)2

US ARMY MEDICAL RESEARCH INSTITUTE OF CHEMICAL DEFENSE_4 ABERDEEN PROVING GROUND, MARYLAND 21010-5425

USAM.RICD-TR-89-08

00

TOM ANALYTICAL METHOD DEVELOPMENT ANDt-- IN VIVO AND IN VITRO EXPOSURE STUDIES OF

BIS- CMPIFLUOROMETHYL)DISULFIDE

-- LEC 19Ming L. Shih~SEP 01 189 John R. Smith

June 1989

Approved for public release; distribution unlimited

US ARMY MEDICAL RESEARCH AND DEVELOPMENT COMMANDFORT DETRICK. MARYLAND 21701-5012

89 8 31 032

DISPOSITION INSTRUCTIONS

Destroy this report when no longer needed. Do not return tothe originator.

The findings in this report are not to be construed as anofficial Department of the Army position unless so designated byother authorized documents.

In conducting the work described in this report, theinvestigators adhered to the "Guide for the Care and Use ofLaboratory Animals" as promulgated by the Committee on Revisionof the Guide for Laboratory Animal Facilities and Care of theInstitute of Laboratory Animal Resources, National ResearchCouncil.

The use of trade names does not constitute an officialendorsement or approval of the use of such commercial hardware orsoftware. This document may not be cited for purposes ofadvertisement.

11

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAIGE

r ~Form ApprovedlREPORT DOCUMENTATION PAGE I0MB4 No 0704-0188_____________________] Exp Date Jan 30. 1986

Ia. REPORT SECURITY CLASSIFICATION 1b. RESTRICTIVE MARKINGSUNCLASSIFIED

2a SECURITY CLASSIFICATION AUTHORITY 3. DISTRIBUTION /AVAILABILITY OF REPORTApproved for public release; distribution

2b DECLASSIFICATION /DOWNGRADING SCHEDULE unlimited

4. PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANI7ATION REPORT INUMBER(S)

USAMRICD-TR-89-08 USAMRICD-TR-89-086a NAME OF PERFORMING ORGANIZATION 6 b OFFICE SYMBOL 7a NAME OF MONITORING ORGANIZATIONUS Army Medical Research (if applicable) US Army Medical Research Institute ofInstitute of Chemical Defense ISGRD-UV-VA Chemical Defense, SGRD-VU-RC

6c. ADDRESS (City, State, and ZIP Code) 17b. ADDRE'iS (City, State, and ZIP Code)

Aberdeen Proving Ground, MD 21010-5425 Aberdeen Proving Ground, MID 21010-5424

Ba. NAME OF FUNDING ISPONSORING 8b. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION (If applicable)

8c. ADDRESS (City, State, and ZIPCode) 10. SOURCE OF FUNDING NUMBERSPROGRAM PROJECT ITASK WORK UNIT

ELMET O.NO NovACSINOI61102A 3MI61102B ISlIA 389ELE1N N D.L (ICCclSde N Ouiy C as iia i n

Analytical Method Development and In Vivo and In Vitro Exposure Studies of

12 PERSONAL AUTHOR(S)Shih, M.L.; Smith, J.R.

71.PGCON13a, TYPE OF REPORT 13b TIME COVERED 114 DATE OF REPORT (Yeair, Month, Day) 1.PG ONTechnical IFROM Jun 88 TO Oct _88 IJune 1989 2316. SUPPLEMENTARY NOTATION

17. COSATI CODES 18. SUBJECT TERMS (Continue on reverse if necessary and identify, by block number)

L FIELD GROUP SUB-GROUP Bis5( trifluorometbyl)dislufide,' Gas Chromatography-,06 04/11 Electron capture; Binding, Partition Ratio, In Vivo,06 15 In Vitro ý 1 _-

19. ABSTRACT (Continue on reverse if necessary and identify by block number)

A GCIEC analytical method was developed to quantitate bis- (trif luoromethyl)disulf ideaccurately in exhaled air of exposed animals. Using the headspace sampling techniquequalitative determination of unbound compound! in biological sample can be achieved.Partition ratio in plasma was much higher than in water (4.26 vs. 0.14). In vitro studiesindicated that irrever-sible binding and interaction could occur when bis-( tr-ifl-uoromethyl)-disulfide was present in the blood. The unbound portion disappeared at a faster rate fromRBC than from plasma. Bis-( trif luoromethyl)disulf ide might -have penetrated the lung tissueand entered the blood stream, but we did not detect any unbound compound in blood fromexposed animals. The results could be explained by irreversible binding and sequesteringaction by blood components.

20 DISTRIBUTION /AVAILABILITY OF ABSTRACT Z1 ABSTRACT SECURITY CLASSIFICATION12 UNCLASSIFIEDIUNLIMIT.ED 0 SAME AS RPT CDTIC USERS 1N.q~pF

22a. NAME OF RESPONSIBLE INDIVIDUAL --- 22b TELEPHONE (include Area Cd)22c. OFFICE SYMBOLGerald P. Jaax, LTC, VC (3012 671-194 7gn-ly

DID FORM 1473, 84 MAR 83 APR edition may be used until exhausted SECURITY CLASSIFICATION OF THIS PAGEAll other editions are obsolete. UN~CLASS IFIED

PREFACE

The work described in this report was initiated in June 1988 andfinished in October 1988 under protocol 1-03-87-000-A-422 at the US AnTyMedical Research Institute of Chemical Defense (USAMRICD) and was authorizedunder US-UK-CA memorandum of understanding on chemical-biological defense(MCU) dated June 1981. All laboratory data were recorded in laboratorynotebook No. 040-88.

ACKNOLELE=

The authors would like to express their sincere appreciations to LTCDavid T. Zolock for his valuable discussions and suggestions, Steve Musserand Edward M. Jakuboski for the nass spectrometry work, and leonardBuettner for supplying the corpound bis-(trifluorcmethyl)disulfide.

U ' ,;, •.td

I .. ......... ...... ------

iii

LIST OF FIGURES

Fiiure No.

1 Standard Calibration Curve for Air Sample 42 Standard Calibration Curve for Aqueous and

Plasma Sample 63 GC/EC Cromatogram of Bis-(trifluorcmethyl)disulfide 84 GC/EC iruem0atogram of CS2 95 GC/EC Circatograni of Headspaoe of CoMtrol RBC 116 GC/EC Qirxnatogrwn of Headspace of RBC Spiked

with Bis- tritluorcuethyl) disulfide 127 GC/EC Chramatogram of Headspace of RBC from

Rats Injected Intraperitoneally withBis- (trifluoranethyl) disulfide 13

8 In Vitro Exposure Study of RBC and Plasma toBis- (trifluoromethyl) disulfide 14

V

INTROUCTION

The Analytical Chemistry Branch of USAMRICD was tasked to develop ananalytical method for determining the presence of ccmpourn bis-(trifluorarethyl)disulfide in either exhaled air or blood samples taken fromanimals exposed to the caqpouni This organamlogen ccmxxuwd was firstsynthesized in 1959 (1). A pesticide action was reported in 1970 by Dearand Gilbert (2): the vapor of one pound per thousand cubic feetconcentration caused 100% mortality in beetles, larvas, and larvae eggs.The chlorine analog was relatively nrntoxic. The cummr biochemicalmechanism which explains toxicities in all species has not been identified.This report presents an analytical method to quantitate the unbound campoundin either biological matrices or exhaled air. The analysis results of bloodsanples frcm exposed anir5zls as well as degradation phencaneno observed inin vitro exposure studies might lead to a better understandiin of thebiochemical basis of the towicity of this compound. Partitions of theccwpound in water and blood were also studied.

MATERIALS AND METHODS

We selected the headspace samrpling technique and the gas chrcmatographinstnumeant using an electron-capture detector (GC/EC) as the analyticalmethod because of the volatility and the characteristic halogen functionalgroups of the compound. Flame photametric and mass detectors were also usedwhen necessary for structure confirmation. The headspace samplingtechnique is an established technique for quantitating volatile canpourds. inbiological matrices. This technique eliminates the ieed for an exhaustivesample extraction procedure because volatile ompcurnds can be driven off thebiological matrix when samples are heated and air spaoe above the matrix canbe sampled and injected directly onto the GC column.

Material

A Hewlett Packard gas chramatograph model 5880 equipped with an electroncapture detector was used. A GS-Q megabore column (30 m x 0.53 mm I.D.)installed in the GC oven was maintained at an isothermal tenperature of1200 C and the helium carrier gas had a flow rate of 10 ml/min. The injectorteupexature was off and detector temperature was set at 2500 C usingargon/methane as makeup gas at a flow of 20 ml/Inin. The Finnigan massspectrcmeter model 4500 was operated in chemical ionization mode and same GCconditions were used.

Gaseous bis-(trifluorcmethyl)disulfide was cbtained fro PR ResearchChemicals (Gainesville, FL) in gas cylinder. The liquid form used in thesestudies was made by oooling the vapor transferred from the cylinder. A 100%purity was stated in the coapany data sheet, but we identified about 6%impurities including carbon disulfide, carbonyl sulfide, and chloroform.

1

1. GC/EC Calibration airve for Air Sanple

The response versus concentration calibration curves for air sanplesusing GC/EC detection system were established as follows. One litercalibrated gas sampling bags (Calibrated Insrtwts Irnc., Ardsley, N.Y.10502) were inlated with a pulse rxmP (same source). Differentcoicentraticns of bis-(trifluoramethyl)disulfide solutions were prepared inacetonitrile and 10 uL of each sandard solution was added to the gas bag.The final concentrations of bis-(trifluoruaethyl)disulfide in the gas bagswere 13.4, 28.9, arn 48.6, and 152.0 ug/L. Five microliters of the airsanple was injected onto the GC column.

2. GC/EC Calibration Curve for Aqueous and Plasma Sample Using HeadspaceSawpling Technique

To spike water and plasma samples the neat bis-(trifluoromethyl)-disulfide was dissolved in acetonitrile at 41 ug/mL concentration andvarious microliter volumes of this stock solution were spiked to I mL ofwater or sheep plas to prepare standard solutions of conoentrationsrarqing from 82 to 820 nr/mL. Each spiked standard solution was kept in atwo mL GC autoswapler glass vial that was crirp-sealed with a teflon discand metal cap. The vials were placed in a constant temperature water bathmaintained at 370C. The standard solutions we-e incubated jiediatelyfollowing spiking. Two microliters of the headspaoe was sampled andinjected into the GC at 5 and 15 minutes incubation time.

3. Rat Intraperitoneal Absorption Studies

Three groups of four rats each were injected intraperitoneally withneat bis-(trifluorcmethyl)disulfide liquid at doses of 0.5, 5, and 50 uL andbeheaded 1-4 hours after exposure. Each group of rats was kept in a closedchamber, and the exhaled air was pumped into one liter gas sampling bag foranalysis. Blood samples were collected in rubber stoppered serum tubes andcentrifuged inmediately in a refrigerated centrifuge (10C) to separateplasa from red blood cells. One mL of whole blood, plasma, or red bloodcells was placed in a two mL GC autosampler glass vial sealed and incubatedas above. Fifty uL of the air above the sample was injected into the gascaroinator~

4. Sheep Inhalation Stbdies

Sheep were exposed to bis-(trifluoramethyl)disulfide at 5 pimconcentration in breathing air for 10 miniutes. Blood sanples were dLawn andcentrifuged immmliately in a refrigerated centrifug to separate plasma frcmred blood cells and analyzed as above.

5. Partition Ratio in Water

The partition ratio of bis-(trifiuorcmtthyl)disulfide in water wasdetermined by adding 10 uL of the neat co ound to 50 mL water in a 60 mL

2

serum vial. The vial was crimp-sealed and ontents wre stirred with asmall magnetic stirrer for 30 minutes at rocm temperature. Equal injectionvolumes, 0.6 uL of liquid phase or air phase, were injected into GC. Thepeak areas of water phase versus air phase were used to derive the partitionratio of this ccnmonrd in water.

6. In Vitro __posure Studies

Bis-(trifluorcmethyl)disulfide was dissolved in acatcnitrile at aconcentration of 7.56 mg/mL. Ten microliters of the solution was added to 1mL of sheep plasma or red blood cell to achieve a final cncrentration of75.6 ug/mL and incubated at 37°C. Water was used as a cotrol. One uLvolume of headspace was taken periodically and analyzed over a five-hcurtime period.

RESULTS AND DISCUSSION

I. Stanrd Calibration Curve for Air Sample

Table 1 lists the linearity, intra- and inter-day repeatabilities ofthe detector response. The values of the correlation coefficient of thecalibration curve were determined on tbree different days and were not lessthan 0.997. The coefficient of variation (CV; standard deviation over mean)for repetitive injections rarned from 1.7 to 7.9%. The averaged regressionline for the calibration curve was plotted in Figure 1 with a slope andintercept values of 833.4 and -4898 respectively.

Table 1Linear Calibration Curve for Air Sample

Peak Area CountDay 1 Day 2 Day 3 Inter-Day

Conc(u/L) Me•an CM n Mean CA n Mean CV% n -..an C n13.4 9573 7.9 6 9755 1.8 3 8895 7.3 3 9407 4.8 328.9 20145 6.4 6 20402 3.1 3 19830 4.0 3 20125 1.4 348.6 29688 2.7 6 31760 5.2 3 29424 5.9 3 30290 4.2 3

152.0 121146 1.7 6 121737 2.7 3 126235 2.6 3 123039 2.2 3

r2 0.997 0.998 0.997 0.997slope 818.6 819.6 862.1 833.4

intercept -4575 -3862 -6256 -4898

3

0

o~ N '

4-d

'-4LL L

00TlCI-

asosa- aIV)e)

2. Standard Calibration Curve for Aqueous and Plasma Samples UsingHfeadspce Samplin rD T±inque

Table 2 presents the peak area response versus xrnentration at 5 and315 minute incubation times fram aqueous or plaza samples using theheadspace sanpling tedinique.

Table 2Linear Calibration Curve for Aqueous and Plawa Sample

Peak Area Counts5 minutes 15 minxrtes

conc (ng/ml) H2 0 Plaa %* H20 Plasma

82 16273 ND** 17725 ND

123 26277 420 2 22573 ND164 33778 7376 22 30466 ND287 60258 12929 21 54538 ND410 74136 - 80946 ND615 106174 16682 16 103758 ND820 152446 23916 16 147585 ND

r 2 0.997 0.926 0.997slope 176 26.8 174intercept 4083 1815 3190

* Percent peak area count in the headspace of plasma versus H20 ** NDstands for not detectable

The linear regression plot of detector response versus concentrationsof spiked water and plasma samples using headspace analysis is shown inFigure 2.

Bis-(trif1uoramethyl)disulfide was very easily and cxmpletely driveninto the head space frcm the water solutias. This was proven when theaqueous pase was injected on the oolumn and no compound cwuld be detectedin the aqueous phase after the sample was heated. 1aroduible slopes wereobserved for the 5- and 15- minute plots for the water sales, andcorrelation coefficients for the linear regression line (r7) ware 0.997. Incontrast the correlation coeffikient for the plasma samples at 5-winuteincubation time was poor (0.926). At 15 minutes no free oczrpcid could bedeatected in the -headspace of any plasim samples.

As seen in Table 2 the peruet peak area count of the plaDma headspaoesample was negligible at low concentrations. Owe percent maintainv1 betwd16 to 21% (average value 19% of total spiked ammt) between theconcentration 164 to 820 ng/mL. This seems to indicate that bis-

5

cc

C13)-0.0

CCu

CZ0

QO N 0 00 (-0

(,Ol asudsa- uaa )T0

(trifluoromethyl)disulfide was irreversibly bound to plsmaa initially andhence very little released to the headspace. At higer xntraticms bothbinding and partitioning processes were involved. The average percent ofbis-(trifluorcuethyl)disulfide remaining in plama was 81% from 164-820ng/iL. Ths equated to a partition ratio of 4.26 in plasma. The absence offree cmxxnrx in the headspace at 15 minutes indicated that somes trim interaction other than binding occurred between the compoundand plasma xzxrkents during this time period.

3. Partition Ratio of Bis-(trifluoruet!tyl)disulfide in Water

Table 3 shows the peak area response of repetitive injections of anequal volume of the air phase and water phase. The partition ratio of bis-(trifluormethyl)disulfide in water was found to be 0.14 by dividing theaverage peak area of water over air.

Table 3Partition Ratio of Bis-(trifluorcmethyl)disulfide betwen Air and Water

Peak Area CountAauecus Phase Air ase

Average 1101188 7994186SD 63502 1511775CV 5.8% 14.4%

No. of Injection 4 5

Partition Ratio = 1101188/7994186 = 0.14

The partition ratio is defined as the equilibrium ratio between the

concentration in water versus in air as in equation (I).

Partition Ratio = Concentration in water/Ccncentration in air (I)

The concentration in water is the sawe as the solubility in water(Swater). The volumes occupied by the water and air are 50 and 10 mLrespectively. The density of the ccupound is 1.52 and 10 uL was added. 7heamount present in the air would be the total amount added m s the amountdissolved in the water as expressed in equation (2).

AAMOnt in air = 1.52 (mVuL) x 10 uL - Sw (mJmL) x 50 inL (2)

And the concantration in air can be expressed as the acmunt in air(equation 2) over the volum of the air space.

CeMontratiCn in Air (MJ•rL) = (15.2 - Sater x 50)/10 (3)

Therefore, equation (1) can be rewritten as equation (4);

7

Figure 3. GC/EC Chro-nattogarn of Bi-tilooety~iufd

CNI

V),

Bis-.(trifluoromethyl)disulfide Standard in Air 2.3 X 10(-4) mg/mi

8

Figure 4. GC/EC Chromatogram of CS2

-- U"U

Partition ratio = 0.14 = Swater/C(15. 2 - Swater x 50)/10) (4)

By solving the above equation the solubility in water was calculated tobe 0.12 Mn/mL.

We found, however, that a definitive solubility value in water wasdifficult to assess because the conpouni was in oonstant dynamic equilibriumbetween the air and water phases and had a terdancy to escape to the airphase. If a solution of bis-(trifluoromethyl)disulfide was left open, verylittle compound could be detected. In a closed syst when partitionequilibrium was reached the solubility in water would be varied with theconcentration in the air space. Therefore, the partition ratio wciuld be amore meaningful constant.

4. Analysis of Blood Samples and Exhaled Air from Rat Injected IP

Figures 3 and 4 show that the retention times of bis-(trifluorcmethyl) disulfide and CS2 are 5.72 and 5.91 minutes respectively inthe GC/EC chrcmatograms. The peak at 5.16 minutes seen in the chromatogramof control blood (Figure 5) was not a sulfur compound as confirmed by flamephotcmetric detector, but structural determination was not amenable for massspectrometer identification. In a spiked blood sarple this unidentifiedpeak and CS2 peak were buried under bis-(trifluorcmethyl)disilfide andappeared as only one single peak (Figure 6) and reappeared when bis-(trifluorcmthyl)disulfide was sequestered after 91 minutes irKjbation(Figure 7). Headspace analyses of all the blood samples from exposed ratsdid not show the presence of bis-(trifluoramethyl)disulfide at the retentiontime of 5.7 minutes (Figures 8), but OOS and CS2 peaks were present. Traceamount of CS2 and CoS were also detected in the exhaled air by EC.Approximately 1 ug/mL of CS2 detected in the blood. Threfore, it probablywas an impurity of the neat material and not a biotransformed netabolite.

5. Analysis of Blood Samples frca Sheep Inhalation Studies

In the sheep inhalation study we did not detect any parent caupord inthe head space of plasma, RBC, or whole blood samples. In the expired airsamples small amounts of CS2 and C0O were detected using mass spectrauetrictechniques.

6. In Vitro Exposure Studies

The observation of irreversible binding and fast elimination of uricRrdbis- (trifluoranethyl) disulfide in plasma led us to use a system of iruchhigher concentration to cbserve the kinetics over a longer period of time.The time course of disappearance of bis-(trifluoromethyl)disulfide frm theheadspace is shown in Figure 8 at a semi-logarithm scale. The sameconcentration in water was used as the control. In the water control aoxnstant level of conpound was present in the headspace, inlicating noleakage in the system or the possible reaction between bis-(trifluororethyl)disulfide and water. In the plaauwa sanples the nbo.-dozound d initially at a slower rate and then a faster ae over

10

Figure 5. GC/EC Chromatogram of Headspace of Control RBC

ýD- !4

cvJi Zi

Rd

Control Fiat RBC

11

NI.:

lOu

u =5

o -o

cu

00

5.4

C)

S0 V*

u -E

Figure 7. GC/EC Chromatogram of Headspace of RBC from Rats InjectedIntraperitoneally with Bis-(trifluoromethylI'disuIfide

%.0

'0 U

013

0T LCU)7

1- _ 0.L

U)4wI

I-'N

00

03-Il

-CD

UI UI

asoda- uaj ., J

the five-hour period. Similar biphasic kinetics were cbserved in RBCsainples, and rates in RBC were faster than in plasma.

In the same sheep exposure study results obtained usin an infra-redtechnique indicated that about 27% of a given dose was retained by sheep andonly 73% was recovered in the edhaled air (3). Pathological studiesindicated damage in endoothelial and red blood cell miirbanes (4). Our dataon blood sanples from exposed animals did not indicate the preseno ofunbound oound. Bt the i yjr studies strogly sqport thepossibility that the compound c ould penetrate through lung tissue as well asinteract with, and be sequest by, blood cor ments without beingdetected through headspace analysis.

JNCI1USIONS

This report demonstrates that a gas chrmatograPhy/electron capturemethod can reproducibly quantitate exhaled air sarples or aqueous sanpleswhen the headspace sampling technique is used. The lower detection limitis 13 ppb in air sample or 70 pg on column injection. For blood samplesonly qualitative determination of unbound compcund can be made because ofthe irreversible binding between blood and the cumpound and degradation ofbis-(trifluorcmethyl)disulfide when the samples are aged.

The partition ratio of bis-(trifluorcmethyl)disulfide was 0.14 in waterand 4.26 in plasma.

3In vitro exposure studies indicated the occurrerce of irreversiblebinding and biphasic degradation of bis-(trifluorumethyl)disulfide in plasmaand red blood cells.

In animal exposure studies we could not detect any unbound cmqxxl inthe headspace of either blood samples frum i.p. dosed rats and fram sheepexposed through inhalation or exhaled air from the dosed rats. CS2 and ODSwere detected in the exhaled air of exposed animals. We believe they werefrum the inpurities present in the original ccmpuund and not the result ofbiotransformation of the parent ouqxunrd because the amount detected was ata trace level compared to the dose given and because the iupIrities '--realso present in spiked water samples. No other volatile chemicaldwxzrposition products were observed during the exposure period.

15

REFERENCES

1. Tullock, C. W. US Patent 2,884,453, "Reaction of Metal Fluorideswith Ticpkcene and Pezrdlorcmethyl Mercaptan," Apr. 28, 1959.

2. Dear, R.E.A., Gilbert, E.E., US Patent 3,505,459, "Fumigation with

Bis(trifluoranethyl) Disulfide," Apr. 7, 1970.

3. Braue, E.H., USAMRICD, personal corinication.

4. Petrali, J.P., USAMRICD, personal cominication.

17

Distribution List

Addresses Copies Addresses Copies

Defense Technical Information 12 CommanderCenter US Army Research Institute of

ATIW: DTIC-DDAC Environmental MedicineCameron Station, Bldg 5 Bldg 42Alexandria, VA 22314-614S Natick, MA 01760-5007

Commander 2 CommandantUS Army Medical Research US Army Chemical School

and Development Command ATTN: ATZN-CM-CFort Detrick, MD 21701-5012 Fort McClellan, AL 36205

HQDA(DASG-HCD) 1 DirectorWashington, DC 20310 Armed Forces Medical

Intelligence CenterDirector 1 Fort Detrick, MD 21701-5004Walter Reed Army Institute of

Research CommanderBldg 40 US Army Institute of DentalWashington, DC 20307-5100 Research

Bldg 40Commander I Washington, DC 20307-5100Letterman Army Institute

of Research CommanderBldg 1110 US Army Institute of SurgicalPresidio of San Francisco, CA Research

94129-6800 Bldg 2653Fort Sam Houston, TX 78234-6200

Commander 1US Army Aeromedical Research Commandant

Laboratory Academy of Health SciencesATIN: Scientific Information Ctr US ArmyP.O. Box 577 ATIN: HSHA-CDCFort Rucker, AL 36362-5000 Fort Sam Houston, TX 78234-6100

Commander 1 CommandantUS Army Biomedical Research Academy of Health Sciences

and Development Laboratory US ArmyBldg 568 ATmN: HSHA-C2MFort Detrick, MD 21701-5010 Fort Sam Houston, TX 78234-6100

Commander 1 Mr Thomas R. DashiellUS Army Medical Research Institute Director, Environmental and

of Infectiou's Disease Life SciencesBldg 1425 Office of the Deputy Under SecretaryFort Detrick, MD 21701-5011 of Defense (Rsch & Adv Technology)

Room 3Di29Washington, DC 20301-2300

19

Comande r 1 Department of Health andUS Army Training and Doctrine Human Services

Command National Institutes of HealthATTN: ATMD The National Library of MedicineFort Monroe, VA 23651 Serial Records Section

8600 Rockville PikeCommander I Bethesda, MD 20894US Army Nuclear and Chemical

Agency Stemson Library7500 Backlick Road Academy of Health SciencesBldg 2073 Bldg 2840, Rn 106Springfield, VA 22150-3198 Fort Sam Houston, TX 78234-6100

Biological Science Division 1 US Army Research OfficeOffice of Naval Research ATTN: Chemical and BiologicalArlington, VA 22217 Sciences Division

P.O. Box 12211Executive Officer 1 Research Triangle Park, NCNaval Medical Research Institute 27709-2211Naval Medicine ComandNational Capital Region AFOSR/NLBethesda, MD 20814 Bldg 410, Rm A217

Bolling AFB,, DC 20332USAF School of Aerospace 1

Medicine/VN CommanderCrew Technology Division US Army Chemical Research,Brooks AFB, TX 78235-5000 Development & Engineering Ctr

ATIN: SMCCR-MISCommander 27 Aberdeen Proving Ground, MDUS Army Medical Research 21010-5423

Institute of Chemical DefenseATTN: SGRD-W-ZA

SGRD-UV-ZBSGRD-UV-ZS (2 copies)SGRD-UV-RC (5 copies)SGRD-UV-R (13 copies)SGRD-UV-AXSGRD-UV-DSGRD-UV-PSGRD-UV-VSGRD-UV-Y

Aberdeen Proving Ground, MD21010-5425

20