active metabolite 6-monoacetylmorphine (6-MAM), which is then deacetylated to morphine. ..... Determination of morphine, morphine-3-glucuronide and mor-.
Journal of Analytical Toxicology,Vol. 23, November/December 1999
Determination of Opiates and Cocaine and Its Metabolites in BiologicalFluidsby High-Performance Liquid Chromatographywith ElectrosprayTandem Mass Spectrometry A. Cailleux, A. te Bouil, B. Auger, G. Bonsergent, A. Turcant, and P. Allain* Laboratoire de Pharmacologie-Toxicologie, CHU, 4 rue Larrey, 49033 Angers, France
Abstract A rapid, sensitive, and specific method for the determination of opiates and cocaine and metabolites in urine, plasma, and blood was established. A one-step extraction followed by liquid chromatography-electrospray ionization tandem mass spectrometry operating in multiple reaction monitoring mode was used. Two chromatographic runs were performed, each in less than 6 rain. The lower limit for accurate quantitative determination was 5 pg/L for cocaine and metabolites and 10 pg/L for opiates. Linearity was obtained from 10 to 1000 pg/L. Intraday (n = 6) and interday (n = 6) precisions and recoveries (n = 6) were determined at 10 or 25, 100, and 1000 pg/L concentrations. Precisions with a coefficient of variation less than 15% were obtained. Recoveries between 85 and 115% were determined.
Introduction The use of unspecific immunological screening for drugs-ofabuse testing of biologicalsamples has led to a need for specific, sensitive, and simple confirmation methods to distinguish illegal from legal use of drugs. Opiates are a group of compounds including drugs of abuse (heroin) and legal drugs (morphine, codeine, codethyline, pholcodine). Their metabolism gives the same compound (morphine). Heroin is rapidlydeacetylatedto its active metabolite 6-monoacetylmorphine(6-MAM),which is then deacetylatedto morphine. Heroin use is detected by the presence of 6-MAMand morphine in biologicalfluids. Immunoassay is not able to differentiatebetween opiates, and confirmation of positive results requires use of mass spectrometric methods with identification and quantitation of each opiate. Several gas chromatographic-mass spectrometric (GC-MS) (1-10) methods have been developed for the analysis of opiates. These methods provide specificityand high selectivitybut need time-consuming extrac-
tion, purification, and derivatizationprocedures. Recently,some authors (11,12) have described liquid chromatography-mass spectrometry (LC-MS)methods for morphine determination. Cocaine (benzoylmethylecgonine)is a psychotropicdrug with a long history of human consumption. Cocaine (13) is rapidly metabolized in the body to two major metabolites, benzoylecgonine (BZE)and methylecgonine(ME),and to a lesser extent, norcocaine and ecgonine derivatives. The coabsorption of cocaine and alcohol produces another cocaine analogue, cocaethylene (CE). A pyrolysis product, anhydromethylecgonine (AME), is formed when cocaine is smoked. The most commonly used method for cocaine testing is benzoylecgonine detection by immunoassay. Although this method is reliable, all positive results must be confirmed using a nonimmunological method generally based on chromatographic separation combined with specific detection (14-20). Various chromatographic methods involvinghigh-performanceliquid chromatography(HPLC)with either IN or direct fluorimetric detection and GC-MS coupling have been described.This last method is very sensitive but time consuming because it often requires a preliminaryderivatization of analytes. The method presented here describesthe use of a liquid chromatograph-electrospray tandem mass spectrometer (LC-ESIMS-MS) as a confirmatorytool for the presence of drug of abuse. Electrospray ionization is a gentle process involvingprotonation at a basic functional group in a molecule.The protonated molecular ion is usually the predominant ion, and limited fragmentation is observed. In quadrupole MS-MS, the protonated molecular ion is selectedby the first quadrupoleQ1. Formation of fragment ions occurs in the collisioncell Q2 of the MS. Fragment ions are monitored by the last quadrupole Q3. This analytical technique, which consists of electrospray ionization followedby selection of a molecular ion/fragment ion transition for each drug, offersvery high selectivityin drug analysis. We report here a rapid, simple, sensitive,and specificLC-ESIMS-MS analysis of opiates and cocaine and metabolites using specifictransition for each compound of interest.
*Authorto whom correspondenceshouldbe addressed.
620
Reproduction(photocopying)of editorialcontentof thisjournalis prohibitedwithoutpublisher'spermission.
Journal of Analytical Toxicology, Vol. 23, November/December 1999
Materials and Methods Materials Morphine, 6-MAM,pholcodine, codeine, norcodeine,codethyline, the internal standard nalorphine, cocaine,benzoylecgonine, methylecgonine, and cocaethylene were purchased from Sigma Chemicals (St. Louis, MO). AME (1 g/L in acetonitrile) and trideuterated analogues of BZE (100 mg/L in methanol) and ME (100 mg/Lin acetonitrile)were obtainedfromRadianPromochem. All solvents (HPLCgrade) and other chemicals (formic acid, ammonium formate, ammonia) were provided by Merck (Darmstadt, Germany).
5.5e5
A
28 ~.2
;.2 1.2e5'
5.0e5
1.1e5
Drug-freeurine, plasma, and blood were collectedfrom healthy subjects. Stock solutions of each drug (1g/L) were prepared in methanol and stored at -20~ A stock mixture solution containing either opiates or cocaine and metabolites, each at 100 mg/L in methanol, was prepared and stocked in the same conditions. Working standards were prepared with deionized water from methanolic stock solutions to obtain concentrations of 0.1, 0.25, 1, 2.5, and 10 mg/L in analytes. The internal standard solution containing nalorphine and trideuterated analogues of BZE and ME, each at 2.5 rag/L, was prepared in methanol and stored at-20~ For the validation of the method, the samples were prepared by adding the working standards to obtain a concentration of each drug at 0 to 1000 B 1Jg/Lfor the linearity study and 10 or 25, 100, and 10001Jg/L for the other studies. Each sample consisted of 250 I~L.
4.5e5 1.0e5 4.0e5 9.0e4
3.5e5 r162
3.0e5 e.
2.5e5
"~ O. 0
8.0e4
l~
7.0e4
r cr .~
6.0e4, 5.0e4,
2.0e5
4.0e4 1.5e5 3.0e4 1.0e5 2.0e4 5.0e4,
I .Oe4
..........
Ill.~l J~uJ.L.._
2~o nVz (amu)
I, I lira.NrlIIII~LIIII..L~ I bill .l,lllllllld YI~'iII'1I,lIllt, i
J. "
"
200 m/z (amu)
Figure 1. Electrospray ionization mass spectra of morphine: A, MS and B, MS-MS.
Table I. MRM Transitions for Opiates and Cocaine and Metabolites Protonatedmolecularion
Fragmention
Compound
(amu)
(amu)
Morphine Codeine 6-Monoacetylmorphine Pholcodine Norcodeine Codethyline Nalorphine Cocaine Benzoylecgonine Methylecgonine Cocaethylene Anhydromethylecgonine
286.2 300.2 328.2 399.2 286.2 314.2 312.2 304.2 290.2 200.2 318.2 182.2
165.1 165.1 165.1 114.1 165.1 165.1 165.1 182.2 168.2 182.2 196.2 122.2
Instrumentation Analysiswas performed on a triple quadrupole AP1300 Perkin Elmer SCIEX(Thornhill, Canada) MS equipped with an atmospheric pressure ionization source via an ionspray interface. For LC, a 200 binary pump and an autosampler ISS 200 Perkin Elmer were used. Drugs were separated on a spherisorb 5 RP 8S (100 x 2.l-ram i.d., 5 lJm, AppliedBiosystems).The mobile phase was a mixture of water and acetonitrile containing 0.1% HCOOHand 2mM of HCOONH4.The proportion of water/acetonitrilewas 80:20 for opiates and was slightly modifiedto 50:50 for cocaine and metabolites. Solvent flowwas set at 400 IJL/min. MS setting was optimized for opiates and for cocaine and metabolites to give optimum ion yield. Ionization of analytes was obtained in positive mode. For opiates, the electrospray voltage was adjusted to 5.2 kV,the orifice ring to 60 V,and the fragmentation energy to 50 eV. For cocaine and metabolites, the electrosprayvoltage was set at 5.4 kV,the orifice ring at 20 V,and the fragmentation energy at 27 eV. Fragmentation was obtained using nitrogen as collision gas. The MS was operated in the multiple reaction monitoring mode (MRM).
Samplepreparation To 250 lJL of urine, plasma, or bloodwere added 25 ~L of a solution of internal standards (nalorphine, trideuterated BZE, and trideuterated ME) at 2.5 rag/L, 100 IJLof ammonia buffer (1M,pH 9), and 1.25 mL of an organic phase consisting of CHClJisopropanol (95:5,v/v).After agitation on a rotative shaker for 10 rain and centrifugation,the organic phase was evaporated to dryness. Disposable 2-mL microvialswere used for sample preparation. The residue was dissolvedin 100 IJL of water/acetonitrile (5:1, v/v), and 20 IJL was 621
Journalof Analytical Toxicology,Vol. 23, November/December 1999
39~ 2.206'
A
14.2
B
2.106"
8.005
injected into the LC-MS-MS. Two chromatographic runs were operated, one with opiate conditions, the other with cocaine conditions.
2.0065.505
1.906" 1.806-
5.005
Results and Discussion
1.7061.608-
4.505
1.5061.408-
4.005 A
1.3o6
I:1. rJ
3.508
"6 J
"6 r
3.005
_=
_a
1.2o6 1.1o6 1,006 9.005
2.805
8.005
2.~8.
7.006 8.005
1.808.
S.005 4.008,
1.005' 3.005' 2.008'
8.004,
1.0o5' 200
400
200
m/z (ainu)
m/z (ainu)
Figure 2. Electrosprayionization massspectraof pholcodine: A, MS and B MS-MS.
3000 2800" 2600, 2400' 2200' 2.0
2000" 1800' vO
~
1600'
C ,~O 1400.
_=
12001000. 800600" 400200"
\^
~, . . . . . . . 3
Time (mln) Figure 3. MRM chromatogramof morphine (1.3 min) and norcodeine(2.0 min). The m/z 286.2 ion was isolated and collisionally dissociatedto produce m/z 165.1.
622
Electrospray MS-MS Opiates and cocaine mass spectra were obtained in positive ionization mode. The electrospray process was able to transform the molecules from solution into ions in the gas phase. This ion evaporation produced for each compound an abundant protonated molecular ion (MH§ For each analyte, this ion (parent ion) is selected in the first MS Q1 and fragmented in the collision cell Q2 with nitrogen gas. Generation of fragments required a collision energy greater for the protonated opiates (50 eV) than for cocaine and metabolites (27 eV). One resulting product ion (daughter ion) for each compound of interest is monitored in the second MS Q3. MRM provides less specificity than a complete MS-MS spectrum but increases specificity over a single stage of MS analysis. For each analyte, generation of an MRMsignal at the detector requires that two criteria be met: the mass-to-charge ratios of both the parentand the daughter ions must satisfy the selected values. The MRM experiments used for detection and quantitation are reported in Table I. Fragmentation of protonated opiates produced the same 165.1 amu fragment ion except pholcodine, which fragmented in an ion at 114.1 ainu. Fragmentation of cocaine and ME protonated molecular ions produced an abundant ion at 182.2 amu. BZE, CE, and AME protonated molecular ions fragmented at 168.2, 196.2, and 122.2 amu, respectively. Examples of MS spectrum obtained in single and in tandem MS are given in Figures 1 and 2. Chromatography As opiates and cocaine and metabolites are not eluted using the same mobile phase, two specific chromatographic runs were performed on batches of samples. Chromatographic elution of opiates was achieved in 5 rain. Separation of the different analytes was not obtained, and some compounds, codeine, norcodeine, and nalorphine, coeluted using the chromatographic conditions described. However, because specific detection was used, individual MRM chromatograms showed any interference for these compounds. Two analytes, morphine and norcodeine, were monitored using the same transition286.2/165.1
Journal of Analytical Toxicology, Vol. 23, November/December 1999
amu, so they needed baseline chromatographic resolution for their identification as shown in Figure 3. Chromatographic elution of cocaine and its metabolites was obtained in 6 rain. Cocaine, BZE, and trideuterated BZE and ME and trideuterated ME coeluted and were differentiated using their specific transition.
Validation Opiates quantitation was performed using nalorphine as internal standard. Cocaine, BZE, and CE quantitation and AME and ME quantitation were obtained using trideuterated BZE and trideuterated ME as internal standards, respectively. Because urine samples have heterogenous composition, an artificial matrix was prepared consisting ofa 1 g/L creatinine and 9 g/L NaC1 water solution. Linearity, limit of quantitation, intraday and interday precisions were studied with this matrix instead of urine. For recovery, drug-free urine, plasma, and blood were used. Validation of the procedure is reported in Table II. Linearity The linearity of the method was tested at seven concentrations: 0, 10, 25, 50, 100, 200, 500, and 1000 Ug/L. Greater values were not assayed. Samples with concentrations greater than 1000 IJg/L were diluted before extraction. The procedure exhibits linearity for all compounds except AME, which is not correctly extracted using our conditions. Correlation coefficients greater than 0.995 have been obtained for the calibration curves of each analyte. Limit of quantitation This limit has been established using drug-free samples to which opiates and cocaine and metabolites were added to
obtained concentrations of I to 10 Ug/L in analytes. Six extractions of the same spiked matrix sample were analyzed. The quantitation limit was found to be 5 ug/L for cocaine and metabolites and 10 Ug/L for opiates, based on a coefficient of variation lower than 20%. Limit of quantitation could be improved by increasing the sample volume, but 250 IJL sample size was chosen in order to use disposable microvials and to avoid problems due to possible sample contaminations.
Precision ~vo parameters of precision were investigated: intraday and interday precision. Interday (n = 6) and intraday (n = 6) precisions were obtained on 5-mL samples spiked at three levels of concentration (10 or 25, 100, and 1000 Ug/L).The mean values, standard deviations, and variation coefficients are shown in Table II. Recoveries Recoveries were tested on six different urine, plasma, and blood samples. Recoveries between 85% and 115% were obtained for all compounds except for pholcodine and AME.
Conclusions Reliable monitoring of drugs of abuse requires a specific, sensitive, rapid, and robust analytical procedure. The LC-ESI-MS-MS method reported here meets all of these criteria. The one-step extraction is very easy and rapid and does not need any additional purification or derivatization. Each chromatographic run takes only 5-6 rain. Automatic injections are possible onto the
Table II. Intraday and Interday Precision for Cocaine and Metabolites and for Opiates
Intraday precision* spike levels Compound
10 (25) + pg/L
100 pg/L
1000 pg/L
10 (25)+pg/L
Interdayprecision*spikelevels 100 pg/L
Morphine
26.8 • 1.9 7.2 22.3 + 2.3 10.5 24.7 • 2.0 8.0 21 + 1.1 5.2 25.3 • 1.2 4.8 22 • 1.1 5.0 10.9 _+0.6 5.3 9.450.3 3.0 9.7 + 1.6 16.5 10.3 • 0.5 4.4
92.7 _+5.5 6.0 87 __.2.2 2.5 99.8 • 2.5 2.6 87.3 • 6.2 7.2 97.5 + 3.7 3.8 85.7 • 3.6 4.2 99.2 • 7.1 7.1 98.9 • 1.5 1.5 97.9 • 2.6 2.7 98.7 • 7.8 7.1
949 + 60 6.3 870 +_63 7.3 976 • 36 3.1 896 • 42 4.7 926 • 50 5.4 829 • 36 4.3 1037 • 33 3.1 993 • 17 1.8 1046 + 13 1.2 1040 _+38 3.7
28.2 + 2.6 9.4 23.6 + 2.9 12.2 23.8 • 3.4 14.1 20.7 • 3.7 17.8 24.3 + 2.2 8.9 23 • 2.6 11.3 10.0 • 0.9 9.1 9.1 • 1.0 11.3 9.8 • 1 11.3 9.6 • 0.7 7.4
104.8 _+10.4 9.9 94.5 • 11.4 12.0 94.3 + 5.2 5.5 89.5 + 11.2 12.5 98.5 -+ 5.9 5.9 90.8 • 7.4 8.1 102.6 • 6.2 6.1 96.2 • 8.2 8.5 100.6 • 1.6 1.6 100.4 • 4.4 4.4
Codeine 6-MAM Pholcodine Norcodeine Codethyline Cocaine 8ZF ME CE
1000 pg/L 970 + 75 7.7 948 + 54 5.7 932 -+ 53 5.7 996 • 147 14.2 923 + 76 8.2 889 • 75 8.5 1053 • 75 7.1 988 • 59 6.0 1038 + 68 6.6 1046 • 66 6.3
* Mean plus or minus standarddeviation concentration and percentcoefficient of variation; n = 6. Lowest level spike for opiates is 25 pg/L.
623
Journal of Analytical Toxicology,Vol. 23, November/December1999
chromatographic column. The ability to determine each compound of interest using its protonated molecular ion, its major product ion, and its LC retention time simultaneously avoids interferants because coeluting and interfering matrix components are excludedfrom detection. This great specificitypermits reduction of the time required for sample preparation by eliminating tedious cleanup of extracts. Only two compounds, pholcodine and anhydroecgoninemethylester,are not well quantitated at low levels.But, because pholcodine is present in high concentration when absorbed for antitussive purpose, and because anhydroecgoninemethylester is only one of the metabolites obtained when cocaine is smoked, the described method can be routinely used for determination of drug addiction.
References I. E.J. Cone, W.D. Darwin, and W.R Buchwald. Assay for codeine, morphine, and ten potential urinary metabolites by gas chromatography-mass fragmentometry. J. Chromatogr. 275:307-318 (1983). 2. A. Solans, R. de la Torre, and I. Segura. Determination of morphine and codeine in urine by gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 8:905-909 (1990). 3. G.F. Grinstead. A closer look at acetyl and pentafluoropropionyl derivatives for quantitative analysis of morphine and codeine by gas chromatography/mass spectrometry. J. AnaL ToxicoL 15:293-298 (1991). 4. E.J. Cone and W.D. Darwin. Rapid assay of cocaine, opiates and metabolites by chromatography-mass spectrometry. J. Chromatogr. 580:43-61 (1992). 5. D.C. Fuller and W.H. Anderson. A simplified procedure for the determination of free codeine, free morphine and 6-acetylmorphine in urine. J. Anal. Toxicol. 16:315-318 (1992). 6. W. Huang, W. Andolino, and W.L. Hearn. A solid phase extraction technique for the isolation and identification of opiates in urine. J. AnaL Toxicol. 16:307-310 (1992). 7. H. Maurer and K. Pfleger. Screening procedure for the detection of opioids, other potent analgesics and their metabolites in urine using a computerized gas chromatographic-mass spectrometric technique. J. Anal. Chem. 317:42-52 (1994). 8. R. Wasels and F. Belleville. Gas chromatographic-mass spectrometric procedures used for identification and determination of morphine, codeine and 6-monoacetylmorphine. J. Chromatogr. A 674: 225-234 (1994). 9. Y. Galliard, G. P~pin, P. Marquet, P. Kintz, M. Deveaux, and P. Mura. Identification et dosage de la benzoylecgonine, cocaine, m~thylecgonine-ester, codeine, morphine et 6-ac~tylmorphine dans le sang total. Toxicorama 8:17-22 (1996).
624
10. L.A. Broussard, L.C. Presley, T. Pittman., R. Clouette, and G.H. Wimbish. Simultaneous identification and quantitation of codeine, morphine, hydrocodone, and hydromorphone in urine as trimethylsilyl and oxime derivatives by gas chromatography-mass spectrometry. Clin. Chem. 43:1029-1032 (1997). 11. N. Tyrefors, B. Hyllbrant, L. Ekman, M. Johansson,and B. LangstrOm. Determination of morphine, morphine-3-glucuronide and morphine-6-glucuronide in human serum by solid-phase extraction and liquid chromatography--massspectrometrywith electrospray ionization. J. Chromatogr. A 729:279-285 (1996). 12. M.J. Bogusz, R.D. Maier, M. Erkens,and S. Driessen. Determination of morphine and its 3- and 6-glucuronides, codeine, codeine- glucuronide and 6-mono-acetylmorphine in body fluids by liquid chromatography atmospheric pressure chemical ionization mass spectrometry. J. Chromatogr. 703:115-117 (1997). 13. N. Tyrefors, B. Hyllbrant, L. Ekrnan,M. Johansson,and B. Langstr0m. Determination of morphine, morphine-3-glucuronide and morphine-6-glucuronide in human serum by solid-phase extraction and liquid chromatography-mass spectrometry with electrospray ionisation. J. Chromatogr. A 729:279-285 (1996). 14. A.J. Jenkinsand E.J.Cone. Pharmacokinetics:drug absorption, distribution, and elimination. In Drug Abuse Handbook, S.B. Karch, Ed. CRC Press,Boca Raton, FL, 1998, pp 184-187. 15. J.O. Svensson.Determination of benzoylecgonine in urine from drug abusers using ion pair high performance liquid chromatography. J. Anal. Toxicol. 10:122-124 (1986). 16. C.L. Williams, S.C. Laizure, R.B.Parker, and J.J.Lima. Quantification of cocaine and cocaethylene in canine serum by high-performance liquid chromatography. J. Chromatogr. B 681: 271-276 (1996). 17. L. Virag, B. Mets, and S. Jamdar. Determination of cocaine, norcocaine, benzoylecgonine and ecgonine methyl ester in rat plasma by high-performance liquid chromatography with ultra-violet detection. J. Chromatogr. B 681:263-269 (1996). 18. F. Tagliaro, C. Antonioli, Z. De Battisti, S. Ghielmi, and M. Marigo. Reverse-phasehigh performance liquid chromatography determination of cocaine in plasma and human hair with direct fluorimetric detection. J. Chromatogr. A 674:207-215 (1994). 19. M.R. Harkey, G.U Henderson, and C. Zhou. Simultaneous quantification of cocaine and its major metabolites in human hair by gas chromatography/ionization massspectrometry. J. Anal. Toxicol. 15: 260-265 (1991). 20. E.J.Cone, M. Hitlsgrove, and W.D. Darwin. Simultaneous measurement of cocaine, cocaethylene, their metabolites and "crack" pyrolysis products by gas chromatography-mass spectrometry. Clin. Chem. 40:1299-1305 (1994). 21. R Kintz, C. Sengler,V. Cirimele, and R Mangin. Evidenceof crack use by anhydroecgonine methylester identification. Hum. Exp. Toxicol. 16:123-127 (1997).
Manuscript received December 3, 1998; revision received February 10, 1999.
