Mass Spectrometric Determination of Cocaine and its Biologically Active Metabolite, Norcocaine, in Human Urine Satya P. Jinda1,t Theresa Lutz and Per Vestergaard Rockland Research Institute, Orangeburg, New York 10962, USA

A gas chromatographic mass spectrometric assay has been developed for the determination of cocaine and its pharmacologically active metabolite, norcocaine, in human urine. ['H3]Cocaine and [*H,]norcocaine were used as internal standards. The assay utilizes selective focusing to monitor in a gas chromatographic effluent the molecular ions of cocaine, ['H3]cocaine and the fragment ions of trifluoroacetylated norcocaine, ['H3]norcocaine generated by electron impact ionization. The assay can measure 2 ng ml-' each of cocaine and norcocaine with about 5% precision. The curves, relating the amounts of cocaine and norcocaine added to control urine per 'fixed' amounts of their labeled analogs, versus the appropriate ion intensity ratios are straight lines with nearly zero intercepts and slopes of 0.98f0.01 and 0.98+0.02, respectively. The methodology is used for the analysis of urinary cocaine and norcocaine from three human subjects who received 100 mg cocaine-HCL intravenously.

Park Forest South, Illinois), dimethylformamidedimethyl-[*H6]acetal (Tri-Deuter-8, Pierce, Rockford, Illinois), trifluoroacetic anhydride (Aldrich Chemical Co., Inc., Milwaukee, Wisconsin) were used without further purification. All solvents were of analytical grade. Silanized tubes with screw caps were used for the urine extractions. Urine samples of cocaine patients were processed as soon as obtained. [2H3]Cocaine (O-[2H3]methyl) was synthesized by treatment of benzoylecgonine with tri-deuter-8 using an established procedure for esterification of carboxylic acids." [2H3]Norcocaine (O-[2H3]methyl was prepared by N-demethylation of labeled cocaine using a known procedure' for the synthesis of unlabeled norcocaine from cocaine. Both the labeled compounds showed satisfactory mass spectral characteristics. Selected ion monitoring analysis (SIM) of [2H3]cocaine showed the presence of an ion equivalent to 98.5% 1 0 . 2 % [2H3]cocaineand an ion equivalent to 1.2% &0.2% of [2Ho]cocaine. A similar analysis of [2H3]norcocaine showed the presence of an ion equivalent to 98.7% & 0.2% [2H3]norcocaineand an ion equivalent to 1.6% k 0.3% of [2H,,]norcocaine.

INTRODUCTION

Cocaine, a tropane type alkaloid, was the first local anesthetic to be discovered; its chronic administration can cause convulsions, hypothermia and mydriasis. Furthermore, its ability to block the reuptake of noradrenaline into nerve endings, makes it a powerful central nervous system stimulant. Deesterificationl-' and N - d e m e t h y l a t i ~ n ~are . ~ the principal known pathways for the biotransformation of cocaine in animals and presumably in man. Hawks et al. provided conclusive evidence for the presence of norcocaine in the brain of monkeys dosed with cocaine.8 Borne and co-workers,' studied the pharmacological characteristics of norcocaine and found these to be comparable with that of cocaine itself. There has been a massive resurgence of cocaine abuse in the 1970s; consequently, there is a need for a sensitive and specific assay for this important drug and its metabolites in biological fluids."." This report describes a gas chromatographic mass spectrometric assay of cocaine and its pharmacologically active metabolite, norcocaine, in human urine. Selected ion m ~ n i t o r i n g ' ~ -was ' ~ used to measure urinary excretion of cocaine and norcocaine in human subjects after Extraction of cocaine and norcocaine from urine administration of cocaine. Co~aine-(O-[~H~]methyl) and norcocaine-(O-[2H3]methyl)were synthesized and T o urine (1 ml) were added appropriate amounts of used as internal standards. [2H3]cocaine and [2H3]norcocaine(typically 144 ng and 56 ng) as internal standards. The urine was adjusted to pH 8.5 with 1 N N g 4 0 H EXPERIMENTAL and extracted twice with 5 ml of cyclohexane. The organic fractions were combined, 1 ml 0.1 N HCL was added and the solution was shaken for 15 min. The Materials organic layer was discarded, aqueous phase was adjusted to p H 8.5 with 1 N N H 4 0 H and extracted twice with Analytical grade cocaine hydrochloride (U.S.P.C. 5 ml of cyclohexane. The organic fractions were Rockville, Maryland), benzoylecgonine (Techman, Ina. combined, dried with sodium sulphate, filtered and the solvent was evaporated at 40 "C under a gentle stream of t Author to whom correspondence should be addressed CCC-0306-042X/78/0005-0658$03.00 658 BIOMEDICAL MASS SPECTROMETRY, VOL 5, NO. 12, 1978

@ Heyden & Son Ltd, 1978

DETERMINATION OF COCAINE AND NORCOCAINE IN URINE

N2. The residue was treated with trifluoracetic anhydride at 25 "C for 45 min. After this period, the material was taken to dryness under a gentle stream of N2. The residue, containing cocaine and norcocaine-TFA, was reconstituted in benzene (50 pl). An aliquot (-1 pl) of this solution was injected into the gaschromatograph mass spectrometer for the selected ion monitoring assay. Recovery of cocaine and norcocaine, added to control human urine was studied at the 10ngmlF' and 7.0 ng ml-' level, respectively.

Instrumentation Mass spectrometr was performed using an LKB 9000 GCMS ~ y s t e m l ~ . 'equipped ~ with the multiple ion monitor/peak matcher accessory for exact mass measurements. Gas chromatography was performed on a 1.8 m glass column (2 mm i.d.), packed with 1.5'10 OV-1 on gas chrom Q 100/200 mesh. The column temperature was maintained at 205 "C, flash heater at 230 "C, separator at 235 "Cand the ion source at 250 "C. The accelerating voltage was 3.5 kV in the scan mode and 3.0 kV in the SIM mode, the ionization potential was 70 eV and the trap current was set at 60 FA. The magnetic field was kept constant by focusing on the background ion (column bleed) at m / e 281 and the additional voltages were 245 V, 213 V, 737 V and 695 V for measuring ion intensities at m / e 303,306,263 and 266, respectively. The retention times of cocaine and norcocaine-TFA were 2.5 min and 4.0 min.

RESULTS A N D DISCUSSION The mass spectrum of cocaine [Fig. l(a)] shows a molecular ion at m / e 303, base peak at m / e 182 and a peak of medium intensity at m / e 105 (benzoyl cation). A priori,

one would expect some localization of charge on the N atom, also to a lesser extent on the two carbony10 atoms in the molecular ion of cocaine." The former ion, in a typical B-cleavage process, loses a stable, stereochemically suitably placed, benzoate radical to give an even electron ion of maximum relative i n t e n ~ i t y . ' ~ 'The ~' molecular ion with charge on the carbony10 atom of the benzoyl moiety, fragments to give a highly delocalized and stable benzoyl cation, while the charge on the carbony10 atomof the carbomethoxy moiety results in a fragment ion at m / e 272. A reasonable mechanism for the fragmentation pattern is proposed in Scheme 1. Exact mass measurements of the ions in question indicate the correct elemental compositions; however, to provide further evidence concerning the proposed pathways, we have examined the mass spectra of several cocaine analogs. The peak at m / e 182 in the spectrum of cocaine is shifted to m / e 185 in the mass spectrum [Fig. l(b) of [2H3]cocaine (0-CD3) as well as in the spectrum" of another labeled cocaine analog, [2H3]cocaine (N-CD,); furthermore, as expected, in the mass spectrum of benzoylecgonine ethyl ester" the above peak is shifted to m / e 196. The peak at m / e 272 in the spectrum of cocaine is shifted to m / e 275 in the spectrum of [2H3]cocaine (N-CD3)11 but not in the spectrum of [2H3]cocaine (O-CD3) [Fig. l(b)], clearly establishing the loss of methoxy radical from the molecular ion. The ion at m / e 105 in the mass spectrum of cocaine also appears at the same m / e in all the abovementioned analogs of cocaine; however, as expected, this peak is shifted to m / e 119 in the spectrum of p-methyl benzoyl ecgonine methyl ester. (Scheme 1). The mass spectrum of norcocaine-TFA shows a fragmentation pattern containing seven fragment peaks above m / e 100 with intensity greater than 5% of the base peak (Fig. 2(a)]. Again, assuming charge localization on the N atom and

-

O= C CF3

105

I

COOCH,

(0)

loo-

CH,

(a)

0

I

11

N

C-0-CH,

0

&-:-0

50

0

50-

105

303

--8

272

190

8 I

0

II

I.

,I I

P

I.

5

1.

1 W

.-

O=C- CF,

t

W

?

100

0 -

CH,

0

I

II

;

I

N

10~,(b;[5

W,O-F-Q

[L W

50

50-1

I05

L I00

0

I

306

201

z72

I

I00

200

300

400

m /e

Figure 1. Mass spectra of (a) cocaine; (b) [*H,lcocaine.

@ Heyden & Son Ltd, 1978

COOCD,

300

200

400

m/e

Figure 2. Mass spectra of (a) norcocaine-TFA and (b) ['H& norcocaine-TFA.

BIOMEDICAL MASS SPECTROMETRY, VOL. 5, NO. 12, 1978 659

S.

P. JINDAL, T. LUTZ A N D P. VESTERGAARD

'

C-OCH,

0

Cocaine

I

J

mle

303.1461

m/e

\

303.1461

m/e

303.1461

I

1

I

0 -

II

CH3

C-OCH,

m/e 272.1279

m/e 182.0186

m/e 105.0334

Scheme 1. Postulated structures of El mass fragments of cocaine. Exact mass measurements reported here are accurate within *12 ppm.

ie carbony10 atoms in the molecular ion, and following ie established ion fragmentation mechanisms,20 all the iajor peaks in the spectrum can be rationalized kheme 2). Though the exact mass measurements of the uious ions indicate the correct elemental composions, additional evidence for the proposed frag-

mentation pathways was sought from analogous norcocaine derivatives. The peak at m / e 263 in the spectrum of norcocaine-TFA is shifted to m / e 266 in the spectrum of labeled norcocaine-TFA [Eig. 2(b)] and as expected, appears at m / e 363 in the spectrum of norcocaine-HFB. Again, in agreement with the proposed fragmentation

Norcocaine-TFA

m/e 385.2123

0

m/e 385.2123

0

I1

II

4

H

O

m/e 263.0766

--*

He?-CCF,

-

CH,O-C

m/e 194.0811

I

I m/e 385.2123

II

0

0

-+

Cpj--cc~,

11

0 m l e 164.0319

II

0 m/e

163.0236

mle 385 Scheme 2. Postulated structures of El mass fragments of norcocaine-TFA.

io

BIOMEDICAL MASS SPECTROMETRY, VOL 5,

NO. 12,

1978

@ Heyden & Son Ltd, 1978

DETERMINATION OF COCAINE AND NORCOCAINE IN URINE

Cocaine odded (ng)

0

Figure 3. Curve relating amount of cocaine (0-255ng ml-')

added to control urine versus the ion intensity ratio at m / e 303/306.For these samples 144 ng ml-' of ['H3]cocaine was added as internal standard in each sample.

pattern, the peak at m / e 194 in the spectrum of norcocaine-TFA as well as in the spectrum of norcocaineHFB is shifted to m / e 197 in the spectrum of ['H3]norcocaine-TFA [Fig. 2(b)]. The ions at m / e 163 and 164 in the spectrum of norcocaine-TFA and labeled norcocaine-TFA are shifted to higher mass by 100 amu in the spectrum of norcocaine-HFB confirming the proposed fragmentation pattern. Control human urine samples subjected to the described procedure for cocaine and norcocaine analysis showed no significant background at m / e 303,306,263 and 266. Known amounts of cocaine and norcocaine, along with their labeled analogs in 'fixed amounts', were added to control urine and processed as described above. Cocaine and norcocaine were quantitated from the ratios of ion intensities at m / e 303/306 and m / e 263/266 respectively. Analysis of cocaine data (Fig. 3) gave a slope of 0.98 0.01 and an intercept of 0.16* 0.2. Similarly norcocaine data (Fig. 4) gave a slope of 0.98 0.02 and an intercept of 0.2*0.15. These data affirm a simple linear relationship between the appropriate ion intensity ratios and concentration of cocaine and norcocaine and excludes significant isotopic fractionation in any of the physicochemical steps involved in the assay. Six samples of control human urine containing 10ngml-' of cocaine and 7.0ngml-' of norcocaine were analyzed by the above method using 7.5 ng ml-' of ['H3] cocaine and 5.6 ng ml-' of ['H3] norcocaine as internal standards. The analysis of these samples gave values of 9.7*0.3 ngml-' for cocaine and 6.9*

*

*

100 Norcocoine added (ng)

Figure 4. Curve relating amount of norcocaine (0-160ng ml-')

added to control urine versus the ion intensity ratio at m / e 263/266.For these samples 1 1 2 ng ml-' of ['H3]norcocaine was added in each sample.

0.2 ng ml-' of norcocaine [Fig. 5(a)]. These samples were assayed in duplicate; in this set exactly the same amounts were taken as above but the internal standards were added after the extraction. The recoveries for these samples, based on comparison of the appropriate ion intensity ratios of the two sets were 80% f 7 % for cocaine and 78% f8% for norcocaine. It should, however, be pointed out that the use of these labeled internal standards averts possible errors in analysis caused by variable extraction efficiencies, glassware and G C column adsorption. The precision of the assay is better than 5 % at the 10 ng mi-' level. Urine specimens obtained from adult patients after administration of a single dose of cocaine hydrochloride (100 mg administered intravenously) were evaluated. After the cocaine dose, three consecutive 2 h urine samples were analyzed from each of the three patients (Table 1). Although there is a considerable variability,in excretion patterns of these materials between patients, nevertheless, a few generalizations can be made. In the first 6 h following the dose, the amount of cocaine recovered in the urine (-1y0) is comparable with that reported earlier.3 The cumulative 0-6 h mean fSD urine concentration for cocaine is 1.5 f 0.8 pg ml-' urine. This value is comparable with the 0-8 h urine value 1.O f 0.1 pg ml-' for cocaine reported by Wallace et in adult patients receiving cocaine as a local anesthetic (250 mg cocaine-HCL). Furthermore, it

Table 1. Urinary cocaine and norcocaine levels of cocaine patients Cocaine (ng rnl-'~ Patient

A Total pgb B Total pg C Total pg

1

3376* 70" 236.4 6284f 78 578.2 843.4k9.2 210.9

2

1778*32 129.8 1012*21 177.2 105.7rt4.9 25.4

Norcocaine (ng rnl-')

3

148.5*7 14.4 150* 5.2 38.2 60.8*2.7 15.5

45.5* 3.8 3.18 87.0k3.8 8.0 4.6rt0.6 1.15

49.6*3.6 3.63 18.3k2.2 3.20 17.6* 0.9 4.22

11.8* 1.5 1.14 6.4*0.9 1.63 1.0*0.2 0.26

Three consecutive 2 h collective urine specimens (1,2,3) were obtained following intravenous administration of 100 mg of cocaine HCI to three patients. a Mean of duplicate determinations. Represents the concentration x volume of urine excreted.

@ Heyden & Son Ltd, 1978

BIOMEDICAL MASS SPECTROMETRY, VOL. 5, NO. 12, 1978 661

S . P. JINDAL, T. LUTZ AND P. VESTERGAARD

C)

n1

I63

Retention tlme (man)

I64

Figure 5. Selective ion chromatograms for cocaine ( m l e 303). [2H3]cocaine ( m l e 306) and norcocaine-TFA ( m l e 263). [2H3]norcocaine-TFA ( m l e 266). tnternal standard ----[2H3]cocainem l e 306, [2H3]norcocaine mle 266; urine [2Ho]cOcainem l e 303, [ZHo]norcocainem l e 263. (a) Control urine containing 10 ng ml-’ of cocaine and 7.0 ng ml-’ of norcocaine; internal standard concentration [2H3]cocaine, 7.5 ng ml-’ and norcocaine5.6 ng mi-’. (b)Urineobtainedfrom subjectA(2-4 h). Internal standard concentrations were 144 ng ml-’ of [2H3]cocaine and 56 ng ml-’ of [2H3]norcocaine, respectively. Cocaine was found to be 1778*32 ng m;-’ while norcocaine was in the concentration of 49.6* 3.6 ng mi- . (c) Urine obtained from subject C (2-4 h). Internal standard concentrations were 44 ng ml-’ of [2H3]cocaine and 28 ng ml-’ of [2H3]norcocaine. Cocaine was found to be 105.7k4.9 ng ml-’ while norcocaine was found to be 17.6*0.9 ng ml-’.

appears that most of the cocaine is excreted from the human body during the first 4 h after administration. This result is of significance in medico-legal cases where the laboratories are provided with 24 h urine samples of suspected cocaine abusers. The urine sample may be too diluted for cocaine detection.” The N-demethylation route of cocaine metabolism, leading to the formation of norcocaine, must be of minor importance. However, very much Eike cocaine, most of the norcocaine is excreted in the urine in the first 4 h after the administration of cocaine. The assay for cocaine and norcocaine presented here is sensitive, specific and probably applicable to other body fluids and tissues with minor changes. With near zero leak current in the ion source and better than 50% recoveries, an assay sensitivity of 2 ng ml-’ each of cocaine and norcocaine is possible. The assay presented here is 20-100-fold more sensitive than the recently reported GC assay for cocaine. Furthermore the specificity of the above method remains a distinct Results from the control human urine as well as from urine of patients show good specificity of the assay; the

I97 194

0

4

0

0

4

0

0

,

,

4

0

266

263

0

4

0

Retention time ( m i d

Figure 6. Selective ion chromatographs (a-d) at various rnle values for urinary norcocaine, with added [2H3]norcocaine, obtained from the extract of patient A (2-4 h).

selected ion recordings obtained from biological extracts, shown in Fig. 5, are clean and symmetrical peaks. The entire mass spectrum of urinary cocaine, with added [2H3]cocaine, could be recorded and it shows expected doublets at mle 303-306, mle 182-185 and a singlet at mle 105, a common fragment ion from the two isotopic species. The mass spectrum of urinary norcocaine-TFA could not be recorded in the scan mode. Nevertheless, the presence of all the characteristic ions of norcocaine-TFA could be ascertained by a systematic selective ion monitoring analysis (Fig. 6) or urinary norcocaine, with added [2H3]norcocaine. The selective ion monitoring analysis [Fig. 6(a-d)] shows characteristic doublets with constant ion intensity io at m/e 100-103,194-197,263-266 and at mle 385-388. Also of interest, are singlets of increased relative ion intensities at mle 105 [Fig. 6(a)], 163 and at mle 164 [Fig. 6(b)], the common fragment ions from the two isotopic species.

Acknowledgements Dr K.G Verebey of New York State Office of Drug Abuse Services, Testing and Research Laboratory, Brooklyn, New York is gratefully acknowledged for providing the urine samples of patients. Mr Sidney Bernstein is acknowledged for making illustrations. This work was supported by grant D A 00327 and by the Department of Mental Hygiene, State of N e w York, Office of Research.

REFERENCES 1. F. Montesinos, Bull. Narcotics, 17, 11 (1965). 2. R. V. Ortiz, Ann. Fac. Quiun. 15, 15 (1966). 3. F. Fish and W. D. C. Wilson, J. Pharm. Pharmacol. 21, 135 (1969). 4. N. N. Valanju, M. M. Baden, S.N. Valanju, D. Mulligan and S. K. Verma. J. Chromatogr. 81, 170 (1973). 5. B. Testa and P. Jener, Drug Metabolism. Chemical and Biochemical Aspects, p. 140. Marcel Dekker, New York (1976). 6. J. Axelrod and J. Cochin, J. Pharmacol. Exp. Ther. 121, 107 (1957). 7. E. G. Leighty and A. F. Fentiman, Res. Commun. Chem. Parhol. Pharmacol. 8, 65 (1974). 8. R. L. Hawks, 1. J. Kopin, R. W. Coburn and N. B. Thoa, Life Sci. 15, 2189 (1974).

662 BIOMEDICAL MASS SPECTROMETRY, VOL 5, NO. 12, 1978

9. R. F. Borne, J. A. Bedford, J. L. Buelke, C. B. Craig, T. C. Hardin, A. H. Kibbe and M. C. Wilson, J. Pharm. Sci. 66, 119 (1977). 10. B. S. Finkle and K. L. McCloskey, in Cocaine, ed. by R. C. Peterson and R. C. Stillman, p. 153 NlDA Research Monograph #13, US Government Printing Office, Washington, DC (1977). 11. S. P.Jindal and P. J. Vestergaard, J. Pharm. Sci. in press. 12. C. C. Fenselau, Appl. Specrrosc. Rev. 28,305 (1974). 13. S.D. Nelson and L. R. Pohl, in Annu. Rep. Med. Chem. 12,319 (1977). 14. C. G. Hammar, B. Holmstedt and R. Ryhage, Anal. Biochem. 25, 53 (1968). 15. J. P. Thenot and E. C. Horning, Anal. Lett. 5, 519 (1972).

@ Heyden & Son Ltd, 1978

DETERMINATION OF COCAINE AND NORCOCAINE IN URINE 16. J. E. Wallace, H. E. Hamilton, D. E. King. D. T. Bason, H. A. Schwertner and S. C. Harris, Anal. Chem. &, 36 (1976). 17. 8. Holmstedt and L. Palmer, Adv. Biochem. Psychophdrmacol. 7, 1 (1973). 18. H. Budzikiewicz, C. Djerassi and D. H. Williams, Structure Elucidation of Natural Products by Mass Spectrometry, Vol. 1, p. 219. Holden Day, San Francisco (1974). 19. C. A. Grob, Bull. SOC.Chim. Fr. 1360 (1960).

@ Heyden & Son Ltd, 1978

20. F. W. McLafferty, Interpretation of Mass Spectra, 2nd Edn, 194, p. 40. W. A. Benjamin Reeding, Massachusetts (1977). 21. M.J Kogan, K. G. Verebey, A. C. DePace. R. B. Resnick and S. J. Mule, Anal. Chem. 49, 1965 (1977). 22. D. L. Von Minden and D. Amato Anal. Chem. 49,1974 (1977). Received 21 February 1978 @ Heyden & Son Ltd, 1978

BIOMEDICAl MASS SPECTROMETRY, VOL. 5,

No. 12,

1978 663