Sulfur mustard (HD) exposure can be verified with an assay developed at the Medical Research Institute of Chemical Defense (MRICD).
1 In general, mustard cannot be simply assayed from urine because of its reactive nature. Thiodiglycol (TDG) (2,2'-thiodiethanol) is one of the in vivo degradation product of bis(2-chloroethyl)sulfide (HD)
3 and can be used to confirm an exposure. TDG is itself subject to chemical and enzymatic transformations. A recent TDG assay demonstrated the existence of control urines with less than 1 nanogram (ng)/milliliter (ml).
4 However, analyte recoveries were only 50 to 70 percent. In the method presented here, detection of TDG after derivatization with heptafluorobutyric anhydride (HFBA) is achieved by using a gas chromatograph (GC) coupled with a mass selective detector (MSD). The lowest quantifiable concentration is 5.0 ng/ml. Thiodipropanol (TDP) is used as a stabilizer and octa-deuterated thiodiglycol (d8-TDG) as an internal standard. Through the use of spiked urine standards and the internal standard, a linear regression plot is used to determine TDG concentrations in urine samples.
2-2. Materials and methods
a. Materials. Urine specimens are collected from individuals suspected of exposure to mustard. Control urines should have less than 1 ng/ml TDG before use in standard preparations. Internal standard, d8-TDG, was obtained from Ash Stevens Incorporated (Detroit, MI). Other materials were obtained commercially. A listing of chemicals and vendors is presented in
b. Supplies and equipment. The procedure requires the use of various pieces of equipment and some common supplies which are listed as follows.
(1) Nitrogen and helium gases, UHP grade (Matheson Gases and Equipment, Montgomeryville, PA).
(2) Polypropylene microcentrifuge tube (Elkay Products Incorporated, Shrewsbury, MA).
(3) N-EvapTM (Organomation Association Incorporated, South Berlin, MA), an evaporator equipped with oil bath and gas nozzles.
(4) Centra-MTM centrifuge (IEC Company, Needham Heights, MA), rotor speed 13,200 revolutions per minute (rpm)/centrifugal force 15,600 x g.
(5) Mixer (Thermolyne Maxi MixTM; Thermolyne Corporation, Dubuque, IA).
(6) GC/MSD, 5890/5970B (Hewlett-Packard, San Fernando, CA).
(7) DB-5 bonded-phase capillary column (J and W Scientific, Folsom, CA), 20 meters, 0.18 millimeter (mm) inner diameter (I.D.), 0.40 micrometer (mm) film thickness.
(8) Pipettes and tips (Rainin Instrument Incorporated, Woburn, MA).
(9) Polyethylene scintillation vials (20 ml capacity).
The chemicals involved in these procedures are toxic. Follow all safety precautions listed below.
c. Hazards involved. TDG, d8-TDG, TDP, and HFBA will be handled in a fume hood. The derivatizing reagent, HFBA, is extremely reactive and toxic. Care must be taken when handling this reagent. Gloves, safety glasses, and a lab coat will be worn when handling chemicals or urine samples. Any further concerns are addressed adequately by the Material Safety Data Sheets.
d. Safety requirements.
(1) Ventilation. A general chemical fume hood will be used.
(2) Clothing. A standard lab coat, a pair of safety glasses, and latex gloves are required.
(3) First aid and fire fighting equipment. There will be standard first aid support and an eyewash station nearby. A standard chemical fire extinguisher will be placed near the work area in case of fire.
e. Preparation of solutions.
(1) TDP. In a plastic scintillation vial or other suitable plastic container, weigh out between 12 and 16 milligrams (mg), then add water until a final concentration of 1 mg/ml is reached. This solution is stable and can be used over a period of 3 months if stored under refrigeration.
(2) D8-TDG. In a plastic vial weigh out approximately 16 mg and dilute in water to a final concentration of 1 mg/ml. The stability data of TDG in aqueous media are not available. The stock solution of 1 mg/ml should be kept at -70 oC for long-term storage. The working solution of 0.1 mg/ml should be made fresh daily.
(3) Enzyme type H-1, b-glucuronidase. In a plastic vial weigh out 250 mg and add 10 ml of water. Gentle mixing in a closed vial is preferred in order to prevent foaming. This solution may be used for up to 1 week if kept frozen after each use.
(4) TDG. In a plastic vial weigh out approximately 16 mg and dilute in water to a final concentration of 1 mg/ml. The stability data of TDG in aqueous media are not available. The stock solution should be made fresh daily or kept at -70 oC for short-term storage.
f. Preparation of the standards and samples.
(1) Preparation of the TDG standards.
(a) Pipette 0.1 ml of the 1 mg/ml TDG solution and add 9.9 ml water. This gives a concentration of 10 micrograms (m g)/ml.
(b) Pipette 5.0 ml of 10 mg/ml TDG and add an additional 5 ml of water in a separate vial. Label this 5 mg/ml.
(c) Pipette 1.0 ml of 10 mg/ml TDG and add an additional 9 ml in a separate vial. Label this 1 mg/ml.
(d) Pipette 1.0 ml of 5 mg/ml and add an additional 9 ml in a separate vial. Label this 500 ng/ml.
(e) Pipette 1.0 ml of 1 mg/ml and add an additional 9 ml in a separate vial. Label this 100 ng/ml.
(f) Pipette 1.0 ml of 0.5 mg/ml and add an additional 9 ml in a separate vial. Label this 50 ng/ml.
(g) Pipette 1.0 ml of 0.1 mg/ml and add an additional 9 ml in a separate vial. Label this 10 ng/ml.
(h) Label the microcentrifuge tubes that are to be used to contain the standards as 0, 1, 5, 10, 50, 100, and 500 ng. Pipette 1.0 ml of a control urine into each tube. The control urine should contain less than 1 ng/ml of TDG. If water is used instead of urine, a non-linear standard curve will be produced.
(i) Pipette 0.1 ml of 5 mg/ml into the tube labeled 500 ng.
(j) Pipette 0.1 ml of 1 mg/ml into the tube labeled 100 ng.
(k) Pipette 0.1 ml of the 500, 100, 50, 10 ng/ml solutions to the tubes labeled 50, 10, 5, and 1 ng respectively.
(2) Preparation of the samples. Label the microcentrifuge tubes that are to be used for the samples appropriately. Pipette 1 ml of the urine sample to each tube.
(3) Enzyme digestion. Pipette 0.1 ml each of enzyme reagent, d8-TDG 0.1 mg/ml, and TDP 1 mg/ml solutions to each tube. Vortex the tubes and let them sit at room temperature for 1 hour.
(4) Derivatizing the samples.
(a) Add 40 microliters (µl) of 1.0 normal solution (N) hydrochloric acid (HCl) to each tube and check the pH with pH paper. Continue to add HCl in 10 ml increments until the pH is in the range of 2 to 3. If the pH falls below 2, use 1.0 N sodium hydroxide (NaOH) to adjust the pH.
(b) Place the tubes in the oil bath at a temperature of 90 oC. A gentle flow of nitrogen gas should be used to aid in the evaporation to complete dryness (approximately 1 hour). Remove the tubes from the oil bath.
(c) Add to each tube 1 to 2 molecular sieves.
(d) Pipette 400 µl HFBA and 400 ml ethyl acetate into each tube and allow the reaction to proceed at room temperature for 1 hour. Vigorously agitate each tube every 15 minutes. The sides of the tube can be scraped with a pipette tip to promote thorough derivatization.
(e) Centrifuge for 15 minutes and decant off the supernatant to a new tube.
(f) Add to the residue 200 ml each of HFBA and ethyl acetate. Vortex and derivatize again at 60 oC for 30 minutes.
(g) Centrifuge for 15 minutes. Decant the supernatant off to be combined with the supernatant from first derivatization.
(h) Evaporate the combined supernatant to dryness at 90 oC under nitrogen (about 30 minutes).
(i) Add 100 ml ethyl acetate to each tube and vortex. Centrifuge each tube for 10 minutes.
(j) Inject 1 to 2 ml of the supernatant to the GC/MSD.
g. GC/MSD parameters.
(1) The GC parameters are as follows:
(a) The injector port temperature is 220 oC.
(b) The transfer line temperature is 265 oC.
(c) The oven temperature is programmed as: initial temperature is kept at 45 oC for 1.1 minute, increased to 110 oC at a rate of 40 oC/minute, then to 125 oC at 3 oC/minute, and finally increased to 265 oC at 40 oC/minute and kept constant at 250 oC for 1 minute.
(d) The total inlet flow is set at 50 ml/minute.
(e) The split delay time is set at 0.20 minutes.
(f) The column head pressure is set at 12 pounds per square inch.
(g) The septum purge is set at 2 ml/minute.
(2) The MSD parameters are listed below.
(a) Data acquisition is set for selected ion monitoring.
(b) The solvent delay is set at 8 minutes.
(c) From 8.0 to 10.0 minutes, the ions mass to charge ratio (m/z) 300 and 301 representing TDG and m/z 307 and 309 representing d8-TDG are monitored. The dwell time for each ion is set at 20 milliseconds (msec), resulting in a cycling time of 7.2 cycles/second.
(d) From 10.0 to 11.0 minutes, the ions m/z 328 and 542 representing TDP are monitored. The dwell time for each ion is set at 20 msec, resulting in a cycling time of 10.0 cycles/second.
h. Analysis of chromatogram.
(a) After injecting the samples, the data editor will be used to integrate the peaks in the chromatograph. Retention times of 9.38, 9.42, and 10.5 minutes have been obtained for d8-TDG, TDG, and TDP under these conditions.
(b) Integration of the located peak is enhanced by narrowing the time window. When integrating, the threshold is set at 2 or lower for best results. When the integration is complete, the peaks at m/z 300 and m/z 301 will be within 0.04 minutes of the m/z 309 peak.
(c) The area count is used in analyzing the data. The peak area ratio is calculated by dividing the sum of the m/z 300 and m/z 301 areas by the area of the m/z 309 peak. This gives a peak area ratio which when plotted against concentration will yield a positive slope.
(2) Sensitivity and linearity. The range of sensitivity for quantitation is from 5.0 ng/ml to 500 ng/ml using a standard curve with points plotted at 0, 1, 5, 10, 50, 100, and 500 ng/ml. Linearity is found by using linear regression analysis in Lotus 123TM. An r2 value of at least 0.995 must be obtained before the curve can be used with confidence.
(3) Precision and accuracy. This assay was found to be accurate within a range of 5.0 ng/ml to 500 ng/ml. The standard curve is non-linear below 10 ng/ml. Precision varies with concentration ranging from 10 percent relative standard deviation (RSD) (percent RSD = standard deviation (SD)/mean x 100) at 10 ng/ml to 2 percent at 500 ng/ml.
i. Quality control.
(1) Tune the instrument each day prior to use.
(2) Change the septum in the injector port every day.
(3) Inject the samples first and then the standard curve starting with the blank and then proceeding from the lower to the higher concentrations.
(4) The source in the MSD should be cleaned periodically.
(5) Carrier gas should be flowing through the column at all times.
2-3. Results and discussion
a. Derivatization of TDG was necessary for two reasons. First, due to a low molecular weight (122 atomic mass unit) its presence would be obscured by the natural components of urine. Secondly, the polar functional group of TDG caused poor chromatographic peak shape, low sensitivity, and a low relative abundance of molecular ion. Detection limits for underivatized TDG in ethyl acetate were between 500 to 1000 ng on column. TDG can easily be esterified with acyl chlorides other than many anhydrides as was demonstrated by Black and Read.
4 The derivatized esters varied in their stability and usefulness in assays. For example, both the trifluoro and heptafluoro derivatives were prepared and analyzed by GC/mass spectrometer (MS) at MRICD. The HFBA derivative was more stable than the trifluoromethyl anhydride (TFA) derivative. The HFBA derivative produced analytically useful fragments at high molecular weights ranging from m/z 241 to m/z 301. Fragments at m/z 300 and 301 represented the loss of one -OCOC3F7 group leaving most of the analyte molecule intact. Therefore the identity of the HFBA derivatized analyte was confirmed by the characteristic fragments. However, the molecular ion of both the HFBA and TFA derivatives had low relative abundances (i.e., for the HFBA derivative the M+ was only 5 to 10 percent of the base at m/z 241) as was also noted by Black and Read. HFBA was still the derivative of choice in the analysis of TDG by electron ionization positive ion detection despite the low molecular ion abundance, because the m/z 300 and 301 fragments were structurally informative and possessed high relative abundances of 70 percent and 50 percent, respectively. Analogous fragmentations of the deuterated internal standard produced ions at m/z 307 and 309. The retention time of the bis (heptafluorobutyl derivative) was approximately 9.4 minutes with the deuterated internal standard eluting 0.04 minutes earlier.
b. TDP was used as a stabilizer to decrease binding effects of the analyte.
c. Because of its polar and very hydrophilic nature, TDG has been very difficult to extract from biological matrices. The strategy at MRICD was to derivatize the sample with excess HFBA after initial acidification and drying steps. The derivatized TDG was then more lipophilic and would extract into ethyl acetate. Amines and amino acids would form salts that would be insoluble in organic solvents. Samples were first made acidic (pH 2 to 3) to both stabilize the TDG during the drying step and to increase amine salt formation. Moderately acidic solutions of TDG are stable to many reactions even at temperature in excess of 100 oC.
5 The extent of derivatization was measured to be 95 percent disubstituted, 4 percent monosubstituted, and 1 percent unreacted. Using this sample preparation scheme it was possible to quantify TDG from 5 to 500 ng in spiked rat or human urine.
d. This assay method was validated in the analysis of urine samples collected from rats after subcutaneous injection and guinea pigs after vapor exposure of neat mustard. Mustard at 750 and 450 mg/kilogram (kg) was injected into rats and produced no overt symptoms. Urine was collected and analyzed from both exposure groups and from a control group. All urines were collected over TDP. Results for the 24, 48, and 116 hour samples are shown in
. These results were gathered before the deuterated internal standard was available; therefore TDP was the internal standard. The control groups did not contain measurable amounts of TDG. Trace levels of TDG (1 to 15 ng/ml) could be measured for up to a week post-exposure.
e. Guinea pig vapor exposure produced TDG levels of 34.4 to 297 ng/ml.
f. Thirty volunteer human urine samples were collected and assayed. Results indicated levels were below the detection limits of the assay. These results agreed with Black and Read.
4 Despite the limited survey of human control levels, the verification of mustard exposure is possible if samples can be stabilized and analyzed fast enough to ensure integrity.
Table 2-1. List of supplies and suggested vendors
|Ethyl acetate 99.5%
|Hydrochloric acid 1.0 N
|Molecular sieves (5A)
|Heptafluorobutyric anhydride (HFBA)
|Sodium hydroxide 1.0 N
Table 2-2. Thiodiglycol levels in rat urine after mustard exposure
||TDG concentration (ng/ml)
|HD dose (mg/kg)
1 Jakubowski, E.M., C.L. Woodard, N.M. Mershon, and T.W. Dolzine. "Quantitation of Thiodiglycol in Urine by Electron Ionization Gas Chromatography-Mass Spectrometry," J. Chromatog. 528 (1990), pp. 184-190.
2 Davison, C.D., R.S. Roman, and P.K. Smith. "Metabolism of Bis-b-Chloroethyl Sulphide (Sulphur Mustard Gas)," Biochem. Pharmacol. 7 (1961), pp. 65-74.
3 Roberts, J.J. and G.P. Warwick. "Studies of the Mode of Action of Alkylating Agents--VI. The Metabolism of Bis-2-Chloroethyl Sulphide (Mustard Gas) and Related Compounds," Biochem. Pharmacol. 12 (1963), pp. 1329-1334.
4 Black, R.M. and R.W. Read. "Detection of Trace Levels of Thiodiglycol in Blood, Plasma, and Urine Using Gas Chromatography-Mass Spectrometry-Electron-Capture Negative Ion Chemical Ionization," J. Chromatog. 449 (1988), pp. 261-270.
5 Reid, E.E., ed. Organic Chemistry of Bivalent Sulphur, Vol II, Chemical Publishing Co., N.Y. (1960).