Amphetamine Adducts of Melanin Intermediates Demonstrated by

8 Aug 2001 - The use of hair as a matrix for the determination of a history of drug abuse is becoming increasingly widespread. Melanin has been shown ...
0 downloads 0 Views 104KB Size
Download PDF
Chem. Res. Toxicol. 2001, 14, 1339-1344

1339

Amphetamine Adducts of Melanin Intermediates Demonstrated by Matrix-Assisted Laser Desorption/ Ionization Time-of-Flight Mass Spectrometry David J. Claffey and James A. Ruth* Program in Molecular and Environmental Toxicology, School of Pharmacy, The University of Colorado Health Sciences Center, Denver, Colorado 80262 Received May 23, 2001

The use of hair as a matrix for the determination of a history of drug abuse is becoming increasingly widespread. Melanin has been shown to play a key role in the incorporation of drugs in hair. The mechanism of this incorporation and the nature of the interaction remains poorly understood. Cationic drugs, such as amphetamine, are thought to be ionically bound to melanin; however, their inextricability has led to the suggestion that they may be covalently bound to a great degree. Identification of covalent adducts remains elusive due to the insoluble polymeric nature of melanin. We succeeded in identifying several such adducts by matrixassisted laser desorption/ionization-time-of-flight mass spectrometry (MALDI-TOF) analysis of the products of in vitro synthesis of melanin in the presence of amphetamine. Amphetamine was incubated with L-DOPA and mushroom tyrosinase under a stream of oxygen. After 1 h, a signal at m/z 281.1324 (n ) 1, R ) H) was observed. After 2 h, the major adduct mass visible in the spectrum was at m/z 470.1074. This appeared to be derived from the monodecarboxylation of a minor adduct at m/z 514.1245 (n ) 2, R ) CO2H). A totally decarboxylated adduct was also observed at m/z 426.1448 (n ) 2, R ) H). These were identified as amphetamine adducts of indole quinones. Corroboration of their identity was obtained by observing the mass shifts with deuterated L-DOPA and amphetamine analogues. Accurate mass measurements using the reflectron mode of the MS showed that the smaller adduct was within 14 ppm, and the larger adducts were within 70 ppm of their theoretical monoisotopic masses. Postsource decay experiments agreed with our structural assignments.

Introduction Over the course of the past decade, the use of hair as a matrix for the determination of a history of drug abuse has become increasingly common. Its power rests in its long window of detection, of months, rather than days in the case of urine (1). However, little is known of the exact mechanism as to how exogenous substances are deposited and retained in hair (2). The key role of pigmentation has been extensively documented for numerous drugs, with black hair generally being able to store significantly more drug than nonpigmented hair (3-8). These and other studies have lead to concerns of racial bias in hair testing (9-11) and called into question its validity. The ability of hair testing to distinguish between drug that has been deposited by actual drug use as opposed to passive external deposition has also been questioned (12, 13). Clearly, more information is needed as to how drugs are deposited and retained in hair and in particular the role that melanin plays needs to be clarified. Many drugs, including amphetamine, have been shown to have a reversible binding affinity to preformed melanin (14-16). Recent studies have shown that extensive irreversible binding occurs when the drugs are present during in vitro melanin synthesis (17). This too has been found in the case of in vivo experiments with [14C]-Lamphetamine (18, 19). Palumbo et al. succeeded in * To whom correspondence should be addressed. Phone: (303) 315 7569. Fax: (303) 315 6281. E-mail: [email protected].

isolating and characterizing a 2-thiouracil adduct of a melanin intermediate (20). This “bound” portion of the drug is not accounted for in routine hair screening. Its potential is enormous. Not only will it help in the understanding of the mechanistic role of pigmentation in drug deposition in hair, but since adducts are only formed during the course of melanin synthesis, the possibility of adduct formation upon passive drug exposure is greatly reduced. Such an intricately bound biomarker of drug use would be less easily removed from hair than “free drug” by normal hygiene practices, chemical hair treatment (21, 22), or by one of the “drug cleaning” shampoos now available (23). Until recently, it was impossible to directly detect these melanin adducts due to the inextricability of melanin and the insolubility of the larger oligomeric adducts. Recently, we successfully used matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry (MALDI-TOF MS) to detect covalent adducts of nicotine and cotinine

10.1021/tx0155303 CCC: $20.00 © 2001 American Chemical Society Published on Web 08/08/2001

1340

Chem. Res. Toxicol., Vol. 14, No. 9, 2001

Claffey and Ruth

Figure 1. Representative MALDI-TOF spectrum of the reaction mixture after incubation of l -DOPA and tyrosinase in distilled water for 4 h.

with a melanin intermediate (24). In this study, we extended this approach to look for amphetamine adducts of melanin.

Materials and Methods L-3,4-Dihydroxyphenylalanine

(L-DOPA), tyrosinase (from mushroom), and preformed melanin (from Sepia officinalis) were purchased from Sigma Chemical Co. (St. Louis, MO). L-DOPAd3 (ring-d3, 98%) was purchased from Cambridge Isotope Laboratories (Andover, MA). Amphetamine-d5 and -d8 were purchased from Radian International (Austin, TX). Melanin was synthesized in vitro as described by Bertazzo et al. (25). Here, L-DOPA (1 mg/mL) was treated with mushroom tyrosinase (1000 units/mL) in distilled water at 37 °C. Additional sample sets were prepared under the same conditions except either L-amphetamine, d-amphetamine, D-amphetamine sulfate, amphetamine-d5, or amphetamine-d8 were added at a 1:1 L-DOPA:drug ratio at the beginning of the oligomerization reaction. In each instance, a black precipitate formed within 60 min. Aliquots (10 µL) were removed from the incubation mixture at 1 and 4 h. An additional sample set was prepared by substituting L-DOPA for L-DOPA-d3. Blank sample sets were prepared by similarly carrying out the incubations with Lamphetamine in the absence of L-DOPA and/or tyrosinase. Each 1 µL sample aliquot was mixed with 1 µL of matrix solution. The matrix was prepared weekly by dissolving 3 mg of R-cyano-4-hydroxycinnamic acid (CHCA, Sigma Chemical Co.) in 1 mL of 70:30 acetonitrile/water (deionized, 20 MΩ cm) mixture containing 0.1% trifluoroacetic acid. A deuterated matrix solution was similarly prepared by washing CHCA three times with D2O. The matrix was dissolved in a 70:30 acetonitrile/D2O mixture containing 0.1% trifluoroacetic acid. A stainless steel MALDI target plate was used. The mixture was allowed to dry at room temperature and pressure. A PerSeptive Biosystems Voyager-DE PRO MALDI-TOF mass spectrometer (Applied Biosystems Inc., Foster City, CA) equipped with delayed extraction and a nitrogen laser (337 nm, with a focal diameter of 25 µm) was used for all analyses. The flight tube length in the reflector mode is 2 m. Spectra were acquired in the positive-ion reflectron mode using an accelerating voltage of 20 kV. The reflectron mode was used because the increased resolution aids in separating low-molecular mass sample ions from matrix ions and increases mass accuracy. Mass spectra were analyzed over the range of m/z 100-2500 by averaging the data from 64 laser shots. Multiple mass determinations (n ) 5-10) were made on each sample to provide a mean and standard deviation for each of the masses of interest. Internal calibration across the mass range was performed and employed the [M + H]+ ions of CHCA (m/z 190.0504 and 379.0930). Postsource decay (PSD) fragment ion analysis was performed on masses of interest by averaging 75 spectra in six segments

(decrement ratio of 0.75). The final PSD spectrum was produced by the Voyager Data Explorer software by stitching the individual segments together. LC-MS/MS analyses were performed on a PE Sciex API-3000 triple quadrupole mass spectrometer with a turbo ionspray source, interfaced with PE Sciex 200 autosampler and HPLC system. Flow injection analysis was performed with C18 column guard in isocratic 75% methanol in water containing 10 mM ammonium acetate with a flow rate of 200 µL/min and sample injection volume of 20 µL. The mass spectrometer settings used were: turbo ionspray temperature, 300 °C; needle spray voltage, 4500 V; declustering potential, 30 V; focus plate, 175 V; collision energy, 30 V; and collision gas (N2) density set at 10.

Results Figure 1 is a representative spectrum of a sample of the incubation of L-DOPA with tyrosinase after 4 h. Signals for the various dihydroxyindole oligiomers were detected at m/z 441 (n ) 3), 589 (n ) 4), and 736 (n ) 5). This assignment was confirmed when L-DOPA-d3 was similarly incubated with tyrosinase (Figure 2). The adduct masses shifted to 445, 594, and 742. When L-amphetamine is added at the start of the incubation a new signal at m/z 281.1324 corresponding to an amphetamine adduct is visible after 1 h. At 4 h this signal appeared to increase along with signals for higher molecular weight amphetamine adducts at m/z 426.1152, 470.1074, and 514.0979 (Figure 3). These four signals were assigned the structures shown in Figure 4, which corresponded closely to the theoretical masses for these compounds (Table 1). A mass shift for all four signals to m/z 283.1468, 430.1588, 474.1417, and 518.1318 was noted when amphetamine was incubated with L-DOPA-d3 and tyrosinase (Figure 5) indicating that a single indole nucleus is present in the smallest adduct and two such nuclei are present on the other three. The exact molecular weights closely match the theoretical weights for the proposed structures (Table 2). Similar incubation of L-DOPA with deuterated amphetamine analogues clearly demonstrated that the signal at m/z 281.1324 also contained an amphetamine nucleus. This signal disappeared in favor of a signal at m/z 286.1594 when amphetamine-d5 (Figure 6) and m/z 289.1789 when amphetamine-d8 (Figure 7) was used. Again the exact molecular weights matched closely with the theoretical weights of such adducts (Table 2). The other three adduct peaks did shift initially by several

MALDI-TOFMS of Amphetamine-Melanin Adducts

Chem. Res. Toxicol., Vol. 14, No. 9, 2001 1341

Figure 2. Representative MALDI-TOF spectrum of the reaction mixture after incubation of l -DOPA-d3 and tyrosinase in distilled water for 4 h.

Figure 3. Representative MALDI-TOF spectrum of the reaction mixture after incubation of amphetamine with L-DOPA and enzyme in distilled water for 4 h.

Figure 4. Proposed adduct structures. Table 1. Statistical Information on the Nondeuterated Adducts experimental (m/z) n standard deviation calculated (m/z) error (ppm)

281.1324 10 0.0089 281.1285 -13.8

426.1152 8 0.0028 426.1448 69.6

470.1074 10 0.0040 470.1347 58.2

514.0979 10 0.0040 514.1245 51.8

mass units when deuterated amphetamine analogues were used; however, after 24 h, only protonated analogues of the higher molecular weight adducts were detected. We hypothesize that the insolubility of these larger adducts in the aqueous medium brought them into more intimate contact with the highly reactive melanin surface, thus effecting the exchange of deuterium for hydrogen.

When deuterated matrix solution was used with a sample of the incubation of amphetamine, L-DOPA and enzyme, the major signal at m/z 470.1074 shifted to m/z 474.1417 in agreement with our proposed structure with four exchangeable hydrogens at the amino and carboxylic acid groups. The major fragment ions from postsource decay analysis of m/z 281.1324 (Figure 8) and its deuterated analogues are shown in Table 3. In each case, a signal for the protonated amino indole quinone (m/z 163) together with the protonated phenyl propene group (m/z 119) (Figure 9) and their deuterated analogues at m/z 124 and 127 were detected. The use of deuterated matrix resulted in the exchange of two deuterons on the secondary amines, resulting in the protonated dideuterated amino indole quinone signal (m/z 165). Fragmentation of 281 on an LC-MS resulted in the same major fragment ions at m/z 163, 119 and 91. PSD analysis of the larger adducts (Table 4) was less informative due to their greater resistance to fragmentation. Their major fragmentation pathway was through the loss of formic acid from ions m/z 470 and 514 and loss of the benzyl and phenyl propylene groups. No adducts were detected when L-amphetamine was incubated in water as described above in the absence of tyrosinase and/or L-DOPA. No adducts were detected

1342

Chem. Res. Toxicol., Vol. 14, No. 9, 2001

Claffey and Ruth

Figure 5. Representative MALDI-TOF spectrum of the reaction mixture after incubation of l -DOPA-d3, amphetamine and tyrosinase in distilled water for 4 h.

Figure 6. Representative MALDI-TOF spectrum of the reaction mixture after incubation of l -DOPA, amphetamine-d5, and tyrosinase in distilled water for 4 h.

Figure 7. Representative MALDI-TOF spectrum of the reaction mixture after incubation of l -DOPA, amphetamine-d8 and tyrosinase in distilled water for 4 h. Table 2. Statistical Information on the Deuterated Adducts experimental (m/z) n standard deviation calculated (m/z) Error (ppm)

283.1468 5 0.0108 283.1411 -20.6

286.1594 10 0.0016 286.1571 -8.0

when similar incubations were performed with preformed melanin. Similar spectra to that shown in Figure 3 were

289.1789 10 0.0011 289.1743 -15.8

430.1588 5 0.0406 430.17 26.1

474.1417 5 0.0240 474.1599 38.4

518.1318 5 0.0197 518.1497 34.6

obtained upon incubation with D-amphetamine sulfate and D-amphetamine free base.

MALDI-TOFMS of Amphetamine-Melanin Adducts

Chem. Res. Toxicol., Vol. 14, No. 9, 2001 1343

Figure 8. Representative PSD spectrum of m/z 281.129

Figure 9. PSD fragmentation pathway. Table 3. PSD Fragments matrix

amphetamine analogue

adduct (m/z)

d0 d5 d8 d0 d5 d8

281.1324 286.1594 289.1789 283 288 291

nondeuterated deuterated

major ions (m/z) 163 163 163 165 165 165

119 124 127 119 124 127

91 93 97 91 93 97

Table 4. PSD Fragments precursor Dopa Dopa d3

adduct (m/z) 426 470 514 430 474 518

major fragment ions (m/z) 310 452 496 315 430 474

293 426 470 252 119 428

281 252 215 200 119 91 281 176 163 119 238 136 119 91 119 91 91 119

Discussion In the course of investigating the mechanisms of drug incorporation into hair, we have discovered that the interaction of a range of drugs and endogenous substances appeared to be partially irreversible (5, 26-28). It is possible that this apparent irreversibility observed in in vivo studies is a function of structural barriers in the hair matrix or in the membranes of the melanosomes in desiccated keratinocytes of the hair shaft preventing release of radiolabled drugs associated with melanin. However, recent in vitro polymerization studies indicate that drug-melanin association is chemical, perhaps covalent in nature (17). Therefore, a study was undertaken to provide direct evidence of a drug-melanin association. This proved successful with our first drug of choice, nicotine. We found that it formed an adduct with the melanin intermediate L-DOPA. We extended this work to the drug amphetamine. Previously, Harrison et al. (19)

found that between 25 and 80% of radioactivity remained in the melanin when hair from animals, that were previously administered 14C-labeled amphetamine, was exhaustively digested. They speculated that amphetamine may be biosynthetically incorporated into the melanin; however, the means of detecting such incorporation was limited until recently due to the insoluble polymeric nature of melanin. MALDI-TOF MS appeared to be suited to this task as it can detect relatively high molecular weight material without the need for it to be in solution. As was seen in our previous studies, when L-DOPA was incubated with tyrosinase, a black precipitate was observed which displayed similar physical characteristics to naturally occurring eumelanin (24). When amphetamine was added at the beginning of the incubation a signal at m/z 281.1324, corresponding to an amphetamineindole adduct, was seen after 1 h. This signal became more intense as melanogenesis progressed and at 3 h a large signal at m/z 470.1074 was observed which corresponds to a larger molecular weight adduct (Figure 3). The exact molecular mass of the m/z 281.1285 and its postsource decay spectrum are consistent with an indole adduct of amphetamine. The use of the isotopically labeled L-DOPA and amphetamine analogues, further demonstrate that m/z 281.1285 consists of both an amphetamine and an L-DOPA related component. PSD yielded amino indole quinone and phenyl propene fragment ions (Figure 9). The identity of these fragments was confirmed with the observation of a deuterated amino indole quinone when deuterated matrix was used, and a deuterated phenyl propylene fragment when deuterated amphetamine was used (Table 3). Similar spectra were obtained from the LC-MS/MS of 281, confirming that these signals were not due to MALDI-TOF MS artifacts. Confirming the proposed structures of the larger adducts (m/z 426.1152, 470.1074, and 514.0979) was more difficult. When L-DOPA-d3 was employed in the polymerization the signals shifted up by 4 mass units. However, when the deuterated amphetamine analogues were used, the major signal shifted by only 2 mass units and was accompanied by many smaller signals. When these samples were reanalyzed 24 h later the signals observed were similar to a nondeuterated sample. We speculate that these larger insoluble molecules are interacting on the melanin surface. The PSD spectrum of this material

1344

Chem. Res. Toxicol., Vol. 14, No. 9, 2001

demonstrated that the three signals for the larger adducts were related to one another. Formic acid was the main neutral loss from the two larger adducts. This was also observed in the PSD spectrum of the two larger adducts from the incubation of amphetamine with LDOPA-d3. All these adducts decayed to a protonated phenyl propene demonstrating that they contain an amphetamine nucleus. These materials were not detected upon LC-MS analysis of the incubate. This may be due to their insolubility in the mobile phase. These adducts are remarkable in that they differ structurally from previously described adducts of nicotine (24) and thiouracil (20). In the case of nicotine and thiouracil, oligomerization appeared to have stopped upon adduction to DOPA and no larger adducts were detected. However, in this study larger adducts were readily detected. This is reflected in the very different type of adduct to that described previously (20, 24). Nicotine and thiouracil reacted directly with the DOPA quinone. However, amphetamine appears to react with the indole after the DOPA quinone formation. Knowledge of these chemical pathways will lead to better understanding of the deposition of drugs in hair and may increase the acceptability of hair as a substrate for forensic analysis (29). One of the major variables in using hair for drug analysis is that only the “free” (unbound) parent drug or their metabolites are substrates for analysis. These compounds are easily extracted by aqueous or organic solutions or by acid or base digest (26). This same pool of drugs have been found to be extractable by normal hygiene practices, chemical treatment (21, 22), or by one of the ‘drug removal’ shampoos now available (23). It can also be argued that these easily extractable drugs could have been deposited in the hair externally (12, 13). As demonstrated in this in vitro study, these drug adducts can only be formed in the presence of enzyme and melanin precursor and therefore could be a definitive biomarker for internal as opposed to external drug deposition. Work is progressing in the identification of these adducts in in vivo models.

Acknowledgment. This work was supported by U.S. Public Health Service Grant DA09545. We are indebted to Dr. Joseph Zirrolli of the Biochemical Mass Spectrometry Facility (BMSF) for the LC-MS spectra on the m/z 281 adduct. Mass spectral analyses were undertaken at the BMSF, a shared resource on the UCHSC campus. The authors thank Prof. Mark Duncan and Ms. Donna Dehn for valuable suggestions on the manuscript.

References (1) DuPont, R. L., and Baumgartner, W. A. (1995) Drug testing by urine and hair analysis: complementary features and scientific issues. Forensic Sci. Int. 70, 63-76. (2) Wennig, R. (2000) Potential problems with the interpretation of hair analysis results. Forensic Sci. Int. 107, 5-12. (3) Hubbard, D. L., Wilkins, D. G., and Rollins, D. E. (2000) The incorporation of cocaine and metabolites into hair: effects of dose and hair pigmentation. Drug Metab. Dispos. 28, 1464-9. (4) Stout, P. R., Claffey, D. J., and Ruth, J. A. (2000) Incorporation and retention of radiolabeled S-(+)-and R-(-)-methamphetamine and S(+)- and R(-)-N-(n-butyl)-amphetamine in mouse hair after systemic administration. Drug Metab. Dispos. 28, 286-91. (5) Stout, P. R., and Ruth, J. A. (1999) Deposition of [3H]cocaine, [3H]nicotine, and [3H]flunitrazepam in mouse hair melanosomes after systemic administration. Drug Metab. Dispos. 27, 731-5.

Claffey and Ruth (6) Slawson, M. H., Wilkins, D. G., and Rollins, D. E. (1998) The incorporation of drugs into hair: relationship of hair color and melanin concentration to phencyclidine incorporation. J. Anal. Toxicol. 22, 406-13. (7) Gygi, S. P., Wilkins, D. G., and Rollins, D. E. (1997) A comparison of phenobarbital and codeine incorporation into pigmented and nonpigmented rat hair. J. Pharm. Sci. 86, 209-14. (8) Gygi, S. P., Joseph, R. E., Jr., Cone, E. J., Wilkins, D. G., and Rollins, D. E. (1996) Incorporation of codeine and metabolites into hair. Role of pigmentation. Drug Metab. Dispos. 24, 495-501. (9) Kelly, R. C., Mieczkowski, T., Sweeney, S. A., and Bourland, J. A. (2000) Hair analysis for drugs of abuse. Hair color and race differentials or systematic differences in drug preferences? Forensic Sci. Int. 107, 63-86. (10) Cone, E., and Joseph, R. (1996) The potential for bias in hair testing for drugs of abuse Drug Testing in hair, pp 69-94, CRC Press, Boca Raton, FL. (11) Henderson, G. L., Harkey, M. R., Zhou, C., Jones, R. T., and Jacob, P. (1998) Incorporation of isotopically labeled cocaine into human hair: race as a factor. J. Anal. Toxicol. 22, 156-65. (12) Smith, F. P., and Kidwell, D. A. (1996) Cocaine in hair, saliva, skin swabs, and urine of cocaine users children. Forensic Sci. Int. 83, 179-89. (13) Mieczkowski, T. (1997) Distinguishing passive contamination from active cocaine consumption: assessing the occupational exposure of narcotics officers to cocaine. Forensic Sci. Int. 84, 87-111. (14) Knorle, R., Schniz, E., and Feuerstein, T. J. (1998) Drug accumulation in melanin: an affinity chromatographic study. J. Chromatogr. B 714, 171-9. (15) Shimada, K., Baweja, R., Sokoloski, T., and Patil, P. N. (1976) Binding characteristics of drugs to synthetic levodopa melanin J. Pharm. Sci. 65, 1057-60. (16) Bathory, G., Szuts, T., and Magyar, K. (1987) Studies on the melanin affinity of selegiline (deprenyl) and other amphetamine derivatives. Pol. J. Pharmacol. Pharm. 39, 195-201. (17) Claffey, D. J., et al. Manuscript submitted for publication. (18) Wepierre, J., Cohen, Y., Rapin, J. (1975) Accumulation de l’e´phedrine, de la Nore´phedrine, de l’amphetamine et de la tyramine marque´es du carbone 14 dans les yeux pigmente´s et non pigmente´s de deux races de rats. Arch Int. Pharmacodyn. Ther. 218, 342-350. (19) Harrison, W. H., Gray, R. M., and Solomon, L. M. (1974) Incorporation of D-amphetamine into pigmented guinea-pig hair. Br. J. Dermatol. 91, 415-8. (20) Palumbo, A., d’Ischia, M., Misuraca, G., Iannone, A., and Prota, G. (1990) Selective uptake of 2-thiouracil into melanin-producing systems depends on chemical binding to enzymically generated dopaquinone. Biochim. Biophys. Acta 1036, 221-7. (21) Yegles, M., Marson, Y., and Wennig, R., (2000) Influence of bleaching on stability of benzodiazepines in hair. Forensic Sci. Int. 107, 87-92. (22) Jurado, C., Kintz, P., Menendez, M., and Repetto, M. (1997) Influence of cosmetic treatment of hair on drug testing. Int. J. Legal Med. 110, 159-63. (23) Rohrich, J., Zorntlein, S., Potsch, L., Skopp, G., and Becker, J. (2000) Effect of the shampoo Ultra Clean on drug concentrations in human hair. Int. J. Legal Med. 113, 102-6. (24) Dehn, D. L., Claffey, D. J., Duncan, M. W., and Ruth, J. A. (2001) Nicotine and cotinine adducts of a melanin intermediate demonstrated by matrix-assisted laser desorption/ionization time-offlight mass spectrometry. Chem. Res. Toxicol. 14, 275-9. (25) Bertazzo, A., Costa, C. V., Allegri, G., Favretto, D., and Traldi, P. (1999) Application of matrix-assisted laser desorption/ionization mass spectrometry to the detection of melanins formed from dopa and dopamine. J. Mass Spectrom. 22, 287-97. (26) Claffey, D. J., Stout, P. R., and Ruth, J. A. (2000) A comparison of sodium hydroxide and sodium sulfide digestion of mouse hair in the recovery of radioactivity following systemic administration of [3H]-nicotine and [3H] flunitrazepam. J. Anal. Toxicol. 24, 548. (27) Stout, P. R., Dehn, D., and Ruth, J. A. (1998) Deposition and retention of radiolabeled serum constituents in hair following systemic administration. Drug Metab. Dispos. 26, 900-6. (28) Stout, P. R., and Ruth, J. A. (2000) Histologic localization of serum constituents 45Ca2+, 36Cl-, [14C]urea and [35S]cysteine in forming hair after systemic administration. Drug Metab. Dispos. 28, 1137. (29) Huestis, M. A. (1996) Judicial acceptance of hair tests for substances of abuse in the United States courts: scientific, forensic and ethical aspects. Ther. Drug Monit. 18, 456-9.

TX0155303