MILITARY MEDiCINE, 170, 12:1053, 2005

Rapid Identification of Dengue Virus by ReverseTranscription-Polymerase Chain Reaction Using

Field-Deployable Instrumentation

Guarantor: James C. McAvin. MSContributors: James C. McAvin. MS*; Elizabeth M. Escamilla, MS*: Maj Jamie A. Blow. USAF+:Michael J. Turell. PhD+: Miguel Quintana. PhD^: Maj David E. Bowles, USAF§: Col James A, Swaby. USAF§:Maj William J. Barnes, USAF*: Col William B. HufT. USAP; Kenton L. Lohman. PhD*: Maj Daniel H. Atchley, USAF*;Lt Col John R. Hickman, USAF*; Lt Col Debra M. Niemeyer.

Dengue virus universal and dengue serotype 1 to 4. fiuoro-genic probe hydrolysis (TaqMan), reverse transcription-poly-merase chain reaction assays were developed for screeningand serotype identification of infected mosquito vectors andhuman sera using a field-deployable, fluorometric thermocy-cler. Dengue universal and dengue 1 to 4 serotype assay invitro sensitivity and specificity results were 100% concordantwhen tested with total nucleic acid extracts of multiple strainsof dengue serotype 1 to 4, yellow fever, Japanese encephalitis.West Nile, and St. Louis encephalitis viruses. The in vitrosensitivity and specificity results for all five assays were con-cordant when tested with a blind panel of 27 dengue virus-infected mosquitoes, 21 non-dengue (yellow fever. West Nile,or St. Louis encephalitis) flavivirus-infected mosquitoes, and11 uninfected mosquitoes and with clinical specimens con-sisting of a human serum panel of eight dengue viremic and 31non-dengue-infeeted febrile patient serum samples. No cross-reaction occurred with vector species or human genomic DNA.Sample processing and polymerase chain reaction required<2 hours.

DIntroduction

engue fever (DF) and the more severe forms of the disease,i.e., dengue hemorrhagic fever (DHF) and dengue shock

'Epidetniological Sttrveillance Dmsion, U.S. Air Forre Institute for Environment,Safety, and Occupational Health Analysis. Brooks Air Force Base (now designated AirForce Institute for Operational Health. Brooks City-Base). San Antonio. TX 78235-5237.

^Virology Division, U.S. Army Medical Research Institute for Infectious Disease,Fort Detrick, Frederick. MD 21702 5011.

iEnvironmental Science Division. U.S. Army Center for Health Promotion andPrevenlalive Medicine-West, Fort Lewis, WA 98433-9500.

§U.S. Air Force Force Protection Battlelab, Lackland Air Force Base, San Antonio,TX 78236-0119.

|U.S. Air Force Office of the Surgeon General, Boiling Air Force Base, Washington.DC. 20332-0103.

IJoint Program Executive OlTtce for Chemical and Biological Defense. FallsChurch, VA 22041-3203.

This manuscript was cleared through the Technical Publication/Presentation Con-trol Record, Brooks City-Base, TX for open publication, January 2004. The conclu-sions and opinions expressed in this document are those of the authors. They do notreflect the ofTiciai position of the U.S. government. Department of Defense, JointProgram OfTtce for Chetnical and Biological Defense, U.S. Air Force, U.S. Army, or theAssociation of Militar>' Stirgeons cf the United States.

Mention of a trade name, vendor, proprietary product, or specific equipment is notan endorsement, a guarantee, or a warranty by the Department of Defense, Army, orAir Force and does not imply an approval to the exclusion of other products or vendorsIhat also may be suitable.

This mantiscript was received for review iti Januar\' 2004. The revised manuscriptwas accepted for publication in December 2004,

syndrome, occur in tropical and subtropical regions globallythrough infection by one or more of four viral serotypes (dengueserotypes 1 - 4).' Dengue virus is a member of the genus Fiavivi-rus. family FlaviWridac, that is transmitted in a cycle that pri-marily involves humans and mosquito vectors, most signifi-cantly Aedes aegypti in developing urban regions and A. aegyptias well as Aedes albopictus in semiurban areas and ruralregions,'"''The prevalence of dengue virus is now comparable tothat of malaria, making DF the most significant mosquito-bomewal disease, threatening two-iifths of the world's human pop-ulation.*̂ '̂ It is estimated that 50 to 100 million people areaffected annually with DF and 300,000 with DHF/dengue shocksyndrome.''

Dengue ylrus surveillance and DF/DHF diagnoses areproblematic. Symptoms are usually nonspecific, and serolog-ical analyses or virus isolation can take 1 week or more.''^'^Antibody cross-reaction occurs across the Flaviviridae family,creating ambiguity.'°" To augment clinical laboratory diag-nostic capability, dengue virus reverse transcription (RT)-polymerase chain reaction (PCR) assays have been developedfor use with laboratory-based instrumentation.'^"'" In areas inwhich clinicai laboratory diagnostics are not available, field sur-veillance is an essential element in achieving timely assess-ments of transmission risk and time-critical implementation ofappropriate tnosquito control measures and clinical responsesin a potential outbreak situation. Rapid identification of circu-lating virus serotypes and causative serotypes in prexious out-breaks is fundamental to risk assessment for DHF.

Given the expansion of expeditionary operations worldwide.U,S. Armed Forces are at risk for acquiring a variety of infec-tions, including dengue. After World War II. a dengue pan-demic in Southeast Asia marked expanded emergence of thevirus, with epidemics caused by multiple serotypes. '̂ With nolicensed vaccine available. DF and other febrile diseases re-main a threat to U.S. military operations, as obsen'edthroughout the history of conflict'^ and most recently duringOperation Restore Hope.̂ " This threat drives the need forimproved far-forward surveillance and rapid diagnosis of fe-brile illnesses. In the past several years, PCR has been shownto be a military field-worthy technique for rapid identificationof biological agents. '̂"^^ In this work, we describe a PCR-based assay system for rapid, sensitive, specific screening ofdengue virus and serotype identification in mosquito vectorsand human sera, using deployable instrumentation.

1053 Military Medicine. Vol. 170. December 2005

Report Documentation Page Form ApprovedOMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.

1. REPORT DATE 1 DEC 2005

2. REPORT TYPE N/A

3. DATES COVERED -

4. TITLE AND SUBTITLE Rapid identification of dengue virus by reverse transcription-polymerasechain reaction using field-deployable instrumentation, Military Medicine170:1053 - 1059

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) McAvin, JC Escamilla, EM Blow, JA Turell, MJ Quintana, M Bowles,DE Swaby, JA Barnes, WJ Huff, WB Lohman, KL Atchley, DHHickman, JR Niemeyer, DM

5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) United States Army Medical Research Institute of Infectious Diseases,Fort Detrick, MD

8. PERFORMING ORGANIZATIONREPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release, distribution unlimited

13. SUPPLEMENTARY NOTES

14. ABSTRACT Dengue virus universal and dengue seroype 1 to 4, fluorogenick probe hydrolysis (Taqman) reversetranscription-polymerase chain reaction assays were developed for screening and serotype identification ofinfected mosquito vectors and human sera using a field-deployable, fluorometric thermocycler. Dengueuniversal and dengue 1 to 4 serotype assay in vitro sensitivity and specificity results were 100% concordantwhen tested with nucleic acid extracts of multiple strains of dengue serotype 1 to 4, yellow fever, Japaneseencephalitis, West Nile, and St. Louis encephalitis viruses. The in vitro sensitivity and specificity results forall five assays were concordant when tested with a blind panel of 27 dengue virus-infected mosquitoes, 21non-dengue (yellow fever, West Nile, or St. Louis encephalitis) flavivirus-infected mosquitoes and 11uninfected mosquitoes and with clinical specimens consisting of a human serum panel of eight dengueviremic and 31 non-dengue-infected febrile patient serum samples. No cross-reaction occurred with vectorspecies or human genomic DNA. Sample processing and polymerase chain reaction required < 2 hours.

15. SUBJECT TERMS dengue, rapid identification, PCR, reverse trancription, field deployable

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT

SAR

18. NUMBEROF PAGES

8

19a. NAME OFRESPONSIBLE PERSON

a. REPORT unclassified

b. ABSTRACT unclassified

c. THIS PAGE unclassified

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

1054 Rapid identiUcation of Dengue Virus by RTPCR

Methods

Primer and Probe Design

Given that an established. TaqMan-based, [Synthetic Genet-ics. Rock\ille. Maryland), RT-PCR. universal dengue virus assaywas not obtainable from the literature, primer and probe se-quences were selected by aligning homologous genomic regionsof serotypes 1 to 4 that excluded other clinically significantflaviviruses. Alignments were compared visually using theClustal algorithm^-' in the MegAlign program of DNA Star soft-ware (Perkin Elmer. Norwtilk. Connecticut).^^ Maximally con-served oligonucleotide sequences were chosen from dengue vi-rus type 1 to 4 genomes downloaded from GenBank (accessionnumbers U88536, M19197, M93130, and AF326825, respec-tively). Yellow fever. Japanese encephalitis. West Nile, and St.Louis encephalitis virus-type strain genomic sequences wereaiigned and \1sually evaluated to validate heterology wilh uni-versal primer and probe sequences (GenBank accession num-bers X03700/K02749, M18370, M12294/M10103. andAF242895, respectively). Universal primer and probe sequencescan be requested through the corresponding author. Serotype-specific primer and probe sequences were obtained from theliterature (DEN-l,'^ DV-2,'3 DEN-3,'2 and DEN-4'2|. and heter-ology was validated as described above. All probes reported hereare dual fluorogenic labels designed with a 5' reporter dye.6-carboxytluorescein. and a 3' quencher dye, 6-carboxytetram-ethylrhodamine,''̂ ''

Universal and serotype-specific primers and probes sequenceheterology with genomic sequences of closely related speciesthrough diverse genera was validated by BLAST databasesearch (BLAST, Madison, Wisconsin). '̂ Melting temperatureswere quantified and the absence of significant primer dimeriza-tions and secondary structure (hairpin) formations was con-firmed with PrimerExpress software (PE Applied Biosystems.Foster City. California). Primers and probes were synthesizedand quality control was conducted commercially (Synthetic Ge-netics, San Diego, California). Assays were designated DU-JCM.DEN-1, DEN-2, DEN-3, and DEN-4, respectively.

Flavivirus Reference Strains

To optimize RT-PCR assay sensitivity and specificity, nucleicacid extracts of reference strains of dengue virus serotypes andother flaviviruses were obtained from the Southwest Foundationfor Biomedical Research (San Antonio, Texas) and the Center forTropical Diseases. University of Texas Health Science Center(Galveston. Texas). Species and strain nucleic acid extracts in-cluded dengue 1 (Hawaii. Mochizuki. and 106), dengue 2(SI6803, NGC, VEN2, 131. Mara-3. Mex44. ig2913. and102954). dengue 3 (CH53489. H87. and 53402). dengue 4(H241. Dakar NC599. and VEN4). yellow fever (Asibe. TVP834,and TVP8249). Japanese encephalitis (JKT9092, PhAnlO91,and JKT8442). West Nile vims (Crow 397-99. TVP3545, andTVP8345). and St, Louis encephalitis (Fort Washington,TVP8314, TVP8344. and LA2001).

Assay Optimization

Relative sensitivity levels of each assay were optimized withsix logarithmic dilutions of each dengue serotype. prepared bydiluting nucieic acid extracts 1:10 in RT-PCR-grade water with 2

U of RNase inhibitor. The specificity of each assay was testedwith undiluted samples and four logarithmic dilutions of eachheterologous dengue serotype and other flaviviruses, preparedby dilution as described above.

Mosquito Panels

To validate RT-PCR assay in vitro sensitivity and specificity,mosquito panels were prepared at the Virology Division. U.S.Army Medical Research Institute for Infectious Disease (USAM-RIID") (Fort Detrick, Maryland), One- to 6-day-old. adult, female,A. aegypti mosquitoes were inoculated intrathoracically^" withone of the following viruses: dengue 1 (Hawaii), dengue 2(S16803). dengue 3 (CH53489). dengue 4 (H241). yellow fever(Asibe). West Nile (Crow 397-99), or St, Louis encephalitis (FortWashington). Mosquitoes were held in cardboard cages, pro-vided with a carbohydrate source (either apple slices or a gauzepad soaked in a 10% sucrose solution) and a water-soakedcotton pledget, and held at 26T for 7 days. Mosquitoes werethen killed by exposure to -20°C for 5 to 10 minutes, and one tosix legs were removed for analyses. Legs were triturated ingrinding diluent (10% heat-inactivated fetal bovine serum inmedium 199 with Earle's salts, NaHCO .̂ and antibiotics) andtested for the presence of virus by plaque assay on LLC-MK-2cells. One virus-inoculated or one unexposed mosquito bodywas added to each pool of uninfected mosquitoes. These wereplaced in sterile, 1.5-mL. Eppendorf tubes and triturated in 750/xLofTRIzol-LS (Life Technologies, Rock\111e. Maryland), A panelwas established, labeled under code at USAMRIIE), and shippedon dry ice to the Epidemiological Surveillance Diiision. U.S. AirForce Institute for Environment. Satety. and OccupationalHealth Analysis. Brooks Air Force Base (San Antonio. Texas), fornucleic acid extraction and blind RT-PCR analyses (Table I),

Clinical Specimens

Human blood specimens were collected during a DF epidemicin Tegucigalpa, Honduras, in February 2002. by the U.S. ArmyCenter for Health Promotion and Preventative Medicine-West incollaboration with the Houduran Ministry of Health (HMOH).from a total of 39 febrile and convalescent patients 0 to 5 daysafter the onset of fever and after defervescence, respectively. TheHMOH confirmed the presence of dengue virus by serologicaltesting (Table II). and tbe specimens were archived at -20°C.Sera were subsequently thawed, and 500-;U.L aliquots of eachspecimen were prepared and labeled under code by the HMOHand shipped on dry ice to the Air Force Institute for Environ-ment, Safety, and Occupational Health Analysis, Serum sam-ples were stored at -20°C for 2 weeks, thawed, maintained at4'̂ C for 72 hours during nucleic acid extractions and blindvalidation testing with RT-PCR assays, and archived at -20°C.Before RT-PCR analyses, each serum sample had undergone atleast two freeze-thaw cycles. The duration of exposure of thespecimens to ambient temperatures throughout collection, pro-cessing, serological analyses, and preparation for shipping isunknown.

RNA Preparation

Individual mosquitoes and mosquito pools were homogenizedin 500 ,uL and 1.000 /xL of TRIzol (Life Technologies), respec-tively, with sterile. RNase-free pestles and 1,5-mL tubes, ho-

Militarj' Medicine. Vol. 170, December 2005

Rapid Identification of Dengue Virus by RT-PCR 1055

TABLE I

RT-PCR RESULTS OF DENGUE VIRUS ASSAY TESTING OF INFECTED MOSgUITOES

Panel

101104105106109

no112113114117118120121123124125127128401402403404405406407408409410411412413414415416417418419420201202203204205206207208209210211212213214215216217218219220221222223224225226Til228

Virus

Dengue 1Dengue 4NoneDengue 1Dengue 4NoneDengue 4Dengue 1NoneDengue 4NoneDengue 1NoneDengue 4Dengue 1NoneDengue 4Dengue 1Dengue 3ADengue 2Dengue 3BNoneDengue 2Dengue 3ADengue 2Dengue 3BNoneNoneDengue 3ADengue 3BDengue 2NoneDengue 2Dengue 3BDengue 3ANoneDengue 3BDengue 3AYellow feverSt, Louis encephalitisWest NileDlluantYeilow feverSt, Louis eneephalitisWest NileDiluantWest NileDlluantYellow feverSt, Louis encephalitisWest NileSt, Louis encephalitisWest NileDlluantYellow feverSt. Louis encephalitisYellow feverDiluantWest NileSt. Louis encephalitisYellow feverDiluantWest NileYellow feverSt. Louis encephalitisDlluant

Strain

HawaiiH241

HawaiiH241

H241HawaU

H241

Hawaii

H241Hawaii

H241HawaiiH87SI 6803CH53489

S16803H87SI6803CH53489

H87CH 53489SI 6803

SI 6803CH 53489H87

CH53489H87AsibeFort WashingtonCrow 397-99

AsibeFort WashingtonCrow 397-99

Crow 397-99

AsibeFort WashingtonCrow 397-99Fort WashingtonCrow 397-99

AsibeFort WashingtonAsibe

Crow 397-99Fort WashingtonAsibe

Crow 397-99AsibeFort Washington

Plaque-FormingUnits per Leg

33,61 uninfected3.23.61 uninfeeted32.91 uninfected2.61 uninfected31 uninfected3.331 uninfected3.1323.93.61 uninfected4.23,63,641 uninfected1 uninfected2,844,31 uninfeeted4,33,92,21 uninfected3345

>503,95

>50

>4,503,7

>4,0>4,5>4,0>4,5

04

>4,03,10

>4.5>4,0

40

>4.54

>4.00

DU-JCM

PositivePositive

—PositivePositivePositivePositivePositivePositivePositive

—Positive

PositivePositive

—PositivePositivePositivePositivePositive

—PositivePositivePositivePositive

—.—

PositivePositivePositive

—PositivePositivePositive

—PositivePositive

——

—

—

—

—

—

——

DEN-1 DEN-2

Positive —.——

Positive ——— —— —

Positive —Positive —

——

Positive —

—Positive —

——

Positive —— __— Positive— —— —— Positive— —^ Positive— —— —

— Positive—— Positive—,—— ———

,— ——

— —

—

—

™ _

— —

_ _

— —

_ _

— —

DEN-3

——

Positive

Positive

Positive

Positive

PositivePositivePositive

.

PositivePositive

—PositivePositive

—

. ,—

_

_—

-_—

—_

___

—

DEN-4

-_Positive

Positive

Positive

Positive

Positive

Positive

.

_

__—

__

_

_

_

-

Military Medicine. Vol, 170, December 2005

1056 Rapid Identification of Dengue Virus by RT-PCR

TABLE II

RT-PCR RESULTS OP DENGUE VIRUS ASSAY TESTING OF HUMAN SERUM

No.

123456789

101112131415161718192021222324252627282930313233343536373839

Sample

02-293502-293802-294502-294802-294302-296402-296502-295802-296302-296802-293402-295502-295702 296002-293202-293102-293302-294202-292102-288802-288202-287802-297202-290902-297802-211702-211502-172002-223802-296902-214402-211402-210102-212502-184602-287402-211602-287702-2869

Colleetion Date

7ni2om7/7/20025/7/20027/7/20026/7/20023/7/20024/7/20027/7/20023/7/20023/7/20026/7/20027/7/20026/7/20023/7/20026/7/20027/7/20025/7/20027/7/20025/7/20023/7/20025/7/20024/7/20023/7/20022/7/200208/07/0227/06/0227/06/0223/06/02i/7/20023/7/200230/06/0229/06/0226/06/021/7/200225/06/024/7/200227/06/024/7/20024/7/2002

Serological Test

—————

——

Dengue 2——

Dengue 2Dengue 2

———

_—

Dengue 2—————

Dengue 2—

Dengue 2—

Dengue 2—

Dengue 2——————

DU-JCM

—_—

—

——

Positive——

PositivePositive

———————

Positive—————

Positive—

Positive—

Positive—

Positive——————

DEN-1

—

——————————

———————————————————————————

DEN-2

———

————

Positive——

PositivePositive

———————

Positive—————

Positive—

Positive—

Positive————————

DEN-3 DEN-4

— —— —— —— —_ _— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —— —

All patient samples were collected during the acute febrile phase.

mogenates were cleared by centrifugation in a tabletop centri-fuge at 12,000 rpm for 60 seconds, and total nucleic acidextracts were prepared according to the manufacturer's instruc-tions. Each extract was suspended in 100 p.L of RNase-free. PCRgrade water with 2 U of RNase inhibitor.

Human serum specimens were thawed at 4°C, and 140 pL ofeach specimen were suspended in 560 fiL of the buffer AVL/carrier RNA component of a preformatted, total nucleic acidextraction kit (QlAamp viral RNA mini kit. Qiagen. Valencia,California). Extracts were prepared and suspended in 60 pL ofelution buffer with the manufacturer's centrifugation protocol.

Reaction Conditions

Preliminary' assay optimizations and cross-reaction testingwere conducted with a laboratory-based, real-time thermocy-cler. the LightCycler (Idaho Technology. Salt Lake City.

with Preformatted reagents [Titan one-tube RT-PCR kit. RocheMolecular Biochemicals, Indianapolis. Indiana). Assays werereoptimized for the Ruggedized Advanced Pathogen Identifica-tion Device (R,A,P,I.D.) (Idaho Technology) with a proprietarybuffer system (Idaho Technology), and sensitivity and specificityvalidation testing was completed.

Master mixture reaction solution was prepared. 18'/,tL vol-umes were dispensed into optical capillary tubes, and 2 ^iL ofRNA extract from specimens and control samples were added (or2 ,uL of PCR-grade water for no-template control samples). Cap-illaries were placed in a tabletop centrifuge and centrifuged for10 seconds at 3,000 rpm, to drive the assay mixture to thebottom of the tube. Master mixture components were 0,20 mMconcentrations of dATP, dTTP. dGTP. and dCTP (Idaho Technol-ogy). 2,5 mM Mn(AOc) (Roche Molecular Biochemicals), 2,5 U ofTth polymerase (Roche Molecular Biochemicals). and 20% (v/v)

Militarv Medicine, Vol. 170. December 2005

Rapid Identification of Dengue Virus by RT-PCR 1057

proprietary 5x RT buffer [Idaho Technology). The dengue virusuniversal assay (DU-JCM) forward primer concentration was0.50 /AM. reverse primer 0.50 fM. and TaqMan probe 0.50 ^M.DEN-1. DEN-2, DEN-3, and DEN-4 had the same primer andprobe concentrations; the forward primer coneentration was0.50 ixM. reverse primer 0.90 ^M. and TaqMan probe 0.10 jiM.

A standardized RT-PCR thermal cycling protocol was estab-lished, consisting of RT at 40T for 30 minutes followed by initialcDNA denaturation al 94''C for 2 minutes and PCR for 45 cyclesof 94''C for 0 seconds for template denaturation and 60^C for 20seconds for combined annealing and primer extension. A singledata point at the end of each annealing-extension cycle wascollected and reported as TaqMan probe fluorescence releasedby 5'-nuclease activit>' during primer extension. Fluorimetergains were set at 8, 2, and 2 for channels 1. 2. and 3, respec-tively. Protocols for all five assays were identical with the excep-tion of DEN-3, for which the extension temperature was 66X.The criterion for a positive result was a significant increase influorescence over background levels, as defined by an algorithmprovided in the R.A.P.I.D analytical software [Roche MolecularBiochemicals).

Results

Assay Sensitivity and Specificity Testing with FlavivinisReference Strains

Preliminary assay sensitivity and specificity evaluations wereconducted with a knoun panel of total nucleic acid extracts fromdengue serotype 1 - to 4-infected A. aegypti inoculated with threedifferent strains of dengue 1. eight strains of dengue 2. threestrains of dengue 3. and three strains of dengue 4 and a cross-reactivity test panel consisting of totai nucleic acid extracts frommultiple strains of other Flaviviridae [three strains each of yel-low fever. Japanese encephalitis, and West Nile and four strainsof St. Louis encephalitis). In \itro sensitivity and specificity ofeach assay were 100% concordant. No cross-reactivity was ob-served with vector species [data not shown).

Assay Sensitivity and Specificity Testing with MosquitoPanels

Testing of assay in vitro sensitivity and specificity was accom-plished with a blind panel of 27 dengue-infected mosquitoes [sixdengue 1-infected mosquitoes, five dengue 2. 10 dengue 3. andsix dengue 4). 21 non-dengue-infected mosquitoes [seven yellowfever virus-infected A. aegypti and seven each of West Nile andSt. Ix>uis encephalitis virus-infected Cufex spp.). and 11 unin-fected mosquitoes, with 27 positive control samples and 32negative control samples [Table I). Diluent samples were notincluded in statistical analyses. In the case of serotype-specificassay analyses, heterologous serotypes were considered as neg-ative specimens. Assay results were as follows: DU-JCM, sensi-tivity of 100% [27 of 27 samples) and specificity of 94% [30 of 32samples): DEN-1. sensiti\ity of 100% [six of six samples) andspecificity of 98% [52 of 53 samples): DEN-2. sensitivit\' of 100%[five of five samples) and specificity of 100% (54 of 54 samples);DEN-3. sensitivity of 100% [10 of 10 samples) and specificity of98% [48 of 49 samples); DEN-4. sensitivity of 100% [six of sixsamples) and specificity of 100% (53 of 53 samples). The DU-JCM assay reported two false-positive results [panels 110 and

114). Dengue 1 and 3 assays reported one false-positive resulteach (panels 114 and 413. respectively) (Table I). The findingthat both DU-JCM and DEN-1 assays reported panel 114 asdengue virus positive indicates experimental error. Because oneoccurrence of a false-positive result was observed when theDEN-3 assay was tested against dengue 2 [panel 413). addi-tional testing will be performed to further delineate observedresults. No cross-reaction was observed with vector speciesgenomie DNA and medium diluent. Sample processing and RT-PCR required <2 hours.

Assay Sensitivity and Specificity Testing with ClinicalSpecimens

Testing was aceomplished with a blind panel of eight dengueviremic [dengue 2) and 31 non-dengue-infected febrile patientserum specimens [Table II). Dengue virus universal assay invitro sensitivity was 100% [eight of eight samples) and specificitywas 100% (31 of 31 samples). DEN-2 assay in vitro sensitivitywas 88% (seven of eight samples) when tested against the hu-man serum panel, with one false-negative result, and specificitywas 100% [31 of 31 samples) when tested against non-dengue-infected specimens. Dengue 1, 3. and 4 assays displayed nocross-reactivity with dengue 2 serotype. Nucleic acid integrity ofthe human serum samples presented an unknown variable be-cause of a lack of history regarding collection conditions, trans-portation and storage temperatures, and number of freeze-thawcycles. Human genomic DNA displayed no detectable fiuores-cence above background levels. Sample processing and real-time RT-PCR required <2 hours.

Discussion

We have described a rapid, sensitive, specific dengue virusRT-PCR assay system for screening and serotype identificationof infected mosquito vectors and human sera. The augmenta-tion of traditional laboratory-based testing with field-deployableRT-FCR can enhance dengue virus surveillance capability andprovides a promising aid in rapid clinical laboratory diagnosticsin fixed-site and deployed medical locales.

Development of pathogen identification assays with field-du-rable instrumentation fulfills a fundamental requirement in therealization of a complete field-deployable assay platform. Al-though the R,A.P.1.D provides field-deployable. real-time PCRinstrumentation, an omnipotent, field-deployable. assay plat-form requires additional components, including thermostableassay reagents and control nucleic acid, and field-formattedsample preservation and nucleic acid isolation technologies.Therefore, in 2001, the Department of Defense established theJoint Biological Agent Identification and Diagnostic System pro-gram to develop a complete field-ready system for rapid medicalsurveillance and diagnostic testing at fixed-site and deployedlocations.'"'The system and components (protocols and assays)will be cleared by the Food and Drug Administration for patientcare use. Our experiences described below emphasize the needfor a standardized system approach.

Sun'eillance studies conducted under austere conditionspresent logistical and operational constraints that make it im-practical to transport, store, and prepare PCR reagents andcontrol nucleic acid. Where resources for cold chain mainte-

Miiitary Medicine, Vol. 170. December 2005

1058 Rapid Identification of Dengue Virus by RT-PCR

nance are unreliable or nonexistent, unknown variables in en-zymatic activit}' and control nucleic acid integrity result. More-over, sample integrity comes into question when samplecollections are made distant to the laboratory or support facili-ties, Exposure to ambient temperature and freeze-thaw occur-rences during transportation often present unknown variablesin nucleic acid integrity of the sample. Also, nucleic acid extrae-tlon and preservation reagents designated as hazardous mate-rials present logistical constraints because of transportationand shipping restrictions,

To meet logistical and operational constraints, we have fo-cused most intensely on the most vulnerable components of thefield-deployable assay platform, namely, PCR enzymes and con-trol nucleic acid template. The inherent fragilit\' of RNA makesmaintaining RT-PCR target template integrity an especiallydaunting task. To ensure the validit}' of field-formatted assays,we are integrating thermostable, hydrolytic enzyme-shielded.Armored RNA control template (Ambion RNA Diagnostics, Aus-tin. Texas) and lyophilized master mixture reagents (Idaho Tech-nology) into our research protocols. These products proWdemaster mixture reagents and positive control samples that areeasily transported, are field-sustainable, and require only hy-dration and addition of sample template for analyses. In addi-tion to the value of Armored RNA in meeting logistical andoperational constraints, there is also promise for development ofquantitative \1ral assays." Our preliminary laboratory evalua-tion of commercial, pre formatted, thermostable, nucleic acidisolation reagents has shown promise for improved field-col-lected sample processing as well as sample stabilization, partic-ularly for RNA-based viruses, Therefore, the lessons learnedfrom this study and subsequent studies, along with probe in-formation and protocols, wili be provided to the Joint ProgramExecutive Office for Chemical and Biological Defense, as stipu-lated in the Critical Reagents Program multilaboratory agree-ment. Under this agreement, information and materials pro-vided by government laboratories are evaluated for use indevelopment of assays, protocols, and procedures for joint ef-forts, including the Joint Biological Agent Identification andDiagnostic System program (unpublished data).

In summaiy. this article illustrates the potential operationalutility of using dengue virus assays for rapid, sensitive, andspecific screening and serotype identification in mosquito vec-tors and human sera, on field-deployable instrumentation. Thissystem also shows promise for development into a fully deploy-able package for epidemiological sur\^eiilance of dengue \1rusand a field-expedient method to augment traditional clinicallaboratory diagnostic techniques.

Acknowledgments

Thanks go to Dr. Philip Armstrong. Southwest Foundation for Biomed-ical Research (San Antonio. Texas), and Dr. Robert Tesh. Center forTropical Diseases. tJiiiversity oi Texas Health Science Center (Galveston.Texas), lor providing flavivirus nucleic acid extracts.

References

I. Gubler tXI: Dengue and dengue hetnorrhagif fever. Cliii Microbio) Rev 1998; 11:480-96.

2. Black WC, Bennett KE, Gorrochotegui-Estaianie N. et al: Fla\iv1nis susceptibilityin Aedps aegypti Arch Med Res 2002: 33: 379-88.

3. Kow CY. Kooii LL, Yin PF: Detection of dengue vinises tn field caught male Aedesaegypti and Aedes atbopictus (Diptera: Culicidae] in Singapore by (ype-speclflcPCR. J Med Entomol 200i: 38: 475-9,

4. Rosen L: Dengue hemorrha(;ir fever. Bull Soc Palhol Exoi i996: 89: 91-4,5. Tu WC, Chen CC. Hou RF: Ultraslruftural studies on ihe reproductive systeni of

male Aedes aeyypli (Diptera: CulicLdae) Infected with dengue 2 \irus. J MedEntomol 1998: 35: 71-6,

6. Centers for Disease Control and Prevention: Dengue fever. Available at ht tp: / /www.cdc.gov/ncldod/dvbld/dengue/index, htm: accessed November 4. 2003.

7. World Health Organization: Dengue and dengue haemorrhaglc fever: faet sheet117. Available ai http://www.who.int/mediacentre/factsheets/fsl l7/en/index.html: accessed November 4. 2003.

8. Gibbons RV. Vaughn DW: Dengue: an escalating problem. BMJ 2002: 324:1563-6,

9. Hayes EB. Gubler DJ: Dengue and dengue hemorrhagic fever. Pediair Infect DisJ 1992: II: 311-7.

10. Schwartz E, Milegulr F. Grossman Z. Mendelson E: Evaluation of ELISA-hasedsero-diagnosis of dengue fever in travelers. J Clin Virol 2000: 19: 169-73,

11. Koraka P, Zellar H, Niedrig M. Osterhaus AD, Groen J: Reactiiity of serumsampies from patients wilh a flavivirus infection measured by immunolluores-cence assay and ELISA. Microbes Infect 2002; 4: 1209-15.

12. Callahan JD. Wu SJ, Dion-Schultz A, et al: Development and evaluation ofserotype- and group-specific iluorogeiiic reverse-transcriptase PCR (TaqMan)assays for dengue virus. J CUn Microhiol 2001: 39: 4119-24.

Ki. Houng HH, Hrit7 D. Kanesa-thasan N: Quantitative detection of dengue 2 virususing iluorogenic RT-IK^R based on 3 -noncoding sequence, J Virol Methods2000:86: 1-11.

14. HoungHH, Chung-Ming Chen R.Vaughn DW. Kanesa-thasanN: Development ofa iluorogenic RT-PCR system for quantitative identiiication of dengue virus se-rotypes i-4 using conserved and serotype-specilic3' noncoding sequences. J Vi-rul Methods 2001; 95: 19-32.

15. LaueT, Emmerich P. Schmttz H: Detection ofdengue\1rus RNA in patients afterprimary or secondary dengue infection by using the TaqMan automated ampli-fication syslem. J Clin Microblol 1999: 37: 2.543-7.

16. Warrilow D. Northill JA, Pyke A. Smith GA: Single rapid TaqMan fluorogenicprobe based PCR assay that detects all four dengue serotypes, J Med Virol 2002:66: 524-8.

17. SudiroTM. IshikoH. GreenS. et al: Rapiddlagnosisofdengueviremia by reversetranscripiase-polymerase chain reaction using 3'-noncoding region universalprimers. Am J Trop Med Hyg 1997: 56: 424-9.

18. fJrostenC. GottigS. Schillings, et al: fiapid detection and quantification of RNAof Eboia and Marburg vimses. i-assa vims. Crimean-Congo hemorrhagic fevervims. Rift Valley lever virus, dengue \1ms, and yellow fever \1ms by real-timereverse-transcriptase-PCR. J Clin Microbiol 2002: 40: 2323-30.

19. institute of Medicine: Strategtes to Protect the Health of Deployed U.S. Forces:Medical SunTillance, Record Keeping, and Risk Reduction, pp 21-32. Wash-ington, DC. Institute of Medicine. 1999, Available at http://www.nap,edu/openbook/0309066379/html/2 l-30.htm: accessed January 2. 2004.

20. Sharp TW, Wallace MR, Hayes CG. et al: Dengue fever in U,S. troops duringOperation Restore Hope, Somalia. 1992-1993, Am J Trup Med Hyg 1995; 53:89-94.

21. Niemeyer D, Jaffe R. Wiggins L: Feasibility determination for use of polymerasechain reaction in the U.S. Air Force Air-Transport ah le Hospital field emironmeni:lessons leamed. Milit Med 2000; 165: 816-20.

22. Niemeyer D: Rapid Biological Agent Identification Kenny Class Initiative AfterInitiative Report. Lackland Air Force Base. TX, Air Force Force Protection Battle-lab. 2000.

23. MaeDonald R, Cao T, Borschel R: Multiplexing lor detection of multiple biologicalwarfare agents shows promise In the field, Milit Med 2001; 166; 237-9.

24. Tliompson JD, Hi^ins DG. Gibson TJ, Clustai W: Improving the sensitivity ofprogressive multiple sequence alignment through sequence weighting, position-spe{'i(ir gap penalties and weight matrix choice. Nucleic Acids Res 1994: 1:4673-80.

25. Clewley JP. Megalign AC: The multiple alignment module of l.asergene. MethodsMolBiol 1997; 70: 119-29.

26. Wittwer CT, Herrmann MG, Moss AA, Rasmussen RP: Continuous iluorescencemonitoringof rapid cycle DNA amplification, BloTechniqiies 1997:22: 130-8,

Military Medicine. Vol. 170. December 2005

Rapid Identification of Dengue Vims by RT-PCR 1059

27, Altschul SF, Gish W, MUler W, Myers EW, Lipman DJ: Basic local alignmentsearch lool, J Mol Biol 1990; 15: 403-10,

28, Rosen L, Gubler D: The use of mosquitoes to detect and propagate dengueviruses. Am J lYop Med liyg 1974; 23: 1153-60,

29, Wittwer CT, Ririf KM, Andre RV, David DA, Gundry RA. Balis UJ: The LightCy-cier: a microvohime multisample nuorimeter with rapid temperature control.BioTechniques 1997; 2: 176-81.

30, Deparlmenl of the Anny; Joint Biological Afjenl IdendficaLlon and DiagnosticSystem pre solicitation notice. Federal Business Opportunities SolicitationDASG60 02 R 0008, Available al littp://v.'\w,eps,g<iv/servlet/Solidtation/R/USA/ SMDC/ DASO60/ DASG60-02 -R-0008,

31, Pasloske BL, Waikerpeach CR, Obemioeller RD, Winkler M. DuBois DB: ArmoredRNA technology for production of ribonuilease resistant \1ral RNA controls andstandards, J Clin Microbiul 1998; 36: 3590-4,

The road iseasier v/\th theright directions

wvsnv.menopause.org

NAMSTHE NORTH AMERICANMENOPAUSE SOCIETY

Tfie premier resource forwomen and healthcare providers

I I I )

Military Medicine, Vol, 170, December 2005