Characterization of the stereoselective biotransformation of ketamine to norketaminevia...

Regula Theurillat1

Monika Knobloch2

Olivier Levionnois2

Paula Larenza2

Meike Mevissen3

Wolfgang Thormann1

1Department of ClinicalPharmacology,University of Bern,Bern, Switzerland

2Department of ClinicalVeterinary Medicine,University of Bern,Bern, Switzerland

3Division of VeterinaryPharmacology & Toxicology,University of Bern,Bern, Switzerland

Characterization of the stereoselectivebiotransformation of ketamine to norketamine viadetermination of their enantiomers in equineplasma by capillary electrophoresis

A robust CE method for the simultaneous determination of the enantiomers of keta-mine and norketamine in equine plasma is described. It is based upon liquid–liquidextraction of ketamine and norketamine at alkaline pH from 1 mL plasma followed byanalysis of the reconstituted extract by CE in the presence of a pH 2.5 Tris-phosphatebuffer containing 10 mg/mL highly sulfated b-CD as chiral selector. Enantiomer plasmalevels between 0.04 and 2.5 mg/mL are shown to provide linear calibration graphs.Intraday and interday precisions evaluated from peak area ratios (n = 5) at the lowestcalibrator concentration are , 8 and , 14%, respectively. The LOD for all enantiomersis 0.01 mg/mL. After i.v. bolus administration of 2.2 mg/kg racemic ketamine, the assayis demonstrated to provide reliable data for plasma samples of ponies under isofluraneanesthesia, of ponies premedicated with xylazine, and of one horse that receivedromifidine, L-methadone, guaifenisine, and isoflurane. In animals not premedicatedwith xylazine, the ketamine N-demethylation is demonstrated to be enantioselective.The concentrations of the two ketamine enantiomers in plasma are equal whereasS-norketamine is found in a larger amount than R-norketamine. In the group receivingxylazine, data obtained do not reveal this stereoselectivity.

Keywords: Capillary electrophoresis; Chirality; Enantiomers; Enantioselective N-demethylation;Isoflurane; Stereoselective metabolism; Sulfated b-cyclodextrin; Xylazine

DOI 10.1002/elps.200500059

1 Introduction

Ketamine ((R,S-2-(2-chlorophenyl)-2-methylamino)cyclo-hexanone, for chemical structure see Fig. 1) is an intrave-nous analgesic and dissociative anesthetic drug widelyused in clinical practice of man and animals. Due to itshallucinogenic effects at subanesthetic doses, effectsthat are similar to those of phencyclidine, it is also subjectof abuse and is taken intranasally, injected, smoked, oringested after mixing into drinks. The pKa value of keta-mine is 7.5 and it is thus partially positively charged atphysiological pH. The partition coefficient expressed asthe log P (octanol/water) value is 3.1. In humans, theplasma half-life of the unchanged drug is about 3 h and20–50% of it is bound to plasma proteins [1, 2]. Ketamineis chiral and exists as two enantiomers (Fig. 1). It is inter-

Figure 1. Chemical structures of (A) S-ketamine(R = CH3) and S-norketamine (R = H) and (B) R-ketamine(R = CH3) and R-norketamine (R = H) presented as mirrorimages.

esting to note that this drug belongs to the rather fewexamples of chemical compounds for which the sign ofrotation reverts when the free base is transformed into thehydrochloride salt with preservation of the absolute con-figuration. For the free base, (2)-ketamine and (1)-keta-mine were found to possess the S- and R-configuration,respectively, whereas for the hydrochloride salt, (1)-keta-mine and (2)-ketamine have S- and R-configuration [3].Although S-ketamine is more potent than R-ketamine andexhibits a greater clearance and faster anesthetic recovery

Correspondence: Professor Wolfgang Thormann, Departmentof Clinical Pharmacology, Murtenstrasse 35, CH-3010 Bern,SwitzerlandE-mail: [email protected]: 141-31-632-4997

Abbreviations: sulfated-�-CD, highly sulfated b-CD; IST, inter-nal standard

3942 Electrophoresis 2005, 26, 3942–3951

© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Electrophoresis 2005, 26, 3942–3951 Enantioselective analysis of ketamine and norketamine in plasma 3943

compared to the racemate, it is typically administered asracemate with both pharmacokinetic and pharmacody-namic enantioselectivity. The in vivo and in vitro metabo-lisms of ketamine have been studied in humans and var-ious animal species. Ketamine undergoes extensivehepatic first-path metabolism, with the major pathway in-volving N-demethylation to norketamine (for chemicalstructure see Fig. 1), a chiral compound that may alsocontribute to the analgesic effects following ketamineadministration. In humans, the CYP3A4, CYP2B6, andCYP2C9 are believed to be the cytochrome P450 enzymesinvolved in the oxidative N-demethylation of ketamine.Other metabolites formed include dehydronorketamineand various free and glucuronidated hydroxylated deriva-tives of ketamine and norketamine [4–7].

Although used extensively in horses, the pharmacokineticdata and clinical analgesic effects of S- and R-ketamineare poorly documented for this species [8]. Furthermore, inanalogy to the findings in human medicine, application ofS-ketamine instead of racemic ketamine could be of greatclinical interest as treatments with S-ketamine revealedmilder psychological side effects and more rapid recoveryof psychomotor function and performance. Researchefforts at our University are currently under way to assessthe use of racemic ketamine versus S-ketamine for anes-thesia and pain therapy in horses, work that requires themonitoring of plasma levels of ketamine and norketamineenantiomers. As of today, ketamine and norketamineenantiomers in human and animal plasma have mainlybeen determined by HPLC using expensive, enantioselec-tive columns that are based upon immobilized a1-acid gly-coprotein (AGP) [8–11] or cellulose tris(3,5-dimethylphe-nylcarbamate)-coated silica-gel (Chiralcel OD) [12]. Anenantioselective GC method that features a b-CD-basedchiral stationary phase has also been reported [13].

During the past few years chiral separations by capillaryelectromigration methods have been studied extensivelyand were successfully applied to bioanalytical drug mon-itoring [14–16]. For enantiomeric separation under electro-kinetic conditions, a chiral selector (such as a CD, a crownether, a protein, or a bile acid, to name buta few) and properbuffer conditions (pH, ionic strength, solvent, additives,etc.) are required. CE represents an accurate technologywhich has found its place in a cost- and environment-con-trolled age. Compared to enantioselective HPLC, CE pro-vides higher efficiency and is simpler, faster, and cheaper,and consumes a much smaller amount of organic solvents.Enantioselective CE not only represents a complementarytool to the widely applied chromatographic methods, it alsooffers the possibility of bringing chiral separations andanalyses into the routine arena. Enantiomeric separationsof ketamine by CE in the presence of various native and

modified CDs have been studied in many laboratories,work in which ketamine was employed as model drug [3,17–23]. Negatively charged CDs have been found to betterresolve the ketamine stereoisomers than neutral CDs, thehighest resolution being reported for highly sulfated b-CD(sulfated-b-CD) in an acidic buffer [23]. No paper describ-ing the simultaneous CE separation of the enantiomers ofketamine and norketamine and no report about the CEanalysis of norketamine and/or ketamine enantiomers inbiological samples were found. Thus, the use of CE withsulfated-b-CD for the separation of ketamine and norketa-mine enantiomers and their analysis in equine plasma wasinvestigated. In this paper, the conditions and specifica-tions of an enantioselective CE-based assay for theseenantiomers in equine plasma is reported. The method isshown to be robust, to be suitable to measure enantiomerdrug levels in presence of xylazine, and to reveal totalketamine and norketamine concentrations that comparewell to those determined by CE in absence of the chiralselector. Furthermore, with the analysis of ketamine andnorketamine enantiomers in the blood of a horse and sev-eral ponies, the assay is demonstrated to provide insightinto the enantioselective N-demethylation of ketamine.

2 Materials and methods

2.1 Chemicals and standard solutions

All chemicals used were of analytical grade. Racemicketamine hydrochloride was obtained from the pharmacyof the Inselspital (Bern, Switzerland), and racemic norket-amine hydrochloride was purchased as methanolic solu-tion (1 mg/mL of the free base) from Cerilliant (Round Rock,USA). Lamotrigine was obtained from The WellcomeFoundation (London, UK). Tris, HCl (37%), and isopropanolwere from Merck (Darmstadt, Germany), H3PO4 (85%) andethylacetate were from Fluka (Buchs, Switzerland), anddichloromethane was purchased from Biosolve (Val-kenswaard, The Netherlands). Sulfated-b-CD (7–11 molsulfate/mol b-CD, batch 07222HO with an optical rotationof 69.47) was obtained from Sigma-Aldrich Chemie(Schnelldorf, Germany). An aqueous stock solution ofketamine hydrochloride (250 mg/mL) and the methanolicsolution of norketamine were used to prepare a standardsolution containing 25.00 mg/mL of each norketamineenantiomer and 21.67 mg/mL of each ketamine enantio-mer. An aqueous solution of lamotrigine (125 mg/mL) wasemployed as internal standard solution. Bovine plasmawas received from cattle at the local slaughter house and itwas used for the preparation of calibration and controlsamples. Drugs i.v. administered to the animals includedracemic ketamine (Ketasol 100®, solution containing100 mg/mL ketamine hydrochloride), S-ketamine (not


CE

and

CE

C

3944 R. Theurillat et al. Electrophoresis 2005, 26, 3942–3951

commercially available in Switzerland, 60 mg/mL S-keta-mine hydrochloride), and xylazine (Xylasol®, 20 mg/mLxylazine hydrochloride), preparations that were obtainedfrom Dr. E. Gräub (Bern, Switzerland).

2.2 Samples

Blood samples analyzed stemmed from an adult patienthorse that received an i.v. bolus of 2.2 mg/kg racemicketamine (premedication with i.v. administration ofL-methadone (50 mg/kg), romifidine (50 mg/kg), and guaife-nisine (about 50 mg/kg) 10 min before ketamine adminis-tration and isoflurane anesthesia that was commencedabout 1 min after ketamine administration) and from sixhealthy Shetland ponies that received either an i.v. bolusof 2.2 mg/kg racemic ketamine or 1.1 mg/kg S-ketamineunder stable anesthesia with isoflurane in oxygen at eachindividual monitored anesthesia care (MAC) [24] or 10 minafter premedication with an i.v. bolus of 1.1 mg/kg xyla-zine. Plasma samples were frozen at 2187C until furtheruse. For the preparation of calibrators, 2, 10, 20, 50, and100 mL aliquots of the standard solution were added to1 mL of bovine plasma producing samples with 0.05,0.25, 0.50, 1.25, and 2.50 mg/mL of each norketamineenantiomer, respectively (0.04, 0.22, 0.43, 1.44, and2.17 mg/mL of each ketamine enantiomer, respectively).The fortified bovine plasma calibrators were made freshlyfor each set of analyses. Controls were prepared batch-wise. Fifty milliliters of a control was produced via com-bining an appropriate amount of an independently pre-pared 0.2 mg/mL racemic ketamine hydrochloride stand-ard solution and the methanolic racemic norketaminesolution with bovine plasma. Aliquots of about 1 mL werefrozen in plastic vials at 2207C. All frozen samples wereslowly defrosted and vortex mixed prior to analysis.

2.3 Sample preparation

The freshly prepared calibrators, defrosted controls(1 mL), and animal plasma samples (1 mL) were pipettedinto 10 mL glass tubes and combined with 20 mL of theinternal standard solution, 100 mL of 1 M NaOH, and 6 mLof a dichloromethane/ethylacetate (75/25% v/v) solventmixture. The tubes were closed, shaken for about 10 min,and centrifuged at about 15006g for 5 min. The aqueousphase (upper phase) and the protein cake were carefullyremoved with a glass pipette (under vacuum) and theorganic phase was decanted into a glass tube with arounded bottom. After acidification with 25 mL of con-centrated HCl, the organic solvent was evaporated undera gentle stream of air at 377C and the residue was redis-solved in 150 mL of tenfold diluted, pH 2.5 Tris-phosphatebuffer that did not contain the chiral selector (see below).

2.4 Electrophoretic instrumentation andrunning conditions

A P/ACE MDQ CE system (Beckman Coulter, Fullerton,CA, USA), equipped with an untreated fused-silica cap-illary (Polymicro Technologies, Phoenix, AZ, USA) of50 mm ID and 38.5 cm (28 cm to the detector) totallength, was used. Samples were introduced from 0.5 mLpolypropylene vials by applying a positive pressure of1 psi (1 psi = 6894.76 Pa) for 5 s. The temperature controlof the circulating cooling fluid of the capillary cartridgewas set to 30.07C and the temperature of the sample traywas kept at 20.07C. The applied voltage was 220.0 kV(reversed polarity; current: about 245 mA) and the runtime was 10 min. The PDA detector at 195 nm wasemployed for solute detection. If not stated otherwise,the buffer was composed of 50 mM Tris, phosphoric acid(pH 2.5), and 10 mg/mL sulfated-b-CD. Fresh buffer wasprepared every day. Solute quantitation was based uponinternal calibration using peak areas for data evaluation.In the morning, the capillary was rinsed with 0.1 M NaOHand bidistilled water (10 min each). Before each experi-ment, the capillary was sequentially rinsed with 0.1 M

NaOH (1 min), water (1 min), and running buffer (2 min).For achiral analyses, a 50 mm ID capillary of 50.5 cm(40 cm to the detector) total length was used. The volt-age was 12 kV (normal polarity; current about 8.1 mA) andthe run time was 8.5 min. The buffer was composed of50 mM Tris, phosphoric acid (pH 6.3), and 5% v/v iso-propanol. All other conditions were the same as for thechiral assay.

2.5 Statistical data treatment and graphicalpresentations

Statistical analyses were performed using the Sigma-Stat for Windows version 1.0 (Jandel, Corte Madera,CA, USA). Input groups were compared with the Mann–Whitney rank sum test. Graphical presentations weremade with SigmaPlot for Windows version 2.01 (Jan-del). Chemical structures were drawn with MDL ISIS/Draw 2.5 (MDL Information Systems, San Leandro, CA,USA).

3 Results and discussion

3.1 Analysis of ketamine and norketamineenantiomers in plasma extracts

Enantiomeric separation of ketamine by CE in the pres-ence of CDs as chiral selectors has been demonstrated inmany laboratories. Neutral CDs were reported to providemodest resolution of ketamine enantiomers [17–22].



Negatively charged CDs were found to better resolvethese stereoisomers, the highest resolution being report-ed for sulfated-b-CD in an acidic Tris-phosphate buffer[23]. Thus, the use of this configuration was investigatedin our laboratory to separate the enantiomers of ketamineand norketamine in plasma extracts. Employing a pH 2.6phosphate buffer (without Tris) containing 5 mg/mL sul-fated-b-CD in an untreated fused-silica capillary andreversed polarity revealed excellent separation of theenantiomers of ketamine and norketamine, indicating thatthe migrating drug complexes have a negative charge(strong complexation). Under these conditions, xylazinewas found to comigrate with the firstly detected ketamineenantiomer (S-ketamine, see below). With a 50 mM Tris-phosphate buffer (pH 2.5), xylazine was determined tomigrate in front of S-ketamine. Furthermore, an increaseof the CD concentration led to reduced detection timesand improved the peak shapes. In order to reach runtimes of 10 min, 10 mg/mL sulfated-b-CD had to beemployed. Typical electropherograms obtained withplasma extracts are presented in Fig. 2. Graph A of Fig. 2illustrates that there is no interference in the time windowin which the compounds of interest are detected. Panel Bshows data obtained with blank bovine plasma that wasfortified with 1.25 mg/mL of each norketamine and1.08 mg/mL of each ketamine enantiomer, and panel Crepresents data from a pony plasma sample that wascollected 8 min after bolus injection of 2.2 mg/kg racemicketamine and premedication with isoflurane. Panel Ddepicts the temporal behavior of the current of theexperiment of panel C.

Peak assignment for the ketamine enantiomers was pos-sible via reanalysis of a plasma extract that was spikedwith the S-ketamine injection solution (increase of firstketamine peak, data not shown) and via analysis of aplasma collected from a pony that was treated withS-ketamine (Fig. 3, panel A). Reanalysis of the latterextract after fortification with racemic ketamine andracemic norketamine (Fig. 3, panel B) and the assumptionthat there is no stereochemical conversion during theN-demethylation of ketamine led to the identification of allfour enantiomers. For both compounds, the S-enantio-mers were detected ahead of the corresponding R-enan-tiomers.

Electropherograms obtained with plasma samples ofponies and one horse that received racemic ketaminerevealed two small peaks that are assumed to be keta-mine metabolites (see Figs. 2C (first peak shown only,marked with asterisk), 4B (peaks marked with asterisks)).Based on the work of Wainer et al. [4, 13], one couldassume that these are the peaks for S- and R-dehy-dronorketamine. Further evidence is given by the fact that

Figure 2. Typical electropherograms obtained for CEanalysis of extracts prepared from (A) blank bovineplasma, (B) blank bovine plasma fortified with 1.25 mg/mLof each norketamine and 1.08 mg/mL of each ketamineenantiomer, and (C) a pony plasma that was collected8 min after i.v. bolus injection of 2.2 mg/kg racemic keta-mine and premedication with isoflurane. (D) depicts thetemporal behavior of the current of (C). Concentrations ofS-ketamine, R-ketamine, S-norketamine, and R-norketa-mine in the pony plasma were determined to be 0.95,0.99, 0.45, and 0.35 mg/mL, respectively. IST refers tolamotrigine, the internal standard, that was added duringextract preparation. The asterisk in (C) marks a peak of anunidentified ketamine metabolite.

only the first peak is obtained for samples analyzed afteradministration of S-ketamine (Fig. 3A, peak marked withasterisk). It is important to note that the small peakdetected shortly before 10 min in the electropherogramspresented in Fig. 3 does not represent a ketaminemetabolite (see also Fig. 4B). Due to the unavailability of adehydronorketamine standard, however, the assignmentof the peaks could not be undertaken.



Figure 3. (A) CE data of an extract prepared from a ponyplasma that was collected 4 min after i.v. bolus injection of1.1 mg/kg S-ketamine and premedication with isoflurane.Concentrations of S-ketamine and S-norketamine weredetermined to be 0.89 and 0.23 mg/mL, respectively.(B) Data obtained after fortification of the plasma extractwith racemic ketamine (about 1.13 mg/mL) and racemicnorketamine (about 1.3 mg/mL each). IST refers to lamo-trigine, the internal standard, that was added duringextract preparation. The asterisk marks a peak of an un-identified ketamine metabolite.

3.2 Assay specifications

Recovery was assessed in triplicate analyses on two con-centration levels using control samples (fortified bovineplasma). One sample comprised ketamine and norketa-mine enantiomer drug levels of 0.22 and 0.25 mg/mL,respectively, the other one 1.73 and 2.00 mg/mL, respec-tively. For all four compounds and the IST, recoveries weredetermined to be between 70 and 80%. Table 1 sum-marizes the analytical characteristics of the five-level inter-nal calibration. All calibration graphs were found to be linear(r2 . 0.9929) with F values .400 (p , 0.001). Furthermore,the y-intercepts were observed to be significantly smallerthan the peak area ratio produced by the smallest calibratorand were negligible in most cases. The detection limit wasdetermined to be about 0.01 mg/mL and the LOQ was takenas 0.03 mg/mL. Assay precision data are summarized inTable 2. For an enantiomer concentration of about 0.25 mg/mL, intraday RSD values are shown to be between 5.1 and7.8% (n = 5), whereas corresponding interday values arebetween 11.1 and 13.5% (n = 5). For an enantiomer con-centration of around 2 mg/mL, all intraday and interday RSDvalues were found to be ,10% (n = 5). Detection timeswithin a day of analysis were found to be stable (RSD ,2%(n = 5) for all enantiomers and the internal standard). Smalldifferences of the chiral selector concentration in the run-ning buffer resulted in appreciable differences in detectiontimes from day to day (compare data ofFigs. 2–4). Based onthe internal calibration used and frequent renewal of thecalibration data, however, these variations did not have animpact on quantitation.

3.3 Determination of ketamine and norketamineenantiomers in the presence of xylazine andother drugs employed for premedication

Xylazine (N-(2,6-dimethylphenyl)-5,6-dihydro-4H-1,3-thiazine-2-amine) is an a-2 adrenergic agonist used assedative, analgesic, and muscle relaxant in veterinaryscience. In anesthesia, xylazine is often employed forinitial sedation of the animal prior to administration of

Table 1. Calibration data (n = 5)a)

Slope y-intercept r2 F value

Mean, /(mg/mL) RSD, % Mean RSD, % Mean RSD, % F p

S-Ketamine 0.731 5.17 0.036 97.5 0.997 0.28 .439 ,0.001R-Ketamine 0.761 3.38 0.031 113.7 0.996 0.30 .424 ,0.001S-Norketamine 0.899 6.59 0.045 117.9 0.996 0.28 .423 ,0.001R-Norketamine 0.971 7.60 0.040 143.2 0.997 0.15 .659 ,0.001

a) Concentration values and peak area ratios were taken as x-axis and y-axis, respectively.



Table 2. Precision data

Intraday (n = 5) Interday (n = 5)

Low High Low High

Ketamine enantiomers S/R S/R S/R S/RTarget concentration, mg/mL 0.22/0.22 1.73/1.73 0.22/0.22 1.73/1.73Mean, mg/mL 0.26/0.24 1.89/1.91 0.28/0.27 2.10/2.10RSD, % 5.08/7.80 1.67/1.45 13.32/11.11 8.53/6.82Norketamine enantiomers S/R S/R S/R S/RTarget concentration, mg/mL 0.25/0.25 2.0/2.0 0.25/0.25 2.0/2.0Mean, mg/mL 0.24/0.24 1.92/1.94 0.29/0.30 2.13/2.13RSD, % 6.59/5.39 4.28/5.82 13.45/12.87 7.99/8.78

Figure 4. Electropherograms of samples from a ponythat were collected (A) about 2 min after i.v. administrationof 1.1 mg/kg xylazine (8 min before ketamine injection)and (B) 4 min after i.v. bolus injection of 2.2 mg/kg racemicketamine. Concentrations of S-ketamine, R-ketamine,S-norketamine, and R-norketamine in the pony plasma ofpanel B were determined to be 0.47, 0.47, 0.49, and 0.45,respectively. Asterisks in (B) mark peaks of unidentifiedketamine metabolites.

Figure 5. Electropherograms of the achiral assay withextracts of (A) a blank bovine plasma sample fortified with1.73 mg/mL racemic ketamine and 2 mg/mL racemic nor-ketamine, (B) a pony plasma sample collected 2 min afteri.v. administration of 1.1 mg/kg xylazine, and (C) the ponyplasma collected 4 min after i.v. bolus injection of 2.2 mg/kg racemic ketamine. Concentrations of ketamine andnorketamine were determined to be 0.99 and 0.93 mg/mL,respectively. (D) depicts the temporal behavior of thecurrent of (C).



ketamine and thus can be present in blood samples dur-ing the initial phase of anesthesia with ketamine. Xylazineis a base that is extracted at alkaline pH in the same wayas ketamine. Not surprisingly, plasma extracts of samplesobtained from ponies that were pretreated with xylazinerevealed the presence of xylazine in the electro-pherograms. Using the enantioselective CE assay, xyla-zine was found to migrate as a rather broad peak justahead of S-ketamine (Fig. 4). Proper detection of S-keta-mine was thus somewhat hampered, but could beaccomplished without any difficulties as xylazine wasfound to become eliminated very quickly (compare dataof the two time points presented in Fig. 4). Furthermore,using the achiral assay (for conditions see Section 2.4),xylazine was detected as sharp, well separated peak infront of ketamine (Fig. 5). It is important to note thatpanels B and C of Fig. 5 were obtained with the sameextracts whose enantioselective data are depicted inpanels A and B of Fig. 4, respectively.

In order to prove the validity of the enantioselective datain presence of xylazine, drug levels of 46 plasma samplescontaining xylazine and ketamine were determined inboth assay formats and the results were compared sub-sequently. The comparative total ketamine and norketa-mine drug concentrations are presented in panels A andB, respectively, of Fig. 6. Good linear correlations withregression coefficients r of 0.993 and 0.972 for ketamineand norketamine, respectively, and with very few outlierswere obtained. The y-intercepts were determined to bevery small (0.039 and 0.008 mg/mL, respectively). Theslopes were calculated to be 0.978 and 0.942, respec-tively, indicating that drug levels determined with thenonchiral assay (mean drug levels of 1.55 and 0.73 mg/mL, respectively) were slightly smaller compared to those

obtained with the enantioselective method (mean druglevels of 1.62 and 0.76 mg/mL, respectively). Using theMann–Whitney rank sum test revealed that there is not astatistically significant difference between the compara-tive data sets (p values of 0.761 and 0.569, respectively).Similar comparative data were obtained for plasma sam-ples that did not contain any xylazine (data not shown).

The adult horse whose plasma samples were analyzedwas treated with L-methadone, romifidine, guaifenisine,and isoflurane (see Section 2.2). Analysis of the plasmaextracts with the chiral CE assay did not reveal any inter-ferences (panel A of Fig. 7) and electropherogramsobtained with samples that contained ketamine and nor-ketamine were comparable to those observed withponies that received isoflurane as premedication (com-pare panel C of Fig. 7 with panel C of Fig. 2). It is importantto note that the data of panel C were monitored on aseparate day. The detection time difference is due to asmall difference of the chiral selector concentration in therunning buffer (see above).

3.4 Enantioselectivity of ketamineN-demethylation

Ketamine and norketamine drug level data determined inthe plasma of a pony and a horse who received 2.2 mg/kgracemic ketamine are presented in Fig. 8. The data repre-sent cases with different premedications, namely, a ponypretreated with isoflurane (data of panels A and D), thepatient horse pretreated with L-methadone, romifidine, andguaifenisine that received isoflurane shortly after ketamineadministration (panels B and E), and the same pony pre-treated with xylazine (panels C and F). Ketamine drug

Figure 6. Comparative druglevels of (A) ketamine and (B)norketamine determined bychiral and nonchiral CE in 46samples drawn from ponies thatwere pretreated with xylazine.Solid lines represent correlationgraphs determined by linearregression analysis, the brokenlines describe the 95% predic-tion interval around the regres-sion line, and the dotted linesare the 95% confidence intervalfor the regression line.



Figure 7. Electropherograms of samples from an adultpatient horse that were collected (A) shortly before keta-mine injection (about 10 min after administration of pre-medication), (B) 2 min after i.v. bolus injection of 2.2 mg/kg racemic ketamine, and (C) 12 min after ketamineadministration. Concentrations of S-ketamine, R-keta-mine, S-norketamine, and R-norketamine in (B) weredetermined to be 2.25, 2.29, 0.28, and 0.25 mg/mL,respectively. Corresponding values for (C) are 0.68, 0.78,1.00, and 0.37 mg/mL, respectively. The asterisk in (C)marks a peak of an unidentified ketamine metabolite.

concentrations are shown to be highest at the first timepoint after drug injection and the peak concentration ofnorketamine was observed to occur around 4 min afterketamine administration. These data further reveal equalconcentrations of the two ketamine enantiomers in allthree cases and at all times shown. Furthermore, the for-mation of the two norketamine enantiomers appears tooccur at the same rate which is manifested by the obser-vation of equal amounts of the two enantiomers up toabout 4 min after ketamine administration. Thereafter, ahigher amount of S-norketamine compared to R-norket-amine was observed for the cases without xylazine pre-

medication, a finding that confirms the enantioselectivityof ketamine N-demethylation in the horse previouslyreported by Delatour et al. [8]. Analysis of 24 pony plasmasamples collected within 2–16 min after administration ofracemic ketamine (total of six ponies, anesthetized withisoflurane) resulted in a mean S-/R-ketamine concentra-tion ratio of 0.985 (RSD: 5.93%, median: 0.975, range:0.910–1.200) and a mean S-/R-norketamine concentra-tion ratio of 1.221 (RSD: 18.09%, median: 1.155, range:0.920–1.730). Using the Mann–Whitney rank sum testrevealed that there is a statistically significant differencebetween the two data sets (p , 0.0001). Comparabledata were obtained for the 4–16 min time points only(n = 18). With xylazine as premedication, however, statis-tical analysis of comparable data sets revealed not a sta-tistically significant difference between the S-/R-enantio-mer drug level ratios of ketamine and norketamine(p = 0.317). The mean S-/R-ketamine concentration ratiowas calculated to be 1.10 (RSD: 22.36%, median: 1.05,range: 0.67–1.60) and the mean S-/R-norketamine con-centration ratio was 1.21 (RSD: 34.13%, median: 1.09,range: 0.88–3.00). These data suggest that xylazine hasan impact on the stereoselectivity of the N-demethylationof ketamine.

4 Concluding remarks

The enantioselective CE assay described in this paper isshown to be robust and provides an analytical method tomonitor ketamine and norketamine enantiomer con-centrations .0.01 mg/mL in liquid–liquid extracts of 1 mLaliquots of equine plasma. Compared to HPLC methodsdescribed in the literature, sample extraction in the CE-based assay is simpler and the running costs are con-siderably lower. A CE capillary column does not cost morethan about 10 $ and can typically be employed for hun-dreds of injections. Small amounts of chemicals and sol-vents are only required. Furthermore, total ketamine andnorketamine concentrations monitored with the enantio-selective CE assay compare well to those determined byCE in absence of the chiral selector. After pretreatment ofsix ponies with isoflurane and subsequent application ofan i.v. bolus of 2.2 mg/kg racemic ketamine, the ketamineN-demethylation is demonstrated to be enantioselective.The same was found to be true for an adult horse thatreceived L-methadone, romifidine, and guaifenisine priorto and isoflurane anesthesia shortly after ketamineadministration. The concentrations of the two ketamineenantiomers in plasma are equal whereas S-norketamineis typically found in a larger amount than R-norketamine.Data obtained after premedication of the ponies withxylazine do not reveal this stereoselectivity. Further workis required to assess the impact of xylazine on the me-



Figure 8. (A–C) Ketamine and(D–F) norketamine drug levelsdetermined in plasma samplesof (A, D) a pony that was pre-treated with isoflurane, (B, E) anadult horse that was pretreatedwith L-methadone, romifidine,and guaifenisine and receivedisoflurane anesthesia shortlyafter ketamine administration,and (C, F) the same pony pre-treated with xylazine. Sampleswere drawn shortly before keta-mine administration and after i.v.bolus injection of 2.2 mg/kgracemic ketamine at the indi-cated time points. Symbollegend is presented in (B).

tabolism of ketamine. The developed enantioselective CEassay is currently being employed as an analytical tool forour research efforts in which the use of racemic ketamineversus S-ketamine for anesthesia and pain therapy inhorses is being assessed.

The authors gratefully acknowledge the support providedby Professor Urs Schatzmann. Xylazine, racemic keta-mine, and S-ketamine for i.v. drug administration werekindly supplied by Dr. E. Gräub (Bern, Switzerland). Thiswork was partly sponsored by the Swiss National ScienceFoundation.

Received January 25, 2005Revised February 18, 2005Accepted February 18, 2005

5 References

[1] Baselt, R. C., Cravey, R. H., Disposition of Toxic Drugs andChemicals in Man, 4th Edn., Chemical Toxicology Institute,Foster City, CA 1995, pp. 412–414.

[2] Moffat, A. C., Osselton, M. D., Widdop, B., Galichet, L. Y.(Eds.), Clarke’s Analysis of Drugs and Poisons in Pharmaceu-ticals, Body Fluids and Postmortem Material, 3rd Edn., Phar-maceutical Press, London, UK 2004, pp. 1152–1153.

[3] Chankvetadze, B., Burjanadze, N., Breitkreutz, J., Bergander,K., Bergenthal, D., Kataeva, O., Fröhlich, R., et al., J. Sep. Sci.2002, 25, 1155–1166.

[4] Williams, M. L., Wainer, I. W., Ther. Drug Monit. 2002, 24, 290–296.

[5] Yanagihara, Y., Kariya, S., Ohtani, M., Uchino, K., Aoyama, T.,Yamamura, Y., Iga, T., Drug Metab. Dispos. 2001, 29, 887–890.



[6] Hijazi, Y., Boulieu, R., Drug Metab. Dispos. 2002, 30, 853–858.

[7] Kharasch, E. D., Labroo, R., Anesthesiology 1992, 77, 1201–1207.

[8] Delatour, P., Jaussaud, P., Courtot, D., Fau, D., J. Vet. Phar-macol. Ther. 1991, 14, 209–212.

[9] Geisslinger, G., Menzel-Soglowek, S., Kamp, H.-D., Brune,K., J. Chromatogr. 1991, 568, 165–176.

[10] Svensson, J.-O., Gustafsson, L. L., J. Chromatogr. B 1996,678, 373–376.

[11] Rodriguez Rosas, M. E., Patel, S., Wainer, I. W., J. Chroma-togr. B 2003, 794, 99–108.

[12] Yanagihara, Y., Ohtani, M., Kariya, S., Uchino, K., Aoyama,T., Yamamura, Y., Iga, T., J. Chromatogr. B 2000, 746, 227–231.

[13] Williams, M. L., Mager, D. E., Parenteau, H., Gudi, G., Tracy,T. S., Mulheran, M., Wainer, I. W., Drug Metab. Dispos. 2004,32, 786–793.

[14] Blaschke, G., Chankvetadze, B., J. Chromatogr. A 2000,875, 3–25.

[15] Zaugg, S., Thormann, W., J. Chromatogr. A 2000, 875, 27–41.

[16] Hadley, M. R., Camilleri, P., Hutt, A. J., Electrophoresis 2000,21, 1953–1976.

[17] Aturki, Z., Fanali, S., J. Chromatogr. A 1994, 680, 137–146.

[18] Koppenhöfer, B., Epperlein, U., Christian, B., Lin, B., Ji, Y.,Chen, Y., J. Chromatogr. A 1996, 735, 333–343.

[19] Fanali, S., Camera, E., Chromatographia 1996, 43, 247–253.

[20] Amini, A., Sorman, U. P., Lindgren, B. H., Westerlund, D.,Electrophoresis 1998, 19, 731–737.

[21] Lu, W. Z., Cole, R. B., J. Chromatogr. B 1998, 714, 69–75.

[22] Matsunaga, H., Tanimoto, T., Haginaka, J., J. Sep. Sci. 2002,25, 1175–1182.

[23] Cherkaoui, S., Veuthey, J.-L., J. Pharm. Biomed. Anal. 2002,27, 615–626.

[24] Levionnois, O. L., Spadavecchia, C., Kronen, P. W., Schatz-mann, U., abstract in Proceedings of Association of Veter-inary Anesthetists (AVA), Autumn Meeting 2004, Vienna2004, p. 74.


Characterization of the stereoselective biotransformation of ketamine to norketaminevia...

Documents

Transcript of Characterization of the stereoselective biotransformation of ketamine to norketaminevia...