c radiological r a r table of contentsyong liang zhao, chang piao and tom k. hei..... 69 mutation(s)...

CENTER FOR RADIOLOGICAL RESEARCH • ANNUAL REPORT 2001

1

Table of ContentsTABLE OF CONTENTS ...................................................................................................................................................... 1INTRODUCTION ................................................................................................................................................................ 4STAFF PHOTO ................................................................................................................................................................... 5STAFF LISTING ................................................................................................................................................................. 6STAFF NEWS ..................................................................................................................................................................... 7THE COLUMBIA COLLOQUIUM AND LABORATORY SEMINARS ................................................................................ 8WEB SITES ........................................................................................................................................................................ 8ACKNOWLEDGEMENTS ................................................................................................................................................... 8

RESEARCH REPORTS (for collaborating institutions see the box following Table of Contents)

PHYSICS, BIOPHYSICS, AND MODELING

Laser Ion Source Development for the Columbia University MicrobeamAlan W. Bigelow, Gerhard Randers-Pehrson, and David J. Brenner ............................................................................... 9

Are there Geometric Associations between Different Human Chromosomes?David J. Brenner, in collaboration with Karin M. Greulich-Bode, Martina Brückner, Michael N. Cornforth, BradfordLucas and Rainer K. Sachs .......................................................................................................................................... 11

Biomarkers Specific to Densely-Ionizing (High-LET) RadiationsDavid J. Brenner, M. Prakash Hande and Charles R. Geard, with Nadia Okladnikova, Ludmilla Burak and TamaraAzizova ...................................................................................................................................................................... 12

Do Low Dose-Rate Bystander Effects Influence Domestic Radon Risks?David J. Brenner, in collaboration with Rainer K. Sachs .............................................................................................. 14

MICROBEAM & BYSTANDER STUDIES

Induction of p21/WAF1 in Microbeam Irradiated and Bystander Normal Human FibroblastsBrian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Gerhard Randers-Pehrson, & Charles R. Geard.. 17

Intra-Nuclear Dynamics of Mre11 in Human Cells Following Microbeam Irradiation with α-ParticlesAdayabalam S. Balajee and Charles R. Geard.............................................................................................................. 18

SCGE Detection of DNA Strand Breaks and Oxidized Base Lesions Induced by Microbeam Irradiation ofHuman Nuclei with Defined Number of Alpha ParticlesAdayabalam S. Balajee, Brian Ponnaiya, Manoor P. Hande, Gloria Jenkins-Baker, Stephen Marino, Gerhard Randers-Pehrson and Charles R. Geard ..................................................................................................................................... 20

Modulation of Adaptive Response in Alpha Particle Induced Bystander MutagenesisHongning Zhou, Gerhard Randers-Pehrson, Eric J. Hall and Tom K. Hei .................................................................... 22

Novel Approaches with Track Segment Alpha Particles and Cell Co-Cultures in Studies of Bystander EffectsCharles R. Geard, Gloria Jenkins-Baker, Stephen A. Marino, Gary Johnson and Brian Ponnaiya .................................. 23

Investigation of a Radiation-Induced Bystander Effect Using Co-Culturing ProtocolsBrian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Stephen Marino and Charles R. Geard ............... 24

Investigation of the Role of Cell Type Specificity in the Induction of a Bystander EffectBrian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Stephen Marino and Charles R. Geard ............... 26

Induction of Chromosomal Aberrations in γ-Irradiated and Bystander FibroblastsBrian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Stephen Marino and Charles R. Geard ............... 27

Detection of Chromosomal Instability in Co-Cultured γ-Irradiated and Bystander Human FibroblastsBrian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Stephen Marino and Charles R. Geard ............... 28

Induction of DNA Repair and Signal Transduction Proteins Triggered by Ionizing Radiation in Bystander CellsAdayabalam S. Balajee and Charles R. Geard.............................................................................................................. 29

CELLULAR STUDIES

Ataxia Telangiectasia Fibroblasts with Extended Lifespan through Telomerase Expression Retain TheirCellular CharacteristicsTej K. Pandita, Sonu Dhar and Arun Gupta, with Lauren D. Wood, Fred Levine, Jerry W. Shay and Jean J. Y. Wang .. 33


2

Spontaneously Immortalized Cell Lines Obtained from Adult ATM Null Mice Retain Sensitivity to IonizingRadiation and Exhibit a Mutational Pattern Suggestive of Oxidative StressSonu Dhar, Girdha G. Sharma and Tej K. Pandita, in collaboration with B.M. Gage, D. Alroy, C.Y. Shin,O.N. Ponomareva, M.J. Thayer and M.S. Turker ......................................................................................................... 34

Extension of Lifespan by Transfection of hTERT in Normal Human Mammary Epithelial CellsLi Liu, Yong L. Zhao, Chang Q. Piao and Tom K. Hei................................................................................................. 35

Immortalization of Primary Human Bronchial Epithelial Cells by Overexpression of Human TelomeraseCatalytic Subunits (hTERT)Chang Q. Piao, Yong L. Zhao, Li Liu and Tom K. Hei................................................................................................. 37

Monitoring Tumor Progression in a Radiation and Estrogen-Induced Breast Cancer ModelGloria M. Calaf, Debasish Roy and Tom K. Hei .......................................................................................................... 38

Effect of Retinol on Radiation and Estrogen-Induced Neoplastic Transformation of Human Breast EpithelialCellsGloria M. Calaf and Tom K. Hei, in collaboration with Nancy J. Emenaker ................................................................. 40

Susceptibility of Human Breast Tissue to Neoplastic Changes Induced by Organophosphorous PesticidesGloria M. Calaf and Tom K. Hei, in collaboration with Gertrudis Cabello.................................................................... 42

Arsenic Induces Oxidative DNA Damage in Mammalian CellsSu Xian Liu, An Xu and Tom K. Hei, in collaboration with Maris Kessel and Regina Santella ..................................... 43

Role of Mitochondrial Oxidants as Regulators of Mutagenicity of Arsenite in Mammalian CellsSu Xian Liu and Tom K. Hei....................................................................................................................................... 45

Development of a Flow Cytometric Assay for the Quantification of CD59 Mutations in Human-Hamster Hybrid(AL) CellsAn Xu, Haiying Hang, Hongning Zhou, Raheel Ansari and Tom K. Hei....................................................................... 46

Extranuclear Targets in the Genotoxicity of Asbestos in Mammalian CellsAn Xu, Suxian Liu and Tom K. Hei............................................................................................................................. 47

Mutagenicity of Crocidolite Asbestos in Mammalian Cells is Associated with Nitric Oxide ProductionAn Xu and Tom K. Hei ............................................................................................................................................... 48

CYTOGENETIC STUDIES

Use of Multicolor Fluorescence in Situ Hybridization (m-FISH) to Detect Radiation-Induced ChromosomeAberrations in Human CellsM. Prakash Hande and David J. Brenner...................................................................................................................... 51

Radiation Induced Inter-Arm Exchanges Detected by Fluorescence in Situ Hybridization Using ChromosomeSingle Arm-Specific ProbesM. Prakash Hande, Adayabalam S. Balajee, Charles R. Geard and David J. Brenner .................................................... 53

Aberrant Hrad9 Expression Influences Telomere Behavior and Ionizing Radiation-Induced ChromosomalInstabilitySonu Dhar, Wei Zheng, Girdhar G. Sharma, Kevin M, Hopkins, Howard B. Lieberman and Tej K. Pandita.................. 55

Extra-Chromosomal Telomeric DNA in Cells from Atm-/- Mice and Patients with Ataxia-TelangiectasiaM. Prakash Hande and Adayabalam S. Balajee, in collaboration with Andrei Tchirkov1, Anthony Wynshaw-Boris andPeter M. Lansdorp....................................................................................................................................................... 56

MOLECULAR STUDIES

Mrad9 Knockout Mouse ES Cells are Sensitive to Ionizing RadiationHoward B. Lieberman and Kevin M. Hopkins, in collaboration with Alexandra L. Joyner and Wojtek Auerbach ......... 61

Identification and Characterization of a Paralogue of Human Cell Cycle Checkpoint Gene HUS1Haiying Hang and Howard B. Lieberman, in collaboration with Yuzhu Zhang and Roland L. Dunbrack, Jr. ................. 62

hSIR2SIRT1 Functions as an NAD-Dependent p53 DeacetylaseTej K. Pandita, in collaboration with Homayoun Vaziri, Scott K. Dessain, Elinor Ng Eaton, Shin-Ichiro Imai, Roy A.Frye, Leonard Guarente and Robert A. Weinberg ........................................................................................................ 63

Influence of PTEN on Telomere Stability and Telomerase ActivityGirdhar G. Sharma, Janusz Puc, Sonu Dhar, Ramon Parson and Tej K. Pandita............................................................ 64


3

hTERT Association with Telomeres Correlates with Reduction in Spontaneous Chromosome Damage andEnhancement of DNA RepairGirdhar G. Sharma, Arun Gupta, Sonu Dhar and Tej K. Pandita................................................................................... 65

14-3-3σ Influences Telomere Stability and ATM Kinase ActivityArun Gupta, Sonu Dhar and Tej K. Pandita, in collaboration with Dea-Sick Lim and Michael B. Kastan ...................... 67

ECK Protein Kinase as a Transcriptional Target of P53 in Signaling Apoptosis and Tumor SuppressionCynthia Y. Liu and Yuxin Yin..................................................................................................................................... 68

Overexpression of Betaig-H3 Gene Suppresses Tumorigenicity in Radiation Induced Tumorigenic HumanBronchial Epithelial CellsYong Liang Zhao, Chang Piao and Tom K. Hei ........................................................................................................... 69

Mutation(s) at Exon 3 of β-Catenin Preventing β-Catenin-GSK-3β Interaction: A possible Role in Radiation-Induced Breast Cancer ProgressionDebasish Roy, Gloria M. Calaf and Tom K. Hei .......................................................................................................... 72

STUDIES RELATED TO RADIOLOGY AND RADIATION THERAPY

Risks of Radiation-Induced Cancer from Pediatric CTDavid J. Brenner, Carl E. Elliston, Eric J. Hall and Walter E. Berdon........................................................................... 75

THE RADIOLOGICAL RESEARCH ACCELERATOR FACILITY

An NIH-Supported Resource Center (WWW.RARAF.ORG)Director, David J. Brenner, Ph.D., D.Sc.; Manager, Stephen A. Marino, M.S.; Chief Physicist, Gerhard Randers-Pehrson,Ph.D. .......................................................................................................................................................................... 77

THE RADIATION SAFETY OFFICE

Radiation Safety Office Staff .................................................................................................................................... 83Introduction .............................................................................................................................................................. 85Overview ................................................................................................................................................................... 85Summary of Services................................................................................................................................................. 86Itemized Services....................................................................................................................................................... 86

ACTIVITIES AND PUBLICATIONS

Professional Activities ............................................................................................................................................... 93Publications ............................................................................................................................................................... 95

COLLABORATING DEPARTMENTS AND INSTITUTIONS

Individuals from the following departments and institutions (listed alphabetically) collaborated with (l) or werementored by (n) Center for Radiological Research staff in the above research abstracts. For individual attribution seespecific reports:

Collaborating Columbia University Departments:

• Department of Environmental Health Sciences, JosephMailman School of Public Health.

• Department of Neurology.• Department of Pathology.• Department of Physiology and Cellular Biophysics.• Department of Radiology, Div. of Pediatric Radiology.• Institute of Cancer Genetics.

Collaborating Institutions:

• Illinois Institute of Technology, Il.• Institute for Cancer Research, Fox Chase Cancer Cen-

ter, Philadelphia, Pa.• Oregon Health Sciences University, Portland, Or.• Skirball Institute, New York University, N.Y., N.Y.

• Southern Urals Biophysics Institute, Ozyorsk, Russia.• St. Jude Children's Research Hospital, Memphis, Tn.• Technical University of Munich, Germany.• Terry Fox Laboratory, British Columbia Cancer

Agency, Vancouver, BC.• University of California, Berkeley, Ca.• University of California, San Diego, Ca.• University of Tarapaca, Arica, Chile.• University of Texas Medical Branch, Galveston, Tx.• University of Texas Southwestern Medical Center,

Dallas, Tx.• V.A. Medical Center, Pittsburgh, Pa.• Whitehead Institute for Biomedical Research, Cam-

bridge, Ma.

§ The Bronx High School of Sciences, New York, N.Y.

4

IntroductionThe Center for Radiological Research of Columbia Uni-

versity epitomizes a multidisciplinary approach toward anunderstanding, at both basic and applied levels, of the bio-logical effects of ionizing radiation as they relate to humanhealth.

The raison d’etre of the Center for Radiological Re-search of Columbia University is to foster a multi discipli-nary approach towards understanding the biological conse-quences of ionizing radiation exposures. The staff of TheCenter includes professionals from fields as diverse as mo-lecular biology, cell biology, radiation physics, computa-tional physics, engineering and radiation oncology.

This report gives a brief account of both research initia-tives and academic activities during the past year.

Perhaps the most exciting area of research continues toinvolve the single particle microbeam facility. This uniquefacility has made it possible to perform a number of projectsthat directly challenge some long-standing dogmas in radia-tion biology. For example, mutations are observed in cellswhere α-particles traversed only the cytoplasm, with nonepassing through the nucleus. In another area, the bystandereffect has been demonstrated unequivocally for a range ofbiological endpoints including chromosomal aberrations,cell lethality, mutation, oncogenic transformation and geneexpression. The bystander effect is defined to be a biologicaleffect in cells that are not themselves traversed by a chargedparticle, but that are close neighbors of cells that are. Thischallenges the traditional view that heritable effects requireddirect damage to DNA.

A publication in The American Journal of Roentgenol-ogy reported on the first realistic estimates of radiation-related cancer risks for children receiving helical CT scans.This paper published in a professional journal led to a front-page story in “USA Today” followed by interviews on NBCand ABC television. This publicity is likely to be beneficialsince The Society of Pediatric Radiology (amongst others)have subsequently held meetings and workshops, as well aspublishing recommendations on how to lower radiationdoses from helical CT scans in children.

We were pleased to be involved in organizing the “5th

International workshop on microbeam probes of cellularradiation response” that was held in Stresa, Italy, immedi-ately before the 13th symposium on microdosimetry. Wewere also privileged to organize a workshop, sponsored byNIH, and entitled, “Probing individual cells: applications tosignaling, structure and function.”

The productivity of the Center continues at a high level,as evidenced by a steady stream of scientific papers in peer-

reviewed journals, including several in high profile journals.Members of staff are frequently invited to participate in na-tional and international meetings, and are frequently calledupon to serve as consultants, reviewers or site visitors bygovernment agencies.

The teaching activities of the Center include the teachingof Radiation Biology and Radiation Physics to undergradu-ates, medical students, and graduate students in the Schoolof Public Health, and residents in both Radiology and Ra-diation Oncology. n

We mourn the passing during the year of JulianaDenekamp, known to all her friends as “Julie.” In fact,Julie and I were both born in Wales, only a few miles(and a few years) apart – but we never met until we werein London, many years later. The photograph was takenat a microdosimetry meeting, while Julie was Director ofThe Gray Laboratory and I had recently been appointedas Director of The Center for Radiological Research.Julie was a frequent visitor to New York and held thetitle of Adjunct Professor of Radiation Oncology at Co-lumbia. We sadly miss her collaboration and most deeplyregret her passing at such an early age from the scourgethat she had devoted her professional career to studying.

Eric J. Hall

In Memoriam


5

CENTER FOR RADIOLOGICAL RESEARCHFACULTY AND STAFF

Front row (l-r): Ms. Monique Rey, Dr. Tom Hei, Dr. Charles Geard, Dr. Eric Hall, Dr. David Brenner,Dr. Howard Lieberman, Ms. Mary Coady.

2nd row: Dr. Yuxin C. Liu, Ms. Sonu Dhar, Dr. Su-Xian Liu, Ms. Xiaojian Wang, Dr. Gloria Calaf, Dr.Aiping Zhu, Dr. Alexander Dymnikov, Dr. Alan Bigelow, Ms. Heidy Hernandez, Ms. Diana Morri-son.

3rd row: Dr. Tej Pandita, Mr. Gary Johnson, Dr. Jaime Rubin, Mr. Robert Archigian, Mr. David Cu-niberti, Mr. Carl Elliston, Ms. Gloria Jenkins-Baker, Dr. Gerhard Randers-Pehrson, Dr. GirdharSharma, Mr. Kevin Hopkins, Dr. Adayabalam Balajee.

Back row: Dr. Haiying Hang, Dr. Brian Ponnaiya, Dr. Li Liu, Mr. Moshe Friedman, Dr. Debasish Roy,Dr. Yuxin Yin, Dr. Prakash Hande, Dr. Arun Gupta, Dr. Hongning Zhou, Dr. Lubomir Smilenov,Mr. Mutian Zhang, Mr. Stephen Marino, Dr. Yong-Liang Zhao, Dr. Chang-Q. Piao.


6

CENTER FOR RADIOLOGICAL RESEARCHFACULTY AND STAFF

RESEARCH STAFF

ERIC J. HALL, D.Phil., D.Sc., FACR, FRCR, HigginsProfessor of Radiation Biophysics, Professor ofRadiology and Radiation Oncology (Director)

CHARLES R. GEARD, Ph.D., Professor of ClinicalRadiation Oncology (Associate Director)

DAVID J. BRENNER, Ph.D., D.Sc., Professor ofRadiation Oncology and Public Health(Environmental Health Science) (Director ofRARAF)

TOM K. HEI, Ph.D., Professor of Radiation Oncologyand Public Health (Environmental Health Science)

HOWARD B. LIEBERMAN, Ph.D., AssociateProfessor of Radiation Oncology

HAIYING HANG, Ph.D., Assistant Professor ofRadiation Oncology

TEJ K. PANDITA, Ph.D., Assistant Professor ofRadiation Oncology

YUXIN YIN, M.D., Ph.D., Assistant Professor ofRadiation Oncology

GERHARD RANDERS-PEHRSON, Ph.D., ResearchScientist

ADAYABALAM BALAJEE, Ph.D., AssociateResearch Scientist

GLORIA CALAF, Ph.D., Associate Research ScientistM. PRAKASH HANDE, Ph.D., Associate Research

ScientistBRIAN PONNAIYA, Ph.D., Associate Research

Scientist SATIN G. SAWANT, Ph.D., Associate Research

Scientist (until Feb. 2001) LUBOMIR SMILENOV, Ph.D., Associate Research

ScientistYONG-LIANG ZHAO, Ph.D., Associate Research

ScientistKEVIN HOPKINS, M.S., Senior Staff AssociateSTEPHEN A. MARINO, M.S., Senior Staff AssociateCHANG-QING PIAO, M.D., Senior Staff AssociateCARL ELLISTON, M.S., Staff AssociateSONU DHAR, B.S., Staff AssociateSU-XIAN LIU, M.D., Staff AssociateAN XU, B.S., Staff AssociateHONGNING ZHOU, Ph.D., Sr. Staff AssociateAIPING ZHU, M.D., Staff Associate

POST-DOCTORAL FELLOWS

ALAN BIGELOW, Ph.D., Post-Doctoral ResearchScientist

JAIN FEN GUO, Ph.D., Post-Doctoral ResearchScientist (until June 2001)

ARUN GUPTA, Ph.D., Post-Doctoral ResearchScientist

YUXIN C. LIU, Ph.D., Post-Doctoral ResearchScientist

DEBASISH ROY, Ph.D., Post-Doctoral ResearchScientist

GIRDHAR SHARMA, Ph.D., Post-Doctoral ResearchScientist

MASAO SUZUKI, Ph.D., Post-Doctoral ResearchScientist (until Feb. 2001)

VISITING SCIENTISTS

ALEXANDER DYMNIKOV, Ph.D., Radiation SafetyPhysicist

DESIGN AND INSTRUMENT SHOP

GARY W. JOHNSON, A.A.S., Senior Staff Associate(Director)

DAVID CUNIBERTI, B.A., Instrument MakerROBERT ARCHIGIAN, Instrument Maker

TECHNICAL STAFF

JOSE GARCIA, Laboratory Assistant (retired Dec.2001)

GLORIA JENKINS-BAKER, B.A., Research WorkerXIAOJIAN WANG, M.S., Research WorkerMUTIAN ZHANG, B.A., Technician BRONALD BAKER, B.S., Sr. Technician

ADMINISTRATIVE AND SECRETARIAL STAFF

MONIQUE REY, B.A., Center AdministratorMARY COADY, Administrative CoordinatorMOSHE FRIEDMAN, B.A., Administrative AssistantHEIDY HERNANDEZ, Jr. AccountantDIANA MORRISON, Administrative Assistant


7

Center for Radiological Research Staff NewsDr. Eric Hall is currently President of the International

Association of Radiation Research. He continues to serve asSenior Biology Editor of the International Journal of Ra-diation Oncology, Biology, Physics.

Drs. Hall and Brenner are both Councilors of the Na-tional Council on Radiological Protection. Both continue toserve on Committee 1-6 of NCRP on linearity, which pub-lished NCRP report 13 during the past year.

Dr. Hall has been awarded the RSNA World Wide Web-Based Educational Program Grant from the RadiologicalSociety of North America Research and Education. Thisgrant is designed to provide an opportunity for scientists andphysicians in the radiological sciences to develop educa-tional materials specifically for widespread distribution viathe Internet. The subject of Dr. Hall’s project is “Web-BasedEducational Program for Diagnostic and InterventionalRadiologists: Radiobiology, Radiation Protection, and Risksvs. Benefits.”

Dr. Tom K. Hei finished his 4-year term as a member ofthe Chemical Pathology Study Section in June and has beenasked to serve as an ad hoc member. He also chaired severalother NIH ad hoc reviews and participated in NCI intramuralreview.

Dr. Hei’s recent study on the role of oxyradicals in thegenotoxicity of arsenic, published in the February 13th issueof the Proceedings of the National Academy of Sciences,was featured in Research Highlights from the National In-stitute of Environment Health Sciences as well as in theJournal of P & S.

Miss Maris Kessel, a high school senior from the BronxSchool of Sciences under the mentorship of Professor TomK. Hei, successfully competed in the Intel Science TalentSearch with her project entitled “Induction of OxidativeDNA damage by Arsenic in Mammalian Cells.” She won asemifinalist title as well as a $2,000 scholarship.

Dr. Hei’s study on the biological consequence of cyto-plasmic irradiation was recently featured in the magazineLife Extension. In addition, his recent finding on single alphaparticle induction of a bystander mutagenic effect, published

in the December issue of the Proceedings of the NationalAcademy of Sciences, was accompanied by a press releasearticle from the Journal of P&S.

Dr. Howard B. Lieberman has accepted an invitation toserve as a Member of basic and preclinical Subcommittee Cof the NCI Initial Review Group. He also continues to serveas a Member of Scientific Review Panel “A” of the IsraelCancer Research Fund, a private organization that supportsbiomedical research related to cancer.

Dr. Richard Gewanter, Chief Resident in Radiation On-cology, completed a laboratory research elective under thementorship of Dr. Charles Geard. Dr. Gewanter has accepteda staff physician appointment in the Department of Radia-tion Oncology.

The picture of telomeric signals at chromosomes ends onthe cover of Human Molecular Genetics (Vol. 10, No. 5)2001, was contributed by Dr. M. Prakash Hande (see repro-duction at the end of the Publications section). n

(L to R): Dr. Charles Geard, Dr. Tom Hei and Dr. RichardGewanter.

Dr. Hall with office staff (L to R): Monique Rey, HeidyHernandez, Dr. Eric Hall, Mary Coady and Diana Morrison.

(L to R): Dr. M. Prakash Hande, Dr. David Brenner, Dr.Tamara Azizova, Olga Danilova, Dr. Ludmilla Burak and Dr.Charles Geard (Southern Urals Biophysics Institute, Ozyorsk,Russia).


8

The Columbia Colloquium and Laboratory SeminarsAt intervals of approximately one month during the aca-

demic year, a regular colloquium has been held to discussongoing research. Dr. Howard Lieberman organized themand scheduled the speakers. These were attended by the pro-fessional staff, graduate students, and senior technical staffof this Laboratory and RARAF, as well as scientists fromother departments of the College of Physicians & Surgeonsinterested in collaborative research. Attention has focused onrecent findings and future plans, with special emphasis onthe inter-disciplinary nature of our research effort.

During the year, we have been pleased to welcome a

number of visitors who have presented formal seminarsand/or spent time discussing ongoing research with variousmembers of the Laboratory. These have included Drs. Du-bravka Krilov, University of Zagreb, Croatia; Carmel Moth-ersill, Dublin Institute of Technology, Ireland; Frank Ellis,Consultant Emeritus, Oxford, UK; Alexandra Miller, ArmedForces Radiobiology Research Institute, Bethesda, Md.;Maria Jasin, Memorial Sloan-Kettering Cancer Center, NewYork; Kenshi Komatsu, Hiroshima University, Japan; Man-zoor A. Bhat, Mount Sinai School of Medicine, New York;and Yossi Shiloh, Tel Aviv University, Israel. n

Web Sites• Department of Radiation Oncology ................. http://cpmcnet.columbia.edu/dept/radoncology• Center for Radiological Research .................... http://cpmcnet.columbia.edu/dept/radoncology/crr• Radiological Research Accelerator Facility ..... http://www.raraf.org• Radiation Safety Office ................................... http://cpmcnet.columbia.edu/dept/radsafety n

AcknowledgmentsResearch at the Center for Radiological Research, College of Physicians & Surgeons of Columbia Univer-sity, is supported by competitively awarded grants from:

♦ American Cancer Society.

♦ Ataxia Telangiectasia Children's Project.

♦ Avon.

♦ Department of Defense (US Army).

♦ Department of Energy, Low Dose Radiation Research Program, Biological & Environmental Research.

♦ Department of Health and Human Services, National Institutes of Health:

• National Cancer Institute (Program Project (PO1) and Individual Research Grants (RO1s)).

• National Center for Research Resources (P41).

• National Institute of Environmental Health Sciences (RO1s).

• National Institute of General Medical Sciences (RO1).

• National Institute of Neurological Disorders and Stroke (RO1).

♦ Herbert Irving Comprehensive Cancer Center of Columbia University.

♦ National Aeronautics and Space Administration.

♦ Radiological Society of North America. n


9

Laser Ion Source Development for theColumbia University Microbeam

Alan W. Bigelow, Gerhard Randers-Pehrson and David J. Brenner

Columbia University’s Radiological Research Accelera-tor Facility (RARAF) conducts fundamental investigationsinto the radiobiological effects on mammalian cells throughcontrolled single-particle single-cell microbeam irradiation.The ion source that is currently used on our 4.2 MV Van deGraaff particle accelerator, a duoplasmatron, ionizes atomsfrom the gaseous phase and is suitable for alpha particle ir-radiation experiments (1). However, to extend the linearenergy transfer (LET) range of our experiments, highlycharged heavy ions are necessary. Expectations are that alaser ion source will enable a range of ions from hydrogen toaround iron with an approximate LET range of 10 to 4,500keV/m.

Laser ion sources have been developed and are used atseveral particle accelerator laboratories (2). Many of theseion sources share a common mechanism of plasma genera-tion through laser ablation of a solid target. Particles areevaporated from a target using a focused high-energy pulsedlaser. Plasma electrons are heated by the laser radiation totemperatures up to several hundreds of eV. High chargestates are produced by electron-ion collisions. The tempera-ture of the plasma and the consequent final ion charge-statedistribution strongly depend on the laser power density onthe target (3). A useful trait common among contemporarylaser ion sources is the directional nature of the plasmaplume; ion extraction is preferred along the direction normalto the target.

The laser ion source development at RARAF is based ona prototype built from components of the laser operated ionsource (LOIS) used by Hughes at the University of Arkansas(4). The prototype construction involved reviving and opti-mizing a 1970’s Holobeam 5050Q Nd:YAG laser and as-sembling a vacuum system component containing the ionsource target material, a cylindrical electrostatic analyzer(ESA), two Einzel lenses (ion optical focusing elements),and a microchannel plate (MCP) particle detector. In orderto attain proof of principle while maintaining result compari-son capabilities, the prototype construction followed specifi-cations for the most recent version of LOIS (5). A schematicof the laser ion source prototype is shown in Fig. 1.

The laser beam enters the vacuum system through a win-dow and passes through a lens (focal length f = 12.2 cm) andimpinges on the target at an 8.5-degree angle of incidence.In the Q-switched mode and with two stages of amplifica-tion, specifications for the Holobeam laser indicated maxi-mum energy E = 850 mJ, pulse duration t = 15 ns, andwavelength λ = 1.064 µm (6). Laser spot scores on black-ened paper provided a beam diameter measurement D = 0.38cm. From optics theory, one can calculate the focused beamspot diameter d = 1.22 λf/D = 42 µm. This leads to a power

density idealization of 4.1 x 1012 W/cm2.A motor continually turns a cylindrical target at six min-

utes per revolution so that each laser pulse strikes a freshsurface. The diameter of the target is 6.15 cm and has a 19.3cm circumference. The diameter of an ablation crater inaluminum is 0.25 mm. Hence, there is room for 770 individ-ual craters about the target circumference. For one targetrotation, the system frequency is 770/(6 min) = 2.1 Hz. Thelaser repetition rate is set to this frequency. After one targetrotation, the target is translated 0.25 mm.

Plasma expansion occurs over a drift distance of 70 cm;this distance fits within the range of typical drift distancesfor laser ion sources according to Sharkov (3). Furthermore,the drift distance leads to an increased temporal ion pulsewidth and to a reduction in subsequent plasma effects, suchas, plasma shielding and arcing, when the ions enter anelectrostatic field region (3,7). Afterwards, the plasma plumeenters an electrostatic analyzer consisting of a pair of cylin-drical plates that bend the ions through 180o. The analyzer,tuned for energy per charge, will reduce considerably thebeam load on the accelerator vacuum system. For infinitelytall coaxial cylinders, electrostatic theory states that the ionsare analyzed by kinetic energy E = kVAz where k is theanalyzer constant ~5.62 eV/V, VA is the voltage across theanalyzer and z is the ion charge state (8). Next along the ionoptical path, two Einzel lenses are available to guide iontrajectories towards the MCP particle detector. The MCPaxial position matches the eventual position of the particleaccelerator entrance aperture. Attached to the MCP are aseries of signal detection electronics and a computer thatruns a digitized waveform acquisition program in LabVIEW(9).

With an operational laser ion source, research was

Fig. 1. Schematic of laser ion source prototype used todemonstrate proof of principle.


10

steered towards the intricate details of laser-induced plas-mas. Ion signal production does depend on target surfacecondition and laser power density. There was a significantreduction in high charge state yield when plasma was gener-ated from a target surface that had been ablated. So, for highcharge state production, it is important to coordinate targetadvancement with laser ablation. As laser power densityincreased, the average charge state increased. At lower laserpower densities, for instance, there is a peak for Al+1; it dis-sipates in favor of higher charge states as laser power den-sity increases.

Critical plasma details for the laser ion source design in-clude the energy and angle distribution of the most highlycharged ions. It is necessary to know where the greatestabundance of these ions resides in order to guide them to theentrance of the particle accelerator. Hughes reported a radialdistribution in an extracted laser-plasma ion beam where thefastest ions, also those with the highest charge states, resideon the outside of the beam (4); space-charge-repulsion ef-fects were used to explain the trend. Data generated from thelaser ion source prototype, however, suggest that the greatestabundance of the highly charged ions was slower (less en-ergy) and was concentrated along an emission directionnormal (perpendicular) to the target material. This wasachieved using magnetic coil deflectors to sample the angu-lar emission of the plasma plume.

Ion trajectories through the laser ion source prototypewere simulated with an ion optics computer program,SIMION (10). The simulation consisted of virtual ion opticalcomponents arranged on an ion optics workbench. Ionsflown through the system provided ion trajectory data usefulin characterizing the source emittance. For a confidencecheck, TOF values from these data compared well withkinematic theory.

Ideally, the laser ion source emittance should match wellwith the 1/8-inch diameter entrance aperture to the particleaccelerator; this calls for an ion optical system with point-to-point focusing. Use of a cylindrical ESA followed by twoEinzel lenses prohibited this condition. Voltage configura-tions on these lenses could focus the ions in either the verti-cal or the horizontal plane, but not simultaneously.

The laser ion source development for the Columbia Uni-versity microbeam incorporates a spherical ESA. This spe-cific type of analyzer has double focusing capabilities withpoint-to-point focusing in both the horizontal and verticalplanes. Spherical ESA theory and fringing field effects arewell documented in Wollnick’s treatment of electrostaticprisms (11). Guided by spatial limitations in the particle ac-celerator and by a desired plasma expansion drift distance of70 cm the ESA dimensions were narrowed to a 24-degreebend with a 2.7-inch radius. This geometrical solution wasfound by applying Barber’s rule; the object point, the centerof curvature, and the image point lie on a straight line (12).A diagram of the laser ion source inside its spatial confinesis shown in Fig. 2.

The laser ion source to be implemented on the ColumbiaUniversity microbeam will include a contemporary laser.Factors important to the laser purchase decision are powerand repetition rate requirements. With today’s laser technol-

ogy, pulsed lasers still provide greatest power. As for repeti-tion rate, the desired experimental error of controlled particleirradiation restricts one particle irradiation per 100 lasershots. Repetition rates greater than 1000 Hz would keep thelaser pulsing frequency from being a rate limiting step.Costly commercial Ti:Sapphire lasers can meet both theserequirements. Budget friendly commercial Nd:YAG lasersprovide the power requirement, but, they are limited to 100Hz repetition rates. Combinations of multiple Nd:YAG la-sers, synchronized with temporal offsets could reduce arepetition rate handicap.

Outlining the laser ion source implementation, a 100 HzNd:YAG laser will be mounted in front of the Van de Graaffaccelerator, parallel to and alongside the charged particlebeam line. The light beam from the laser will enter the baseof the accelerator through an existing window. Inside theaccelerator, the light will pass through one of the insulatingsupport tubes to the terminal where it will be directed towardthe window of the ion source vacuum chamber. With har-monic generation, a variety of laser wavelengths (1064, 532,and 355 nm) leads to flexibility concerning potential at-tenuation in the accelerator’s insulating gas and wavelength-dependent ablation yields (13). The insulating gas is pres-ently a mixture of N2-CO2 at 10 atm. and it is possible thatSF6 will be used in the future. Ultimately, the vacuumchamber component to the laser ion source will be a modularunit, interchangeable with the duoplamatron gas ion source.

References

1. Randers-Pehrson G, Geard CR, Johnson G, Elliston CDand Brenner DJ, Radiat. Res. 156:210-214, 2001.

2. Fournier P et al., Rev. Sci. Instrum. 71:1405, 2000.3. Sharkov B, in Handbook of Ion Sources, edited by Wolf

B (CRC Press, Boca Raton, 1995) p. 149.4. Hughes RH et al, J. Appl. Phys. 51:4088, 1980.5. Miller RD, Wattuhewa G, Hughes RH, Pederson DO,

and Ye XM, Phys. Rev. B, 45:12019-12027, 1992.6. Miller RD, PhD thesis, The University of Arkansas,

1990.7. Brown IG, in The Physics and Technology of Ion

Sources (Wiley, New York, 1989) pp. 11-12.8. Ye XM, PhD thesis, The University of Arkansas, 1986.

AcceleratorColumn

PlasmaPlume

LaserTarget

LaserOptics

SphericalElectrostaticAnalyzer

Laser Beam

Fig. 2. Top view diagram of the proposed laser ion source in-side the particle accelerator terminal dome.


11

9. National Instruments Corporation, 11500 N Mopac Ex-pwy, Austin, TX 78759-3504

10. Idaho National Engineering and Environmental Labora-tory, Idaho Falls, ID 83415

11. Wollnik H, in Focusing of Charged Particles, edited bySeptier A (Academic Press, Inc., Orlando, 1967) pp. 163-

202.12. Wollnik H, in Optics of Charged Particles (Academic

Press, Inc., San Diego, 1987) p. 98.13. Kools JCS, in Pulsed Laser Deposition of Thin Films,

edited by Chrisey DB and Hubler GK (Wiley, NewYork, 1994) p. 457. n

Are there Geometric Associations betweenDifferent Human Chromosomes?

David J. Brenner, in collaboration with Karin M. Greulich-Bode and Martina Brückner,1

Michael N. Cornforth and Bradford Lucas,2 and Rainer K. Sachs3

There has been some attention paid to the question ofwhether particular chromosomes are more or less sensitivethan others to radiation. Typically this is done by countingthe number of times a particular chromosome is involved inan aberration, and taking into account the relative DNAcontent of that chromosome. The results have been contro-versial, with some hints thatthe larger chromosomes maybe slightly less sensitive thanaverage (after factoring inchromosome size), and thatsmaller chromosomes may bemore sensitive than average.Others, however, have main-tained that there is no system-atic variation in average inter-chromosomal sensitivity -though there are clearlyvariations in sensitivity withinchromosomes.

Here, whilst also address-ing this first question, we ask a rather different, independent,question: Are there systematic variations in the pairwisegeometric association between particular chromosomes? Putanother way, are any specific pairs chromosomes that are, onaverage, preferentially located either nearer or further fromone another than would be expected if they were randomlylocated in the nucleus? We approach this question by askingif radiation damage in a particular chromosome is morelikely than average to interact with damage from some otherparticular chromosome. If so, this would most likely reflect ageometric association (or anti-association) between thosetwo chromosomes.

We analyze here data from two different groups (Techni-cal University of Munich, University of Texas Medical

1 Technical University of Munich, Germany.2 University of Texas Medical Branch, Galveston, Tx.3 University of California, Berkeley, Ca.

Branch), who measured the yields of exchange-type chro-mosome aberrations between each pair of chromosomes inthe human genome (276 pairs for females, 253 pairs formales, 231 autosome pairs). Specifically the data were gen-erated using the multiplex-FISH (mFISH) technique, inwhich each chromosome is “painted” a different color

through in-situ hybridization,permitting measurement ofthe number of metaphase cellswhich have a particular chro-mosome (color) pair interact-ing, as indicated by the rele-vant mFISH color junctions(see, for example, Fig. 1). Thenull hypothesis of the study isthat the pair frequencies arerandom; specifically, “ran-dom” here means that, aftercorrecting for the differentoverall radiation sensitivitiesof the different chromosomes,

a damaged chromosome is equally likely to interact with anyother damaged chromosome.

Nine sets of mFISH data were used to investigate ran-domness of chromosome aberration formation in humanlymphocytes subjected to sparsely ionizing radiation in vitro;these data sets were for 7 different individuals (2 female, 5male) whose blood was irradiated with x or γ rays to dosesof 2, 3, or 4 Gy. We were able to combine the results for allthe individuals and all the doses, because statistical testing(Kruskal-Wallis singly-ordered test) on the nine data sets(each containing 276 ordered data points [253 for females])gave no evidence that these 9 ordered data sets were notidentically distributed, one from another (of course in com-paring the results between males and females, only the auto-some data were used).

Being able to combine these various data sets allows usto draw conclusions which may be relevant for all humanindividuals, and also significantly increases our statistical

Fig. 1.


12

power. Some results are shown in Fig. 2. The figure indi-cates those pairs of chromosomes which interact more fre-quently than would be expected if the chromosome-chromosome interactions were purely random. The pointsrepresent the 22 autosomes, and the line widths are propor-tional to the deviation from randomness for the given pair ofautosomes. Large deviations are indicated by solid lines,smaller deviations by dotted lines, and those chromosomepairs for which there was no significant increase over ran-domness are shown as not connected.

If we place into the same cluster all the chromosomesconnected, directly or indirectly, by solid lines (a techniquecalled sparsification), the resulting groupings are(9,16,18,19,20); (1,2,15); (6,7,12); and (8,13) with the re-maining eight chromosomes not clustered. Because χ2 tests,with high statistical power, suggest overall randomness forthe pairwise associations, these clusters are to be regardedwith caution. For example, in some scenarios genuine geo-metric associations would be expected to show transitivity(if A is close to B and B is close to C then A should be closeto C) while the diagram gives no indications of transitivity.

But if there is any clustering the indicated combinations areamong the most likely candidates.

In summary it was found that pair participation does notshow statistically significant deviations from a random dis-tribution. If there are geometric associations of differentchromosomes in interphase human lymphocyte nuclei, theassociations are too weak, and/or are too variable from cellto cell, and/or involve DNA regions too small to show upsignificantly as radiation-induced interchanges (i.e. at thewhole-chromosome level with any two homologues indis-tinguishable). We have, however, derived candidate chromo-some clusters, relevant if there are nonetheless some weakgeometric associations.

By summing the results for pairs we also investigated theindividual sensitivity of different chromosomes. As illus-trated in Fig. 3, there was a tendency for large chromosomesparticipate less frequently than expected from Monte Carlocomputer simulations based on DNA content, randomness,and the standard breakage-and-reunion aberration formationmodel, and for small chromosomes to participate more fre-quently. n

Biomarkers Specific to Densely-Ionizing(High-LET) Radiations

David J. Brenner, Nadia Okladnikova,1 M. Prakash Hande, Ludmilla Burak,1 Charles R. Geard and Tamara Azizova1

The overall goal of this research is to test the hypothesisthat a specific chromosomal biomarker exists and is detect-able for past exposure to high-LET radiation such as pluto-nium alpha particles, even in the presence of other mutagensor clastogens, such as tobacco, organic chemicals, or gammarays. Such a biomarker would significantly increase the

1 Southern Urals Biophysics Institute, Ozyorsk, Russia.

power of epidemiological studies of individuals exposed todensely-ionizing radiations such as alpha particles (e.g., ra-don, Pu workers) or neutrons (e.g., DOE/NRC workers, air-line personnel).

The proposed biomarker is the ratio of induced intra-armto inter-chromosomal aberrations – the H value. In otherswords, the prediction is that densely ionizing radiations suchas alpha particles or neutrons will give a uniquely high yield

Fig. 2.

0

50

100

150

200

250

300

350

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

chromosome

part

icip

atio

ndatatheory

Fig. 3.


13

of intra-arm exchanges (paracentric inversions), relative toother mutagens / clastogens such as x-rays or chemicals.Both theoretical and experimental studies have suggestedthat this ratio will be significantly different between l highLET radiation and any other mutagen /clastogen. Subsequentto our initial (1997) predictions, three groups (Munich,Livermore and Columbia) have all validated this LET fin-gerprint in vitro, though for unstable chromosome aberra-tions, which are of limited practical utility.

A key aspect of this research is that, only in the last twoyears, have efficient techniques become available for meas-uring stable intra-arm exchange aberrations (MBAND). Weare one of the first groups worldwide, to use this technique.

In collaboration with the Southern Urals Biophysics In-stitute in Ozyorsk, we are making FISH-based H-value

measurements (in particular paracentric inversions) on pe-ripheral blood lymphocytes from individuals in the Mayakworker cohort who were exposed to a wide range of pluto-nium exposures, gamma ray exposures, and chemical car-cinogens. The large number of individuals involved, togetherwith the broad spectrum of exposures, means that this cohortoffers a unique opportunity for rigorous hypotheses testing.We are using:

1) DAPI analysis, in which metaphase spreads werestained with DAPI and aberrations are scored to determinethe baseline frequency of chromosome aberrations, mainlyasymmetric exchanges.

2) Fluorescence in situ hybridization (FISH) analysis,using arm-specific FISH probes, in which frequencies ofintra-chromosomal exchanges (pericentric inversions) andinter-chromosomal exchanges (dicentrics and translocations)are estimated using two color FISH with chromosome-arm-specific probes.

3) mFISH (multi-colored FISH) analysis, in which 24-color FISH (painting each human chromosome in differentcolor) is performed to detect complex aberrations induced byexposure to ionizing radiation.

4) mBAND (multicolor banding FISH) analysis, inwhich high resolution multicolor banding FISH is beingstandardized to detect simple and complex aberrations in-cluding peri- and, central to this project, paracentric inver-sions.

So far we have samples from a total of 30 irradiated in-dividuals and 7 controls (see Table I).

To date, we have analyzed 4 individuals with either highor intermediate plutonium exposures, and have just startedanalyzing low-plutonium/high γ individuals. All analyses arebeing done both with mFISH and mBAND. The results areshown in Table II.

Table I.

Yearof

birthSex Culture

number

Externalradiation

dose,cGy

Pu-239body

burden,nCi

1925 f 8288 412.9 1871927 m 8293 200 2481926 f 8295 522.54 51919 f 8296 507.06 01927 m 8298/7 466.87 01926 m 8261 423.1 5021931 m 8259 312.5 1651924 m 8266 367.91 1601923 f 8302 345 4921937 m 8303 247.6 2841925 m 8268 222.1 01926 m 8273 310.9 01924 m 8274 226.82 01930 m 8275 244.13 01924 m 8276 246.59 01928 m 8277 246.48 01928 m 8278 318.14 01931 m 8279 206.76 01936 m 8282 57.38 3631933 f 8283 78.02 438.21925 f 8284 58.91 3051926 f 8285 63.62 4521933 m 8287 273.97 01919 m 8258 22.25 429.01927 f 8252 0 404.01924 M 8253 27.61 10.01934 M 8254 37.816 -1939 M 8255 42.75 41.01938 M 8256 27.61 60.01939 M 8257 9.93 23.01937 82371924 82381978 8239 0 01958 8240 0 01926 8242 0 01957 8244 0 0

Table II.mFISH results:

Sam

ple

ID

No.

of c

ells

anal

yzed

Rec

ipro

cal

Non

-re

cipr

ocal

Com

plex

Dic

entr

ic

Frag

men

ts

Oth

erFu

sion

s

8252 105 10 6 1 2 12 08258 105 6 6 3 1 2 08256 109 5 4 1 2 3 08257 110 3 6 1 0 1 4

mBAND results:

Sam

ple

ID

No.

of c

ells

Para

cent

ric

inve

rsio

n

Peri

cent

ric

inve

rsio

n

Com

plex

inve

rsio

n

Inse

rtio

n

Inte

rstit

ial

dele

tion

Ter

min

alde

letio

n

8252 115 3 1 1 1 2 28258 116 1 0 0 1 1 18256 112 2 0 0 1 5 18257 113 2 0 1 0 8 3


14

The key finding in our results so far is the major excess(p <0.01) of paracentric inversions (intra-arm) relative topericentric inversions (inter-arm) for high-LET exposed in-dividuals, as shown above. Specifically, we have found 8paracentric inversions and only one pericentric inversion.This reflects the mechanistic underpinning for the currentstudy, i.e., that DNA double strand breaks that occur closetogether (e.g. on the same chromosome arm) are more likelyto result in an exchange than when the two breaks are furtherapart (e.g. on different arms). While we had predicted thistheoretically for high-LET radiation, and this effect has sub-sequently been seen in-vitro for unstable aberrations (ofcourse, unstable aberrations are of less interest for retro-spective exposure assessment), this is the first experimentaldemonstration for stable aberrations, and the first from anirradiated human population (see Fig. 1).

Thus the evidence so far is that this yield of paracentricinversions is indeed the “fingerprint” which we had hy-pothesized. While we are only in the preliminary phases ofour analyses of individuals exposed only to gamma rays, wehave not so far seen any paracentric inversions and, theoreti-cally, we expect to see very few. n

Do Low Dose-Rate Bystander EffectsInfluence Domestic Radon Risks?

David J. Brenner, in collaboration with Rainer K. Sachs1

Radon1risks derive from exposure of bronchio-epithelialcells to high-LET alpha particles. Alpha particle exposurecan result in bystander effects, where irradiated cells emitsignals resulting in damage to nearby unirradiated bystandercells. This can result in non-linear dose-response relations,and inverse dose-rate effects. Domestic radon risk estimatesare currently extrapolated from miner data which are at bothhigher doses and higher dose rates, so bystander effects onunhit cells could play a large role in the extrapolation ofrisks from mines to homes. We therefore extend an earlierquantitative mechanistic model of bystander effects to in-clude protracted exposure, with the aim of quantifying thesignificance of the bystander effect for very prolonged expo-sures.

A model of high-LET bystander effects, originally de-veloped to analyze oncogenic transformation in vitro (2), hasbeen extended to low dose rates. The model considers radia-tion response as a superposition of bystander and linear di-rect effects. It attributes bystander effects to a small sub-population of hypersensitive cells, with the bystander contri-bution dominating the direct contribution at very low acutedoses but saturating as the dose increases. Inverse dose-rate 1 University of California, Berkeley, Ca.

effects are attributed to replenishment of the hypersensitivesubpopulation during prolonged irradiation. The essentialfeatures of the approach are summarized in Figure 1.

The model was fitted to dose- and dose-rate dependentradon-exposed miner data (2), and gives a reasonable fit tothe data (Fig. 2). Parameters from the fit suggest that onedirectly-hit target bronchio-epithelial cell can send bystandersignals to about 50 neighboring target cells.

The estimated parameter values from this fit were used toextrapolate the miner data to lower doses, and for a 60-yearexposure period. The results are shown in Fig. 3: For thecomparatively short miner exposures (solid curve; for illus-trative purposes, we use a duration of 6 y, the average timeof miner exposure in the data), the dose-response relation islinear at very high doses (where the direct effect dominates).It can be seen, however, that at intermediate doses, wherethe bystander response starts to become important, the 6-yrexposure (solid) curve become non linear and curves down-wards. At these intermediate doses the risks from a 6-yr ex-posure (dashed line) are larger than for a 60-yr exposure(solid line) – the inverse dose-rate effect. At still lowerdoses, dose rate effects become small, so the 6-yr exposureand the 60-yr-exposure produce the same risk.

Figure 3 also shows a linear extrapolation of the miner

Fig. 1. A paracentric inversion from individual 8252.

8252(exposed)

Paracentric inversion (Complex)

True color(merged color)

Pseudocolor

Normalchromosome 5

Paracentric inversion (Complex)


15

0 10 20 30Duration (y)

50-100 WLM


100-200 WLM


>800 WLM


400-800 WLM


0

50

100

150

Exc

ess

Rel

ativ

e R

isk 200-400 WLM


0

5

10

15

Exc

ess

Rel

ativ

e R

isk < 50 WLM

100

150

Exc

ess

Rel

ativ

e R

isk

data in which the effects ofdose rate are ignored. It canbe seen that ignoring dose-rate effect and simply using alinear extrapolation from theminer to the domestic situa-tion would result (using ourestimated parameters) in anunderestimation of the low-dose radon risk by about afactor of 4.5. This underes-timation is comparable to thecorresponding empiricallyestimated factor in the BEIR-VI report of ~3.7.

For comparison, alsoplotted in Fig. 3 are resultsfrom various domestic radoncase-control studies. Thespread and uncertainties ofthe results are such that theyare consistent with both thecurrent mechanistically-based low-dose-rate / low-dose extrapolation, theBEIR-VI phenomenologicallow-dose-rate / low-doseextrapolation, and also the“naïve” low-dose extrapola-tion from miner data whichignores the effect of doserate. It is important, however,to note that these data typi-cally represent above-average cumulative radonexposures, and that, assum-ing low-dose linearity, mostradon-related deaths will beat still lower cumulative ex-posures.

The main conclusions ofthis analysis are as follows:1. At high doses, the modelpredicts saturation effects and inverse dose-rate effectsin the bystander response. At sufficiently low doses, inagreement with general microdosimetric arguments, thepredicted response is linear in dose and independent ofdose-rate.2. Parameter estimates based on applying the model todose- and dose-rate dependent miner data suggest that asingle directly-hit target bronchial basal cell can sendbystander signals to about 50 neighboring cells.3. The model parameter values obtained from thisanalysis of epidemiological data, in as much as they canbe compared with parameter values obtained from in-vitro analyses, are significantly different. Thus modelparameters estimated from analysis of in-vitro studiescannot necessarily be applied to the in-vivo situation.4. The high-dose saturation and inverse dose-rate ef-

Fig. 1. Cartoon illustrating the main results regarding the interplay of risk between dose anddose rate: The small boxes represent collective, supra-cellular targets, defined by the property that ahit on any target cell nucleus in the collective target results in bystander signal to all cells in thatcollective target. Our estimates suggest ~50 target cells / collective target, but, for clarity, each col-lective target is shown as containing just two cells. In a few cases, a collective target may contain ahypersensitive cell, shown here as solid. The average number of alpha particle hits is labeled D inthis cartoon to emphasize its proportionality to dose, and the bystander response – number of hithypersensitive cells, is labeled R.

A. Panel represents a dose which is “very low” in the sense that most collective targets are nothit, and the chance for two alpha particles in one collective target is negligible. The bystander re-sponse is 1.

B. To illustrate the effect of dose rate on very-low dose risks, we split the same very low doseinto two separate fractions. The pattern of hypersensitive cells can change between fractions, but itis seen that a very low total dose will produce the same average response, R. Thus at very low doses,inverse dose-rate effects are negligible.

C and D. Here the dose is twice as large, but it remains low in the sense that the chance of 2 hitsper collective target is negligible. In agreement with general microdosimetric arguments, the re-sponse is also doubled, i.e. is linearly proportional to dose, and dose-rate effects remain negligible.

E. At a high acute dose, the chance of more than one alpha particle per collective target is nolonger negligible and this panel represents the case where an average of 4 hits occurs per collectivetarget. For acute doses, 4 alpha particles in one collective target are no more effective, in terms ofthe bystander response, than one alpha particle; the bystander response therefore increases less rap-idly than linearly with dose because of “saturation” – some of the alpha particles are “wasted.”

F. If the high dose is split into two fractions separated by a time interval (long enough for hyper-sensitive cells to be replaced), the response is doubled, i.e. there is an inverse dose-rate effect at highdoses.

Overall, comparing panel E with panel B shows that a linear extrapolation of risk from a highacute dose to low dose and low dose rate may underestimate this risk, in this schematic case by afactor of 4, due to saturation and to inverse dose-rate effects in the bystander response.

then later

then later

then later

A: very low dose & high dose rate

C: low dose & high dose rate

E: high dose & high dose rate

B: very low dose & low dose rate

D: low dose & low dose rate

F: high dose & low dose rate

D=2

R=1

D=4

R=2

D=32

R=4

D=2

R=1

D=4

R=2

D=32

R=8

Fig. 2.


16

fects in the bystander response suggest that a linear extrapo-lation from miner data which does not properly take intoaccount dose rate effects would underestimate the domesticradon risk by about a factor of four – a value comparable tothe empirical estimate applied in the recent BEIR-VI reporton radon risk estimation.

It is important to stress that we have in no sense“proven” the relevance of bystander phenomena to low-doseradon risks; rather we have described a mechanistic modelwhich is parsimonious in its number of parameters (fourparameters, making the model potentially highly testable),and which is consistent with a large body of epidemiologicaland laboratory data.

In conclusion, bystander effects represent a plausiblequantitative and mechanistic explanation of inverse dose-rate effects by high-LET radiation, resulting in dose-response relations which are non linear and which feature acomplex interplay between the effects of dose and exposuretime. The model presented here provides a potential mecha-nistic underpinning for the empirical exposure-time correc-tion factors applied in the recent BEIR-VI report on domes-tic radon risk estimation.

References

1. Brenner DJ, Little JB, Sachs RK, The bystander effect inradiation oncogenesis, II. A quantitative model, Radia-tion Research 155:402-408, 2001.

2. Lubin JH et al, Radon-exposed underground miners andinverse dose-rate (protraction enhancement) effects,Health Physics 69:494-500, 1995. n

Fig. 3.

10-4 10-3 10-2 10-1 100 101 102

Number of alpha particle traversals

10-4

10-3

10-2

10-1

100

101

102

Exc

ess

Rel

ativ

e R

isk

60 yr exposure(model-basedextrapolation)

miners, model fit 6-yr exposure

linear extrapolation from miner data,ignoring dose rate

measured domestic exposure risks

10-2 10-1 100 101 102 103 104Cumulative exposure (WLM)


17

Induction of P21/WAF1 in Microbeam Irradiated andBystander Normal Human Fibroblasts

Brian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Gerhard Randers-Pehrson and Charles R. Geard

There is now considerable evidence that the effects ofionizing radiation are not confined to the irradiated cell.Several studies have reported that non-irradiated cells adja-cent to those that are hit, also demonstrate several biologicalresponses observed in irradiated cells. Previous studies inthis laboratory have demon-strated the induction ofp21/WAF1 in irradiated andbystander cells using single-cell RT-PCR. Here we pres-ent data of the time kineticsof the induction ofp21/WAF1 following mi-crobeam irradiation with 5and 10 α-particles usingimmunofluorescent staining.

Normal human fibro-blasts (Clonetics) were ei-ther stained with 50 nMHoechst 33342 or 100 nMCell Tracker Orange (CTO,Molecular Probes) for 30minutes, followed by anincubation in fresh mediafor 30 minutes. Both sets ofcells were then trypsinized, counted, mixed in a 1:1 ratio,and seeded onto microbeam dishes at a density of 500 cellsper dish (250 cells of each type). Following microbeam irra-diation, dishes were fixed at 30 minutes, 1, 2 and 3 hours in100% ice cold methanol. In situ detection of p21/WAF1protein induction in hit and bystander cells was performed aspreviously described. The fluorescent images were captured

using a Hammatsu camera and an inverted Olympus micro-scope and fluorescence intensity per cell was measured us-ing Image ProPlus.

As can be seen in figure 1, the fluorescence of the celltracker orange stain made it possible to clearly distinguished

irradiated (non-stained) andbystander (orange fluores-cence) cells.

Following microbeam ir-radiation with 5 α-particles,there was an induction ofp21/WAF1 that was apparentat the earliest time point, 30minutes (Figure 2A). Thisinduction ranged between 1.5and 2 fold that of time-matched controls and seemedto increase up to 2 hours postirradiation and decline there-after. As can be seen therewas considerable variationamong individual cells, whichis similar to that reported forsingle-cell RT-PCR. By-stander cells also demon-

strated an increased p21/WAF1 signal (Figure 2B), althoughthese increases remained relatively constant over all timepoints; at around 1.25 fold higher than time-matched con-trols.

Following microbeam irradiation with 10 α-particlesthere was in increase in p21/WAF1 signal between 1.5 and2.5 fold higher than that seen in controls (Figure 3A). This

Fig. 2. Fold increase in p21/WAF1 over controls in normal human fibroblasts microbeam irradiated with 5 α-particles (A) and by-stander human fibroblasts (B). Each gray bar represents a single cell, and the black bars represent the mean of all cells at that time point.

Fig. 1. FITC labeled p21/WAF1 (yellow/green fluores-cence) in microbeam irradiated (non-stained) and bystander(orange fluorescence) normal human fibroblasts. Cells werefixed 1 hour following irradiation with 10 α-particles.

0

0.5

1

1.5

2

2.5

3

3.5

Time Post Irradiation

Fold

incr

ease

ove

r Con

trol

s

30 Minutes 1 Hour 2 Hours 3 Hours

A

0

0.5

1

1.5

2

2.5

3

3.5

Time Post-irradiation

Fold

Incr

ease

ove

r Con

trol

s


B

Time Post-irradiation Time Post-irradiation30 Minutes 1 Hour 2 Hours 3 Hours 30 Minutes 1 Hour 2 Hours 3 Hours


18

continued up to 3 hours post irradiation. This pattern wasdifferent from the 5 α-irradiated population that demon-strated a peak response at 2 hours post irradiation. Again,bystander cells also have elevated p21/WAF1 signals (Fig-ure 3B). The induction of p21/WAF1 in bystander cells didnot seem to be dependent on the number of α-particles de-livered to the irradiated cells (5 or 10).

These data complement previous results using single cellRT-PCR that have demonstrated the induction of p21/WAF1in both microbeam irradiated and bystander human fibro-blasts. Furthermore, the data are in keeping with previousresults that indicate the absence of a dose response in thebystander response following α-irradiation. n

Intra-Nuclear Dynamics of Mre11 in Human CellsFollowing Microbeam Irradiation with α-Particles

Adayabalam S. Balajee and Charles R. Geard

Exposure of cells to ionizing radiation (IR) induces DNAsingle strand breaks (SSBs), double strand breaks (DSBs),base damage and DNA-protein crosslinks. Recent studieshave suggested that the spectrum of lesions induced in theinterphase nuclei largely depends both on the quality of ra-diation exposure and on the higher order chromatin struc-ture. Among the wide variety of DNA lesions induced by IR,DSBs are considered to be significant as they lead both tocell mortality and stable genetic alterations if left unrepaired.Eukaryotic cells possess two major pathways to deal withthe repair of DSBs: (i) non-homologous end joining repairpathway and (ii) the recombinational repair pathway. Bothof these repair pathways involve a number of replication,repair and recombination proteins, which interact with eachother and sequentially assemble at the site of DNA lesions.Although substantial progress has been made in under-standing the functional complexities of the various proteinsin response to DNA damage, the precise participation ofeach of these proteins in the nuclear environment is far from

clear. As compared to conventional gamma and X-ray irra-diation, introduction of lesions at defined sub-cellular sitesby microbeam irradiation aids an understanding of the pre-cise mode of action of the various protein complexes in DSBrepair. Using such a facility developed at RARAF for mi-crobeam α-particle irradiation, we have set out to examinethe intra-nuclear dynamics of Mre 11 in human cells in re-sponse to DNA strand breaks. We have chosen Mre 11 be-cause the Mre 11 complex comprised of Mre 11, Rad50 andthe gene product Nijmegen breakage syndrome, Nbs1 playsan important role in the repair of DSBs (1).

Human primary fibroblast cells lines (MRC5 and WI38)used in this study were procured from Coriell Cell reposi-tory, Camden, New Jersey. The layout and microbeam irra-diation procedure have been previously described (2).Briefly, plateau phase cells were trypsinized and approxi-mately 500-600 cells were seeded in dishes specially de-signed for microbeam irradiation. The cells were stainedwith a 50 nM solution of Hoechst 33342 for 30 min prior to

0

0.5

1

1.5

2

2.5

3

3.5

Time Post Irradiation

Fold

incr

ease

ove

r Con

trol

s


A

0

0.5

1

1.5

2

2.5

3

3.5

Time Post-irradiationFo

ld In

crea

se o

ver C

ontr

ols


B

Fig. 3. Fold increase in p21/WAF1 over controls in normal human fibroblasts microbeam irradiated with 10 alpha particles (A) andbystander human fibroblasts (B).

30 Minutes 1 Hour 2 Hours 3 Hours 30 Minutes 1 Hour 2 Hours 3 Hours


19

irradiation. The cells were irradiated with 1, 2, 4, 8 and 12α-particles and the cells were either fixed immediately inice-cold methanol or allowed to recover for various lengthsof time (30 min, 2hr, 4hr, 8hr and 24 hr). The cells that werestained with Hoechst 33342 and exposed to UV served ascontrol.

Prior to immunostaining, cells were briefly washed in asolution of PBS containing 0.5% Triton X-100, whichgreatly improved the accessibility of Mre 11 antibody. Thecells were washed in TBST (20 mM Tris-HCl pH 7.4, 137mM NaCl and 0.2% Tween 20) and incubated for 30 min inTBST containing 5% non-fat dried milk (NFDM). The cellswere washed in TBST and incubated with rabbit monoclonalMre 11 antibody (Calbiochem, 1:50 dilution in TBST-5%NFDM) for 1h at 37°C. The cells were washed three timesof 5 min each in TBST buffer, incubated for 1h with bi-otinylated secondary antibody and finally with avidin-TexasRed antibody. The cells were washed in TBST and dehy-drated in 70% and 95% ethanol. DNA was counterstainedwith DAPI (0.1µg/ml prepared in Vectashield mountingmedium; Vector laboratories) and covered with a coverglass.

In unirradiated control cells, Mre 11 staining was not in-tense and numerous small Mre 11 foci were found homoge-

nously throughout the interphase nuclei. The staining patterndid not differ in the control cells irrespective of the differentincubation times. The intensity of Mre 11 staining was mar-ginally higher in some of the cells irradiated with 2α-particles. In cells treated with 4α-particles and higher, in-tense Mre 11 staining was observed even at the earliest timepoint analyzed (30 min). The quantitation of fluorescenceintensity of Mre 11 measured in 50 randomly chosen irradi-ated cells exhibited approximately 3-4 fold increase as com-pared to control cells. The induction of Mre 11 foci wasclearly observed at 2hr after irradiation. In cells irradiatedwith 4 and 8α- particles, 50-60 distinct focal sites of Mre 11were detected. The number of focal sites increased to morethan 100 at 4hr after irradiation. The focal sites were farmore intense as compared to 2hr post-irradiation and thefocal sites of Mre 11 gradually declined at 8 hr and 24 hr.The pattern of Mre 11 foci formation in control and irradi-ated cells is shown in Fig. 1.

In addition to Mre 11, we have also examined the timecourse of Serine15 phosphorylation of p53 in 12α-particleirradiated cells. Phosphorylated p53 is detected as prominentfoci 30 min after IR and persisted up to 2hr after irradiation.The focal pattern of p53 was greatly diminished at 8hr afterirradiation and at this time point, p53 was more homogene-

Fig. 1. Pattern of distribution of Mre 11 foci induced by targeted nuclear irradiation of α-particles in the primary human fibroblastcells (MRC5). The cells were fixed in methanol and immunostained for Mre 11. Avidin-Texas Red conjugated antibody was used todetect the spatial distribution of Mre 11 in the interphase nuclei.

0α, 2hr 4α, 2hr 8α, 2hr

0α, 2hr 4α, 2hr 4α, 4hr


20

ous in distribution and the intensity of p53 was much lesscompared to earlier time points. P53ser15 intensity in the irra-diated cells reached the level of control cells at 24hr afterirradiation. The time course of p53 foci formation correlatedwell with the formation of Mre 11 foci. We are currentlyexamining the Mre 11 foci formation in cells derived fromradiosensitive Ataxia telangiectasia patients. The ATM geneproduct phosphorylates Nbs1, a component of Mre 11 com-plex, in a S-phase specific manner in response to DNA dam-age and it would be interesting to determine the kinetics ofMre 11 formation in AT cells. Clearly however, the Mre 11foci found after microbeam irradiation are not confined tosites at or near the nuclear centroid where the alpha particlespassed through the nucleus. We are planning to characterizethe spatial distribution as well as the co-localization of Nbs1

and Rad50 with Mre 11 in microbeam irradiated cells. Wealso intend to use the protein kinase and DNA repair inhibi-tors to determine the formation and persistence of Mre 11foci in microbeam-irradiated cells. In addition to Mre 11complex, additional factors involved in the non-homologousend joining and recombination repair pathways will also beanalyzed in future.

References

1. Mirzoeva OK and Petrini JH, Mol. Cell. Biol., 21:281-288, 2001.

2. Hei TK et al, Proc. Natl. Acad. Sci. (USA) 94:3765-3770,1997. n

SCGE Detection of DNA Strand Breaks and OxidizedBase Lesions Induced by Microbeam Irradiation of

Human Nuclei with Defined Number of Alpha ParticlesAdayabalam S. Balajee, Brian Ponnaiya, M. Prakash Hande, Gloria Jenkins-Baker,

Stephen A. Marino, Gerhard Randers-Pehrson, Charles R. Geard

Ionizing radiation (IR) induces a complex spectrum oflesions including single strand breaks (SSBs), double strandbreaks (DSBs), base damage and DNA-protein crosslinks inthe genomic DNA. Among them, DSBs are considered to bethe most important lesion as the persistence and mis-repairof DSBs lead to cell mortality and stable genetic alterations.Recent studies have indicated that the intra-nuclear forma-tion of DSBs largely depends on the track structure of radia-tion. DSBs generated by high LET radiations appear to benon-random and the distribution of DSB is dependent onhigher order chromatin structure. To determine the inductionof lesions in relation to higher order chromatin structure ininterphase nuclei, precise targeted irradiation with a definednumber of charged particles is highly desirable. Such anapproach would enable us to understand the impact of chro-matin structure on lesion induction and repair in differentsub-cellular components. Using the microbeam facility de-veloped at RARAF for α-particle irradiation, it has beenshown that targeted cytoplasmic irradiation has the potentialto induce mutations in mammalian cells (1). Induction of“bystander” mutagenic effects has also been noticed in unir-radiated cells (2). Elucidation of the molecular cause for themutagenicity and bystander effects require the characteriza-tion of the DNA lesions induced by α-particle irradiation.Determination of the types of DNA lesions and their repairkinetics induced by targeted nuclear and cytoplasmic α-particle irradiation may provide insights towards under-standing the mechanistic basis for mutagenicity/bystander

effects caused by microbeam irradiation.The comet assay, also known as single cell gel electro-

phoresis (SCGE), has been shown to be extremely sensitivefor the analysis of induction and repair of DNA strandbreaks and oxidative DNA lesions. This assay is particularlyuseful for microbeam experiments where the sample size isoften restricted. In the alkaline version of this assay, a dilutesuspension of cells mixed with low melt agarose is lysed atpH 10 to release the nucleoids. The nucleoids are subse-quently incubated in an alkaline electrophoresis buffer (0.3M NaOH; 1mM EDTA pH 12.1 to 13.1) to facilitate denatu-ration, unwinding and the exposure of alkali-labile sites. Thenicked, relaxed and denatured DNA is then subjected toelectrophoresis and the extent of DNA damage is reflectedby the migration of the DNA from the nuclei resembling thetail of a comet.

We have used the comet assay to detect the DNA strandbreaks and oxidative lesions induced by targeted irradiationof defined number of alpha particles in the nuclei of humancells. For this purpose, primary fibroblast cells derived fromNormal individuals (WI38 and MRC5) and radiosensitiveAtaxia telangiectasia patients (GM2052C, GM5823C) wereused. Briefly, plateau phase cells were trypsinized and ap-proximately 1000 cells were seeded in dishes specially de-signed for microbeam irradiation. The cells were stainedwith a 50 nM solution of Hoechst 33342 for 30 min prior toirradiation. In the pilot experiments, we have irradiated thenuclei of normal diploid fibroblasts (WI38) with 1, 2, 4 and


21

8 α-particle and analyzed the induction and repair of alltypes of DNA lesions by alkaline SCGE at two pH condi-tions (12 and 13.1). Alkaline unwinding was done for 30min and the electrophoresis was carried out for 30 min (25Volts and 300 mA). After electrophoresis, comets were neu-tralized for 15 min in 0.4M Tris pH 7.4 and stained withCyber green (0.1 µg/ml in 10mM Tris/1mM EDTA bufferpH 7.5). The comets (50-100 in number) were randomlychosen and captured using the Axioplan 2 Imaging micro-scope. The length as well as tail intensity was measured us-ing Euclid comet software. The initial induction of strandbreaks detected under both pH conditions was essentially thesame and there was a clear dose dependent increase in theinduction of strand breaks and apurinic/apyrimidinic sites inWI 38 cells. The distribution of tail moment observedamong the comets of irradiated cells clearly revealed a dosedependent effect as compared to Hoechst 33342 stained andUV exposed control cells where a small fraction (20%) alsoexhibited the tail moment. In order to assess the repair ofstrand breaks, irradiated cells were allowed to recover for 3hrs and the cells were subsequently processed for the cometassay. The tail moment observed 3hr after irradiation wasessentially similar to that of unirradiated control cells indi-cating the efficient repair of strand breaks and AP sites. Therepresentative comets showing the extent of initial inductionof DNA damage and repair induced by 4, 8 and 12 α-parti-cles are given in Fig.1.

The tail moment of control and irradiated cells at 0hr and3hr for 4, 8 and 12 α-particles as well as the distribution oftail moment among the comets are illustrated in Fig. 2.

Experiments comparing the repair kinetics of the DNAlesions induced by microbeam irradiation in Normal and ATcells are in progress. Initial experiments revealed that theinitial induction of lesions by 8 alpha particles is similar inprimary as well as SV-40 transformed Normal and AT fi-broblast cells.

Although the SCGE is very sensitive, we could not dis-criminate the initial DNA damage induced by less than 4 α-particles. Alkaline comet assays reflect DNA strand breaksand AP sites, but a lack of detection of all the AP sites and

of base damage under alkaline conditions is noticed when amodified SCGE assay is used. The enzymatic digestion ofnucleoids with endonuclease III (Endo III) and Formamidopyrimidine glycosylase (FPG) prior to electrophoresis hasgreatly improved the sensitivity of the assay in detecting theinduction and repair of different oxidized base lesions suchas AP sites, thymine glycol, urea, 5-hydroxy-6-hydrothy-mine, 5,6-dihydrouracil, uracil glycol, alloxan, 5-hydroxy-6-hydrouracil, uracil glycol, 5-hydroxy-5-methylhydantoin, 5-hydroxycytosine, 5-hydroxyuracil, methylartonylurea, thy-mine ring fragmentation product and 8-Oxoguanine.

In order to identify the different spectrum of lesions in-duced by microbeam irradiation, we have characterized thecomet assay coupled with the enzymatic digestion of nucle-oids with endo III. As compared to the comets without EndoIII treatment, Endo III treated comets demonstrated a 1.5 – 2fold increase in the tail moment demonstrating the increased

Fig. 1. The induction of DNA damage and repair induced byα-particle irradiation.

Fig. 2. The distribution of tail moment among the comets of control and irradiated cells at 0hr and 3hr.

0 α

4 α

8 α

12 α

0hr 3hr


22

sensitivity in detecting some of the oxidized base lesions andAP sites. We are beginning to use this approach to detect thedamage induced by a single alpha particle irradiation. Ourfuture goal is to increase the sensitivity of this assay to de-tect the damage induced by a single α-particle. In addition tousing the alkaline comet assay, attempts have been made tocharacterize the double strand breaks by alkaline unwindingfollowed by neutral gel electrophoresis. We also intend touse the neutral lysis and neutral gel electrophoresis to deter-mine the relative proportion of single strand breaks versusthe double strand breaks. We are planning to use this assayto determine the effect of cytoplasmic irradiation on nuclearDNA damage and also the impact of cytoplasmic vs. nuclearirradiation on bystander effects. We wish to investigate by-

stander effects in p53 mutant cell lines as well as in human,mouse and hamster cell lines that are defective in DSB re-pair proteins. Additionally, we would like to determine DNAdamage and repair in defined genomic sequences by a com-bination of SCGE and fluorescence in situ hybridization(FISH) techniques.

References

1. Wu LJ et al, Proc. Natl. Acad. Sci. (USA) 96:4959-4964,1999.

2. Zhou H et al, Proc. Natl. Acad. Sci. (USA) 97:2099-2104, 2000. n

Modulation of Adaptive Response in Alpha ParticleInduced Bystander MutagenesisHongning Zhou, Gerhard Randers-Pehrson, Eric J. Hall and Tom K. Hei

The risk of developing radiation-induced cancer has tra-ditionally been estimated from cancer incidence amongJapanese A-bomb survivors. These data provide the bestestimate of cancer risk over the dose range from 20 to 250cGy. The cancer risk at doses below 20cGy, however, re-mains uncertain and has been the subject of controversy fordecades in the absence of definitive data. Two conflictingphenomena appear to be important for low dose radiationand have potential to impact on the shape of the dose re-sponse relationship. First, there is the bystander effect, theterm used to describe the biological effects observed in cellsthat are not themselves traversed by a charged particle, butare neighbors of cells that are. Second, there is the adaptiveresponse, whereby exposure to a low level of DNA damagerenders cells resistant to a subsequent high dose exposure.

Ever since X-rays were shown to induce mutation inDrosophila and maize, it has been accepted dogma that thedeleterious effects of radiation, such as mutation and carci-nogenesis, were due mainly to direct damage to DNA. Evi-dence is now emerging that extranuclear or extracellulartargets are extremely important in mediating the genotoxiceffects of radiation (1). Using a precision charged particlemicrobeam, we reported that cells that had been lethally ir-radiated with alpha particles could induce mutagenesis inneighboring cells not directly hit by the particles, and thatmutant induction depended on cell-cell communication (2).However, exposure to high doses of alpha particles is anunlikely scenario in environmental exposures to radon. Toextend this observation, we found that a single alpha particletraversal of a small fraction of AL cells (10-20%) induced amutagenic response similar to that occurring when 100% of

the cells in the population were hit, and that gap junctionmediated cell-cell communication plays an important role inthe process (3).

The adaptive response is characterized by a reduction ofradiobiological response in cells pretreated with a low doseradiation followed by exposure to a challenging higher dose.Numerous experimental data have shown the existence ofsuch a response with a variety of endpoints such as micronu-clei, sister chromatid exchange, and mutation (4,5). How-ever, the molecular nature of the mechanism underlying theadaptive response is not yet elucidated. The working hy-pothesis is that an inducible molecular process is triggeredby low doses and leads to cell protection against the delete-rious effects of a subsequent irradiation.

Although these two conflicting phenomena appear to beimportant, there is very limited data available comparing thebystander effect versus adaptive response (6). In our presentstudy, AL cells plated in microbeam dishes two days beforeas described (2,3) were exposed to a 0.1 Gy dose of 250 kVX-rays from a Westinghouse Coronado X-ray machine, op-erating at 10mA, with 0.2mm copper and 1mm aluminumfilter (dose rate was 25.0 cGy/min). Four hours later, 10% ofrandomly selected cells were irradiated with a single alphaparticle using the Columbia precision particle microbeam.After alpha particle traversal, the cells were kept in micro-beam dishes for 2 days before replating in culture flasks.Determination of the mutant fraction was carried out as de-scribed before (2,3). Our preliminary data showed that themutant yield from the population where 10% of randomlyselected cells were irradiated with single alpha particle de-creased significantly if the cells were pretreated with a low


23

dose of X-rays (Figure 1). The results show that in the pres-ence of low dose radiation stress, the bystander mutagenesisis modulated by the adaptive response. Though the mecha-nism(s) is unclear, it is likely that when cells are exposed tolow dose X-rays, they initiate a series of self-preservationmechanisms that diminish their ability to respond to by-stander signaling.

References

1. Mothersill C and Seymour C, Radiation-induced by-stander effects: past history and future directions (Re-view), Radiat. Res. 155:759-67, 2001.

2. Zhou H, Randers-Pehrson G, Waldren CA, Vannais D,Hall EJ and Hei TK, Induction of a bystander mutageniceffect of alpha particles in mammalian cells, Proc. Natl.Acad. Sci. (USA) 97:2099-104, 2000.

3. Zhou H, Suzuki M, Randers-Pehrson G, Vannais D,Chen G, Trosko JE, Waldren CA and Hei TK, Radiationrisk to low fluences of alpha particles may be greaterthan we thought, Proc. Natl. Acad. Sci. (USA) (in press),2001.

4. Upton AC, Radiation hormesis: data and interpretations(Review), Crit. Rev. Toxicol. 31:681-695, 2001.

5. Rigaud O and Moustacchi E, Radioadaptation for genemutation and the possible molecular mechanisms of theadaptive response (Review), Mutat. Res. 358:127-134,1996.

6. Sawant SG, Randers-Pehrson G, Metting NF and HallEJ, Adaptive response and the bystander effect inducedby radiation in C3H10T1/2 cells in culture, Radiat. Res.156:177-180, 2001. n

Novel Approaches with Track Segment Alpha Particlesand Cell Co-Cultures in Studies of Bystander Effects

Charles R. Geard, Gloria Jenkins-Baker, Stephen A. Marino, Gary Johnson and Brian Ponnaiya

Here we report on a protocol that has been devised to as-sess one aspect of bystander cell responsiveness, specificallyeffects on non-contacting cells where passage of responsibleagents through medium is required. We take advantage ofthe short ranges of low energy charged particles to ensurethat all energy is deposited in culture medium or one popu-lation of cells, with the co-cultured separated cells beingbystanders. Further, since the majority of bystander studiesevaluate responses in one cell type we allow for cells ofsimilar or disparate origins to be co-cultured to determinethe specificity or lack thereof in cellular responses to insult.We find that irradiation of medium alone can elicit responsesin some non-hit cells, but responses are enhanced when onepopulation of cells is targeted. These responses vary betweencell types with fibroblasts being quite responsive, whereasepithelial cells are not. That is, the bystander effect does notrequire cells to be in contact and directly communicating andthere are profound differences between cells of differentorigins.

Regular dishes for track segment experiments consist of

stainless steel rings of 35mm internal diameter [as for stan-dard tissue culture plastic dishes] with 6 µm thick mylarepoxied to one surface. However if ports of 3.8 mm diameterare drilled through the 9.5 mm deep and 3.4 mm thick sidesof the ring opposite each other then mylar can be epoxied toboth surfaces [Fig. 1]. The dishes are then baked and alcoholsterilized, which also serves to tauten the mylar. The portscan accommodate the luer fitting of a standard hypodermicneedle or of appropriate plugs and hence cells in mediumcan be introduced and allowed to attach to one mylar sur-face. After cells have attached to one surface most mediumis aspirated off and the dish turned over for introduction ofanother population of cells in medium for attachment to theother mylar surface. The culture vessel is then completelyfilled with medium taking care to ensure no air bubbles re-main. The dishes with cells growing on both surfaces, whichcan be of similar or disparate origins, only constrained bythe appropriateness of the cell culture medium, are thenplaced in the incubator until track segment irradiationthrough one surface. A variant of this approach takes the

1 2 3 4

Indu

ced

CD

59- M

utan

ts

per 1

05 Sur

vivo

rs

0

50

100

150

2001. Control2. 1α, 10%3. 0.1 Gy X-ray, 1α, 10%4. 0.1 Gy X-ray

Fig. 1. Mutation fraction obtained from population of ALcells in which 10% of whose nuclei were traversed by singlealpha particle. Four hours before alpha particle irradiation, cellswere treated with 0.1 Gy X-ray. Data were pooled from 3 inde-pendent experiments.


24

same stainless steel rings and machines down the inner sur-face to a 38mm diameter, except for a 1mm thick and 1.2mmwide lip. Stainless steel inserts are then machined to sit onthese lips. These 2.2 mm thick medical grade inserts are ma-chined with holes of 28 mm diameter into which Costartranswells can be inserted [Fig. 1]. The transwells have ta-pering sides with a polyester microporous membrane at-tached to the lower surface of 24mm diameter. The mem-brane readily allows for cell attachment and growth and haspore sizes of 3 µm, which readily permits the passage ofculture medium. A similar procedure is followed as with thedouble-sided mylar. Cells are grown on the mylar surface [ornot] and on the surface of the transwell inserts. The cells areseparated by a minimum distance of 1mm, with the intro-duction of spacers allowing for the amount of medium be-tween hit and bystander cells to be varied over a ten-foldrange in five steps. The distance of 1mm is sufficient to en-sure that no charged particles accelerated on the RARAFtrack segment facility actually reach the superimposed by-stander cells.

The potential of medium irradiation alone to initiateclastogenic changes in the form of micronuclei and to altercellular progression through the cell cycle was assessed.Human fibroblasts that are lifespan extended [BJ-1/h-tert]and human epithelial cells that are similarly lifespan ex-tended [RPE/h-tert] were cultured as exponentially growingpopulations on the non-irradiated surface of both dish typesand the medium irradiated with doses of 0, 0.1, 1, 10, and100 Gy of 90 keV/µm alpha particles. Following irradiationbromodeoxyuridine at 5 µM was added to the medium to betaken up into the replicating DNA of S-phase cells, and cellswere fixed in situ with ice-cold methanol at times from 2hr[to assess % of S-phase cells at the time of irradiation] to 72

hr post irradiation. The incorporation of BrdU into cells wasmonitored by use of an antibody against such substitutedDNA and cell DNA was stained with DAPI. The fraction ofBrdU containing nuclei was determined as a function of timeas was the frequency of cells showing post-mitotic micronu-clei using a microscope with incident mercury bulb illumi-nation and a filter cube to discriminate non-BrdU containingnuclei [blue fluorescing] from nuclei with incorporatedBrdU [yellow-green fluorescing].

The frequencies of micronuclei in non-hit fibroblasts andepithelial cells following medium irradiation for the 24, 48and 72 hr sampling periods indicated that for one individualsampling time differences between treatments was signifi-cant. For the combined results micronuclei in fibroblastswere increased above expected levels [based on frequenciesin control cells], whereas the epithelial cells did not showsuch an increase. Differences between the two cell types intheir responses were profound. Fractions of cells with BrdUuptake [% labeled cells] were assessed at 24, 48 and 72 hrpost irradiation. In the control cells, for both cell types, thefraction of labeled cells increases with time, with only minornon-consistent differences from control for the bystandercells at any time period.

Overall, high dose medium irradiation alone can initiatelimited clastogenic changes but not cytostatic changes, witha clear difference between the response of fibroblast andepithelial cells. This approach to evaluate aspects of cellbystander responsiveness using cell co-cultures, or to assessmedium responsiveness at the time of irradiation foregoingmedia transfers, can efficiently use charged particle tracksegments over an extensive LET range. This then allows forthe discrimination of the factor/s responsible and for thedetermination of their LET dependence. n

Alphaparticles

Mylar surface withbystander cells

Mylar surface withcells to be irradiated

Port withPlug

Stainlesssteel ring

Transwell withbystander cells

Spacer (of variablethickness)

AlphaparticlesA B

Media

Fig. 1.


25

Investigation of a Radiation-Induced Bystander EffectUsing Co-Culturing Protocols

Brian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Stephen A. Marino and Charles R. Geard

Previous studies in this laboratory using the charged par-ticle microbeam have demonstrated the induction of micro-nuclei in fibroblast cells that were adjacent to alpha particleirradiated cells. To determine the suitability of the co-culturing approach to study the bystander effect, initialstudies focused on the detection of the induction of micronu-clei in irradiated and bystander fibroblasts.

Immortalized human fibroblasts (BJ1-htert) were irradi-ated with 0,0.1 (~1 per nucleus), 1 and 10 Gy α-particles (90keV/µm) and hit and bystander cells were examined forBrdU uptake and micronuclei incidence at 24, 48 and 72hours post irradiation. As expected at 24 hours post irradia-tion there was a dose dependent reduction in the number ofBrdU positive cells in the irradiated population (Fig. 1). Thiswas apparent at 48 and 72 hours as well. In addition, fibro-blast bystander cells also demonstrated small though con-sistent delays in cell cycle progression.

Micronuclei frequencies observed in the irradiated andbystander populations are presented in Table I. Yields ofmicronuclei in the irradiated populations were 2-4 foldhigher than that of controls. The peak frequency was ob-served in the populations that received 1 Gy. The popula-tions that received 10 Gy did not have significantly higheryields of micronuclei, which would be expected given thelarge delays in cell cycle progression observed. Bystandercells also had increased micronuclei frequencies at all doses.These increased yields were about 2 fold higher that controlsat all doses, which is similar to the observations in the mi-crobeam experiments. The similarities between these results

and those of the microbeam experiments indicate that this isa suitable approach to study the bystander effect.

Having detected the bystander effect in co-cultured fi-broblasts, experiments were then conducted to determinewhether a similar effect could be induced in co-culturedepithelial cells. Immortalized retinal epithelial cells (RPE-htert) were seeded onto mylar mylar rings and irradiatedwith 0,0.1,1 and 10 Gy α-particles (90 keV/µm). Epithelialcells seeded onto transwells served as bystander populations.Irradiated and bystander cells were examined for BrdU up-take and micronuclei incidence at 24, 48 and 72 hours postirradiation.

As can be seen in Fig. 2, there was a dose dependent de-crease in the number of cycling cells in the irradiated popu-lations at all times points. Interestingly, unlike the bystanderfibroblasts, the bystander epithelial cells did not appear todelayed in progression through the cell cycle. Even at theearliest time point examined, 24 hours, there were no appar-ent differences between bystanders to irradiated populationsand controls. Frequencies of micronuclei observed inepithelial irradiated and bystander populations are presentedin Table II. Again, there was a 2-4 fold increase in micronu-clei yields in the irradiated populations, with the exceptionof the population that received the highest dose. Importantly,micronuclei frequencies in epithelial bystander populationswere not similar to those seen in fibroblast bystander popu-lations. All of the epithelial bystanders had frequencies ofmicronuclei comparable to that seen in the controls. Thesedata are at odds with microbeam results that indicated the

020406080

100 72 Hours

0

20

40

60

80

100

0 0.1 1 10

24 Hours 48 Hours

Fig. 1. BJ1-htert – BrdU uptake (% labeled cells) in irradiated ( ) and bystander ( ) cells at 24, 48 and 72 post irradiation (n=5).

% L

abel

ed C

ells

Dose (Gy)

0

20

40

60

80

100

0 0.1 1 10

Dose (Gy)

0

20

40

60

80

100

0 0.1 1 10

Dose (Gy)

24 Hours 48 Hours 72 Hours

0

50

100

150

0 0.1 1 10

24 Hours

0

50

100

150

0 0.1 1 10

48 Hours

0

50

100

150

0 0.1 1 10

72 Hours

Fig. 2. RPE-htert – BrdU uptake (% labeled cells) in irradiated ( ) and bystander ( ) cells at 24, 48 and 72 post irradiation (n=5).

% L

abel

ed C

ells

Dose (Gy)

24 Hours 48 Hours 72 Hours


26

induction of micronuclei in bystander epithelial cells.Data presented here indicate that the co-culturing tech-

niques developed in the laboratory are suitable for studyingthe bystander effect and may useful in examining details ofthe phenomenon when used alone or in parallel with micro-beam experiments. n

Investigation of the Role of Cell Type Specificity in theInduction of a Bystander Effect

Brian Ponnaiya, Gloria Jenkins-Baker, Mutian Zhang, Alan Bigelow, Stephen Marino and Charles R. Geard

Previous reports have suggested that the radiation-induced bystander effect is cell type specific. We have seenthat irradiated fibroblasts can induce micronuclei in by-stander fibroblasts, but bystander epithelial cells are refrac-tory to irradiated epithelial cells. In addition, it has been seenthat media from irradiated epithelial cells can reduce platingefficiencies in fibroblasts but not vice versa. We have usedthe co-culturing protocols discussed earlier to determine ifthis is true in our system. Epithelial cells (RPE-htert) andfibroblasts (BJ1-htert) were seeded on either side of doublesided mylar dishes and co-cultured in epithelial cell mediafor 48 hours prior to irradiation. One side (either epithelial orfibroblast) were irradiated with 0.1, 1 or 10 Gy α-particlesand returned to the incubator for 1 hour. Cells from eithersurface (irradiated and bystander) were then trypsinized,seeded in chambered slides in media containing BrdU (final

concentration 1µM) and fixed at 24, 48 and 72 hours post-irradiation. Dividing cells, as indicated by BrdU uptake,were identified using a FITC-labeled anti-BrdU antibody(Becton Dickinson) following the manufacturers recommen-dations. Cells were counterstained with DAPI and werescored for presence of BrdU uptake (green fluorescence) andmicronuclei. Micronuclei yields at 24, 48 and 72 hours werepooled and are presented in Tables I and II.

The incidences of micronuclei observed in irradiatedepithelial cells and bystander fibroblasts are presented inTable I.

As expected, irradiation with 0.1 and 1 Gy resulted in 2and 4 fold increases in micronuclei yields respectively. Thefrequency of micronuclei in the epithelial population ex-posed to 10 Gy was similar to that of controls which wouldbe expected given that these cells do not readily re-enter the

Table II.

Frequencies of micronucleiobserved in co-cultured human epithelial cells

and bystanders following α-irradiation.

Dos

e (G

y)

No.

of C

ells

Scor

ed

Expe

cted

No.

of

Mic

ronu

clei

a

No.

of

Mic

ronu

clei

Scor

ed

Irradiated Epithelial Cells0 5791 115

0.1 3570 1971 5967 386

10 4993 85Bystanders of Irradiated Epithelial Cells

0 6527 - 1180.1 3064 55 761 6151 111 106

10 5986 108 100a Expected frequencies of micronuclei for bystander cellsbased on corresponding control values in the absence of abystander effect.

Table I.

Frequencies of micronucleiobserved in co-cultured human fibroblasts

and bystanders following α-irradiation.D

ose

(Gy)

No.

of C

ells

Scor

ed

Expe

cted

No.

of

Mic

ronu

clei

a

No.

of

Mic

ronu

clei

Scor

ed

Irradiated Fibroblasts0 6357 123

0.1 7014 2561 7021 44910 5854 155Bystanders of Irradiated Fibroblasts0 6150 134

0.1 6929 146 212*1 6138 129 216*10 6743 142 217*

a Expected frequencies of micronuclei for bystander cellsbased on corresponding control values in the absence of a by-stander effect. * indicates observed frequencies are signifi-cantly different from expected values as determined by the χ2

test (p<0.001).


27

cell cycle. Further, only the fibroblast population that wasco-cultured with epithelial cells that received 10 Gy demon-strated significantly elevated yields of micronuclei.

Micronuclei frequencies observed in irradiated fibro-blasts and co-cultured bystander epithelial cells are pre-sented in Table II.

Irradiation with 0.1 and 1 Gy resulted in a 1.5 to 3.5 foldincreases in micronuclei incidences over that seen in con-trols. At the highest dose (10 Gy) the incidence remainedsimilar to controls. These results are similar to those previ-ously reported for irradiated fibroblasts. The bystanderepithelial cell populations demonstrate elevated frequenciesof micronuclei. All epithelial populations that were bystand-ers to irradiated fibroblasts had about 2 fold higher yields of

micronuclei than the control epithelial population. Theseincreases in the bystander populations do not appear to bedose dependent.

Taken together, these results would suggest that epithe-lial cells are less capable of inducing a bystander effectwhen irradiated, but that they are capable of responding toirradiated fibroblasts which results in the induction of mi-cronuclei in the bystander epithelial cells. This is in contrastto other reports that suggest that irradiated epithelial cellscan induce a bystander effect in bystander fibroblasts but notvice versa. Experiments are on going to examine whetherthis cell-type specificity extends to other epithelial cell linesas well. n

Induction of Chromosomal Aberrations in α-Irradiatedand Bystander Fibroblasts.


Early studies of the bystander effect reported that fol-lowing exposure to low fluences of alpha particles, elevatedfrequencies of sister chromatid exchanges were observed inmore cells than were likely to be traversed by a particle.Here we present data from experiments designed to investi-gate the induction of chromsomal aberrations in irradiatedand bystander fibroblasts using co-culturing techniques.

Immortalized human fibroblasts (BJ1-htert) seeded onboth surfaces of double-sided mylar dishes were co-cultured

for 48 hours such that both populations were in plateauphase at the time of irradiation. Cells on one side were irra-diated with 0.1 or 1 Gy α-particles and the two sides wereseparated 1 hour post-irradiation. Both, irradiated and non-irradiated (bystander) populations were reseeded onto 60mm dishes and colcemid was added at 24, 27 or 30 hourspost irradiation to arrest cells at the first metaphase post ir-radiation. Metaphases were prepared according to standardprotocols, Giemsa stained and scored for gross chromosomal

Table II.

Frequencies of micronuclei observed inco-cultured α-irradiated fibroblasts

and bystander epithelial cells.

Dos

e (G

y)

No.

of C

ells

Scor

ed

Expe

cted

No.

of

Mic

ronu

clei

a

No.

of

Mic

ronu

clei

Scor

ed

Irradiated Fibroblasts0 2755 57

0.1 2880 801 2554 171

10 2872 60Bystander Epithelial Cells of Irradiated Fibroblasts

0 2025 320.1 2020 32 651 2045 33 71

10 2091 33 79

Table I.

Frequencies of micronuclei observed inco-cultured α-irradiated epithelial cells

and bystander human fibroblasts.

Dos

e (G

y)

No.

of C

ells

Scor

ed

Expe

cted

No.

of

Mic

ronu

clei

a

No.

of

Mic

ronu

clei

Scor

ed

Irradiated Epithelial Cells0 3534 58

0.1 3479 1001 3587 249

10 3553 37Bystander Fibroblasts of Irradiated Epithelial Cells

0 3335 610.1 3443 62 521 3328 60 77

10 3435 62 98


28

aberrations.As would be expected, frequencies of chromosomal ab-

errations in the irradiated populations were increased in adose dependent manner (Figure 1). The populations thatreceived 0.1 and 1 Gy had 0.3 and 1.3 aberrations per cellrespectively. These aberrations were almost all chromo-some-type aberrations, including dicentrics, excess acentricfragments and acentric rings. The frequencies of chromatid-type aberrations in all irradiated populations were compara-ble to that seen in controls, which would be expected giventhat these populations were in the G0/G1 phase of the cellcycle at the time of irradiation.

Increased frequencies of chromosomal aberrations werealso detected in the bystander populations. However, unlikethe irradiated populations, bystander populations had in-creased yields of chromatid-type aberrations (Figure 2).

Furthermore, all the chromatid-type aberrations observedwere of the simple type, namely breaks and gaps.

These studies demonstrate the suitability of co-culturingtechniques to investigate the bystander effect, especially instudies of end points that require large numbers of irradiatedand bystander cells. Since these studies were performed us-ing Giemsa staining that detects gross chromosomal aberra-tions, it remains possible that the bystander populationscontained symmetric chromosome-type aberrations that re-mained undetected. Current studies are aimed at examiningthis possibility using whole chromosome painting tech-niques. This study also implies that irradiated cells do notrespond to the factor(s) that are responsible for the elevatedfrequencies of replication/post-replication type aberrations inbystander cells. n

Detection of Chromosomal Instability in Co-Culturedα-Irradiated and Bystander Human Fibroblasts


There are now several reports on the induction of de-layed chromosomal aberrations (genomic instability) fol-lowing both high and low LET irradiation. De novo chromo-some-type as well as chromatid-type aberrations have beendectected several cell generations post irradiation. Here wepresent data on the delayed appearance of chromatid typesaberrations in both α-irradiated and bystander human cells.

Immortalized human fibroblasts (BJ1-htert) seeded oneither surface of double-sided mylar dishes were co-culturedfor 48 hours such that both populations were in palteau

phase at the time of irradiation. Cells on one side were irra-diated with 0.1,1 or 10 Gy α-particles and the two sideswere separated 1 hour post-irradiation. Both irradiated andnon-irradiated (bystander) populations were reseeded onto60 mm dishes and examined for chromosomal aberrations at15 and 20 population doublings post irradiation using stan-dard Giemsa staining.

By 15 population doublings all aberrations induced bythe initial irradiation had been removed from the irradiatedpopulations, and the frequencies of chromosome-type aber-

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 0.1 1

Dose (Gy)

Abe

rrat

ions

/Cel

l

Fig. 1. Frequencies of chromosomal aberrations observedin human fibroblasts following α-irradiation.

Fig. 2. Frequencies of chromatid-type aberrations in fibro-blasts bystanders that were bystanders to α-irradiated fibro-blasts.

0

0.05

0.1

0.15

0.2

0.25

0 0.1 1

Dose (Gy)

Abe

rrat

ions

/Cel

l


29

rations in these cells were comparable to those observed inthe controls. As can be seen in Figures 1 and 2, chromatidaberration frequencies in all control populations remainedbetween 0.05 and 0.07 aberrations per cell. At 15 populationdoublings post irradiation, aberration frequencies in the irra-diated populations ranged from 0.07 to 0.09 per cell. At 20doublings, these frequencies were between 0.09 and 0.14 percell (Figure 1).

Similar increases in chromatid aberration frequencieswere also observed in the bystander cells (Figure 2). Aber-ration frequencies in the bystander populations were be-

tween 0.08 and 0.14 per cell at the time points assayed.Studies are ongoing to examine both irradiated and by-stander populations for chromosomal aberrations not readilyobserved by Giemsa staining (eg. Translocations) usingchromosome painting techniques.

The results presented here support the link between theradiation-induced phenomena of genomic instability and thebystander effect. Interestingly, the level of response at latercell generations is not distinguishable between irradiated andnon-irradiated bystander cells. n

Induction of DNA Repair and Signal TransductionProteins Triggered by Ionizing Radiation

in “Bystander” CellsAdayabalam S. Balajee and Charles R. Geard

The “bystander effect” (BE) is an interesting biologicalphenomenon that has attracted a great deal of attention in thefield of radiation biology in recent times. BE is the result ofthe ability of the cells directly affected by an agent to conveythe manifestation of the damage to other cells that are notdirectly targeted thereby eliciting a response similar to thatof targeted cells. Although the molecular mechanism of BEis poorly understood at this moment, this multifaceted phe-nomenon may play a significant role in the therapy of tu-mors and in carcinogenesis. A better understanding of themolecular steps involved in BE is desirable for modulationand evaluation of protocols which would improve the effi-cacy of radio and chemotherapy treatments.

The BE can be triggered either by direct cellular interac-tion with the targeted cells or through factors released fromthe targeted cells (1). Recent studies have suggested thatsome of the factors released from the targeted cells may be

dependent on the transactivation potential of p53 indicatinga potential role for signal transduction proteins in mediatingthe BE. The increased levels of sister chromatid exchange,micronuclei, p21 and p53 proteins in bystander cells indicatethe involvement of a complex DNA damage response path-way in mediating the BE. Identification and characterizationof the DNA repair and signal transduction proteins involvedin BE may help in understanding the molecular steps in-volved in this complex phenomenon. In an attempt to char-acterize the factors involved in the BE, we have studied theexpression of some of the DNA repair and signal transduc-tion proteins in human primary fibroblast cells using ioniz-ing radiation as a DNA damaging agent.

The primary fibroblast cell lines (MRC5 and WI38) de-rived from healthy individuals were obtained from CoriellCell repository, Camden, New Jersey. All the cells wereroutinely maintained in 2X Eagle’s minimal essential

Fig. 1. Delayed chromatid-type aberrations observed in α-irradiated human fibroblasts at 15 ( ) and 20 ( ) populationdoublings post-irradiation.

00.020.040.060.080.1

0.120.140.16

0 0.1 1 10

Dose (Gy)

Abe

rrat

ions

/Cel

l

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0 0.1 1 10

Dose (Gy)

Abe

rrat

ions

/Cel

l

Fig. 2. Delayed chromatid-type aberrations observed in fi-broblasts bystander to α-irradiated human fibroblasts at 15 ( )and 20 ( ) population doublings post-irradiation.


30

medium (E-MEM) supplemented with 15% fetal bovineserum, vitamins, essential amino acids, non-essential aminoacids and antibiotics (Gibco BRL). The cultures were main-tained at 37°C in a humidified 5% CO2 atmosphere. WI38cells in plateau phase were irradiated with 10 Gy of γ-raysusing a 137Cs source delivering a dose rate of 0.98 Gy/min(Gamma cell 40, Atomic Energy of Canada, Canada). Thecells were kept at 37°C for 1 hr and the conditioned mediumfrom irradiated cells was transferred to plateau phase MRC5cells. Soluble and insoluble proteins were isolated (2). Atdifferent post-incubation times, 6-8X106 treated and mocktreated cells were lysed for 8 min on ice in 500 µl of buffer I(10mM Tris-HCl, pH 7.4, 2.5 mM MgCl2, 1mM PMSF and0.5% Nonidet P-40). The cell lysates were centrifuged at3000 rpm for 5 min and the soluble proteins (supernatant)were transferred to a fresh tube. The pellet fraction contain-ing the detergent insoluble proteins was lysed for 20 minwith 200 µl of buffer II (25mM Sodium phosphate bufferpH7.4, 0.5M NaCl, 1mM EDTA, 0.5% Triton X-100, Glyc-erol 10%, 5 mM MgCl2 and 1 mM PMSF). The proteinswere recovered by centrifugation at 12,000 rpm for 5 min.Protein concentration was determined by Pierce protein as-say kit. Aliquots of 20 µg of soluble and insoluble proteinswere fractionated by 4-20% polyacrylamide gradient gelelectrophoresis and blotted onto PVDF membrane followingthe standard protocol (Novex). The membrane was incu-

bated with TBST (20 mM Tris-HCl, pH 7.4, 137 mM NaCl,0.2% Tween) buffer containing 5% NFDM for 60 min atroom temperature. Antibodies to APE (apurinic/apyrimidinicendonuclease), RPA (replication protein A; p32 subunit),p53 and c-fos were used at a dilution of 1:1000 in phosphatebuffered saline containing 0.2% Tween 20 and 5% non-fatdried milk and (PBST-5%NFDM) for 1h. The membranewas washed three times (5 min each) in PBST buffer. Themembrane was incubated with horseradish peroxidase- con-jugated (HRP) secondary antibody (Vector Laboratories,1:2000 dilution in PBST-5% NFDM) for 1h followed byrepeated washing in PBST buffer. The signal was detectedusing the chemiluminescence technique (ECL) following themanufacturer’s protocol (Amersham). Immunodetection ofactin (Mouse IgM, 1:2000 dilution; Oncogene) was used as apositive control for equal loading of the proteins in the gel.The band intensity was measured using Scion Image Pro-gramme (National Institutes of Health, USA). Immunofluo-rescence for APE, PCNA and p53 was carried out on by-stander MRC 5 cells grown on chamber slides. After differ-ent treatment times cells were extracted for 10 min at 4°Cwith buffer 1 and fixed in acetone:methanol (1:1). The cellswere immunostained with primary and FITC conjugatedsecondary antibodies using a standard protocol (3).

Increased levels of p53, PCNA, APE, c-fos and RPAwere observed in bystander MRC5 cells both by immuno-

Fig. 1. Immunoblot analysis of soluble and insoluble pro-teins in “bystander” MRC5 cells. (1a)Time course kinetics ofvarious DNA repair and signal transduction proteins in thesoluble and insoluble proteins. Histogram showing the foldincreases in the induction of APE, RPA, c-fos and p53 proteinsin the soluble (1b) and insoluble (1c) protein fractions isolatedfrom “bystander” MRC5 cells.

Fig. 1a.

Fig. 1b.

Soluble Insoluble

RPA p32

6 0.5 2 4 6 6 0.5 2 4 6

Post-incubation time (h)

0Gy 10Gy 0Gy 10Gy

Ref-1

c-fos

Phosphorylated

forms

p53

Fig. 1c.

Post-treatment time

Post-treatment time

Fold

incr

ease

ove

r con

trol

Fold

incr

ease

ove

r con

trol


31

fluorescence and western blot analyses. The induction wasquite rapid and occurred at the earliest time point of 30 minafter treatment of MRC5 cells with conditioned mediumfrom irradiated WI38 cells. Although all of the factorsshowed elevated expression in bystander cells, there wasvariation in the time course kinetics of their induction inboth the soluble and insoluble protein fractions. The induc-tion of the redox factor APE was quite significant in thesoluble protein fraction as compared to the insoluble frac-tion, which showed a number of phosphorylated forms in thehigh molecular weight regions. Immunofluorescence analy-sis also showed a 2-fold increase in the fluorescence inten-sity of APE as compared to the control cells treated withmedium of non-irradiated cells. Western blot indicated thatthe APE level in the soluble fraction peaked at 30 min aftertreatment and there was a 4 fold increase as compared to thecontrol cells. APE level gradually declined with increasingpost-incubation times yet showed a 2-fold increase overcontrol cells. The induction of c-fos, like APE, was quitesignificant in the soluble fraction and c-fos level peaked at30 min and persisted with the increasing post-treatmenttimes. RPA also showed a higher level of induction at 30min in both insoluble and soluble protein fractions, whichsubsequently declined with increasing post-incubation times.RPA in the insoluble fraction showed a bi-phasic kineticswith the highest induction at 30 min followed by a decline at4hr and a subsequent increase in induction at 6hr. The p53protein detected only in the insoluble protein fraction wasinduced at all the time points but the peak of induction wasat 4hr after treatment unlike RPA, c-fos and APE. The timecourse kinetics of RPA, APE, c-fos and p53 as well as theirrelative induction are shown in Fig. 1a, b and c. The immu-nofluorescence staining of APE in control and bystandercells is shown in Fig. 2.

The induction levels of APE and p53 in the proteins iso-lated from directly hit WI38 cells were found to be identicalto the “bystander” MRC5 cells. Attempts are being made tocharacterize and identify more DNA repair and signal trans-duction proteins that respond to BE. We are also planning toevaluate the potential of BE in a number of radiosensitivehuman, mouse and hamster cell lines that have mutations inAtaxia telangiectasia mutated (ATM) gene, DNA dependentprotein kinase (DNA-PK), Ku 70/80 and p53. Analysis ofthe response of these mutant cells to BE is expected to yieldvaluable information on the molecular mechanism of BE inmammalian cells. Additionally, we are also planning to em-ploy the comet assay to determine the type of DNA lesionsinduced by BE and the kinetics of repair of such lesions.This would be helpful to find out the cause(s) of genomicinstability manifested in the form of sister chromatid ex-change and micronuclei in the “bystander” cells. Identifica-tion of a redox factor, APE as a potential inducer in by-stander cells raises the possibility that modulation of theredox potential in the intra-nuclear environment may play akey role in mediating the BE. In addition to being a redoxfactor, APE is a critical enzyme in the base excision repairof oxidative DNA damage. Keeping this in mind, we wouldattempt to look at the response of other key base excisionrepair enzymes in the “bystander” cells.

References

1. Iyer R and Lehnert BE, Arch. Biochem. Biophys. 376:14-25, 2000.

2. Prosperi E, Stivala LA, Sala E, Scovassi AI and BianchiL, Exp. Cell Res. 205, 320-5, 1993.

3. Balajee AS and Geard CR, Nucleic Acids Res. 29:1341-1351, 2001. n

Fig. 2. Immunodetection of APE in “bystander” MRC5 cells.

APE (FITC)

Control Treated*, 30 min Treated*, 3h

*W138 cells were irradiated with 10 Gy of γ-rays and incubated for 1 hr. The medium from the irradiated cells weretransferred to confluent MRC5 cells. The cells were incubated for 3h, fixed and immunostained for AP-Endonucleaseantibody.

CELLULAR STUDIES

33

Ataxia Telangiectasia Fibroblasts with ExtendedLifespan through Telomerase Expression

Retain Their Cellular CharacteristicsLauren D. Wood,1 Sonu Dhar, Arun Gupta, Fred Levine,1 Jerry W. Shay,2 Jean J.Y. Wang1 and Tej K. Pandita

Ataxia telangiectasia is an autosomal recessive disordercharacterized by progressive cerebellar degeneration, pre-mature aging, growth retardation, gonadal atrophy, immuno-deficiency, high sensitivity to ionizing radiation (IR), geno-mic instability and cancer predisposition. Cells derived fromA-T patients require higher levels of serum factors, exhibitcytoskeletal defects and undergo premature senescence inculture. The premature senescence of primary cells from A-T patients make it difficult to study the cellular and molecu-lar basis of A-T defects in culture. Previously, several linesof immortalized human A-T cells have been created throughthe expression of SV40 T-antigen or infection with humanEBV. However, the presence of the oncoproteins has com-plicated the interpretation of the data obtained. The ATMgene product has a role in cell cycle progression followingirradiation; however, the majority of the oncoproteins util-ized to immortalize cell lines block the function of the p53and Rb genes, both critical in the establishment of cell cyclecheckpoint. Furthermore, previous studies have implicatedATM in activating the transcription factor NF-κB followingexposure to ionizing radiation. However, a recent report hasshown that IR does not activate NF-κB in primary A-T cells,and attributed the previous finding to the effect of T-antigen(1). These controversies emphasize the importance of usingprimary A-T deficient cells to study ATM-regulated path-ways. Towards this end, we ectopically expressed hTERT inA-T patient cells and extended their life span.

Cells infected with control or hTERT-expressing viruseswere passaged continuously following drug selection.Within 15 PD, uninfected and control-infected (hTERT-)GM02052 and GM05823 cells ceased proliferating. Thecells increased in size and exhibited an altered morphology,consistent with changes associated with replicative senes-cence. Moreover, the majority of the control cells,(hTERT-) GM02052 clones (PD 42, PD 40), stainedstrongly positive for senescence-associated β-galactosidaseactivity. The level of SA-β-Gal staining was similar to thatseen in senescent normal human fibroblasts. In contrast,hTERT-expressing A-T and normal cells that were at similaror greater population doublings proliferated to confluency,maintained their original morphology, and demonstratedmuch less SA-β-Gal activity.

Since accelerated loss of telomeres is reported in A-Tcells, we determined whether expression of hTERT couldcompensate for this loss. By Southern-blotting, we found 1 University of California, San Diego, Ca.2 University of Texas Southwestern Medical Center, Dallas, Tx.

differences in average TRF sizes when comparing DNAderived from cells with and without ectopic expression ofhTERT in A-T cells. However, this analysis only yields anapproximation of the population of TRFs generated, anddoes not monitor ends of individual chromosomes. We spe-cifically examined this question by performing in situ hy-bridization, which allows us to examine the telomeres ofindividual chromosome ends. FISH for telomeric repeats inmetaphase cells was done by using a telomere specific Cy3labeled (CCCTAA)3 peptide nucleic acid probe. Fifty meta-phase chromosome spreads from cells with (80 PD) andwithout (38 PD) hTERT were included and analyzed. Ahigher proportion of chromatid ends in A-T cells withouthTERT (about 8% per metaphase) have less telomere spe-cific fluorescent signals as compared to the A-T cells withhTERT.

To determine whether the activation of telomerase byectopic expression of hTERT influences the stability oftelomeres, we examined the frequency of cells with telomereassociations in A-T fibroblasts with and without hTERT. A-T fibroblasts (GM05823) with hTERT expression have 0.13telomere associations per metaphase compared to GM05823without hTERT, which has 0.135 telomere associations permetaphase. Another A-T fibroblast cell line (GM02052) has0.08 telomere associations per metaphase, compared to0.075 telomere associations per metaphase in GM02052without hTERT. These observations suggest the telomereend-association defect in A-T fibroblasts might not be a di-rect result of short telomeres since it was not corrected byhTERT.

A hallmark characteristic of A-T cells is the loss of DNAdamage induced cell cycle checkpoints. To determinewhether telomerase activity affects the DNA damage in-duced cell cycle response in A-T fibroblasts, hTERT+GM02052 cells were exposed to 8 Gy of ionizing radiation.Fourteen hours following exposure, cell cycle analysis wasperformed. Normal fibroblasts expressing hTERT showed asignificant decrease in S phase entry indicating G1 arrest,while there was not a significant decrease in S phase entry inirradiated hTERT+GM02052 cells. Similar results were ob-served in the hTERT+ GM05823 cells following 2 Gy ra-diation dose. The loss of the G1 checkpoint seen is similar tothe parental GM02052 and GM05823 cells indicating thatthis phenotype is preserved in hTERT immortalized A-Tcells.

In addition to the G1 checkpoint, ionizing radiationcauses a transient inhibition of DNA replication. Cells fromA-T individuals and Atm deficient mouse fibroblasts exhibit


34

radioresistant DNA synthesis (RDS) (2). Normal fibroblastsexpressing hTERT showed an inhibition in DNA synthesissimilar to their parental cells following IR exposure. In con-trast, both the parental and hTERT+A-T cells (GM02052and GM05823) showed radioresistant DNA synthesis. Theseresults indicate that hTERT expression extends the prolif-erative life span of A-T cells without altering the funda-mental phenotype, as characterized by loss of IR-inducedcell cycle checkpoints.

We further determined whether the defect in the repair ofchromosomes is retained in ectopically hTERT expressingA-T cells. No difference in the G1 type of aberrations wasfound between the A-T fibroblasts with and without hTERT.The levels of chromosome aberrations in the separate A-Tcell lines were four-fold higher than the normal cells,indicating the defective G1 repair is not corrected in A-Tcells by the ectopic expression of hTERT. We also foundthat the presence of telomerase had no effect on the numbersof G2 chromosomal breaks or gaps. These results indicatethat the chromosomal repair following IR is still defective inhTERT expressing A-T fibroblasts.

Cells derived from A-T patients are more sensitive toionizing radiation as compared to cells from normal patients.Both parental and hTERT expressing A-T fibroblasts(GM02052 and GM05823) were exposed to various doses ofionizing radiation. The presence of hTERT slightly im-proved the survival of the GM02052 cells but not ofGM05823. This probably reflects the fact that the GM02052cells were assayed only 8-10 doublings prior to senescence,

when clonogenicity in the absence of treatment is alreadycompromised in the hTERT- control cells.

We have demonstrated that expression of the catalyticsubunit of telomerase (hTERT) in primary A-T patient fi-broblasts can rescue the premature senescence phenotype.Ectopic expression of hTERT does not rescue the radiosen-sitivity or the telomere fusions in A-T fibroblasts. ThehTERT+AT cells also retain the characteristic defects incell-cycle checkpoints, and show increased chromosomedamage before and after ionizing radiation. Although A-Tpatients have an increased susceptibility to cancer, the ex-pression of hTERT in A-T fibroblasts does not stimulatemalignant transformation. These immortalized A-T cellsprovide a more stable cell system to investigate the molecu-lar mechanisms underlying the cellular phenotypes ofAtaxia-telangiectasia.

References

1. Ashburner BP, Shackelford RE, Baldwin Jr AS andPaules RS, Cancer Res., 59:5456-5460, 1999.

2. Barlow C, Hirotsune S, Paylor R, Liyanage M, EckhausM, Collins F, Shiloh Y, Crawley JN, Ried T, Tagle Dand Wynshaw-Boris A, Cell 86:159-171, 1996.

3. Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP,Morin GB, Harley CB, Shay JW, Lichtsteiner S, andWright WE, Science 279:349-352, 1998.

4. Painter RB, and Young BR, Proc. Natl. Acad. Sci. (USA)77:7315-7317, 1980. n

Spontaneously Immortalized Cell Lines Obtained fromAdult ATM Null Mice Retain Sensitivity to Ionizing

Radiation and Exhibit a Mutational PatternSuggestive of Oxidative Stress

B.M. Gage,1 D. Alroy,1 C.Y. Shin,1 O.N. Ponomareva,1 Sonu Dhar, Girdha G. Sharma,Tej K. Pandita, M.J. Thayer,1 M.S. Turker1

The mouse homolog, Atm (ataxia-telangiectasia mutant),was mapped to mouse chromosome 9C, which is syntenic tohuman 11q22-23. Atm shows 84% amino acid identity and91% similarity with ATM. Sequence comparisons revealedthat ATM and Atm are members of a family of genes in-volved in cell cycle regulation (TOR1, TOR2, and MEC1 ofSaccharomyces cerevisiae and rad3 of Schizosaccharomycespombe), telomere length monitoring (TEL1 of S. cerevisiae),meiotic recombination (MEC1 of S. cerevisiae and mei41 ofDrosophila melanogaster), and DNA repair (DNA-PKCS), 1 Oregon Health Sciences University, Portland, Or.

supporting a role for ATM in DSB repair and cell cyclecontrol. A mouse model of Ataxia-telangiectasia (A-T) cre-ated by several laboratories by disrupting the murine Atmlocus using gene targeting. The phenotype of mice homozy-gous for the disruption of Atm is remarkably similar to thehuman AT phenotype.

The study of A-T has benefited significantly from mousemodels. While these models have proven useful for in vivostudies, cell cultures from Atm null embryos have been re-ported to grow poorly and then senesce. We initiated pri-mary cultures from adult ears and kidneys of Atm homozy-gous mice. The ear cells are fibroblast-like in appearance

CELLULAR STUDIES

35

and the kidney cells are primarily epithelial, which demon-strates that immortalization can occur for at least two differ-ent cell types. The only modification used in creating theAtm cell lines was to supplement the early primary cultureswith medium conditioned from a permanent kidney epithe-lial cell line. However, because we did not try to create per-manent cell lines in non-conditioned medium it can not bestated definitely that this supplementation is required for theestablishment of the permanent Atm cell lines. Regardless,conditioned medium was not required for continued growthof the immortalized cultures. We found that these culturesimmortalized readily without loss of sensitivity to ionizingradiation and other Atm related cell cycle defects. Althoughwe did not grow the mass cultures for prolonged periods oftime to demonstrate unequivocally that these cultures wouldnever senesce, it is noted that the mutational study requiredtwo sequential cloning steps. The first step was to obtainsubclones that were used for the isolation of mutant cells andthe second was to expand the mutant cells for a molecularanalysis. These two cloning steps when combined required aminimum of 50 population doublings. A mutational analysisfor loss of expression of an autosomal locus showed thationizing radiation had a mutagenic effect. Interestingly,some spontaneous mutants exhibited a mutational patternthat is characteristic of oxidative mutagenesis. This result isconsistent with chronic oxidative stress in Atm null cells,which we reported recently. The results demonstrate that

permanent cell lines can be established from the tissues ofadult mice homozygous for Atm and that these cell lines willexhibit expected and novel consequences of this deficiency.

We have shown that permanent cell lines can be estab-lished readily from adult tissues of mice that are homozy-gous for Atm. These cells retain sensitivity to ionizing ra-diation and cell cycle defects characteristic of Atm cellsdemonstrating that they are useful for further study of theeffects of Atm deficiency. Previous studies demonstratedthat mice homozygous for Atm do not exhibit elevated mu-tant frequencies in vivo for kidney and ear tissues. This re-sults suggested that a mutator phenotype was not inherent invivo, at least in the absence of internal or external stresses. Acomplete molecular analysis of the extent and types of mu-tations that cause Aprt deficiency in vivo in the Atm nullmice was not possible because heterozygosity for polymor-phic markers on chromosome 8 was limited for most of themice tested. The only exception was for one mouse cell line,which did not yield a significant number of Aprt mutants butdid yield a kidney cell line that was used in recent studies.This cell line has a near diploid number of chromosomesincluding diploidy for chromosomes 8, which bears the Aprtgene. The observation of discontinuous loss of heterozygos-ity in spontaneous Aprt mutants derived from the Atm cellsprovides a new tool to study this interesting mutational pat-tern. n

Extension of Lifespan by Transfection of hTERT inNormal Human Mammary Epithelial Cells

Li Liu, Yong L. Zhao, Chang Q. Piao and Tom K. Hei

Primary human and other mammalian cells generallyhave a limited proliferative life span when cultured in vitroand enter a non-dividing state termed senescence, which ischaracterized by altered gene expression. The strict replica-tive life span of human cells are postulated to serve as animportant barrier to malignant transformation, indicating thatcancer cells must overcome this obstacle and reach replica-tive immortality before malignant transformation. The pre-dominant theory of senescence is a response to telomereshortening, which occurs as a result of incomplete replica-tion of the end of chromosomes. Human telomeres consistsof repeats of the sequence TTAGGG/CCCTAA at chromo-some ends. These sequences are synthesized by the ribonu-cleoprotein enzyme telomerase with its RNA templete. Te-lomerase is active in germ cells and most cancer cells, but itis not expressed in most human somatic cells. The humantelomerase reverse transcriptase subunit (hTERT) has beencloned recently. Ectopic expression of hTERT resulted inlife span extension or immortalization in various cells in-

cluding human fibroblasts, retinal pigment epithelial cells,endothelial cells and T lymphocytes etc. (1, 2). Here weshow that the life span of human mammary epithelial cells(HMECs) were extended by transfection of hTERT.

Normal human mammary epithelial cells (HMEC) werepurchased from Clonetics (Walkersville MD). The cellswere cultured in MEGM medium (Clonetics) In our labora-tory, the normal cells can only be passaged 20-25 populationdoublings in normal culture before reaching growing crisis.The cells at passages 5 were transfected with retrovial con-struct pBabest2 containing cDNA encoding hTERT obtainedfrom Dr. Vaziri (1). pBabe2 and control vector pBAbe weretransfected into the highly efficient and helper-free packag-ing cell line Phoenix-A (ATCC) cultured in 100mm dishes.When the cells reached approximately 80% confluence, 4 µgretroviral plasmid DNA in 10ml Medium were introducedby LipofectAMINE PLUS reagent (GibcoBRL), accordingto the manufacturer’s instructions At 48 hs post-transfection,the viral-containing medium was harvested and the virus


36

titre was determined using NIH3T3 cells. Titres of >3x106

transducing units per ml were obtained. The HMECs wereinfected with the viral supernatant of pBabest2 and pBabeseparately in the presence of 4µg/ml polybrene. The G418resistant colonies formed in both pBabest2 and pBabe in-fected cells after 2 weeks transfection, and were continu-ously cultured. Senescence occurred in the pBabe trans-fected cells (MvT) at the same time as normal cells, how-ever, the cells from mixed clones transfected with pBabest2(MhT) are continuously growing and have reached morethan 56 PDs with a normal morphology and exuberant pro-liferation (Fig. 1, 2). Telomerase activity (TRAP) was de-tected by TRAPEZE Telomerase Detection Kit (Intergen)according manufacturer’s instruction. The results showedthat telomerase activity could not be detected in HMEC andMvT cells, but MhT cells expressed a high levels of te-lomerase activity (Fig. 3). The hTERT mRNA expressionwas also analyzed by RT-PCR using primers HT-1 (5'-AAGTTCCTGCACTGGCTGATGAG-3') and HT-5 5'-TCGTAGTTGAGCACGCTGAACAG-3' to amplify a 377bp hTERT mRNA fragment (2). The results are consistentwith TRAP detection, no band was found in HMEC cells, a377bp band was appeared in MhT cells (Fig. 4). Telomerelengths was significantly elongated in MhT cells comparedwith MvT and HMEC cells examined by Telo TAGGG Te-

lomere length Assay kit (Roche Molecular Biochemical)according to the manufacturer’s instruction (Fig 5).

In conclusion, our results show that the life span ofHMECs can be extended up to more than 56 PDs withoutcrisis by ectopic expression of hTERT. Studies of gene ex-pression by cDNA array in MhT cells compared withHMECs and telomerase expression regulated by arsenite inthis cell model are underway.

References

1. Vaziri H and Benchimol S, Reconstitution of telomeraseactivity in normal human cells leads to elongation oftelomeres and extended replicative life span, Curr. Biol.8:297-282, 1998.

2. Meyerson M, Counter CM et al, hTERT2, the putativehuman telomerase catalytic subunit gene, is up-regulatedin tumor cells and during immortalization, Cell 90:785-795, 1997. n

Fig. 1. Cell morphology: 1. HMECs at passage 5 beforetransfection; 2. MhT cells at 56 PDs after transfection; 3. MvTcells at 10 PDs after transfection.

Fig. 2. Growth curve of MhT cells compared with MvTcells.

Fig. 3. Telomerase activity: 1.HMECs; 2. MvT cells; 3. The extractof MhT cells was heated by 65oC for10min.; 4. MhT cells.

Fig. 4. mRNA expression ana-lyzed by RT-PCR: 1. Molecularmarker; 2. HMECs; 3. MhT cells.

Fig. 5. Telomere length: 1. Mo-lecular marker; 2. HMECs; 3. MvTcells at 10 PDs after transfection; 4.MhT cells at 10 PDs after transfection.

CELLULAR STUDIES

37

Immortalization of Human Bronchial Epithelial Cellsby Expressing hTERT

Chang Q. Piao, Yong L. Zhao, Li Liu and Tom K. Hei

Carcinogenesis is a multi-step process requiring a num-ber of cellular changes. Primary human epithelial cells arehighly refractory to malignant transformation in vitro. Cel-lular immortalization is thought to be an important stepalong the pathway to cancer formation. It is of interest toimmortalize normal human epithelial cells for carcinogene-sis studies.

When normal human cells are cultured in vitro, theyproliferate only a limited number of times and enter a stateof growth arrest termed senescence. It has been proposedthat telomere shortening may act as a molecular clock thatdetermines the number of times that a cell has divided.When telomeres are shortened to a critical length, cellularsenescence occurs. Recent studies indicate that introductionof the core catalytic subunit of telomerase (hTERT) intonormal human fibroblast results in restoration of telomeraseactivity and life-span extension (1, 2). In this study, we in-troduced hTERT into normal human bronchial epithelialcells to monitor possible their extension of life-span andimmortalization.

Primary cultures of small airway epithelial cells (SAEC)were purchased from Clonetics (Walkersville, MD.) the cellswere cultured in SAGM medium (Clonetics). In our labora-

tory, these cells can only be passaged for 20-25 populationdoublings in normal cultures. Introduction of hTERT intoSAEC cells was achieved by retrovirus-mediated gene trans-fer (see Liu et al previous). Colonies formed in SAEC cellstransfected with the pBabet2 construct, but there was notcolony formation in the cells transfected with the vectorpBabe. All of the clones are resistant to 400µg/ml G418.One of the clones (SVT-1) was continuously cultured, andhas reached more than 150 PDs. This cells shows a highproliferative rate with population doubling time ~36 hs.Control cells transfected by the vector senesced at 20-25PDs when cultured under the same conditions. We concludethat SVT-1 has became an immortal cell line. The SVT-1cells express higher telomerase activity and increase in te-lomere length compared with parental cells detected by theTRAPEZE Telomerase Detection Kit (Intergen) and Te-loTAGGG Telomere Length Assay kit (Roche) accordingmanufacture’s instructions (Fig. 1, 2). Karyotype of SVT-1cells is near diploid (Fig. 3), and no p16 protein alterationswere found by western blotting analyzing (Fig. 4). The SVT-1 and their parental cells were irradiated with 3Gy doses ofγ-rays, and p21 expression was analyzed in the proteins iso-lated at 4 and 6 hrs post-irradiation by western blotting. p21

Fig. 1. Telomerase activity detected

by TRAPEZE Telomerase Detection Kit:1. Parental SAEC, 2. SVT-1cells wereheated in 85oC for 10 min telomerase wasinactive, 3. SVT-1 cells.

Fig. 2. Telomere length analyzed byTeloTAGGG Telomere Length Assay: 1.Molecular weight marker, 2. Positivecontrol from kit, 3.Negative control noDNA was added, 4. SAEC cells, 5. SVT-1cells.

Fig. 3. Karyotype of SVT-1.


38

expression showed the same increase in parental and SVT-1cells (Fig. 5). This result indicates that p53 function is nor-mal in the SVT-1 cells. The SVT-1 cells are anchorage de-pendent by soft agar analysis and non-tumorigenic in nudemice. This cell line will be useful in studying mechanisms ofcarcinogenesis and telomere function in human bronchialepithelial cells. Studies of malignant transformation of SVT-1 cells by arsenite are underway.

References

1. Vaziri H, Benchimol S, Current Biology 8:279-82, 1998.2. Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP,

Morin GB, Harley CB, Shay JW, Lichtsteiner S andWright WE, Science 279:349-352, 1998. n

Monitoring Tumor Progression in a Radiation andEstrogen-Induced Breast Cancer Model

Gloria M. Calaf, Debasish Roy and Tom K. Hei

Breast cancer is a complex disease in which numerousgenetic aberrations occur. It is well accepted that cancerarises in a multistep fashion where exposure to environ-mental carcinogens is a major etiological factor. It is unclearwhich of these abnormalities are causative of breast tumori-genesis. It is possible that specific abnormalities may berequired in the progression from a normal to an invasivetumor cell line, therefore, knowledge of specific geneticchanges is critical to an understanding of the disease. Sev-eral experimental and clinical studies have focused on therole of certain genes in mammary tumorigenesis. Thus, agreat deal of progress has been made in understanding theoncogenes which are involved in the pathogenesis of breastcancer, allowing the development of new markers for prog-nosis and potentially the identification of new targets fortherapy.

Fra-1 (closely related to c-fos). It is an immediate earlygene, the expression of which is induced by growth factorsand mitogens (1). The MDM2 gene appeared as a new typeof proto-oncogene through the ability of its product to neu-tralize the activity of tumor suppressors (2, 3). The maintumor suppressor protein targeted by MDM2 is p53 which itis known to be mutated in more than 50% of human tumorsand is a key element in the protection against tumor devel-opment. PTEN and tsg 101 are tumor suppressor genes

which exhibit a high frequency of mutations in cancer, in-cluding breast (4, 5).

The aim of the present work was to analyze the expres-sion of oncoproteins that are frequently altered in breast can-cer in a radiation-induced and estrogen-dependent cancermodel. We developed a model in which the spontaneously-immortalized MCF-10F breast epithelial cells were irradi-ated with high-LET α-particles (150 keVµm) (6). Such cellswere transformed by two doses of 60 cGy α particles in thepresence of 17β estradiol and showed gradual phenotypicchanges including altered morphology, increased cell prolif-eration, anchorage independent growth and invasive capa-bilities and tumorigenicity in the nude mice (Table I).

Quantification of the immunofluorescent image and im-munoperoxidase protein staining showed an increase in Fra-1 and MDM2 protein expression (Figures 1 and 2) that wasdetected in all the irradiated populations when comparedwith the control MCF-10F cells. Such an increase was irre-spective of E treatment. However, there was no significantdifference in the levels of Fra-1 and MDM2 expression be-tween cells irradiated with either a single or double 60cGydose of α particles with or without E-treatment. The tumori-genic cell line 60cGy+E/60cGy+E and Tumor 2 had signifi-cantly (P<0.05) greater protein expression than the other celllines.

Fig. 4. p16 protein expression by Western blotting: 1. Pa-rental SAEC , 2. SVT-1 cells. Fig. 5. p21 protein expression by Western blotting: 1. Pa-

trental SAEC, 2. SAEC from 4 hrs post-irradiation, 3. SAECfrom 6 hrs post-irradiation, 4. SVT-1 cells, 5. SVT-1 from 4 hrspost-irradiation, 5. SVT-1 from 6 hrs post-irradiation.

CELLULAR STUDIES

39

There was a decrease in the tsg101 and PTEN proteinexpression (Figures 3 and 4) in the irradiated cell lineswhether or not the cells were transformed or tumorigenic incomparison to MCF-10F cell line. Figures 3 and 4 representthe protein expression of the tsg101 and PTEN protein ex-pressions of irradiated MCF-10F cells with or without treat-ment with estrogens. The tumorigenic cell line60cGy+E/60cGy+E and Tumor2 cell lines showed a signifi-cant decrease (P< 0.05), in the expression level of these twosuppressor genes in comparison to the control MCF-10Fcells and other irradiated cells.

In summary, these results showed a progression in phe-notypic and molecular characteristics, as cells became tumo-rigenic from the effect of high-LET radiation and estrogens.Such changes can be used to monitor the events indicative oftransformation in human breast epithelial cells.

References

1. Philips A, Teyssier C, Galtier F, Rivier-Covas C, ReyJM, Rochefort H and Chalbos D, Fra-1 expression levelmodulates regulation of activator protein-1 activity byestradiol in breast cancer cells, Mol. Endocrinol. 12:973-985, 1989.

2. Haines DS, Landers JE, Engle LJ and George DL, Physi-cal and functional interaction between wild-type andmdm2 proteins, Mol. Cell. Biol. 14:1171-1178, 1994.

3. Chen CY, Oliner JD, Zhan Q, Fornave Jr AJ and Vogel-stein B, Interactions between p53 and MDM2 in amammalian cell cycle checkpoint pathway, Proc. Natl.Acad. Sci. (USA) 91:2684-2688, 1994.

4. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI,Puc J, Miliaresis C, Rodgers L, Mc Combie R, BignerSH, Giovanella BC, Ittmann M, Tycko B, Hibshoosh H,Wigler MH and Parsons R, PTEN, a putative protein ty-rosine phosphatase gene mutated in human brain, breast,and prostate cancer, Science 275:1943-1947, 1997.

5. Li L and Cohen SN, Tsg101: a novel tumor susceptibilitygene isolated by controlled homozygous functionalknockout of allelic loci in mammalian cells, Cell 85:319-329, 1996.

6. Calaf G and Hei TK, Establishment of a radiation andestrogen-induced breast cancer model, Carcinogenesis21 (4):769-776, 2000.

(Figures 1-4 are on the following page.)

Table I.

Cel

l lin

es

Anc

hora

geIn

depe

nden

cy2

Inva

sion

3

Tum

orig

enic

ity4

MCF10F - - -

MCF10F+E - - -

60cGy1 + + -

60cGy+E + + -

60cGy/60cGy + + -

60cGy/60cGy+E + + -

60cGy+E/60cGy+E ++ ++ ++

Tumor 25 ++ ++ ++

1. Dose in cGy; number of exposures; estradiol 17 β treat-ment (10-8M)(E).

2. Colony-forming efficiency in agar fluctuated from 1-3%.3. Invasive characteristics of control and treated MCF-10F

cells were scored 20 h after plating onto matrigel basementmembranes using modified Boyden's chambers constructedwith multi-well cell culture plates and cell culture inserts.Positive signs represent the results in relation to the numberof cells that crossed the filters.

4. Positive signs represent the results in relation to the num-ber of animals that formed tumors in the nude mice.

5. Tumor 2: Cell line derived from 60cGy+E/60cGy+E afterinjection in the nude mouse.


40

4

MC

F-10

F

MC

F-10

F+E

60cG

y

60cG

y+E

60cG

y/60

cGy

60cG

y/60

cGy+

E

60cG

yE/6

0cG

yE

Tum

or 2

0

20

40

60

80

100

120

140 PTEN (Index of peroxidase)

PTEN(fluorescence)

Rel

ativ

e in

tens

ity

PTE

Npr

otei

n ex

pres

sion

MC

F-1

0F

MC

F-10

F+E

60cG

y

60cG

y+E

60cG

y/60

cGy

60cG

y/60

cGy+

E

60cG

y+E

/60c

Gy+

E

Tum

or 2

0

20

40

60

80

100

120

140tsg 101

Inde

x of

pero

xida

sest

aini

ng

tsg

101

prot

ein

expr

essi

on

Rel

ativ

e in

tens

ity

MD

M2

prot

ein

expr

essi

on

2

MC

F-10

F

MC

F-10

F+E

60cG

y

60cG

y+E

60cG

y/60

cGy

60cG

y/60

cGy+

E

60cG

yE/6

0cG

yE

Tum

or 2

0

50

100

150

200

250

300MDM2 (Index of peroxidase)

MDM2 (fluorescence)

1 MC

F-1

0F

MC

F-10

F+E

60cG

y

60cG

y+E

60cG

y/60

cGy

60cG

y/60

cGy+

E

60cG

y+E

/60c

Gy+

E

Tum

or 2

0

50

100

150

200

250

300FRA-1(Index of peroxidase)

FRA-1 (fluorescence)

Inde

x of

pero

xida

sest

aini

ng

and

rela

tive

fluor

esce

nce

inte

nsity

FR

A-1

pro

tein

exp

ress

ion

3

Figs. 1-4. Quantification of immunofluorescent imaging of Fra-1 (Fig. 1), MDM2 (Fig. 2), tsg (Fig. 3) and PTEN (Fig. 4) proteinexpression expressed by α-particle-irradiated and estradiol 17 β-transformed human breast epithelial cells lines after being exposed toretinol. Protein expression was determined by immunofluorescent staining, visualized by using confocal microscopy, and quantified bya computer program that gives the area and the intensity of the staining. Index of peroxidase is determinated by considering the inten-sity and the percentage of cells stained with peroxidase staining. n

CELLULAR STUDIES

41

Effect of Retinol on Radiation and Estrogen-InducedNeoplastic Transformation of Human Breast

Epithelial CellsGloria M. Calaf, Nancy J. Emenaker1 and Tom K. Hei

There have been a number of investigations in animalmodels as well as epidemiological studies in humans, dem-onstrating the chemoprotective effects of selected vitamins(1-3). Clinical, epidemiological and experimental findingshave provided evidence of a role for free radicals in the eti-ology of cancer (4, 5). It is enhanced in many disease statesand by carcinogen exposure contributing widely to cancerdevelopment. The free radical-scavenging vitamins, such asall-trans-retinol (retinol) have been shown to protect againstcancer (1). The aim was to understand whether or no freeradicals can participate in the breast carcinogenic process.

We developed a model in which the spontaneously im-mortalized MCF-10F breast epithelial cells were irradiatedwith high-LET α-particles (150 keVµm) (3). Such cells weretransformed by two doses of 60 cGy α-particles in the pres-

1 Department of Physiology and Cellular Biophysics.

ence of 17-β estradiol and showed gradual phenotypicchanges including altered morphology, increased cell prolif-eration, anchorage independent growth and invasive capa-bilities and tumorigenic in the nude mice (Table I).

Here we report the effect of retinol on the production ofH2O2 in the MCF-10F and irradiated and E-treated cells.MCF-10F cells irradiated with two doses of 60cGy α-particles in the presence of E (60cGy+E/60cGy+E) and thetumor cells derived from such cell line, named Tumor 2 aswell as the line derived from the latest. Each of the cell typesexamined had significantly (p<0.001) elevated H2O2 pro-duction levels compared to MCF-10F control cells by usingAmplex Red assay. Retinol treatment reduced such produc-tion in all cell types, including the MCF-10F control(p<0.001) (Figure 1).

The expression of oncoproteins was altered after retinoltreatment (Figures 2-5). There was a decrease in PCNA(Figure 2), mutant p53 (Figure 3) and Fra-1 (Figure 4) pro-tein expression and up regulation of RB (Figure 5) in all thecell lines in comparison to MCF-10F control, as detected byimmunofluorescent staining and quantified by confocal mi-croscopy. These studies showed that retinol treatment re-duced free radical production and induced alterations at theprotein level on a high LET radiation and estrogen-inducedbreast cancer model.

Fig. 1. Quantification of H2O2 (concentration µM) pro-duced by α-particle-irradiated and estradiol 17-β-transformedhuman breast epithelial cells lines in the presence of retinol.

Table I.

Cel

l Lin

es

Anc

hora

geIn

depe

nden

cy2

Inva

sion

3

Tum

orig

enic

ity4

MCF10F - - -

MCF10F+E - - -

60cGy+E/60cGy+E1 + + +

Tumor 25 ++ ++ ++

Tumor 36 ++ ++ ++

1. Dose in cGy; number of exposures; estradiol 17-β treat-ment (10-8M)(E).

2. Colony-forming efficiency in agar fluctuated from 1-3%.3. Invasive characteristics of control and treated MCF-10F

cells were scored 20 h after plating onto matrigel basementmembranes using modified Boyden's chambers constructedwith multi-well cell culture plates and cell culture inserts.Positive signs represent the results in relation to the numberof cells that crossed the filters.

4. Positive signs represent the results in relation to the numberof animals that formed tumors in the nude mice.

5. Tumor 2: Cell line derived from 60cGy+E/60cGy+E afterinjection in the nude mouse.

6. Tumor 3: Cell line derived from Tumor 2 after injection inthe nude mouse.

H2O

2pr

oduc

tion

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8No Retinol Plus Retinol

H2O

2pr

oduc

tion

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8No Retinol Plus Retinol


42

References

1. Moon RC, Vitamin A, retinoids and breast cancer, Ad-vanced Experimental and Medical Biology 364:101-107,1993.

2. Gerster H, Anticarcinogenic effect of common carote-noids, International Journal of Vitamins and NutritionalResearch 63:93-121, 1993.

3. Van Poppel G, Carotenoids and cancer: an update with

emphasis on human intervention studies, EuropeanJournal of Cancer 29:1335-1344, 1993.

4. Ames BN, Shigenaga MK and Hagen TM, Oxidants,antioxidants, and the degenerative diseases of aging,Proceedings of National Academy of Science (USA)90:7915-7925, 1993.

5. Calaf G and Hei TK, Establishment of a radiation andestrogen-induced breast cancer model, Carcinogenesis21 (4):769-776, 2000. n

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

50

100

150

200

250

300

350No Retinol (Fra-1)

Plus Retinol (Fra-1)

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

50

100

150

200

250

No Retinol Plus Retinol

Rel

ativ

e flu

ores

cenc

e in

tein

tens

ity

Rb

prot

ein

expr

essi

on

Rel

ativ

e flu

ores

cenc

e in

tein

tens

ity

Fra

-1 p

rote

in e

xpre

ssio

n

4 5

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

50

100

150

200

250

300

350No Retinol Plus Retinol

Rel

ativ

e flu

ores

cenc

e in

tein

tens

ity

PC

NA

pro

tein

exp

ress

ion

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

50

100

150

200

250

300No Retinol (p53)

Plus Retinol (p53)

Rel

ativ

e flu

ores

cenc

e in

tein

tens

ity

p53

pro

tein

exp

ress

ion

Fig. 2-5. Quantification of immunofluorescent imaging of PCNA (Fig 2), p53 (Fig. 3), Fra-1 (Fig. 4) and Rb (Fig. 5), protein ex-pression expressed by α-particle-irradiated and estradiol 17β-transformed human breast epithelial cells lines after being exposed toretinol. Protein expression was determined by immunofluorescent staining, visualized by using confocal microscopy, and quantified bya computer program that gives the area and the intensity of the staining.

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

50

100

150

200

250

300

350No Retinol Plus Retinol

Rel

ativ

e flu

ores

cenc

e in

tein

tens

ity

PC

NA

pro

tein

exp

ress

ion

MC

F-10

F

MC

F10F

+E

60cG

y+E

/60c

Gy+

E

Tum

or 2

Tum

or 3

0

50

100

150

200

250

300No Retinol (p53)

Plus Retinol (p53)

Rel

ativ

e flu

ores

cenc

e in

tein

tens

ity

p53

pro

tein

exp

ress

ion

2 3

CELLULAR STUDIES

43

Susceptibility of Human Breast Tissue to NeoplasticChanges Induced by Organophosphorous Pesticides

Gloria M. Calaf, Gertrudis Cabello1 and Tom K. Hei

The World Health Organization has reported that breastcancer has become the most common cancer in womenthroughout the world. The cause for the majority of humantumors has been attributed to exposure to environmentalcarcinogens, tobacco smoking, pesticides, radon alpha parti-cles, ultraviolet light radiation, pollutants and drugs. Mortal-ity and incidence studies have reported slightly increasedrates for the following cancers: breast, non-Hodgkin's lym-phoma, Hodgkin's disease, multiple myeloma, leukemia,prostate, and ovary among others. Furthermore, in vivo andin vitro data have shown that environmental xenobiotics(e.g., DDT, polychlorinated biphenyl, 4-nonylphenol, 4-octylphenol and many others) can promote mammary cancer(1).

Parathion [Diethyl 0-(4-nitrophenyl)-phosphorothyoate)]is an organophosphorous pesticide that is widely used with ahigh potential for human exposure. We have recently re-ported that parathion and malathion induced 14.3% and24.3% of mammary carcinomas in Sprague Dawley rat, re-spectively (2). Since there is some controversy over muta-genic and/or genotoxic potential of pesticides for humans, itis important to evaluate its possible effects on carcinogenesisin the human breast by using in vitro systems.

Cancer in humans and animals is likely to develop via amultistage process and one of the steps in this sequenceprobably includes amplification of oncogenes and alterationsin tumor suppressor genes, resulting in alteration of proteinexpressions. A complex set of genetic abnormalities hasbeen detected in breast cancer, including amplification ofoncogenes and mutation of the tumor suppressor genes(BRCA1, BRCA2, and p53 among many others).

It is accepted that uncontrolled cell proliferation is thefirst step in cell carcinogenesis. Expression of c-myc is gen-erally found in association with cellular proliferation. Am-plification of c-myc is a common genetic alteration in breastcancer and one third of breast carcinomas are associatedwith enhanced c-myc mRNA expression. PCNA is a non-histone intranuclear protein that serves as a cofactor forpolymerase delta during the DNA synthesis stage of the cellcycle. The detection reflects the proliferative activity of thecells. It is synthesized in early G1 and S phases of the cellcycle and is associated with the nuclear region where theDNA synthesis takes place. The activation or amplificationof c-Ha-ras oncogene has also been implicated in the pro-gression of human breast cancer. The p53, another importanttumor suppressor gene in normal cells is involved in regula-tion of the normal cell cycle; when it is inactivated, leads touncontrolled cell proliferation. Among other oncogenes im- 1 University of Tarapaca, Arica, Chile.

plicated in breast cancer c-ErbB2 has been found to be asso-ciated with the progression of the disease.

We tested parathion for the effects on cell proliferation,and invasiveness with the immortalized human breastepithelial cell line, MCF-10F. Our data indicated that para-thion significantly (P<0.05) increased cell proliferation after5 days in culture (Figure 1), and induced anchorage inde-pendency, as well as invasiveness (Figure 2) after 10 pas-sages. On the other hand, increased PCNA, c-myc, c-Ha-ras,p53 and c-erbB2 (Figure 3) protein expression of parathion-

0 2 3 4 50

10000

20000

0100 ng/ml300 ng/ml600 ng/ml

Days

Tot

al #

cel

ls/d

ish

Fig. 1. Effect of several doses of parathion (100, 300 and600 ng/ml) on cell proliferation of MCF-10F cells.

MCF-10F-control MCF-10F-parathion

0

500

1000

1500

2000

2500

Tota

l # c

ells

/filte

r

Fig. 2. Effect of parathion (100 ng/ml) on invasive capa-bilities of MCF-10F cells after 20 hrs in Boyden’s chamber.


44

treated MCF-10F cells in comparison to the control MCF-10F cell line, as shown by peroxidase staining. A tumori-genic cell line (cell line irradiated twice with 60cGy γ-particle in the presence of estrogens, 60+E/60cGy+E) wasused in these studies (3). Fluorescence was analyzed andquantified by confocal microscopy and quantified by com-putational studies corroborated the results of c-Ha-ras andmutant p53 protein expression (Figure 4).

References

1. Falck F, Ricci AJ, Wolf MS, Godbold J and Deckers P,Pesticides and polychlorinated biphenyl residues inhuman breast lipids and their relation to breast cancer,Arch. Environ. Health. 47:143-146, 1992.

2. Cabello G, Valenzuela M, Rudolph I, Hrepic N andCalaf G, A rat mammary tumor model induced by theorganophosphorous pesticides, parathion and malathion,possibly through acetylcholinesterase inhibition, EnvironHealth and Perspect 109(5):471-479, 2001.

3. Calaf G and Hei TK, Establishment of a radiation andestrogen-induced breast cancer model, Carcinogenesis21(4):769-776, 2000. n

Arsenic Induces Oxidative DNA Damagein Mammalian Cells

Maris Kessel,1 Su Xian Liu, An Xu, Regina Santella2 and Tom K. Hei

Arsenic, as trivalent arsenite (As3+) or pentavalent arse-nate (As5+), is naturally occurring and ubiquitously presentin the environment. Epidemiological data have shown thatchronic exposure of humans to inorganic arsenical com-pounds is associated with liver injury, peripheral neuropathy,and an increased incidence of cancer of the lung, skin, blad-der, and liver. However, the mechanism(s) underlying itscarcinogenicity remains unknown. The United States Envi- 1 The Bronx High School of Sciences, New York, N.Y.2 Department of Environmental Health Sciences, Joseph Mailman

School of Public Health, Columbia University.

ronmental Protection Agency has placed arsenic at the top ofits Superfund contamination list (1). Although the watersupplies in the United States are generally low in arsenic,there have been reports of arsenic contamination of groundwater in the Southwest with levels in the hundreds, and infew cases, more than 1,000 µg/liter, a level that is 20 timeshigher than the current U.S. maximum allowed level of 50µg/liter. Biologically, the trivalent sodium arsenite is signifi-cantly more active than the pentavalent sodium arsenate.

Arsenic is unusual because it is one of the few demon-strated human carcinogens that have not been shown to in-duce tumors in laboratory animals thus far. In the absence of

MCF-10F MCF-10F+Parathion0

50

100

150

200

250

300

c-Ha-ras mutant p53

Rel

ativ

e in

tens

ity p

rote

in e

xpre

ssio

n

(flu

ores

cenc

e st

aini

ng)

3000

MC

F-10

F

MC

F-10

F+Pa

rath

ion

60cG

y+E

/60c

Gy+

E

00.5

11.5

22.5

33.5

4

4.55

p53

PCNA

c-myc

c-erbB2

c-Ha-ras

R

elat

ive

inte

nsity

pro

tein

exp

ress

ion

(

pero

xida

se s

tain

ing)

Fig. 3. Effect of parathion on MCF-10F cells. Quantifica-tion of peroxidase staining of PCNA, c-myc, c-Ha-ras, p53, andc-erb B-2 protein expression, comparied to tumorigenic cells.

Fig. 4. Effect of parathion on MCF-10F. Quantificationimmunofluorescent staining of c-Ha-ras and p53 protein ex-pression visualized by using confocal microscopy.

CELLULAR STUDIES

45

an animal model to study the mechanisms of arsenic-inducedcarcinogenicity, in vitro studies have been conducted to pro-vide information on the cellular mechanisms involved. Al-though arsenic has been shown to be inactive as a gene mu-tagen at either the hypoxanthine guanine phosphoribosyltransferase (HPRT) or ouabain loci, it is a potent clastogen inmany cell types and induces sister chromatid exchanges andchromosomal aberrations in both human and rodent cells inculture. One possible scenario is that arsenic induces mostlymultilocus deletions that are incompatible with cell survivalwhen selected at gene loci located in closed proximity toother essential genes. Using the AL mutagenic assay systemwhich is sensitive in detecting multilocus deletions, we haveshown that this is, in fact, the case with the trivalent sodiumarsenite (2,3). At an equivalent dose of arsenite, the mutantyield at the CD59 locus was ~35 fold higher than the corre-sponding HPRT locus. In addition, the majority of theCD59− mutants induced were multilocus deletions. In thepresent study, we examined the effects of the antioxidantenzymes catalase and superoxide dismutase on mutagenicityof arsenite at the CD59 locus of the human hamster hybrid(AL) cells. We further characterized the induction of theoxidative DNA damage product, 8-hydroxydeoxyguanosine(8-OHdG) as well as the modulating effects of sulfhydryldepletion on the induction and types of mutations induced inarsenite treated cells.

Figure 1 shows the surviving fraction and induced CD59mutations in AL cells treated with either a 1.5 or 2.0 µg/mldose of sodium arsenite for 24 hr, with or without concurrentexposure to SOD (400 U/ml). The normal plating efficiencyof AL cells used in these studies ranged from 81 to 89%.Over the range of arsenic concentration examined, the doseresponse survival of AL cells was consistent with our previ-ously published data (2). Addition of SOD to the culturemedium had essentially no effect on the clonogenic survivalof control cells. In contrast, SOD treatment significantlyreduced the clonogenic toxicity of arsenic at both the 1.5µg/ml dose (0.31 ± 0.06 versus 0.44 ± 0.08 with SOD, p

<0.05) and 2 µg/ml dose level (0.18 ± 0.07 versus 0.62 ±0.13 with SOD, p<0.025). The average number of spontane-ous CD59− mutants per 105 survivors in AL cells used forthese experiments averaged 46 ± 10. Treatment of cells witha 1.5 and 2.0 µg/ml dose of arsenite resulted in induced mu-tant fractions (total mutant fraction minus background) of 95± 24 and 125 ± 35, respectively. While SOD treatment byitself induced no CD59− mutations, its presence in the cul-ture medium during arsenic treatment reduced the mutantfractions by 3.2 and 2 fold to 30 ± 8 and 64 ± 16, respec-tively for the 1.5 and 2.0 µg/ml dose treatment (Figure 1).

The possible contribution of hydrogen peroxide in thecytotoxic and genotoxic effects of arsenite was ascertainedusing catalase in the culture medium as shown in Figure 2.Similar to the findings with SOD, catalase, at a concentra-tion of 5000 U/ml, significantly reduced the clonogenic tox-icity of arsenic at both the 1.5 µg/ml dose (0.39 ± 0.06 ver-sus 0.84 ± 0.12 with SOD, p <0.025) and 2 µg/ml dose level(0.28 ± 0.07 versus 0.64 ± 0.09 with SOD, p<0.025). Like-wise, catalase treatment reduced the mutagenicity of arsenicin AL cells, being more effective at the lower dose of thenaturally occurring metalloid.

If generation of reactive oxygen species is one of themajor pathways for arsenic-mediated genotoxicity, then itshould be expected to induce specific DNA lesions consis-tent with oxidative damages. One of the most common oxi-dative DNA lesions is 8-hydroxy-2'-deoxyguanosine (8-OHdG). Using a monoclonal antibody specific for 8-OHdGcoupled with immunoperoxidase staining (4), we determinedthe formation of the oxidative DNA damage product in ALcells treated with a 4 µg/ml dose of sodium arsenite for 24hr. 8-OHdG was localized mainly in the nucleus of bothcontrol and fiber treated cells. Although a faint, backgroundstaining was evident in the control cultures, treatment of ALcells with arsenic resulted in a dose dependent increase in 8-OHdG levels. Quantification of staining intensity from 50-80 randomly selected cells treated with a 4 µg/ml dose ofarsenite indicated a 2.1 fold increase in staining intensity

Fig. 1. Effects of exogenous SOD (400 U/ml) on inducedmutant fractions in AL cells treated concurrently with gradeddoses of sodium arsenite for 24 hr. The survival fraction of thevarious treatment groups is shown above each bar. Data werepooled from 3-5 experiments. Error bars represent ± SD.

0

50

100

150

200

1 2 3 4 5Indu

ced

CD

59-M

utan

ts p

er 1

05Su

rviv

ors

S.F.=0.18

S.F.=0.31

S.F.=0.62

S.F.=0.44

S.F.=0.97

1. 2.0 µg/ml Arsenite2. 1.5 µg/ml Arsenite3. 2.0 µg/ml Arsenite + 400 U/ml SOD4. 1.5 µg/ml Arsenite + 400 U/ml SOD5. 400 U/ml SOD

0

50

100

150

200

1 2 3 4 5Indu

ced

CD

59-M

utan

ts p

er 1

05Su

rviv

ors

S.F.=0.18

S.F.=0.31

S.F.=0.62

S.F.=0.44

S.F.=0.97


0

50

100

150

200

250

300

1 2 3 4 5Indu

ced

CD

59-M

utan

ts p

er 1

05Su

rviv

ors

S.F.=0.28

S.F.=0.39

S.F.=0.64

S.F.=0.85

S.F.=0.96

1. 2.0 µg/ml Arsenite2. 1.5 µg/ml Arsenite3. 2.0 µg/ml Arsenite

+ 5000 U/ml Catalase4. 1.5 µg/ml Arsenite

+ 5000 U/ml Catalase5. 5000 U/ml Catalase

0

50

100

150

200

250

300

1 2 3 4 5Indu

ced

CD

59-M

utan

ts p

er 1

05Su

rviv

ors

S.F.=0.28

S.F.=0.39

S.F.=0.64

S.F.=0.85

S.F.=0.96




Fig. 2. Effects of exogenous catalase (5,000 U/ml) on in-duced mutant fractions in AL cells treated concurrently withgraded doses of sodium arsenite for 24 hr. The survival fractionof the various treatment groups is shown above each bar. Datawere pooled from 3 experiments. Error bars represent ± SD.


S.F.=0.18

S.F.=0.31

S.F.=0.62

S.F.=0.44

S.F.=0.97

S.F.=0.28

S.F.=0.64

S.F.=0.39

S.F.=0.96

S.F.=0.85





46

above background (Figure 3). In the presence of the antioxi-dant enzymes, the relative staining intensity decreased froman arbitrary unit of 333 to 207 and 212 in the presence ofSOD and catalase, respectively. SOD and catalase by them-selves, however, had little or no effect on the formation of 8-OHdG among control AL cells yielding a staining intensityin arbitrary units of 156 and 162, respectively.

Arsenic is a well-established human carcinogen basedon epidemiological studies. However, the mechanism(s)underlying its carcinogenicity remains unclear. The lack ofsuitable animal models as well as a poor understanding of itscarcinogenic/genotoxic mechanism hamper accurate riskassessment of the health effects of arsenite on both humansand animals, and necessitates reliance on in vitro studies toilluminate the cellular and molecular pathways involved.Using the human hamster hybrid (AL) cells assay that is effi-cient in the recovery of multilocus deletions, we showedpreviously that arsenic is indeed a potent gene and chromo-somal mutagen. Herein we present evidence that the muta-genicity is largely mediated by oxyradicals.

Reactive oxygen species such as superoxide anions, hy-droxyl radicals and hydrogen peroxides are the intermediatesformed during oxidative metabolisms. The observations thatantioxidants such as dimethyl sulfoxide, catalase, and N-acetyl cysteine reduce the in vitro biological activities of

arsenic suggest that active oxygen species may contribute tothe carcinogenic/mutagenic process of the metalloid. Thedeleterious effect of oxygen toxicity is normally held incheck by the delicate balance between the rate of generationof these radicals and their removal by various antioxidantenzymes. Superoxide dismutase catalyzes the dismutation ofsuperoxide anions while catalase removes hydrogen perox-ides and prevents the subsequent formation of hydroxylradicals. These hydroxyl radicals are far more damaging tocells than other radical species and have been associatedwith arsenic induced mutagenicity. Our findings that cata-lase and SOD can reduce the mutagenic potential of arsenicin mammalian cells are consistent with our recent findingsusing ESR spin trapping assay that arsenite increases thelevels of superoxide-driven hydroxyl radicals in mammaliancells (3).

A s a naturally occurring metalloid, arsenic is a seriousenvironmental concern world-wide, because of the largenumber of known contamination sites and millions of peopleat risk from drinking arsenic-contaminated water. A betterunderstanding of the mutagenic/carcinogenic mechanisms ofarsenic should provide a basis for a better interventionalapproach in both treatment and prevention.

References

1. USEPA, Soil Screening Guidance Technical BackgroundDocument, Office of Solid Waste and EmergencyResponse, EPA/540/R-95/128, United StatesEnvironmental Protection Agency, Washington, DC,1996.

2. Hei TK, Liu SX and Waldren CA, Mutagenicity ofarsenic in mammalian cells: Role of reactive oxygenspecies, Proc. Natl. Acad. Sci. (USA) 95:8103-8107,1998.

3. Liu SX, Athar M, Lippai I, Waldren CA and Hei TK,Induction of oxyradicals by arsenic: implications formechanisms of genotoxicity, Proc. Natl. Acad. Sci (USA)98:1643-1648, 2001.

4. Xu A, Wu LJ, Santella R and Hei TK, Role of reactiveoxygen species in the mutagenicity and DNA damageinduced by crocidolite fibers in mammalian cells, CancerResearch 59: 5615-5624, 1999. n

Role for Mitochondrial Oxidants as Regulators ofMutagenicity of Arsenite in Mammalian Cells

Su-xian Liu, Mercy M. Davidson1 and Tom K. Hei

The biological effect of arsenite, such as mutagenesis, 1 Department of Neurology.

are mediated by reactive oxygen species (ROS). The majorintracellular ROS intermediate is hydrogen peroxide, whichis synthesized from superoxide anion (•O2

− ) and further

Fig. 3. Effects of exogenous catalase (5,000 U/ml) on in-duced mutant fractions in AL cells treated concurrently withgraded doses of sodium arsenite for 24 hr. The survival fractionof the various treatment groups is shown above each bar. Datawere pooled from 3 experiments. Error bars represent ± SD.

0

100

200

300

400

1 2 3 4 5 6

Rel

ativ

e St

aini

ng I

nten

sity

of 8

-OH

dG 1. Control2. 400 U/ml SOD3. 5000 U/ml Catalase4. 4.0 µg/ml Arsenite 5. 4.0 µg/ml Arsenite

+ 400 U/ml SOD6. 4.0 µg/ml Arsenite

+ 5000 U/ml Catalase

0

100

200

300

400

1 2 3 4 5 6

Rel

ativ

e St

aini

ng I

nten

sity

of 8

-OH

dG 1. Control2. 400 U/ml SOD3. 5000 U/ml Catalase4. 4.0 µg/ml Arsenite 5. 4.0 µg/ml Arsenite

+ 400 U/ml SOD6. 4.0 µg/ml Arsenite

+ 5000 U/ml Catalase

1. Control2. 400 U/ml SOD3. 5000 U/ml Catalase4. 4.0 µg/ml Arsenite

5. 4.0 µg/ml Arsenite + 400 U/ml SOD6. 4.0 µg/ml Arsenite + 5000 U/ml Catalase

CELLULAR STUDIES

47

metabolizes into the highly reactive hydroxyl radical. In thisstudy, we examined the involvement of mitochondria in thearsenite-induced ROS and mutagenesis in AL cells.

The technique of cytochalasin B induced cell enucleationand fusion of mammalian cells in culture has provided aconvenient and direct technology to determine in a widevariety of systems if the mammalian cell cytoplasm canmodify nuclear gene activity.

Cytoplasm from tumor cells was shown to transfer tumo-rigenic properties when fused with karyoplasts from normalcells, suggesting that unknown cytoplasmic factors maysupport the formation of malignant phenotypes. Currentlythere is increasing evidence that mitochondrial mutations

and structural-functional abnormality are associated withvarious tumors. It has been shown that some types of chemi-cal carcinogens bind preferentially to mtDNA, suggestingthat mtDNA mutations might be involved in the carcino-genic process. If mitochondria are the critical target in medi-ating arsenite-induced genotoxicity, it will be able to detectCD 59− mutation may be expressed once treated cytoplastsare fused with the nuclei.

Exponentially growing AL cells (2x105) were plated onto35 mm dishes. After 24 hr incubation, the cells were enucle-ated by centrifugation at 6700 rpm for 21 min at 370C inmedium containing 10 µg/ml of cytochalasin B. Stainingwith a karyophilic dye revealed that enucleation efficiencywas at least 95%. The cytoplasts remain attached to the sub-strate, and the karyoplasts can be recovered in a pellet at thebottom of the centrifuge tube.

Examination of the ROS production by arsenite-treatedenucleated AL cells shows that a 2µg/ml dose of arsenite-exposed enucleated AL cells had a average fluorescence in-tensity per cell, 1.94 - 2.25 fold that of control during thefirst 20 min. and about 34%-46% (average 37.7%) relative toarsenite-treated whole cells.

To determine whether arsenic-treated cytoplasts (withoutnucleus) are capable of mediating mutagenicity at the CD59−

locus when rescued by fusion with karyoplasts (reconstruc-tion of cultured AL cells from cytoplasts and karyoplastparts), the cytoplasts were treated with arsenite 1.5 µg/ml for3.5 hr, then fuse with karyoplasts at a ration of 3:1 using50% polyethylene glycol (PEG). Figure 1 shows the muta-tion yield induced by arsenite-treated cytoplasts was 2.4 foldof the non-treated cytoplasts. Among the mutants induced byarsenite-treated cytoplasts the percentage of markers lostwere significantly increased, particularly for the loss of mul-tiple markers (Figure 2).

Results emerging from these studies, therefore, demon-strate that mitochondrial stress signaling in AL cells modu-lates nuclear gene expression. n

Fig. 2. Mutant spectra of CD59− mutants induced by fusion arsenite-treated cytoplasts with karyoplasts. Each line represents thespectrum for a single, independent mutant. Blank spaces depict missing markers. The absence or presence of markers among the mu-tants was determined by multiplex PCR.

0

1 0 0

2 0 0

3 0 0

4 0 0C

D59

mut

ants

per

10

su

rviv

ors

-5

Non-t reat ed cy t oplast s

Arsenit e-t reat ed cy t op last s(1 .5 µg / ml 3 .5 hr)

Figure 1. Mutation induction at CD59- locusFig. 1. Mutation induction at CD59− locus by fusion of ar-

senite-treated cytoplasts with karyoplasts. Each data point rep-resents an average of four experiments. Error bars represent ±SEM.

Arse nit e 0 .5 mg / ml 2 5 µM BSO

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

Non-t re at e d c y t op las t s arse nit e -t re a t e d c y t o plas t sNon-treated Cytoplasts Arsenite-treated CytoplastsArse nit e 0 .5 mg / ml 2 5 µM BSO

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

Non-t re at e d c y t op las t s arse nit e -t re a t e d c y t o plas t sNon-treated Cytoplasts Arsenite-treated CytoplastsArse nit e 0 .5 mg / ml 2 5 µM BSO

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

Non-t re at e d c y t op las t s arse nit e -t re a t e d c y t o plas t sArse nit e 0 .5 mg / ml 2 5 µM BSO

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

APO

-A1

CAT

WT

PTH

M1C

1(S1

)

RAS

Non-t re at e d c y t op las t s arse nit e -t re a t e d c y t o plas t sNon-treated Cytoplasts Arsenite-treated Cytoplasts


48

Development of a Flow Cytometric Assay forthe Quantification of CD59 Mutations in

Human-Hamster Hybrid (AL) CellsAn Xu, Haiying Hang, Hongning Zhou, Raheel Ansari and Tom K. Hei

The human-hamster hybrid (AL) cell system is a well-established in vitro model for detecting mutagens that inducemostly large, multilocus deletions, including ionizing irra-diation, heavy metals, and asbestos fibers (1-3). The AL cellscontain a standard set of CHO chromosomes and a singlehuman chromosome 11, which expresses several cell surfaceproteins including CD59 encoded by the CD59 gene at11p13.5. CD59 is a widely distributed, glycosylphosphati-dylinositol (GPI)-anchored cell surface protein, which actsas an inhibitor of complement (4). Because only a small partof 11p1 is required for the viability of AL cells, mutations inthe human chromosome 11 ranging in size up to 140 Mbp ofDNA can be detected. After exposure to mutagens, the sur-viving AL cells are either wild-type, CD59+, or mutated toCD59− . Using rabbit serum-complement plus anti CD59antibody, mutants lacking CD59 antigen can form colonies,while wild type cells are killed. However, it has been sug-gested that even mutated cells can maintain small amountsof wild type CD59 protein for several generations, whichresults in less efficient mutation detection. Furthermore, thespontaneous mutants are not easy to be removed. At the

same time, it usually takes 2-3 weeks to complete the wholeanalysis.

A modified mutation assay by flow cytometry has beenperformed to determine the yield of CD59− mutants. Forimmunophenotypical quantification of CD59− mutants, 106

cells were incubated with phycoerythrin-conjugated mousemonoclonal anti-CD59 antibody on ice for 30 to 60 mins.Then, the cells were rinsed with FACS buffer twice andfixed with 0.5% paraformaldehyde. Fluorescence of the cellswas then analyzed on a FACSCaliburTM instrument (BectonDickinson).

Figure 1 shows the mutation yields induced by gradeddoses of MNU in AL cells, which were determined by eithercomplement-antibody mutation assay or flow cytometry. Itwas shown that there was a dose dependent increase in themutation yield induced by MNU. However, the frequency ofCD59 mutants determined by flow cytometry was muchhigher than complement-antibody mutation assay, whichwas consistent with the findings of mutation induced by γ-rays (data not shown). This could be due to the fact that thedirect-labeled fluorescence assay by flow cytometry is moresensitive than complement-antibody mutation assay. Al-though there are several advantages to the flow cytometryassay, such as relative cheapness, rapidity, and no requiredcomplement, its still necessary to provide further evidence toestablish this method including setting up the gate and de-termining the mutation spectrum.

References

1. Waldren CA, Correll L, Sognier MA and Puck TT,Measurement of low levels of X-ray mutagenesis inrelation to human disease, Proc. Natl. Acad. Sci. (USA)83:4839-4843, 1986.

2. Liu SX, Athar M, Lippai I, Waldren C and Hei TK,Induction of oxyradicals by arsenic: implication formechanism of genotoxicity, Proc. Natl. Acad. Sci. (USA)98:1643-1648, 2001.

3. Hei TK, He ZY, Piao CQ and Waldren CA,Mutagenicity of mineral fibers, NATO ASI series A23:319-326, 1990.

4. Bodian DL, Davis SJ, Morgan BP and Rushmere NK,Mutational analysis of the active site and antibodyepitopes of the complement-inhibitory glycoproteinCD59, J. Exp. Med. 185:507-516, 1997. n

Concentration of MNU

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Indu

ced

mut

atan

ts p

er 1

05

0

500

1000

1500

2000

2500

Complement-antibody mutation assayFlow cytometry

Fig. 1. Effects of MNU on the mutation yield in AL cellsdetected by either complement-antibody mutation assay or flowcytometry.

CELLULAR STUDIES

49

Extranuclear Targets in the Genotoxicity of Asbestosin Mammalian Cells

An Xu, Suxian Liu and Tom K. Hei

Asbestos is a common name given to a group of hydratedfibrous silicates with unusual physical and chemical proper-ties, which have been widely used for industrial purposes.However, exposure to asbestos is known to associate withthe development of asbestosis, bronchogenic carcinoma, andmalignant mesosthelioma. Crocidolite is the most carcino-genic form of asbestos, containing up to 26% iron by weight.Based on the oxidation of the fluorescent probe 5',6'-dhloro-methyl-2',7'-dicholorodihydrofluorescein CM-H2DCFDA byreactive oxygen species (ROS) using confocal microscopy,previous studies have demonstrated the induction of ROS infiber-treated human-hamster hybrid (AL) cells. Furthermore,it has been shown that ROS were located mainly in the cyto-plasm, especially in spherical organelles, both in the controland fiber-treated cell pretreated with CM-H2DCFDA. Butthese experiments did not specifically identify the origin andthe types of these radicals.

It is generally accepted that mitochondria is the majorsource of intracellular ROS, which are enhanced by electrontransport inhibitors such as ischemia-reperfusion, rotenone,antimycin A, and diphenyleneiodonium. It is possible thatmitochondrial membrane damage induced by asbestos couldtrigger a cascading event in ROS production involving lipidperoxidation. Alternatively, peroxynitrite anions generatedas a result of mitochondrial damage can also be involved.Recent evidence has indicated that stimulation of an NADHor NADPH oxidase and/or conversion of xanthine dehydro-genase to xanthine in the cytoplasm can occur after contactof fibers with the plasma membrane and the subsequent up-take by the cells. At the plasma membrane-bound NDPHcomplex and oxidase of the respiratory chain, molecularoxygen is converted into superoxide anions, which is subse-quently converted into H2O2 and OH.

To evaluate the role of mitochondria in the muatagenic-ity of crocidolite, enucleated AL cells were exposed to croci-dolite at a dose of 4µg/cm2 in the presence or absence of0.5% DMSO for 3.5hrs. Then, enucleated AL cells werefused with non-treated nuclei. Figure 1 shows that the muta-tion yield induced by crocidolite in fused cells was morethan 2 fold that of the control, which could be dramaticallydecreased by concurrent treatment with DMSO. Further-more, using confocal scanning microscopy together withCM-H2DCFDA enucleated cells exposed to fibers exhibiteda higher fluorescent level when compared to the control,indicative of higher intercellular oxidant levels. These dataprovide a direct link between mitochondria and the muta-genicity of crocidolite in mammalian cells. n

Mutagenicity of Crocidolite Fibers in Mammalian Cells isAssociated with Nitric Oxide Production

An Xu and Tom K. Hei

Asbestos has been recognized as a human carcinogensince the 1950s, which results in restriction on its use in theUnited States. However, the danger of developing asbestosrelated diseases appears to extend beyond that of a simple

occupational hazard since it has been documented in familymembers of asbestos workers, in individuals living in theneighborhood of industrial sources of asbestos, and in someschools and public buildings where asbestos is being used as

Fig. 1. Mutation yields of CD59 induced by crocidolite inthe presence or absence of DMSO infused cells.

0

50

100

150

200

250

1 2 3 4

Indu

ced C

D59

- m

utan

ts p

er 1

05 sur

vivo

rs

1. Control2. Control + 0.5% DMSO3. Crocidolite4. Crocidoliete + 0.5% DMSO


50

insulation. Re-suspension of materials from asbestos con-taining ceilings has been shown to be the main source ofasbestos pollution in old, poorly maintained buildings. Thecontinued discovery of routes through which the generalpublic may be exposed to asbestos suggests a long-term, lowdosage exposure of a large number of people.

Crocidolite, the most carcinogenic form of asbestos, hasbeen shown to increase the production of nitric oxide (NO)in rat alveolar macrophages, human lung epithelial cells, andinterleukin-1β-stimulated rat mesothelial cells. However, therole of NO in the mutagenicity of crocidolite is not clear.NO is produced in many cell types and involved in a widespectrum of activities, such as control of blood pressure andnerve transmission. NO is synthesized through the conver-sation of L-arginine to citralline, which is catalyzed by nitricoxide synthases (NOS). In the presence of superoxide anion,NO is rapidly converted into nitrite, nitrate, and peroxyni-trite . In vivo and in vitro studies have suggested that highconcentrations of NO and its metabolites can damage pro-teins, lipids, and DNA. The latter include deaminated bases,DNA cross-links, oxidized bases, and single-strand breaks,which can lead to mutations (1).

To evaluate the contribution of NO to crocidolite muta-genesis, human-hamster hybrid (AL) cells were exposed tocrocidolite either in the presence or absence of NG-methyl-L-arginine (L-NMMA), which has been shown to competi-tively block the activity of NOS in various cell lines (2). It isshown in figure1 that there is a dose dependent decrease ofthe mutation yield induced by crocidolite in the presence ofgraded doses of L-NMMA. Concurrent treatment of AL cellswith crocidolite fibers at a dose of 4µg/cm2 and L-NMMA ata concentration of 15µM dramatically suppressed the muta-tion yield by more than 3-fold. Furthermore, concurrent ex-posure of AL cells with D-NMMA, the enantiomer of L-NMMA, which does not inhibit the function of NOS (3), hasless effect on the yield of mutants at equal doses (data notshown). The doses of both L-NMMA and D-NMMA usedhere have been shown to be nontoxic and nonmutagenic in

mammalian cells. These results strongly suggest that freeradicals (especially NO), as being mediators of the muta-genic response to crocidolite exposure. We further deter-mined the generation of NO from crocidolite treated cellsusing 2,3-diaminonaphthotriazole (DAN), which reacts withnitrite to form the fluorescent product 1H-naphthotriazole(4). As shown in figure 2, crocidolite fibers induced a dose-dependent increase in the production of NO. These data pro-vide further corroborating evidence for the mutagenic effectof crocidolite fibers on mammalian cells being mediatedthrough a pathway involved in NO production and modula-tion.

References

1. Zhuang JC, Lin C, Lin D and Wogan GN, Mutagenesisassociated with nitric oxide production in macrophages,Proc. Natl. Acad. Sci. (USA) 95:8286-8291, 1998.

2. Olken NM, Osawa Y and Marletta MA, Charaterizationof the inactivation of nitric oxide synthase by NG-methyl-L-arginine: evidence for heme loss, Biochem.33:14784-14791, 1994.

3. Rees DD, Palmer RM, Schulz R, Hodson HF, MoncadaS, Characterization of three inhibitors of endothelialnitric oxide synthase in vitro and in vivo, Br. J.Pharmacol. 101:746-52, 1990.

4. Fernandez-Cancio M, Fernandez-Vitos EM, Centelles JJ,Imperial S, Sources of interference in the use of 2,3-diaminonaphthalene for the fluorimetric determination ofnitric oxide synthase activity in biological samples, Clin.Chim. Acta. 312:205-212, 2001. n

1 2 3 4 5 6

Indu

ced

CD

59- M

utan

ts p

er 1

05 S

urvi

vors

0

50

100

150

200

250

1. Control2. 5 µM L-NMMA3. 15 µM L-NMMA4. Fibers5. Fibers + 5µM L-NMMA6. Fibers + 15µM L-NMMA

Fig. 1. Dose response of L-NMMA on the mutation yieldinduced by crocidolite in mammalian cells.

Control 10ug/ml 30ug/ml 60ug/ml

Rel

ativ

e flu

ores

cenc

e in

tens

ity

0

100

200

300

400

500

Fig. 2. Induction of nitric oxide by crocidolite in mammal-ian cells.

CYTOGENETIC STUDIES

51

Use of Multicolor Fluorescence in Situ Hybridization(mFISH) to Detect Radiation-Induced Chromosome

Aberrations in Human CellsM. Prakash Hande and David J. Brenner

Chromosome painting has been the technique of choiceto detect the symmetrical exchanges or chromosomal trans-locations induced by ionizing radiation. This technique usesfluorescence in situ hybridization protocol where smallnumbers of chromosomes are painted to visualize the simplechromosome exchanges. Though this is a very efficienttechnique to paint large number of cells, it is difficult toidentify the complex aberrations produced by exposure toionizing radiation. Therefore, the recently developed multi-color fluorescence in situ hybridization (mFISH, ref 1-3)techniques involving painting all the human chromosomes indifferent colors have circumvented the difficulties involvedin single or two-three color painting. mFISH involves acombinatorial painting technique that can be used to pseudo-color each chromosomal homologue a unique color, therebyeliminating many of the difficulties encountered whenpainting a limited subset of chromosomes. Until recently, itwas not possible to analyze intra-chromosomal rearrange-ments (paracentric/pericentric inversions) as standard FISHprobes as well as the probes used in mFISH and spectralkaryotyping (SKY) paint the entire chromosome in the samecolor. However, the recently developed mBAND FISH tech-nique yields high resolution multi-color banding based onregion-specific partial chromosome paints.

Cell culture and culture conditions

Human primary fibroblasts derived from four individuals(WI38, MRC-5, IMR-90 and IMR-91) were procured fromCoriell Cell Repository (Camden, NJ) and utilized in thestudy. Cells at early passages were grown in complete me-dium (E-MEM supplemented with 15% FBS, essentialamino acids, non-essential amino acids, vitamins and antibi-otics). Plateau phase cells were exposed to different doses ofgamma radiation (0, 0.5 Gy, 1 Gy, 2 Gy, 3 Gy, 4 Gy) andwere allowed to complete one cell cycle after irradiation andthen harvested for metaphase preparations using routine pro-cedures. Cells were harvested at 40 – 48 h after irradiationfollowing a 2-hour colcemid treatment. After hypotonicswelling (0.075 M KCl) for 15 minutes at 37°C, the cellswere fixed in methanol: acetic acid (3:1). Cell suspensionwas dropped on to a wet, clean slide for use in FISH. Forlong-term storage, the fixed cells were kept at –20°C untiluse. Based on the yield of metaphase chromosomes, two celllines were selected for further experimental analysis (WI38and MRC-5).

Analysis by multicolor FISH (mFISH)

The multicolor fluorescence in situ hybridization(mFISH) uses various fluorescence dyes to represent differ-ent painting probes at the same time. This offers the simulta-neous presentation of all 24 different human chromosomeswith a single hybridization. The detection of at least 24 dif-ferent chromosome painting probes is realized with fivevaricolored fluorochromes. Each paint is labeled with one ofthe five fluorochromes or a unique combination of them(combinatorial labeling). The separation of different excita-tion and emission spectra is guaranteed by appropriate filtersets (DAPI, DEAC, FITC, Spectrum Orange, Texas Red andCy5). The resulting unequivocal color signature for eachchromosome enables the analysis of hidden or complexchromosome aberrations was well as the composition ofmarker chromosomes. The procedure followed was that pro-vided by the manufacturer of the 24Xcyte mFISH kit (Meta-systems, Germany). A number of steps were taken to treatthe metaphase spreads prior to mFISH hybridization. TheFISH procedure as such is composed of four steps: The de-naturation of probe and target, the hybridization, the wash-ing and detection and then the analysis. In the last 6 months,the investigator has standardized the multicolor (24-color)FISH in the laboratory and the representative image from anormal human metaphase spread is shown in Fig. 1a. In thisfigure, the colors are merged and presented. Since it is notpossible to represent all colors in the image, the softwareassociated with the Metasystems multicolor system presentsthe same image in computer-generated pseudocolors (Fig.1b) whereby one can differentiate all the painted chromo-somes in different colors.

A complex chromosome aberration (dicentric-transloca-tion between three different chromosomes) following expo-sure to ionizing radiation in human fibroblasts is shown inFig. 2a (merged color) and in Fig. 2b (psuedocolor). mFISHrequires three to four days of hybridization per slide. Oncethe slides are hybridized, a minimum of 30 minutes is re-quired for the proper analysis of a single metaphase. For theanalysis of 100 metaphases, therefore, a minimum of 10days is necessary for a complete analysis.

Analysis by multicolor banding FISH (mBAND FISH)

The use of mBAND FISH (color bar code on a particularchromosome; ref: 4-5) was tested to determine the inter- andintra-arm chromosome exchanges. Region-specific partialchromosome paints were purchased from Metasystems


52

(Germany). Region-specific partial chromosome paints(RPCP) were generated by microdissection. For each regionspecific library eight to ten chromosome fragments wereexcised. The respective regions were isolated with extendedglass needles and the DNA was amplified by DOP-PCR.Altogether, seven overlapping microdissection DNA librar-ies of chromosome 5 were constructed, two within the p-armand five within the q-arm. Hybridization, post hybridizationwashes and signal detection of the 24-color mix as well as ofthe RPCPs was carried out following standard protocols.

Microscopic analysis was performed using an AxioplanII imaging microscope (Zeiss, Germany) equipped with anHBO 100 mercury lamp and filter sets for DAPI, DEAC,FITC, Spectrum Orange, Texas Red and Cy5. Images werecaptured and processed using the isis/mFISH imaging sys-tem (Metasystems, Germany). The software controls themotorized filter revolver as well as the excitation and emis-sion filter wheels, thus automating the capture process com-pletely. Once again, mBAND FISH requires three to four

days of hybridization per slide. At a time, two slides can beprocessed for mBAND FISH. That is the limitation of theprotocol. Once the slides are hybridized, a minimum of 30minutes is required for the proper analysis of a single meta-phase as explained above.

An example of mBAND FISH for human chromosome 5is given in Fig 3a (merged color). In Fig. 3b, the same imageshown in Fig 3a is presented as pseudocolor. Any change inthe banding pattern will be considered as aberration either inthe form of peri-centric or para-centric inversions. The irra-diated samples will be hybridized and will be analyzed forperi- and para-centric inversions.

References

1. Speicher MR, Ballard SG and Ward DC, Karyotypinghuman chromosomes by combinatorial multifluor FISH,Nat. Genet. 12:368-375, 1996.

2. Eils R, Uhrig S, Saracoglu K, Sätzler K, Bolzer A, Peter-

Fig. 2. Metaphase spread from irradiated human fibroblasts. a) True or merged color profile b) Pseudo color image: Arrow pointsto a complex chromosome exchange involving chromosomes 1, 14 and 17 (dicentric and non-reciprocal translocation). Note that thesame aberration cannot be visible clearly in the merged color profile but could be readily detected in the pseudocolor profile.

Fig. 1. Multicolor (24-color) FISH (mFISH): Metaphase chromosomes from normal human fibroblasts after multicolor FISH(mFISH). a) True or merged color profile from different fluorochromes used. b) Computer generated pseudocolor image of the samemetaphase spread. Note each chromosome is painted in a different color.

Fig. 1a Fig. 1b

Fig. 2a Fig. 2b

CYTOGENETIC STUDIES

53

sen I, Chassery JM, Ganser M, Speicher MR, An opti-mized, fully automated system for fast and accurateidentification of chromosomal rearrangements by multi-plex-FISH (MFISH), Cytogenet. Cell Genet. 82:160-171,1998.

3. Greulich KM, Kreja L, Heinze B, Rhein AP, WeierHUG, Brückner M, Fuchs P and Molls M, Rapid detec-tion of radiation-induced chromosomal aberrations inlymphocytes and hematopoietic progenitor cells by

mFISH, Mutat. Res. 452:73–81, 2000.4. Chudoba I, Plesch A, Lörch T, Lemke J, Claussen U,

Senger G, High resolution multicolor-banding:a newtechnique for refined FISH analysis of human chromo-somes, Cytogenet. Cell Genet. 84:156-160, 1999.

5. Johannes C, Chudoba I and Obe G, Analysis of X-ray-induced aberrations in human chromosome 5 using high-resolution multicolour banding FISH (mBAND), Chro-mosome Res. 7(8):625-633, 1999. n

Radiation Induced Inter-Arm Exchanges Detected byFluorescence in Situ Hybridization Using Chromosome

Single Arm-Specific ProbesM. Prakash Hande, Adayabalam S. Balajee, Charles R. Geard and David J. Brenner

Radiation-induced chromosomal aberrations are believedto be the result of mis- or non-repair of chromosome breaks.These breaks can be efficiently repaired and such a restitu-tion of chromosomes cannot be detected under the light mi-croscope at metaphase. Mis-repair or mis-joining of breaksleads to the formation of exchange aberrations betweenchromosomes (dicentrics and translocations) or within a

particular chromosome (rings and inversions) (for review:1,2). The primary motivations for study of chromosome ab-errations are the association of chromosome structuralchanges with cancer and their importance to the three mainapplications of radiobiology: biodosimetry, carcinogenesisrisk estimation and radiotherapy.

Fig. 3a

Fig. 3. Multicolor (24-color) banding FISH (mBAND FISH): Human chromosome 5 showing (a) merged and (b) pseudocolorbands following high-resolution multicolor banding fluorescence in situ hybridization. Any change in the location of the bands withinthe chromosomes following irradiation will be detected.

Fig. 3a Fig. 3b


54

Cell culture and culture conditions

Human primary fibroblasts derived from four individuals(WI38, MRC-5, IMR-90 and IMR-91) were procured fromCoriell Cell Repository (Camden, NJ) and utilized in thestudy. Cells at early passages were grown in complete me-dium (E-MEM supplemented with 15% FBS, essentialamino acids, non-essential amino acids, vitamins and antibi-otics). Plateau phase cells were exposed to different doses ofgamma radiation (0, 0.5 Gy, 1 Gy, 2 Gy, 3 Gy, 4 Gy) andwere allowed to complete one cell cycle after irradiation andthen harvested for metaphase preparations using routine pro-cedures. Cells were harvested at 40 – 48 h after irradiation

following a 2-hour colcemid treatment. After hypotonicswelling (0.075 M KCl) for 15 minutes at 37°C, the cellswere fixed in methanol: acetic acid (3:1). Cell suspensionwas dropped on to a wet, clean slide for use in FISH. Forlong-term storage, the fixed cells were kept at –20°C untiluse. Based on the yield of metaphase chromosomes, two celllines were selected for further experimental analysis (WI38and MRC-5).

Dose response curve (DAPI analysis)

Asymmetrical exchanges such as dicentrics, rings, frag-ments are being scored and compiled based on the DAPIstaining. The dose-response curve for dicentric frequencyobtained following irradiation is given in Figure 1. Dicen-tric frequency was detected based on the DAPI staining; thedicentric yield per cell followed a linear-quadratic model:ydic = 0. 0.07x2+0.003x+0.01 with an r2 value of 0.9976. Thelinear-quadratic model was the best fit for the dicentric fre-quency as was the case in earlier classical experiments,though there exists a controversy. Based on the aberrations,dicentrics, rings and fragments, detected, total aberrationsinduced followed a linear-quadratic function as well beingytotal = 0.17x2-0.02x+0.06 with an r2 value of 0.9986

FISH analysis using chromosome arm-specific probes

Chromosome arm specific painting probes (AmericanLaboratory Technologies, USA) were used in FISH to detectinter-arm exchanges. The technique has been optimized tostain the p and q arms of the selected chromosomes (e.g. 1,2, 3) in different colors to detect exchanges occurring be-tween the two arms of the chromosomes. In this analysis,

Fig. 1. Chromosome aberrations induced by gamma-radiation in human fibroblasts. Aberrations were detected in theDAPI stained metaphase chromosomes.

Fig. 2. Two color FISH: (a). Partial metaphase spread from normal human fibroblasts painted with arm specific probes for chromo-some 3. The p-arm is painted in green and the q-arm in red. Other chromosomes are visualized by the blue counter-stain DAPI. (b). Acomplete pericentric inversion (arrow) is seen in the partial metaphase spread from 3-Gy irradiated human fibroblasts.

0

0.5

1

1.5

2

2.5

3

0 1 2 3 4 5Radiation Dose (Gy)

Freq

uenc

y/ce

ll

dicentricsfragmentsringstotal aberrations

Fig.1

l

l

n

n

nnl

u

pppp

Fig. 2a Fig. 2b

u dicentrics n fragmentsp rings l total aberrations

l

l

l

n

nn

u

pppp

uu

u

Radiation Dose (Gy)

Freq

uenc

y/ce

ll

CYTOGENETIC STUDIES

55

any aberration involving a particular arm will be identified(inter-arm exchanges, translocations with other chromo-somes, color junctions, dicentrics and rings) and scored fortheir frequency following irradiation.

Arm specific probes for both the p and q arms of chro-mosomes were purchased from American Laboratory Tech-nologies. The FISH was done according to the manufac-turer’s instructions. Briefly, the slides were washed withPBS and denatured for 5 minutes at 75°C using 70% form-amide, 2 X SSC and 50 mM Phosphate buffer (pH 7.0). Hy-bridization was performed overnight at 37°C. Biotin wasdetected using avidin-FITC and biotinylated goat-anti-avidinantibodies. Detection of digoxigenin was achieved usinganti-digoxigenin TRITC. Finally, slides were embedded withVectashield mounting medium (Vector labs) containing 0.15µg/ml DAPI as blue counter-stain. The painted slides wereobserved using a Zeiss Axioplan 2 imaging fluorescencemicroscope.

In Fig 2a, a partial metaphase spread derived from anunirradiated sample shows homologous chromosomespainted with arm specific probe for chromosome 3. Dualcolor FISH using chromosome arm specific probes for 3p

(green) and 3q (red) on human chromosomes b) A meta-phase showing pericentric inversion between the p- (paintedgreen) and q-arm (painted red) of chromosome 3 followingexposure to 3 Gy of gamma radiation.

Frequencies of different types of aberrations involvingchromosomes #3 and #2 induced in human fibroblasts(WI38) by gamma-irradiation were detected using chromo-some arm specific probes (Tables I and II). Frequency per100 cells is given and a minimum of 500 cells per dosegroup was analyzed.

References

1. Savage JRK, Update on target theory as applied to chro-mosomal aberrations, Environ. Mol. Mutagen. 22:198-207, 1993.

2. Natarajan AT, Balajee AS, Boei JJ, Darroudi F, Domin-guez I, Hande MP, Meijers M, Slijepcevic P, VermeulenS and Xiao Y, Mechanisms of induction of chromosomalaberrations and their detection by fluorescence in situhybridization, Mutat. Res. 372:247-258, 1996. n

Table I. 3PQ

Translocations Pericentric InversionsRadiationDose (Gy)

AberrantCells Rec Ter Dic F Simple Complete Complex

Total ColorJunctions

0 0.38 0 0 0 0.38 0 0 00.5 1.65 0.50 0.33 0.33 0.66 0 0 0 1.651 3.91 1.36 1.02 1.36 2.21 0.34 0.17 0 5.952 6.75 2.08 2.08 2.25 3.11 0.35 0.35 0.52 10.553 10.54 3.51 2.99 4.57 5.10 0.53 0.88 0.88 18.284 15.52 5.92 4.48 4.64 7.68 0.8 1.44 0.8 26.40

Rec = Reciprocal Translocations; Ter = Terminal Translocations; Dic = Dicentric Chromosomes; F = Fragments.

Frequencies of aberrations involving chromosomes #3 induced in human fibroblasts (WI38) by gamma-irradiation as detected usingchromosome specific probes. Frequency per 100 cells is given and a minimum of 550 cells per dose group was analyzed.

Table II. 2PQ

Translocations Pericentric InversionsRadiationDose (Gy)

AberrantCells Rec Ter Dic F Simple Complete Complex

Total ColorJunctions

0 0.42 0 0 0 0.42 0 0 00.5 2.25 0.69 0.35 0.52 1.04 0 0 0 2.601 4.61 1.54 1.71 1.02 1.88 0.17 0.17 0 7.002 11.32 2.79 2.44 3.14 3.83 0.52 0.70 0.70 14.293 15.22 5.43 3.80 5.07 7.61 0.91 1.27 0.91 25.364 17.89 6.13 5.11 5.28 8.01 1.19 2.04 1.36 30.49

Rec = Reciprocal Translocations; Ter = Terminal Translocations; Dic = Dicentric Chromosomes; F = Fragments.

Frequencies of different types aberrations involving chromosomes #2 induced in human fibroblasts (WI38) by gamma-irradiation as de-tected using chromosome specific probes. Frequency per 100 cells is given and a minimum of 500 cells per dose group was analyzed.


56

Aberrant Hrad9 Expression Influences Telomere Behaviorand Ionizing Radiation-Induced Chromosomal Instability

Sonu Dhar, Wei Zheng, Girdhar G. Sharma, Kevin M. Hopkins, Howard B. Lieberman and Tej K. Pandita

The DNA damage checkpoint function pathways, similarto that of telomeres, are conserved among Saccharomycescerevisiae, Schizosaccharomyces pombe, Drosophila mela-nogaster, Caenorhabditis elegans, and mammals. In S.pombe, the protein products of several checkpoint rad genes(rad1+, rad3+, rad9+, rad17+, rad26+, and hus1+) play cru-cial roles in sensing changes in DNA structure and some ofthem are involved in maintenance of telomeres. Genomestability is also maintained by the telomeres, as these chro-mosome terminal structures protect the chromosomes fromfusions or degradation. The rad3 gene product, like Tel1 andMec1, is involved in telomere metabolism. Recently, it hasbeen shown that the hRad9 protein is constitutively phos-phorylated in undamaged cells and undergoes hyperphos-phorylation upon treatment with ionizing radiation (1). Thishyperphosphorylation is dependent on ATM. Since ATMand some of its down stream targets have been shown to beinvolved in telomere metabolism, we were interested to ex-amine whether aberrant hRad9 expression influences telo-mere stability and telomerase activity.

We used two different approaches to determine whetherhRad9 function influences telomere stability. First, we over-expressed hRad9 and monitored telomere fusions as well aschromosome aberrations before or after treatment with ion-izing radiation. We found that overexpression of hRad9 haddrastic effects on telomere stability. However, when mutanthRad9 was expressed in human cells, no effect on telomerestability was found, suggesting the lack of a dominant-negative effect. Furthermore, we examined whether overex-pression of hRad9 influences DNA repair; interestingly, wefound that cells overexpressing hRad9 have higher G2 typechromosomal aberrations. In contrast, cells with ectopicallymutant hRad9 expression did not show any difference inresidual DNA damage.

Differences in the frequency of chromosome end-to-endassociations between cells with and without overexpressionof hRad9 are statistically significant. Since chromosomeend-to-end associations may lead to anaphase bridge forma-tion, cells without colcemid treatment were analyzed for thistype of aberration. Cells overexpressing hRad9 displayed an8-fold higher frequency of anaphase bridges, as compared toparental cells.

These data suggest that due to occasional losses of telo-mere function, chromosome end associations are formed andthese associations are not resolved in cells with excesshRad9. To further examine how overexpression of hRad9 islinked with loss of telomere function, we monitored the sizesof terminal restriction fragments. By Southern-blotting, wefound no significant differences in TRF sizes. However, thisanalysis only yields an appraisal of the population of TRFs

generated, and does not monitor ends of individual chromo-somes. We therefore performed FISH for telomeric repeatsin metaphase cells by using a telomere specific Cy3 labeled(CCCTAA)3 nucleic acid probe. A significantly higher pro-portion of chromatid ends in cells with excess hRad9 (about11% of telomeres per metaphase) have less telomere specificfluorescent signals as compared to the parental cells. Theseobservations suggest that the chromosome end-to-end fu-sions observed in cells when hRad9 is overexpressed corre-lated with loss of telomeric repeats. However, telomere sig-nals were seen in about 18% of fusion sites, indicating thattotal loss of telomeres is not required for telomere fusions.

A difference in telomere end fusions might also be due toalterations in telomerase activity. Using the TRAP-ELISAassay method, no significant differences in telomerase activ-ity between cells with or without high levels of hRad9 werefound, indicating that the overall activity of telomerase is notaffected by overexpression of hRad9.

The present studies reveal that cells overexpressinghRad9 have frequent chromosome breakage with completeloss of telomeric repeats, as well as chromosome end fu-sions, ring chromosomes, reciprocal and unbalanced translo-cations. It is important to note that these chromosomal aber-rations occurred in cells with normal p53 function and undernormal growth conditions without genotoxic challenges. Thefact that the p53-mediated G1 checkpoint is operative inthese cells is underscored by our findings that γ-irradiationinduces comparable frequencies of observable chromosomeaberrations when cells overexpressing or not overexpressinghRad9 were irradiated while in G1. However, there is a clearand significant difference in such aberrations if log-phasecells were irradiated and mitoses, occurring shortly thereaf-ter, were monitored. At 90 min after irradiation, there is aseveral-fold difference in the frequencies of aberrations permetaphase between the cells. These observations suggestthat cells need a specific level of hRad9, or on the other handthat overexpression of hRad9 leads to aberrant cell cyclecheckpoints, resulting in loss of telomeres. This is similar tothat observed recently in cells deficient in the function ofanother cell cycle checkpoint gene, “14-3-3σ” (2).

References

1. Chen MJ, Lin YT, Lieberman HB, Chen G and Lee EY,ATM-dependent phosphorylation of human Rad9 is re-quired for ionizing radiation-induced checkpoint activa-tion, J. Biol. Chem. 276:16580-16586, 2001.

2. Dhar S, Squire JA, Hande MP, Wellinger RJ, PanditaTK, Inactivation of 14-3-3sigma influences telomere be-havior and ionizing radiation-induced chromosomal in-stability, Mol Cell Biol. 20:7764-7772, 2000. n

CYTOGENETIC STUDIES

57

Extra-chromosomal Telomeric DNA in Cells from Atm-/-

Mice and Patients with Ataxia-telangiectasiaM. Prakash Hande, Adayabalam S. Balajee, Andrei Tchirkov,1 Anthony Wynshaw-Boris2 and Peter M. Lansdorp1

Ataxia-telangiectasia (AT) is a pleiotropic inherited dis-ease characterized by neurodegeneration, cancer susceptibil-ity, immunodeficiency, genetic instability, radiation sensi-tivity and premature aging. ATM (Ataxia telangiectasiamutated), the gene responsible for AT, is thought to play acrucial role in a signal transduction network that modulatescell cycle checkpoints, genetic recombination, apoptosis, andother cellular responses to DNA damage. Cells derived fromAT patients (1) and ATM-deficient mice show slow growthin culture and premature senescence (2). Cells from AT in-dividuals display abnormalities in culture such as cytoskele-tal defects and hypersensitivity to ionizing radiation. Theyalso show chromosome abnormalities in the form of end-to-end associations involving telomeres (3). Telomeres, con-sisting of (TTAGGG)n repeats and associated proteins, pro-tect chromosomes from end fusions, incomplete replicationand from exonuclease degradation. Telomere shorteningmay play an important role in tumorigenesis and aging. Pe-ripheral blood lymphocytes from AT-patients show acceler-ated telomere shortening (4). In this study, we have evalu-ated the potential role of ATM gene in telomere function andchromosome integrity in wild type and ATM knockout mice.Our results indicate that the functional inactivation of ATMleads to telomere shortening, chromosome instability and theoccurrence of extrachromosomal fragments of telomericDNA suggestive of an important role for the mammalianATM gene in maintaining telomere integrity.

To study telomeres in mice lacking the ATM gene, telo-mere length was measured using quantitative fluorescence insitu hybridization (Q-FISH) (5) in metaphase chromosomesfrom primary cells derived from littermate Atm+/+, Atm+/- andAtm-/- mice (129/SvEv and 129/SvEv/BlSW genetic back-ground). Atm-/- splenocytes from 129/SvEv mice showedextensive telomere shortening (Fig. 1A) corresponding to a35-40% loss of telomeric DNA (from 44kb in wild type to27 kb in Atm-/- mice). Heterozygous Atm+/- mice also dis-played a significant reduction (20-25%) in telomere lengthcompared to their Atm+/+ littermates. A similar degree oftelomere shortening was observed in cultured fibroblastsfrom lung and skin of Atm-/- compared with wild-type Atm+/+

129/SvEv mice (not shown). Telomeres in cells from129/SvEv/BlSW Atm-/- splenocytes were very heterogeneousin length and also significantly shorter than in Atm+/+ con-trols (Fig. 1B). Chromosome ends without detectable repeatswere frequently observed in Atm-/- but not Atm+/+ cells ofmice from both genetic backgrounds (Fig. 1A,B). 1 Terry Fox Laboratory, British Columbia Cancer Agency, Van-

couver, BC.2 School of Medicine, University of California, San Diego, Ca.

To localize telomeres and to detect the number and dis-tribution of telomere signals in mitotic cells, fibroblasts de-rived from skin or lung tissues of both Atm+/+ and Atm-/-

mice were grown on coverglasses. Cells at a semi-confluentstage were fixed and hybridized with telomeric PNA probe.Deconvolution microscopy was used to visualize the telo-mere signals in the interphase nuclei of such cells. The re-constructed 3D images were used to count the number oftelomere spots in each nucleus. Surprisingly, Atm-/- mousefibroblasts showed more than the expected number of telo-mere signals in interphase nuclei (Fig. 2B, C). Cells fromAtm+/+ mouse displayed an average of 74.5 ± 2.0 telomeresignals in their nuclei, close to the expected 80 (40 chromo-somes, 2 ends each; Fig. 2A, D). In contrast, Atm-/- mousefibroblasts showed approximately 1.3-2 times more telomerespots depending on the threshold fluorescence intensity usedto detect telomere spots (127-181 spots). Possibly, the sur-plus telomere signals represent broken telomere fragments.Alternatively, a high proportion of Atm-/- interphase cellscould have been in S-phase or telomeres in Atm-/- cells couldbe replicated earlier than those in wild type cells. To studythese possibilities further, we performed parallel measure-ments of DNA and TTAGGG repeats by flow-cytometry.The results of this analysis were compatible with a modestincrease in the percentage of Atm-/- fibroblasts in the S- andG2/M phase of the cell cycle. The data were furthermoresuggestive of early DNA replication of a subset of telomeresin cells from both normal and Atm-/- mice.

To further study the excess number of telomere spotsseen in interphase cells, we performed telomere FISH onmetaphase chromosomes from fibroblasts that were fixed insitu to preserve cytoplasm (see Fig. 2). In the cells fromAtm-/- mice, we observed that more than 80% of the mitoticmetaphase cells displayed excess telomere signals that wereclearly dissociated from the bulk of chromosomal DNA (Fig.2C, D). Such extra-chromosomal spots observed were rarelyobserved in Atm+/+ cells (Fig. 2A, B). The presence of extratelomere signals that were not associated with chromosomesbut located within the cell was also noticed in 60-75% of themetaphases derived from bone marrow cells of Atm-/- mice.Extra chromosomal telomere signals observed in the meta-phase cells of Atm-/- mice correlated with the increased telo-mere spots observed in the interphase nuclei in these cells.The extra spots could represent broken telomeres or defec-tive replication intermediates such as single strand G-richrepeats. However, no signals were observed in Q-FISH ex-periments if heat denaturation was omitted from the proto-col, suggesting a double strand nature of the extra chromo-somal telomere signals.


58

Primary fibroblasts from three normal individuals andfour different AT patients were also analyzed to study thepresence of extra-chromosomal telomeric DNA in humancells. The fibroblasts were grown on chamber slides andhybridized with Cy3-labelled PNA-telomeric probe. Al-though the presence of extra-chromosomal telomeric DNAin cells from human AT patients was not as pronounced as incells from Atm-/- mice, about 40 to 50% of the cells analyzedshowed a significantly increased number of telomere fluo-rescence spots not associated with the chromosome endsrelative to normal fibroblasts. A metaphase spread from ahomozygous AT patient also shows the presence of extra-chromosomal telomeres in cells from both AT patients andAtm-/- mice supports an important role of the ATM gene inmaintaining telomere integrity. This notion is supported bycontrol experiments with PNA probes specific for alphoidsequences showing bright staining at centromeres but noextrachromosomal signals in AT deficient cells from eitherhuman or murine origin (results not shown).

The mechanism responsible for the presence of extra-chromosomal telomeric DNA in cells with defective ATMobserved in this study is not known. It is possible that theloss of ATM leads to defective replication of telomericDNA, which results in breakage of telomeric DNA. Oneintriguing possibility is that telomeric DNA is particularlyvulnerable to breakage. In the absence of ATM, broken sin-gle or double strand breaks in telomeres may not be detectedand/or repaired possibly resulting in the observed fragmentsand accelerated telomere shortening.

Our observations suggest that telomeric DNA fragments

accumulate in Atm-/- cells, possibly due to defective replica-tion, recombination or repair of telomeric DNA. ATM isclearly involved in the p53-mediated DNA damage responsefollowing telomere fusions and chromosome breakage inmammalian cells. ATM is also a part of p53- and p21-mediated cell cycle checkpoint pathways in response to dou-ble strand DNA breaks. ATM could then participate in themaintenance of telomeres and telomere length in culturedcells. Irrespective of the precise mechanisms involved, theexcess telomere signals observed in Atm-/- cells reported heresupport a major role of ATM in the replication and mainte-nance of the telomeric DNA in mammalian cells.

References

1. Shiloh Y, Ataxia-telangiectasia and the Nijmegenbreakage syndrome: related disorders but genes apart,Annu. Rev. Genet. 31:635-632, 1997.

2. Barlow C, Hirotsune S, Paylor R, Liyanage M, EckhausM, Collins F, Shiloh Y, Crawley JN, Ried T, Tagle D etal, Atm-deficient mice: A paradigm of Ataxiatelangiectasia, Cell 86:159-171, 1996.

3. Pandita TK, Pathak S and Geard CR, Chromosome endassociations, telomeres and telomerase activity in ataxiatelangiectasia cells, Cytogenet. Cell Genet. 71:86-93,1995.

4. Metcalfe JA, Parkhill J, Campbell L, Stacey M, Biggs P,Byrd, PJ and Taylor MR, Accelerated telomereshortening in ataxia telangiectasia, Nat. Genet. 13:350-353, 1996.

Fig. 1. (A) and (B) Frequency distributions of telomere fluorescence in 129/scEv (A) and 129/svEv/BlSW (B) mice with and with-out the ATM gene. Data were collected from Q-FISH studies on metaphase spreads of primary mouse splenocytes of the indicatedgenotype. The mean and standard deviation for each group of mice is shown. * Indicates p=<0.001 in Mann-Whitney Rank Sum testcomparing the telomere length distribution in cells from Atm-/- and Atm-/+ with that in cells from Atm+/+ mice. One Telomere Fluores-cence Unit corresponds to an estimated 1 kb of telomeric DNA.

Telomere Fluorescence (TFU)

A B 129/svEv 129/svEv/BISw

Freq

uenc

y

CYTOGENETIC STUDIES

59

5. Hande MP, Samper E, Lansdorp P and Blasco MA,Telomere length dynamics and chromosomal instability

in cells derived from telomerase null mice, J. Cell Biol.144:589-601, 1999. n

Fig 2. Telomere FISH using FITC labeled (CCCTAA)3 probe on murine metaphase fibroblasts grown on microscope slides. Intactmetaphase chromosomes are easily observed in Atm+/+ fibroblasts (A-B) and Atm-/- fibroblasts (C-D) using PI as a counterstain. Sepa-rately captured PI (A and C) and FITC images were superimposed (B and D). Note the abundant signals obtained with the telomere probethat are not associated with chromosomes within the cell boundary. Asterisks point to chromatid breaks and arrows could represent telo-meres at the breakage sites.

MOLECULAR STUDIES

61

Mrad9 Knockout Mouse ES Cells are Sensitive toIonizing Radiation

Howard B. Lieberman and Kevin M. Hopkins, in collaboration with Alexandra L. Joyner1 and Wojtek Auerbach1

The S. pombe rad9 gene plays important roles in pro-moting resistance to radiations and chemicals that damageDNA, and also regulates the associated checkpoints (1).Furthermore, the yeast and cognate human gene productshave pro-apoptotic function (2,3). A murine version of thegene, Mrad9, has been isolated (4) and used to make knock-out ES cells to test the function of the corresponding geneproduct in a mammalian cell environment. As indicated inFig. 1A, Southern blot analysis demonstrates that heterozy-gous and homozygous deleted Mrad9 cells were obtained.To verify this at the protein level, cell extracts were analyzedby western blotting for the presence of Mrad9 protein. Asindicated in Fig. 1B, the homozygous Mrad9 deleted cellsare completely devoid of Mrad9 protein and the heterozy-gous cells have about half as much, supporting the Southernresults and indicating that a homozygous Mrad9 knockoutcell is viable.

The WT as well as heterozygous and homozygousMrad9 deleted ES cells were assessed for gamma-ray sensi-tivity. As demonstrated in Fig. 2, the homozygous Mrad9deleted cells are dramatically more sensitive to gamma-raysthan the WT cell population, indicating that as predictedfrom previous yeast studies the gene product plays a veryimportant role in promoting radioresistance. In addition,Mrad9+ was introduced into the homozygous null cells anddemonstrated the predicted complementation (Fig. 2), thusfurther verifying that it is the Mrad9 null causing the radio-sensitivity. Furthermore, the heterozygous cells are also sig-nificantly more radiosensitive than the WT population.These studies establish a role for Mrad9 in modulating ra-dioresistance. Moreover, since the Mrad9+/- cells are alsomore radiosensitive than the isogenic WT population, this

1 Skirball Institute, New York University, N.Y., N.Y.

investigation suggests that individuals bearing heterozygousor homozygous mutant alleles of this gene may be geneti-cally predisposed to deleterious health effects associatedwith exposure to radiations or other agents that damageDNA. Investigations that directly address this issue are inprogress.

Fig. 1. Southern (A) and western (B) blot analyses indicating that homozygous Mrad9 knockout ES cells are viable. Symbols: +/+,Mrad9 WT; +/-, Mrad9 heterozygous deleted; -/-, Mrad9 homozygous deleted. Southern blot: Mrad9 WT and deleted (del) fragmentsare indicated. Western blot: 80 µg of total cell protein were loaded per well. Position of Mrad9 protein is indicated. Actin serves as theinternal loading control.

1086420.001

.01

.1

1

10

100

+/+ mrad9+/- mrad9-/- mrad9-/- mrad9 + rad9 cDNA

Dose, Gy

Per

cent

Sur

viva

l

Fig. 2. Gamma-ray sensitivity measured by colony formingassays. Points represent the average of 3 trials each; bars repre-sent standard deviations.

A B

+/+ mrad9+/- mrad9-/- mrad9-/- mrad9 + rad9 cDNA

Mrad9

Actin

Mrad9 del

Mrad9+


62

References

1. Lieberman HB, Hopkins KM, Laverty M and Chu HM,Molecular cloning and analysis of Schizosaccharomycespombe rad9, a gene involved in DNA repair and muta-genesis, Mol. Gen. Genet. 232:367-376, 1992.

2. Komatsu K, Hopkins KM, Lieberman HB and Wang HG(first two authors contributed equally, last two are corre-sponding authors), Schizosaccharomyces pombe Rad9contains a BH3-like region and interacts with the anti-apoptotic protein Bcl-2, FEBS Letters 481:122-126,

2000.3. Komatsu K, Miyashita T, Hang H, Hopkins KM, Zheng

W, Cuddeback S, Yamada M, Lieberman HB and WangHG, Human homologue of S. pombe Rad9 interacts withBCL-2/BCL-xL and promotes apoptosis, Nature CellBiol. 2:1-6, 2000.

4. Hang H, Rauth SJ, Hopkins KM, Davey SK and Lieber-man HB, Molecular cloning and tissue-specific expres-sion of Mrad9, a murine orthologue of the Schizosaccha-romyces pombe rad9+ checkpoint control gene, J. Cell.Physiol. 177:241-247, 1998. n

Identification and Characterization of a Paralogue ofHuman Cell Cycle Checkpoint Gene HUS1

Haiying Hang, Yuzhu Zhang,1 Roland L. Dunbrack, Jr. ,2 and Howard B. Lieberman

A paralogue of the human cell cycle checkpoint geneHUS1 has been identified and designated HUS1b. It encodesa 278 a.a-long protein, 48% identical and 69% similar toHUS1 (Fig. 1). Mouse and rat orthologues of HUS1b have

1 Dept. of Biological, Chemical and Physical Sciences, Illinois

Institute of Technology, Il.2 Institute for Cancer Research, Fox Chase Cancer Center, Phila-

delphia, Pa.

also been detected by a BLAST search. HUS1b is expressedvariably in many human tissues, and the tissue-specific lev-els observed parallel those for HUS1 (Fig. 2). A HUS1-RAD1-RAD9 protein complex is thought to form a

HUS1b 1 MKFRAKITGKGCLELFIHVSGTVARLAKVCVLRVRPDSLCF...GPAGSG 47 HUS1 1 MKFRAKIVDGACLNHFTRISNMIAKLAKTCTLRISPDKLNFILCDKLANG 50

HUS1b 48 GLHEARLWCEVRQ.GAFQQFRMEGVSEDLDEIHLELTAEHLSRAARSAAG 96 HUS1 51 GV...SMWCELEQENFFNEFQMEGVSAENNEIYLELTSENLSRALKTAQN 97

HUS1b 97 ASSLKLQLTHKRRPSLTVAVELVSSLGRARSVVQDLPVRVLPRRVWRDCL 146 HUS1 98 ARALKIKLTNKHFPCLTVSVELLSMSSSSRIVTHDIPIKVIPRKLWKDLQ 147

* ** *** ** **HUS1b 147 PPSLRASDASIRLPRWRTLRSIVERMANVGSHVLVEANLSGRMTLSIETE 196 HUS1 148 EPVVPDPDVSIYLPVLKTMKSVVEKMKNISNHLVIEANLDGELNLKIETE 197 RAD1*********** *HUS1b 197 VVSIQSYFKNLGNPPQSAVGVPENRDLESMVQVRVDNRKLLQFLEGQQIH 246 HUS1 198 LVCVTTHFKDLGNPPLASESTHEDRNVEHMAEVHIDIRKLLQFLAGQQVN 247 RAD1

HUS1b 247 PTTALCNIWDNTLLQLVLVQEYVSLQYFIPAL~ 278 HUS1 248 PTKALCNIVNNKMVHFDLLHEDVSLQYFIPALS 280

Fig. 1. Alignment of human HUS1b and HUS1 proteins. HUS1b and HUS1 share homology across their entire amino acid se-quences. Underlined residues may be involved in binding RAD1. Residues potentially critical for the interaction have asterisks on top.These regions were chosen based on sequence alignment with PCNA and the order of arrangement of HUS1, RAD9, and RAD1 in amodel of the trimer complex. Darkly shaded areas highlight identical a.a. whereas lightly shaded regions denote a.a. with similar physio-chemical properties.

MOLECULAR STUDIES

63

PCNA-like structure, important for cell cycle checkpointfunction. However, HUS1b directly interacts with RAD1,but not RAD9 or HUS1, while HUS1 can bind RAD1,RAD9 and another molecule of HUS1, suggesting thatHUS1b cannot simply substitute for HUS1 in the complex.

HUS1b is less conserved evolutionarily than HUS1. Fur-thermore, overexpression of HUS1b but not HUS1 in humancells induces clonogenic cell death (Fig. 3). We suggest thatHUS1b and HUS1 play distinct but related roles in regulat-ing cell cycle checkpoints and genomic integrity. n

hSIR2SIRT1 Functions as an NAD-Dependentp53 Deacetylase

Homayoun Vaziri,1 Scott K. Dessain,1 Elinor Ng Eaton,1 Shin-Ichiro Imai,1 Roy A. Frye,2 Tej K. Pandita,Leonard Guarente1 and Robert A. Weinberg1

The Silent Information Regulator (SIR2) family of genesrepresents a highly conserved group of genes present in thegenomes ranging from archaebacteria to a variety of eukary-otes. The encoded SIR proteins are involved in diverse proc- 1 Whitehead Institute for Biomedical Research, Cambridge, Ma.2 V.A. Medical Center, Pittsburgh, Pa.

esses from regulation of gene silencing to DNA repair. Theproteins encoded by members of the SIR2 gene family showhigh sequence conservation in a 250 amino acid core do-main. The best characterized gene in this family is S. cere-visiae SIR2, which is involved in silencing HM loci thatcontain information specifying yeast mating type, telomere

Fig. 2. Comparison between HUS1b and HUS1 expressionin human tissues. A poly(A) RNA blot was consecutively hy-bridized with 23P-labeled HUS1b RNA (Top), HUS1 RNA (Mid-dle) and β-actin cDNA (Bottom). After each hybridization, theblot was allowed to expose X-film before treatment to strip offthe probe. HUS1b and HUS1 expression patterns vary in parallelin the human tissues examined.

Fig. 3. Clonogenic cell death caused by overexpressionof HUS1b. pcDNA3.1/Hygro(+) plus pFLAG-CMV-2-HUS1b,pFLAG-CMV-2-HUS1 and pFLAG-CMV-2-GFP, respec-tively, were co-transfected into 293T cells followed by incuba-tion in medium containing hygromycin B for 10 days. Colonieswere counted and survival levels were calculated relative to thenumber of colonies formed from pFLAG-CMV-2-HUS1 trans-fected cells as the standard. The results are an average of threeindependent experiments. Error bars indicate standard devia-tion.

HUS1 HUS1b GFP

0

20

40

60

80

100

120

140

0.46%

Surv

ival

Lev

el (%

)


64

position effects, and cell aging (1).In mammalian cells, one of the primary mediators of the

DNA damage response is the p53 protein. Following DNAdamage, the p53 protein is protected from rapid degradationand acquires transcription-activating functions, thesechanges being achieved largely through post-translationalmodifications. There are at least 13 different residues both atthe N and C terminal portions of p53 protein that are phos-phorylated by various kinases. Transcriptional activation ofp53 protein in turn upregulates promoters of a number ofgenes including p21WAF1 that monitor cell cycle progression.

Analogs of NAD that inhibit endogenous ADP-ribosylases are also able to reduce induction of p21WAF1 inresponse to DNA damage and to overcome p53-dependentsenescence (2). In addition, it was demonstrated that p53protein can bind to the NAD-dependent poly-ADP-ribosepolymerase. We reasoned that other NAD-dependent en-zymes, notably Sir2, which is known to be involved inlifespan regulation in lower eukaryotes, might also be able tointeract with p53. For these reasons, we identified the humanhomolog of Sir2 and investigated its possible function inregulating the p53-dependent DNA damage response path-ways in mammalian cells.

We have observed that hSir2 directly binds the humanp53 protein in vitro and in vivo and specifically deacetylatesthe K382 residue of p53. One of the observed functionalconsequences of this deacetylation is an attenuation of thep53 protein’s activity as a transcription factor operating atthe p21WAF1 promoter. In another cellular context, in whichthe DNA damage response leads to apoptosis, hSir2 expres-sion attenuates the p53-dependent apoptotic response.

We propose that in mammalian cells, signals indicatingthe successful completion of DNA repair are relayed viahSir2 to acetylated proteins like p53 that have been charged

with the task of imposing a growth arrest following DNAdamage. In some way, these signals enable hSir2 to reversepart or all of the damage-induced activation of p53 as a tran-scription factor by deacetylating the K382 residue of p53.By doing so, hSir2 reduces the likelihood of subsequentapoptosis and, at the same time, makes it possible for cells tore-enter the active cell cycle, enabling them to return to thephysiological state that they enjoyed prior to sustainingdamage to their genomes.

Inactivation of the p53 signaling pathway is involved inthe pathogenesis of most if not all human tumors. In abouthalf of these tumors, mutation of the p53 gene suffices toderail function. In some of the remaining tumors, loss ofp14ARF, which acts to down-regulate p53 protein levels, hasbeen implicated (3). The present observations suggest that athird mode by which an incipient cancer cell can rid itself ofsome of p53’s functions may involve over-expression ofhSir2 which, like the other two genetic strategies, results inthe inactivation of both the cytostatic and pro-apoptoticfunctions of p53.

References

1. Guarente L, Diverse and dynamic functions of the Sirsilencing complex, Nat. Genet. 23:281-5, 1999.

2. Vaziri H, West MD, Allsopp RC, Davison TS, Wu YS,Arrowsmith CH, Poirier GG and Benchimol S, ATM-dependent telomere loss in aging human diploid fibro-blasts and DNA damage lead to the post-translationalactivation of p53 protein involving poly(ADP-ribose)polymerase, Embo. J. 16:6018-33, 1997.

3. Lohrum MA and Vousden KH, Regulation and activa-tion of p53 and its family members, Cell Death Differ.6:1162-8, 1999. n

Influence of PTEN on Telomere Stabilityand Telomerase Activity

Girdhar G. Sharma, Janusz Puc,1 Sonu Dhar, Ramon Parson,2 and Tej K. Pandita

PTEN is a tumor suppressor protein that functions, inlarge part by dephosphorylating the lipid second messengerphosphatidylinositol-3,4,5-trisphosphate (PIP3) and by do-ing so antagonizing the action of phosphoinositide 3-kinase.PTEN is also known as MMAC1/TEP1. Mutations of PTENare frequently found in a variety of human cancers, includ-ing brain, breast, endometrial, prostate and kidney tumors.Germline mutations of PTEN cause Cowden syndrome, a 1 Institute of Cancer Genetics.2 Department of Pathology.

multiple hamartoma condition associated with high inci-dence of breast, brain, and thyroid neoplasia. Despite ho-mology to protein phosphatases, PTEN functions bydephosphorylation of the lipid second messenger,PI(3,4,5)P3, a product of phosphatidylinositol 3' kinase(PI3'K) activity, and negatively regulating survival signalingmediated by protein kinase B/Akt (PKB/Akt). Loss of PTENleads to elevated levels of PI3,4,5P3 and consequent Aktactivation. PTEN structural domains include an amino-terminal phosphatase domain, a lipid-binding C2 domainand a 50 amino acid carboxy terminal tail that contains a

MOLECULAR STUDIES

65

PDZ binding sequence. It has been shown that phosphoryla-tion of the PTEN tail negatively regulates its function as anantagonist of PI3K signaling.

Mice lacking PTEN indicate that PTEN negativelyregulates neural stem cell proliferation. They have enlarged,histoarchitecturally abnormal brains, which are the conse-quences of enhanced cell division with decreased cell death,and enlarged cell size. Neurosphere cultures show a greaterproliferation capacity for PTEN-/- CNS stem/progenitor cells,which is due, at least in part, to a shortened cell cycle. Wewere interested in examining whether inactivation of PTENinfluences telomere stability and telomerase activity. Todetermine if PTEN was necessary for the telomere metabo-lism, mouse embryonic fibroblasts were generated from awild-type embryo, a heterozygous PTEN mutant embryo(MEF PTEN+/-), and a homozygous PTEN mutant embryo(two independent MEF PTEN-/- cell lines).

Genome stability is in part maintained by the telomeres,as these chromosome terminal structures protect chromo-somes from fusions or degradation. Human telomeres con-tain long stretches of a tandemly arranged hexameric se-quence, TTAGGG, bound by specific proteins. Shorteningor loss of telomeric repeats is correlated with chromosomeend-to-end associations that could lead to genomic instabil-ity and gene amplification. To determine the influence ofinactivation of PTEN gene on the frequency of chromosomeend-to-end associations two hundred metaphases were ex-amined. PTEN-/- cells had 1.9 chromosome end-to-end asso-ciations per metaphase whereas their parental PTEN cellshad 0.12 chromosome end-to-end associations per meta-

phase. The differences in the frequency of chromosome end-to-end associations between PTEN-/- and PTEN+/+ cells arestatistically significant. Since chromosome end-to-end asso-ciations may lead to anaphase bridge formation, cells with-out colcemid treatment were analyzed for anaphase bridges.Three hundred cells at anaphase were examined for bridges.PTEN-/- cells displayed an at least 8 fold higher frequency ofanaphase bridges as compared to PTEN+/+ cells.

These data suggest that due to occasional losses of telo-mere function, chromosome end associations are formed andthese associations are less likely to be resolved in PTEN-/-

cells. To determine whether end fusions are consequences ofthe loss of telomeres we performed FISH for telomeric DNArepeats in metaphase cells by using a telomere specific Cy3labeled (CCCTAA)3 peptide nucleic acid probe. A signifi-cantly higher proportion of chromatid ends in PTEN-/- cells(about 8% of telomeres per metaphase) have less telomerespecific fluorescent signals as compared to the PTEN+/+

cells. These observations suggest that the chromosome end-to-end fusions observed in PTEN-/- cells correlated withlosses of telomeric repeats. However, telomere signals wereseen in about 12% of fusion sites, indicating total loss oftelomeres is not required for telomere fusions.

A difference in the frequency of telomere fusions mightalso be due to alterations in telomerase activity. Using theTRAP-ELISA assay method, a significant difference in te-lomerase activity between PTEN-/- and 1 PTEN+/+ cells wasfound, indicating that the overall activity of telomerase isalso affected by the lack of the PTEN protein. n

hTERT Association with Telomeres Correlates withReduction in Spontaneous Chromosome Damage

and Enhancement of DNA RepairGirdhar G. Sharma, Arun Gupta, Sonu Dhar and Tej K. Pandita

Biochemical and genetic studies have established an as-sociation between telomere maintenance and extended life-span mediated through the expression of the telomerasecatalytic subunit (hTERT). Ectopic expression of TERT pre-vents replicative senescence in several cell types includingfibroblasts and epithelial cells. It may also exert anti-apoptotic action at an early stage of the cell death processprior to mitochondrial dysfunction and caspase activation. Ithas been proposed that telomere shortening during humanreplicative aging generates antiproliferative signals whichmediate p53-dependent G1 arrest as is observed in senescentcells. It has been found that telomerase expression sup-presses senescence-associated genes in Werner syndrome

cells but it is not known if such suppression correlates withgenomic integrity.

We compared profiles of gene expression in normal hu-man fibroblasts (HFF) with and without ectopic hTERT ex-pression. We used stable human fibroblasts (HFF+hTERT)that were generated by infection with hTERT expressingretrovirus. The transcriptional profile of such cells wereanalyzed by monitoring 5,776 expressed sequence tags(ESTs) on oligonucleotide arrays. Several possible pairingsfor hybridization for each of the cultures (HFF andhTERT+HFF) at population doublings (PDs) of 20, 30, 40,50 and 60 were performed. Sixty-seven (about 1.1%) ESTsreproducibly exhibited significant changes in expression.


66

The threefold differences in gene expression in cells withoutectopic hTERT expression compared to cells with ectopicexpression appeared after cells (HFF-hTERT) pass 40 PDs.When we used the early passage HFF-hTERT cells, we didnot find any significant differences in the gene expression.We suggest that hTERT protein through its interaction withtelomeres, generates a signal to regulate the expression of avariety of genes required for extended life span. There areseveral ESTs whose function are not yet known which alsoshowed altered expression. Many of these results were con-firmed by either Northern blot analysis or reverse transcrip-tion-polymerase chain reaction (RT-PCR). mRNA from cellswith and without hTERT expression were used for confir-mation of microarray results by northern or RT-PCR.

To determine whether alteration in gene expression cor-relates with the interaction of the TERT protein with thetelomeres, time course interaction of TERT protein withTTAGGG sequences was investigated. By using the chro-matin immunoprecipitation (ChIP) procedure, we found thatcells with hTERT expression have TERT protein bound tothe telomeres.

At senescence, cells are viable and metabolically activebut no longer divide. These cells are in G1 phase of the cellcycle. This arrest in cell division is associated with at leastsome dysfunctional telomeres and accumulation of geneticdamage. We then wanted to determine if ectopic hTERTexpression not only stabilized telomeres but also reducedspontaneously occurring chromosomal damage. By using thecombination of premature chromosome condensation (PCC)technique with in situ hybridization, we examined directly ininterphase cells whether hTERT expression stabilizes telo-meres and reduces spontaneously induced chromosomedamage. Cells with ectopic hTERT expression had only 0.04telomeric associations per G1 cell, which is 12-fold less thanthe cells without ectopic hTERT expression (0.5 telomericassociations per G1 cell). The difference in telomeric asso-ciations in interphase cells with and without hTERT werestatistically significant as demonstrated by the student t-test(P<0.001). Spontaneous chromosome breaks in G1-phasecells without ectopic hTERT expression occur at 0.2 breaksper cell, which is 20-fold higher than the cells with ectopichTERT expression with 0.04 break per cell (P<0.001).

The higher mRNA levels of some of the DNA repairgenes suggest that genome stabilization in cells with ectopichTERT expression could be due to enhanced repair capacityof the cells. To determine whether there is any difference inDNA repair efficiency in cells with and without ectopichTERT expression, we examined DNA repair after treatmentwith ionizing radiation and cisplatin. Ionizing radiation pro-duces DNA strand breaks whereas cisplatin produces DNAadducts. Exponentially growing isogenic cells with andwithout hTERT expression were treated with 15 Gy ofgamma rays and were allowed to repair at 37oC for varioustime periods prior to analysis for the residual levels of DNAstrand breaks. Cells in logarithmic growth without ectopichTERT expression exhibited a slow rate of DNA repairalong with higher residual DNA damage, whereas cells withectopic hTERT expression had a fast rate of DNA repair andless residual DNA damage.

The most frequent damage that occurs in vivo is DNA

adduct formation and its deficiency in repair is linked withthe age of an individual. DNA interstrand crosslinks pro-duced by the incubation of human peripheral blood mononu-clear cells with cisplatin (CDDP) are repaired faster inyoung donors (age 25 years) as compared to individuals ofan elderly group (age above 75 years). Those from the eld-erly group have significantly higher mean levels ofcrosslinking after CDDP treatment and this could be indica-tive of impaired DNA repair capacity in the cells from theelderly group. Because telomere shortening also correlateswith advanced age and hTERT expression stabilizes telo-meres, we investigated whether hTERT expression has aninfluence on DNA adduct repair. The extent of DNA inter-strand crosslinking was measured up to 24 hr after treatmentwith CDDP. The rate of formation of DNA cross-linksreached a peak levels 10-15 hr after treatment in cells withand without ectopic hTERT expression. However a sharpcontrast in the repair process was observed (Fig. 4). DNAdamage was rapidly repaired in cells with ectopic hTERTexpression after 15 hr following exposure to CDDP, whileless change in interstrand cross-linking was observed in cellswithout hTERT expression. These results suggest thathTERT expression has a pronounced effect on the repair ofDNA adducts.

To test whether hTERT protein is directly involved in theprocess of DNA end joining, we used an in vitro assay thatinvolves the rejoining of DNA ends of plasmid in cell ex-tracts with and without telomerase activity. hTERT proteinin cell extracts was depleted by immunoprecipitation by us-ing different concentrations of hTERT antibody. We con-firmed complete immunodepletion of hTERT by westernanalysis and by conducting a telomerase activity functionalassay. Further, telomerase activity was measured by TRAP-ELISA procedure and no activity was detected in the de-pleted extracts (data not shown). The hTERT immunode-pleted extracts were used to assay for the DNA end joiningactivity. No difference in DNA end joining was found inextracts with and without hTERT protein. We further ana-lyzed whether addition of hTERT antibody could influencethe DNA end joining kinetics. These results suggest thathTERT protein itself does not directly interact with the DNArepair machinery.

Inactivation of mTERT leads to the telomere instabilitybut only after several generations. To determine if such in-activation influences chromosomal recombination, we ex-amined the formation of bouquets (also known as telomereclustering), synaptonemal complexes and sex vesicles inmeiocytes of mTERT null and control mice. It has been sug-gested that bouquet formation might support the alignmentof homologs prior to their synaptic pairing. We found nodifference in the bouquet formation between the spermato-cytes of mTERT null and control mice. The synaptonemalcomplex formation as well as sex vesicle formation wasnearly identical in mTERT null and control mice. These re-sults suggest that mTERT does not have direct influence onthe in vivo meiotic recombination process.

In summary, using cDNA microarray analyses we identi-fied genes whose expression was altered in HFF+hTERTcells as opposed to their parental HFF-hTERT cells. Chro-matin immunoprecipitation with hTERT antibody demon-

MOLECULAR STUDIES

67

strated that hTERT interacts with telomeric DNA before asignificant alteration in gene expression is observed. HowhTERT protein alters the gene expression is not clear at pre-sent. Because hTERT protein interacts with the telomeres, itis possible that such interaction may generate the signals to

down regulate senescence-associated gene expression and toup regulate DNA repair-associated genes. Our findings sug-gest that up regulation of DNA repair genes could be thepossible reason for the telomere stability and decrease inspontaneous chromosome damage. n

14-3-3σ Influences Telomere Stabilityand ATM Kinase Activity

Arun Gupta, Dea-Sick Lim,1 Sonu Dhar, Michael B. Kastan1 and Tej K. Pandita

Recent data indicate that the expression of 14-3-3σ islost in 94% of breast tumors (1). At the functional level, the14-3-3σ protein has been implicated in the G2 checkpoint.For instance, its association with different kinases in thecytosol and on the nuclear membrane may contribute to ki-nase activation during intracellular signaling, and the proteinappears to sequester the mitotic initiation complex, cdc2-cyclinB1, in the cytoplasm after DNA-damage (2). The latterprevents cdc2-cyclin B1 from entering the nucleus, wherethe protein complex could normally initiate mitosis. Thus,14-3-3σ has been implicated in maintaining a post-DNA-damage G2-arrest, thereby allowing for DNA repair. Suchcell cycle checkpoints are considered to be the guardians ofgenome integrity, with their abrogation contributing to re-duce genomic stability. Genome stability is also maintainedby the telomeres. Shortening or loss of telomeric repeats iscorrelated with chromosome end-to-end associations thatcould lead to genomic instability and gene amplification.

Cells in which both alleles of 14-3-3σ are inactivatedgrow slowly and exhibit decreased cell survival after gammaray treatment. It is thus possible that damaged DNA is notrepaired appropriately in these cells. To determine the influ-ence of inactivation of 14-3-3σ gene on the frequency ofchromosome end-to-end associations two hundred meta-phases were examined. 14-3-3σ-/- cells had 1.9 chromosomeend-to-end associations per metaphase whereas their paren-tal 14-3-3σ+/+ cells had 0.12 chromosome end-to-end asso-ciations per metaphase. The differences in the frequency ofchromosome end-to-end associations between 14-3-3σ-/- and14-3-3σ+/+ cells are statistically significant. Since chromo-some end-to-end associations may lead to anaphase bridgeformation, cells without colcemid treatment were analyzedfor anaphase bridges. Three hundred cells at anaphase wereexamined for bridges. 14-3-3σ-/- cells displayed an at least a8 fold higher frequency of anaphase bridges as compared to14-3-3σ+/+ cells. These results indicate that inactivation ofthe 14-3-3σ gene enhances the frequency of observablechromosome end-to-end associations. This pheonotype is 1 St. Jude Children's Research Hospital, Memphis, Tn.

simlar to that found in cells derived from ataxia telangiecta-sia individuals. We next compared ATM kinase activity in14-3-3σ-/- and parental 14-3-3σ+/+ cells. We found that thelevels of ATM protein were similar in 14-3-3σ-/- and paren-tal 14-3-3σ+/+ cells. When ATM kinase activity was exam-ined, we found that 14-3-3σ-/- cells have relatively lowerATM kinase activity as compared to parental 14-3-3σ+/+

cells. Interestingly, we also found that 14-3-3σ-/- cells aredefective in the enhancement of ATM kinase in response toionizing radiation treatment. These observations suggest that14-3-3σ protein has some functions similar to that of theATM protein. The fact that the p53-mediated G1 checkpointis operative in these cells is underscored by our findings thatγ-irradiation induces comparable frequencies of observablechromosome aberrations when 14-3-3σ-/-, 14-3-3σ+/- cells or14-3-3σ+/+ cells were irradiated while in G1. When the G2-induced aberrations were monitored for the parental cellsover time after irradiation, there is a decline in the frequencyof aberrations, indicating successful repair of at least someof the induced damage. For 14-3-3σ-/- cells on the otherhand, the frequencies of aberrations increased in the sametime span, indicating very inefficient repair. Consistent withthese data, survival studies demonstrate that 14-3-3σ-/- cellsare more sensitive to γ-irradiation.

References

1. Ferguson AT, Evron E, Umbricht C, Pandita TK, Her-meking H, Marks J, Futreal A, Stampfer MR and Suku-mar S, High frequency of hypermethylation at the 14-3-3σ locus leads to gene silencing in breast cancer, PNAS(USA) 97:6049-6054, 2000.

2. Chan TA, Hermakin H, Lengauer C, Kinzler KW andVogelstein B, 14-3-3σ is required to prevent mitotic ca-tastrophe after DNA damage, Nature 401:616-620, 1999. n


68

ECK Protein Kinase as a Transcriptional Target of p53in Signaling Apoptosis and Tumor Suppression

Cynthia Y. Liu and Yuxin Yin

The tumor suppressor p53 plays a fundamental role incell cycle checkpoint control, cell death induction, and ge-netic stability. As a sequence-specific DNA binding protein,p53 activates transcription through binding to specific DNAconsensus sequences. p53 is a negative regulator of cell cy-cle progression and functions as a checkpoint protein incontrolling the transition from G1 to S phase of the cell cy-cle. p53 also functions as a cell death mediator and is re-quired for cellular apoptotic response to oxidative stress (1).In recent years, much progress has been made towards un-derstanding the mechanism by which p53 mediates pro-grammed cell death. It is known that p53 transactivationplays an essential role in induction of apoptosis, suggestingthat p53 downstream genes may be involved in this process.p53 acts as both a transcriptional activator and repressor.Some transcriptional targets of the p53 protein are involvedin regulation of cell growth, DNA repair or cell death proc-esses. However, there is still uncertainty as to whether andhow any of these known p53-regulated genes act as a tumorsuppressor and function in p53-mediated tumor suppression.Thus, it is likely that there are additional unidentified p53-regulated genes that play roles in p53-mediated apoptosisand tumor suppression. We have recently identified ECK(epithelial cell kinase) as a new target of p53 function insignaling apoptosis and tumor suppression.

ECK is a transmembrane tyrosine kinase structurally re-lated to the EPH subfamily of receptor protein-tyrosine ki-nases (RPTK). ECK protein has an external domain with anN-terminal signal peptide, a transmembrane domain, and acytoplasmic domain, which includes a canonical protein-tyrosine kinase catalytic site in the intracellular region of theprotein. ECK is widely expressed in a variety of human tis-sues. The human ECK maps to 1p36.1, which is a regionthat is frequently deleted in neuroblastoma, melanoma andother types of human cancers. Studies of ECK in cancerhave concentrated primarily on the levels of ECK in clinicalsamples and cell models. The evidence for the role of ECKin tumorigenesis is limited and controversial. It was recentlyreported that activation of ECK/EphA2 receptor tyrosinekinase blocks the Ras/MAPK cascade in a variety of celltypes and inhibits cell proliferation (2). However, there havebeen reports that there are high levels of ECK in some tumorspecimens and cancer cell lines (3).

To explore the process of p53-mediated apoptosis, weused DNA microarrays to identify new p53 target genes thatare involved in the apoptotic process in the EB-1 cell line, ap53 inducible system (4). One of the identified clones con-tains a sequence identical to the gene for human ECK. Toconfirm these observations, we used the human ECK codingsequence as a probe for Northern analysis of ECK expres-

sion in the p53-inducible systems. We observed that there isa low level of ECK transcript in p53-null H1299 cells. Theexpression of ECK is greatly increased in p53ER cells in thepresence of 17β-estradiol that activates the transactivity ofp53, suggesting that ECK is induced by p53 at the transcrip-tional level. To extend our observation of positive regulationof ECK by p53, we used the EB-1 cell line from EB cellsderived from a human colon cancer. EB-1 cells contain astable-transfected wild-type p53 transgene under the controlof the metallothionein promoter and express wild-type p53upon administration of zinc chloride (ZnCl2). These cellsundergo apoptosis following serum starvation in the pres-ence of p53. We observed that ECK mRNA is increased inEB-1 cells in the presence of ZnCl2 only. Significantly, ECKexpression is increased even more in the EB-1 cells treatedwith both ZnCl2 and serum starvation. These results suggestthat ECK is upregulated by p53 and induction of ECK isenhanced when cells undergo apoptosis.

To determine whether ECK expression is responsive togenotoxic damage, we treated embryo fibroblasts derivedfrom mice with differing p53 status with gamma-irradiationor H2O2. ECK protein is not induced by γ-radiation, whichcauses cell cycle arrest at G1. However, ECK is markedlyincreased by H2O2, which induces cell death by apoptosis(2). The response of ECK to oxidative stress is dependent onp53 because ECK is not significantly induced following oxi-dative stress in p53-/-MEFs. Interestingly, ECK protein isincreased by γ-irradiation in p53-/-MEF cells, suggesting thatECK is repressed by p53 when normal cells are exposed toγ-irradiation.

In order to understand the molecular basis for regulationof ECK by p53, we cloned the regulatory region of the hu-man ECK gene and made luciferase reporter pGL3/ECK-luciusing the ECK promoter. To examine luciferase activity, theluciferase reporters were transfected into H1299 cells alongwith a pCMV vector as a control, or with a pCMV/wild-typep53 expression vector (pCMV/wtp53). We found that theactivity of the pGL3/ECK-luci is greatly induced by wild-type p53, but not by mutant p53. The fold induction of theECK promoter activity is similar to that of the p21 promoteractivity by wild-type p53. However, induction of the ECKpromoter activity by p53 is much higher than that of bothMDM2 and Bax luciferase promoters. These results suggestthat the human ECK promoter is highly responsive to p53function.

To determine the role of ECK in the apoptotic process,we constructed a human ECK expression vector by cloning a2.9 kb human ECK cDNA containing the full coding se-quence in frame into a mammalian expression vector drivenby a human cytomegalovirus (pcDNA3/hygro, Invitrogen).

MOLECULAR STUDIES

69

The resulting ECK expression vector, pcDNA3/hECK, wastransfected into MCF-7 cells, which were then selected withhygromycin B for isolating stable clones. We obtained twocontrol clones and three stable clones expressing high levelsof ECK protein, designated as MCF-7/pcDNA3, MCF-7/ECK-1, -7 and -12. Interestingly, during culture of thesecells, we observed that approximately 30% of MCF-7/ECK-1 and MCF-7/ECK-7 cells were susceptible to spontaneouscell death. The nature of the cell death is apoptosis, whichwas determined by TUNEL assay. These results indicate thatECK may function directly as a cell death mediator. Becauseoverexpression of ECK causes apoptosis and inhibits an-chorage-independent growth of MCF-7 cells, it is possiblethat ECK influences tumorigenesis in vivo. To determinewhether ECK inhibits tumor formation of breast cancer cellsin animal models, we tested MCF-7 cells with or withoutectopic ECK for their ability to form tumors in nude mice.We chose MCF7/pcDNA3 as a control and two MCF-7clones overexpressing ectopic ECK: MCF7/hECK-1 andMCF7/hECK-7. To conduct the tumorigenicity assay innude mice, the cells with or without ectopic ECK were in-oculated subcutaneusly into the back region of athymic nudemice. Each group includes 6 nude mice for statistical analy-sis. We found that 6 of 6 mice inoculated with MCF-7 cellsproduced tumors and 5 of 6 mice inoculated withMCF7/pcDNA3 produced tumors. In contrast, there was notumor formation in any mice inoculated with eitherMCF7/hECK-1 or MCF7/hECK-7 cells up to 8 weeks post-

injection. Our results demonstrate that p53 is a transcrip-tional activator of ECK and that ECK is a new member ofp53-downstream family. Since ECK causes apoptosis andsuppresses tumor suppression, we propose that ECK is a celldeath receptor in the p53 pathway.

References

1. Yin Y, Terauchi T, Solomon GG, Aizawa S, RangarajanPN, Yazaki Y, Kadowaki T and Barrett JC, Involvementof p85 in p53-dependent apoptotic response to oxidativestress, Nature 391:707-710, 1998.

2. Miao H, Wei BR, Peehl DM, Li Q, Alexandrou T,Schelling JR, Rhim JS, Sedor JR, Burnett E and Wang B,Activation of EphA receptor tyrosine kinase inhibits theRas/MAPK pathway, Nat. Cell Biol. 3:527-530, 2001.

3. Zelinski DP, Zantek ND, Stewart JC, Irizarry AR andKinch MS, EphA2 overexpression causes tumorigenesisof mammary epithelial cells, Cancer Res. 61:2301-2306,2000.

4. Zhao R, Gish K, Murphy M, Yin Y, Notterman D,Hoffman WH, Tom E, Mack DH and Levine AJ, Analy-sis of p53-regulated gene expression patterns using oli-gonucleotide arrays, Genes & Dev. 14:981-993, 2000.

5. Yin Y, Solomon G, Deng C and Barrett JC, Differentialregulation of p21 by p53 and Rb in cellular response tooxidative stress, Mol. Carcino. 24:15-24, 1999. n

Overexpression of Betaig-H3 Gene SuppressesTumorigenicity in Radiation-Induced Tumorigenic

Human Bronchial Epithelial CellsYong L. Zhao, Chang Q. Piao and Tom K. Hei

Tumor growth and metastasis is a multistep process in-volving in cell adhesion, extracellular matrix (ECM) degra-dation and cell migration. Betaig-h3 is a secreted proteininduced by transforming growth factor-β (TGF-β) in humanadenocarcinoma cells as well as other human cell types. Al-though transfection of Betaig-h3 gene into CHO (ChineseHamster Ovary) fibroblasts markedly reduces their ability toform tumors in nude mice, its regulation in human tumors isnot clear. Previous data showed that mutations or alteredexpression of this gene are involved in corneal dystrophyand osteogenesis in humans. In addition, Betaig-h3 protein isa component of ECM in lung, bladder and skin, which pro-motes adhesion and the spreading of dermal fibroblasts invitro and mediates cell adhesion by interacting with α3β1integrin in human corneal epithelial cells. These data suggestthat Betaig-h3 protein is involved in cellular adhesion and

imply an important role for this gene in the process of hu-man tumor progression.

Although in vitro transformation studies with humancells are highly desirable in studying the molecular eventsassociated with malignant conversion, such studies, thus far,have not been successful with primary human epithelialcells. Using papillomavirus-immortalized human bronchialepithelial (BEP2D) cells, we have previously shown that α-particles can malignantly transform these cells in a stepwisefashion before they become tumorigenic and metastatic innude mice (1). It should be stated that although these cellsare immortalized, they do not possess any other transformedphenotypes and only after carcinogen treatment and ex-tended subculturing do transformed/tumorigenic phenotypesemerge in a sequential fashion. The BEP2D cell model is,therefore, useful in studying the genetic events involved in


70

tumor progression. In the present study, we show that recov-ery of Betaig-h3 expression in radiation-induced tumor cellssignificantly suppresses their in vivo tumorigenicity. Thisfinding provides strong evidence that Betaig-h3 many func-tions as a tumor suppressor in humans.

Tumorigenic BEP2D cells were derived previously fromtreatment of exponentially growing BEP2D cells with a sin-gle 60 cGy dose of alpha-particles. Tumor larger than 1 cmin diameter were resected from nude mice and used to es-tablish independently-generated cell lines (TL1-TL5). TheBEP2D cells and tumor cell lines were maintained in serum-free LHC-8 medium supplemented with growth factors asdescribed previously. Primary human bronchial epithelial(NHBE) cells were grown in BEBM medium purchasedfrom Clonetics.

Beatig-h3 cDNA was cloned and Sequenced. The firststrand cDNA was synthesized from 0.2 µg poly(A)+ RNAisolated from NHBE cells using Superscript II reverse tran-scriptase and oligo(dT) primer (Gibco). Human Betaig-h3cDNA was then PCR-amplified and subcloned into the Hin-dIII and XbaI-digested pRc/CMV2 expression vector (Invi-trogen). The sequence analysis showed that the protein se-quence is 100% identical with the report in Gene Bank ex-cept several modified nucleotide sites such as 698 (C→ G),1667 (T→ C) and 1118 (C→ T). TL1 Tumor cells were thentransfected with either pRc/CMV2-Betaig-h3 or pRc/CMV2(2 µg/dish) for 24hr using lipofectamin (Gibco) according tothe manufacturer′s instruction. The cells were split at 1:10and cultured in the medium containing 500 µg/ml of theG418 (Gibco) for 21 days. Colonies were isolated usingcloning ring and maintained in the presence of 300 µg/ml ofG418. The anchorage-independence growth and tumori-genicity in nude mice were determined as follow: Anchor-age-independence assays were performed by plating the Be-taig-h3-transfected and control BEP2D cells in 0.35% agaron the layer of 0.7% agar. Colonies ≥ 10 cells in numberwere counted after 4 weeks. Tumorigenicity assay was per-formed as described previously (1). Either Betaig-h3- orempty vector-transfected tumor cells were injected subcuta-

neously into nude mice at the sites of flanks. Tumors werepalpated and measured with calipers and tumor volume cal-culated as [longest diameter × (shortest diameter)2] × 0.5.Control animals were inoculated with either control BEP2Dor with radiation-induced TL1tumor cells. For each cell line,two independent experiments were performed.

By using cDNA array techniques, we identified a seriesof genes that were differentially expressed in radiation-induced tumor cells relative to parental BEP2D cells (2).Among these genes, Betaig-h3 expression was markedlydecreased in tumor cells. The result was further confirmedby Northern blot using mRNAs obtained from different pas-sages of transformed cells and five tumor cell lines (Figure1). In early passage cells (one week after radiation), nochanges in Betaig-h3 expression were found compared withcontrol BEP2D cells. However, the expression of Betaig-h3was downregulated by 2.4 fold in late passage cells (justbefore inoculating into nude mice) and between 7.5-9 fold inall five tumor cell lines examined. These results indicate thatdecreased expression of Betaig-h3 gene might be related tothe acquisition of a malignant phenotype of BEP2D cells.

To examine the significance of Betaig-h3 downregula-tion in malignant conversion, we recovered the expression ofBetaig-h3 gene in a representative tumor cell line (TL1) withpRc/CMV2-Betaigh3 vector. Two G418-resistant colonies(TL1-clones 18 and TL1-clone 28) that expressed differentlevels of Betaig-h3 were chosen for further studies. From theNorthern and Western blot results (Figure 2), the parentalTL1 and TL1-pRc/CMV2 cells expressed similar levels ofBetaig-h3, which were lower than control BEP2D cells. Af-ter Betaig-h3 transfection, the expression of this gene inTL1-clone 18 was recovered to a level similar to that ofcontrol BEP2D cells, whereas TL1-clone 28 had a 4 foldhigher expression level. However, expression of Betaig-h3gene on mRNA and protein levels was similar between pri-mary bronchial epithelial (NHBE) cells and control BEP2Dcells.

Fig. 1. Nothern blot analysis of Betaig-h3 gene in controlBEP2D cells, early passage cells (1 week post radiation), latepassage cells (just before inoculation into nude mice) and fivetumor cell lines (TL1-TL5). The blots were hybridized to 32P-labeled Betaig-h3 cDNA probes. After stripping, the membraneswere rehybridized to human β-actin probe.

Fig. 2. mRNA and protein levels of Betaig-h3 gene deter-mined by Northern blot and immunobloting (IB) in normalNHBE, control BEP2D, TL1 and Betaig-h3-transfected tumorcells.

MOLECULAR STUDIES

71

We also checked colony-forming efficiency in soft agar(Table I). The result showed that there was no significantdifference between parental TL1 and TL1-pRc/CMV2 cells(2.39% and 2.28%, respectively with P > 0.05). However,TL1-clone 18 and TL1-clone 28 cells resulted in a signifi-cant lower ability of anchorage independent growth withcolony-forming efficiency in agar of 0.29% and 0.34%, re-spectively.

To determine whether the recovery of Betaig-h3 geneexpression suppresses tumor formation in vivo, we injectedsubcutaneously 5 × 106 of each of the following cell typesinto nude mice: control BEP2D cells, TL1 tumor cells, TL1-pRc/CMV2 and Betaig-h3-transfected cells (clone 18 and28). The tumor volumes were measured weekly during theexperiments. As shown in table 1, no tumors (0/8 mice) werefound in parental BEP2D cells after monitoring for morethan 20 weeks. However, in 8/8 mice that were injected witheither TL1 or TL1-pRc/CMV2 tumor cells progressively

growing tumors developed with average volumes of 1021.8± 330.7 mm3 and 970.2 ± 295.6 mm3, respectively. In con-trast, 4/8 mice with TL1-clone 18 and 5/8 mice with TL1-clone 28 cells formed only small nodules (P < 0.01). Tumorgrowth was then significantly suppressed in tumor cells afterBetaig-h3 transfection .

It has been suggested that Betaig-h3 protein affects cell-ECM interaction through regulation of integrin receptor.Using cDNA arrays, we found that α5β1 integrin receptorwas overexpressed in radiation-induced tumor cells (data notshown). To determine whether the expression of integrinreceptor α5β1 is regulated by Betaig-h3 gene, we checkedits mRNA and protein levels in Betaig-h3 transfected tumorcells. As shown in Figure 3, expression of α5 and β1 sub-units was 5 and 3 fold higher, respectively, in parental TL1and TL1-pRc/CMV2 cells than in control BEP2D cells.However, after transfecting Betaig-h3 gene into TL1 cells,expression of α5β1 integrin (clone 18 and 28 cells) de-creased to level of control BEP2D cells. This data were fur-ther confirmed by immunoprecipitation using monoclonalantibody for α5 and β1.

Altered cell-matrix interaction is an essential prerequisitestep in the invasive and metastic cascade. Our finding ofsimilar levels of Betaig-h3 expression detected in normalNHBE and immortalized BEP2D cells suggests that loss ofBetaig-h3 expression occurs during the late stage of tumorprogression. In this study, we provide evidence that recoveryof Betaig-h3 expression in TL1 tumor cells significantlyinhibits colony-forming efficiency in soft agar, and tumorgrowth in nude mice relative to parental tumor cells. This isthe first evidence that Betaig-h3 gene functions as a tumorsuppressor in human epithelium-derived tumor model.

References

1. Hei TK, Piao CQ, Willey JC, Thomas S and Hall EJ,Malignant transformation of human bronchial epithelialcells by radon-simulated alpha-particles, Carciongenesis15(3):431-437, 1994.

2. Zhao YL, Piao CQ, Hall EJ and Hei TK, Mechanisms ofradiation-Induced neoplastic transformation of humanbronchial epithelial cells, Radiat. Res. 155:230-234,2001. n

Fig. 3. mRNA and protein levels of α5β1 determined byNorthern blot and immuno-precipitation (IP) in TL1 and Betaig-h3-transfected tumor cells.

Table I.

Suppression of anchorage-independent growth and tumorigenicity by Betaig-h3 gene.

Cell type CFE in soft agar (%)‡ Tumors/total mice Tumor volumeat 4 weeks (mm3)

BEP2D cells 0.02 0/8 TL1 tumor cells 2.39 ± 0.29 8/8 1021.8 ± 330.7TL1-pRc/CMV2 2.28 ± 0.37 8/8 970.6 ± 295.6

TL1-Betaig-h3 clone 18 0.29 ± 0.05† 4/8 9/16 86.7 ± 32.3†

TL1-Betaig-h3 clone 28 0.34 ± 0.07† 5/8The colonies in soft agar were counted and tumor volumes were measured as in Materials and Methods. † P < 0.01, compared withparental tumor cells. ‡ CFE, Colony Forming Efficiency.


72

Mutation(s) at Exon 3 of β-Catenin Preventingβ-Catenin-GSK-3β Interaction: A Possible Role in

Radiation-Induced Breast Cancer ProgressionDebasish Roy, Gloria M. Calaf and Tom K. Hei

β-Catenin is a multifunctional protein involved in cell-cell adhesion, by strengthening the linkage of cadherin andα-catenin to the actin cytoskeleton. It is also involved inWingless/Wnt signaling pathway whose inappropriatereactivation leads to tumorigenesis. Intracellular concentra-tions of β-Catenin are mainly regulated by degradation,which is probably initiated by interaction with adenomatouspolyposis coli (APC) tumor suppressor protein and by phos-phorylation at serine and/or threonine residues of codons 33,37, 41 and 45 of exon 3 through the functional interactionwith glycogen synthase kinase-3β (GSK-3β) (1-2). Missensemutations involving serine/threonine residues in exon 3 of β-Catenin had been reported in a different series of tumors (3).

Mutations of APC or β-Catenin and activation of theWnt signal inhibits GSK-3β activity and induces β-Cateninstabilization within the cell. Accumulated β-Catenin maytranslocate into the nucleus, where it could serve as a tran-scriptional factor through binding with high-mobility groupbox factors of the Tcf-Lef family and stimulate tumor for-mation by increasing c-myc, cyclin D1 etc., activity (Figure1). These observations have implied a possible associationbetween disruption of the cadherin-catenin system whichleads to tumorigenesis (4-5).

The neoplastic transformation of HBEC (human breast

epithelial cells) in vitro represents a successful model forobtaining step-by-step knowledge on the molecular and cel-lular alterations that may contribute to the tumorigenicmechanisms. Therefore, to assess the effect of ionizing ra-diation, particularly of high-LET (linear energy transfer)radiation, on the progression of human breast carcinogene-sis, we have developed a model system of irradiated, trans-formed and tumorigenic MCF-10F cell lines with gradeddoses of high-LET radiation (6).

Results from mutation spectrum of β-Catenin obtainedby direct sequencing of the amplified β-Catenin gene prod-uct of exon 3 in irradiated and tumorigenic MCF-10F celllines showed a significant and consistent alterations of vari-ous base substitutions (Figure 2). These substitutions weredirectly proportional to the doses of radiation and had moredeleterious effects when given in combination with 17-βestradiol (E). No mutation was detected in GSK-3β gene inany of the cell lines tested in this model. Also immunopre-cipitation study had confirmed that β-Catenin-GSK-3β com-plex formation was disrupted in tumorigenic and tumor celllines due to this mutation (Figure 3). This situation ulti-mately leads to over-expression of GSK-3β in the system(7).

It is concluded that β-Catenin plays an important role as

Fig. 1. Diagrammatic representation of the movement of β-Catenin as a signaling molecule within different cellular compartments.P = phosphorylation, MUT = mutated.

MOLECULAR STUDIES

73

a signaling molecule during radiation- and estrogen-inducedbreast cancer progression.

References

1. Polakis P, The oncogenic activation of β-Catenin, Curr.Opt. Genet. & Dev. 9:15-21, 1999.

2. Iwao K, Nakamori S, Kameyama M, Imaoka S,Kinoshita M, Fukui T, Ishiguro S, Nakamura Y andMiyoshi Y, Activation of β-Catenin gene by interstitialdeletions involving exon 3 in primary colorectalcarcinomas without adenomatous polyposis colimutations, Cancer Res. 58:1021-1026, 1998.

3. Rubinfeld B, Robbins B, El-Gamil M, Albert I, Porfiri E,Polakis P, Stabilization of β-Catenin by genetic defectsin melanoma cell lines, Science (Washington DC)275:1790-1792, 1997.

4. Schlosshauer PW, Brown SA, Eisinger K, Yan Q,Guglielminetti ER, Parsons R, Ellenson LH, KitajewskiJ, APC truncation and increased β-Catenin levels in ahuman breast cancer cell line, Carcinogenesis21(7):1453-1456, 2000.

5. Morin PJ, Sparks AB, Korinek V, Barker N, Clevers H,Vogelstein B, Kinzler KW, Activation of β-Catenin-Tcfsignaling in colon cancer by mutations in β-Catenin orAPC, Science 275:1787-1789, 1997.

6. Calaf G and Hei TK, Establishment of a radiation- andestrogen-induced breast cancer model, Carcinogenesis21(4):769-776, 2000.

7. Roy D, Calaf G and Hei TK, Profiling of differentiallyexpressed genes induced by high linear energy transferradiation in breast epithelial cells, Mol. Carcinogen.31:192-203, 2001. n

1 <------No Change-------->227<------------Mutation------------->298<-------------No Change---------->360

219<------------------------------------------------------------------------------------------------------->26810F => CAGAATGCAG TTTTGAGAAC TAAAAAGTTA GTGTCTAATA GTTTAAATAA10F+E => C T C A60α => G T A A60α+E => C T A A60α/60α=> G T A A60α/60E=> C T A A60E/60E=> C T A TT-2 => G C A T

269<------------------------------------------------------------->29810F => AATGTTGCGG TGAACAAAAC ATACTCATAG10F+E => C C A C T C60α => T C A C T C60α+E => T C A C T C60α/60α=> T A A A T C60α/60E=> T A A A T C60E/60E=> T A C A C CT-2 => T A C A C A

Fig. 2. Mutation spectrum of exon 3 of β-Catenin gene in various irradiated, tumorigenic and tumor cell lines generated from im-mortalized MCF-10F cell lines. T-2 = tumor-2.

Fig. 3. Western immunoblotting showing the GSK-3β im-munoblot hybridize with β-Catenin primary antibody. L1 →MCF-10F; L2 → MCF-10F+E; L3 → MCF-10F+60α; L4 →MCF-10F+60α+E; L5 → MCF-10F+60/60α; L6 → MCF-10F+60α/60α+E; L7 → MCF-10F+60α+E/60α+E; L8 → Tu-mor-2; M → Cruz Marker molecular weight standards. β-Catenin-GSK-3β complex is formed around 125-130 KD.

RADIOLOGY AND RADIATION THERAPY

75

Risks of Radiation-Induced Cancer from Pediatric CTDavid J. Brenner, Carl D. Elliston, Eric J. Hall, Walter E. Berdon1

The use of CT has increased rapidly in the past two dec-ades, fueled in part by the development of helical CT. Forexample, the estimated annual number of CT examinationsin the United States rose approximately sevenfold from 2.8million in 1981 to 20 million in 1995. By their nature, CTexaminations contribute disproportionately to the collectivediagnostic radiation dose to the population; for example, inBritain it has been estimated that approximately 4% of diag-nostic radiology procedures are CT examinations, but theircontribution to the collective dose is approximately 40%.The corresponding data for a large US hospital are that 11%of radiological procedures are CT, giving rise to 70% of thecollective dose.

Evaluated lifetime cancer mortality risks per unit dose asa function of age at exposure are given both by the NationalAcademy of Sciences Biological Effects of Ionizing Radia-tions committee and by the International Commission onRadiological Protection, as shown in Figure 1. Both arebased on relative risk models that depend on sex, age at ex-posure, and time since exposure, and inherently assume alinear extrapolation of risks from intermediate to low doses.

Our basic technique is to multiply age-dependent lifetimecancer mortality risks (per unit dose) by estimated age-de-pendent doses produced by various CT examinations.

Figure 2 shows the estimated lifetime cancer mortalityrisk attributable to a single CT examination performed atdifferent ages. The combination of larger doses and in-creased lifetime risks in pediatric CT result in a sharp in-crease in estimated risk relative to adult CT. Results areshown for two of the most common routine CT examina-tions, CT of the head and CT of the abdomen.

To generate an estimate of the absolute numbers of can-cer deaths attributable to CT examinations, we first appliedthe 1995 United States rate of CT examinations (91/1000population per year) to the current United States population;second, we subdivided this rate, assuming that 40% of CTexaminations are of the head and 20% are abdominal; third,we further subdivided this rate into 5-year age-at-examination intervals, assuming the age distribution for CTexaminations from the 1989 British survey; and finally, weapplied lifetime mortality risks as a function of age at CT, ascalculated here.

On the basis of these assumptions, the predicted totalnumber of deaths attributable to 1 (current) year of CT ex-aminations in the United States is approximately 700 fromhead examinations and approximately 1800 from abdominalexaminations, of which approximately 170 and 310, respec-tively, would be attributable to head and abdominal CT ex-aminations in individuals who were less than 15 years at thetime of examination. In both cases, childhood CT examina- 1 Department of Radiology, Division of Pediatric Radiology.

tions contribute significantly to the overall estimated CT-related potential cancer mortality.

On the basis of the standard models applied here, thelifetime cancer mortality risks attributable to radiation froma pediatric CT examination are estimated to be considerablyhigher than for adults.

Although the absolute estimated risks that we have pro-jected are quite high, the percentage increase in the cancermortality rate over the natural background rate is very low.For example, of the approximately 600,000 children lessthan 15 years old who are estimated to undergo CT eachyear in the United States, approximately 140,000 will ulti-mately die of cancer. Thus, the estimated projected 500 CT-related deaths represents a small (≈0.35%) percentage in-crease over this background. This small estimated relativerisk suggests that detection of an increased risk in an epide-

Fig. 1.

0 10 20 30 40 50 60 70 80

Age at CT Examination (Years)

Est

imat

ed L

ifetim

e A

ttri

buta

ble

Ris

k (%

)

AbdominalHead

0.00

0.05

0.10

0.15

200 mAs

Fig. 2.


76

miologic study would not be easy.Specifically, the dose delivered in most pediatric CT ex-

aminations could potentially be reduced by reducing themilliampere-seconds either manually or automatically andby increasing the pitch. Various authors have suggested thatpediatric CT exposures (i.e., milliampere-seconds) could bereduced by at least 30-50% relative to adult exposures to

obtain essentially the same information; such reductionswould result in a corresponding decrease in radiation risks(whatever they might be) by the same factors.

Of course, in most situations in which pediatric CT isused, the risk-benefit balance is strongly tilted toward bene-fit, which may explain why reduce exposure settings are notroutinely used for pediatric CT. n


76

RARAF Staff Photo

RARAF staff (l-r): Dr. Alan Bigelow, Dr. Brian Ponnaiya, Ms. Gloria Jenkins-Baker, Dr. GerhardRanders-Pehrson, Dr. Charles Geard, Mr. Stephen Marino, Dr. David Brenner, Dr. AlexanderDymnikov, Mr. Mutian Zhang.


77

The Radiological Research Accelerator FacilityAN NIH-SUPPORTED RESOURCE CENTER – WWW.RARAF.ORG

Director: David J. Brenner, Ph.D., D.Sc.Manager: Stephen A. Marino, M.S.

Chief Physicist: Gerhard Randers-Pehrson, Ph.D.

Research Using RARAF

Interest in the “bystander” effect, in which only somecells are irradiated and there is a response greater than wouldbe expected for the fraction of cells irradiated, continued at ahigh level again this year. Several experiments with a varietyof endpoints have been undertaken to determine the size ofthe effect and the mechanism by which it is transmitted.There is evidence for both direct cell-cell communicationthrough cell membranes and indirect, longer-range commu-nication through some release into the cell medium. In someexperiments, the unirradiated cells can be identified due to adifferent staining and scored di-rectly. Both the microbeam and thetrack segment facilities continue tobe utilized in various investigationsof this phenomenon. The single-particle microbeam facility providesprecise control of the number andlocation of particles, but is some-what limited in the number of cellsthat can be irradiated. The tracksegment facility provides broadbeam irradiation that has a randompattern of charged particles but al-lows large numbers of cells to beirradiated.

The experiments performed atRARAF during the period May 1,2000 through April 30, 2001 and thenumber of days each was run in thisperiod are listed in Table I. Fourteendifferent experiments were run dur-ing this 12-month period, about thesame as the average for 1995-2000.Ten experiments were undertakenby members of the CRR, supportedby grants from the National Insti-tutes of Health (NIH), the Depart-ment of Energy (DOE) and theAvon Products Foundation. Fourexperiments were performed byoutside users, supported by grantsand awards from the NIH and theMinistry of Education, Science,Sports and Culture of Japan. Briefdescriptions of these experimentsfollow.

Investigations involving the on-

cogenic neoplastic transformation of mouse C3H 10T½ cells(Exp. 73) were continued by Satin Sawant of the CRR. Us-ing the microbeam facility, a fraction of the cell nuclei wasirradiated through the nucleus to observe the bystander ef-fect. Irradiation of 10% of the cells with 4 or more alphaparticles yielded less than 90% clonogenic survival and atransformation rate significantly higher than 10% of the ratewhen all the cells were irradiated. This indicates the pres-ence of a bystander effect. Cells first given a low dose of250 kVp X rays before microbeam irradiation showed higherclonogenic survival compared to the corresponding popula-tions treated with same number of alpha particles through

Table I.

Experiments Run at RARAF May 1, 2000 - April 30, 2001Exp.No. Experimenter Institution Exp.

Type Title of Experiment DaysRun

73 S. SawantE. Hall CRR Biology Neoplastic transformation of C3H 10T½

cells by specific numbers of α particles 21.5

84 W. Morgan(C.R.Geard)

Universityof Maryland Biology Genomic instability using specific

numbers of α particles 2.0

92 S. Amundson NIH Biology Functional genomics of cellularresponse to high-LET radiation 1.0

94 B. PonnaiyaC.R. Geard CRR Biology

Single cell responses in hit andbystander cells: single-cell RT-PCR

and protein immunofluorescence14.5

100 T. Kumaravel(B. Ponnaiya) NIH Biology

Comet assay of normal and AtaxiaTelangiectasia cells irradiated withspecific numbers of alpha particles

4.0

101 K. Komatsu(H. Zhou)

HiroshimaUniv. Biology Bystander effect of Ataxia

Telangiectasia cells 2.0

102 H. ZhouT.K. Hei CRR Biology Mutagenesis of alpha particle traversal in

normal human bronchial cells 7.5

103 G. JenkinsC.R. Geard CRR Biology Damage induction and characterization

in known hit versus non-hit human cells 14.5

104 A. XuT.K. Hei CRR Biology

Role of reactive oxygen species incytoplasmic irradiation by alphaparticles using CM-H2DCFDA

15.0

106 B. PonnaiyaC.R. Geard CRR Biology Track segment alpha particles, cell

co-cultures and the bystander effect 3.5

107 H. Zhou T.K. Hei CRR Biology Effects of irradiated medium with or

without cells on bystander cell response 2.0

108H. Zhou

M. SuzukiT.K. Hei

CRR Biology Modulation of adaptive response inalpha-particle-induced bystander effects 7.0

109 A. BalajeeC.R. Geard CRR Biology DNA damage induction in microbeam-

irradiated cells assessed by the comet assay 3.5

110H. ZhouD. Roy

T.K. HeiCRR Biology

Identification of molecular signalsof alpha particle-inducedbystander mutagenesis

6.0

Note: Names in parentheses are CRR members who collaborated with outside experimenters.


78

their nuclei.William Morgan of the University of Maryland, in col-

laboration with Charles Geard of the CRR, continued use ofthe microbeam facility to investigate genomic instability(Exp. 84), this time using hamster cells containing humanchromosome 4. A fraction of the cells, stained so they couldbe located by the microbeam imaging system, was irradiatedin their nuclei with alpha particles and the neighboring unir-radiated (unstained) cells cultured to observe genomic insta-bility in the human chromosome.

An experiment employing cDNA microarray technology(Exp. 92) by Sally Amundson of the National Institutes ofHealth (NIH) was continued this year to study bystandereffects at the level of gene expression. Aluminum half-discsplaced under the Mylar-bottomed cell culture dishes for thetrack segment facility allowed irradiation of the “inducer”portion of the dish with alpha particles having an LET of125 keV/um while shielding the bystander cells. RNA har-vested from the separate halves of the dishes after allowing 4or 24 hours incubation was used for cDNA microarray hy-bridization to monitor differences in gene expression. In thispreliminary experiment, a pattern of changes in gene expres-sion for unirradiated bystander cells was observed that wasdistinct from but overlapping with that induced in the di-rectly irradiated cells. Comparison of the genes regulated bydirect irradiation with those responding to bystander signal-ing should identify candidates for genes transmitting andregulating the bystander signal.

Two studies involving the bystander effect were contin-ued by Brian Ponnaiya and Charles Geard of the CRR. Aprotocol has been developed in which a single cell can beobserved for gene expression using reverse transcriptionpolymerase chain reaction (RT-PCR) (Exp. 94). Copies ofDNA segments are created by reverse transcription fromRNA produced by the cell. The DNA is then amplified byPCR until enough material is available for gel electrophore-sis. This method permits observation of individual responsesto radiation instead of just the average response of a largenumber of cells. A fraction of the cells is stained with a nu-clear dye that fluoresces blue, used for the microbeam facil-ity to observe nuclei during irradiation. The others arestained with a cytoplasmic dye that fluoresces orange andare not irradiated because they are not visualized in the mi-crobeam system. Irradiated cells and unirradiated (by-stander) cells each can be identified after irradiation by usingtwo excitation wavelengths and observing the different colorstains. Individual cells are selected using a micromanipulatoron the off-line microscope system of the microbeam facility.Immunofluorescence is used to quantify p21/WAF1 induc-tion in irradiated and unirradiated cells as a function of thenumber of particle traversals through the cell nuclei and timeafter irradiation. The majority of bystander cells show a re-sponse, indicating that the signal from hit cells is wide-spread. These are the first studies wherein the bystander ef-fect has been directly visualized rather than inferred. Theother investigation (Exp. 106) involves use of the track seg-ment facility for broad-beam charged particle irradiations ofhuman fibroblasts and epithelial cells immortalized withtelomerase. Stainless steel rings have Mylar epoxied to both

sides, cells are plated on both inner surfaces and the volumeis filled with medium. Cells on one surface are irradiatedwith 4He ions; cells on the opposite surface are unirradiatedbecause the particle range is much too short. This eliminatesall possibility of cell-cell contact. Some experiments in-volved no cells on the irradiated surface, looking for anyeffects of irradiating only the cell medium. Cells are ob-served in situ at 4 time periods up to 72 hours after irradia-tion with doses from 0.1 to 100 Gy of 125 keV/µm 4He ions.Plateau phase cells are scored for cell cycle delay and mi-cronucleus production while log phase cells are scored forchromosomal aberrations. Yields of micronuclei in by-stander cells ranged from 1.2- to 1.8-fold higher than con-trols, with no clear increase with alpha particle number.

A study of the effects of 4He ions on normal and AtaxiaTelangiectasia human fibroblasts using the comet assay(Exp. 100) by T. Kumaravel of the National Institutes ofHealth continued with the collaboration of Brian Ponnaiyaof the CRR. This procedure, like the single-cell PCR assay(Exp. 94), is a way to observe effects in individual cells.Because the cells are irradiated using the microbeam facility,the number of 4He ion traversals is known, so variability inresponse is solely due to individual variability in the cellsand the stochastic nature of the radiation.

Kenshi Komatsu of Hiroshima University in Japan, incollaboration with H. Zhou of the CRR, has also studied thebystander effect on Ataxia Telangiectasia (AT) cells (Exp.101) using the microbeam facility. Much evidence indicatesthat p53 may play a crucial role in the bystander effect. Atmis a kinase for the phosphorylation of several proteins in-cluding p53 and seems to be the sensor of DNA damage orcenter of signal transduction. AT cells lack Atm and there-fore could provide useful information on the role of p53 inthe bystander effect. In current experiments, cell survivalwas measured for radiation sensitive AT cells that received 5alpha particles in each nucleus and normal Hx cells that re-ceived 10 alpha particles. In addition, the bystander effectfor HPRT mutation in Hx cells was determined by irradiat-ing 10% of the cell nuclei with 20 alpha particles. Futureexperiments will be performed to observe bystander effectsin the AT cells.

Hongnig Zhou and Tom Hei of the CRR are using thesingle-particle microbeam facility for several experiments,most investigating the bystander effect. One study involvesthe mutagenesis of normal human bronchial cells with alphaparticles (Exp. 102). The cells have been irradiated with asingle alpha particle in each nucleus using the microbeamfacility and examined for HGRT mutations. Because yield isvery low, not enough cells have been irradiated yet to form adetermination of the effect. However, a mutation rate abovebackground has been observed using the track segment fa-cility, where orders of magnitude more cells can be irradi-ated. Doses of alpha particles were used that produced anaverage of 1-8 particle tracks per cell nucleus. A secondstudy is examining adaptive response in bystander effects(Exp. 108) in human-hamster hybrid (AL) cells. After low-dose X-ray irradiation, 10% of the cells are traversed by 1 or20 alpha particles. There is a decrease in the bystander effectfor mutation when neighbor cells are traversed by one parti-


79

cle and a somewhat smaller decrease for traversal by 20 par-ticles. With Debasish Roy of the CRR, they are trying toidentify the molecular signals of cell-cell communication inbystander mutagenesis (Exp. 110). AH1-9 cells (AL cellsplus a hygromycin resistance vector in the human chromo-some 11) were transfected with either a dominant negativeconnexin 43 vector, which shut down gap junction commu-nication, or with connexin 43-expressing vector. Using themicrobeam facility, 20% of the cells were irradiated with asingle alpha particle. The data indicate that AL cells con-taining the connexin 43 vector expressed a higher bystandermutagenic yield than that of vector control. In contrast, therewas no significant mutagenic effect observed among ALcells containing the dominant negative connexin 43 vector.These studies provide clear evidence that irradiated cellswith alpha particles can induce bystander mutagenic re-sponse in non-irradiated neighboring cells, and gap junctioncell-cell communication plays a critical role in mediatingsuch a bystander mutagenesis.

Damage induction in irradiated and bystander cells isbeing studied by Gloria Jenkins and Charles Geard of theCRR (Exp. 103). Using the microbeam facility, specificnumbers of alpha particles are delivered to normal humanfibroblasts. Bystander cells are stained as described above inExperiment 106 to distinguish them from those irradiated. Insitu immunofluorescence is used to assay the levels of p53,p21, mdm2 and a number of other damage response proteinsin both the irradiated and unirradiated cells. The totalamount of protein can be determined as a function of thenumber of particles delivered to the irradiated cells and thefraction of cells irradiated.

An Xu and Tom Hei of the CRR have continued investi-gating the role of nitric oxide (NO), an important bioregula-tory molecule, in mediating the mutagenicity of cytoplasmicirradiation (Exp. 104) using the microbeam facility. AL cellswere irradiated with 8 4He ions through the cytoplasm in thepresence or absence of L-NMMA, which has been shown tocompetitively inhibit nitric oxide synthases (NOS). Pre-treatment with L-NMMA suppressed mutation induction by~3-fold to near background level. In contrast, the treatmenthad no effect on the mutagenic yield in cells irradiated by 2alpha particles through the nucleus.

Using the track segment facility, Hongning Zhou andTom Hei of the CRR have investigated the effects on by-stander cells that are not in contact with irradiated cells(Exp. 107). As in Experiment 106, cells were plated on op-posite sides of a special dish with two Mylar surfaces. ALcells were irradiated with doses from 0.1 to 100 Gy of 4Heparticles with an LET of 125 keV/µm. For some dishes, onlythe cell medium (no cells on irradiated side of dish) wasirradiated. Bystander cells were examined 1 and 48 hoursafter irradiation for mutation at the S1 locus. Cell survival inthe bystander cells after 48 hours was reduced by 20% forthe highest dose to the irradiated cells but no statisticallysignificant change in mutation was observed at any dose.Neither survival nor mutation was significantly different for1 hour of co-culture after irradiation. This implies that cyto-toxic factors were released into the cell medium but thesehad minimal effect on mutation in AL cells.

Adayabalam Balajee and Charles Geard have begun in-vestigations of damage induction in human fibroblasts andAtaxia Telangiectasia cells using the comet assay (Exp.109). Cells irradiated through the nucleus using the micro-beam facility show increasing initial damage with increasingnumbers of particles. After 3 hours, much of the damage isrepaired. Future experiments will focus on damage in by-stander cells using some of the identification methods de-scribed above in Experiment 106.

Accelerator Utilization and Operation

Accelerator usage is summarized in Table II. Use of theaccelerator for radiobiology and associated dosimetry in-creased by about 15% over last year and was ~40% higherthan the average for 1995-2000. Over 90% of the acceleratoruse for all experiments was for microbeam irradiations. Be-cause of the relatively low number of cells that can be irradi-ated in a day, these experiments often require considerablebeam time to obtain sufficient biological material, especiallyfor low probability events such as transformation and muta-tion. In addition, there continues to be interest in “bystander”experiments that produce low yields even for normally fre-quent responses.

No utilization of the accelerator by radiological physicsand chemistry occurred this past year. Two ongoing projects,one for physics and the other for chemistry, are scheduled torun again in November of this year and there are at least twomore physics experiments under discussion.

Use of the accelerator for online development doubledover last year and was 50% higher than 1998-99 due to theinstallation of the new 90° bending magnet and beam com-ponents and to increased testing and modification of the mi-crobeam focusing system.

We have continued to minimize the time spent for in-spections of the radiation safety system by not inspectingthose systems and target stations that are rarely used. Ofcourse, any facility is inspected before it is put back into use.

Accelerator reliability was somewhat better than normalthis year. Maintenance and repair time was about 20% lowerthan last year and slightly less than our long-term average.No major repairs or modifications to the accelerator wereperformed. A vacuum leak in one of the sections of the ac-celeration tube is a problem that has troubled us for a coupleof years and will require a permanent repair, possibly nextyear. Whenever we open the accelerator, we have to reseal a

Table II.

Accelerator Use, May 2000 - April 2001Percent Usage of Available Days

Radiobiology and associated dosimetry 42% Radiological physics and chemistry 0% On-line facility development and testing 38% Off-line facility development 3% Safety system 2% Accelerator-related repairs/maintenance 9% Other repairs and maintenance 6%


80

portion of one of the 17 glue joints in one of the seven accel-eration tube sections. Once sealed it poses no problem untilthe next opening, when the changes in tank pressure disturbthe seal. Since there are no spares, we are dismantling dam-aged tube sections that were removed previously in order torecover the electrodes. After these electrodes have been re-conditioned, new tube sections will be assembled usingspare insulators. The assembly process requires heat andpressure to create a strong bond between the insulators to theelectrodes using vinyl sealing material. The special oven atBrookhaven National Laboratory that was used for this pro-cess no longer exists. We will have to either build a new oneand assemble the tube sections ourselves, or send the partsout for assembly to one of the few companies that build ac-celeration tubes.

Development of Facilities

The considerable development of the single particle mi-crobeam facility is described here briefly:• Testing of the single electrostatic quadrupole quadrupletcontinued, using the existing microbeam facility. The beamnow has been reduced down to 11 µm diameter for an objectaperture with a diameter of 40 µm or smaller and the posi-tion is quite steady. Further alignment, modifications andtesting will be done to obtain the calculated demagnificationfactor of 4 so that an object aperture diameter of 8-10 µmwill produce a final beam spot diameter of 2-2.5 µm.• In order to focus the ions better, a “phase space sweeper”was constructed. The beam at the object aperture of the mi-crobeam lens system has a correlation between angular di-vergence and position in the beam, which interfered withfocusing. An electrostatic wobbler was made from sectionsof the ceramic rods from the first lens prototype and posi-tioned below the object aperture. The beam spot, which isconsiderably larger than the object aperture, is movedaround by voltages applied to pairs of electrodes that are atright angles to each other. The voltage on each pair of elec-trodes is varied with a different frequency, selected so thatthere is no repetitive pattern to the motion of the beam.• A fixture was constructed to rotate the ceramic rods usedfor the quadrupole electrodes so that their surfaces couldundergo ion implantation. Six ceramic rods were implantedwith tantalum ions by Dr. Ian Brown of the Lawrence Ber-keley National Laboratory. These rods were substituted forthe ones in the lens that still were exhibiting some sparkingat high voltage. This modification has essentially eliminatedsparking and has allowed the voltage on the lens to be in-creased quickly to the working value of ~15 kV.• The test laser system obtained from the University ofArkansas has been used successfully to extract ions pro-duced by the laser pulses on aluminum, iron and lead targets.Tests using the analyzing and focusing systems providedinformation on the yields of different charge states as afunction of emission angle, laser power density and targetsurface condition.• The new 90º bending magnet was received in January ofthis year and positioned between the exit of the Van deGraaff and the switching magnet to direct the charged parti-

cle beam to the floor above (Fig. 1). It is capable of bendingthe heavy ion beams that will be eventually produced by thelaser ion source under development. Because there is nowless room in this region, several new, more compact beamline components were purchased and installed to replaceexisting ones. Whereas the older components used o-ringseals, the new components all use metal vacuum seals thathave resulted in better vacuum. The magnet was aligned andits bending and focusing characteristics were checked. Acable tray - filled with cables for the magnets and radiationsafety system – which was originally directly above thebeamline had to be relocated, with the cables still in it. Theentire operation lasted five weeks.• Construction of a new microbeam laboratory on the floorover the exit of the Van de Graaff was completed in June ofthis year. The room was designed to be essentially the sameas the existing microbeam lab and has already been equippedwith a cell handling bench, an incubator and the table for themicrobeam apparatus. This will house the next-generationmicrobeam facility with an ultimate beam diameter of <0.5µm.

Microbeam Workshop

We organized a workshop “Probing Individual Cells:Applications to Signaling, Structure and Function” held inBethesda, Maryland on March 12-14, 2001. This was moti-vated by the recent rapid development of new technologiesto study biological responses in individual cells, as well asnew technologies (in particular single-particle/single-cellmicrobeams) for inserting perturbations into individual cellsor parts of cells.

The Workshop provided a forum to assess how suchtools may enrich “conventional” structural biology studies of

Fig. 1. The new 90° bending magnet on its stand is in thecenter of the picture. The switching magnet is on the right and theVan de Graaff on the left. The vertical beam-port flange extendsjust above the cable tray near the top of the picture. Almost allthe beam line components between the Van de Graaff and theswitching magnet were replaced.


81

multi-component signaling and DNA repair complexes orchromatin, and to discuss future directions, both technologi-cal and biological, in these fields. The meeting brought to-gether individuals from a number of very different fields,and much cross-fertilization took place of ideas and tech-nologies in this emerging area of basic cancer research.

Approximately 55 scientists attended the workshop. Themeeting proceedings have been published in the journal Ra-diation Research (vol. 156: 434-445, 2001).

Personnel

• The Director of RARAF is Dr. David Brenner. The Vande Graaff accelerator is operated by Mr. Stephen Marino andDr. Gerhard Randers-Pehrson.• Dr. Alan Bigelow, a postdoctoral fellow, is continuingthe development of the laser ion source as part of his duties.• Dr. Alexander Dymnikov, an expert on ion beamtransport who left RARAF in April 2000, returned as a part-time Visiting Research Scientist in June of this year. He isassisting in the design of the electrostatic lenses for themicrobeam facility.• Mr. Mutian Zhang continues as a full-time acceleratortechnician.• Dr. Charles Geard, the Associate Director of the CRRand biologist on the RARAF grant, spends much of eachworking day at RARAF. In addition to his own research, heis collaborating with several outside users on experimentsusing the single-particle microbeam facility.• Dr. Satin Sawant, an Associate Research Scientist whospent all his time at RARAF primarily doing experimentsutilizing the microbeam facility, left in March of this year.• Dr. Brian Ponnaiya, formerly a postdoctoral fellow andnow an Associate Research Scientist, works at RARAF full-time performing experiments using the track segment andmicrobeam irradiation facilities.• There is one full-time biology technician, Ms. GloriaJenkins.• Two new postdoctoral fellows are expected to arrive inthe year: Dr. Oleg Belyakov and Dr. Stephen Mitchell. Theywill both work full time at RARAF.• Biologists from the Center for Radiological Research notsupported by the RARAF grant spend various amounts oftime at the facility in order to perform experiments:

Recent Publications of Work Performed at RARAF(2000-2001)

1. Balajee AS and Geard CR, Chromatin bound PCNAcomplex formation triggered by DNA damage occurs in-dependent of the ATM gene product in human cells, Nu-cleic Acids Res. 29:1341-1351, 2001.

2. Brenner DJ, Extended Abstracts: Proceedings of the 4thInternational Workshop: Microbeam probes of cellularradiation response, Radiat. Res. 153:220-238, 2000.

3. Brenner DJ and Sachs RK, Are bystander effects relevant

for domestic radon exposure risk estimation?, Int. J. Ra-diat. Biol. (accepted for publication 2001).

4. Geard CR, Jenkins-Baker G, Marino SA and Ponnaiya B,Novel approaches with track segment alpha particles andcell co-cultures in studies of bystander effects, 13thSymposium on Microdosimetry, Stresa, Italy, May 26-June 1, 2001, to be published in Radiat. Prot. Dosim.

5. Geard CR, Jenkins-Baker G, Ponnaiya B and Randers-Pehrson G, Microbeam irradiation of cytoplasmic re-gions of normal human fibroblasts, Int. J. Radiat. Biol.(accepted for publication following revision).

6. Geard CR, Jenkins-Baker G, Ponnaiya B and Randers-Pehrson G, Hit cell fraction and bystander cell responsesin normal human fibroblasts, Radiat. Res. (in press).

7. Geard CR, Jenkins-Baker G, Ponnaiya B and Randers-Pehrson G, Irradiation of inter-cellular medium with mi-crobeam directed alpha particles, Radiat. Res. (submit-ted).

8. Hong J, Craig WW and Hailey CJ, Laboratory tests onneutron shields for gamma-ray detectors in space, Nucl.Inst. Meth. Phys. Res., Section A 452:192-204, 2000.

9. Marino SA and Johnson GW, A microdosimetry cham-ber for low-energy x-rays, Proceedings of the 13th Sym-posium on Microdosimetry, Stresa, Italy, May 26-June 1,2001, to be published in Radiat. Prot. Dosim.

10. Milligan JR, Aguilera JA, Paglinawan RA, Ward JF andLimoli CL, DNA strand break yields after post-high LETirradiation incubation with endonuclease-III and evi-dence for hydroxyl radical clustering, Int. J. Radiat. Biol.77:155-164, 2001.

11. Ponnaiya B, Jenkins-Baker G, Brenner DJ, Hall EJ,Randers-Pehrson G and Geard CR, Biological responsesin individual known microbeam irradiated and non-irradiated bystander cells, Cancer Res. (submitted 2001).

12. Randers-Pehrson G, Microbeams, Microdosimetry andSpecific Dose, 13th Symposium on Microdosimetry,Stresa, Italy, May 27-June 1, 2001, to be published inRadiat. Prot. Dosim.

13. Sawant SG, Randers-Pehrson G, Geard CR, Brenner DJand Hall EJ, The bystander effect in radiation oncogene-sis, Mutation Res. (submitted, 2001).

14. Sawant SG, Randers-Pehrson G, Metting NF and HallEJ, Adaptive response and the bystander effect inducedby radiation in C3H 10T1/2 cells in culture, Radiat. Res.156:177-180, 2001.

15. Sawant SG, Randers-Pehrson G, Metting N and Hall EJ,Can adaptive response alter the bystander effect in C3H10T1/2 cells?, 13th Microdosimetry Symposium, Stresa,Italy, May 26-June 1, 2001, to be published in Radiat.Prot. Dosim.

16. Zhou H, Randers-Pehrson G and Hei TK, Studies of by-stander mutagenic response using charged particle mi-crobeam, Radiation Research 153:236-237, 2000.

17. Zhou HN, Suzuki M, Geard CR and Hei TK, Effects ofirradiated medium with or without cells on bystander cellresponses, Mutat. Res. (accepted for publication). n

RADIATION SAFETY OFFICE

83

RADIATION SAFETY OFFICE 2000-2001

Standing (l-r): Mr. Roman Tarasyuk, Mr. Tom Juchnewicz, Mr. Ahmad Hatami, Dr. Jacob Kamen, Mr.Salmen Loksen, Dr. Ilya Pitimashvili, Mr. Dae-In Kim.

Seated (l-r): Ms. Diana Morrison, Ms. Milvia Perez, Ms. Raquel Rodriguez, Ms. Yvette Acevedo.Not pictured: Mr.Jason Bergman, Dr. Alexander Dymnikov, Mr. Bruce Emmer, Mr. Clifford Jarvis, Ms.

Jennifer Kirchherr, Ms. Olga Loukhton.

PROFESSIONAL STAFF

SALMEN LOKSEN, M.S., CHP, DABR; Director,Radiation Safety Officer

AHMAD HATAMI, M.S., DABR , DABMP; AssistantDirector

THOMAS JUCHNEWICZ, M.S., DABR; AssistantRadiation Safety Officer

JACOB KAMEN, Ph.D., NRRPT, CHP; AssistantRadiation Safety Officer

BRUCE EMMER, M.S., DABMP, DABR; PhysicistILYA PITIMASHVILI, Ph.D.; Radiation Protection

SupervisorDAE-IN KIM, M.S.; Junior Physicist

TECHNICAL STAFF

CLIFFORD JARVIS, B.S.; Chief TechnicianOLGA LOUKHTON, M.S.; Acting Chief TechnicianROMAN TARASYUK; Technician BJENNIFER KIRCHHERR, B.S.; Technician B

SECRETARIAL STAFF

YVETTE ACEVEDO; Administrative AideRAQUEL RODRIGUEZ; Clerk BMILVIA PEREZ; Clerk BJASON BERGMAN, B.S.; Clerk A

CONSULTING STAFF

ALEXANDER DYMNIKOV, Ph.D.; Physicist


85

Radiation Safety Office 2000-2001INTRODUCTION

On May 19, 1957, the President of Columbia Universitydistributed a memo entitled “Directive to All University De-partments Having a Source of Ionizing Radiation,” advisingall parties of the expanded function of the Radiation SafetyCommittee.

Later, a notice entitled “Radiation Safety Guide for Co-lumbia University,” dated February 10, 1959, named PhilipM. Lorio as the Health Physics Officer for University De-partments and Laboratories other than the College of Physi-cian & Surgeons, where Dr. Edgar Watts was the namedHealth Physics Officer. The Chairman of the RadiationSafety Committee was Dr. Gioacchino Failla, who initiatedthe Radiological Research Laboratory in the Department ofRadiology of Columbia-Presbyterian Medical Center(CPMC).

By agreement between The Presbyterian Hospital in theCity of New York (PH) and Columbia University (CU), theRadiation Safety Office (RSO) was established as anautonomous unit in 1962 for the purpose of maintaining ra-diation safety. The Joint Radiation Safety Committee(JRSC), appointed by the Medical Board of CPMC and theVice President for Health Sciences of Columbia University,is charged with the responsibility of defining and ensuringenforcement of proper safeguards in the use of sources ofionizing radiation.

Dr. Harald H. Rossi, Director of the Radiological Re-search Laboratories, was appointed Chairman of the JointRadiation Safety Committee. Under his direction, this com-mittee developed a “Radiation Safety Code and Guide,” theadministration of which is assigned to the Radiation SafetyOfficer. Dr. Eric J. Hall, the present Director of the Centerfor Radiological Research, now chairs the JRSC.

The present Radiation Safety Office came into existencethrough an agreement made on February 12, 1991 betweenNew York State Psychiatric Institute (NYSPI), the Collegeof Physicians and Surgeons of Columbia University (P&S),and The Presbyterian Hospital in the City of New York(PH). This agreement combined several overlapping clinicaland educational programs, including all programs for en-suring radiation safety. On December 16, 1996, Mr. SalmenLoksen was appointed Director of the Radiation Safety Of-fice and Radiation Safety Officer.

The Radiation Safety Office reports to and advises theColumbia-Presbyterian Medical Center and New York StatePsychiatric Institute Joint Radiation Safety Committee,which meets on a quarterly basis. For administrative pur-poses, the Radiation Safety Office reports to Dr. RichardSohn, Associate Dean for Research Administration and Di-rector of Grants and Contracts. It participates in the reviewof research protocols for the Radioactive Drug ResearchCommittee under the jurisdiction of the U.S. Food and DrugAdministration.

Radiation Safety Office staff are Columbia Universityemployees. New York Presbyterian Hospital, Columbia

University College of Physicians and Surgeons, and NewYork State Psychiatric Institute fund the Radiation SafetyOffice budget via a cost sharing payback arrangement.

A full-asset merger between The Presbyterian Hospitalin the City of New York and New York Hospital on Decem-ber 1, 1997, created a single entity known as New YorkPresbyterian Hospital with facilities in two major Manhattanlocations, Columbia Presbyterian Medical Center at West168th Street in Washington Heights and New York WeillCornell Center at East 68th Street on the Upper East Side.However, this merger did not effect the independent opera-tion of the CPMC Radiation Safety Office.

OVERVIEW

Columbia-Presbyterian Medical Center is a large healthsciences campus with extensive teaching, research and clini-cal facilities that use sources of ionizing radiation. The goalof the Radiation Safety Office is to provide adequate protec-tive measures for patients, visitors, students, faculty and staffon campus, and for the general community at large againstexposure to these sources, and to ensure that any dose re-ceived from ionizing radiation is kept “As Low As Reasona-bly Achievable” (ALARA).

The Radiation Safety Office is responsible for ensuringcompliance with federal, state and city regulatory agencies.These regulatory agencies, which mandate rules, regulations,and guidelines, include:

• United States Food and Drug Administration• United States Nuclear Regulatory Commission• New York State Department of Environmental Conser-

vation• New York State Department of Health• New York City Department of Health, Office of Ra-

diological Health.The Radiation Safety Office ensures compliance with

all regulatory requirements and guidelines for the use ofradioactive material and radiation producing machines bymeans of training, education, consultation and a program ofaudits and inspections of facilities. These measures are re-quired pursuant to CPMC Radioactive Materials Licenserequirements and conditions.

The Radiation Safety Office is responsible for main-taining and updating licenses authorizing the use of radioac-tive materials and registrations of radiation producingequipment. Licenses include the New York City Departmentof Health, Office of Radiological Health, Broad Scope Re-search and Broad Scope Human-Use licenses, and specificlicenses for a number of facilities, including Cyclotron,Gamma-Knife and Cobalt-60 Teletherapy units. Registra-tions include New York City Department of Health registra-tions for X-ray equipment and for operation of medical ac-celerators. In addition the Radiation Safety Office maintainsthe New York State Department of Environmental Conser-vation Radiation Control Permit for controlled discharge ofradioisotopes to the environment.


86

Both the New York City Department of Health and theNew York State Department of Environmental Conservationconduct periodic inspections and audits of Columbia-Presbyterian Medical Center and New York State Psychiat-ric Institute facilities operating under their licenses or per-mits. The Radiation Safety Office works continuously toensure that regulatory violations are prevented and to ensurethose that do occur are swiftly corrected.

The Radiation Safety Office also ensures compliancewith the rules and regulations of the “Radiation Code andGuide of Columbia-Presbyterian Medical Center and NewYork State Psychiatric Institute.”

SUMMARY OF SERVICES

The Radiation Safety Office provides the followingprimary services:

• Radiation safety services to:§ Columbia University§ College of Physicians & Surgeons§ New York Presbyterian Hospital§ New York State Psychiatric Institute§ Allen Pavilion Facility§ Cyclotron Facility§ Radioligand Laboratory§ PET Net Pharmaceuticals, Inc.§ Audubon Biomedical Science and Technology Park

(Audubon I)§ Russ Berrie Medical Science Pavilion (Audubon II).

• Radiation safety support for clinical procedures per-formed at New York Presbyterian Hospital-Columbia-Presbyterian Center.

• Routine and specialized laboratory inspections for Hu-man and Non-Human Use.

• Leak testing and inventory of sealed sources.• Gas clearance time measurements and calculations for

areas where radioactive gasses are used.• Area and roof monitoring for all related areas.• Monitoring and evaluation of radioisotope effluent dis-

charges to the atmosphere and sewer system.• Calibration of radiation survey instruments.• Receive, ship and track radioactive material packages,

and wipe test packages for radioactive contamination.• Pick-up, storage and disposal of radioactive waste from

laboratories and hospital facilities.• Supervise and assist with cleanup of contaminated ar-

eas.• Emergency response, including weekends and after

hours, in event of radiation accidents.• Review and evaluation of Human and Non-Human Use

Protocols for the Joint Radiation Safety Committee andthe Radioactive Drug Research Committee for compli-ance with federal, state and local regulatory require-ments.

• Evaluation of education, training and experience of Re-sponsible Investigators seeking to obtain radioactivematerials and/or operate radiation-producing equipmentfor laboratory or hospital use.

• Initial and annual refresher training to personnel in-volved in handling radioactive materials or operating

radiation-producing equipment.• Personnel radiation dose monitoring and investigation

of overexposure reports.• Bioassay testing, including personnel thyroid uptake

and urinalysis.• Review and authorization for purchase, and monitoring

and quality assurance testing of non-radiology dentaland medical X-ray equipment.

• Consultation for radiation shielding requirements.• Provide consultation for pregnant radiation workers.

ITEMIZED SERVICES

The statistical data detailed below are for the fiscal year,July 1, 2000 through June 30, 2001. Instances of RadiationSafety Office support, activities, incidents and response,include those from the last annual report, December 2000,through December 2001.

NYC ORH LICENSES, AUDITS AND INSPECTIONS

1. The Radiation Safety Office continued to maintain sev-eral City of New York Radioactive Materials Licenses: Li-cense No. 75-2878-01 (Broad Scope Human Use), LicenseNo. 92-2878-02 (Teletherapy), 74-2878-03 (Non-HumanUse), License No. 52-2878-04 (Cyclotron Facility), LicenseNo. 93-2878-05 (Gamma Knife), and City of New YorkTherapeutic Radiation LINAC Unit Certified RegistrationNo. 77-0000019.2. The Radiation Safety Office performed 418 routine ra-diation safety inspections and audits of Columbia Universityand New York State Psychiatric Institute research laborato-ries using radioactive materials, and results were communi-cated to Responsible Investigators. A total of 124 deficien-cies were followed up with correction of the cited deficien-cies.3. The Radiation Safety Office completed 835 quarterlyand annual inspections and audits of Columbia-PresbyterianMedical Center and New York State Psychiatric Instituteclinical facilities using radioactive materials. The inspectionsand audits are to ensure compliance with City of New YorkRadioactive Materials License conditions and with RCNYArticle 175, Radiation Control. The facilities audited in-clude: New York Presbyterian Hospital Nuclear Cardiology,Nuclear Medicine, Allen Pavilion Nuclear Cardiology andAllen Nuclear Medicine.4. Audits of non-radiology X-ray facilities were conductedin July and August 2001. The departments audited includedAnimal Care, Endoscopy, Interventional Cardiology, PainManagement, Surgery and Urology. Letters detailing auditresults were sent of the departments involved and copies sentto Medical Physics.5. All quarterly and annual inventory and leak testing forall the human-use departments, i.e., Cyclotron Facility, Al-len Pavilion Nuclear Medicine, Anesthesiology, MilsteinNuclear Medicine and Research (VC 11 alpha sources, irra-diators, etc.) were done and found to be in compliance. Leaktest certificates were created for sealed sources in the abovelocations.6. Ventilation was measured in 83 of the 195 fume hoods


87

in areas where gases or aerosols were used or spilled, andclearance times were posted. Adjustments were made to theair supply and exhaust system to negative pressure condi-tions.7. As required by 6 NYCRR Part 380 and the conditions ofour Permit, the Radiation Safety Office reviewed controlledsewer disposal of aqueous radionuclides. Quarterly and an-nual discharges for all isotopes were well below the concen-tration limits of 6 NYCRR Part 380-11.7 Table II.8. In July 2001 the New York City Department of Health,Office of Radiological Health, conducted a re-inspection offacilities and an audit of records of the Gamma Knife Facil-ity in the Department of Radiation Oncology operating un-der Radioactive Materials License No. 93-2878-05. This re-inspection verified the implementation of recommendationsmade following the inspection of February 28, 2001. Areaschecked for compliance were: internal and external audits ofthe Gamma Knife, the requirement for second physicschecks in calculations, quality assurance for coordinate errorand linearity, and that more than one physicist be named onthe license. The inspector found no violations, but he made arecommendation that the Gamma Knife QA manual be fur-ther improved by including individual subsections specifi-cally addressing authorized users, treatment review and sign-off, treatment planning, dose calculation and verification,monthly spot checks, and annual calibration. He further rec-ommended that future external ACR audits contain a partevaluating the Gamma Knife program explicitly.9. In addition to regular inspection of research laborato-ries, the Radiation Safety Office investigated all major spills,incidents, misadministrations and anomalous exposures, andprovided timely notice of reportable incidents to the NewYork City Department of Health, Office of RadiologicalHealth.10. In August 2001 the Radiation Safety Office submitted tothe New York City Department of Health, Office of Radio-logical Health an application to amend Radioactive Materi-als License No. 74-2878-03 (Non-Human Use). Theamendment request is to add to the License authorization topossess SMP Model PET Gallium/Gernanium-68 sealedsources for calibration of a Concorde MicroPET small labo-ratory scanner. The amendment request was reviewed andapproved by a quorum of the Joint Radiation Safety Com-mittee on September 26, 2001 and was granted by the Officeof Radiological Health on October 2, 2001.11. From September 4, 2001 through September 19, 2001,the New York City Department of Health, Office of Radio-logical Health conducted a comprehensive inspection offacilities and an audit of records for all activities conductedunder the City of New York Radioactive Materials LicenseNo. 75-2878-01 (Broad Scope Human Use). The activitiesconducted under this license include all clinical patient man-agement and human subject research using radioactive mate-rials at New York Presbyterian Hospital, Columbia Univer-sity Health Sciences Campus and the New York State Psy-chiatric Institute. The City inspectors audited patient charts,including that of an I-131 therapy patient for thyroid carci-noma then in progress. At the exit interviews conducted onSeptember 20, 2001, the inspectors stated that they hadfound no violations, but made recommendations with regard

to documenting HDR annual checks, documenting dose cal-culation algorithms, and that the responsible radiation ther-apy physicist be named on this License and on other Li-censes and Registrations.12. The Radiation Safety Office obtained from the City ofNew York, Department of Health, Office of RadiologicalHealth Amendment 17 of Radioactive Materials License 75-2878-01 (Human-Use) authorizing Columbia-PresbyterianMedical Center the possession and use of the Novoste Beta-Cath System employing Strontium-90/Yttrium-90 sealedsources for intravascular brachytherapy. In addition to ob-taining the License Amendment, the Radiation Safety Officesent a board-certified member of its professional staff toreceive didactic Novoste training and to participate in three(3) patient cases under the supervision of the Novoste repre-sentative, in order to be certified by Novoste to perform theMedical Physicist/Radiation Safety Officer function as amember of a Novoste authorized VBT team for intravascularbrachtherapy. The Radiation Safety Office continues to sup-port Radiation Oncology and Interventional Cardiology inthis role.

NYS DEC PERMITS, AUDITS AND INSPECTIONS

13. The Radiation Safety Office continued to maintain NewYork State Department of Environmental ConservationPermit No. 2-6201-00005/00006.14. As required by 6 NYCRR Part 380 and the conditions ofour NYSDEC Radiation Control Permit, the RadiationSafety Office has reviewed atmospheric discharges of posi-tron-emitting radioisotopes from the Cyclotron Facility andthe Radioligand Laboratory. All atmospheric dischargeswere within the maximum annual quantity authorized.15. The 2001 Annual Report to the New York State De-partment of Environmental Conservation (DEC) was sub-mitted in a timely manner and favorably received by theDEC.16. The Radiation Safety Office received from theNYSDEC Radiation Section a June, 2001 communicationthat radioactive discharges for 2000 were in compliance withthe effluent limits of our Permit and the requirements ofSection 380-5.1(a).17. On August 31, 2001, the draft copy of a Permit Modifi-cation Application for our NYSDEC Radiation Control Per-mit was sent via Federal Express to the Radiation Section inAlbany, New York, and on September 3, 2001 a final draftwas sent to the Regional Permit Administrator in Long Is-land City, New York. The permit was received on January17, 2002. The 105 page Application:

• Requested an increase in the Maximum QuantityAuthorized for C-11 at Emission Point G-2 (Radioli-gand Lab stack, Milstein Hospital Building roof-top)from 16.00 to 20.00 Curies per year.

• Demonstrated that with this increased Permit Limitpublic doses will remain below the USNRC “ConstraintLimit” of 10 mrem/year.

• Confirmed that the ability to synthesize C-11 com-pounds is of vital concern to on-going Human-Use re-search programs and patient management at the MedicalCenter.


88

18. On October 4, 2001, the New York State Department ofEnvironmental Conservation conducted an inspection offacilities at the Columbia-Presbyterian Medical Center oper-ating under DEC Permit No. 2-6201-00005/00006, Radia-tion Control Permit No 167-3. On November 7, 2001, theRadiation Safety Office received a letter from theN.Y.S.D.E.C. stating that within the scope of their inspec-tions, operations at CPMC were in compliance with6NYCRR Part 380 and the conditions of the CPMC Permit.

ROUTINE ACTIVITIES

19. The Radiation Safety Office provided radiation safetysupport for 94 brachytherapy patients and nine I-131 radio-pharmaceutical therapy patients receiving treatment from theNew York Presbyterian Hospital Departments of NuclearMedicine and Radiation Oncology. This support includedroom preparation, distribution of personnel radiation do-simeters, performance of patient and room surveys, postinginstructions in patient rooms, entering instructions in patientcharts, patient discharge surveys, room decontamination andremoval of patient generated wastes for decay-in-storage anddisposal.20. The Radiation Safety Office received and distributed3,236 radioactive packages with a total activity of 28.42Curies, excluding Nuclear Medicine and Radiation Oncol-ogy shipments. For all shipments the Radiation Safety Officeconducts package surveys, ensures correct distribution toAuthorized Users, and maintains inventory control and asso-ciated records.21. The Radiation Safety Office reviewed Applications forthe Use of Radioisotope and issued isotope orders for thepurchase of S-35, P-32, Ir-192, and I-125. These four are themajor isotopes purchased.22. The Radiation Safety Office reviewed Applications toAdminister Radioactivity to Humans submitted to the CPMCRadioactive Drug Research Committee (RDRC) and/or theCPMC Joint Radiation Safety Committee. All applicationswere approved, some with modifications. In addition, 14new and 62 renewed Responsible Investigator Applicationsfor Non-Human Use of Radioactivity were reviewed andapproved.23. An officer of the Radiation Safety Office participates asan Ad Hoc Member of the Animal Care Protocol ReviewCommittee, reviewing all procedures using radionuclides inanimal research. Protocols involving the use of radioactivematerials in animals were approved.24. The Radiation Safety Office performed 55 routine ani-mal radiation surveys in the Institute of Comparative Medi-cine in order to minimize contamination in animal facilitiesand cages, protect Animal Care staff, and ensure proper dis-posal of animal carcasses with radioactivity.25. The Radiation Safety Office provided 325 calibrationand maintenance services for 501 radiation survey instru-ments used throughout the Columbia-Presbyterian MedicalCenter and New York State Psychiatric Institute. The Radia-tion Safety Office maintains a supply of portable surveyinstruments available for loan to Responsible Investigators.26. The Radiation Safety Office maintained a program foremergency response, including updating a list of individual

and group pager numbers and procedures for Security per-sonnel to contact Radiation Safety Office staff at any time inthe event of an emergency.27. The Radiation Safety Office participates as part of theColumbia University Health Sciences Division (CUHSD)Emergency Management Plan Task Force. The EmergencyManagement Plan is necessary in event that any significantoccurrence disrupts the normal day-to-day operation atCUHSD, including University research activity and/or em-ployee safety. The objective of the plan is to utilize Univer-sity resources in an effective manner should interruption ofan essential service occur. The plan provides written policiesand procedures to be implemented in event of emergenciesincluding radiation spills, chemical spills, transit disruption,utility shutdown, etc. A number of meetings were held inorder to formulate policies, and an Emergency ManagementPlan document was completed.28. On December 14, 2000 and June 14, 2001, the Radia-tion Safety Office shipped the following amounts of Low-Level-Radioactive-Waste: 92 drums of 30 gallon LiquidScintillation Vial wastes totaling 10.31 cubic meters weigh-ing 6,266 kilograms, with a total activity of 82.919 mCi,shipped for disposal by Perma-Fix of Gainesville, Florida.Additional LLRW disposed of on-site by the RadiationSafety Office in fiscal year 2000-2001 totals: 306 drumscontaining 34.78 cubic meters of short half-life researchwastes which were held for decay-in-storage and ultimatelycleared for landfill disposal as regular trash: 5,380 liters oflow-activity aqueous research wastes assayed and disposedof by controlled sewer disposal; 379 thirty-gallon “blackbags” containing waste removed form patient rooms, heldfor decay in storage, and ultimately disposed of as “red bag”patient waste.29. On December 28, 2001, the Radiation Safety Officeassisted in the acceptance testing of the new Ximatron Ra-diotherapy Simulator installed in the Department of Radia-tion Oncology. The Radiation Safety Office conducted theradiation safety portion of the required testing, which in-cluded an inspection of the X-ray unit and facilities for com-pliance with physical and procedural requirements of RCNYArticle 175 and a radiation safety survey of the controlledand public areas in the vicinity of the new installation. Allinitially calculated doses to workers and general public ap-pear to be well below the RCNY Article 175 limits.

TRAINING

30. Pursuant to Article 175 of the New York City HealthCode, the following radiation safety courses and trainingsessions were presented from July 2000 through June 2001:

• 12 initial training seminars for a total of 555 individuals.• 12 refresher seminars for a total of 239 individuals.• 12 nursing seminars for a total of 375 individuals.• 2 special dental seminars for a total of 140 individuals.• 2 training sessions for Radiology residents.• 3 training sessions for the Facility department.

31. The Radiation Safety Office conducted an extensiveseries of in-service and refresher-training sessions focusedon the specific radiation safety requirements of users ofpositron emitting isotopes in research laboratories. Re-


89

searchers were trained in the health physics of PET isotopes,stressing the high energy of annihilation radiation, the highgamma constant, the short half-life, the sharp energy depo-sition curve and the concerns for beta skin dose in compari-son to standard nuclear medicine isotopes such as Tc-99mand I-123. They were instructed in proper levels of shield-ing, specific handling techniques and implements and rec-ords and reports required by our License and Article 175.32. For employees who could not attend the regularlyscheduled classes, the Radiation Safety Office designed andimplemented a self-study program including the use ofvideotapes available at the Health Sciences Library. A pass-ing grade on the quiz administered after viewing the videoqualified an employee working in Non-Human Use applica-tions to be issued a radiation monitor badge. If the individ-ual’s employment involves human use of radioactive mate-rial, a passing grade on the quiz results in obtaining a tempo-rary badge until the next regularly scheduled training ses-sion.33. A certified Diagnostic Radiological Physicist from theRadiation Safety Office presented the Fellow’s Lecture tothe radiology residents and staff.34. The Radiation Safety Office participates as part of theColumbia University Health Sciences Division (CUHSD)Institutional Health and Safety Council (IHSC). The IHSChas encouraged the utilization of Web-based resources toprovide information, education and training to personnel.The Radiation Safety Office continues development of itsWeb page to improve dissemination of information andcommunication with Responsible Investigators and membersof the CPMC community (see http://cpmcnet.columbia.edu/dept/radsafety).35. The Radiation Safety Office made progress toward thegoal of developing a web-based radiation safety trainingprogram by obtaining authorization from a number of ven-dors to use their video training on the Radiation Safety Of-fice website. As a part of these steps, the Radiation SafetyOffice and Dr. Hall, Chairman of the JRSC, engaged in pre-paring material to be utilized in a web-based training pro-gram customized to the requirements of CPMC.36. The Radiation Safety Office participated in planningdiscussions with personnel in Electronic Research Admini-stration in order to facilitate the integration of the RadiationSafety Office computerized database with the RASCALsystem.

PERSONNEL MONITORING

37. The Radiation Safety Office distributed approximately8,800 personnel radiation dosimeters each quarter, includingmonthly and quarterly badges. A total of about 35,200 do-simeters were distributed and collected annually. To main-tain dosimetry records the Radiation Safety Office usesdedicated computers with direct modem access to the ven-dor’s (Landauer, Inc.) database.38. The Radiation Safety Office received Annual Occupa-tional Exposure Reports (NRC Form 5) from Landauer, Inc.for the year 2000 and mailed these reports to radiation work-ers as required by the New York City Board of Health regu-lations.

39. The Radiation Safety Office notified 149 personnel withALARA Level 1 readings, and investigated 45 cases ofALARA Level 2 readings as provided by our personnel ra-diation dosimetry vendor. Particular attention was paid tooccupational groups typically at or exceeding ALARA limitsfor extremity or collar badges, i.e., workers and researchersat the Cyclotron Facility, Angiography, Cath Lab, and phy-sicians in the PET Suite.40. The Radiation Safety Office performed 70 thyroid bio-assays on radiation workers using isotopes of iodine includ-ing I-125, I-123 and I-131.41. During fiscal year 2000-2001, 16 employees completeddeclaration of pregnancy forms. The Radiation Safety Officeprovided them with health physics counseling about riskfactors and additional monitoring of the fetuses for the ges-tation period, and continues to closely follow their exposurereports.42. In April 2001 the Radiation Safety Office requested theNew York City Department of Health, Office of Radiologi-cal Health to provide written authorization to remove anincorrect dose from an employee’s Occupational Dose Rec-ord. The reported dose of 5,461 mrem to the whole body wasdetermined to be the result of a therapeutic radiation oncol-ogy procedure.

X-RAY

43. The Radiation Safety Office has responsibility forQuality Assurance of Dental X-ray units, as well as respon-sibility for acceptance testing and radiation safety surveys onall new units installed at the following locations:

• Morningside Dental Associates – 9 intraoral units, 1panorex unit, and 1 cephalographic unit.

• Ambulatory Care Networked Corporation (ACNC) – 2intraoral units.

• Babies Hospital OR – 1 portable intraoral unit.• Vanderbilt Clinic Teaching & Research Area – 2

panorex units, 2 cephalous-graphic units, and 15 intra-oral units.

• Dentcare Clinic (Intermediate School 183) – 1 intraoralunit.

• New York State Psychiatric Institute – 1 intraoral unit.• Columbia Eastside – 6 intraoral units, 1 panorex unit,

and 1 cephalographic unit.• Columbia North – 5 intraoral units, 1 panoramic unit.

The quality assurance program is designed, to optimize theradiological safety and clinical quality of dental radiography,based on recommendations for quality assurance that havebeen promulgated by a number of professional organiza-tions, including the National Council on Radiation Protec-tion and Measurements (NCRP), the Bureau of RadiologicalHealth of the Food and Drug Administration, the AmericanCollege of Radiology (ACR), and the American Academy ofDental Radiology Quality Assurance Committee.44. X-ray units in the Vanderbilt Clinic 7th floor teachingarea were recalibrated and retested at the request of the Co-lumbia University School of Dental and Oral Surgery toprovide more uniform radiation output among the variousteaching X-ray units. New ESE technique charts were pro-vided for the new “F” speed dental film.


90

45. The Radiation Safety Office provided support to Facili-ties Management by performing preliminary shielding esti-mates of the proposed Vanderbilt Clinic Dental TeachingFacility. The Radiation Safety Office also provided guidanceand support to the architects for Columbia UniversityHealthcare to ensure that adequate radiation shielding thatmeets regulatory requirements would be provided.46. The Radiation Safety Office conducted trials of addi-tional testing equipment for X-ray quality assurance. Theequipment was purchased to enable the Radiation SafetyOffice to continue to perform the best possible quality assur-ance evaluations of X-ray equipment.47. The Radiation Safety Office met with CPMC’s ChiefHospital Physicist of Radiology to discuss standards forquality assurance audits of non-radiology X-ray facilities. Inaddition, there was discussion about the responsibility formandated QA activities of non-radiology equipment, such asperforming lead apron integrity checks and view-box in-spections.48. In association with CPMC’s Department of Radiology,the Radiation Safety Office maintained a radiation safetyinspection and audit program for non-radiology X-rayequipment at CPMC to assure compliance with regulatoryrequirements. The audit program included evaluation ofcompliance with Quality Assurance requirements and proce-dures, attendance of employees at radiation safety trainingsessions, and compliance with regulatory requirements foruse and timely return of personnel radiation dosimeters.Prior to the audit a form was sent to each non-radiology X-ray facility requesting a list of individuals responsible forperforming QA/QC functions and an inventory list of all X-ray equipment and film processors.49. In January 2001 the Radiation Safety Office receivedOffice of Radiological Health (ORH) Information Notice2001-1 from the City of New York, Office of RadiologicalHealth. The notice stated that the New York City HealthDepartment is requesting the Office of Radiological Healthto locate, register and regularly inspect analytical X-rayequipment located within the City of New York. Examplesof such equipment are: electron microscopes, analytic X-rayunits, X-ray fluorescence units, particle accelerators and X-ray machines. The Office of Radiological Health in turn re-quested that the Radiation Safety Office send the NYCORHa list of such equipment at the Columbia-Presbyterian Medi-cal Center, including the type and location of each instru-ment and the individual(s) responsible for the instrument(s).The completed list was sent as requested.50. In March 2001 the Columbia University School ofDental and Oral Surgery submitted plans for a renovation ofthe north side of the VC-9 Dental Clinic to the RadiationSafety Office for review. The Radiation Safety Office re-viewed the plans and the area became operatories for thePrograms in Periodontics and Endodontics.51. In July 2001 the Radiation Safety Office provided as-sistance to the Departments of Endoscopy and Bronchos-copy in planning and preparations for the relocation of threefluoroscopic units from the Presbyterian Hospital Buildingto new facilities on the thirteenth floor of the Atchley Pavil-ion. The Radiation Safety Office performed an X-ray Radia-tion Protection Survey of these fluoroscopic units immedi-

ately upon their installation, and a report was submitted it tothe City of New York Department of Health that all publicradiation doses were found to be in compliance with RCNYArticle 175 requirements.

PET ISOTOPE PRODUCTION

52. The Radiation Safety Office provides radiation safetysupport for the new Columbia University Radioligand Labo-ratory Facility for the synthesis of PET imaging radiophar-maceuticals. During the construction phase of the facility theRadiation Safety Office provided assistance in the designand specification of the radioisotope exhaust and effluentmonitoring systems.53. The Radiation Safety Office is involved in the prelimi-nary planning for the proposed second Cyclotron facility.The Radiation Safety Office is providing professional healthphysics consultation with regard to all radiation safety as-pects of the new facility. This includes shielding design andevaluation for the Cyclotron and radiochemistry laboratoriesand design and specification of the radioisotope exhaust andeffluent monitoring systems, including the design and speci-fication of radioactive waste gas hold-up and treatmentequipment.54. The Radiation Safety Office completed the process ofacceptance testing the Radioligand Laboratory stack moni-toring system purchased from Bicron. Because of problemsidentified by Radioligand Laboratory and Radiation SafetyOffice personnel, Bicron engineers visited the RadioligandLaboratory in order to replace hardware and update soft-ware. Acceptance testing confirmed that the system meets allspecifications provided by Bicron, and that all of the hard-ware and software contracted to be delivered was workingsatisfactorily.55. In February 2001 after protracted negotiations the Ra-diation Safety Office and Michael Reich, Senior ContractsOfficer, Columbia University, entered into agreement withSaint-Gobain Crystals (Bicron) on terms for fulfillment ofthe legacy installation warranty for the Cyclotron stackmonitoring system assumed by Bicron when it took over theobligations of the original Shalco contract with Columbia.The estimated cost of replacing the Shalco installed stackmonitoring system without the settlement is approximately$40,000. This settlement also includes replacement of theRadioligand Laboratory stack monitoring system Ion Cham-ber with a NaI Scintillation Detector, due to failure of theIon Chamber to meet the manufacturer’s specifications. TheRadiation Safety Office thanks Mr. Reich for his invaluableassistance in the above matter and commended him to theJRSC.56. In April 2001 the Radiation Safety Office provided ra-diation safety support to the New York Presbyterian Hospi-tal Facilities Department for the periodic filter replacementand scheduled quarterly preventive maintenance for the Cy-clotron Facility, Radioligand Laboratory and PET Suite ex-haust systems. Periodic filter replacement and quarterlyPM’s are required by the Conditions of our NYSDEC Ra-diation Control Permit.57. In April 2001 the Radiation Safety Office providedtechnical supervision and assistance to the New York Pres-


91

byterian Hospital Facilities Department in preparing the Mil-stein Hospital Building rooftop site for the installation ofnew Bicron NaI stack monitoring systems. Preparations in-cluded installation of electrical service, installation of com-munication cable conduits, hole drilling, installation ofmounting flanges and erection of scaffolding at the Cyclo-tron Facility, Radioligand Lab and PET Suite exhaust stacklocations.58. In April 2001, Roger Moroney, CHP, Regional HealthPhysicist and Ashok Dhar, Radiation Safety Officer of PETNet Pharmaceuticals, Inc., visited the Columbia-PresbyterianMedical Center for a tour of the Cyclotron facility, PETSuite and Radioligand Laboratory. A productive meetingcovered areas of mutual concern, including the contractualagreement between the Columbia University and PET NetPharmaceuticals to provide radiation safety support for PETNet employees and facility operation. In August 2001 twoofficers of the Radiation Safety Office visited the PET Netoperation and the Pharmalogic facilities at Albany, NewYork. They were provided with comprehensive tour of theCyclotron facilities including the roof, stack monitoringsystem and daily operation procedures. They reviewed thestack monitoring calculations and analyzing the emissiondata. The Radiation Safety Office has continued its contactswith PET Net Pharmaceuticals since the visit, exchanginginformation and discussing areas of interest.

MISCELLANEOUS

59. The Radiation Safety Office provided Health Physicssupport to the Department of Radiation Oncology with re-gard to the delivery, installation and required radiation safetysurveys for the replacement of the Phoenix #23 Cobalt-60teletherapy source.60. The Radiation Safety Office upgraded waste monitoringsystems and installation of additional systems at new loca-tions throughout Columbia- Presbyterian Medical Centerand the New York State Psychiatric Institute. The systemsare network capable.61. The Radiation Safety Office assisted Tribeca Pharma-ceuticals in preparing an application for a New York CityDepartment of Health, Office of Radiological Health, Radio-active Materials License for Non-Human Use. The applica-tion was submitted on December 8, 2000, and the licensewas approved on January 25, 2001.62. In February 2001 the Radiation Safety Office filed forrenewal of the State of South Carolina Transportation Permitfor radioactive waste. The renewed Permit was received onMarch 1, 2001 and is valid until the end of the calendar year.On November 12, 2001, the Radiation Safety Office filed forrenewal of the State of South Carolina Transportation Permitfor Radioactive waste for the year 2002. The renewal Permitwas received on December 28, 2001.63. The officers and technical staff of the Radiation SafetyOffice attended a training session on Hazardous ChemicalWaste Management given by a certified Hazardous Materialspecialist from the Columbia University EnvironmentalHealth & Safety Office. At the end of the session all at-tendees were given a certificate of accomplishment.64. On June 28, 2001, in a meeting of the Institutional

Health & Safety Council, there was a discussion about therequirement of the special permit granted by the New YorkCity Board of Estimates for the development of the Auduboncommercial laboratory building (Mary Woodard LaskerBiomedical Research Building), to the effect that ColumbiaUniversity perform inspections of each laboratory in thebuilding at least four times per year, and that copies of allinspection reports must be sent to Community Board 12 andthe Manhattan Borough President’s Office. Shortly after themeeting, Columbia University’s Associate General Counselcontacted the Radiation Safety Office and EnvironmentalHealth and Safety Offices to arrange for a conference call todiscuss this requirement. On September 17, 2001 an officerof the Radiation Safety Office met with the Associate Gen-eral Counsel in order to review the inspection reports for theMary Woodard Lasker Biomedical Research Building per-taining to radioactive materials. These reports are a condi-tion of Special Permit for the Lasker Research Building, andtherefore must be sent to Community Board 12 and theManhattan Borough President’s Office.65. As a result of the tragic events on September 11, 2001,the Radiation Safety Office immediately attempted to con-tact the New York City Health Department, Office of Ra-diological Health to offer assistance and inquire about thewell being of their personnel located in the downtown Man-hattan area. We received a list of pager and home telephonenumbers of ORH personnel in order to maintain communi-cations in the event of communications disruption. The Ra-diation Safety Office staff was on an emergency standby forany possible radiation related issues.66. In July 2001 two officers of the Radiation Safety Officeattended a day-long Hazardous Waste Management trainingsession under the Resource Conservation and Recovery Act(RCRA) at the Russ Berrie Building, arranged by the Co-lumbia University Environmental Health & Safety Depart-ment.67. Eyewash stations were purchased for all RadiationSafety Office waste storage areas. The installation of aneyewash station at the Russ Berrie radioactive waste storagefacility had already been completed.68. The Radiation Safety Office was advised that as ofAugust 14, 2001 Nuclear Diagnostic Laboratory (NDL) Inc.,the previous radioactive waste broker to CPMC, has beenordered by the state Department of Labor to desist from ac-cepting new waste for storage. The staff of the RadiatiojnSafety Office was prescient in changing its waste brokerlong before NDL’s problems were overtly apparent. Al-though the Radiation Safety Office presently maintains apermit to ship very long-lived isotopes with Envirocare inUtah, most of the radioactive waste is now handled inter-nally by the method of decay in storage. This method notonly keeps the institution independent, but also helps theinstitution financially and keeps it immune from legal issuesin the long term.

RSO PERSONNEL AND FACILITIES

69. There were a number of changes in personnel in theRadiation Safety Office. All of the Assistant RadiationSafety Officers are now certified in various areas of medical


92

physics and health physics. The broad range of experienceand certifications of the radiation safety officers in the Ra-diation Safety Office, including Medical Nuclear Physics,Therapeutic Radiological Physics, Diagnostic RadiologicalPhysics and Health Physics enables the office to provide afull range of professional services to Columbia UniversityHealth Science division, New York Presbyterian Hospitaland the New York State Psychiatric Institute.70. Mr. Ahmad Hatami, Assistant Director of the RadiationSafety Office, recently completed his board certification inNuclear Medicine Physics by the American Board of Radi-ology (ABR).71. Dr. Jacob Kamen recently completed certification by theAmerican Board of Health Physics and is now a CertifiedHealth Physicist (CHP).72. In addition, there were changes in the technical staffing

in the Radiation Safety Office. We are fortunate at this timethat all of the staff hired met or exceeded all of the require-ments for the positions filled. The Radiation Safety Officerecently hired a temporary replacement for the Chief Tech-nologist, Ms. Olga Loukhton, who has a Masters Degree inphysics. The Technician B position was filled by Ms. Jenni-fer Kirchherr, B.S. The part-time Clerk A position was filledby Mr. Jason Bergman, B.A. The part-time AdministrativeAssistant position was filled by Ms. Darnell Adams, B.S.,Ed.M.73. At present the Radiation Safety Office is housed in anoffice space on the basement level of the William BlackMedical Research Building. There are plans for the Radia-tion Safety Office to relocate to a larger office suite on the4th floor of the School of Public Health Building. n

PROFESSIONAL ACTIVITIES

93

Professional ActivitiesADAYABALAM S. BALAJEE, Ph.D.

MemberAmerican Association for Advancement of ScienceRadiation Research Society

ReviewerMutation ResearchAdvances in Space ResearchMedical Science Monitor

ALAN BIGELOW, Ph.D.Member

American Physical SocietyHonors

Best Poster Award, 9th International Conference on IonSources, September 2001

DAVID J. BRENNER, Ph.D., D.Sc.Adjunct Faculty

Univ. of California, Berkeley, Miller Professor (2001)Member

Columbia University Radiation Safety Committee,Chairperson

Columbia University Committee on Response to Radia-tion Terrorism, Chairperson

National Council on Radiation Protection and Measure-ments (NCRP)

NCRP Committee 1-6 on Risk LinearityASTRO Refresher Course, Program CommitteeJoint Task Force on Vascular Radiation TherapyRadiation Research Society, Policy CommitteeInternational Symposium on Microdosimetry, Organiz-

ing CommitteeInternational Workshop on Radiation Risk Research in

Southern Urals, Organizing CommitteeInternational Conference on Radiation Damage and its

Modification, Organizing CommitteeTV and Radio appearances on the subject of pediatric CT

examinations

GLORIA M. CALAF, Ph.D.Adjunct Faculty

University of Tarapaca; Faculty of Sciences; Departmentof Biology and Health, Arica, Chile, Adjunct Prof.

MemberBiology Society of ChileMastology Society of ChileChilean Society of CitologyChilean Society of CancerNew York Academy of SciencesTissue Culture AssociationInternational Association of Breast Cancer ResearchAmerican Association of Cancer ResearchSociety of Experimental Biology and Medicine

Radiation Research SocietyReviewer

British Journal of Cancer

CHARLES R. GEARD, Ph.D.Member

American Society of Therapeutic Radiology and Oncol-ogy (ASTRO)

Environmental Mutagen SocietyAssociate Member Radiobiology Advisory Team

(AMRAT) of the Armed Forces Radiobiology Re-search Institute (AFRRI)

Columbia University, Faculty Council, Voting MemberAdvisory Committee on Radiobiology, Brookhaven Na-

tional LaboratoryEditorial Work

International Journal of Radiation Biology, EditorialBoard

ReviewerBritish Journal of CancerMutation ResearchRadiation Research

ERIC J. HALL, D.Phil., D.Sc., FACR, FRCRMember

American Board of Radiology, Radiotherapeutic Writ-ten-Test Committee

National Academy of ScienceAmerican Society of Therapeutic Radiology and Oncol-

ogy (ASTRO)Radiation Research SocietyInternational Stereotactic Radiosurgery Society: Member

of the BoardAmerican Radium Society, Program Comm. ChairmanInternational Association of Radiation Research, Pres.Columbia University, College of Physicians & Surgeons

Cancer Center, Internal Advisory Committee/Execu-tive Committee

Columbia-Presbyterian Medical Center, Joint RadiationSafety Committee, Chairman; Radioactive Drug Re-search Committee, Chairman

National Council on Radiation Protection and Measure-ments, Committee 1, Member

Editorial WorkIntl. Journal of Radiation Oncology Biology PhysicsInternational Journal of Brachytherapy

M. PRAKASH HANDE, Ph.D.Member

Indian Association for Radiation Protection, IndiaAssociation of Medical Physicists of India, IndiaIndian Society for Radiation Biology, IndiaEnvironmental Mutagen Society


94

Radiation Research SocietyNew York Academy of SciencesAmerican Association for the Advancement of Science

HAIYING HANG, Ph.D.Member

Radiation Research SocietyGrant

RSNA scholar grant

TOM K. HEI, Ph.D.Adjunct Faculty

Department of Radiological Health Science, ColoradoState University, Fort Collins, Co., Adjunct Professor

Department of Ion Beam Bioengineering, Chinese Acad-emy of Sciences, Hefei, China, Adjunct Professor

MemberChemical Pathology Study Section, 1998-2001Pathology C Study Section, Chairman, Ad Hoc Review

PanelUICC Fellowship, Ad Hoc Review PanelIntramural Site Visit, Frederick Cancer Center, Bethesda,

Md.Radiation Research SocietyAmerican Association for Cancer ResearchEnvironmental Mutagen SocietyOxygen Society

ReviewerProceedings of the National Academy of SciencesBritish Journal of CancerCancer ResearchCarcinogenesisRadiation ResearchEnvironmental Health Perspective

Editorial WorkAdvances in Space Sciences, section editor

Student MentoringMaster degree students of Environmental Health Sci-

ences, Columbia UniversityNew York City High School science students for Intel

Science project

HOWARD LIEBERMAN, Ph.D.Member

Summer Research Program for NYC Secondary SchoolScience Teachers, Columbia University, AdvisoryBoard

American Association for the Advancement of ScienceAmerican Society for MicrobiologyEnvironmental Mutagen SocietyGenetics Society of AmericaRadiation Research SocietyWeb-Site Committee of the Radiation Research Society,

ChairmanSigma XiTheobald Smith Society

Reviewer Grants:

Basic and Preclinical Subcommittee C of the NCI InitialReview Group, Member

Joint Center for Radiation Therapy Foundation, HarvardMedical School, Reviewer

Israel Cancer Research Fund Grant Review Panel A,Member

Manuscripts:Intl Journal of Radiation Oncology, Biology and PhysicsNature GeneticsNucleic Acids ResearchRadiation Research

STEPHEN A. MARINO, M.S.Member

Columbia University Radiation Safety CommitteeRadiation Research Society

Guest ScientistBrookhaven National Laboratory, Upton N.Y.

TEJ K. PANDITA, Ph.D.Member

American Association for the Advancement of Science.American Association of Cancer Research.Radiation Research SocietyThe American Society of MicrobiologyDepartment of Defense, Prostate Cancer Study Section

ReviewerCancer ResearchCarcinogenesisClinical Cancer ResearchFASEB JournalInternational Journal of Radiation Oncology, Biology

and PhysicsInternational Journal of Radiation BiologyMolecular and Cell BiologyMutation ResearchNature GeneticsNeoplasiaOncology ReportsOncogenePNAS, USARadiation Research

BRIAN PONNAIYA, Ph.D.Member

Radiation Research SocietyReviewer

International Journal of Radiation BiologyRadiation Research

YONG-LIANG ZHAO, Ph.D.Member

Radiation Research Society n

PUBLICATIONS

95

Publications1. Balajee AS and Geard CR, Chromatin bound PCNA

complex formation triggered by DNA damage occursindependent of the ATM gene product in human cells,Nucleic Acids Res. 29:1341-1351, 2001.

2. Brenner DJ, Is the linear-no-threshold hypothesis ap-propriate for use in radiation protection? Favouring theproposition, Radiat. Prot. Dosimetry 97(3):279-82,2001.

3. Brenner DJ and Miller RC, Long-term efficacy of in-tracoronary irradiation in inhibiting in-stent restenosis,Circulation 103(9):1330-2, 2001.

4. Brenner DJ, Elliston CD, Hall EJ and Berdon WE,Estimates of the cancer risks from pediatric CT radia-tion are not merely theoretical: comment on“Point/counter-point: in x-ray computed tomography,technique factors should be selected appropriate to pa-tient size; Against the proposition,” Med. Phys.28(11):2387-8, 2001.

5. Brenner DJ, Elliston CD, The potential impact of by-stander effects on radiation risks in a Mars mission, Ra-diat. Res. 156(5 Pt 2):612-7, 2001.

6. Brenner DJ, Little JB and Sachs RK, The bystandereffect in radiation oncogenesis: II. A quantitative model,Radiat Res. 155(3):402-8, 2001.

7. Brenner DJ, Okladnikova N, Hande P, Burak L,Geard CR and Azizova T, Biomarkers specific todensely-ionising (high LET) radiations, Radiat. Prot.Dosimetry 97(1):69-73, 2001.

8. Cabello G, Valenzuela M, Vilaxa A, Durán V, RudolphI, Hrepic N and Calaf G, A rat mammary tumor modelinduced by the organophosphorous pesticides, Parathionand Malathion, possibly through acetylcholinesteraseinhibition, Environmental Health & Perspectives109(5):471-479, 2001.

9. Calaf G and Hei TK, Oncoprotein expression in humanbreast epithelial cells transformed by high LET radia-tion, Int J Radiat Biol 77:31-40, 2001.

10. Calaf G, Russo J and Alvarado ME, Morphologicalphenotypes in neoplastic progression of benz(a)pyrene-treated breast epithelial cells, Journal of Submicro-scopic Cytology and Pathology 32(4):535-545, 2001.

11. Calaf G, Russo J, Tait L, Estrada S and Alvarado ME,Morphological phenotypes in neoplastic progression ofhuman breast epithelial cells, Journal of SubmicroscopyCytology and Pathology 32(1):83-96, 2000.

12. Chen MJ, Lin YT, Lieberman HB, Chen G and LeeEY, ATM-dependent phosphorylation of human Rad9 isrequired for ionizing radiation-induced checkpoint acti-vation, J. Biol. Chem. 276:16580-16586, 2001.

13. DiFagagna FD, Hande MP, Tong W-M, Roth D, Lans-dorp PM, Wang ZQ and Jackson SP, Effects of DNAnon-homologous end-joining factors on telomere lengthand chromosome stability in mammalian cells, CurrentBiology 11:1192-1196, 2001.

14. Elder RT, Song XQ, Chen M, Hopkins KM, Lieber-man HB and Zhao YL, Involvement of rhp23, a S.

pombe orthologue of the human HHR23A and S. cere-visiae RAD23 nucleotide excision repair genes, in cellcycle control and protein ubiquitination, Nucl Acids Res30(2):581-91, 2002.

15. Emenaker NJ, Calaf G, Cox D, Basson MD and Qure-shi N, Short chain fatty acids inhibits invasive humancolon cancer by modulating uPA, TIMP-1, TIMP-2,Mutant p53, Bcl-2, BAX, p21 and PCNA protein ex-pression in an in vitro cell culture model, The Journal ofNutrition 131(11 Suppl):3041S-6S, 2001.

16. Gage BM, Alroy D, Shin CY, Ponomareva ON, Dhar S,Sharma GG, Pandita TK, Thayer MJ and Turker MS,Spontaneously immortalized cell lines obtained fromadult Atm null mice retain sensitivity to ionizing radia-tion and exhibit a mutational pattern suggestive of oxi-dative stress, Oncogene 20(32):4291-7, 2001.

17. Gatei M, Shkedy D, Khanna KK, Uziel T, Shiloh Y,Pandita TK, Lavin MF and Rotman G, Ataxia-telangiectasia: chronic activation of damage-responsivefunctions is reduced by alpha-lipoic acid, Oncogene20(3):289-94, 2001.

18. Hall EJ, CT Scanning; risk versus benefit, invited edi-torial, J. Radiol. Prot. 20:347-348, 2000.

19. Hall EJ, Do no harm: Normal tissue effects, a reviewarticle, Acta. Oncologica. 40:913-916, 2001.

20. Hall EJ, Genomic instability, bystander effect, cyto-plasmic irradiation and other phenomena that mayachieve fame without fortune (Editors: Cirio R, Cuci-notta FA, Durante M), Prodeedings of the 1st Interna-tional Workshop on Space Radiation Research & 11thAnnual NASA Space Radiation Health Investigators’Workshop, Arona, Italy, May 2000. Physica. Medica.Vol. XV11: Supp. 1, 21-25, 2001.

21. Hall EJ, Radiation, the two-edged sword: Cancer risksat high and low doses. The Cancer Journal 6:343-350,2000.

22. Hande MP, Balajee AS, Tchirkov A, Wynshaw-Boris Aand Lansdorp PM, Extra-chromosomal telomeric DNA incells from Atm-/- mice and patients with ataxia-telangiectasia, Human Molecular Genetics 10:519-528(with cover picture), 2001.

23. Hande MP, Balajee AS, Techirkov A, Wynshaw-BorisA and Lansdorp P, Extra-chromosomal telomeric DNAin cells from Atm-/- mice and patients with Ataxia-telangiectasia, Hum. Mol. Genet. 10:519-528, 2001.

24. Hei TK, Zhao YL, Roy D, Piao CQ, Calaf G and HallEJ, Molecular alterations in tumorigenic human bron-chial and breast epithelial cells induced by high LET ra-diations, Advances in Space Research 27:411-419,2001.

25. Hei TK, Zhou HN, Wu LX, Randers-Pehrson G,Waldren C and Geard CR, Radiation induced geno-toxic damage: from cytoplasm to nucleus and the by-stander phenomenon, in: Free Radicals in Chemistry,Biology and Medicine (Yoshikawa T, Toyokuni S, Ya-mamoto Y and Naito Y ed.), OICA press, London, pp


96

241-247, 2000.26. Hsu HL, Gilley D, Galande SA, Hande MP, Allen B,

Kim SH, Li GC, Campisi J, Kohwi-Shigematsu T, ChenDJ, Ku acts in unique way at the mammalian telomere toprevent end joining, Genes and Development 14:2807-2812, 2000.

27. Karmakar P, Balajee AS and Natarajan AT, Analysis ofrepair and PCNA complex formation by ionizing radia-tion in human fibroblast cell lines, Mutagenesis 16:225-232, 2001.

28. Liu SX, Athar M, Lippai I, Waldren CA and Hei TK,Induction of oxyradicals by arsenic: implications formechanisms of genotoxicity, Proc. National AcademyScience (USA) 98:1643-1648, 2001.

29. Pandita TK, The role of ATM in telomere structure andfunction, Radiat Res. 156(5 Pt 2):642-7, 2001.

30. Piao CQ and Hei TK, Gene amplification and micro-satellite instability induced by tumorigenic human bron-chial epithelial cells by alpha particles and heavy ions,Radiation Research 155:263-267, 2001.

31. Piao CQ, Zhao YL and Hei TK, Analysis of p16 andp21 expression in tumorigenic human bronchial cellsinduced by asbestos, Oncogene 20:7301-7306, 2001.

32. Ponomarev AL, Cucinotta FA, Sachs RK and BrennerDJ, Monte Carlo predictions of DNA fragment-sizeddistributions for large sizes after HZE particle irradia-tion, Phys. Med. 17 Suppl 1:153-6, 2001.

33. Ponomarev AL, Cucinotta FA, Sachs RK, Brenner DJ,Peterson LE, Extrapolation of the DNA fragment-sizedistribution after high-dose irradiation to predict effectsat low doses, Radiat. Res. 156(5 Pt 2):594-7, 2001.

34. Proietti De Santis L, Garcia CL, Balajee AS, BreaCalvo GT, Bassi L and Palitti F, Transcription coupledrepair deficiency results in increased chromosomal aber-rations and apoptotic death in the UV61 cell line, theChinese hamster homologue of Cockayne’s syndromeB, Mutat. Res./DNA Repair 485:121-132, 2001.

35. Randers-Pehrson G, Geard CR, Johnson G, EllistonCD and Brenner DJ, The Columbia University single-ion microbeam, Radiat. Res. 156(2):210-4, 2001.

36. Roy D, Calaf G and Hei TK, Frequent allelic imbal-ance on chromosome 6 and 17 correlate with radiation-induced neoplastic transformation of human breastepithelial cells, Carcinogenesis 22(11):101-108, 2001.

37. Roy D, Calaf G and Hei TK, Profiling of differentiallyexpressed genes induced by high-LET radiation inbreast epithelial cells, Molecular Carcinogenesis31:192-203, 2001.

38. Sawant SG, Randers-Pehrson G, Geard CR, Bren-ner DJ and Hall EJ, The bystander effect in radiationoncogenesis: I. Transformation in C3H 10T1/2 cells invitro can be initiated in the unirradiated neighbors of ir-radiated cells, Radiat. Res. 155:397-401, 2001.

39. Sawant SG, Randers-Pehrson G, Metting NF andHall EJ, Adaptive response and the Bystander Effectinduced by radiation in C3H 10T1/2 cells in culture,Radiat. Res. 156:177-180, 2001.

40. Schmeissner PJ, Xie H, Smilenov LB, Shu F, Marcan-tonio EE, Integrin functions play a key role in the dif-ferentiation of thymocytes in vivo, J. Immunol.167(7):3715-3724, 2001.

41. Smilenov LB, Brenner DJ and Hall EJ, Modest in-creased sensitivity to radiation oncogenesis in ATMheterozygous versus wild-type mammalian cells, Can-cer Res. 61(15):5710-5713, 2001.

42. Suzuki M, Piao CQ, Hall EJ and Hei, TK, Cell killingand chromatid damage in primary human bronchialepithelial cells irradiated with accelerated Fe ions, Ra-diation Research 155:432-439, 2001.

43. Suzuki M, Piao CQ, Zhao YL and Hei TK. Karyotypeanalysis of tumorigenic human bronchial epithelial cellstransformed by chrysotile asbestos using chemically in-duced PCC technique. International Journal of Mo-lecular Medicine 8:43-47, 2001.

44. Tong W-M, Hande MP, Lansdorp PM and Wang Z-Q,DNA strand break-sensing molecule PARP cooperateswith p53 in telomere function, chromosome stability andtumor suppression, Molecular and Cellular Biology21:4046-4054, 2001.

45. Vaziri H, Dessain SK, Ng Eaton E, Imai SI, Frye RA,Pandita TK, Guarente L and Weinberg RA,hSIR2(SIRT1) functions as an NAD-dependent p53deacetylase, Cell Oct 19;107(2):149-59, 2001.

46. Wood LD, Halvorsen TL, Dhar S, Baur JA, PanditaRK, Wright WE, Hande MP, Calaf G, Hei TK, LevineF, Shay JW, Wang JJY and Pandita TK, Characteriza-tion of ataxia telangiectasia fibroblasts with extendedlife-span through telomerase expression, Oncogene20:278-288, 2001.

47. Zhao YL, Piao CQ, Hall EJ and Hei TK, Mechanismof radiation induced transformation of human bronchialepithelial cells, Radiation Research 155:230-234, 2001.

48. Zhou HZ, Suzuki M, Randers-Pehrson G, Vannais D,Trosko JE, Waldren CA and Hei TK, Radiation risk atlow doses may be greater than we thought, Proc. Na-tional Academy Science (USA) 98:3857, 2001. n

PUBLICATIONS

97

Human Molecular Genetics 10: 519-528 2001Cover picture by Hande MP, Balajee AS, Tchirkov A, Wynshaw-Boris A and Lansdorp PM

Extra-Chromosomal Telomeric DNA in Cells from Atm-/- Mice and Patients with Ataxia-Telangiectasia

c radiological r a r table of contentsyong liang zhao, chang piao and tom k. hei..... 69 mutation(s)...

Documents