenotype (definingthe primary immunodeficiency). The most striking exam-ple of such conventional primary immunodeficiencies isreticular dysgenesia, an exceedingly rare disorder associ-atedwith agranulocytosis and alymphocytosis, resulting inearly-onset vulnerability to virtually all microorganismsand a rapidly fatal outcome in the absence of hematopoieticstem cell transplantation.2 Immunodeficiency is com-monly ruled out in patients with a single severe infectiousdisease (even if recurrent or life-threatening) and normalroutine immunologic workup (searching for signs ofinherited or acquired immunodeficiency). Self-contradic-tory titles in the medical literature, such as ‘‘Fatal infectionin an immunocompetent individual,’’ remain common.Unusual infectious diseases are often described as idio-pathic, demonstrating caution and a desire to avoid thedirect incrimination of the patient’s genetic background.However, some infections typically caused by weaklyvirulent (opportunist) microbes have been found tobe associated with a high frequency of familial forms,parental consanguinity, or both, suggesting Mendelianpredisposition. This group of nonconventional primaryimmunodeficiencies is characterized by a very narrowspectrum of opportunistic infections limited to one micro-bial genus or species possibly, but not necessarily, recur-rent in otherwise healthy patients with no detectableimmunologic abnormality on initial investigation.3 Thesediseases do not fit easily into the classical classification ofprimary immunodeficiencies.

Nonconventional primary immunodeficiencies includethe syndromes of Mendelian susceptibility to mycobacte-rial diseases (OMIM 209950,4 first described in 1951) inpatients with mutations in the IL-12/23–IFN-g circuit(first identified in 1996)5-8; recurrent invasive diseasecaused by Neisseiria species in patients with mutationsaffecting the terminal components of complement (C5 toC9) forming the membrane attack complex (first described

From aLaboratoire de Genetique Humaine des Maladies Infectieuses,

Universite de Paris Rene Descartes-INSERM U550, Faculte de Medecine

Necker, Paris; bUnite d’Immunologie et d’Hematologie Pediatriques,

Hopital Necker Enfants Malades, Paris; cService d’Immunologie

Clinique, Hopital Saint Louis, Paris; dKlinik fur Kinderheilkunde,

Klinikum der J.W. Goethe Universitat, Frankfurt; and eService de

Pediatrie, Centre Hospitalier Intercommunal de Creteil, Creteil.

Our laboratory is supported in part by grants from the BNP-Paribas and

Schlumberger foundations, the Institut Universitaire de France, and the EU

grant QLK2-CT-2002-00846.




Reprint requests: Jean-Laurent Casanova, MD, PhD, Laboratoire de Genetique

Humaine des Maladies Infectieuses, Universite de Paris Rene

Descartes-INSERM U550, Faculte de Medecine Necker, Paris 75015,

France, EU. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.053

426

Basic

andclin

icalim

munology


in 1974)9,10; isolated chronic mucocutaneous candidiasis(OMIM 114580, first described in 1969), which remainsunexplained genetically11,12; epidermodysplasia verruci-formis with disseminated warts caused by human papil-lomaviruses belonging to group B1 (OMIM 226400, firstdescribed clinically in 1922 and subsequently shown to bea Mendelian trait [1939] conferring susceptibility topapillomaviruses [1946-1966]) in patients with mutationsin EVER1 and EVER2 (first described in 2002)13,14; andX-linked lymphoproliferative syndrome caused byEpstein-Barr virus (OMIM 308240, first described in1975) in patients with mutations in SAP (first describedin 1998).15,16 Needless to say, the dichotomy betweenconventional and nonconventional conditions is some-what artificial because there is really a continuum betweenthese 2 extremes.17 Patients with a recently describedconventional primary immunodeficiency, IL-1 receptor-associated kinase 4 deficiency, are particularly susceptibleto Streptococcus pneumoniae,18,19 and conversely, pa-tients with mutations in the IL-12–IFN-g axis are alsosusceptible to Salmonella species.20 In any event, neitherthe identification of a cellular phenotype nor that of thecausal gene suggested the existence of an underlyingprimary immunodeficiency in patients with nonconven-tional primary immunodeficiencies. Instead, primary im-munodeficiency diagnosis was based on the relatively lowvirulence of the microbe and the seemingly Mendelianinheritance of predisposition to severe disease.

It would not be wise to limit the group of nonconven-tional primary immunodeficiencies to these 5 Mendeliansyndromes, to patients presenting unexplained infectionscaused by weakly virulent opportunistic microorganisms,or even to patients with recurrent infections caused bymore virulent pathogens. There is good reason to believethat other human conditions reflect currently unknownMendelian primary immunodeficiencies. First, geneticepidemiologic studies searching for familial forms andparental consanguinity have not been carried out for mostinfectious, autoimmune, and allergic clinical syndromes.Second, neither the absence of familial cases nor thelack of consanguinity are sufficient to exclude Mendeliandefects, and sporadic cases might reflect a genetic lesion,as illustrated by the first genetic lesion discovered inhuman subjects, trisomy 21, in patients with Downsyndrome.21 Third, the virulence of microorganisms isalso a continuum, and many pathogenic microbes, suchas Mycobacterium tuberculosis, are actually innocuous inmost human beings. A number of common infectiousdiseases are likely to reflect nonconventional primaryimmunodeficiencies in at least a fraction of patients.Consistent with this view, mycobacterial diseases causedby weakly virulent BCG species were described asidiopathic infections before the identification of defectsin the IL-12–IFN-g circuit.22,23 The identification of thesedefects has led to the recent description of 3 unrelatedfamilies with a purelyMendelian form of predisposition tobona fide tuberculosis,24-26 following on from the obser-vation that IL-12Rb1 deficiency had low penetrance forthe case-definition phenotype of clinical disease caused by

weakly virulent mycobacteria.20,27 Currently unexplainedcandidate infectious diseases include invasive pneumo-coccal disease18,19 and herpes simplex encephalitis,28,29

which have been diagnosed in at least a few patients withconventional immunodeficiencies. Many life-threateninginfectious diseases might well turn out to result from theMendelian inheritance of a specific predisposition, reflect-ing a nonconventional primary immunodeficiency.

Nonconventional primary immunodeficiencies are de-fined on clinical grounds, raising the issue of the classifi-cation of primary immunodeficiencies. In fact, there hasnever been a fully satisfactory classification of primaryimmunodeficiencies.1,30,31 This problem has become in-creasingly acute because of the explosion of knowledgein the field in the last 20 years, with at least 200 conditionsdescribed clinically and more than 100 disease-causinggenes identified. Moreover, many more conventional andnonconventional primary immunodeficiencies are likelyto be identified in the near future. A genetic classificationwas impossible in the early days before identification of thedisease-causing genes. Even today, genetic classificationwould be hindered by the lack of a well-defined temporaland spatial expression pattern for the disease-causinggenes, limiting our understanding of pathogenesis.Furthermore, different clinical syndromesmight be causedby different mutations in the same gene, and the samesyndrome might be caused by different genetic causes.Even if it were possible, a genetic classification would notactually be sufficient because phenotypes are obviouslymore important than genotypes; the chief value of agenotype lies in its ability to account for a given pheno-type. Accordingly, the McKusick catalog of humangenetic disorders is merely a catalog and not a classifica-tion.4 Any classification system for academic and clinicalpurposesmust therefore be primarily phenotypic, althoughimprovements in our understanding of the genetic basisof primary immunodeficiencies might lead to changesin phenotypic classification. The phenotypic definition ofprimary immunodeficiencies is clearly the necessarystarting point for phenotypic classification.

Historically, the identification of agammaglobulinemiain 1952 by Ogden Bruton32 and the subsequent discoverythat its inheritance was X-linked and recessive33 was theorigin of current classifications of primary immunodefi-ciencies on the basis of a combination of immunologicphenotypes and modes of inheritance.1,30,31,34-39 Primaryimmunodeficiencies are commonly classified into disor-ders of T cells, B cells, phagocytes, and complement.They are then further classified according to the modeof inheritance and, when known, genetic cause. Thismethod of classification poses a serious problem of de-finition because it tightly links the concept of primary im-munodeficiency with the observation of an immunologicphenotype. According to this view, even asymptomaticIgA-deficient individuals are immunodeficient, unlike,paradoxically, patients dying of infectious disease with-out immunologic abnormality. Moreover, because manydisease-causing genes are expressed in different cell typesin which mutant alleles might have different effects,



Casanova et al 427


clinical phenotypes, whether infectious, allergic, or auto-immune, are far from being consistent within each of the4 cell type–based groups. What is the clinical similaritybetween defects of C1 inhibitor and C9? Conversely,X-linked agammaglobulinemia (primarily, but not exclu-sively, a B-cell defect) clinically resembles HLA class Ideficiency (often improperly classified as a T-cell defect).This situation also results in paradoxical classifications,with an immunologic phenotype attributed to one cell typebut a genetic defect actually affecting another cell type.For example, CD40L deficiency is generally described asa B-cell defect because of the hyper-IgM syndrome,40 butCD40L is expressed on T cells and is involved in theinteraction of T cells with both B cells and macrophages–dendritic cells. The focus on certain immunologic pheno-types, such as hyper-IgM syndrome, is misleading initself: there are more differences than common pointsbetween patients with mutations in CD40L (expressedon T cells), AID (expressed in B cells), and NEMO(expressed ubiquitously), all of which can result in hyper-IgM syndrome. The current immunologic classificationof primary immunodeficiencies is thus imperfect bothimmunologically and clinically.

An ideal definition and classification of primary immu-nodeficiencies and inborn deficiencies should evidentlyrely on clinical phenotype because this best reflects thephysiologic effect of any deleterious genotype. Indeed,immunodeficiencies in general, whether inherited or ac-quired, should be defined clinically, as opposed to immu-nologically. Would anyone seriously suggest that it wouldbe better to define respiratory failure in terms of epithelialabnormalities rather than the physiologic consequencesof insufficient oxygen inhalation? It is certainly useful toassess various parameters in the course of any organfailure, but the definition and monitoring of organ failuremust be physiologic. Immunodeficiency is a failure toachieve immune function to provide efficient, self-limitedhost defense against the biotic and abiotic environmentwhile preserving tolerance to self. Immunodeficiencies arethus best defined in terms of the diverse forms of life-threatening infections, allergies, or autoimmune reactions.The detection of an identifiable immunologic abnormalityis less important and depends on the tools available to theinvestigator. Immunodeficiencies might also be consid-ered in terms of whether they are inherited or acquired.Most immunodeficiencies are actually idiosyncratic, re-flecting both nature (genetic background) and nurture (theeffect of the environment on the host). An ideal classifi-cation of primary immunodeficiencies should take thisinto account, considering infectious syndromes one by one(and possibly autoimmune and allergic syndromes aswell). For example, primary immunodeficiencies associ-ated with mycobacterial41 or pneumococcal42 diseases arespecifically associated with defects of interaction betweenT-cells and phagocytes (involving in particular the IL-12/23–IFN-g circuit and the respiratory burst) and bacterialsensing and opsonization (involving mucosal inflamma-tion, complement, carbohydrate-specific antibodies, andsplenic macrophages), respectively. This classification is

not operational yet because it awaits investigators in thefield to review all pathogens one by one, perhaps in acollaborative effort (eg, primary immunodeficiencies as-sociatedwith, for example,Pneumocystis species infectionor Toxoplasma species infection). This classification willaddress the most relevant clinical question at the bedside(Which immunodeficiency should the physician considerin a given infected patient?) and the most relevant immu-nologic question at the bench (What role does a particularmolecule play in immunity to infection in vivo?). Ofcourse, the genotype and immunologic phenotype areinvaluable both clinically to tailor treatment options toindividual patients and immunologically to decipher themolecular basis of immune responses.

A corollary of this purely clinical definition and clas-sification of primary immunodeficiencies is that inbornMendelian deficiencies of immunity are more commonthan initially thought. Accordingly, newly described pri-mary immunodeficiencies, such as partial IFN-gR1 andsignal transducer and activator of transcription 1 defi-ciencies, have been shown to be transmitted as autosomaldominant traits in multiplex families, at odds with theclassical view that primary immunodeficiencies are nec-essarily recessive traits because of their severity.43 Not allsevere infectious diseases will be found to reflect aMendelian primary immunodeficiency or to be due tothe inheritance of a major susceptibility gene, as seen inleprosy with mutations in Parkin,44-46 because predispo-sition to infection might display truly polygenic determi-nism. Nevertheless, it will be important in the future todecipher the Mendelian genetic basis of infectious dis-eases. Studies of autoimmune and allergic syndromes arealso likely to reveal novel Mendelian disorders. Once aclinical definition of immunodeficiency is accepted, pa-tients with infectious diseases, allergy, or autoimmunity(in the broad sense of these terms, including angioedema,hemophagocytosis, and autoinflammation) should beconsidered as potential bearers of Mendelian primaryimmunodeficiencies. Accordingly, several primary im-munodeficiencies were recently shown to present purelyas autoimmune,47 autoinflammatory,48 and hemophago-cytosis49 syndromes. Intriguingly, only one Mendeliandisorder, C1 inhibitor deficiency, has thus far been foundto be purely associated with autosomal-dominant angio-edema, a syndrome related to (but possibly different from)allergy.50 Noninfectious immunologic diseases have onlyrecently emerged as a public health problem and do notthreaten mankind as acutely as infections. Records showthat life expectancy in Western Europe in the 18th centurywas about 25 years, whereas life expectancy is currentlyabout 40 years in Sub-Saharan Africa, largely becauseof the burden of infection.17,51 The current longer lifeexpectancy in developed countries primarily reflectsrecent developments in hygiene (preventing infection),vaccines (preventing disease), and antibiotics (preventinga fatal outcome), rather than the intrinsic efficiency of ourimmune system.51 Although the immune system serveswell at the population level, ensuring the reproduction ofspecies, it is the least efficient physiologic system at the


AUGUST 2005

428 Casanova et al

Basic

andclin

icalim

munology

individual level. As indicated by medical and demo-graphic data, most human subjects are immunodeficientand exposed to life-threatening infectious diseases.51

Many might carry a Mendelian primary immunodefi-ciency, being thus perhaps the rule rather than theexception and paradoxically raising hope for scientists,physicians, and patients.

We thank Laurent Abel for critical reading of the manuscript and

other members of the laboratory of Human Genetics of Infectious

Diseases for helpful discussions. We also thank Gerard Orth for

helpful discussions, and we thank 2 anonymous reviewers for their

constructive criticisms.

REFERENCES



2004;114:677-87.

2. Bertrand Y, Muller SM, Casanova JL, Morgan G, Fischer A, Friedrich

W. Reticular dysgenesis: HLA non-identical bone marrow transplants in

a series of 10 patients. Bone Marrow Transplant 2002;29:759-62.

3. Casanova JL, Schurr E, Abel L, Skamene E. Forward genetics of

infectious diseases: immunological impact. Trends Immunol 2002;23:

469-72.

4. McKusick VA. Mendelian inheritance in man. Catalogs of human genes

and genetic disorders. Baltimore (MD): Johns Hopkins University Press;

1998.

5. Newport MJ, Huxley CM, Huston S, Hawrylowicz CM, Oostra BA,

Williamson R, et al. A mutation in the interferon-gamma-receptor gene

and susceptibility to mycobacterial infection. N Engl J Med 1996;335:

1941-9.

6. Jouanguy E, Altare F, Lamhamedi S, Revy P, Emile JF, Newport M,

et al. Interferon-gamma-receptor deficiency in an infant with fatal bacille

Calmette-Guerin infection. N Engl J Med 1996;335:1956-61.

7. Dorman SE, Holland SM. Interferon-gamma and interleukin-12 pathway

defects and human disease. Cytokine Growth Factor Rev 2000;11:

321-33.

8. Casanova JL, Abel L. Genetic dissection of immunity to mycobacteria:

the human model. Annu Rev Immunol 2002;20:581-620.

9. Wurzner R, Orren A, Lachmann PJ. Inherited deficiencies of the

terminal components of human complement. Immunodef Rev 1992;3:

123-47.

10. Wurzner R, Witzel-Schlomp K, Tokunaga K, Fernie BA, Hobart MJ,

Orren A. Reference typing report for complement components C6, C7

and C9 including mutations leading to deficiencies. Exp Clin Immuno-

genet 1998;15:268-85.

11. Kirkpatrick CH. Chronic mucocutaneous candidiasis. Pediatr Infect

Dis J 2001;20:197-206.

12. Lilic D. New perspectives on the immunology of chronic mucocutaneous

candidiasis. Curr Opin Infect Dis 2002;15:143-7.

13. Ramoz N, Rueda LA, Bouadjar B, Montoya LS, Orth G, Favre M.

Mutations in two adjacent novel genes are associated with epidermo-

dysplasia verruciformis. Nat Genet 2002;32:579-81.

14. Orth G. Human papillomaviruses and the skin: more to be learned.

J Invest Dermatol 2004;123:XI-XIII.

15. Morra M, Howie D, Grande MS, Sayos J, Wang N, Wu C, et al. X-linked

lymphoproliferative disease: a progressive immunodeficiency. Annu Rev

Immunol 2001;19:657-82.

16. Nichols KE, Ma CS, Cannons JL, Schwartzberg PL, Tangye SG.

Molecular and cellular pathogenesis of X-linked lymphoproliferative

disease. Immunol Rev 2005;203:180-99.

17. Casanova JL, Abel L. The human model: a genetic dissection of

immunity to infection in natural conditions. Nat Rev Immunol 2004;4:

55-66.



Science 2003;299:2076-9.

19. Ku CL, Yang K, Bustamante J, Puel A, von Bernuth H, Dos Santos O,

et al. Inherited disorders of human Toll-like receptor signalling: immu-

nological implications. Immunol Rev 2005;203:10-20.

20. Fieschi C, Casanova JL. The role of interleukin-12 in human infectious

diseases: only a faint signature. Eur J Immunol 2003;33:1461-4.

21. Lejeune J, Gautier M, Turpin R. [Study of somatic chromosomes from

9 mongoloid children]. C R Hebd Seances Acad Sci 1959;248:1721-2.

22. Casanova JL, Jouanguy E, Lamhamedi S, Blanche S, Fischer A.

Immunological conditions of children with BCG disseminated infection.

Lancet 1995;346:581.

23. Casanova JL, Blanche S, Emile JF, Jouanguy E, Lamhamedi S,

Altare F, et al. Idiopathic disseminated bacillus Calmette-Guerin

infection: a French national retrospective study. Pediatrics 1996;98:

774-8.

24. Altare F, Ensser A, Breiman A, Reichenbach J, Baghdadi JE, Fischer A,

et al. Interleukin-12 receptor beta1 deficiency in a patient with abdominal

tuberculosis. J Infect Dis 2001;184:231-6.

25. Caragol I, Raspall M, Fieschi C, Feinberg J, Larrosa MN, Hernandez M,

et al. Clinical tuberculosis in 2 of 3 siblings with interleukin-12 receptor

beta1 deficiency. Clin Infect Dis 2003;37:302-6.

26. Ozbek N, Fieschi C, Yilmaz BT, De Beaucoudrey L, Bikmaz YE,

Feinberg J, et al. Interleukin-12 receptor beta 1 chain deficiency in a child

with disseminated tuberculosis. Clin Infect Dis 2005;40:e55-8.

27. Fieschi C, Dupuis S, Catherinot E, Feinberg J, Bustamante J, Breiman A,

et al. Low penetrance, broad resistance, and favorable outcome of

interleukin 12 receptor beta1 deficiency: medical and immunological

implications. J Exp Med 2003;197:527-35.

28. Dupuis S, Jouanguy E, Al-Hajjar S, Fieschi C, Al-Mohsen IZ,

Al-Jumaah S, et al. Impaired response to interferon-alpha/beta and

lethal viral disease in human STAT1 deficiency. Nat Genet 2003;

33:388-91.

29. Niehues T, Reichenbach J, Neubert J, Gudowius S, Puel A, Horneff G,

et al. A NEMO-deficient child with immunodeficiency yet without

anhidrotic ectodermal dysplasia. J Allergy Clin Immunol 2004;114:

1456-62.

30. Stiehm ER, Ochs HD, Winkelstein JA. Immunologic disorders in infants

and children. Philadelphia: Elsevier Saunders; 2004.

31. Ochs H, Smith CIE, Puck J. Primary Immunodeficiencies: a molecular

and genetic approach. New York: Oxford University Press; 2005.

32. Bruton OC. Agammaglobulinemia. Pediatrics 1952;9:722-8.

33. Bruton OC. A decade with agammaglobulinemia. J Pediatr 1962;60:

672-6.

34. Good RA. Historical aspects of immunologic deficiency diseases. In:

Kagen BM, Stiehm ER, editors. Immunologic incompetence. Chicago:

Year Book Medical Publishing; 1971. p. 149-77.

35. Hitzig WH. The discovery of agammaglobulinaemia in 1952. Eur J

Pediatr 2003;162:289-304.

36. Stiehm ER, Johnston RB Jr. A history of pediatric immunology. Pediatr

Res. In press 2005.

37. Stiehm ER. New and old immunodeficiencies. Pediatr Res 1993;

33(suppl):S2-8.

38. Fischer A. Primary immunodeficiency diseases: an experimental model

for molecular medicine. Lancet 2001;357:1863-9.

39. Conley ME. Molecular basis of immunodeficiency. Immunol Rev 2005;

203:5-9.

40. Etzioni A, Ochs HD. The hyper IgM syndrome—an evolving story.

Pediatr Res 2004;56:519-25.

41. Reichenbach J, Rosenzweig S, Doffinger R, Dupuis S, Holland SM,

Casanova JL. Mycobacterial diseases in primary immunodeficiencies.


42. Picard C, Puel A, Ku CL, Casanova JL. Primary immunodeficiencies

associated with pneumococcal disease. Curr Opin Allergy Clin Immunol

2003;3:451-9.

43. Lawrence T, Puel A, Reichenbach J, Ku CL, Chapgier A, Renner E, et al.

Autosomal-dominant primary immunodeficiencies. Curr Opin Hematol

2005;12:22-30.

44. Mira MT, Alcais A, Nguyen VT, Moraes MO, Di Flumeri C, Vu HT,

et al. Susceptibility to leprosy is associated with PARK2 and PACRG.

Nature 2004;427:636-40.

45. Mira MT, Alcais A, Van Thuc N, Thai VH, Huong NT, Ba NN, et al.

Chromosome 6q25 is linked to susceptibility to leprosy in a Vietnamese

population. Nat Genet 2003;33:412-5.



Casanova et al 429


46. Alcais A, Mira M, Casanova JL, Schurr E, Abel L. Genetic dissection of

immunity in leprosy. Curr Opin Immunol 2005;17:44-8.

47. Arkwright PD, Abinun M, Cant AJ. Autoimmunity in human primary

immunodeficiency diseases. Blood 2002;99:2694-702.

48. Hull KM, Shoham N, Chae JJ, Aksentijevich I, Kastner DL. The

expanding spectrum of systemic autoinflammatory disorders and their

rheumatic manifestations. Curr Opin Rheumatol 2003;15:61-9.

49. de Saint Basile G, Fischer A. Defective cytotoxic granule-mediated cell

death pathway impairs T lymphocyte homeostasis. Curr Opin Rheumatol

2003;15:436-45.

50. Davis AE 3rd. The pathophysiology of hereditary angioedema. Clin

Immunol 2005;114:3-9.

51. Cairns J. Matters of life and death. Princeton: Princeton University Press;

1997.


AUGUST 2005

430 Casanova et al

Basic

andclin

icalim

munology

Original articles

Infant home endotoxin is associated withreduced allergen-stimulated lymphocyteproliferation and IL-13 production in childhood

Joseph H. Abraham, ScD,a,b,c Patricia W. Finn, MD,d Donald K. Milton, MD,b,c

Louise M. Ryan, PhD,e David L. Perkins, MD,d and Diane R. Gold, MDb,c Boston, Mass

Background: Infant endotoxin exposure has been proposed

as a factor that might protect against allergy and the

early childhood immune responses that increase the risk of

IgE production to allergens.

Objective: Using a prospective study design, we tested the

hypothesis that early-life endotoxin exposure is associated with

allergen- and mitogen-induced cytokine production and

proliferative responses of PBMCs isolated from infants with a

parental history of physician-diagnosed asthma or allergy.

Methods: We assessed household dust endotoxin at age 2 to 3

months and PBMC proliferative and cytokine responses to

cockroach allergen (Bla g 2), dust mite allergen (Der f 1),

cat allergen (Fel d 1), and the nonspecific mitogen PHA at

age 2 to 3 years.

Results: We found that increased endotoxin levels were

associated with decreased IL-13 levels in response to cockroach,

dust mite, and cat allergens, but not mitogen stimulation.

Endotoxin levels were not correlated with allergen- or mitogen-

induced IFN-g, TNF-a, or IL-10. Increased endotoxin levels

were associated with decreased lymphocyte proliferation after

cockroach allergen stimulation. An inverse, although

nonsignificant, association was also found between endotoxin

and proliferation to the other tested stimuli.

Conclusion: Increased early-life exposure to household

endotoxin was associated with reduced allergen-induced

production of the TH2 cytokine IL-13 and reduced

lymphoproliferative responses at age 2 to 3 years in children

at risk for allergy and asthma. Early-life endotoxin-related

reduction of IL-13 production might represent one pathway

through which increased endotoxin decreases the risk of

allergic disease and allergy in later childhood. (J Allergy Clin

Immunol 2005;116:431-7.)

Key words: Endotoxin, lymphocyte proliferation, cytokine, child-hood, allergy

Childhood allergic diseases, such as asthma and hayfever, are increasing in prevalence, cause chronic illhealth, and are a substantial public health concern indeveloped countries.1,2 Measurable childhood sensitiza-tion to inhaled allergens occurs primarily after the age of3 years,3 but the immunologic underpinnings of allergicdisease and airway inflammation likely develop far earlierin life.4,5 Manifestation of an allergic phenotype likelyresults from the complex interplay of genetic, develop-mental, and environmental influences. Adaptive immuneprocesses, including activation of helper T lymphocytesand subsequent B-lymphocyte activation with IgE isotypeswitching, underlie the process of allergic sensitizationin individuals genetically predisposed to development ofallergic responses.6 Although still controversial,7,8 thereis mounting evidence from animal models9-13 and theepidemiologic literature14,15 suggesting that exposure toendotoxin, a potent activator of innate immunity, mightinfluence subsequent adaptive immune responses to aller-gen. Furthermore, these studies suggest that the timing anddose of endotoxin exposure influence the nature of theimmune response, leading to the hypothesis that early lifemight be a crucial time window during which endotoxinmight reduce the risk of allergy through its influence oninnate immunity and downstream T-cell and B-cell reg-ulation of cytokine and IgE expression.

The immunologic pathway linking endotoxin exposureto adaptive immunity and evidence for its effects on thedevelopment of allergic diseases have recently beenreviewed.7,16,17 Briefly, endotoxin is biologically activeLPS, a primary component of the outer cell membrane ofgram-negative bacteria.18,19 Even minute amounts ofendotoxin provoke innate immune responses in vitro andin vivo.20 The nature of that response, which is onlypartially understood, might depend on the developmental

Abbreviations usedEU: Endotoxin units

OR: Odds ratio

SI: Stimulation index

TLR4: Toll-like receptor 4

From the Departments of aEpidemiology, bEnvironmental Health, andeBiostatistics, Harvard School of Public Health, and cThe Channing

Laboratory, Department of Medicine, and dthe Department of Medicine,

Brigham and Women’s Hospital and Harvard Medical School.


Supported by NIEHS R01 ES-07036; NIEHS 2P30ES00002; NIH/NHLBI

HL07427-23; and AI/EHS35786.

Received for publication December 10, 2004; revised March 28, 2005;

accepted for publication May 9, 2005.


Reprint requests: Diane R. Gold, MD, Channing Laboratory, 181 Longwood

Ave, Boston, MA 02115. E-mail: [email protected]

0091-6749/$30.00


doi:10.1016/j.jaci.2005.05.015

431



stage of the organism (eg, infant or adult) and on thesequence andmode of exposure to endotoxin and allergen.Inhaled endotoxin activates the innate immune systemby binding transmembrane toll-like receptors (TLRs)expressed on macrophages and dendritic cells21 throughIL-12 signaling.22 Given the right timing of exposure andgenotype, these endotoxin-activated antigen-presentingcells might stimulate production of IFN-g and other TH1cytokines.23 Through production of IFN-g by TH1-biasedlymphocytes or through alternative pathways, endotoxinmight downregulate TH2 cytokine (including IL-13)secretion, IgE production, and consequent allergic dis-ease. Although studied extensively in animal models, fewprospective data are available on the effects of home LPSexposure on cytokine production in young children.

Using a prospective birth cohort with a parental historyof allergy or asthma, we explored the hypothesis thatearly-life endotoxin exposure alters immune system re-sponsiveness by examining the relationship betweenhouse dust endotoxin and PBMC responses to allergenstimulation. Specifically, we examined associations be-tween house dust endotoxin levels measured 2 to 3monthsafter the child’s birth and allergen- and mitogen-inducedcytokine and proliferative responses of mononuclear cellsisolated from peripheral blood sampled at age 2 to 3 years.Previously, we had found that increased endotoxin levelsin infancy were associated with decreased risk of eczemain the first year of life, suggesting that endotoxin mighthave a protective effect against allergic disease and thebiologic pathways influencing the risk of allergy.24 Wehypothesized that endotoxin levels would be positivelycorrelated with levels of IFN-g, a TH1 cytokine, andinversely associated with levels of the TH2 cytokineIL-13, which can mediate isotype switching to IgE.25

We further hypothesized that endotoxin levels would beassociated with decreased proliferative responses afterallergen stimulation.

METHODS

Description of cohort

The Epidemiology of Home Allergens and Asthma study is a

longitudinal birth cohort study of environmental predictors of allergy

and asthma development. A description of the recruitment of study

participants and study protocol has been previously published.26 In

brief, 505 children from 499 families with a parental history of asthma

or allergy were enrolled in a birth cohort study designed to examine

the effects of allergen exposure in early life on the development

of asthma. The Brigham and Women’s Hospital Human Research

Committee approved the study. Informed consent was obtained from

the parents for blood collection and longitudinal follow-up. Mothers

in the greater Boston metropolitan area delivered at a large Boston

hospital were screened with the following questions: (1) Have you

ever had asthma, hay fever, or allergies? (2) Has the biologic father

of your child ever had asthma, hay fever, or allergies? Mothers

responding yes to either question were asked to complete a screening

questionnaire. Families were not approached if the index child was

premature, had a major congenital anomaly, or was in the neonatal

intensive care unit or if the mother was less than 18 years old or could

not speak English or Spanish. Informed consent was obtained from

the parents for blood collection and longitudinal follow-up. The

Brigham and Women’s Hospital Institutional Review Board ap-

proved the study protocol.

Environmental sampling and endotoxinmeasurements

The home sampling protocol has been described previously.27 A

trained research assistant visited participants’ homes within 2 to 3

months of the child’s birth. During these visits, conducted between

1994 and 1996, detailed demographic, socioeconomic, parental

disease history, and home characteristics questionnaires were com-

pleted, and standardized dust sampling was conducted in various

sites within the home, including the family room. Dust samples to be

used for endotoxin assays were stored desiccated at 220C until

extraction.

Endotoxin activity of dust samples was determined by using the

kinetic Limulus amebocyte lysate assay with resistant-parallel-line

estimation, as previously described.28-30 The Limulus amebocyte

lysate was supplied by BioWhittaker (Walkersville, Md). Reference

standard endotoxin was obtained from the United States

Pharmacopoeia, Inc (Rockville, Md), and control standard endotoxin

was supplied by Associates of Cape Cod (Woods Hole, Mass).

Results were reported in endotoxin units (EU) per milligram of dust

adjusted to account for lot-to-lot variation in Limulus amebocyte

lysate sensitivity to house dust endotoxin and referenced to the

reference standard endotoxin EC5 and EC6 (US Pharmacopoeia, Inc;

1 ng of EC5 and EC6 = 10 EU).30 Because of the prioritization

schema for assaying dust samples, the availability of endotoxin

measurements was conditional on there being sufficient dust to first

assay for home allergens and fungi. As such, for 19 of the 115

subjects with biomarker outcome data, family room dust endotoxin

levels were unavailable.

PBMC responses

At 2 to 3 years of age, blood sampling and analysis was conducted

in a subgroup of the study participants (n = 115). As previously

described, selection of this subgroup was based on the home allergen

levels measured during the initial home visit.31,32 The goal in

choosing these subjects was to maximize variability in early-life

exposure to the allergens with which their cells were to be stimulated.

PBMCs were isolated from this blood sample by using Ficoll-

Hypaque centrifugation.33 Fresh cells were incubated in media;

media containing either 30 mg/mL cockroach allergen (Bla g 2), 30

mg/mL house dust mite allergen (Der f 1), or 1000 U/mL cat allergen

(Fel d 1); or media plus 10 mg/mL PHA. Optimal stimulant

concentrations used for the assay were determined in a prior dose-

response analysis.31

At 24 and 60 hours after the initiation of stimulation, supernatants

were harvested, and cytokine concentrations were quantified by

means of ELISA (Endogen, Woburn, Mass). On the basis of prior

optimization for detection of cytokine levels, IL-10 and TNF-a were

measured in the 24-hour samples, and IFN-g and IL-13 were

measured in the 60-hour samples. The lower limits of detection for

cytokine assays were as follows: IFN-g, less than 2 pg/mL; IL-13,

less than 7 pg/mL; IL-10, less than 3 pg/mL; and TNF-a, less than 5

pg/mL. Because supernatant quantities were limited, cytokine assays

were prioritized, with IFN-g levels being measured first. As a result,

fewer subjects have observations of IL-13, IL-10, and TNF-a

levels.32

After incubation of PBMCs with allergen or mitogen for 72 hours,

1 mCi of tritiated thymidine was added to each well. After incubation

for an additional 8 hours, the cells were harvested, and tritiated

thymidine uptake was determined by means of b-counting.


AUGUST 2005

432 Abraham et al

Basic

andclin

icalim

munology

Proliferation was quantified by calculating the stimulation index (SI)

for each stimulant with the following formula:

SI¼ðMeanvalueof tritiatedthymidineuptakeforstimulatedsamplesÞ=ðMeanvalueof tritiatedthymidineuptakeforunstimulatedsamplesÞ:31

IgE measurements

Serum samples were assayed for total IgE and specific IgE

antibodies to dust mite, cat, cockroach, and ovalbumin by using the

UNICAP System (Pharmacia, Uppsala, Sweden). IgE increase was

defined either as an IgE specific response (0.35 IU/mL) to at least

one allergen or a total IgE level of greater than 60 IU/mL for age 2 to 3

years.

Statistical methods

All statistical analyses were conductedwith SAS version 8.2 (SAS

Institute Inc, Cary, NC). In primary analyses investigating lympho-

proliferative response as an outcome, both endotoxin and SI were

treated as continuous variables. The distribution of SIs and endotoxin

levels were log10 transformed for linear regression analyses to

improve the normality of residuals and facilitate interpretation of

model results. The relationships between log10-transformed SIs,

log10-transformed endotoxin, and potential covariates were assessed

by using ordinary least-squares univariate andmultivariate regression

models. The estimates for the relationship of the log of endotoxin

to SI were used to calculate the percentage difference in SI for a

doubling of endotoxin. This enabled us to translate the estimates into

an easier-to-understand outcome that was scaled by a realistic (within

the range of our data) increase in endotoxin level. We also considered

SI as a dichotomous categorical by using cutoff points that have been

considered as positive PBMC responses to allergen or mitogen

stimuli in other studies (allergen responses, SI > 3; PHA responses,

SI > 153).31

Our focus on the relation of endotoxin to IL-13 production

presented statistical challenges because of the skewed distribution of

the cytokine levels and because of the significant proportion of values

below the limit of detection. First, we used nonparametric correlation

analyses to relate continuous dust endotoxin with continuous cyto-

kine levels. For nonparametric testing, cytokine observations below

the lower limits of detection were assigned very low but nonmissing

values (0.001 pg/mL) to be included in the analysis and tested the

sensitivity of our findings to the choice of number assigned to

the observations below the limit of detection. We then assessed the

relation between endotoxin and the odds of havingmeasurable IL-13.

Finally, in secondary analyses we tested for trends in the proportion

of cytokine observations above the lower limit of detection when

grouped by tertile of endotoxin level.

Cockroach (Bla g 1 or 2), dust mite (Der f 1), and cat (Fel d 1)

allergen levels were classified as follows: Bla g 1 or 2 of 0.05 U/g or

greater, Fel d 1 of 1mg/g or greater, andDer f 1 of 2mg/g or greater).31

Reported household income was classified as being less than or

greater than $50,000.

RESULTS

Of the 505 children initially enrolled in the study, 7were followed for less than 5 months in the first year oflife. Of the 498 with follow-ups, 115 had lymphocyteproliferation measurements at age 2 to 3 years, of whom96 also had family room dust endotoxin measurements.Compared with the 402 subjects without both lymphocyteproliferation and endotoxin data, the subsample withboth (n = 96) had a greater proportion of boys (Table I).No other selection bias was detected. Endotoxin in the96 family room dust samples had a geometric mean of 95EU/mg (geometric SD, 1.8 EU/mg) and a median level of

TABLE I. Characteristics of children in the cohort

Variable

Subjects with measurement

of endotoxin and PBMC

outcomes* (n = 96)

Subjects with no measurement

of both endotoxin and PBMC

outcomes (n = 402)

Total

(n = 498)

Sex, n

Male 60 (63%) 208 (52%) 268 (54%)

Female 36 (38%) 194 (48%) 230 (46%)

Race-ethnicity, n

White or Asian 79 (82%) 324 (81%) 403 (81%)

Other 17 (18%) 78 (19%) 95 (19%)

Income, n

<$50,000 27 (28%) 106 (26%) 133 (27%)

$50,000 68 (71%) 283 (70%) 351 (70%)

Family room Bla g (1 or 2), n

0.05 U/g 25 (26%) 84 (21%) 109 (22%)

<0.05 U/g 69 (72%) 291 (72%) 360 (72%)

Family room Der f 1, n

2 mg/g 43 (45%) 190 (47%) 233 (47%)

<2 mg/g 53 (55%) 206 (51%) 259 (52%)

Family room Fel d 1, n

1 mg/g 61 (64%) 234 (58%) 295 (59%)

<1 mg/g 34 (35%) 147 (37%) 181 (36%)

Cold before blood draw, n

Yes 24 (25%) – –

No 71 (75%) – –

*The total number varied according to specific outcomes, but 96 children had endotoxin and lymphocyte proliferation data.



Abraham et al 433


101 EU/mg (Table II). The distributions of lymphocyteproliferation and cytokine levels for this subsample weresimilar to those previously demonstrated for the largersample (Table II).31,32

We found no confounders of the association of endo-toxin with either the lymphoproliferative response or thecytokine production by stimulated PBMCs after consid-ering living room allergen level, report of dog or cat, race-ethnicity, household income, or cold before blood draw(data not shown). As previously reported for the largercohort, cockroach allergen (Bla g 1 or 2), cat allergen(Fel d 1), and dust mite allergen (Der f 1) levels measuredin family room dust samples were not significantlyassociated with home endotoxin levels.34

Association of house dust endotoxin and SIs

Weobserved a consistent association between increaseddust endotoxin measured in the first months of a baby’slife and decreased SIs at age 2 to 3 years, although thisassociation was only significant for the response to cock-roach allergen (Table III). Despite the absence of measur-able confounders, to adjust for potential independentinfluences on the lymphoproliferative responses, we ad-justed for indicators of age at the time of blood draw, reportof a cold in the week before blood draw, household in-come level, and house dust allergen levels in the allergen-specific proliferation models in addition to reporting theunivariate models.31 The multivariate-adjusted relation-ships between endotoxin and the SIs ranged from24% to217%, being strongest and statistically significant for theSI after cockroach allergen stimulation. When dichoto-mized at previously defined cutoff points,31 positivecockroach-induced lymphocyte proliferative responses(SI > 3) were also inversely associated with endotoxinlevels (odds ratio [OR] of Bla g 2 SI 3 for a doubling infamily room endotoxin level: OR, 0.53; 95% CI, 0.27-1.049; P = .07). Although the estimates for Der f 1 SIwere negative, the analyses using SI as a dichotomousoutcome were not supportive of a significant associationof endotoxin with a lower odds of Der f 1 SI of greaterthan 3 (P = .9) or Fel d 1 SI of greater than 3 (P = .7).

Association of house dust endotoxin andstimulation-induced cytokine production

By using rank-based correlations, increasing house dustendotoxin levels measured early in life were inverselycorrelated with allergen-induced IL-13 but not IL-13induced by PHA (Table IV). For a doubling of endotoxinlevels, children had significantly reduced odds of havingmeasurable-detectable allergen-induced IL-13 levels inresponse to all 3 allergens, with no association betweenendotoxin- and PHA-induced IL-13 (Table V). Withincreasing tertiles of endotoxin, a smaller percentage ofchildren had measurable allergen-induced IL-13 levels(above the limits of detection), although this trend wasonly statistically significant for the response to Der f 1 (seeFig E1 in the Online Repository in the online version ofthis article at www.mosby.com/jaci). We did not observecorrelations between allergen- or mitogen-induced levelsof IFN-g, TNF-a, and IL-10 in the supernatants of theisolated PBMCs and endotoxin (data not shown). Wefound no evidence of an association between PHA-induced IL-13 and endotoxin.

DISCUSSION

In this prospective cohort study of young children witha family history of allergic disease, we observed thatincreased levels of family room dust endotoxin in infancywere associated with decreased allergen-induced IL-13production by mononuclear cells isolated from peri-pheral blood at age 2 to 3 years. We also found thatendotoxin levels were associated with a downregulationof allergen- andmitogen-stimulated lymphocyte prolifera-tion. Through the interactions of its Lipid A moiety withreceptors such as TLR4, endotoxin is hypothesized tomodulate the innate immunity pathway and, through thosepathways, to influence adaptive immunity16,17 with re-duced production of cytokines, such as IL-13, as seen inour study. This in turn might result in subsequent reduc-tion of IgE levels, allergy, and allergic disease. In ourcohort at this age, we have previously demonstrated an

TABLE II. Distribution of endotoxin, IL-13, and PBMC lymphocyte proliferative response levels*

Variable Minimum Median 75th Percentile Maximum Detectable, n

Endotoxin (EU/mg), n = 96 19.3 101.2 138.4 518.8 96 (100%)

IL-13 (pg/mg), n = 67

Bla g 2 <LD <LD 9.6 207.2 32 (48%)

Der f 1 <LD 8.6 30.5 397.3 47 (70%)

Fel d 1 <LD <LD 4.7 137.6 26 (39%)

PHA <LD 1425 2255 3812 64 (96%)

SI

Bla g 2 0.9 4.1 5.7 53.6 94 (100%)

Der f 1 0.4 3.3 5.9 32.5 95 (100%)

Fel d 1 0.3 2.2 4.4 27.5 96 (100%)

PHA 14.4 138.6 285.5 1205.1 96 (100%)

<LD, Below the limits of detection.

*For the purpose of ranking/nonparametric testing, values below the limits of detection were assigned the value 0.001 pg/mL.


AUGUST 2005

434 Abraham et al

Basic

andclin

icalim

munology

association between lower IL-13 levels and reduced risk ofallergy/increased IgE levels.32 Increased endotoxin levelswere also associated with reduced risk of allergy, but thisassociation was not statistically significant. A doubling ofendotoxin in family room dust was associated with an ORof 0.71 (95% CI, 0.39-1.31) for increased IgE levels, asdetermined by positive RAST results or high total IgElevels among 97 children with data on endotoxin in earlylife and IgE levels at age 2 to 3 years. In a previous re-port from this study, we demonstrated that higher infantendotoxin levels were associated with a significantlylower risk of eczema in the first year of life.24 The datawe present in this article support the hypothesis that theassociation of increased home endotoxin exposure withprotection against childhood allergic disease might bemediated in part through reduced TH2 cytokine expressionin early life.

Researchers in Europe have observed an associationbetween a farming lifestyle and lower prevalence of al-lergic disease in children and hypothesized that increasedexposure to endotoxin in infancy might be responsiblefor this association through an early-life influence oninnate immune development and its relation to adaptiveimmunity, cytokine production, and IgE production.14,35,36

In a cross-sectional study of German, Austrian, and Swisschildren ages 6 to 13 years, endotoxinmeasured in the dustof children’s mattresses was inversely associated withasthma, atopic sensitization, and hay fever.37 Increasedbed dust endotoxin levels were associated with decreasedLPS-induced production of IFN-g, IL-10, IL-12, andTNF-a. The researchers interpreted these findings as aglobal downregulation of the immune response. In con-trast, we found a specific endotoxin-related downregula-tion of a TH2 cytokine, IL-13, which can mediate isotypeswitching to IgE25,38 without an influence on IFN-g,IL-10, or TNF-a production. Our study contrasts with themulticenter European study in many ways. We stimulatedour PBMCs with allergen, our study was prospective, ourcohort was predisposed to allergy by virtue of familyhistory, and our endotoxin levels, measured in infancy,were relatively low. In a small US cross-sectional study ofinfants with repeated wheeze, allergen-sensitized childrencame from homeswith higher endotoxin levels, and higherhome endotoxin levels were correlated with increased

proportions of IFN-g–producing CD4 T cells.15 In thisstudy the investigators did not evaluate cytokine secretionof lymphocyte proliferative response of antigen- or mito-gen-stimulated PBMCs.

Although this is, to the best of our knowledge, the firstreport of a prospective association of endotoxin withdiminished IL-13 production in children, some animalmodels have demonstrated similar associations. However,in animal models endotoxin exposure with TLR4 agonistshas been shown to both diminish and enhance IFN-gproduction, IL-13 production, and allergic responses.12

The directionality and nature of the immune and allergicresponses appears to be dependent on many factors,including the type of model, dose of allergen or TLR4agonist, timing of allergen or TLR4 agonist, use or not ofadjuvant, and type of species or murine strain.9,11,39-42 Inone murine model prenatal exposure to LPS resulted inincreased neonatal IFN-g secretion and decreased neona-tal IL-13 production after OVA exposure, with no protec-tion against airway responsiveness.43 In another murinemodel, Velasco et al12 showed that pulmonary adminis-tration of Lipid A before allergen sensitization decreasedeosinophilia, bronchoalveolar lavage fluid IL-13 levels,serum IgE levels, airway hyperresponsiveness, and thenumber of CD41 cells in the lung. Lipid A administrationduring allergen challenge diminished eosinophilia. Incontrast, Delayre-Orthez et al42 showed that LPS duringchallenge enhanced allergen-induced eosinophilia but didnot enhance allergen-induced IgE or airway hyperrespon-siveness. A recent study compared the effects of Lipid A interms of maturity of the immune system, species, dose,and timing by examination of human cord and adultPBMCs and murine cells in vitro and in vivo.13 Lipid Ainduced primarily a time- and dose-independent produc-tion of IFN-g.

In both human and animal models, the effects ofenvironmental endotoxin exposure, are likely to be de-pendent on the dose and timing of exposure and on thedevelopmental stage and genotype.44 These factors mightexplain seemingly contradictory epidemiologic findings ofendotoxin effects on allergic disease.7,16 As an irritant, inthose with or without asthma, endotoxin might increasethe severity of symptoms.31,45 Previously, we found thatendotoxin was associated with increased risk of wheeze

TABLE III. Percentage change in lymphocyte proliferative response (SI) for a doubling of family room dust

endotoxin levels

Univariate* Multivariatey

Stimulus % Difference 95% CI % Difference 95% CI

Cockroach allergen (Bla g 2) 215 227 to 21 217 228 to 24

Dust mite allergen (Der f 1) 213 230 to 8 215 231 to 5

Cat allergen (Fel d 1) 25 221 to 15 24 221 to 16

PHA 211 228 to 10 29 227 to 12

*The percentage difference in SI for a doubling of endotoxin was calculated as follows: ð10 ½log10ð2Þ ðbÞ21Þ 100, where b is the endotoxin effect estimate

for the univariate model: log10ðSIÞ ¼ b01b1½log10ðEndotoxinÞ. Also see the Methods section.

Same as the univariate model: adjusted for age at blood draw, report of a cold before blood draw, household income, and family room dust allergen level for

allergen-induced proliferation models.



Abraham et al 435


and repeated wheeze in the first year of life in this cohort;this might be secondary to its irritant effects.29 In thesiblings of the subjects in this cohort, the association ofendotoxin with wheeze decreased over time, and as thechild grew older, earlier-life endotoxin appears to have anull or even protective effect on wheeze risk, perhapsbecause the wheeze was more related to allergic inflam-mation.46 This is consistent with the hypothesis thatendotoxin might increase the risk of irritant wheeze atthe same time as decreasing the risk of allergy and itspulmonary and extrapulmonary manifestations. In theirstudies of children growing up on farms, Braun-Fahrlanderet al37 have found consistent protective effects of endo-toxin on allergy but an increased risk of wheeze withincreasing endotoxin levels among those without allergy.

We had limited power to fully elucidate pathwaysleading from endotoxin exposure to cytokine production,early allergy, and subsequent allergic disease expression.In this group of young children, when allergy, particularlyallergy to inhaled allergens, is just developing, we foundthat increased levels of endotoxin were associated with alower risk of high IgE levels but that association was notstatistically significant. This might be due to small num-bers (low power) because we were only able to measureboth endotoxin and cytokines on a subset of the cohortor due to the age of the children because the allergicphenotype is not fully evolved by age 2 to 3 years. It isquite possible that a skewing toward TH2 cytokineproduction is an earlier step in the pathway toward fullexpression of the allergic phenotype. The associations be-tween observed family room dust endotoxin and allergen-and mitogen-induced proliferation of lymphocytes weresmall and achieved statistical significance only afterstimulation with cockroach allergen. Therefore we cannot

exclude the role of chance in determining the observedassociation between endotoxin and proliferation afterstimulation with the other allergens and the mitogen.However, the consistently negative associations suggestthat this is not a chance observation. Although confound-ing bias by unmeasured factors is always a possibility, wefound no confounding by other measured exposures(including dog) in our analyses evaluating the associationof endotoxin with detectable IL-13, and we adjusted forconfounders in our analyses, with SI as our outcome.Ideally, an estimate of each subject’s endotoxin exposurewould capture the true temporally and spatially integratedexposure that occurs within the first year of life. Our singlemeasure of endotoxin in the home is likely to represent thistrue exposure with error. Because of the prospective studydesign, differences between the observed measurementsof endotoxin and the subjects’ true endotoxin exposureare most likely nondifferential with respect to PBMC re-sponses. This measurement error would most likely haveattenuated the exposure effect estimates. A final limitationis that we did not evaluate the allergens that we used tostimulate the PBMCs for the possible presence of endo-toxin. Production of IL-13 is likely to be less influenced bytrace LPS than production of cytokines involved in innateimmunity, such as TNF-a. The possibility remains, how-ever, that increased home allergen levels in infancy areassociated with reduced IL-13 production not only inresponse to allergens but also in response to a combinationof allergen and LPS.

In conclusion, in a cohort of children at risk of allergyand asthma, we conducted a prospective study on the re-lation of home endotoxin levels in infancy to subsequentallergen-stimulated lymphocyte proliferative responsesand cytokine production. Our data suggest that increasedearly-life endotoxin exposure might be associated with areduction in subsequent allergen-stimulated lymphocyteproliferation and IL-13 secretion. Endotoxin-inducedreduction in IL-13 secretion might be one early-life stepin the pathway to endotoxin-associated reduction inallergy and allergic disease.

REFERENCES

1. Vollmer WM, Osborne ML, Buist AS. 20-year trends in the prevalence

of asthma and chronic airflow obstruction in an HMO. Am J Respir Crit

Care Med 1998;157:1079-84.

TABLE IV. Correlations between allergen- and mitogen-

induced cytokine levels and family room dust endotoxin

levels*

Stimulus

Cytokine

(pg/mL) Obs.

Correlation

coefficient P value

Bla g 2 IFN-g 80 20.05 .66

IL-13 67 20.31 .01

IL-10 64 0.11 .38

TNF-a 50 20.01 .94

Der f 1 IFN-g 85 0.01 .91

IL-13 67 20.27 .02

IL-10 69 20.08 .51

TNF-a 50 20.09 .54

Fel d 1 IFN-g 80 0.06 .57

IL-13 67 20.21 .08

IL-10 64 20.02 .89

TNF-a 50 20.03 .83

PHA IFN-g 85 0.08 .46

IL-13 67 0.12 .34

IL-10 69 20.05 .67

TNF-a 50 20.14 .35

Obs., Number of observations.

*Rank-based (Spearman) correlation coefficients.

TABLE V. Relative odds of detectable versus nondetect-

able allergen- and mitogen-induced IL-13 for a doubling

of family room endotoxin levels

Cytokine Stimulus Obs. OR 95% CI

IL-13

Bla g 2 67 0.36 0.16-0.81

Der f 1 67 0.27 0.11-0.70

Fel d 1 67 0.44 0.20-0.94

PHA 67 0.89 0.19-4.17

Obs., Number of observations.


AUGUST 2005

436 Abraham et al

Basic

andclin

icalim

munology

2. Holgate ST. The epidemic of allergy and asthma. Nature 1999;402:B2-4.

3. Kulig M, Bergmann R, Klettke U, Wahn V, Tacke U, Wahn U. Natural

course of sensitization to food and inhalant allergens during the first 6

years of life. J Allergy Clin Immunol 1999;103:1173-9.

4. Croner S, Kjellman NI. Natural history of bronchial asthma in childhood.

A prospective study from birth up to 12-14 years of age. Allergy 1992;

47:150-7.

5. Wahn U, von Mutius E. Childhood risk factors for atopy and the

importance of early intervention. J Allergy Clin Immunol 2001;107:

567-74.

6. Donovan CE, Finn PW. Immune mechanisms of childhood asthma.

Thorax 1999;54:938-46.

7. Liu AH. Endotoxin exposure in allergy and asthma: reconciling a

paradox. J Allergy Clin Immunol 2002;109:379-92.

8. Douwes J, Pearce N, Heederik D. Does environmental endotoxin

exposure prevent asthma? Thorax 2002;57:86-90.

9. Tulic MK, Wale JL, Holt PG, Sly PD. Modification of the inflammatory

response to allergen challenge after exposure to bacterial lipopolysac-

charide. Am J Respir Cell Mol Biol 2000;22:604-12.

10. Wan GH, Li CS, Lin RH. Airborne endotoxin exposure and the

development of airway antigen-specific allergic responses. Clin Exp

Allergy 2000;30:426-32.

11. Eisenbarth SC, Piggott DA, Huleatt JW, Visintin I, Herrick CA,

Bottomly K. Lipopolysaccharide-enhanced, toll-like receptor 4-depen-

dent T helper cell type 2 responses to inhaled antigen. J Exp Med 2002;

196:1645-51.

12. Velasco G, Campo M, Manrique OJ, Bellou A, He H, Arestides RS, et al.

Toll-like receptor 4 or 2 agonists decrease allergic inflammation. Am J

Respir Cell Mol Biol 2005;32:218-24.

13. Schaub B, Bellou A, Gibbons FK, Velasco G, Campo M, He H, et al.

TLR2 and TLR4 stimulation differentially induce cytokine secretion in

human neonatal, adult, and murine mononuclear cells. J Interferon

Cytokine Res 2004;24:543-52.

14. Von Ehrenstein OS, Von Mutius E, Illi S, Baumann L, Bohm O, von

Kries R. Reduced risk of hay fever and asthma among children of

farmers. Clin Exp Allergy 2000;30:187-93.

15. Gereda JE, Leung DY, Thatayatikom A, Streib JE, Price MR, Klinnert

MD, et al. Relation between house-dust endotoxin exposure, type 1

T-cell development, and allergen sensitisation in infants at high risk of

asthma. Lancet 2000;355:1680-3.

16. Reed CE, Milton DK. Endotoxin-stimulated innate immunity: a contrib-

uting factor for asthma. J Allergy Clin Immunol 2001;108:157-66.

17. Bellou A, Schaub B, Ting L, Finn PW. Toll receptors modulate allergic

responses: interaction with dendritic cells, T cells and mast cells. Curr

Opin Allergy Clin Immunol 2003;3:487-94.

18. Rietschel ET, Brade H. Bacterial endotoxins. Sci Am 1992;267:54-61.

19. Reeves PR, Hobbs M, Valvano MA, Skurnik M, Whitfield C, Coplin D,

et al. Bacterial polysaccharide synthesis and gene nomenclature. Trends

Microbiol 1996;4:495-503.

20. Imrich A, Ning YY, Koziel H, Coull B, Kobzik L. Lipopolysaccharide

priming amplifies lung macrophage tumor necrosis factor production in

response to air particles. Toxicol Appl Pharmacol 1999;159:117-24.

21. Beutler B, Poltorak A. Positional cloning of Lps, and the general role of

toll-like receptors in the innate immune response. Eur Cytokine Netw

2000;11:143-52.

22. D’Andrea A, Rengaraju M, Valiante NM, Chehimi J, Kubin M, Aste

M, et al. Production of natural killer cell stimulatory factor (interleukin

12) by peripheral blood mononuclear cells. J Exp Med 1992;176:

1387-98.

23. Le J, Lin JX, Henriksen-DeStefano D, Vilcek J. Bacterial lipopolysac-

charide-induced interferon-gamma production: roles of interleukin 1 and

interleukin 2. J Immunol 1986;136:4525-30.

24. Phipatanakul W, Celedon JC, Raby BA, Litonjua AA, Milton DK, Sredl

D, et al. Endotoxin exposure and eczema in the first year of life.

Pediatrics 2004;114:13-8.

25. de Vries JE. The role of IL-13 and its receptor in allergy and

inflammatory responses. J Allergy Clin Immunol 1998;102:165-9.

26. Litonjua AA, Carey VJ, Burge HA, Weiss ST, Gold DR. Parental history

and the risk for childhood asthma. Does mother confer more risk than

father? Am J Respir Crit Care Med 1998;158:176-81.

27. Chew GL, Burge HA, Dockery DW, Muilenberg ML, Weiss ST, Gold

DR. Limitations of a home characteristics questionnaire as a predictor of

indoor allergen levels. Am J Respir Crit Care Med 1998;157:1536-41.

28. Milton DK, Johnson DK, Park JH. Environmental endotoxin measure-

ment: interference and sources of variation in the Limulus assay of house

dust. Am Ind Hyg Assoc J 1997;58:861-7.

29. Park JH, Spiegelman DL, Gold DR, Burge HA, Milton DK. Predictors of

airborne endotoxin in the home. Environ Health Perspect 2001;109:859-64.

30. Milton DK, Feldman HA, Neuberg DS, Bruckner RJ, Greaves IA.

Environmental endotoxin measurement: the Kinetic Limulus Assay with

Resistant-parallel-line Estimation. Environ Res 1992;57:212-30.

31. Finn PW, Boudreau JO, He H, Wang Y, Chapman MD, Vincent C, et al.

Children at risk for asthma: home allergen levels, lymphocyte prolifer-

ation, and wheeze. J Allergy Clin Immunol 2000;105:933-42.

32. Contreras JP, Ly NP, Gold DR, He H, Wand M, Weiss ST, et al.

Allergen-induced cytokine production, atopic disease, IgE, and wheeze

in children. J Allergy Clin Immunol 2003;112:1072-7.

33. Lara-Marquez ML, Deykin A, Krinzman S, Listman J, Israel E, He H,

et al. Analysis of T-cell activation after bronchial allergen challenge in

patients with atopic asthma. J Allergy Clin Immunol 1998;101:699-708.

34. Park JH, Spiegelman DL, Burge HA, Gold DR, Chew GL, Milton DK.

Longitudinal study of dust and airborne endotoxin in the home. Environ

Health Perspect 2000;108:1023-8.

35. Braun-Fahrlander C, Gassner M, Grize L, Neu U, Sennhauser FH,

Varonier HS, et al. Prevalence of hay fever and allergic sensitization in

farmer’s children and their peers living in the same rural community.

SCARPOL team. Swiss Study on Childhood Allergy and Respiratory

Symptomswith Respect to Air Pollution. Clin ExpAllergy 1999;29:28-34.

36. Riedler J, Braun-Fahrlander C, Eder W, Schreuer M, Waser M, Maisch

S, et al. Exposure to farming in early life and development of asthma and

allergy: a cross-sectional survey. Lancet 2001;358:1129-33.

37. Braun-Fahrlander C, Riedler J, Herz U, Eder W, Waser M, Grize L, et al.

Environmental exposure to endotoxin and its relation to asthma in

school-age children. N Engl J Med 2002;347:869-77.

38. Zimmermann N, Hershey GK, Foster PS, Rothenberg ME. Chemokines

in asthma: cooperative interaction between chemokines and IL-13.


39. Gerhold K, Blumchen K, Bock A, Seib C, Stock P, Kallinich T, et al.

Endotoxins prevent murine IgE production, T(H)2 immune responses,

and development of airway eosinophilia but not airway hyperreactivity.


40. Dabbagh K, Lewis DB. Toll-like receptors and T-helper-1/T-helper-2

responses. Curr Opin Infect Dis 2003;16:199-204.

41. Rodriguez D, Keller AC, Faquim-Mauro EL, de Macedo MS, Cunha FQ,

Lefort J, et al. Bacterial lipopolysaccharide signaling through Toll-like

receptor 4 suppresses asthma-like responses via nitric oxide synthase 2

activity. J Immunol 2003;171:1001-8.

42. Delayre-Orthez C, de Blay F, Frossard N, Pons F. Dose-dependent

effects of endotoxins on allergen sensitization and challenge in the

mouse. Clin Exp Allergy 2004;34:1789-95.

43. Blumer N, Herz U, Wegmann M, Renz H. Prenatal lipopolysaccharide-

exposure prevents allergic sensitization and airway inflammation, but not

airway responsiveness in a murine model of experimental asthma. Clin

Exp Allergy 2005;35:397-402.

44. Cook DN, Wang S, Wang Y, Howles GP, Whitehead GS, Berman KG,

et al. Genetic regulation of endotoxin-induced airway disease. Genomics

2004;83:961-9.

45. Michel O, Ginanni R, Duchateau J, Vertongen F, Le Bon B, Sergysels R.

Domestic endotoxin exposure and clinical severity of asthma. Clin Exp

Allergy 1991;21:441-8.

46. Litonjua AA, Milton D, Celedon JC, Ryan L, Weiss S, Gold D. A

longitudinal analysis of wheezing in young children: the independent

effects of early life exposure to house dust endotoxin, allergens, and pets.




Abraham et al 437


Does early EBV infection protect againstIgE sensitization?

Caroline Nilsson, MD,a Annika Linde, MD, PhD,b Scott M. Montgomery, BSc, PhD,c

Liselotte Gustafsson,b Per Nasman, Ph Lic,d Marita Troye Blomberg, PhD,e and

Gunnar Lilja, MD, PhDa Stockholm, Sweden

Background: There is indirect evidence that an increased

infectious burden is associated with a decreased prevalence

of IgE-mediated allergy during childhood.

Objective: To determine whether there is a relation between the

serostatus of 13 different viruses and parentally reported

infections and IgE sensitization in 2-year-old children. To

investigate whether there is an interaction between

cytomegalovirus (CMV) and Epstein-Barr virus (EBV) in

relation to IgE sensitization.

Methods: A total of 246 infants were followed prospectively to

2 years of age with clinical examinations, skin prick test, and

specific IgE analyses and through analysis of seropositivity

against adenovirus, influenza, parainfluenza, respiratory

syncytial virus, CMV, EBV, herpes simplex virus, human

herpesvirus 6, and varicella-zoster virus.

Results: There was some evidence that IgE sensitization (24%)

tended to be more common among children who were

seropositive against few compared with children who were

seropositive against many viruses, but this was not statistically

significant, and there was no consistent trend across the groups.

IgE sensitization was statistically significantly less prevalent at

2 years of age among infants who were seropositive against

EBV but not other viruses (adjusted odds ratio, 0.34; 95% CI,

0.14-0.86). The interaction of seropositivity against both

CMV and EBV antibodies indicated a further reduction in the

risk for IgE sensitization (adjusted odds ratio for interaction,

0.10; 95% CI, 0.01-0.92), indicating effect modification

associated with seropositivity against CMV.

Conclusion: Our results indicate that acquisition of EBV

infection during the first 2 years of life is associated with a

reduced risk of IgE sensitization, and this effect is enhanced

by CMV coinfection. (J Allergy Clin Immunol 2005;116:

438-44.)

Key words: Childhood, CMV, EBV, IgE, infections, sensitization,

serology, viral infections

The increasing prevalence of allergic disease hasbecome a major health problem in the industrialized partsof the world.1 Epidemiological studies have shown thatvarious markers of increased burden of infections areassociated with a decreased prevalence of allergy andasthma during childhood.2,3 Viral infections have beenimplicated in influencing IgE-mediated sensitization, buttheir exact role remains controversial.3,4 It has beensugested that many viruses influence the differentiationof T cells, thus causing an imbalance between TH1 andTH2 immune responses.5 Cytomegalovirus (CMV) andEpstein-Barr virus (EBV) are persistent viral infections, asdemonstrated by frequent presence of the virus in salivaand urine of healthy individuals,6 and may influence theimmune system with respect to the development ofallergy, as recently proposed by Sidorchuk et al.7

The aim of this study was therefore to elucidate theinterplay between CMV and EBV in relation to IgEsensitization among children at 24 months of age. Thestudy was also designed to investigate the association ofIgE sensitization with infectious burden measured asseroprevalence against 13 different viruses, includingCMV, EBV, and respiratory syncytial virus (RSV), aswell as parentally reported infections.

METHODS

Subjects

Families who where expecting a child were asked by the midwife

at the maternity ward whether they were interested in participating in

the study. Only parents who reported they had a history of allergy in

the mother, in both parents, or in neither parent were eligible. The

parents provided a blood sample and underwent skin prick tests

(SPTs). Only parents whose SPT results confirmed their positive or

negative history of respiratory allergy to pollen and/or furred pets

were invited to continue. When evaluated at 24 months of age, 246

children (126 boys and 120 girls) born to the selected parents

participated in the study. One hundred two children had 2 allergic

parents, 75 children had an allergic mother, and 69 children had no

parental history of allergy. All infants were born full-term (>35weeks

of gestation) at hospitals in Stockholm and had birth weights within

the normal range (data not shown). The socioeconomic status of the

From athe Department of Pediatrics, Sachs’ Children’s Hospital, and bthe

Swedish Institute for Infectious Disease Control, Microbiology and Tumor

Biology Center, Karolinska Institute; cthe Clinical Epidemiology Unit,

Department of Medicine, Karolinska Hospital, Karolinska Institute and

Clinical Research Centre, Orebro University Hospital; dthe Royal Institute

of Technology; and ethe Department of Immunology, Stockholm

University.

Disclosure of potential conflict of interest: None to disclose.

Supported by the Swedish Asthma and Allergy Association, Consul Th C

Berg’s Foundation, the Samariten Foundation, Mjolkdroppen, the Vardal

Foundation, the Heart and Lung Foundation, GlaxoSmithKline, Brio AB,

and the Karolinska Institute. Pharmacia Diagnostics AB supplied reagents

for plasma IgE analyses.

Received for publication October 25, 2004; revised April 20, 2005; accepted



Reprint requests: Caroline Nilsson, MD, Department of Pediatrics, Sachs’

Children’s Hospital, Stockholm South Hospital, S-118 83 Stockholm,

Sweden. E-mail: [email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.027

438

Basic

andclin

icalim

munology


Abbreviations used

EBV: Epstein-Barr virus

CMV: Cytomegalovirus

HHV6: Human herpesvirus 6

HSV: Herpes simplex virus

kUA: Kilo Units allergen-specific antibodies

RSV: Respiratory syncytial virus


VZV: Varicella-zoster virus

families was estimated through the father’s occupation grouped ac-

cording to the classification used by Statistics Sweden. Demographic

data are presented in Table I.

The study was approved by the Human Ethics Committee at

Huddinge University Hospital, Stockholm (Dnr 75/97, 113/97), and

the parents provided informed consent.

Clinical evaluation

The children were followed from birth to 2 years of age and were

clinically evaluated at ages 6, 12, 18, and 24 months by 1 pediatrician

(C. N.).

Skin prick testing

Skin prick tests against food and inhalant allergens were

performed among the children at 24 months of age according to the

manufacturer’s recommendation (ALK, Copenhagen, Denmark).

The SPT included food allergens: egg white (Soluprick weight to

volume ratio, 1/100), cod (Soluprick 1/20), peanut, (Soluprick 1/20),

cow’s milk (3% fat, standard milk), and soybean protein (Soja Semp;

Semper AB, Stockholm, Sweden). SPTs were also performed for

inhalant allergens: cat, dog, Dermatophagoides farinae, birch, and

timothy (Soluprick 10 Histamine Equivalent Prick test). All parents

were skin prick tested against the same inhalant allergens as the

children but also against horse, rabbit, and mugwort. Histamine

chloride (10 mg/mL) was the positive control and the allergen diluent

the negative control. The SPT was considered positive if the wheal

diameter was 3 mm after 15 minutes.

Parents’ report of infections

During the observation period, the parents were asked to record

every infection that their child had in a structured diary. This included

symptoms—runny nose, cough, vomiting, diarrhea, fever—and a

doctor’s diagnosis where relevant. The diary consisted of sets of

structured questions, 1 set for each illness. The parents filled in the

form every time the child was ill, marking the correct squares with an

X and recording the date of the illness. Parents confirmed the events

recorded in the diary when they visited the outpatient clinic with the

child at ages 6, 12, 18, and 24 months. In an attempt to record missed

illnesses, at each visit the parents were asked, ‘‘Has your child had

any illness since your last visit?’’ If additional illnesses were

mentioned, they were added to the diary.

Blood sampling

Venous blood samples were collected when the children were 24

months old. Plasma was separated by centrifugation and stored at

270C pending analysis.

Specific IgE

Circulating IgE antibodies against cow’s milk, egg white, peanut,

cod fish, soy bean, wheat, cat dander, dog dander, birch pollen,

timothy pollen, and Dermatophagoides farinae were determined in

plasma with Pharmacia CAP-FEIA (Pharmacia-Upjohn, Uppsala,

Sweden). A positive test was defined as an IgE antibody level0.35

kilo Units allergen-specific antibodies (kUA) per liter.

Classification of the children

In accordance with Johansson et al,8 the child was classified as

IgE-sensitized if at least 1 SPT was positive (3 mm) and/or if spe-

cific IgE against at least 1 of the selected allergens was0.35kUA/L.

To optimize the classification in IgE-sensitized and non–IgE-

sensitized children, the results of the in vivo and in vitro analysis

were combined.

Viral infections/serological methods

The serostatus against 13 viruses was investigated: respiratory

tract infections, including adenovirus, influenza (A/H1, A/H3, and

influenza B), parainfluenza (types 1, 2, 3), and RSV; and herpesvirus,

including CMV, EBV, herpes simplex virus (HSV), human herpes-

virus 6 (HHV6), and varicella-zoster virus (VZV).

IgG against the EBV capsid antigen and HHV6 was determined

according to previously published immunofluorescence assays.9,10

A specific fluorescence in dilution of 1/20 was regarded as a sign of

seropositivity.

For HSV, CMV, and VZV IgG ELISA with purified nuclear

antigens from the respective viruses cultivated in human fetal lung

fibroblasts were used.11,12 The cutoff for seropositivity was an

absorbance of >0.2 at a dilution of 1/100.

IgG antibodies against influenza A/H1, A/H3, and influenza B

were determined with ELISAs by using recombinant influenza

antigens.13 IgG antibodies against parainfluenza (serotypes 1, 2, 3),

RSV, and adenoviruses were measured by ELISA. The viruses were

cultured to full cytopathogenic effect either in human fetal lung

fibroblasts (RSV, adenovirus) orMA 104 cells (monkey kidney cells,

parainfluenza) and prepared mainly by ultracentrifugation and son-

ification of clarified supernatants. For adenovirus, sonicated, infected

cells were used. Preparations from the respective cell lines were used

as control antigen in the assays using cell culture antigen. Optical

densities above 0.3 after subtraction of control antigen activity were

regarded as a sign of past infection for the respiratory viruses.

Statistics

Descriptive statistics were used to characterize the data. x2

Analysis and the Student t test (2-tailed) were used for comparison

of IgE-sensitized and non–IgE-sensitized children where appropriate.

The number of serologically verified infections was normally

distributed (Fig 1), and the parentally reported infections were close

to normally distributed, and they were divided into quarters defined

by quartiles by using the statistical program SPSS 11.0 for Windows

(SPSS Inc, Chicago, Ill). There was variation in the size of the groups

as a result of characteristics of the distribution.

Odds ratios and 95% CIs were calculated for the development of

IgE sensitization. Data were adjusted for background variables by

using multiple logistic regression analysis. Adjustments were made

for sex, parental allergy (none, single-heredity, or double-heredity),

maternal age, parental smoking, furred pets at home, months of birth,

older siblings, duration of breast-feeding, socioeconomic status,

parentally reported infections, and seropositivity against viruses. All

of the measures were modeled as series of binary dummy variables.

The interaction of seropositivity for CMV and EBV was inves-

tigated by using logistic regression, with adjustment for the main

effects.

P values <.05 were considered statistically significant. The data

were analyzed by using Stata 7.0 (Stata Corp, College Station, Tex),

SPSS 11.0 for Windows, and the SAS System for Windows release

8.02 (SAS Institute, Cary, NC).



Nilsson et al 439


RESULTS

IgE sensitization

The children had an average age of 24.1 months (range,22-29) at the 24-month evaluation. Fifty-nine (24%)children were classified as IgE-sensitized. The in vitrotest (allergen-specific IgE in plasma) was positive in 52(22%) infants, whereas 35 (14%) infants had at least1 positive SPT against the selected food and inhalantallergens. The majority (n = 49; 83%) were sensitizedagainst food allergens, and sensitization toward individualallergens was for milk, 13.4%; egg, 7.7%; peanut, 6.1%;dog, 4.9%; wheat, 4.1%; cat, 3.7%; birch, 3.7%; soy,2.0%; fish, 1.2%; timothy, 0.8%; and mite, 0.8%.

There were no statistically significant differences insex, having a furred pet at home, or having smokingparents between IgE-sensitized and nonsensitized children(Table I).

However, the nonsensitized children were statisticallysignificantly more likely to have been born during thesummer than sensitized children and more frequently hadmore than 1 sibling. The sensitized children were statis-tically significantly more likely to have 2 atopic parents.The statistically significant association with short durationof breast-feeding was observed only in the adjustedanalyses, suggesting a complex set of associations or achance finding.

IgE sensitization and the parental reportof infections

The median number of parental reported infections was13 (range, 4-24) during the first 24 months of life. Forevaluation, the infants were divided into 4 groups definedby quartiles: 4 to 9, 10 to 13, 14 to 16, and 17 to 24reported infections. The association with IgE sensitization

TABLE I. Data among infants with positive SPT and/or positive specific IgE and infants with negative SPT/specific IgE

at 24 months of age

Whole cohort IgE-sensitized Nonsensitized OR; 95% CI ORadj; 95% CI

n (%) 246 59 (24.0) 187 (76.0)

Sex

Boy, n (%) 126 (51.2) 34 (27.0) 92 (73.0) 1 1

Girl, n (%) 120 (48.8) 25 (20.8) 95 (79.2) 0.71; 0.39-1.28 0.65; 0.32-1.31

Heredity

Nonheredity; n (%) 69 (28.0) 11 (15.9) 58 (84.1) 1 1

Double-heredity; n (%) 102 (41.5) 32 (31.4) 70 (68.6) 2.41; 1.12-5.20 3.79; 1.50-9.60

Maternal heredity; n (%) 75 (30.5) 16 (21.3) 59 (78.7) 1.43; 0.61-3.34 2.29; 0.83-6.31

Maternal age at delivery

Maternal age,

mean (range)

31.4 (21-44) 31.5 (22-44) 31.3 (21-43)

21-30 y, n (%) 103 (41.9) 22 (21.6) 81 (78.4) 1 1

31-44 y, n (%) 143 (58.1) 37 (25.9) 106 (74.1) 1.29; 0.70-2.35 1.42; 0.67-3.01

Month of birth

Born April-September, n (%) 148 (60.2) 27 (18.2) 121 (81.8) 1 1

Born October-March, n (%) 98 (39.8) 32 (32.7) 66 (67.3) 2.17; 1.20-3.93 2.23; 1.10-4.52

Exposure

Exclusive breast-feeding

0 mo, n (%) 13 (5.3) 2 (15.4) 11 (84.6) 0.61; 0.13-2.89 0.91; 0.13-6.60

0.5-3.9 mo, n (%) 34 (13.8) 10 (29.4) 24 (70.6) 1.40; 0.62-3.19 3.11; 1.10-8.85

4-5 mo, n (%)* 166 (67.5) 38 (22.9) 128 (77.1) 1 1

5.1-10 mo, n (%) 33 (13.4) 9 (27.3) 24 (72.7) 1.26; 0.54-2.95 1.21; 0.45-3.21

Mothers smoking, n (%) 10 (4.1) 1 (1.7) 9 (4.8) 0.34; 0.04-2.75 0.32; 0.03-3.19

Fathers smoking, n (%) 19 (7.7) 7 (11.9) 12 (6.4) 1.96; 0.73-5.24 2.24; 0.68-7.34

Furred pets at home, n (%) 45 (18.3) 9 (15.3) 36 (19.3) 0.76; 0.34-1.68 0.74; 0.28-1.94

Number of older siblings

0, n (%) 135 (54.9) 34 (25.2) 101 (74.8) 1 1

1, n (%) 81 (32.9) 24 (29.6) 57 (70.4) 1.25; 0.68-2.31 1.35; 0.65-2.80

>2, n (%) 30 (12.2) 1 (3.3) 29 (96.7) 0.10; 0.01-0.78 0.07; 0.01-0.65

Socioeconomic statusHigh, n (%) 141 (57.3) 34 (24.1) 107 (75.9) 1 1

Medium, n (%) 39 (15.9) 9 (23.1) 30 (76.9) 0.94; 0.41-2.19 0.97; 0.36-2.64

Low, n (%) 41 (16.7) 10 (24.4) 31 (75.6) 1.02; 0.45-2.28 1.35; 0.49-3.75

Studying, n (%) 11 (4.5) 4 (36.4) 7 (63.6) 1.80; 0.50-6.52 3.39; 0.67-17.06

Not specified, n (%) 14 (5.7) 2 (14.3) 12 (85.7) 0.52; 0.11-2.46 0.52; 0.09-3.14

OR, Odds ratio; ORadj, adjusted odds ratio.

*The recommended time for exclusively breast-feeding is at least 4 months in Sweden.

Grouped by the father’s occupation according to Statistics Sweden (Swedish government for official statistics).


AUGUST 2005

440 Nilsson et al

Basic

andclin

icalim

munology

for each group was evaluated. The sensitized childrenwere fewer in the group with 10 to 13 parentally reportedinfections, but there were no statistically significant dif-ference between the groups, and the trend across thegroups was not consistent (Table II).

IgE sensitization and the frequency ofseropositivity

The frequency of seropositivity at 24 months of ageagainst the selected viruses is presented in Table II. Allchildren apart from 1 had detectable IgG antibodiesagainst at least 1 of the viruses studied, and 1 child haddetectable IgG against all 13. There were no significantassociations between the number of parentally reportedinfections and the number of viruses identified throughserology (data not shown).

The mean number of viruses identified through serol-ogy was 5 (Fig 1). The children were divided into 4 groupsdefined by quartiles. Fifty-five children had antibodiesagainst 0 to 3 viruses, 95 children against 4 to 5 viruses, 42against 6 viruses, and 54 against 7 to 13 viruses, respec-tively. There was some suggestion that children withfewer antibodies were more often sensitized than childrenwith antibodies against many viruses, but this result wasnot statistical significant (Table II), and the pattern did notshow a consistent trend across the groups.

IgE sensitization and seropositivityto individual viruses

Respiratory viruses. The association between pres-ence of antibodies against the individual viruses andIgE sensitization is presented in Table II. There wereno statistically significant associations between IgEsensitization and seropositivity against the 8 airborneviruses—adenovirus, influenza (A/H1, A/H3, and B),parainfluenza (types 1, 2, 3), and RSV. Sixty-three percent(154 children) were seropositive against RSV at 24months of age, and among these, 21% (n = 33) wereIgE-sensitized.

Herpesviruses. Seropositivity against the virusesbelonging to the herpesvirus family—CMV, EBV, HSV,HHV6 andVZV—and IgE sensitization are also presentedin Table II. Seronegativity against EBV was statisticallysignificantly associated with IgE sensitization. Sixty-fourchildren (26%) were seropositive against EBV. Amongthese, 8 (12 %) were sensitized. This was significantly lessthan in the EBV seronegative group (odds ratio, 0.37; 95%CI, 0.16-0.82). The association remained statisticallysignificant after adjustment for all of the potential con-founding factors (Table II).

Cytomegalovirus IgG antibodies were detectable in 96(39%) children. Among these, 27 (28%) were IgE-sensi-tized. There were no statistically significant differences inthe numbers of seropositivity or seronegativity againstCMV when comparing the IgE sensitized and nonsensi-tized infants. However, seropositivity for EBV was morenegatively associated with sensitization among subjectswho were also seropositive for CMV. The odds ratio for

IgE sensitization associated with seropositivity againstEBV among children seronegative against CMVwas 0.75(95% CI, 0.30-1.91) but was reduced to 0.07 (95% CI,0.01-0.57) among children who were seropositive againstboth viruses. Interaction testing revealed a statisticallysignificant synergistic effect (effect modification), pro-ducing an odds ratio for the interaction of both viruseswith IgE sensitization of 0.10 (95% CI, 0.01-0.92) afteradjustment for the main effects.

We also observed an association with sensitizationamong subjects who were seropositive against CMV andseronegative against EBV. Among those seronegativeagainst EBV (n = 182), 26 of 71 were sensitized amongthose infected with CMV, compared with 25 of 111 whowere not infected with CMV. This produces an odds ratioof 1.99 (95% CI, 1.03-3.83), suggesting a modest in-creased risk associated with CMV infection in subjectswho are seronegative for EBV.

The serostatus against the other herpesviruses was notsignificantly associated with IgE sensitivity.

DISCUSSION

The hypothesis that infectious diseases during earlychildhood may have a protective role against the devel-opment of allergy, the hygiene hypothesis, was raised inthe late 1980s.5

In the current study, we did not find strong evidence ofan association between IgE sensitization and the numberof previous infections indicated either by parental reportor by seropositivity. However, IgE sensitization was lessprevalent at 2 years of age among infants who wereseropositive against EBV. The combination of havingboth CMV and EBV antibodies was more stronglynegatively associated with sensitization than would bepredicted by the individual associations of EBV or CMVantibodies alone, indicative of an interactive effect.

The prevalence of EBV antibodies in our study iscomparable with that of other industrialized countries.14

Previous studies of EBV are contradictory in relation todevelopment of atopy. Increased levels of antibodiesagainst EBV were found in children 5 to 18 years old

FIG 1. Frequency of seropositivity against the 13 selected viruses at

24 months of age in IgE sensitized* (n = 59) and nonsensitized

(n = 187) infants. *At least 1 positive SPT 3 mm and/or at least

1 specific IgE against the selected allergens 0.35 kUA/L.



Nilsson et al 441


with clinical signs of atopy compared with their nonatopiccounterparts.15 However, Calvani et al14 reported a higherprevalence of atopy among EBV seronegative children inthe age group 0 to 6 years, corroborating our results.A recent Swedish study (BAMSE) failed to demonstratethat the EBV serostatus in 4-year-old children correlatedwith IgE sensitization.16 In developing countries, whereEBV asymptomatically infects the majority of childrenbefore 3 years of age,17 the prevalence of atopy has beenreported to be lower than in industrialized countries.18 Incombination with our findings, these observations mightindicate an age-dependent role of EBV, rather than EBVinfection per se, in relation to IgE sensitization.

Cytomegalovirus becomes persistent after primary in-fection and seems to induce a TH1 cytokines response.19

CMV is frequently transmitted from mothers to infantsduring pregnancy, at delivery, or via breast milk.20 Thefew studies published on the relation between CMV andallergies are inconclusive.21 In the Swedish BAMSEstudy, no association was found between CMV seropos-itivity and IgE sensitization in 4-year-old children.However, among children with seropositivity againstCMV and seronegativity against EBV, there was a pos-itive association with sensitization to food allergens.7 Thisled us to test the interaction of seropositivity against CMVand EBV in relation to IgE sensitization. This indicatedeffect modification such that the negative association ofEBV seropositivity with sensitization was further en-hanced in children who were also seropositive for CMV.The mechanism for this putative interaction is unknown.

Several plausible explanations are possible, including theidea that IL-10 homologues present in the viruses mightdownregulate the antigen processing/presentation capac-ity of dendritic cells/macrophages and thereby switch offthe host T-cell system, similar to the downregulationobserved for T regulatory cells.22,23 Alternately, bothEBV and CMV can polyclonally activate B cells toproduce antibodies with many different specificities andthereby hinder the capacity of allergens to cross-link theB-cell receptor as seen for helmintic infections.24 Thus,these data and our results provide further support for thehypothesis that specific characteristics of EBV and/orCMV infection, rather than infection per se, might influ-ence the risk of IgE sensitization.

Interestingly, RSV infection, according to serology,was not associated with IgE sensitization at 24 months ofage. Previous studies have suggested that hospitalizationbecause of RSV infection is linked with atopy andinduction of IgE synthesis.25 Discrepancies between dif-ferent studies indicate that the vulnerability to RSV mightbe associated with a propensity for asthma and allergy, andnot that RSV per se causes asthma. Again, the character-istics of infection may be important, and identificationthrough hospitalization suggests more severe acute infec-tion than the majority of children in our study would havehad. Blanco-Quiros et al26 reported that infants who de-veloped severe RSV bronchiolitis had low levels of IL-12,a strong TH1 inducer, in cord blood. We have recentlypublished similar data showing low levels of IL-12 in cordblood of IgE sensitized infants at 24 months of age.27

TABLE II. Seropositivity against the investigated viruses and parentally reported infections in IgE-sensitized and

nonsensitized children

Whole cohort IgE-sensitized Nonsensitized OR; 95% CI ORadj; 95% CI

n 246 59 187

Parainfluenza 1, n (%) 70 (28.5) 16 (27.1) 54 (28.9) 0.92; 0.48-1.76 0.93; 0.46-1.86

Parainfluenza 2, n (%) 30 (12.2) 7 (11.9) 23 (12.3) 0.96; 0.39-2.36 1.14; 0.44-2.99

Parainfluenza 3, n (%) 166 (67.5) 42 (71.2) 124 (66.3) 1.26; 0.66-2.38 1.13; 0.58-2.21

Influenza Panama, n (%) 47 (19.1) 13 (22.0) 34 (18.2) 1.27; 0.62-2.61 1.32; 0.62-2.82

Influenza Texas, n (%) 30 (12.2) 6 (10.2) 24 (12.8) 0.76; 0.30-1.97 0.63; 0.23-1.76

Influenza Beijing, n (%) 92 (37.4) 16 (27.1) 76 (43.8) 0.54; 0.28-1.03 0.52; 0.26-1.01

Adenovirus, n (%) 200 (81.2) 46 (78.0) 154 (82.4) 0.76; 0.37-1.56 0.77; 0.36-1.63

RSV, n (%) 154 (62.6) 33 (55.9) 121 (64.7) 0.69; 0.38-1.26 0.71; 0.37-1.36

HHV6, n (%) 207 (84.2) 47 (79.7) 160 (85.6) 0.66; 0.31-1.40 0.66; 0.30-1.46

HSV, n (%) 24 (9.8) 6 (10.2) 18 (9.6) 1.06; 0.40-2.81 1.10; 0.40-3.02

CMV, n (%) 96 (39.0) 27 (45.8) 69 (36.9) 1.44; 0.80-2.60 1.20; 0.64-2.27

VZV, n (%) 45 (18.3) 8 (13.6) 37 (19.8) 0.64; 0.28-1.45 0.60; 0.25-1.42

EBV, n (%) 64 (26.0) 8 (13.6) 56 (29.9) 0.37; 0.16-0.82 0.34; 0.14-0.86

Frequency of seropositivity

0-3 viruses, n (%) 55 (22.4) 16 (27.1) 39 (20.9) 1.57; 0.71-3.50 1.53; 0.59-3.97

4-5 viruses, n (%) 95 (38.6) 25 (42.4) 70 (37.4) 1.15; 0.47-2.84 0.78; 0.28-2.21

6 viruses, n (%) 42 (17.1) 7 (11.9) 35 (18.7) 1.02; 0.42-2.51 0.99; 0.35-2.80

7-13 viruses, n (%) 54 (22.0) 11 (18.6) 43 (23.0) 1 1

Parentally reported infections

4-9 infections, n (%) 54 (22) 17 (28.8) 37 (19.8) 1.48; 0.64-3.46 1.40; 0.50-3.95

10-13 infections, n (%) 75 (30.5) 13 (22.0) 62 (33.2) 0.68; 0.29-1.61 0.57; 0.21-1.54

14-16 infections, n (%) 62 (25.2) 16 (27.1) 46 (24.6) 1.12; 0.48-2.61 1.11; 0.40-3.08

17-24 infections, n (%) 55 (22.3) 13 (22.0) 42 (22.5) 1 1

OR, Odds ratio; ORadj, adjusted odds ratio.


AUGUST 2005

442 Nilsson et al

Basic

andclin

icalim

munology

These observations might indicate an interaction amongcytokines, RSV, and sensitization during early infancy.

There was some indication that the numbers of sero-logically verified viral infections were inversely correlatedto IgE sensitization, but there was no consistent trendacross the groups and no statistically significant associa-tion. The number of parentally reported infections andseropositivity against the selected viruses did not corre-late. This is not entirely surprising, because many viruses,eg, rhinovirus and corona virus, which are proposed to bethe major causes of upper respiratory infections in infants,were not included in our analysis.28 There is not always arelation between viral infections and diseases, becauseasymptomatic infections are common. The suggestion ofan inverse association between the frequencies of paren-tally reported and serologically verified viral infectionswith sensitization is in agreement with previous publica-tions,29,30 although these authors have mainly studiedclinical signs of allergy. It is possible that the suggestedprotective effect of infections continues to influence thenascent immune system beyond age 2 years, and ourfollow-up was conducted at too young an age to observethe protective effect against allergic disease.

Previous studies have shown mainly indirect evidencefor the importance of infections for the development ofallergy,3,5,29 and these studies have often been retrospec-tive. The strength of our study is that the atopic status ofthe parents was characterized, and information on infec-tions was collected prospectively by using diaries.Importantly, objective serological measurements againstantiviral antibodies were made, and the results wereadjusted for factors that might bias the evaluation.

However, we recognize some limitations. The studypopulation was selected and not population-based.Because we selected children with different family histo-ries of allergy, we believe that the group of children in ourstudy is fairly comparable with children in the generalpopulation. Furthermore, the exact sensitivity for sero-positivity in some of the assays used to evaluate therespiratory infections is not fully known because theyhave been used mainly to detect ongoing infections, so therate of seropositivity is somewhat underestimated, but thisshould not introduce bias. The optimal approach wouldhave been to perform neutralization assays, but serumsamples from infants are inevitably insufficient for testsusing low serum dilutions. In contrast, the sensitivity andspecificity for seropositivity of the assays used for theherpesvirus infections have been proven reliable, as bestdemonstrated by the correlation between initial serostatusand clinical outcome found in transplant recipients.31 Itcould also be argued that analyzing 13 different virusescould produce some associations with IgE sensitization bychance. However, because an association specifically withEBV was evaluated as a result of an a priori hypothesisbased on the results of previous studies,14 the associationsreported for EBV are unlikely to be caused by chance.

In summary, our results indicate that an EBV infectionduring the first 2 years of life is associated with a reducedrisk of IgE sensitization. Some patterns of infection may

confer greater protection, such as EBV infection in theage-defined window of susceptibility with possible mod-ification through interaction with CMV. Development ofmore precise measures of patterns of acute infection andtheir immunological sequel will assist in our understand-ing of how early in life viral infections are implicated inthe etiology of IgE sensitization.

We thank the families who participated in the study.We also thank

Anna Stina Ander, Johan Berggren, Jeanette Harrysson, Lena

Jagdahl, and Monica Nordlund for assistance.

REFERENCES

1. Aberg N, Hesselmar B, Aberg B, Eriksson B. Increase of asthma, allergic

rhinitis and eczema in Swedish schoolchildren between 1979 and 1991.

Clin Exp Allergy 1995;25:815-9.

2. Matricardi PM, Franzinelli F, Franco A, Caprio G, Murru F, Cioffi D,

et al. Sibship size, birth order, and atopy in 11,371 Italian young men.


3. Matricardi PM, Rosmini F, Ferrigno L, Nisini R, Rapicetta M,

Chionne P, et al. Cross sectional retrospective study of prevalence of

atopy among Italian military students with antibodies against hepatitis

A virus. BMJ 1997;314:999-1003.

4. Benn CS, Melbye M, Wohlfahrt J, Bjorksten B, Aaby P. Cohort study of

sibling effect, infectious diseases, and risk of atopic dermatitis during

first 18 months of life. BMJ 2004;328:1223-8.

5. Strachan DP. Hay fever, hygiene, and household size. BMJ 1989;299:

1259-60.

6. Grillner L, Strangert K. A prospective molecular epidemiological study

of cytomegalovirus infections in two day care centers in Sweden: no

evidence for horizontal transmission within the centers. J Infect Dis

1988;157:1080-3.

7. Sidorchuk A, Wickman M, Pershagen G, Lagarde F, Linde A. Cyto-

megalovirus infection and development of allergic diseases in early

childhood: interaction with EBV infection? J Allergy Clin Immunol

2004;114:1434-40.

8. Johansson SG, Hourihane JO, Bousquet J, Bruijnzeel-Koomen C,

Dreborg S, Haahtela T, et al. A revised nomenclature for allergy: an

EAACI position statement from the EAACI nomenclature task force.

Allergy 2001;56:813-24.

9. Linde A, Fridell E, Dahl H, Andersson J, Biberfeld P, Wahren B. Effect

of primary Epstein-Barr virus infection on human herpesvirus 6,

cytomegalovirus, and measles virus immunoglobulin G titers. J Clin

Microbiol 1990;28:211-5.

10. Linde A, Andersson J, Lundgren G, Wahren B. Subclass reactivity to

Epstein- Barr virus capsid antigen in primary and reactivated EBV

infections. J Med Virol 1987;21:109-21.

11. Sundqvist VA, Linde A, Wahren B. Virus-specific immunoglobulin G

subclasses in herpes simplex and varicella-zoster virus infections. J Clin

Microbiol 1984;20:94-8.

12. Sundqvist VA, Wahren B. An interchangeable ELISA for cytomegalo-

virus antigen and antibody. J Virol Methods 1981;2:301-12.

13. Stepanova L, Naykhin A, Kolmskog C, Jonson G, Barantceva I,

Bichurina M, et al. The humoral response to live and inactivated

influenza vaccines administered alone and in combination to young

adults and elderly. J Clin Virol 2002;24:193-201.

14. Calvani M, Alessandri C, Paolone G, Rosengard L, Di Caro A,

De Franco D. Correlation between Epstein Barr virus antibodies, serum

IgE and atopic disease. Pediatr Allergy Immunol 1997;8:91-6.

15. Strannegard IL, Strannegard O. Epstein-Barr virus antibodies in children

with atopic disease. Int Arch Allergy Appl Immunol 1981;64:314-9.

16. Sidorchuk A, Lagarde F, Pershagen G, Wickman M, Linde A. Epstein-

Barr virus infection is not associated with development of allergy in

children. Pediatr Infect Dis J 2003;22:642-7.

17. Cohen JI. Epstein-Barr virus infection. N Engl J Med 2000;343:481-92.

18. Addo Yobo EO, Custovic A, Taggart SC, Asafo-Agyei AP,

Woodcock A. Exercise induced bronchospasm in Ghana: differences



Nilsson et al 443


in prevalence between urban and rural schoolchildren. Thorax 1997;

52:161-5.

19. Rentenaar RJ, Gamadia LE, van DerHoek N, van Diepen FN, Boom R,

Weel JF, et al. Development of virus-specific CD4(1) T cells during

primary cytomegalovirus infection. J Clin Invest 2000;105:541-8.

20. Ahlfors K, Ivarsson SA, Harris S. Report on a long-term study of

maternal and congenital cytomegalovirus infection in Sweden: review of

prospective studies available in the literature. Scand J Infect Dis 1999;31:

443-57.

21. Wu CA, Puddington L, Whiteley HE, Yiamouyiannis CA, Schramm

CM, Mohammadu F, et al. Murine cytomegalovirus infection alters

Th1/Th2 cytokine expression, decreases airway eosinophilia, and

enhances mucus production in allergic airway disease. J Immunol

2001;167:2798-807.

22. Salek-Ardakani S, Arrand JR, Mackett M. Epstein-Barr virus encoded

interleukin-10 inhibits HLA-class I, ICAM-1, and B7 expression on

human monocytes: implications for immune evasion by EBV. Virology

2002;304:342-51.

23. Raftery MJ, Wieland D, Gronewald S, Kraus AA, Giese T, Schonrich G.

Shaping phenotype, function, and survival of dendritic cells by cyto-

megalovirus-encoded IL-10. J Immunol 2004;173:3383-91.

24. Smits HH, Hartgers FC, Yazdanbakhsh M. Helminth infections: protec-

tion from atopic disorders. Curr Allergy Asthma Rep 2005;5:42-50.

25. Schauer U, Hoffjan S, Bittscheidt J, Kochling A, Hemmis S, Bongartz S,

et al. RSV bronchiolitis and risk of wheeze and allergic sensitisation in

the first year of life. Eur Respir J 2002;20:1277-83.

26. Blanco-Quiros A, Gonzalez H, Arranz E, Lapena S. Decreased

interleukin-12 levels in umbilical cord blood in children who developed

acute bronchiolitis. Pediatr Pulmonol 1999;28:175-80.

27. Nilsson C, Larsson A-K, Soderlund A, Gabrielsson S, Troye Blomberg

M, Lilja G. Low numbers of IL-12-producing cord blood mononuclear

cells and immunoglobulin E sensitization in early childhood. Clin Exp

Allergy 2004;34:373-80.

28. Johnston SL. Natural and experimental rhinovirus infections of the

lower respiratory tract [review]. Am J Respir Crit Care Med 1995;152:

S46-52.

29. Celedon JC, Wright RJ, Litonjua AA, Sredl D, Ryan L, Weiss ST, et al.

Day care attendance in early life, maternal history of asthma, and asthma

at the age of 6 years. Am J Respir Crit Care Med 2003;167:1239-43.

30. Lau S, Nickel R, Niggemann B, Gruber C, Sommerfeld C, Illi S, et al.

The development of childhood asthma: lessons from the German

Multicentre Allergy Study (MAS). Paediatr Respir Rev 2002;3:265-72.

31. Ljungman P, Aschan J, Lewensohn-Fuchs I, Carlens S, Larsson K,

Lonnqvist B, et al. Results of different strategies for reducing cytomeg-

alovirus-associated mortality in allogeneic stem cell transplant recipients.

Transplantation 1998;66:1330-4.


AUGUST 2005

444 Nilsson et al

Basic

andclin

icalim

munology

Biased use of VH5 IgE-positive B cells in thenasal mucosa in allergic rhinitis

Heather A. Coker, PhD,a* Helen E. Harries, MBiochem,a Graham K. Banfield, FRCS,b

Victoria A. Carr, RGN,b Stephen R. Durham, MD,b Elfy Chevretton, FRCS,c

Paul Hobby, MSc,a Brian J. Sutton, PhD,a and Hannah J. Gould, PhDa

London, United Kingdom

Background: IgE antibody-producing B cells are enriched in

the nasal mucosa in patients with allergic rhinitis because of

local class switching to IgE. The expressed IgE VH genes also

undergo somatic hypermutation in situ to generate clonal

families. The antigenic driving force behind these events is

unknown.

Objective: To examine the possible involvement of a super-

antigen in allergic rhinitis, we compared the variable (VH) gene

use and patterns of somatic mutation in the expressed IgE

heavy-chain genes in nasal biopsy specimens and blood from

allergic patients and the IgA VH use in the same biopsy

specimens and also those from nonallergic controls.

Methods: We extracted mRNA from the nasal biopsy specimens

of 13 patients and 4 nonallergic control subjects and PBMCs

from 7 allergic patients. IgE and IgAVH regions were RT-PCR

amplified, and the DNA sequences were compared with those

of control subjects. We constructed a molecular model of VH5

to locate amino acids of interest.

Results: We observed a significantly increased frequency of

IgE and IgAVH5 transcripts in the nasal mucosa of the

allergic patients compared with the normal PBMC repertoire.

Within IgE and IgAVH5 sequences in the nasal mucosa, the

distribution of replacement amino acids was skewed toward the

immunoglobulin framework regions. Three of 4 nonintrinsic

hotspots of mutation identified in the VH5 sequences were in

framework region 1. The hotspots and a conserved VH5-specific

framework residue form a tight cluster on the surface of VH5.

Conclusion: Our results provide evidence for the activity of

a superantigen in the nasal mucosa in patients with allergic

rhinitis. (J Allergy Clin Immunol 2005;116:445-52.)

Key words: Human, allergic rhinitis, VH5, superantigen, B lympho-

cyte, mucosa

IgE and its receptors are central to allergic disease,manifested in different target organs, including the nose(allergic rhinitis), the lung (allergic asthma), the skin(atopic dermatitis), and the gut (allergic gastroenteritis).IgE binds to effector and antigen-presenting cells bearingIgE receptors (FceRI, CD23, or both) in mucosal tissuesassociated with the target organs mediating the allergicresponse. In addition, IgE-mediated allergen presentationto T helper cells might lead to renewed IgE antibodysynthesis, epitope spreading, and exacerbation of allergicsensitivity. A large number of genetic and environmentalrisk factors for allergy have been identified, providingbroad insight into the pathogenesis of allergic disease. Thefactors that determine the susceptible target organ indifferent individuals exposed to the same allergens are,however, unknown. We suggest that these factors mightbe localized in the target organ and might include super-antigens.

Snow and coworkers1-5 have observed a bias in therepertoire of IgE heavy-chain variable (IgE VH) regions inasthma, which exhibited the hallmarks of a superantigen.There are 51 VH genes grouped into 7 gene families (VH1-VH7), varying in size from 22 members in VH3 to 1 or2 members in VH5, VH6, and VH7.

6 Each VH region has3 framework regions (FWR1-FWR3) alternating with3 complementarity-determining regions (CDR1-CDR3).The FWRs determine the structural framework for theantigen-binding sites of the CDRs. The CDR sequencesare inherently more prone to somatic hypermutationduring affinity maturation of antibodies in the immuneresponse, whereas the FWR sequences are relativelyconserved.7 In peripheral blood B cells of healthy indi-viduals, the proportion of expressed VH regions fromdifferent VH families generally reflects the size of thefamily.8 In asthma, however, Snow and coworkers foundan overabundant use of the minor VH5 family in IgE in theblood, lung mucosa, and spleen of asthmatic patients1-3

Abbreviations usedCDR: Complementarity-determining region

FWR: Framework region

R/S: Replacement/silent mutation ratio

From aThe Randall Division of Cell andMolecular Biophysics, King’s College

London; bUpper Respiratory Medicine, National Heart and Lung Institute,

Imperial College, London; and cthe Department of Respiratory Medicine

and Allergy, Guy’s Hospital, London.

*Dr Coker is currently affiliated with Clare Hall Laboratories, Cancer Research

UK, South Mimms, Hertfordshire, United Kingdom.

Supported by the Clinical Research Committee Royal Brompton and Harefield

Hospitals NHS Trust. HAC and HEH were supported by BBSRC PhD

studentships, and HG and SRD were supported by project grants from

Asthma UK (grant no. 03/055) and the MRC (grant no. G0200485).

Received for publication August 25, 2004; revised March 23, 2005; accepted



Reprints of this article are not available from the authors.

0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.032

445


and in the blood of the majority of 6 asthmatic patients.5

Overabundant use of VH5was also reported in the blood of3 patients with atopic dermatitis9 but not in another studyof 2 patients with atopic dermatitis.10 By contrast, inthe blood of 2 subjects with peanut allergy, Snow andcoworkers observed an overabundant use of VH1, sug-gesting the activity of a peanut-associated superantigen.11

B-cell superantigens, similar to T-cell superantigens,act by binding to immunoglobulin FWRs, leading toclonal amplification of all members of the family. Becausethe number of CDRs in the B-cell repertoire (millions)is vastly greater than the number of VH families (7),this results in biased use of the selected family.12

Two previously identified B-cell superantigens,Staphylococcus aureus protein A and HIV gp120, selec-tively expand B cells expressing VH3.

12-14 The interactionof protein A with the FWR of VH3 can be seen in thecrystal structure of a complex with the VH3 Fab fragmentof an IgM antibody.15 Superantigen-selected B cells mightalso exhibit a skewed distribution of amino acid substitu-tions away from CDRs toward the FWR.7,16 The frequentoccurrence of point mutation hotspots in DNA sequencesthat are intrinsically more susceptible to hypermutation isa further criterion for CDR- versus FWR-oriented selec-tion; the sequences WRC and WA, which occur morefrequently in the CDRs, are recognized as intrinsichotspots of mutation.17 These 2 features of superantigenselection of B cells were also linked to the VH5 over-abundance in asthma and atopic dermatitis,2,9 which isconsistent with the influence of a superantigen and incontrast to the lack of such influence observed in VH5B cells from normal spleen.18

In a previous study of the expressed VH regions inallergic rhinitis, we presented evidence of local clonalexpansion, somatic hypermutation, and class switching inthe nasal mucosa in patients with allergic rhinitis.19 In thisstudy we present a detailed analysis of IgE and IgA VH

family use in the nasal mucosa in patients with allergicrhinitis.

METHODS

Samples from patients with allergic rhinitisand nonallergic control subjects

Male and female donors with allergic rhinitis aged between 18 and

55 years were recruited for this study. The allergic status of the donors

was assessed on the basis of medical history, skin prick tests, and,

where possible, serum allergen-specific IgE (RAST). Of the 11 tissue

samples from the nasal mucosa, 10 originated from patients with

multiple allergies (CD6, JB7, CM10, HD14, SO16, AP19, SJ24,

TL25, CA30, and SLT1)whowere allergic to grass and also allergens

such as animal dander or house dust mite, to which they could be

perennially exposed. One tissue sample (HD17) originated from a

patient with only grass pollen allergy. The samples were taken

throughout the year, with the patient with only grass pollen allergy

undergoing biopsy within the grass pollen season. Of the 11 nasal

mucosa samples analyzed, 10 were biopsy specimens taken from the

inferior turbinate, and one (SLT1) consisted of a piece of an inferior

turbinate removed by surgery to alleviate nasal obstruction. In

agreement with previous authors,9,19 we were unable to use healthy

subjects as a control group for IgE because PCR amplification of IgE

was only successful in the allergic subjects. Instead, we recruited 4

nonallergic subjects (AA2, MTS3, KB5, and SK6) and used biopsy

specimens from 2 of the allergic patients (TL25 and CA30) plus 2

additional patients with multiple allergies (JC1 and IB4) to examine

VH use in IgA.

Volunteers were recruited from the Royal Brompton Hospital

Allergy Clinic or by advertisement in the local press to donate a nasal

biopsy specimen. None had received immunotherapy, and any

medication was discontinued at least 2 weeks before nasal biopsy.

Biopsies were performed at the Royal Brompton Hospital, London,

United Kingdom, and processed as described previously.20 All such

work had the approval of the local ethics committee and the patients’

written informed consent. Blood samples were taken from 7 of the 16

patients who also donated nasal biopsy specimens (CD6, JB7, CM10,

HD14, SO16, HD17, and AP19). PBMCs, including B cells, were

isolated from these samples, as described previously.19 The tissue

sample from the inferior turbinate resulted from operations performed

at Guy’s Hospital, London, United Kingdom,with the approval of the

Guy’s Research Ethics Committee and also with the patient’s written

informed consent.

Amplification and analysis ofVH region sequences

Total RNA was extracted from both the nasal mucosa and PBMC

samples, cDNA was produced, and VH-Ce sequences were PCR

amplified by using the proofreading Pfu DNA polymerase (Promega,

Madison, Wis) from IgE-positive B cells. The VH-Ce PCR products

were then cloned and sequenced. This entire procedure has been

described in detail previously.19 VH-Ca sequences were amplified

from cDNA in the same way as VH-Ce sequences, replacing the

Ce-specific primers with nested primers specific for Ca: Ca1, 5#-TTTCGCTCCAGGTCACAC-3#; Ca2, 5#-GGGAAGAAGCCCT-GGACCAGGC-3#. The annealing temperature for the second round

of PCRwas adjusted from 65C to 69C. Assignment of the VH genes

and their somatic mutations was carried out according to their

homology with the germline sequences detailed on the VBase

database (www.mrc-cpe.cam.ac.uk).

Only unique sequences were included in the analysis, meaning

that repeat copies of identical sequences and also sequences that

originated from clonally related B cells (determined by an identical

CDR3/FWR4 signature region) were included only once in the

analysis. However, when sequences did originate from related

B cells, each unique mutation isolated from the family members

was included in the mutational analysis.

Statistical significance was determined by the use of x2 analysis

with the Yates correction for continuity (generating a more con-

servative calculation of significance when smaller data values are

used).

Molecular modeling of VH5 antibody structure

Because no VH5 antibody crystal structure is available, a model

was generated for the VH5-51 germline sequence on the basis of the

known structure of the most closely matched antibody (PDB code:

1CGS). The heavy-chain CDR1 and CDR2 (H1 and H2) loop lengths

were identical in both antibody sequences, and the heavy-chain

CDR3 (H3) loop and the light chain (VL) domain structures were

taken directly from 1CGS to produce a complete Fv model. No steric

clashes were detected after substitution of the VH5-51 sequence. The

model was generated by using HOMOLOGY and displayed with

INSIGHT II (Accelrys, Cambridge, United Kingdom).


AUGUST 2005

446 Coker et al

Basic

andclin

icalim

munology

http://www.mrc-cpe.cam.ac.uk/

RESULTS

IgE VH gene use

We examined a total of 62 in-frame VH region se-quences derived from the nasal mucosa of the 11 patientswith allergic rhinitis* and 50 VH region sequences fromthe PBMCs of 7 of these patients. There was a cleardifference in the VH gene use observed in IgE B cells fromthe nasal mucosa of the allergic patients compared withthat expected on the basis of the normal genomic PBMCrepertoire21 and also compared with that observed in IgE-positive PBMCs from the allergic patients (Fig 1).21

There was a highly significant decrease in the use ofVH3 by IgE-positive B cells in the allergic nasal mucosa(34%) compared with that expected on the basis of thenormal repertoire in PBMCs (55%) observed by previousworkers (P < .005, x2 analysis).21 This was not, however,significant when compared with that observed in theallergic PBMCs (50%; P > .1, x2 analysis).

There was also a highly significant increase in the useof VH5 in the nasal mucosa (29%) compared with thatobserved by previous researchers in normal PBMCs(2.9%; P < .005, x2 analysis). The increased use of VH5in the allergic nasal mucosa was significant when com-pared with the 8% VH5 use observed in the allergic IgE-positive PBMCs (P < .025, x2 analysis). There was nosignificant difference in the use of VH5 in the allergicPBMCs compared with that expected on the basisof the normal PBMC repertoire (8% vs 2.9%; P > .1,x2 analysis).

These data were taken from sequences pooled frommultiple patients. The occurrence of VH5 sequencesacross the PBMC samples appeared to be evenly distrib-uted (with the 4 VH5 sequences occurring in 4 of the 7different PBMC samples). This was in contrast to thedistribution of VH5 sequences in the samples from thenasal mucosa, in which the 18 VH5 sequences wereisolated from only 5 of the 11 patients. There was aparticularly high level of VH5 use in the nasal mucosa ofpatient CA30, in whom 11 of 18 sequences were fromunrelated B cells expressing VH5 IgE-positive sequences.However, even when this patient was excluded from theanalysis, the increased VH5 IgE use (14%) in the allergicnasal mucosa compared with that expected on the basis ofthe normal PBMC repertoire was still highly significant(P < .005, x2 analysis), although the significance betweenthe allergic nasal mucosa and allergic PBMCs was lost(P > .1, x2 analysis).

Hotspots of mutation in IgE VH5 sequencesfrom the allergic nasal mucosa

All of the sequences isolated from nasal biopsy spec-imens exhibited evidence of somatic hypermutation.Somatic mutations evident within 17 distinct VH5 se-quences were compared with those from 19 randomlychosen non-VH5 sequences to determine whether therewas evidence to suggest that the overuse of VH5 in theallergic nasal mucosa might have been a consequence ofB-cell selection (eg, by the presence of nonintrinsichotspots of mutation).

The percentage variability of each codon was used toidentify apparent hotspots of somatic hypermutation(Fig 2). The percentage mutation (including both silentand replacement mutations) and the percentage mutationat each nucleotide were also analyzed (data not shown),

FIG 1. VH gene use in allergic nasal mucosa IgE-positive B cells. Sixty-two sequences from 11 allergic nasal

mucosa samples (filled bars) and 50 sequences from 7 allergic PBMC samples (gray bars) were pooled to

compare VH gene use. VH gene use in normal PBMCs (open bars) was also included for comparison.21

*Nucleotide sequences submitted to Genbank, accession numbers AY640472–

AY640533.

Nucleotide sequences submitted to Genbank, accession numbers AY640534–

AY640583.



Coker et al 447


FIG 2. Distribution of somatic mutations across IgE VH. A, Somatic mutations identified in 17 allergic nasal

mucosa VH5 sequences were pooled. Hotspots of mutation were identified on the basis of the percentage

variability at each codon. B, Nineteen non-VH5 sequences randomly chosen from the same cohort of allergic

patients were subjected to the same analysis.


AUGUST 2005

448 Coker et al

Basic

andclin

icalim

munology

although neither identified any further hotspots of muta-tion to that evident from Fig 2.

Of the hotspots of mutation identified in Fig 2, those atMet 40 and Tyr 56 in the VH5 sequences and at codons52a, 56, and 84 in the non-VH5 sequences, when examinedin more detail, were disregarded because the mutationswere spread evenly across the 3 nucleotide positions of thecodon and had very different effects on the amino acid.The apparent hotspots at codon 52b and also codon 82a inthe non-VH5 sequences were also disregarded becausecodon 52b was mutated in both of only 2 sequences thatused this polymorphic codon, and the mutations at codon82a were divided between 3 Asn and 3 Ser residues, suchthat there were insufficient mutations of each for analysis.

The exact nature of the remaining apparent hotspots ofmutation was examined in detail to determine whether thedirection of mutation at each codon reflected the acceptedtrends of somatic hypermutation, as defined by previousauthors.22 Those mutations that conformed to such aprofile were defined as intrinsic, whereas those that clearlydiffered were defined as nonintrinsic and therefore likelyto have been selected in response to antigen (Table I).22

The only hotspot apparent in the non-VH5 sequenceswas a commonly identified hotspot, Ser 31,22 which wasclassified as intrinsic. In contrast, although 9 of the 13remaining hotspots identified in the VH5 sequences alsoappeared to be intrinsic in nature, importantly, 4 apparentnonintrinsic hotspots were also identified. This is consis-tent with the non-VH5 IgEmolecules having been targetedtoward awide range of antigens, whereas the identificationof nonintrinsic hotspots of mutation in the VH5 sequencessuggests that the VH5 IgE molecules were targeted towarda limited number of antigens. The 4 nonintrinsic mutationsevident in the VH5 sequences were present at Lys 23(1),Gly 24(2), and Thr 30(2), each in FWR1, and also atTyr 52(2), which was found in CDR2. According toKabat’s numbering, CDR1 incorporates codons 31through 35, and CDR2 incorporates codons 50 through65. FWR1 incorporates codons 1 through 30, FWR2incorporates codons 36 through 49, and FWR3 incorpo-rates codons 66 through 95.

Analysis of replacement/silent mutationratio values

Because 3 of the 4 apparently nonintrinsic hotspots ofmutation in the VH5 sequences were unconventionallypresent in the framework region, further analysis of theratio of replacement to silent mutations (R/S values) in thedifferent sequences was carried out. The non-VH5 groupof sequences exhibited R/S values that were consistentwith conventional antigen selection, with a significantdifference between the CDR and the FWR (CDR, 3.46;FWR, 1.67; P < .025, x2 analysis). In stark contrast, therewas no significant difference between the R/S values in theCDR and the FWR of the VH5 group of sequences (CDR,2.38; FWR, 2.06; P > .1, x2 analysis).

Because the VH5 FWR R/S value was higher thanexpected and also because 3 of the 4 nonintrinsic muta-

tions were present in FWR1, analysis of the R/S values ofthe individual FWRs in the VH5 group of sequences wasalso carried out. FWR1 exhibited an R/S value of 2.16,FWR2 exhibited an R/S value of 1.06, and FWR3exhibited an R/S value of 2.61, suggesting that althoughFWR2 was conventionally conserved in the VH5 se-quences, FWR1 and FWR3 contributed to the unusualoverall FWR R/S value observed in the VH5 sequences.

TABLE I. The direction of somatic mutation at hotspots

in VH IgE sequences from the allergic nasal mucosa

suggests the presence of 4 nonintrinsic hotspots in

VH5 sequences

Hotspot From To O* Ey

VH5

Lys 23(1) AAG CAG Gln 6 2 Nonintrinsic

GAG Glu 2 4

TAG Stop 0 2

Gly 24(2) GGT GAT Asp 0 2 Nonintrinsic

GCT Ala 5 0

GTT Val 1 4

Ser 28(3) AGC AGA Arg 3 1

AGG Arg 2 2

AGT Ser 4 6

Thr 30(2) ACC AAC Asn 1 1 Nonintrinsic

AGC Ser 4 1

ATC Ile 2 5

Ser 31(2) AGC AAC Asn 7 7

ACC Thr 5 4

ATC Ile 0 1

Ser 31(3) AGC AGA Arg 1 1

AGG Arg 0 1

AGT Ser 6 5

Tyr 32(3) TAC TAA Stop 0 0

TAG Stop 0 1

TAT Tyr 4 3

Tyr 52(2) TAT TCT Ser 1 1 Nonintrinsic

TGT Cys 1 4

TTT Phe 5 2

Asp 53(3) GAT GAA Glu 4 2

GAC Asp 2 4

GAG Glu 1 1

Ser 57 (3) AGC AGA Arg 0 1

AGG Arg 0 1

AGT Ser 7 5

Ala 71(2) GCC GAC Asp 0 1

GGC Gly 0 1

GTC Val 6 4


ACC Thr 1 2

ATC Ile 0 1

Met 89(3) ATG ATA Ile 4 3

ATC Ile 0 2

ATT Ile 1 0

Non-VH5


ACC Thr 0 2

ATC Ile 0 0

*O denotes the observed number of somatic mutations.

E denotes the expected direction of mutation on the basis of previous

research.22



Coker et al 449


IgA VH gene use

If a superantigen was involved in the biased use ofVH5, this might be apparent in isotypes other than IgE.We first examined this possibility by using 2 adjacentbiopsy specimens (CA30A and CA30B) from patientCA30.

As shown in Table II, all 11 IgE VH sequences frombiopsy specimen A but none of the 4 sequences frombiopsy specimen B were VH5. IgA exhibited a similarpattern, with 2 of 12 IgA VH sequences in biopsyspecimen A but none of 10 in biopsy specimen B beingVH5. Combining IgE and IgA results, 57% (13/37) ofsequences in biopsy specimen A were VH5, whereas 0%(0/14) of sequences in biopsy specimen B was VH5. Noneof 13 VH5 sequences in CA30 were related to any of theothers, suggesting polyclonal activation of B cells ex-pressing VH5 sequences locally in the region of biopsyspecimenA. The results for CA30 demonstrate the linkagebetween VH5 bias in IgE and IgA and suggest that a B-cellsuperantigen might influence the selection of VH in a localmanner.

The FWR and CDR R/S values for IgA VH5 and non-VH5 were determined and compared. R/S values forIgA non-VH5 sequences in CA30 were comparable withthose of IgE non-VH5 sequences (CDR, 4; FWR, 1.4)and thus indicative of normal antigen selection. However,as observed for IgE VH5, the FWR and CDR R/S valuesfor IgA VH5 sequences were similar (CDR, 4; FWR,4.3).

On analysis of 73 IgA VH sequences in 2 of the biopsyspecimens from the original cohort (TL25 and CA30) plus2 additional biopsy specimens (JC1 and IB4) from allergicpatients, we observed VH5 to be significantly higher thanin normal PBMCs (14% vs 2.9%; P < .005, x2 analysis).21

The greater abundance of IgA-expressing compared withIgE-expressing B cells in the nasal mucosa allowed us todetermine the IgA VH use in the nasal mucosa of healthynonallergic subjects. Of 55 sequences§ from 4 nonallergicdonors, we found a 9% frequency of VH5, which is alsosignificantly different from that seen in normal PBMCs(P <.05, x2 analysis).21

Location of nonintrinsic hotspots in thestructure of IgE VH5

The 4 nonintrinsic hotspots occurred at Lys 23(1),Gly 24(2), Thr 30(2), and Tyr 52(2). Of these, 3 wereunusually situated within FWR1, with only Tyr 52(2) inCDR2. Therefore it would seem likely that the putativesuperantigen has contact with FWR1 and CDR2, althoughthe R/S values imply that FWR3 might also be involved.The B-cell superantigen protein A has been shown to bindsimilarly to VH3, interacting with FWR1, FWR3, andCDR2.12,13,15,23

Because no crystal structure of a VH5 is available, weconstructed a model on the basis of the known structure ofthe most closely matched antibody. There are 2 VH5genes, VH5-51 and VH5-a, but only the former is ex-pressed in the majority of the population. Fig 324 showsLys 23, Gly 24, and Thr 30 in FWR1 and Tyr 52 in CDR2in our model of VH5-51.

When the locations of the 4 nonintrinsic hotspots areidentified within this model, it is immediately apparentthat they are clustered together at the edge of the conven-tional antigen-binding site, as defined by the 6 CDRs. Lys23 and Gly 24 are adjacent to CDR1, and Thr 30 lies at thevery boundary between FWR1 and CDR1; Tyr 52 lies atthe boundary between FWR2 and CDR2. Also shown inFig 3 is residue Ile 75, a hydrophobic surface residue in anunusually exposed location but conserved in the vastmajority of VH5 sequences. In fact, all VH2, VH3, VH4,and VH6 germline sequences have lysine at this position,whereas it is isoleucine only in VH5. This cluster ofresidues might thus define at least a part of the site at whicha putative superantigen might interact with IgE VH5.

DISCUSSION

In our previous study of somatic hypermutation in VH

sequences in IgE-expressing B cells in the nasal mucosa ofpatients with allergic rhinitis, we observed that 2 of a total

TABLE II. Regional diversification of IgE and

IgA VH sequences

Isotype

Biopsy

specimen

Number of sequences

VH1 VH2 VH3 VH4 VH5 VH6 VH7 Total

IgE CA30A 0 0 0 0 11 0 0 11

CA30B 0 0 1 3 0 0 0 4

IgA CA30A 1 0 8 1 2 0 0 12

CA30B 2 0 6 1 0 1 0 10

FIG 3. Location of nonintrinsic hotspots in the 3-dimensional

structure. Hotspots Lys23, Gly24 and Thr30 (FWR1), and Tyr52

(CDR2) in VH5 are shown in yellow on a homologymodel of VH5-51.

Exposed hydrophobic residue Ile75 is shown in green. CDRs are

indicated, encompassing both Kabat and Chothia24 definitions. The

remaining VH and VL FWRs are shown in dark and light blue.

Nucleotide sequences submitted to Genbank, accession numbers AY971069-

AY971142

§Nucleotide sequences submitted to Genbank, accession numbers AY971012-

AY971068


AUGUST 2005

450 Coker et al

Basic

andclin

icalim

munology

of 3 clonal families were derived from the VH5 family, andwe detected an IgA VH5 clone related to one of thefamilies.19 This hinted at a biased use of VH5 in the tissuesimilar to that found in asthma and atopic dermatitis.1-5,9

We have therefore analyzed VH sequences from a largecohort of patients with allergic rhinitis to test this hypoth-esis.

Our results have revealed a significant bias toward VH5,even when one biopsy (CA30), which exhibited the mostextreme bias, was excluded. A similar trend in the PBMCsfrom a subgroup of these patients did not reach statisticalsignificance. The greater bias in tissue suggests that a localevent caused the VH5 bias. The similar trend in PBMCsmight result from the displacement of VH5-expressingB cells from the tissue and dilution in the pool of normalB cells in the circulation. IgAVH sequences also exhibiteda significant but less extreme bias toward VH5 comparedwith that seen in IgE sequences. The demonstration ofVH5 bias in both IgE and IgA and colocalization in thenasal mucosa support the superantigen hypothesis. Theputative superantigen might bind to the subset of B cellsexpressing VH5, leading to selective proliferation orrescue from apoptosis. The smaller bias in IgA VH5 mightbe due to the TH2 environment of the nasal mucosa, whichfavors class switching to IgE in rapidly dividing B cells.We observed an even smaller IgA VH5 bias in the normalnasal mucosa, which might hint that the bias precedes thedevelopment of allergy. Allergen-specific B cells might befurther selected from this population.

Although there was a significant IgE VH5 bias in the 62sequences analyzed, these sequences came from a minor-ity (5/11) of the patient population. This must not be takento indicate that a superantigen is likely to have been actingin only the 5 patients, however. We have demonstratedpreviously19 and again here, exemplified by the compar-ison of biopsy specimens A and B from patient CA30, thatthere are not necessarily any identical sequences or evenclonal relationships between cells in adjacent 1- to 2-mm3

biopsy specimens. These biopsy specimens (10-20 mg)are less than 1% of the inferior turbinates (average weight,approximately 2 g). Thus sequences derived from a singlebiopsy specimen are not necessarily representative of thetissues in individual patients. Our VH5 bias is based onsampling a population of patients. The absence of VH5sequences in 6 of 11 of the biopsy specimens does notimply that there are none elsewhere in the tissue.Superantigens acting in other regions of the tissue mightstill contribute to the ‘‘VH5-negative patients’’’ symptomsof allergy.

Our comparison of VH5 and non-VH5 R/S values andthe positions of the nonintrinsic mutations within FWR,notably FWR1, provide additional evidence to supportthe superantigen hypothesis. Moreover, the 3 nonintrinsichotspots in FWR1, along with one at the edge of CDR2,are clustered in our model of VH5. These mutations mightincrease the affinity of the putative superantigen for VH5.Similar interactions involving the edge of the conventionalantigen-binding site and adjacent FWR residues have alsobeen observed in other superantigen-antibody complexes:

antibodies that recognize the carbohydrate I/i red bloodcell antigen are restricted to the VH4-34 gene, and residues23-25 in FWR1 are implicated in binding.25 In addition,the crystal structure of the complex between a rheumatoidfactor antibody and its autoantigen, IgG Fc, involvedresidues at the FWR1/CDR1 and FWR2/CDR2 bound-aries, enabling the CDR to remain available for conven-tional antigen binding.26,27

Remarkably, exactly the same features that we haveobserved in IgE VH5 in allergic rhinitis were previouslyreported in asthma.1-5 For example, VH5 represented 29%and 33% of the IgE sequences in this allergic rhinitis studyand a previous asthma study,2 respectively. Exactly thesame nonintrinsic hotspots, Lys 23(1), Gly 24(2), Thr30(2), and Tyr 52(2), were observed in asthma1 as in thisstudy of allergic rhinitis but differing from those in atopicdermatitis (Gly 35 and Thr 95).22 The similarities betweenallergic rhinitis and asthma might not be surprising, giventhat the upper and lower airways are physically contiguousand exposed to some of the same aeroallergens, but areimportant, given the differing susceptibility of someindividuals to asthma and others to rhinitis and thatmany individuals have both conditions. It might befortuitous that the same minor VH family appears to beoverexpressed in both allergic rhinitis–asthma and (at leastin one of the 2 studies) atopic dermatitis. The selection ofdifferent FWR mutations in the IgE VH5 in atopicdermatitis, however, might point to the action of differentsuperantigens.

What might be the putative B-cell superantigen asso-ciated with allergic rhinitis? About 37% of the populationcarries Staphylococcus aureus in the nasal mucosa.28

Certain S aureus enterotoxins are characterized as T-cellsuperantigens that stimulate B-cell proliferation, IgEsynthesis, or both.29-31 Some of these or others might actas B-cell superantigens. If a B-cell superantigen binds aswe suspect to FWR1 of VH5, an allergen might still bindto the CDR on the immunoglobulin. This could haveimportant consequences for the activation of B cellsexpressing VH5 and also the mast cells and dendritic cellsthat capture IgE VH5 in the tissue.

We thank Drs Rebecca Beavil, Pooja Takhar, Lyn Smurthwaite,

and Graham Dunn (King’s College London) for helpful discussions

and practical advice.

REFERENCES

1. Snow RE, Chapman CJ, Frew AJ, Holgate ST, Stevenson FK. Pattern of

usage and somatic hypermutation in the VH5 gene segments of a patient

with asthma: implications for IgE. Eur J Immunol 1997;27:162-70.

2. Snow RE, Djukanovic R, Stevenson FK. Analysis of immunoglobulin E

VH transcripts in a bronchial biopsy of an asthmatic patient confirms bias

towards VH5, and indicates local clonal expansion, somatic mutation and

isotype switch events. Immunology 1999;98:646-51.

3. Snow RE, Chapman, Frew AJ, Holgate ST, Stevenson FK. Analysis of

Ig VH region genes encoding IgE antibodies in splenic B lymphocytes of

a patient with asthma. J Immunol 1995;54:5576-81.

4. Snow RE, Chapman CJ, Holgate ST, Stevenson FK. Clonally related IgE

and IgG4 transcripts in blood lymphocytes of patients with asthma reveal

differing patterns of somatic mutation. Eur J Immunol 1998;26:3354-61.



Coker et al 451


5. Snow RE, Chapman CJ, Stevenson FK. Allergen recognition sites in

immunoglobulin E from patients with asthma. In: Holgate ST, Busse

WW, editors. Lung biology in health and disease. New York: Marcel

Dekker Inc; 1989. p. 323-41.

6. Cook GP, Tomlinson IM. The human immunoglobulin VH repertoire.


7. Chang B, Casali P. The CDR1 sequences of a major proportion of human

germline Ig VH genes are inherently susceptible to amino acid replace-

ment. Immunol Today 1994;15:367-73.

8. Brezinschek H-P, Brezinschek RI, Lipsky PE. Analysis of the heavy

chain repertoire of human peripheral B cells using single-cell polymerase

chain reaction. J Immunol 1995;155P:190-202.

9. Van der Stoep N, Van der Linden J, Logtenberg T. Molecular evolution

of the human immunoglobulin E response: high incidence of shared

mutations and clonal relatedness among e VH5 transcripts from three

unrelated patients with atopic dermatitis. J Exp Med 1993;177:99-107.

10. Edwards MR, Brouwer W, Choi CHY, Ruhno J, Ward RL, Collins AM.

Analysis of IgE antibodies from a patient with atopic dermatitis; biased V

gene usage and evidence for polyreactive IgE heavy chain complemen-

tarity-determining region 3. J Immunol 2002;168:6305-13.

11. Janezic A, Chapman CJ, Snow RE, Hourihane JO, Warner JO, Stevenson

FK. Immunogenetic analysis of heavy chain variable regions of IgE from

patients allergic to peanuts. J Allergy Clin Immunol 1998;101:391-6.

12. Silverman GJ, Goodyear CS. A model B-cell superantigen and the

immunobiology of B lymphocytes. Clin Immunol 2002;102:117-34.

13. Potter KN, Li Y-C, Capra JD. Staphylococcal protein A simultaneously

interacts with framework region 1, complementarity determining region

2, and framework region 3 on human VH3 encoded immunoglobulins.

J Immunol 1996;157:2982-8.

14. Berberian L, Goodglick L, Kipps TJ, Braun J. Immunoglobulin VH3 gene

products: natural ligands for HIV gp120. Science 1993;261:1588-91.

15. Graille ME, Stura A, Corper AL, Sutton BJ, Taussig MJ, Charbonnier

J-B, et al. Crystal structure of a Staphylococcus aureus protein A

domain complexed with the Fab fragment of a human IgM antibody:

structural basis for recognition of B-cell receptors and superantigen

activity. Proc Natl Acad Sci U S A 2000;97:5399-404.

16. Shlomchik MJ, Marshak-Rothstein A, Wolfowicz CB, Rothstein TL,

Weigert MG. The role of clonal selection and somatic mutation in

autoimmunity. Nature 1987;328:805-11.

17. Rogozin IB, Pavlov YI. Theoretical analysis of mutation hotspots and

their DNA sequence context specificity. Mut Res 2003;544:65-85.

18. Ellyard JI, Avery DT, Phan TG, Hare NJ, Hodgkin PD, Tangye SG.

Antigen-selected, immunoglobulin-secreting cells persist in human

spleen and bone marrow. Blood 2004;103:3805-12.

19. Coker HA, Durham SR, Gould HJ. Local somatic hypermutation and

class switch recombination in the nasal mucosa of allergic rhinitis

patients. J Immunol 2003;171:5602-10.

20. Durham SR, Ying S, Varney VA, Jacobson MR, Sudderick RM, Mackay

IS, et al. Cytokine messenger RNA expression for IL-3, IL-4, IL-5 and

granulocyte/macrophage-colony-stimulating factor in the nasal mucosa

after local allergen provocation: relationship to tissue eosinophilia.

J Immunol 1992;48:2390-4.

21. Brezinschek H-P, Dorner T, Monson NL, Brezinshek RI, Lipsky PE. The

influence of CD40-CD154 interactions on the expressed human VH

repertoire: analysis of VH genes expressed by individual B cells of a

patient with X-linked hyper-IgM syndrome. Int Immunol 2000;12:

767-75.

22. Betz AG, Neuberger MS, Milstein C. Discriminating intrinsic and

antigen-selected mutational hotspots in immunoglobulin V genes.


23. Randen I, Potter KN, Li Y, Thompson KM, Pascual V, Forre O, et al.

Complementarity-determining region 2 is implicated in the binding of

staphylococcal protein A to human immunoglobulin VHIII variable

regions. Eur J Immunol 1993;23:2682-6.

24. Chothia C, Lesk AM. Canonical structures for the hypervariable regions

of immunoglobulins. J Mol Biol 1987;196:901-17.

25. Potter KN, Hobby P, Klijn S, Stevenson FK, Sutton BJ. Evidence for

involvement of a hydrophobic patch in framework region 1 of human

V4-34-encoded immunoglobulins in recognition of the red cell I antigen.

J Immunol 2002;16:3777-82.

26. Corper AL, Sohi MK, Bonagura VR, Steinitz M, Jefferis R, Feinstein A,

et al. Structure of human IgM rheumatoid factor Fab bound to its

autoantigen IgG Fc reveals a novel topology of antibody-antigen

interaction. Nat Struct Biol 1997;4:374-81.

27. Sutton BJ, Corper AL, Bonagura V, Taussig MJ. The structure and origin

of rheumatoid factors. Immunol Today 2000;21:177-83.

28. Kluytmans J, van Belkum A, Verbrugh H. Nasal carriage of Staphylo-

coccus aureus: epidemiology, underlying mechanisms, and associated

risks. Clin Microbiol Rev 1997;10:505-20.

29. Hofer MF, Harbeck RJ, Schlievert PM, Leung DYM. Staphylococcal

toxins augment specific IgE responses by atopic patients exposed to

allergen. J Invest Dermatol 1999;112:171-6.

30. Jabara HH, Geha RS. The superantigen syndrome toxin-1 induces CD40

ligand expression and modulates IgE isotype switching. Int Immunol

1996;8:1503-10.

31. Bachert C, Van Zele T, Gavaert P, De Schrijver L, Van Cauwenberge P.

Superantigens in nasal polyps. Curr Allergy Asthma Rep 2003;3:

523-31.


AUGUST 2005

452 Coker et al

Basic

andclin

icalim

munology

Antibody responses against galactocerebrosideare potential stage-specific biomarkers inmultiple sclerosis

Til Menge, MD,a Patrice H. Lalive, MD,a Hans-Christian von Budingen, MD,a,b

Bruce Cree, MD, PhD,a Stephen L. Hauser, MD,a and Claude P. Genain, MDa

San Francisco, Calif, and Zurich, Switzerland

Background: Galactocerebroside, the major glycolipid of

central nervous system myelin, is a known target for

pathogenic demyelinating antibody responses in experimental

allergic encephalomyelitis (EAE), the animal model of multiple

sclerosis (MS).

Objective: To address the importance of anti-galactocerebroside

(a-GalC) antibodies in MS and to evaluate them as biomarkers

of disease.

Methods: a-GalC IgGs were quantified from sera of patients

with MS and in marmoset EAE by a new immunosorbent assay.

Results: We report a significant difference in serum a-GalC

IgG titers between patients with relapsing-remitting (RR)–MS

and healthy controls (HCs; P < .001). The frequencies of

a-GalC antibody-positive subjects (a-GalC titers $ mean

HC titers1 3 SD) are also significantly elevated in RR-MS

compared with HC (40% vs 0%; P = .0033). Immunoaffinity

purified a-GalC IgGs from human serum bind to cultured

human oligodendrocytes, indicating that the ELISA detects

a biologically relevant epitope. Corroborating these findings,

a-GalC antibody responses in marmoset EAE were similarly

found to be specifically associated with the RR forms and not

the peracute or progressive forms, in contrast with other

anti-myelin antibodies (P = .0256).

Conclusion: (1) a-GalC antibodies appear MS-specific and

are not found in healthy subjects, unlike antibodies against

myelin proteins; (2) when present, a-GalC antibodies identify

mostly RR-MS and may be an indicator of ongoing disease

activity. This novel assay is a suitable and valuable method to

increase accuracy of diagnosis and disease staging in MS.


Key words: Galactocerebroside, myelin antigens, autoantibody,multiple sclerosis, experimental allergic encephalomyelitis

Multiple sclerosis (MS) is a chronic immune-mediatedinflammatory demyelinating disease of the central nervoussystem (CNS) characterized by heterogeneity in clinicalpresentation and underlying pathological mechanisms.1

There is currently no easy paraclinical marker to diagnoseMS subtypes and predict disease course accurately with-out lengthy periods of clinical follow-up.

Several myelin autoantigens may serve as targets forthe autoaggressive attack in MS—for example, myelinprotein myelin/oligodendrocyte glycoprotein (MOG), ex-pressed on the outermost lamellae of the myelin sheathand thus readily accessible to the immune machinery;and a major CNS myelin glycolipid, galactocerebroside(GalC), which accounts for 32% of the myelin lipidcontent. Both MOG and galactocerebroside are highlyencephalogenic in various models of experimental auto-immune encephalomyelitis (EAE), the prototypic animalmodel for MS.2-4 Furthermore, passive antibody transfersin myelin basic protein (MBP)–primed animals5-9 andin vitro models have demonstrated the demyelinatingproperties of anti-galactocerebroside (a-GalC) anda-MOG antibodies.10-13 Antibody responses against thesemyelin targets are thus factors that potentially regulate

From athe Department of Neurology, University of California San Francisco;

and bthe Neurologische Klinik der Universitat Zurich.

Supported by grants from the National Institutes of Health (NS4678-01 to

Dr Genain and AI43073-11 to Dr Hauser), the National Multiple Sclerosis

Society (RG3370-A-3 and 3438-A-7 to Dr Genain), the CureMSNow fund,

the Lunardi Supermarkets, Inc, the Nancy Davis Center Without Walls, and

Aventis Pharmaceuticals. DrMenge and Dr Lalive are postdoctoral research

fellows of the National Multiple Sclerosis Society.

Disclosure of potential conflict of interest: T. Menge: named as inventor on

patent application ‘‘Methods to diagnose and prognose multiple sclerosis,’’

filed by University of California San Francisco, which includes data from

this work; received postdoctoral fellowship of the National Multiple

Sclerosis Society (FG 1476-A-1); employed by University of California

San Francisco. P. H. Lalive: named as inventor on patent application

‘‘Methods to diagnose and prognose multiple sclerosis,’’ filed by University

of California San Francisco, which includes data from this work; received

postdoctoral fellowship of the National Multiple Sclerosis Society (FG

1476-A-1); received grant/support from Swiss National Foundation

(PBGEB-102918); employed by University of California San Francisco.

H.-C. von Budingen: none disclosed. B. Cree: none disclosed. S. L. Hauser:

none disclosed. C. Genain: has done consulting work for Aventis

Pharmaceuticals; named as the main inventor on a patent application

‘‘Methods to diagnose and prognose multiple sclerosis,’’ filed by University

of California San Francisco, which includes data from this work; received

grants/support from National Institutes of Health (NS4678-01), National

Multiple Sclerosis Society (RG3370-A-3 and 3438-A-7); research contract

with Aventis Pharmaceuticals; donations from the Cure MS Now

Foundation and the Lunardi Supermarkets, Inc; employed by University

of California San Francisco; on the speakers’ bureau for Biogenidec, Teva

Pharmaceuticals, Serono, Inc.

Received for publication January 7, 2005; revised March 9, 2005; accepted for



Reprint requests: Claude Genain, MD, Department of Neurology,

Neuroimmunological Laboratories, C-440, University of California San

Francisco, 513 Parnassus Ave, San Francisco CA 94143-0114. E-mail:

[email protected].

0091-6749/$30.00


doi:10.1016/j.jaci.2005.03.023

453



Abbreviations usedAM: Acute monophasic

CIS: Clinically isolated syndrome

CNS: Central nervous system

EAE: Experimental allergic encephalomyelitis

GalC: Galactocerebroside

a-GalC: Anti-galactocerebroside

HC: Healthy control

HR: Hazard ratio

MBP: Myelin basic protein

MOG: Myelin/oligodendrocyte glycoprotein

MRI: Magnetic resonance imaging

MS: Multiple sclerosis

PP: Primary-progressive

rMOG: Recombinant rat myelin/oligodendrocyte

glycoprotein (extracellular domain)

RR: Relapsing-remitting

RT: Room temperature

SP: Secondary-progressive

disease phenotype expression in the context of establishedCNS inflammation.

The pathogenic involvement of anti-myelin antibodiesin human MS is less well established, because antibodytiters against the myelin proteins do not unequivocallydiffer between control populations and patients withMS.14-19 However, regardless of pathogenicity, anti-myelin antibodies have recently been proposed as pre-dictive disease markers.20

Here, we examined whether a-GalC antibodies couldserve as disease markers in MS. We demonstrate for thefirst time that significantly elevated titers of a-GalCantibodies are specifically found in relapsing-remitting(RR)–MS, and not in early or progressive forms of thedisease. In strong support of our clinical observations,longitudinal assessment of galactocerebroside reactivityduring the course of relapsing EAE inmarmosets indicatesthat appearance of antibodies against galactocerebroside isdelayed with respect to disease onset.

METHODS

Patients and controls

Sixty-five consecutive patients seen in our MS center, 51 meeting

the diagnostic criteria for clinically definite MS,21 were recruited for

this study: 20 with RR-MS, 15 secondary-progressive (SP)–MS, and

16 primary-progressive (PP)–MS (Table I). In addition, 14 patients

had a clinically isolated syndrome (CIS), ie, a single clinical attack

suggestive of CNS demyelination. Twenty volunteers served as

healthy controls (HCs). Both untreated patients and patients treated

with IFN-b and glatiramer acetate were included in this study, but

those treated with glucocorticoids within 3 months or on immuno-

suppressive therapy within 6 months of phlebotomy were excluded.

Blood was drawn by venipuncture and clotted serum stored at

240C. Informed consent was obtained from the patients and HCs,

and the study was conducted in accordance with Institutional Review

Board approval.

Animals

Callithrix jacchus marmosets were cared for in accordance with

the guidelines of the Institutional Animal Care andUsage Committee.

EAE was induced by immunization with 100 mg human white matter

homogenate as described.22 Plasma samples were obtained from

EDTA-anticoagulated blood at baseline and at intervals of 2 to 4

weeks and stored at240C. The animals were scored every other day

for the development of clinical signs and disability using a previously

published scale.22

a-GalC ELISA

Bovine brain–derived galactocerebroside (Matreya, Pleasant Gap,

Pa) was dissolved in chloroform-methanol (2:1). For coating,

galactocerebroside was air-dried, stepwise resuspended in 65Chot ethanol (50% vol/vol) at a final concentration of 50 ug/mL,

with 100 uL added to wells of Polysorb 96-well microtiter plates

(Nunc, Rochester, NY), and incubated uncovered overnight at room

temperature (RT) for solvent evaporation. Plates were washed with

double-distilled H2O and blocked with 1% BSA (A7030; Sigma, St

Louis, Mo) in PBS (ELISA buffer) for 2 hours at RT. After washing

with PBS and ddH2O, 100 uL of either human serum samples, diluted

1:40 in ELISA buffer, or C jacchus samples, diluted 1:100, were

incubated in triplicate overnight at 4C. Background binding of eachsample was controlled for on blocked wells without coated antigen.

After washing, specific antibody binding was detected by an alkaline

phosphate–labeled goat-anti-human IgG (A9544; Sigma) or by a

horseradish peroxidase–conjugated rabbit-anti-monkey IgG (A2054;

Sigma), diluted in ELISA buffer and incubated for 1 hour at RT. For

human sera, binding was detected by reading the OD at 405 nm in a

microplate reader (SpectraMax; Molecular Devices, Sunnyvale,

Calif) after incubation with paranitrophenyl phosphate (Moss,

Pasadena, Md) for 30 minutes in the dark at RT. The marmoset

assay was developed with 3,3#,5,5#-tetramethylbenzidine (Pierce,

Rockford, Ill) for 15 minutes at RT and the OD read at 450 nm

wavelength.

For specificity and sensitivity controls, a polyclonal rabbit-

anti-bovine galactocerebroside antiserum (G9152; Sigma) was used

and antibody binding detected by a horseradish peroxidase–labeled

goat-anti-rabbit IgG (A0545; Sigma). Quenching experiments were

performed by overnight pre-incubation with solubilized galactocer-

ebroside; galactocerebroside was air-dried and resuspended in 65Chot ethanol at 200 ug/mL and further diluted in ELISA buffer to a final

concentration of 2 ug/mL.

Anti-myelin protein antibody ELISA

C jacchus antibodies against human MBP and recombinant rat

(r)MOG, amino acids 1-12523 were coated to microtiter plates

(Maxisorb;Nunc) overnightwith 1 ug antigen perwell. After washing

and blocking with 3% BSA in PBS plus .05% Tween for 1 hour at

37C, marmoset samples were incubated for 1 hour at 37C and

diluted 1:100 in 3%BSA in PBS plus .05%Tween. Antibody binding

was detected by a peroxidase-labeled rabbit-anti-monkey IgG for

1 hour at 37C.


To express the results of the galactocerebroside assay, a signal-

to-background binding ratio was calculated as the ratio of OD

(signal) over OD (background). Positive controls, ie, a human sample

with strong binding signal, and negative controls, ie, ELISA buffer

only, omitting serum, were included on each plate. For human

samples, samples above the mean binding ratio 1 3 SD for the HC

group were considered positive. In the marmoset assay, samples were

considered positive for a binding ratio above 3 with ODGalC>0.1 and

greater than 3-fold the baseline (unimmunized) sample. Statistical


AUGUST 2005

454 Menge et al

Basic

andclin

icalim

munology

analysis was conducted by using STATA 7.0 (StataCorp LP, College

Station, Tex) and GraphPad Prism 3.0 (GraphPad Software, San

Diego, Calif). Categorical variables were compared by using the x2

test, continuous variables by using ANOVA, and ordinal variables by

using the Kruskal-Wallis test. The Bonferroni method and the Dunn

test were used to determine differences in between groups. Survival

analysis was used to assess time-dependent variables. Because the

binding ratios are not normally distributed, the binding ratio was

transformed by using an inverse ratio to generate a normal distribu-

tion for parametric analysis.

Antibody affinity purification

Human serumwas diluted in 10mmol/L sodium phosphate buffer,

pH 7.0 (SP buffer), and IgG was purified over a protein G column

(HiTrap HP; Amersham, Piscataway, NJ). Bound IgG was eluted

with 100 mmol/L glycine-HCl, pH 2.7, and dialyzed against the

sodium phosphate buffer. For immunoaffinity purification of a-GalC

antibodies, galactocerebroside was dissolved at 5.0 mg/mL in 65Chot methanol and hydrophobically bound to a FF-octyl column

(HiTrap; Amersham) as previously described.24 The IgG fraction was

applied to this column and bound IgG eluted and dialyzed into PBS as

described.

Immunohistochemistry

The human oligodendrocytoma cell line HOG (kind gift of

Dr Glyn Dawson), known to express galactocerebroside,25 was

grown in monolayers. Cells were trypsinized and plated at a density

of 20,000 cells/well onto chamber glass slides (Nunc); fixed in ice-

cold methanol; blocked with 2% BSA and 2% FBS in PBS; and

stained with human serum (1:50), rabbit antiserum (1:50), or 1006-

GalC (30 ug/mL), respectively, diluted in 1% BSA-PBS for 1 hour at

RT and developed with fluorescein isothiocyanate–labeled anti-IgG

secondary antibodies (F3512 for human, F9887 for rabbit; Sigma).

Control slides omitting the first antibodies were included.

RESULTS

Validation of the a-GalC assay

The assay was validated by a rabbit antiserum reactiveto bovine galactocerebroside, with reactivity detectable to

a titer of 1:12,800. Preincubation of the rabbit antiserumwith galactocerebroside solubilized in ELISA buffer(maximal solubility concentration, 2 ug/mL in aqueousbuffer) led to an 85% reduction in signal, proving spec-ificity of the assay. A mAb reactive against MOG (8.18-C5) did not react with the coated galactocerebroside,confirming the purity of the antigen. In serial dilutions oftotal IgG purified on protein G from either the humanpositive control or pooled immune C jacchus sera, thethreshold of detection was 6.25 ug IgG per well. Theinterplate and intraplate coefficients of variation were 15%and 4%, respectively.

Detection of a-GalC IgG in patients with MS

Quantitatively, significant differences in a-GalCantibody titers were found between HC and RR-MS(P < .001) as well as between patients with CIS andRR-MS (P < .05; ANOVAwith Bonferroni correction formultiple comparisons; Fig 1, A). There was a trendsuggesting a difference for the antibody titers betweenSP-MS and HC (P = .092). In contrast, there were nosignificant differences for a-GalC reactivity among theHC, CIS, and PP-MS subgroups. Even if the 2 patientswith the highest binding ratios in the RR-MS group wereexcluded from the calculations, the difference in thereciprocal binding ratio compared with the HC groupremained highly significant (P < .01).

The threshold for positivity was 3.23 and is indicated inFig 1, A (dashed line; see Methods). The frequencies ofpatients with RR-MS identified as a-GalC antibody–positive by this analysis were significantly higher com-pared with HC (40% vs 0%; P = .0033; Fisher exact test

TABLE I. Characteristics of patients with MS and HCs

Variable HC CIS RR-MS SP-MS PP-MS

N 20 14 20 15 16

Sex

Female 10 9 15 9 10

Male 10 5 5 6 6

Age (y)

Median 51.0 36.0* 43.0 45.0 51.0

Range 28-71 23-50 25-61 37-60 40-65

Disease duration

(mo)

Median NA NA 120 120 48

Range NA NA 9-266 24-384 9-216

Expanded Disability

Status Scale

Median NA 1.5 2.0 5.5 4.5

SD NA 0.9 1.5 1.5 1.2

NA, Not applicable.

*P < .05 if compared with HC and P < .01 if compared with PP-MS

(ANOVA with Bonferroni correction for multiple comparisons).

P < .001 and P < .01 if compared with SP-MS or PP-MS, respectively

(Kruskal-Wallis test with Dunn post hoc test for multiple comparisons).FIG 1. Binding ratios and frequencies of a-GalC IgG responses in

human MS and HCs. A, a-GalC IgG binding ratios for each disease

subgroup. Solid lines (—) denote mean binding ratios; dashed line

(--)denotes threshold of detection (mean binding ratio of HC 1 3 SD

(see B). B, Frequencies of a-GalC IgG seropositivity in human sera.



Menge et al 455


with Bonferroni correction for multiple independentcomparisons). Again, there was a trend observed fora-GalC antibody positivity in SP-MS compared with HC(26.7% vs 0%; P = .026; not significant after correctionfor multiple independent comparisons). Other pairwisecomparisons were not significant (Fig 1, B).

Immunoaffinity purification of a-GalC IgGand immunohistochemistry

To assess the specificity and biological relevance of theELISA assay, serum of 1 patient demonstrating a higha-GalC response (#1006) was subjected to immunoaffin-ity purification ofa-GalC IgG. From 50mL serum, 190 uga-GalC IgG (1006-GalC) was extracted by a custom-made galactocerebroside column. 1006-GalC reacted inthe ELISA with a detection limit 62.5 ng specific IgG perwell (0.625 ug/mL), and the signal could be quenched bysoluble galactocerebroside (45% signal reduction). Thesegalactocerebroside-purified IgGs showed staining of thehuman oligodendrocytoma cell line HOG identical to thecontrol rabbit a-GalC specific antiserum (Fig 2, A and B).These results unequivocally show that galactocerebrosidespecific antibodies purified from human serum are cell-surface binding on oligodendrocytes, and indicate that thenewly implemented ELISA assay system likely detectsbiologically relevant antibodies.

Detection of a-GalC IgG responses inmarmoset EAE

Sequential sera of 20 animals immunized with humanwhite matter homogenate were studied. Because of theoutbred nature of the animals, the clinical course of EAE isnot uniform: 9 animals displayed a RR-EAE diseasecourse, and 2 animals did not remit during attacks butprogressively worsened over time (similar to a PP course).Six animals were euthanized at onset of the first attack,termed acute monophasic (AM), and 2 of these had aperacute disease course rapidly progressing to a score of 4.An additional 3 animals were euthanized before the onsetof clinical disease, at the time when pleocytosis waspresent in the cerebrospinal fluid, demonstrating presenceof CNS inflammation. Clinical information is summarizedin Table II.

Antibodies against rMOG andMBPwere detected in allbut 1 of the animals regardless of their disease course,

including the preclinical animals (Table II). In contrast,a-GalC antibodies were detected only in animals withRR-EAE, and not during the first attack of AM-EAE, evenin the severely affected animals or in animals displaying aprogressive course (Table II). However, this could haveresulted from the overall shorter observation period forthese animals (median, 28 and 60 days postimmunizationvs 70 days postimmunization for RR-EAE; Table II).

The a-GalC antibody response appeared significantlylater compared with antibody responses against themyelinprotein rMOG and MBP in RR-EAE: median time lapsebetween immunization and appearance, 70 days fora-GalC vs 45 days for a-rMOG and 27 days for a-MBP(P = .0256; log rank test for equivalence of survivalfunctions). A Cox proportional hazard model showedthat the hazard ratios (HRs) for a-rMOG and a-MBPantibody responses were significantly different from theHR for a-GalC (HR a-rMOG = 5.56, P = .013; HRa-MBP = 12.76, P = .001; Fig 3), indicating that a-GalCantibodies occurred most distant from immunizations andthus onset of EAE in these animals.

DISCUSSION

We present here a reproducible solid-phase assay fordetection of galactocerebroside-specific IgG in humansera. These specific IgG were purified by means of agalactocerebroside immunoaffinity chromatography col-umn and were shown to retain the ability to bind to agalactocerebroside epitope expressed on human oligoden-drocytes and in vitro by ELISA. The assays previouslydescribed to measure such antibodies in MS18,19,26 iden-tified differences between controls and MS for cerebro-spinal fluid, but not serum, even with undiluted serum in asolid-phase radioimmunoassay.18 The most likely expla-nation for the differences we find between HC and MSis the stratification for MS subgroups, which was notexamined in previous studies.18,19 Indeed, comparing allour 65 patients with MS as 1 group with HC showed nosignificant difference in the frequency of antibody-posi-tive patients.

The current a-GalC IgG assay is performed in serum atdilutions of 1:40 and above, which is considerably easierto access than cerebrospinal fluid and can be repeated

FIG 2. Immunostaining of HOG cells with affinity-purified human a-GalC IgG. A, Affinity purified a-GalC IgG

(1006-GalC) at 30 ug/mL. B, Positive control (rabbit a-GalC antiserum) at 1:50 dilution. C, Staining with serum

of 1006-GalC at dilution 1:50. D and E, Negative controls: fluorescein isothiocyanate–labeled anti-human and

anti-rabbit IgG.


AUGUST 2005

456 Menge et al

Basic

andclin

icalim

munology

multiple times. Most significant, serum a-GalC are spe-cific for MS, because they are not encountered in any ofthe controls, and practically never if at all in CIS. Althoughother neurological diseases were not examined, thisfinding at least indicates that, unlike for myelin proteinslike MOG, serum positivity helps to distinguish patientswith MS from healthy individuals. The intergroup differ-ences are very significant, despite the relatively smallnumber of subjects studied. The 65 patients were chosenrandomly in consecutive order of presentation, anda-GalC measurements were performed in a blind fashion.In addition, we could rule out any confounding variablefor age, sex, or disease duration.

These observations imply that a-GalC antibodies canhelp stratify different MS subgroups, namely RR-MS, anovel finding with high clinical relevance. Patients withCIS by definition have had 1 apparent clinical attack,whereas patients with RR-MS are characterized by diseasedissemination in time and space. A high proportion of CISwho present with brain magnetic resonance imaging(MRI) abnormalities will proceed to develop RR-MS,27,28 and indeed, for many of those patients, subclin-ical MS or minor attacks may have been present for aconsiderable period. Thus it can be envisioned thatdetection of a-GalC antibodies may permit staging ofMS forms according to time from the first demyelinatingevent. Because these antibodies appear to be characteristic

of established MS, their detection in patients with earlyMS and CIS could potentially help correct and achieve anearlier diagnosis of definite MS than with conventionalcriteria. The anti-myelin protein antibodies, on the otherhand, have recently been described as potential predictorsof early conversion in patients with CIS.20

Critical for interpretation of our clinical findings in theabsence of longitudinal measurements in human MS was

FIG 3. Time course of a-GalC and a-myelin protein IgG responses in

immunized C jacchus. Serum dilutions, 1:100. (.) denotes onset of

clinical signs; (-;-) denotes a-MBP positivity; (-*-) denotes a-rMOG

positivity; (-n-) denotes a-GalC positivity. Significant levels for

median onset postimmunization (pi) of antibody positivity were

determined by a Cox proportional hazard model.

TABLE II. Characteristics of C jacchus marmoset EAE and antibody status

Animal

Disease

course

Clinical onset

(day PI)

Sacrifice

(day PI)y

Maximal clinical

score (day PI)y

a-GalC

IgG

First day

detected (PI)za-rMOG

IgG

First day

detected (PI)za-MBP

IgG

First day

detected (PI)z

326-91 RR 14 120 4 (120) 1 23 1 63 1 23

106-90 RR 16 112 3 (96) 1 80 1 53 1 25

191-92 RR 16 112 2 (20) 1 81* 1 53* 1 27*

378-85 RR 16 112 2 (40) 1 26* 1 26* 1 56*

U062-02 RR 43 97 3 (84) 1 96 1 36 1 36

185-99 RR 7 86 2 (56) 1 70 1 15 1 28

U050-01 RR 21 86 2 (57) 1 70 1 45 1 15

U057-02 RR 32 82 1.5 (73) 1 62 1 62 1 18

U050-00 RR 21 78 3 (38) 1 78 1 29 1 29

Median 6 SD RR 21 6 11 98 6 16 2.0 6 0.8

(97 6 16)

70 6 25 42 6 17 29 6 12

127-93 CP 16 68 4 (56) 2 2 1 54* 1 26*

U052-01 CP 21 52 3 (40) 2 2 1 36 1 18

U025-00 AM 21 61 2 (57) 2 2 1 42 1 28

U023-00 AM 16 31 2 (28) 2 2 1 28 1 28

273-93 AM 12 28 3.5 (24) 2 2 1 28* 1 28*

274-93 AM 16 28 2 (20) 2 2 1 30* 1 30*

U021-99 AM 18 23 1.5 (23) 2 2 1 20 1 20

346-92 AM 17 22 3 (22) 2 2 1 23 1 23

Median 6 SD AM 16 6 3 28 6 15 2.0 6 0.8

(23 6 14)

28 6 8 28 6 4

U053-01 Preclinical NA 38 0 2 2 1 35 1 35

U061-02 Preclinical NA 31 0 2 2 2 2 1 31

U030-00 Preclinical NA 23 0 2 2 1 13 1 20

NA, Not applicable; PI, postimmunization.

*Monthly blood draw only.

P < .001 for timing of euthanasia and maximal clinical scores between RR-EAE vs AM, respectively (2-tailed t test); maximal clinical score or

time of clinical onset were not significantly different (P = .69 and .39, respectively; 2-tailed t test)

Statistical analysis provided in Fig 3.



Menge et al 457


the study of chronic relapsing marmoset EAE, which bestapproximates MS complex pathophysiology. Specifically,we found that the a-GalC responses occurred distinctlyafter disease onset only in animals with RR forms of EAE(Table II; Fig 3). This was in contrast to the a-rMOG anda-MBP responses that occurred in all animals tested, insome cases even before clinical onset. These findings arein line with results from 2 other EAEmodels,29,30 in whicha-GalC antibodies were also present in the early chronicstage of guinea pig EAE29 and occurred after the clinicalonset and after the development of a-MBP antibodies.30

Reactivity against rMOG was not tested in either of thesestudies.

Although the pathophysiological explanation for thedelayed antibody response to galactocerebroside in MSand EAE is not known, several mechanisms may bepostulated. First, glycolipids are not classic, MHC-restricted T-cell antigens but may elicit a TH1 response viaCD1 presentation.31,32 CD1 expression has been demon-strated on astrocytes within MS lesions.33 Glycolipidantigens may be presented to T cells only once detachedfrom the membrane bilayer, yet the degradation of myelinglycolipids by macrophages takes considerably longerthan the breakdown of myelin proteins.34 Second, lipids assuch may be haptens and have to be attached to carrierproteins to elicit an immune response.4 It has been pro-posed that MOGmay serve as a carrier protein interactingwith galactocerebroside within the cell membrane.35

These possibilities all may also explain the low titers ofa-GalC antibodies, which are considerably lower inhuman beings compared with titers of antibodies againstmyelin proteins (Dr Menge, personal observation, ref 15),as in EAE models.29

Antibodies reactive against galactocerebroside mayhave demyelinating properties, at least experimentallyin vitro10,12,13 and in vivo.6,9,36 Although our study did notaim at proving any functional disability associated withthe presence of a-GalC antibodies in human beings, it isinteresting to note that 40% of RR-MS cases studied havedetectable a-GalC reactivity, which could potentially beindicative of a particular RR-MS group in terms of diseasecourse and severity. In addition, a lesser proportion ofpatients with SP-MS than with RR-MS appear to bea-GalC antibody–positive. This could mean that a-GalCautoantibodies predominate during a yet to be definedwindow of time that overlaps between RR-MS andSP-MS, with a tendency to decrease during the neurode-generative stage of SP-MS. Future studies with largernumbers of subjects and longitudinal measurements areneeded to address whether these antibody responses areassociated with clinical (ExpandedDisability Status Scale,progression rate, treatment response) or paraclinical(magnetic resonance imaging burden of lesion) parame-ters, and to establish their prognostic significance.

In conclusion, we have demonstrated that a-GalCantibodies are a predominant phenomenon of RR-MS,and that in a primate disease model, the a-GalC responseoccurs significantly later than a-myelin protein responses.This galactocerebroside assay is available as a paraclinical

investigation, in combination with MRI. In line with otherrecent reports on humoral immunity in MS and EAE,20,37

these novel findings continue to underscore the value ofa-myelin antibody assessment—both protein and glyco-lipid—as biomarkers that will be used in the near future forMS diagnostics, staging, and prognosis.

We thank Salomon Martinez for expert animal work; Jerry

Hernandez, Drew Dover, and Kevin Morgan for help analyzing the

samples, and the clinical coordinators and neurologists at the

University of California San Francisco MS Center for sample

collection.

REFERENCES

1. Steinman L. Multiple sclerosis: a two-stage disease. Nat Immunol 2001;

2:762-4.

2. von Budingen HC, Villoslada P, Ouallet JC, Hauser SL, Genain CP.

Immune responses against the myelin/oligodendrocyte glycoprotein in

experimental autoimmune demyelination. J Clin Immunol 2001;21:

155-70.

3. Moore GR, Traugott U, Farooq M, Norton WT, Raine CS. Experimental

autoimmune encephalomyelitis: augmentation of demyelination by dif-

ferent myelin lipids. Lab Invest 1984;51:416-24.

4. Raine CS, Traugott U, Farooq M, Bornstein MB, Norton WT. Aug-

mentation of immune-mediated demyelination by lipid haptens. Lab

Invest 1981;45:174-82.

5. Genain CP, Nguyen MH, Letvin NL, Pearl R, Davis RL, Adelman M,

et al. Antibody facilitation of multiple sclerosis-like lesions in a

nonhuman primate. J Clin Invest 1995;96:2966-74.

6. Fierz W, Heininger K, Schaefer B, Toyka KV, Linington C, Lassmann

H. Synergism in the pathogenesis of EAE induced by an MBP-specific

T-cell line and monoclonal antibodies to galactocerebroside or a myelin

oligodendroglial glycoprotein. Ann N Y Acad Sci 1988;540:360-3.

7. Schluesener HJ, Sobel RA, Linington C, Weiner HL. A monoclonal

antibody against a myelin oligodendrocyte glycoprotein induces relapses

and demyelination in central nervous system autoimmune disease.

J Immunol 1987;139:4016-21.

8. Linington C, Bradl M, Lassmann H, Brunner C, Vass K. Augmentation

of demyelination in rat acute allergic encephalomyelitis by circulating

mouse monoclonal antibodies directed against a myelin/oligodendrocyte

glycoprotein. Am J Pathol 1988;130:443-54.

9. Morris-Downes MM, Smith PA, Rundle JL, Piddlesden SJ, Baker D,

Pham-Dinh D, et al. Pathological and regulatory effects of anti-myelin

antibodies in experimental allergic encephalomyelitis in mice. J Neuro-

immunol 2002;125:114-24.

10. Fry JM, Weissbarth S, Lehrer GM, Bornstein MB. Cerebroside antibody

inhibits sulfatide synthesis and myelination and demyelinates in cord

tissue cultures. Science 1974;183:540-2.

11. Raine CS, Johnson AB, Marcus DM, Suzuki A, Bornstein MB.

Demyelination in vitro: absorption studies demonstrate that galactocer-

ebroside is a major target. J Neurol Sci 1981;52:117-31.

12. Saida T, Saida K, Silberberg DH. Demyelination produced by experimen-

tal allergic neuritis serum and anti-galactocerebroside antiserum in CNS

cultures: an ultrastructural study. Acta Neuropathol (Berl) 1979;48:19-25.

13. Menon KK, Piddlesden SJ, Bernard CC. Demyelinating antibodies to

myelin oligodendrocyte glycoprotein and galactocerebroside induce

degradation of myelin basic protein in isolated human myelin.

J Neurochem 1997;69:214-22.

14. Lindert RB, Haase CG, Brehm U, Linington C, Wekerle H, Hohlfeld R.

Multiple sclerosis: B- and T-cell responses to the extracellular domain of

the myelin oligodendrocyte glycoprotein. Brain 1999;122:2089-100.

15. Reindl M, Linington C, Brehm U, Egg R, Dilitz E, Deisenhammer F,

et al. Antibodies against the myelin oligodendrocyte glycoprotein and the

myelin basic protein in multiple sclerosis and other neurological diseases:

a comparative study. Brain 1999;122:2047-56.

16. Lampasona V, Franciotta D, Furlan R, Zanaboni S, Fazio R, Bonifacio E,

et al. Similar low frequency of anti-MOG IgG and IgM in MS patients

and healthy subjects. Neurology 2004;62:2092-4.


AUGUST 2005

458 Menge et al

Basic

andclin

icalim

munology

17. Mantegazza R, Cristaldini P, Bernasconi P, Baggi F, Pedotti R, Piccini I,

et al. Anti-MOG autoantibodies in Italian multiple sclerosis patients:

specificity, sensitivity and clinical association. Int Immunol 2004;16:

559-65.

18. Rostami AM, Burns JB, Eccleston PA, Manning MC, Lisak RP,

Silberberg DH. Search for antibodies to galactocerebroside in the serum

and cerebrospinal fluid in human demyelinating disorders. Ann Neurol

1987;22:381-3.

19. Kasai N, Pachner AR, Yu RK. Anti-glycolipid antibodies and their

immune complexes in multiple sclerosis. J Neurol Sci 1986;75:33-42.

20. Berger T, Rubner P, Schautzer F, Egg R, Ulmer H, Mayringer I, et al.

Antimyelin antibodies as a predictor of clinically definite multiple sclerosis

after a first demyelinating event. N Engl J Med 2003;349:139-45.

21. Poser CM, Paty DW, Scheinberg L, McDonald WI, Davis FA, Ebers GC,

et al. New diagnostic criteria for multiple sclerosis: guidelines for

research protocols. Ann Neurol 1983;13:227-31.

22. Massacesi L, Genain CP, Lee-Parritz D, Letvin NL, Canfield D, Hauser

SL. Active and passively induced experimental autoimmune encephalo-

myelitis in common marmosets: a new model for multiple sclerosis. Ann

Neurol 1995;37:519-30.

23. Amor S, Groome N, Linington C, Morris MM, Dornmair K, Gardinier

MV, et al. Identification of epitopes of myelin oligodendrocyte glyco-

protein for the induction of experimental allergic encephalomyelitis in

SJL and Biozzi AB/H mice. J Immunol 1994;153:4349-56.

24. Nakajima H, Katagiri YU, Kiyokawa N, Taguchi T, Suzuki T, Sekino T,

et al. Single-step method for purification of Shiga toxin-1 B subunit using

receptor-mediated affinity chromatography by globotriaosylceramide-

conjugated octyl sepharose CL-4B. Protein Expr Purif 2001;22:267-75.

25. Lily O, Palace J, Vincent A. Serum autoantibodies to cell surface

determinants in multiple sclerosis: a flow cytometric study. Brain 2004;

127:269-79.

26. Ichioka T, Uobe K, Stoskopf M, Kishimoto Y, Tennekoon G,

Tourtellotte WW. Anti-galactocerebroside antibodies in human cerebro-

spinal fluids determined by enzyme-linked immunosorbent assay

(ELISA). Neurochem Res 1988;13:203-7.

27. Frohman EM, Goodin DS, Calabresi PA, Corboy JR, Coyle PK, Filippi

M, et al. The utility of MRI in suspected MS: report of the Therapeutics

and Technology Assessment Subcommittee of the American Academy of

Neurology. Neurology 2003;61:602-11.

28. Brex PA, Ciccarelli O, O’Riordan JI, Sailer M, Thompson AJ, Miller

DH. A longitudinal study of abnormalities on MRI and disability from

multiple sclerosis. N Engl J Med 2002;346:158-64.

29. Tabira T, Endoh M. Humoral immune responses to myelin basic

protein, cerebroside and ganglioside in chronic relapsing experimental

allergic encephalomyelitis of the guinea pig. J Neurol Sci 1985;67:

201-12.

30. Lolli F, Liuzzi GM, Vergelli M, Massacesi L, Ballerini C, Amaducci L,

et al. Antibodies specific for the lipid-bound form of myelin basic protein

during experimental autoimmune encephalomyelitis. J Neuroimmunol

1993;44:69-75.

31. Beckman EM, Porcelli SA, Morita CT, Behar SM, Furlong ST, Brenner

MB. Recognition of a lipid antigen by CD1-restricted alpha beta1 T

cells. Nature 1994;372:691-4.

32. Shamshiev A, Donda A, Carena I, Mori L, Kappos L, De Libero G.

Self glycolipids as T-cell autoantigens. Eur J Immunol 1999;29:

1667-75.

33. Battistini L, Fischer FR, Raine CS, Brosnan CF. CD1b is expressed in

multiple sclerosis lesions. J Neuroimmunol 1996;67:145-51.

34. Lumsden CE. The neuropathology of multiple sclerosis. In: Vinken PI,

Bruyn GW, editors. Handbook of clinical neurology. New York:

Elsevier; 1970. p. 217-309.

35. Bernard CC, Johns TG, Slavin A, Ichikawa M, Ewing C, Liu J, et al.

Myelin oligodendrocyte glycoprotein: a novel candidate autoantigen in

multiple sclerosis. J Mol Med 1997;75:77-88.

36. Rosenbluth J, Schiff R, Liang WL, Dou WK, Moon D. Antibody-

mediated CNS demyelination: focal spinal cord lesions induced by

implantation of an IgM anti-galactocerebroside-secreting hybridoma.

J Neurocytol 1999;28:397-416.

37. Robinson WH, Steinman L, Utz PJ. Protein arrays for autoantibody

profiling and fine-specificity mapping. Proteomics 2003;3:2077-84.



Menge et al 459


Letters to the Editor

Perilesional GM-CSF therapy of a chronic legulcer in a patient with common variableimmunodeficiency

To the Editor:Impaired wound healing characterizes multiple immu-

nodeficiency states, including common variable immuno-deficiency (CVID). Chronic, nonhealing wounds oftenensue, with a considerable associated toll of pain, disfig-urement, disability, and increased medical costs. Thesechronic wounds are notoriously difficult to treat, oftenprompting debridement and skin grafting.

Here we report a patient with CVID with chronic legulcers who responded to perilesional GM-CSF therapy.

The patient was a 68-year-old white man with CVIDwho had been on intravenous immunoglobulin replace-ment for 15 years. He also had diabetes mellitus (type 2),was on chronic steroid therapy for several years (forsteroid-dependent asthma), and more recently was onmethotrexate for chronic inflammatory myositis. In May2002, he developed a right leg ulcer that did not heal andrequired skin grafting.

In January 2004, he developed another ulcer over hisright leg, after minor trauma, in an area distinct from theprevious ulcer. The wound continued to progress despitedebridement and local care, reaching a maximum sizeof 7 cm 3 5 cm (Fig 1, A). Bacterial culture from theulcer (March 2004) grew Pseudomonas aeruginosa andmethicillin-resistant Staphylococcus aureus. Standardtherapy, including local care and topical and systemic anti-biotics, was ineffective. GM-CSF treatment (sargramostim,Leukine; Immunex Corp, Seattle, Wash) was begun inMarch 2004. He received 125 mg in 0.25 mL subcutane-ously at each of 4 sites distributed evenly around thehealthy edge of the ulcer (3, 6, 9 and 12 o’clock). This wasrepeated weekly for 4 weeks.

The ulcer started improving a few days after the firstinjection. After 4 weeks, the patient was instructed to washthe ulcer with GM-CSF (3 mL saline containing 15 mgGM-CSF) twice daily for another 4 weeks (Fig 1, B). Theulcer healed completely in September 2004 (Fig 1,C). Thepatient reported no side effects fromGM-CSF treatment. Itis worth noting that his methotrexate dose was actuallyincreased during this period, thus ruling out the possibilitythat the accelerated wound healing was a result of removalof immunosuppression.

Normal wound healing can be divided into 3 differentphases: inflammation, proliferation, and maturation.1

Inflammation is characterized by an influx of monocytesand neutrophils that remove the debris. This is followed bythe proliferation of epithelial cells, fibroblasts, and endo-thelial cells. The resultant effect is to lay down connectivetissue and restore tissue architecture. During maturation,the wound contracts and gains tensile strength.

GM-CSF has been shown to influence all phases ofwound healing. In addition to its well characterized effectson neutrophil and monocyte proliferation, migration, and

phagocytosis, GM-CSF stimulates the proliferation ofkeratinocytes and the differentiation of myofibroblasts.1-3

GM-CSF also increases wound tensile strength.4 The roleof bacterial infections in chronic wounds is not wellestablished. However, GM-CSF might prove useful inchronically infected wounds. Previous studies have dem-onstrated that GM-CSF enhances phagocyte killing ofseveral microorganisms.5,6 In addition, GM-CSF wasshown in animal studies to promote the healing of infectedwounds.7 Interestingly, GM-CSF appears to be effectivein chronic venous stasis ulcers, which one would notexpect to be associated with decreased immune defense.Da Costa et al8 administered perilesional placebo or GM-CSF in 200-mg or 400-mg doses. The healing rates were57%, 61%, and 19% for the low-dose GM-CSF, high-doseGM-CSF, and placebo, respectively. GM-CSF was alsoused in a patient with CVID with bilateral leg ulcerscaused by dermatofibromas, with a dramatic response.9

Our patient illustrates that in addition to venous stasisulcers and dermatofibroma-associated ulcers, perilesionalGM-CSF can be very effective in the treatment of infectedleg ulcers in patients with CVID.

Ammar Z. Hatab, MDa

Deanna McDanel, PharmD, BCPSb

Zuhair K. Ballas, MDa,c

aDivision of Allergy/Immunology

Department of Internal MedicineC42/E-13, GH

Carver College of Medicine

University of IowabDepartment of Pharmaceutical Care

University of Iowa Hospitals and Clinics and the

College of Pharmacy

University of IowacIowa City VA Medical Center

Iowa City, Iowa

REFERENCES

1. Groves RW, Schmidt-Lucke JA. Recombinant human GM-CSF in the treat-

ment of poorly healing wounds. Adv Skin Wound Care 2000;13:107-12.

2. Hancock GE, Kaplan G, Cohn ZA. Keratinocyte growth regulation by the

products of immune cells. J Exp Med 1988;168:1395-402.

3. Gabbiani G. Modulation of fibroblastic cytoskeletal features during wound

healing and fibrosis. Pathol Res Pract 1994;190:851-3.

4. Jyung RW, Wu L, Pierce GF, Mustoe TA. Granulocyte-macrophage

colony-stimulating factor and granulocyte colony-stimulating factor: dif-

ferential action on incisional wound healing. Surgery 1994;115:325-34.

5. Reed SG, Nathan CF, Pihl DL, Rodricks P, Shanebeck K, Conlon PJ, et al.

Recombinant granulocyte/macrophage colony-stimulating factor activates

macrophages to inhibit Trypanosoma cruzi and release hydrogen peroxide:

comparison with interferon gamma. J Exp Med 1987;166:1734-46.

6. Weiser WY, Van Niel A, Clark SC, David JR, Remold HG. Recombinant

human granulocyte/macrophage colony-stimulating factor activates intra-

cellular killing of Leishmania donovani by human monocyte-derived

macrophages. J Exp Med 1987;166:1436-46.

7. Robson M, Kucukcelebi A, Carp SS, Hayward PG, Hui PS, Cowan WT,

et al. Effects of granulocyte-macrophage colony-stimulating factor on

wound contraction. Eur J ClinMicrobiol Infect Dis 1994;13(suppl 2):S41-6.

8. Da Costa RM, Ribeiro Jesus FM, Aniceto C, Mendes M. Randomized,

double-blind, placebo-controlled, dose-ranging study of granulocyte-

macrophage colony stimulating factor in patients with chronic venous

leg ulcers. Wound Repair Regen 1999;7:17-25.

460

Lette

rsto

theEd

itor

FIG 1. A, Leg ulcer at initiation of GM-CSF therapy. B, Marked improvement after 4 weeks of GM-CSF.

C, Complete healing after completion of therapy.



Letters to the Editor 461

Letters

toth

eEd

itor

9. Siddiqui FH, Biundo JJ Jr, Moore C, Ermitano ML, Ortigas AP,

DeFrancesch F. Recombinant granulocyte macrophage colony stimulating

factor (rhu-GM-CSF) in the treatment of extensive leg ulcers: a case

report. Surgery 2000;127:589-92.

Available online May 24, 2005.doi:10.1016/j.jaci.2005.04.008

Asthma caused by cyanoacrylate used in aleisure activity

To the Editor:Acrylic compounds (acrylates, methacrylates, and

cyanoacrylates) are volatile and chemically reactive agentsused extensively in the manufacture of such productsas adhesives, resins, solvents, and glues and in the healthprofession (dental prostheses and bone cement in ortho-pedics).1 These agents are well known to cause occu-pational asthma,2,3 as well as skin sensitization andirritation.4 Although acrylate glues are widely used inseveral activities of daily life, to our knowledge, there hasbeen only one case reported of their causing respiratorysymptoms out of the workplace.5

We report bronchial asthma caused by cyanoacrylatein a 55-year-old man whose hobby was making miniatureplanes, an activity that required the use of a cyanoacrylateadhesive paste. This exsmoker had never experiencedasthmatic or rhinitis symptoms. A year before being seenat the clinic, he reported acute dyspnea during a weekend,which is when he normally worked on his model planes;this episode required emergency care, followed by a shortcourse of oral and inhaled corticosteroid therapy. Afterthis occasion, he stopped practicing his hobby and did notrequire medication, except for short-acting bronchodilatoroccasionally when his respiratory symptoms were exac-erbated by physical exercise, cold temperature, and heavysmells.

The results of skin prick tests to common aeroallergenswere negative; spirometry showed an FEV1 of 2.9 L(100% of predicted value), a forced vital capacity of 3.5 L(100% of predicted value), and an FEV1/forced vitalcapacity ratio of 83% (normal). Methacholine bronchialresponsiveness was normal (PC20 5 128 mg/mL; normalvalue >16 mg/mL in our laboratory). The subject under-went a specific inhalation challenge (SIC) according toa standardized procedure.6 Results are shown in Fig 1. Ona control day, the patient was exposed to diluent paintby means of nebulization for 30 minutes. Spirometry,methacholine testing, and induced sputum performed afterdiluent exposure produced normal results. On 2 subsequentdays, exposure to cyanoacrylate was carried out by askingthe patient to mimic his leisure activity in a challengeroom, spreading cyanoacrylate glue on a piece of card-board for progressively longer periods of time (totals of4 and 30 minutes of exposure on the 2 days). The testrevealed a typical early late response. Induced sputumperformed before and after SIC demonstrated pronouncedeosinophlia after the cyanoacrylate challenge: eosinophilcounts switched from 0.5% before SIC to 63% at the endof the last day of exposure.

Occupational asthma caused by acrylates has beendescribed often, but in this case exposure took place onlyon occasion during a leisure activity; it is relevant to high-light that in this nonatopic subject specific reactivity to thesensitizer persisted even 1 year after cessation of exposure.This case underlines the sensitization strength of acrylates,proved as well by the fact that very intermittent exposurewas enough to trigger bronchial asthma. This can alsoreasonably explain why the subject was cured afterstopping exposure.

Dr Mona-Rita Yacoub is a postdoctoral fellow supported by

Asthma in the Workplace (Canadian Institutes of Health Research,

Canadian Lung Association, Institut de recherche Robert-Sauve en

sante et securite du travail du Quebec).

We thank L. Schubert for reviewing the manuscript.

Mona-Rita Yacoub, MDCatherine Lemiere, MD, MSc

Jean-Luc Malo, MD

Department of Chest MedicineSacre-Coeur Hospital

5400 West Gouin Blvd

Montreal, Quebec, Canada H4J 1C5

REFERENCES

1. Piirila P, Kanerva L, Keskinen H, Estlander T, Hytonen M, Tuppurainen M.

Occupational respiratory hypersensitivity caused by preparations contain-

ing acrylates in dental personnel. Clin Exp Allergy 1998;28:1404-11.

2. Quirce S, Baeza ML, Tornero P, Blasco A, Barranco R, Sastre J. Occupa-

tional asthma caused by exposure to cyanoacrylate. Allergy 2001;56:446-9.

3. Weytjens K, Cartier A, Lemiere C, Malo JL. Occupational asthma to

diacrylate. Allergy 1999;54:289-90.

4. Kopferschmit-Kubler MC, Stenger R, Blaumeiser M, Eveilleau C,

Bessot JC, Pauli G. Asthma, rhinitis and urticaria following occupational

exposure to cyanoacrylate glues. Rev Mal Respir 1996;13:305-7.

5. Kopp SK, McKay RT, Moller DR, Cassedy K, Brooks SM. Asthma and

rhinitis due to ethylcyanoacrylate instant glue. Ann Intern Med 1985;102:

613-5.

6. Cartier A, Bernstein IL, Burge PS, Cohn JR, Fabbri LM, Hargreave FE,

et al. Guidelines for bronchoprovocation on the investigation of occupa-

tional asthma. J Allergy Clin Immunol 1989;84(suppl):823-9.

Available online June 1, 2005.doi:10.1016/j.jaci.2005.04.015

FIG 1. Results of SIC.


AUGUST 2005

462 Letters to the Editor

Lette

rsto

theEd

itor

Correspondence

Cystic fibrosis gene mutations andchronic rhinosinusitis

To the Editor:In their review of the definition, pathophysiology,

treatment of rhinosinusitis by Meltzer et al,1 there is nodiscussion of the association between mutations in thecystic fibrosis transmembrane conductance regulator(CFTR) gene and chronic rhinosinusitis.2 Most of thesepatients do not meet diagnostic criteria for cystic fibrosisbecause their sweat chloride is not elevated above 60mmol/L and they do not have evidence of 2 disease-causing CFTR gene mutations. However, some of thepatients with CFTR gene mutations and chronic rhinosi-nusitis reported by Wang et al2 did have other features ofcystic fibrosis, such as infertility and infection withPseudomonas aeruginosa. Identification of CFTR genemutations in patient with chronic rhinosinusitis maypotentially be of clinical importance in the future, becausepharmacologic therapies to overcome defective CFTRfunction are under development.3

Clement L. Ren, MD

Division of Pediatric Pulmonology and Allergy

Room 4-3236601 Elmwood Avenue

University of Rochester

Rochester, NY 14642

Editor’s note: This Correspondence has no accompanying reply. The

authors of the Meltzer article chose not to reply, saying that the

correspondence provides interesting information and is worthy of

publication.

REFERENCES

1. Meltzer EO, Hamilos DL, Hadley JA, Lanza DC, Marple BF, Nicklas RA.

Rhinosinusitis: establishing definitions for clinical research and patient

care. J Allergy Clin Immunol 2004;114:S156-212.

2. Wang X, Moylan B, Leopold DA, Kim J, Rubenstein RC, Togias A,

et al. Mutation in the gene responsible for cystic fibrosis and predis-

position to chronic rhinosinusitis in the general population. JAMA 2000;

284:1814-9.

3. Zeitlin PL. Novel pharmacologic therapies for cystic fibrosis. J Clin Invest

1999;103:447-52.


Leukotriene receptor antagonists are notas effective as intranasal corticosteroidsfor managing nighttime symptoms ofallergic rhinitis

To the Editor:I wish to comment on the supplement published in the

November 2004 issue titled ‘‘Allergic Rhinitis AfterHours: The Relevance and Consequences of NighttimeSymptoms.’’1 Nasal congestion associated with allergicrhinitis was identified as an important risk factor forsleep-disordered breathing, sleep fragmentation, daytime

somnolence, and fatigue, and it was noted that nasalcongestion and other rhinitis symptoms follow a circadianrhythm, being more severe at night and early in themorning. Chronotherapy, the timed dosing of rhinitismedications to manage optimally the diurnal variation innasal congestion and other rhinitis symptoms, was dis-cussed, as well as the advantages and disadvantages ofeach medication class in relation to nighttime rhinitissymptoms; however, readers must draw their own con-clusions regarding the most effective medication class.

Three large, double-blind, randomized, placebo-con-trolled trials of montelukast 10 mg (MON), the onlyleukotriene receptor antagonist approved in the UnitedStates for seasonal allergic rhinitis, were discussed in thesupplement.2-4 Compared with placebo, MON adminis-tered once daily at bedtime significantly reduced thenighttime symptom scores and peripheral blood eosin-ophil counts in all 3 trials.

However, in contrast with the 3 large trials cited formontelukast, the intranasal corticosteroid trials discussedvaried considerably in size, scope, and design. The sup-plement did not include a discussion of a recent large,randomized, double-blind, double-dummy, parallel-group trial in 705 subjects with seasonal allergic rhinitisthat compared the effectiveness of a 15-day course ofintranasal fluticasone propionate aqueous nasal spray200 mg (FPANS) with MON, both administered oncedaily in the evening.5 The results of this study showedthat FPANS was consistently superior to MON for alldaytime and nighttime symptoms, including nasal con-gestion. The nighttime symptoms (difficulty going tosleep, nighttime awakenings, nasal congestion on awak-ening) and scoring in this trial were the same as in the3 montelukast trials cited. In addition, a randomized,double-blind, double-dummy, placebo-controlled, paral-lel-group trial in 62 subjects with seasonal allergicrhinitis compared subjects treated with FPANS, MON,MON 1 loratadine 10 mg (LOR), or placebo throughoutthe grass pollen season.6 Subjects assessed their rhinitissymptoms during the study and also underwent nasalbiopsy before and during the season for evaluationof local eosinophilic inflammation. Both MON andMON 1 LOR were less effective than FPANS for con-trol of daytime and nighttime symptoms, including nasalblockage, and for reduction of pollen-induced nasaleosinophilic inflammation on biopsy.

Furthermore, when evaluating nighttime rhinitis treat-ment options, another important consideration is that manypatients with rhinitis have mixed rhinitis, which may in-clude a combination of seasonal allergic, perennial allergic,or perennial nonallergic rhinitis.WhereasMON is currentlyindicated only for relief of seasonal allergic rhinitis, FPANSis indicated for the management of nasal symptoms of all3 of these types of rhinitis.

I agree with the premise that therapy aimed at reducingnighttime nasal congestion is paramount for improving

463

sleep and quality of life, but also that the most effectiveclass of medication (ie, intranasal steroids) should beconsidered as initial therapy for relief of nighttime rhinitissymptoms.

Robert A. Nathan, MDUniversity of Colorado Health Sciences Center

2709 North Tejon

Colorado Springs, CO 80907

Disclosure of potential conflict of interest: Dr Nathan receives

grants/research support from Abbott, Altana, Aventis, AstraZeneca,

Bayer, Berlex, Boehringer Ingelheim, Bristol-Myers Squibb, CIBA

Geigy, Dura, Forest, GlaxoSmithKline, Immunex, Janssen, Parke-

Davis, Pfizer, 3-M Pharmaceuticals, Proctor & Gamble, Roberts,

Sandoz, Sanofi, Schering/Key, Sepracor, Sterling, Tap Pharm,

Wallace, and Wyeth; is a consultant/scientific advisor for AMGEN,

Altana, AstraZeneca, Aventis, Genentech, GlaxoSmithKline, Merck,

Novartis, Pfizer, Schering/Key, Sepracor, and Viropharm; and is on

the speakers’ bureau for AstraZeneca, Aventis, Genentech/Novartis,

GlaxoSmithKline, Pfizer, and Schering/Key.

Editor’s note: This Correspondence has no accompanying reply.

REFERENCES

1. Meltzer EO, editor. Allergic rhinitis after hours: the relevance and conse-

quence of nighttime symptoms. J Allergy Clin Immunol 2004;114:S133-53.

2. Philip G, Malmstrom K, Hampel FC, Weinstein SF, LaForce CF, Ratner

PH, et al. Montelukast for treating seasonal allergic rhinitis: a randomized,

double-blind, placebo-controlled trial performed in the spring. Clin Exp

Allergy 2002;32:1020-8.

3. Nayak AS, Philip G, Lu S, Malice M-P, Reiss TF. Efficacy and toler-

ability of montelukast alone or in combination with loratadine in seasonal

allergic rhinitis: a multicenter, randomized, double-blind, placebo-

controlled trial performed in the fall. Ann Allergy Asthma Immunol

2002;88:592-600.

4. van Adelsberg J, Phillip G, LaForce CF, Weinstein SF, Menten J, Malice

M-P, et al. Randomized controlled trial evaluating the clinical benefit of

montelukast for treating spring seasonal allergic rhinitis. Ann Allergy

Asthma Immunol 2003;90:214-22.

5. Ratner PH, Howland WC, Arastu R, Philpot EE, Klein KC, Baidoo CA,

et al. Fluticasone propionate aqueous nasal spray provided greater im-

provement in daytime and nighttime nasal symptoms of seasonal allergic

rhinitis compared with montelukast. Ann Allergy Asthma Immunol 2003;

90:536-42.

6. Pullerits T, Praks L, Ristioja V, Lotvall J. Comparison of a nasal

glucocorticoid, antileukotriene, and a combination of antileukotriene and

antihistamine in the treatment of seasonal allergic rhinitis. J Allergy Clin

Immunol 2002;109:949-55.


Efficacy of ant venom immunotherapy andwhole body extracts

To the Editor:Golden1 presents a useful review of insect venom

immunotherapy, but we disagree with his conclusionthat imported fire ant (IFA) whole body extract (WBE)has been proven efficacious. Golden1 stresses the needto understand the natural history of sting allergy andthe importance of controlled studies. We add to this theneed for prospective design, adequate randomization, anddouble-blinding.

No prospective controlled study of IFAWBE treatmentefficacy or prospective study of the natural history of IFAallergy has been published. Retrospective studies haveselection bias, and the natural history of allergy can varyenormously between species. Large prospective studieshave found reaction rates on re-exposure to range from70% for the jack jumper ant through 50% for the honeybeeand to 25% for the yellow jacket.2,3 Individuals allergic toIFA who react to multiple simultaneous stings (as oftenoccurs) may experience few reactions when exposed tosmaller doses of venom.

Our randomized, double-blind, placebo-controlled trialof venom immunotherapy (VIT) provides a model thatcould be used to assess IFAWBE.3 How didwe justify ourstudy, and why did 2 respected university ethics commit-tees approve? First, large studies that have demonstratedthe safety of sting challenges after applying health and ageexclusion criteria included a total of 238 patients withsevere (Mueller grade IV) allergy. Second, the efficacydata from 2 controlled trials of VIT to prevent honey beeand vespid sting anaphylaxis are suboptimal by contem-porary standards. One allocated treatment according topatient choice, with outcomes determined by reactionsoccurring outside hospital that were unobserved by theinvestigators. The other was single-blind and stratified byusing factors that do not influence reaction risk. Finally,the efficacy of immunotherapy varies between species,and we could not be sure of the efficacy of jack jumper antVIT.

Without double-blinding, investigators can be misledby personal bias and subjective features such as itch, mildflushing, breathlessness, anxiety, and hypotension asso-ciated with bradycardia and anxiety. It is notable that wegave epinephrine to a patient who appeared to have amoderate reaction to an injection that was later revealedto be placebo.3

The inclusion of a blind placebo group is the only wayto be certain that a sting challenge adequately tests treat-ment efficacy. This is an important ethical issue; we haveobserved the death of a man who believed he was pro-tected by ant WBE, reinforcing the risks of a flawedevidence base. As explained, the exposure to a smallnumber of IFA in a sting challenge may be insufficientto provoke anaphylaxis in many people with allergy.Furthermore, insect handling procedures may lead to adepletion of venom as assessed at the time of venom sacdissection.3

Imported fire ant WBE may be an effective treatment,because extracts contain venom proteins and quality con-trol methods have been developed. However, uncertaintiesremain with regard to the natural history of IFA allergyand whether the dose of venom delivered by IFA WBEextracts is sufficient to confer protection.

Simon G. A. Brown, MBBS, PhD, FACEMa

Robert J. Heddle, MBBS, PhD, FRACP, FRCPAb

Michael D. Wiese, BPharm, MClinPharmc

Konrad E. Blackman, MBBS, FACEMc

aUniversity of Western Australia

Fremantle Hospital


AUGUST 2005

464 Correspondence

Alma Street

Fremantle, WA 6160, AustraliabDepartment of Respiratory Medicine

Flinders Medical Centre

Bedford Park, AustraliacRoyal Hobart Hospital

Hobart, Australia

REFERENCES

1. Golden DB. Insect sting allergy and venom immunotherapy: a model and

a mystery. J Allergy Clin Immunol 2005;115:439-47.

2. Brown SGA, Franks RW, Baldo BA, Heddle RJ. Prevalence, severity, and

natural history of jack jumper ant venom allergy in Tasmania. J Allergy

Clin Immunol 2003;111:187-92.

3. Brown SGA, Wiese MD, Blackman KE, Heddle RJ. Ant venom immu-

notherapy: a double-blind, placebo-controlled, crossover trial. Lancet

2003;361:1001-6.


Reply

To the Editor:I appreciate the comments of Brown et al.1 It was clearly

only by unintended oversight that the exemplary work ofthe authors was not cited in my review that focused ontreatment strategies in the United States.2 Their work onthe natural history of ant venom allergy and their con-trolled trial of ant venom immunotherapy are a model towhich we should aspire.3,4 In contrast, there has been nocontrolled trial of imported fire ant (IFA) whole bodyextract (WBE) immunotherapy, and the natural history ofIFA allergy is unknown.

I am dismayed that my statements were construed toimply that IFA WBE has been proven efficacious.Comparison of IFA WBE and venom in vitro and byskin test suggests that although inferior to the venom,WBE contains sufficient allergen to provide reasonablediagnostic accuracy.5-8 For this reason, I stated, ‘‘For fireant allergy, venom is themost accurate diagnosticmaterial,but WBE have shown adequate diagnostic sensitivity.’’My statement that ‘‘Fire ant immunotherapy is performedwith WBE that contains sufficient venom allergens toprovide reasonable clinical protection’’ was based on thereported content of venom allergens in IFA WBE. I didnot state that WBE was as potent as venom.8 The reportof clinical efficacy of IFA WBE by Freeman et al9 hadthe strengths of prospective sting challenge (instead ofretrospective field sting reports) and a limited (buthighly significant) control group. Still, Freeman et al9

concluded that ‘‘A controlled prospective trial of WBEversus placebo is needed.to help define the naturalhistory of IFA hypersensitivity.’’ Of recent interest arereports of rush immunotherapy with IFA WBE to preventsystemic and large local reactions, but these too wereuncontrolled.10,11

Together, the in vivo and in vitro evidence led to thebelief that unlike the other Hymenoptera, IFA WBE does

contain sufficient venom allergens to have acceptable,albeit suboptimal, efficacy. Also, unlike the wingedHymenoptera WBE, there are relatively few reports oftreatment failure with IFA WBE, and no fatalities. Brownet al1 mention a fatal reaction but not the species of WBEused for treatment or the dose, schedule, and durationof treatment. However, our experience with the wingedHymenoptera WBEs ‘‘demonstrates the value of a com-plete understanding of the natural history of the disease indetermining the efficacy and indications for treatment andthe importance of clinical trials.’’2 As Brown et al1 pointout, prospective design and blind treatment are also criticalto the strength of the evidence.

The lack of IFA venom products for diagnosis andimmunotherapy in the United States is a continuing gap inour repertoire. It is not the ethics but the economics ofclinical trials that has deterred the performance of double-blind, placebo-controlled clinical trials of IFA WBE andvenom products and encouraged the acceptance of theWBEs as the only option in practice. However, thedilemma posed by the authors has attracted the attentionof the Insect Committee of the American Academy ofAllergy, Asthma and Immunology, who have now re-solved to explore the development of a controlled trialof IFA WBE immunotherapy. (Nelson, personal com-munication, March 2005). When the efficacy of currenttreatment has not been proven up to current standardsand the risk of treatment failure is a life-threateningreaction, a controlled trial is clearly justified. Ourthanks to Brown et al1 for exposing this importantissue.

David B. K. Golden, MD

Johns Hopkins Asthma and Allergy Center5501 Hopkins Bayview Blvd

Baltimore, MD 21224

REFERENCES

1. Brown SGA, Heddle RJ, Wiese MD, Blackman KE. Efficacy of ant

venom immunotherapy and whole body extracts. J Allergy Clin Immunol

2005;116:464-5.

2. Golden DBK. Insect sting allergy and venom immunotherapy: a model

and a mystery. J Allergy Clin Immunol 2005;115:439-47.

3. Brown SG, Franks RW, Baldo BA, Heddle RJ. Prevalence, severity and

natural history of jack jumper ant venom allergy in Tasmania. J Allergy

Clin Immunol 2003;111:187-92.

4. Brown SG, Wiese MD, Blackman KE, Heddle RJ. Ant venom immu-

notherapy: a double-blind placebo-controlled crossover trial. Lancet

2003;361:1001-6.

5. Strom GB, Boswell MD, Jacobs RL. In vivo and in vitro comparison of

fire ant venom and fire ant whole body extract. J Allergy Clin Immunol

1983;72:46-53.

6. Paull BR, Coghlan TH, Vinson SB. Fire ant venom hypersensitivity,

I: comparison of fire ant venom and whole body extract in the diagnosis

of fire ant allergy. J Allergy Clin Immunol 1983;71:448-53.

7. Butcher BT, deShazo RD, Ortiz AA, Reed MA. Superiority of

Solenopsis invicta venom to whole body extract in RAST for diagnosis

of imported fire ant allergy. Int Arch Allergy Appl Immunol 1988;85:

458-61.

8. Hoffman DR, Jacobson RS, Schmidt M, Smith AM. Allergens in

Hymenoptera venoms, XXIII: venom content of imported fire ant whole

body extracts. Ann Allergy 1991;66:29-31.



Correspondence 465

9. Freeman TM, Hyghlander R, Ortiz A, Martin ME. Imported fire ant

immunotherapy: effectiveness of whole body extracts. J Allergy Clin

Immunol 1992;90:210-5.

10. Tankersley MS, Walker RL, Butler WK, Hagan LL, Napoli DC, Freeman

TM. Safety and efficacy of an imported fire ant rush immunotherapy

protocol with and without prophylactic treatment. J Allergy Clin

Immunol 2002;109:556-62.

11. Walker R, Jacobs J, Tankersly M, Hagan L, Freeman T. Rush immu-

notherapy for the prevention of large local reactions secondary to

imported fire ant stings [abstract]. J Allergy Clin Immunol 1999;103:

S180.



AUGUST 2005

466 Correspondence

Toll-like receptors and atopyPierre Olivier Fiset, BSc, Meri Katarina Tulic, PhD,

and Qutayba Hamid, MD, PhD, Editors

Editor’s note: This feature, Images in allergy and immunology,

is designed to highlight current concepts of the immunopathol-ogy of allergic diseases and other common immunologically

mediated diseases. The presentation will appear as sets of

images that involve cross-pathology, histopathology, andmolecular pathology and will cover a range of topics of interest

to allergists and immunologists.

The Toll-like receptors (TLRs) are a recently dis-

covered family of receptors involved in the innate

recognition of pathogens. TLRs have much homology

to the IL-1 receptor family and the Drosophila Toll

protein, and at least 10 distinct TLRs have now been

identified in human subjects (Fig 1). TLR ligands are

highly conserved structures and molecules present on

many pathogens, the so-called pathogen-associated

molecular patterns (PAMPs). Some PAMPs are bac-

terial molecules, such as lipopeptides, mannans, LPSs,

flagellin, and CpGDNA. Other PAMPs recognized by

TLRs include virus- and fungus-associated molecules.

Triggering of TLRs leads to expression of many genes

involved in inflammatory responses to pathogens,

leading to cell activation, differentiation, proliferation,

and cell recruitment. Because of their strong immu-

nostimulatory capacities, many PAMPs are currently

studied as potential treatment agents for allergic

diseases.

Because the TLRs are part of the innate immune

system, they are not modified during an immune

response and are passed on to the progeny with little

genetic change. This has prompted genetic studies to

determine whether specific single nucleotide polymor-

phisms in the TLR genes are associated with atopy.

A recent study has suggested a polymorphism in

TLR2 and TLR4 in Europeans to be associated with

decreased atopy, dependent on PAMP exposure

467

Images in

Allergyand

Immunology

From Meakins-Christie Laboratories, Department of Pathology and

Medicine, McGill University, Montreal, Canada.

Received for publication April 21, 2005; accepted for publication

April 22, 2005.


Reprint requests: Qutayba Hamid, MD, PhD, McGill University,

Meakins-Christie Laboratory, 3626 St Urbain St, Montreal, Canada

H2X 2P2. E-mail: [email protected].


0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.034

FIG 1. Ten distinct TLRs exist in human subjects, recognizing

many PAMPs. TLRs can associate as heterodimers changing

their ligand specificity.

FIG 2. A, Prevalence of asthma, atopy, and current hay fever

symptoms in TLR2/216934 in farmers’ children. B, Prevalence

of asthma, atopy, and current hay fever symptoms in TLR4/

14434 in children exposed to high endotoxin concentrations.1

J ALLERGY CLIN IMMUNOL August 2005


468

(Fig 2).1 On the other hand, another study showed no

association of atopy and polymorphisms in TLR2,TLR3, TLR4, and TLR9 in Japanese populations.2

Evidence from epidemiologic studies has shown an

association between high exposure to PAMPs during

early life with decreased levels of atopic diseases and

asthma. This has led to the proposal of the hygiene

hypothesis, which states that lack of a ‘‘pathogenic

pressure’’ (increased hygiene) in early childhood

results in an imbalanced immune system hypersensi-

tive to allergens (Fig 3). Thus atopy is associated with

increases in TH2 cytokines compared with TH1 or

immunoregulatory cytokines. Tulic et al3 have shown

that LPS can cause a proliferation of CD3-positive

cells in the nasal mucosa of children (Fig 4). This was

associated with increases in IL-2–positive, IL-12–

positive, and IFN-g–positive cells without increases

in TH2 cytokines and MBP–positive cells (Fig 5). It

has been also shown that LPS can inhibit allergen-

induced increases in IL-4–positive, IL-5–positive, and

IL-13–positive cells, as well as in MBP-positive and

tryptase-positive cells.4 These effects were determined

to be due to the increase of IL-10, IL-12, and IFN-g

induced by LPS. Additionally, in this same study

TLR4-positive cells were higher and more responsive

to LPS in children compared with adults.

PAMPs are also studied to treat patients with atopic

diseases. In this context PAMPs are used as adjuvants

for current immunotherapy regimens to reduce the

antigen dose needed for therapy and to promote the

development of immunoregulatory mechanisms. For

example, combining CpG DNA with ragweed immu-

notherapy has been shown to provide potential clinical

benefits. In a mouse model of allergy, physical linking

of CpG DNA to ragweed protein inhibited IgE

expression, inhibited IL-5 expression, and promoted

IFN-g expression.5 Physical linking of the CpG DNA

to the ragweed protein enhanced the effects, suggest-

ing that cells reacting to the allergen also have TLR9

FIG 3. The hygiene hypothesis states that exposure to microorganisms (decreased hygiene) during early age

is important for the development of a balanced immune system. Increased hygiene leads to an uncontrolled

TH2 immune response to allergens, resulting in atopic diseases.

FIG 4. Detection of bromodeoxyuridine-positive proliferating

cells colocalized with CD3-positive cells in explants of nasal mu-

cosa of children. The explants were stimulated with LPS (0.1 mg/

mL) for 2 hours (A) and 24 hours (B). Ten percent colocalization

wasseenat2hours,and70%colocalizationwasseenat24hours.3

August 2005 J ALLERGY CLIN IMMUNOL

activated by the CpG DNA (Fig 6). Ragweed CpG

DNA injections in human clinical trials has been

shown to inhibit allergen-induced IL-4 mRNA ex-

pression, IL-5 mRNA expression, and MBP-positive

cell numbers in the nasal mucosa of allergic patients

(Fig 7) and to reduce chest and nasal symptom scores.6

The compound can also increase the number of TLR9-

positive cells in the nasal mucosa (Fig 8). As the role of

PAMPs and TLRs is clarified in atopy and TLRs are

better characterized, development of new therapies to

both prevent atopic diseases and treat existing disease

will be possible.

REFERENCES

1. EderW, KlimeckiW, Yu L, vonMutius E, Riedler J, Braun-Fahrlander

C, et al. Toll-like receptor 2 as a major gene for asthma in children of

European farmers. J Allergy Clin Immunol 2004;113:482-8.

2. Noguchi E, Nishimura F, Fukai H, Kim J, Ichikawa K, Shibasaki M,

et al. An association study of asthma and total serum immunoglobin

E levels for Toll-like receptor polymorphisms in a Japanese popu-

lation. Clin Exp Allergy 2004;34:177-83.

3. Tulic MK, Manoukian JJ, Eidelman DH, Hamid Q. T-cell prolifer-

ation induced by local application of LPS in the nasal mucosa of

nonatopic children. J Allergy Clin Immunol 2002;110:771-6.

4. Tulic MK, Fiset PO, Manoukian JJ, Frenkiel S, Lavigne F, Eidelman

DH, et al. Role of toll-like receptor 4 in protection by bacterial

lipopolysaccharide in the nasal mucosa of atopic children but not

adults. Lancet 2004;363:1689-97.

469

FIG 6. Linking of the allergen to an immunostimulatory CpG

DNA sequence increases the potency of the ragweed CpGDNA

vaccine for immunotherapy. The same antigen-presenting cell

is activated by the complex, through TLR9, to synthesize

cytokines and increase antigen presentation of the allergen.

Allergen presentation and cytokines activate allergen-specific

T cells to change their cytokine profile.

FIG 7.Horseradish peroxidase immunocytochemistry for MBP-

positive cells in sections of patients receiving the ragweed CpG

DNA vaccine for immunotherapy (A) or placebo (B).

FIG 5. IL-12 in situ hybridization of sections taken from nasal

explants without stimulation (A) and stimulated with 0.1 mg/mL

LPS (B). IL-2, IL-12, and IFN-g profile after stimulation with 0.1

mg/mL LPS (C). Open bars indicate no stimulation, and filled

bars indicate LPS stimulation.3


4705. Tighe H, Takabayashi K, Schwartz D, Van Nest G, Tuck S, Eiden JJ,

et al. Conjugation of immunostimulatory DNA to the short ragweed

allergen Amb a 1 enhances its immunogenicity and reduces its

allergenicity. J Allergy Clin Immunol 2000;106:124-34.

6. Tulic MK, Fiset PO, Christodoulopoulos P, Vaillancourt P,

Desrosiers M, Lavigne F, et al. Amb a 1-immunostimulatory

oligodeoxynucleotide conjugate immunotherapy decreases the nasal

inflammatory response. J Allergy Clin Immunol 2004;113:235-41.

Chronic active Epstein-Barr virus infectionof natural killer cells presenting as severe

skin reaction to mosquito bitesSusan E. Pacheco, MD, Stephen M. Gottschalk, MD,

Mary V. Gresik, MD, Megan K. Dishop, MD,Takayuki Okmaura, MD, and Theron G.

McCormick, MD, Guest Editors

Discovered more than 40 years ago, EBV is known

to exhibit tropism for lymphocytes, especially B-cells.

This g herpes virus is capable of immune evasion,

producing latent infections that might lead to B-cell

and other lymphoproliferative diseases. A relatively

unusual target of EBV infection involves natural killer

(NK) cells. Despite varying classifications, a form of

chronic active EBV infection (CAEBV) involving NK

cells presents with severe inflammatory and necrotic

skin reactions considered pathognomonic of EBV1

NK cell lymphoproliferative disease.1-3 Most patients

presenting with this condition are of Asian descent,

and there is no sex predominance.

Fig 1 shows a 7-year-old Latin American boy with

NK cell CAEBV. Typical of this condition is the

presence of bullous and ulcerative skin lesions after

exposure to mosquito bites. In addition, patients

develop high fever, lymphadenopathy in draining

nodes, and marked hepatosplenomegaly. Bullous

FIG 8. In situ hybridization for TLR9 mRNA–positive cells in

sections of patients receiving placebo (A) or the ragweed CpG

DNA vaccine for immunotherapy (B).

From Baylor College of Medicine, Pediatric Allergy and Immunology

Service, Texas Children’s Hospital, Houston, Tex.

Received for publication February 25, 2005; revised April 11, 2005;



Reprint requests: Theron G. McCormick, MD, Baylor College of

Medicine, Pediatric Allergy and Immunology Service, Texas

Children’s Hospital, 6621 Fannin St. FC330.01, Houston, TX



0091-6749/$30.00


doi:10.1016/j.jaci.2005.04.044

FIG 1.

FIG 2.


lesions develop within 24 hours after mosquito expo-

sure and are filled with a sterile fluid; this is followed

by necrotic ulcerations (Figs 2 through 4).

Skin biopsy from a bullous lesion revealed subep-

idermal bullae with a dense dermal infiltrate of eosin-

ophils, lymphocytes, and histiocytes and a negative

gram stain analysis (hematoxylin-eosin stain; Figs 5

and 6).

The unusual reaction to mosquito bites, very high

IgE level (often >10,000 IU/mL), and significant

eosinophilia has prompted the nomenclature of ‘‘hy-

persensitivity reaction.’’ However, this condition does

not meet criteria for an immunologic allergic or

hypersensitivity reaction to mosquitoes on laboratory

or clinical grounds.

471

FIG 4.

FIG 5.

FIG 6.

FIG 7.

TABLE I. Characteristics of NK cell CAEBV

Cell markers CD32, CD561, and/or CD161

Receptor for infection Unknown

Main transforming protein Unknown

Mosquito bite reactions Often present

Target population First 2 decades of life

Associated malignancies Hemophagocytic

lymphohistiocytosis,

NK cell leukemia,

NK cell lymphoma

FIG 3.


472

By current parameters, patients with NK cell

CAEBV disease seem to have normal immunity

before development of the disease. An immunologic

evaluation in this patient revealed normal lymphocyte

proliferation to antigen and mitogens and functional

antibodies to polysaccharide and protein antigens.

Skin biopsy specimens from patients with NK cell

CAEBV related to mosquito bites are significant for

an inflammatory reaction composed primarily of

NK cells (CD561CD3–) expressing EBV DNA by

in situ hybridization (Fig 7).

PBMCs from affected patients often demonstrate

30% to 70% NK cells, most infected with monoclonal

or oligoclonal EBV. In addition, EBV DNA PCR

levels from PBMCs are significantly elevated, with

mean levels of 1042 copies/mg.4 A table distinguishing

NK cell CAEBV is provided (Table I). Aside from

symptomatic care, the optimal treatment option is bone

marrow transplantation.5

REFERENCES

1. Miyazato H, Nakasuka S, Dong Z, Takakuwa T, Oka K, Hanamoto

H, et al. NK-cell related neoplasms in Osaka, Japan. Am J Hematol

2004;76:230-5.

2. TokuraY, Ishihara S, Tagawa S, Naoshiro S, OhshimaK, TakigawaM.

Hypersensitivity tomosquito bites as the primary clinical manifestation

of a juvenile type of Epstein-Barr virus-associated natural killer cell

leukemia/lymphoma. J Am Acad Dermatol 2001;45:569-78.

3. Ohga S, Nomura A, Takada H, Hara T. Immunological aspects of

Epstein-Barr virus infection. Crit Rev Oncol Hematol 2002;44:203-15.

4. Kimura H, Hoshino Y, Kanegane H, Tsuge I, Okamura T, Kawa K,

et al. Clinical and virologic characteristics of chronic active

Epstein-Barr virus infection. Blood 2001;98:280-6.

5. Fujii N, Takenaka K, Hiraki A, Maeda Y, Ikeda K, Shingawa K, et al.

Allogeneic peripheral blood stem cell transplantation for the treat-

ment of chronic active Epstein-Barr virus infection. Bone Marrow

Transplant 2000;26:805-8.


Articles of note. . .

Adverse events after influenza immunization inyoung childrenBecause influenza (Flu) infections cause considerable

morbidity in young children, the Advisory Committee

on Immunization Practices has encouraged health care

providers to give healthy 6- to 23-month-old children

the trivalent Flu influenza vaccine (TFV). However,

concerns have been raised by some parents about

adverse effects of such immunization. This study

reviewed records of the Vaccine Adverse Event

Reporting System (VAERS), a passive surveillance

system begun by the US Food and Drug Adminis-

tration and the Centers for Disease Control and Pre-

vention in 1990. Since 1990, there were 166 reports

of adverse events (AEs) following receipt of the TFV

alone or along with other vaccines in children less

than 2 years old. These AEs have generally been quite

mild, (fever, transient rash). Seizureswere themost com-

mon serious AE, reported in 28 cases. Most of the

seizures occurred along with fever with onset within

2 days after immunization. No sequelae were re-

ported. Although there is probably some underreport-

ing of AE in the voluntary reporting in the VAERS,

these findings suggest that TFV immunization is

generally very safe and well tolerated by young

children.

(McMahon et al. Pediatrics 2005;115:453-9.)

Respiratory syncytial virus infection in elderlyand high-risk adultsRespiratory syncytial virus (RSV) infection, exten-

sively investigated in children, is increasingly recog-

nized as a cause of illness in adults. This study

prospectively investigated all respiratory illnesses in

cohorts of (1) healthy elderly patients (65 years of age

or older), (2) high-risk adults (those with chronic heart,

lung, or airways disease), and (3) patients hospitalized

with acute cardiopulmonary conditions. RSV infec-

tions occurred annually in 3% to 7% of healthy elderly

patients and in 4% to 10% of high-risk adults. The

frequency of RSV infection was at least as great as that

of influenza A in these populations. In the hospitalized

patients, RSV infection and influenza A resulted in

similar lengths of stay, rates of use of intensive care

(15% and 12%, respectively), and mortality (8% and

7%, respectively). RSV infection accounted for 10.6%

of hospitalizations for pneumonia, 11.4% of hospital-

izations for chronic obstructive pulmonary disease,

and 7.2% of hospitalizations for asthma. The authors

concluded that RSV infection is an important illness

in elderly and high-risk adults, with a disease burden

similar to that of nonpandemic influenza A in a

population in which the prevalence of immunization

for influenza is great. One would hope that these

findings will help stimulate a continued search for an

effective, safe RSV vaccine.

(Falsey et al. N Engl J Med 2005;352:1749-59.)

Anaphylactic reaction to lupin flourFlour made from ground-up lupin beans is being

used increasingly as a wheat flour substitute in

some European countries. This case report described

a severe anaphylactic reaction in a 25-year-old woman

shortly after ingestion of a meal containing chicken,

fried potatoes, and onion rings with recovery after

intensive therapy. There was a past history of transient

asthma at age 15 years and a prior anaphylactic

reaction to peanuts. It was found that the breading on

the onion rings contained lupin flour. Subsequent skin

tests to peanuts and a crude extract of lupin flour were

strongly positive. There is probably a 20% to 40%

cross-reacting homology between lupin and one of

the allergens in peanuts. Lupin flour allergy has been

reported mainly in European patients known to be

allergic to other legumes, particularly peanut, soy, or

pea. Indeed, reactions to lupin are one of the most

common types of food-induced anaphylaxis in France.

This report should be kept in mind, because lupin

beans and lupin flour are now becoming available for

consumption in the United States.

(Radcliffe et al. Lancet 2005;365:1360.)

A new mechanism of nonatopic asthma elicited byimmunoglobulin free light chainsA significant proportion of asthmatic individuals are

nonatopic, yet the mechanism by which an asthma

exacerbation is triggered in these individuals is not

known. In this report, the investigators extended their

prior observation that antigen-specific immunoglobu-

lin free light chains (LCs) mediate mast cell–depen-

dent hypersensitivity by examining the role of LCs in a

murine model of nonatopic asthma. In particular, they

use a LC antagonist, the 9-mer F991, and abrogate the

development of airway hyperresponsiveness and pul-

monary infiltration. Using mast cell–deficient mice,

they show that the role of LCs is dependent on mast

cells. Finally, they demonstrate that asthmatic indi-

viduals (both atopic and nonatopic) have elevated sera

BeyondOurPages

Burton Zweiman, MD, & Marc E. Rothenberg, MD, PhD, Editors

J ALLERGY CLIN IMMUNOL August 2005 Page 473

levels of jLC (but not kLC) compared with control

individuals. These results substantiate a new mecha-

nism that might be involved in triggering allergic, but

not IgE-mediated, asthma and suggest that inhibition

of LC-mediated mast cell activation might be thera-

peutically useful.

(Kraneveld et al. Proc Natl Acad Sci U S A 2005;102:

1578-83.)

Suppressive effects of prostaglandin E receptor subtypeEP3—the mechanism of aspirin sensitivity?Prostaglandins (PGs), including PGD2 and PGE2, are

produced during allergic responses, yet PG synthesis

inhibitors (eg, aspirin) are generally ineffective for

asthma. Inasmuch as PGD2 is a potent proinflamma-

tory mediator and smooth muscle constrictor, this

suggests that PGE2 might have an important regula-

tory (or protective) role in allergy. To address this

possibility, the investigators subjected mice with

specific deficiencies in each of the 4 PGE2 receptors

(EP1 through EP4) to an OVA-induced model of

asthma. Notably, mice deficient in EP3 had a marked

increase in multiple aspects of asthma (including

inflammation and TH2 cytokine production), whereas

mice deficient in the other receptors were comparable

to wild-type mice. On the basis of these results sug-

gesting a suppressive role for EP3 signaling, the

investigators examined the impact of an EP3-selective

agonist on the development of asthma. Indeed, the EP3

agonist inhibited airway inflammation, TH2 cytokine

production, and bronchoconstriction, even when it

was administered 3 hours after antigen challenge.

In addition, a significant number of allergen-induced

genes were inhibited by the EP3 agonist. Taken

together, these results call attention to the anti-allergic

role of PGE2 and its EP3 receptor, providing a new

paradigm for therapeutic intervention. Furthermore,

the results provide a possible explanation for aspirin-

sensitive asthma by suggesting that such individuals

might preferentially require the protective effects of

the EP3 pathway (and are thus sensitive to aspirin).

(Kunikata et al. Nat Immunol 2005;6:524-31.)

Dendritic cells are critical for experimental asthmaDendritic cells (DCs) have been shown to have an

important role in sensitization to inhaled allergens, but

their function in ongoing TH2 cell–mediated lung

inflammation is currently unknown. Using an OVA-

induced murine asthma model, the investigators show

that airway DCs acquire a mature phenotype and

interact with CD4 T cells within the lung tissue. To

study whether the DCs contributed to inflammation,

they subsequently depleted the DCs from the airways

using CD11c-diphtheria toxin (DT) receptor trans-

genic mice during the OVA aerosol challenge. In these

mice, DT is only active on CD11c+ cells (primarily

DCs and macrophages). Indeed, administration of DT

to the lungs of these transgenic mice depleted CD11c+

DCs and alveolar macrophages. Notably, DT abol-

ished the characteristic features of asthma, including

eosinophilic inflammation, goblet cell hyperplasia,

and bronchial hyperreactivity. Furthermore, in the

absence of CD11c+ cells, TH2 cells did not produce

IL-4, IL-5, and IL-13 in response to OVA aerosol.

Importantly, in CD11c-depleted mice, eosinophilic

inflammation and TH2 cytokine secretion were re-

stored by adoptive transfer of CD11c+ DCs, but not

by transfer of alveolar macrophages. These findings

identify lung DCs as key pro-inflammatory cells that

are necessary and sufficient for TH2 cell stimulation

during ongoing lung inflammation.

(van Rijt et al. J Exp Med 2005;201:981-91.)

Daily versus as-needed inhaled corticosteroid treatmentof mild, persistent asthmaMost current national guidelines recommend daily use

of controller medications, such as an inhaled cortico-

steroid (ICS), in the treatment of persistent asthma

(PA), even of mild degree. This randomized, double-

blind study investigated whether treatment with daily

ICS (budesonide 200 micrograms bid), or daily

leukotriene antagonist (zafirlukast 20 mg bid) was

more effective than daily placebo with as-needed ICS

use in the treatment of mild PA in 199 adults. After

1 year of treatment, there were no significant differ-

ences in the major asthma outcome (increases in

average morning peak expiratory flow) among the 3

treatment groups. The frequency of acute asthma

exacerbations requiring corticosteroid therapy was

also not significantly different in the 3 groups. There

were greater improvements in patients using ICS daily

than in placebo-treated individuals in prebroncho-

dilator FEV1 (P = .005), PC20 bronchial reactivity

(P < .001), asthma control score (P < .001), sputum

eosinophils (P = .007), and number of symptom-free

days (P = .03). However, the postbronchodilator FEV1

and quality of life scores were not significantly

different between those treated daily with ICS and

those receiving placebo. There were no differences in

any asthma outcome scores between those treated with

daily zafirlukast and those receiving placebo. These

findings indicate that daily ICS treatment improves

some manifestations of mild PA in adults. However,

the authors concluded that daily ICS might not be

needed in such asthmatic individuals; they can instead

be treated with short intermittent courses of inhaled

or oral corticosteroids taken when asthma symptoms

worsen significantly. It is still uncertain whether these

findings can be extended to mild PA in children.

(Boushey et al. N Engl J Med 2005;352:1519-28.)

Page 474 August 2005 J ALLERGY CLIN IMMUNOL

journal of allergy and clinical immunology

Documents