CHIRALITY 7342-348 (1995)

Stereoselectivity of the Carbonyl Reduction of Dolasetron in Rats, Dogs, and Humans

JAMES DOW AND CHRISTIANE BERG Department of Drug Metabolism, Marion Merrell Dour, Strasbourg, France

ABSTRACT The initial step in the metabolism of dolasetron or MDL 73,147EF [(2a,6a,8a,9a~)-octahydro-3-oxo-2,6-methano-W-quinolizin-8-yl lH-indol-3-carboxylate, monomethanesulfonate] is the reduction of the prochiral carbonyl group to give a chiral secondary alcohol "reduced dolasetron." An HPLC method, using a chird column to separate reduced dolasetron enantiomers, has been developed and used to measure enantiomers in urine of rats, dogs, and humans after dolasetron administration. In all cases, the reduction was enantioselective for the (+)-(Rbenantiomer, although the dog showed lower stereose- lectivity, especially after iv administration. An approximate enantiomeric ratio (+/-) of 9 0 10 was found in rat and human urine. The contribution of further metabolism to this enantiomeric ratio was considered small as p r e h n a r y studies showed that oxidation of the enantiomeric alcohols by human liver microsomes demonstrated only minor stereoselectivity. Further evidence for the role of stereoselective reduction in man was obtained from in vitro studies, where dolasetron was incubated with human whole blood. The enantiomeric composition of reduced dolasetron formed in human whole blood was the same as that found in human urine after administration of dolasetron. Enantioselectivity was not due to differences in the absorption, distribution, metabolism, or excretion of enantiomers, as iv or oral adrmnistration of rac-reduced dolasetron to rats and dogs lead to the recovery, in urine, of essentially the same enantiomeric composition as the dose adrmnistered. It is fortuitous that the (+)-(R)- enantiomer is predominantly formed by carbonyl reductase, as it is the more active com- pound. o 1 9 9 ~ Wiley-Liss, Inc.

KEY WORDS: dolasetron, chiral reduction, dolasetron

Dolasetron or MDL 73,147EF [(201,6u,81~,9ap)-octahydro- 3-oxo-2,6-rnethano-W-quinolizin-8-yl lH-indol-3-carboxy- late, monomethanesulfonate] is a novel chemical entity exhib- iting high affinity for 5-HT, receptors.' In animals, the compound was found to be a potent and selective antagonist at the 5-HT3 receptor. In conscious ferrets, dolasetron sup- pressed the vomiting induced by an iv injection of the antican- cer drug cisplatin. Preliminary studies indicate that dolas- etron is an effective antiemetic agent in humans undergoing cancer chemotherapy with highly emetogenic agents., The compound is thus being developed for the treatment of cyto- toxic drug-induced nausea and vomiting. The structure of dolasetron includes a prochiral carbonyl group, which is readily reduced to the corresponding chiral alcohol, reduced dolasetron. Structures of dolasetron and its alcohol metabo- lite, reduced dolasetron, are given in Figure 1.

Previous pharmacokinetic studies in man4,5 showed that following iv administration to healthy male subjects, dolas- etron disappeared extremely rapidly from plasma, and less than 1% of the administered dose was excreted in urine as parent drug. A major metabolite, reduced dolasetron, peaked rapidly in the plasma.

The present study was carried out to assess whether the 0 1995 Wiley-Liss, Inc.

formation of (+)-(Rkenantiomer of reduced

reduction of dolasetron was stereoselective. The enantiomers of reduced dolasetron were measured by chiral HPLC in urine of rat, dog, and man, after iv and oral administration of dolas- etron. Racemic reduced dolasetron was also administered to rats and dogs in order to venfy that any differences in the relative amounts of enantiomers excreted was indeed due to stereoselective reduction, and not due to some other stereo- selective process of drug distribution or excretion.

MATERIALS AND METHODS Compounds

Dolasetron and rac-reduced dolasetron, both as the mesy- late salt, were used for drug administration. rac-Reduced dolasetron was obtained by sodium borohydnde reduction of dolasetron. The absolute configuration of (+)-(R) reduced dolasetron was assigned by X-ray crystallography, and this compound was used as a standard for chiral HPLC separa-

Received for publication October 17, 1994; accepted December 15, 1994. Address reprint requests to Dr. James Dow, Department of Drug Metabolism, Marion MerreU Dow, 16 rue &Ankara, 67080 Strasbourg Cedex, France.

STEREOSELECTIVE REDUCTION OF DOLASETRON 343

90 % / H

(+)-(R)-reduced dolasetron carbonyl reductase

H

Dolasetron

H

(-)-(S)-reduced dolasetron

Fig. 1. Reduction of the prochiral carbonyl group of dolasetron to give the corresponding chiral alcohol, reduced dolas- etron

tions. All the above compounds were synthetized at the Mar- ion Merrell Dow Research Institute, Strasbourg, France. HPLC grade solvents and chemicals were obtained from Merck (Darmstadt, Germany).

Drug Administration Male Sprague-Dawley rats (Iffa Credo, L’Arbresle,

France), weighing around 200 g, were studied after iv or PO administration of a 5 mgkg dose of dolasetron, or rac-reduced dolasetron. Doses were expressed as the free base of the compound. The animals (n = 3 for each compound, and route of administration) were housed in stainless steel metabolism cages and 0-8 h urine was collected into containers cooled with dry ice.

One male Beagle dog (Lessieux, Bray-Lu, France) weigh- ing 12 kg was given the following doses: dolasetron, PO 5 mgkg, iv 4 mgkg; rac-reduced dolasetron, PO 2 mgkg, iv 1 mgkg. There was a 1-month washout period between each adrmnistration. The dog was housed in a stainless steel me- tabolism cage, and 0-8 h urine was collected into containers cooled with dry ice. Again, doses were expressed as the free base of the compound.

For the human study, approval of the study protocol was obtained from an independent ethics committee. Volunteers were entered into the study after clinical examination and with their written consent. Four healthy male volunteers were studied after intravenous administration of 80 mg dolasetron, and urine was collected over 0-24 h. Two healthy male volun- teers were studied after oral administration of 100 mg dolas- etron, and urine was collected over 0-8 h.

HPLC The HPLC apparatus consisted of two Waters Model 510

pumps (Millipore S.A., Molsheim, France), a Waters Model 680 Automated Gradient Controller, a WISP autosampler, and a Waters 990 photodiode array detector.

HPLC was carried out using an Ultron E S - O W column (15 x 0.46 cm i.d.) (Rockland Technologies Inc, Delaware, USA). The mobile phase consisted of acetonitrile/0.025 M Na,HPO, [10/90 (v/v)l, whose pH was adjusted to 6.0 with 85% H:,PO,. Flow rate was 1 ml/min, and UV detection was at 281 nm. The wavelength was also scanned from 210 to 350 nm to allow peak identification and monitor peak purity.

Extraction of Reduced Dolasetron From Urine Urine (1 ml) was made alkaline with an equal volume of 0.25

M borate buffer pH 10.0 and extracted with 5 ml ethyl acetate/ n-hexane (75/25) by shaking on a rotary shaker for 15 min. The phases were separated by centrifugation, and the organic phase removed and evaporated under N2. The residue was redissolved in the HPLC eluant before injection.

Quantitation of Reduced Dolasetron in lirine Extraction efficiencies for the two enantiomers were esti-

mated in control human urine (4 x 1 ml), which was spiked with 2.5 pg/ml of rac-reduced dolasetron and extracted as outlined above. The extracts were injected onto the chiral HPLC column and the peak areas of each enantiomer mea- sured, and compared with that of an unextracted standard of rac-reduced dolasetron. The extraction efficiency for the (+ 1- (R)- and (-)-(S)-enantiomers was 68 and 61%, respectively.

344

AUFS 0.03

0.025

0.02

0.015

0.01

0.005

0

DOW AND BERG

AUFS 0.04

0.03

0.02

0 5 10 15 20 rnin 0 5 10 15 20 rnin

(4 (b)

Fig. 2. Chromatograms of (a) rac-reduced dolasetron and (b) (+)-(Rbenantiomer of reduced dolasetron, obtained with an Ultron ES-OVM HPLC column.

omers by chiral column HPLC was not available, quantitation of reduced dolasetron in urine from rats, dogs, and humans

similar, the calculation of enantiomeric composition was not corrected for recovery.

was done with reference to an external standard (2.5 pg) of rac-reduced dolasetron, which was iniected at the b e ~ n n Incubation of Urine With p-Gkcuronidase and at the end of u k e samples. Urine was extracted as outlined above and the extracts injected onto the chiral HPLC column. The peak area of each enantiomer in urine extracts

Rat, dog and human urine was incubated at 37°C for 48 h using the following conditions: 0.5 ml urine, 0.5 ml 1 M ace- tate pH 5.0, and 100 pl @-glucuronidase (Sigma type HP-2). A

. . . . .

CXLCI lldl bLdllUdlU, dllU dlLCt LUIICLLIULI IUI CXLIdLUUll CUP

ciency the concentration of each enantiomers was deter- mined.

Determination of Enantiomeric Composition The pure (-)-(3-enantiomer of reduced dolasetron was

unfortunately not available at the time this work was carried out. Standards of rac-reduced dolasetron and the pure (+I- (R)-enantiomer were injected onto the chiral HPLC column, and the retention time of each enantiomer was determined.

Peaks corresponding to each enantiomer of reduced dolas- etron, in extracts of rat, dog, and human urine, were identi- fied by their retention time and by their UV spectrum, using the diode array detector. The enantiomeric composition was then calculated by comparing the peak area corresponding to each enantiomer to the total peak area of both enantiomers.

out to cnecK me stamity or me compounas in unne.

Incubation of Dolasetron With Human Whole Blood Human whole blood (4 x 1 ml) was incubated with dolas-

etron, at a fmal concentration of 200 p M , for 2 h at 37°C. The incubation was terminated by the addition of ice cold methanol (1 ml) and the mixture left at -20°C overnight. After thawing, the mixture was centrifuged and the supernatant removed. The supernatant was extracted in the same manner as de- scribed above for urine. The reduction of dolasetron was monitored by reverse-phase HPLC on a Zorbax C, (5 pm) column (25 x 0.46 cm id.) (Shandon, Runcorn, England). The mobile phases consisted of A, 50 mM ammonium acetate/ acetonitrile [90/10 (v/v)l and B, 50 mM ammonium acetatel acetonitrile [40/60 (v/v)l. A linear gradient from 30% B to 100% B over 20 min was used for separations. Flow rate was

STEREOSELECTIVE REDUCTION OF DOLASETRON 345

AUFS 0025

0 02

0 015

0 0 1 .

0 005

nb

Fig. 3. Chual HPLC chromatograms of rat urine obtained after iv administration of (a) dolasetron and (b) rac-reduced dolasetron.

1 ml/min, and UV detection was at 280 nm. The peak corre- spondmg to reduced dolasetron was collected and the eluant evaporated under N,. The residue was redissolved in the HPLC eluant used for chiral HPLC before injection onto the chiral column. Chiral column HPLC separation of reduced dolasetron formed by whole blood was carried out as de- scribed previously.

RESULTS Separation of Enantiomers

Chromatograms of rac-reduced dolasetron and the (+)-(R)- enantiomer, obtained with an Ultron ES-OVM HPLC column, are shown in Figure 2a and b, respectively. The retention times of the (+)-(R)- and (-)-(S)-enantiomers were 9.8 and 15.4 min, respectively. The (+I-(R)-enantiomer was not completely pure, as it contained approximately 5% of the (- )-6)-enantiomer. The UV spectrum of each enantiomer peak was recorded, using the diode array detector, and this was used to confirm the identity of peaks in urine extracts.

Chiral HPLC of Urine Extracts From Rats, Dogs, and Humans

HPLC chromatograms of rat urine after iv administration of dolasetron or rac-reduced dolasetron are shown in Figure 3a and b, respectively. Chromatograms of dog urine were similar

to those of rat and are not shown. An HPLC chromatogram of a urine extract from a subject after iv administration of dolas- etron is shown in Figure 4. As there was some shift in the retention time of enantiomers, their identity was confirmed by comparing their UV spectra with those of standards. The enantiomeric ratio was calculated using these chromato- grams.

Quantitation The total concentration of reduced dolasetron in urine was

calculated by adding the concentration of each enantiomer and expressing this as a percentage of the dose administered. Results after iv administration of dolasetron or rac-reduced dolasetron are shown in Table 1, and those after oral drug administration in Table 2.

Enantiomeric Composition The enantiomeric ratio [(+)-(R)/( -)+)I of reduced dolas-

etron found in rat, dog, and human urine after iv admmstra- tion of dolasetron, showed some differences between the three species, but the (+ )-(R)-enantiomer was always pre- dominant. The dog showed the least stereoselectivity with a ratio of 65/35, whereas in rat this was 82/18 and in man 9416 (Table 1). After iv administration of rac-reduced dolasetron to rat and dog, the enantiomeric composition of reduced dolas-

AUFS 0.025 T

0.02 -

0.015 -

DOW AND BERG

0.005 0 5 10 15 20 min

Fig. 4. Chual HPLC chromatograms of a human urine extract obtained after iv administration of dolasetron.

etron in urine was essentially the same as that of the dose administered (Table 1).

The enantiomeric ratio, after oral administration of dolas- etron, also showed some species differences, but again the (+)-(R)-enantiomer was the most predominant. The dog showed slightly lower stereoselectivity with a ratio of 85/15, whereas in rat this was 91/9 and in man 95/5 (Table 2). Again, after oral administration of rac-reduced dolasetron to rat and dog, the enantiomeric composition of reduced dolasetron in urine was essentially the same as that of the dose adminis- tered (Table 2).

Incubation of Urine With P-Glucuronidase No evidence for stereoselective conjugation of the enanti-

omers of reduced dolasetron was found, as incubation of urine with P-glucuronidase did not appear to change their total con- centration or their relative amounts.

Incubation of Dolasetron With Human Whole Blood Around 50% of dolasetron was reduced after incubation for

2 h with human whole blood. The enantiomeric composition (+/-I of the reduced dolasetron formed by human whole blood was 88/12, which was similar to that found in human urine after administration of dolasetron.

DISCUSSION The resolution of the enantiomers of reduced dolasetron

can also be achieved by formation of diastereoisomers after

derivatization with Mosher’s reagent,‘ but this involves te- dious sample preparation. In the present study, direct chiral resolution of the enantiomers of reduced dolasetron was ob- tained by chiral HPLC, using an Ultron ES-OVM column.

Although the present method used an external standard for quantitation, it was sufficiently robust, as a comparison of concentrations of total reduced dolasetron in human urine, using the present method, compared with a specific GC-MS method7 showed good agreement. Good agreement was also found when amounts of total reduced dolasetron excreted in human urine, found in the present study, were compared with values found by Boxenbaum et al. 4.5

The specificity of the method was enhanced by the use of a diode array detector. This allowed the comparison and identi- fication of enantiomer peaks by their UV spectra. The diode array detector was also useful for the monitoring of peak purity.

Dolasetron is reduced by carbonyl reductase, a cytosolic enzyme. Carbonyl reductases are ubiquitous enzymes which are present in all mammalian species and a wide range of tissues.8 Very low amounts of parent compound were de- tected in human plasma after oral administration of dolas- etron5 due either to first-pass intestinal or hepatic reduction of the compound. Parent drug could be measured in human plasma after iv adrmni~tration,~ but only over the first 2 h after administration; thus, liver, blood, or other tissue carbonyl reductases rapidly reduce dolasetron.

Reduced dolasetron is the major metabolite of dolasetron in man and corresponds to over 20% of the administered dose excreted in 0-24 h urine.’ As reduction of the carbonyl group of dolasetron is very rapid in man,5 it is thought to be the initial step in dolasetron metabolism. Glucuronide and sulfate conju- gates of reduced dolasetron only account for a further 5% of the administered dose,’ thus, in the present study, the failure to detect any increase in the amount of reduced dolasetron after P-glucuronidase treatment of urine may be due to the relatively small amounts of conjugates present in human urine. Furthermore, it is not known whether the p-glucuronidase used exhibits any stereospecificity toward deconjugation of the respective diastereomeric glucuronides. Further metabo- lism of reduced dolasetron, by oxidation at the 5 and 6 posi- tions of the indole ring, also takes place. The total amount (free plus conjugates) of 5’-hydroxy and 6’-hydroxy reduced dolasetron excreted in 0-24 h human urine represents around 7 and 9% of the administered dose, respectively.’ As yet, no chiral HPLC method has been developed to separate the enantiomers of the hydroxylated metabolites of reduced dola- setron.

It could be argued that further metabolism of reduced dola- setron contributes to the enantiomeric ratio (+/-) of 9010 found in urine, especially if this oxidation was stereoselective. Preliminary work, using reverse-phase HPLC, has shown that microsomes prepared from human liver can form both 5’- and 6’-hydroxy reduced dolasetron, when incubated with the (+)-(R) or (-)-(S)-enantiomer of reduced dolasetron. For both enantiomers, the 6‘-hydroxy metabolite was the major compound formed in vitro, which is similar to results found in vivo. The (-)-(S)-enantiomer formed twice as much 6’- hydroxy reduced dolasetron as the (+I-(R)-enantiomer,

STEREOSELECTIVE REDUCTION OF DOLASETRON 347

TABLE 1. Enantiomeric composition of reduced dolasetron excreted in urine of rat and dog (0-8 h) and human (0-24 h) after iv administration of dolasetron or rac-reduced dolasetron

~~ ~

Drug administered % Dose excreted Enantiomeric composition (+I-)

Species Dose as reduced dolasetron of reduced dolasetron

Dolasetron Rat (n = 3) 5 m g k 15.6 t 7.6 Dog (n = 1) 4 m g k 14.9 Human (n = 4) 80 mg 25.1 ? 9.1

82/18 65/35 9416

rac-Reduced dolasetron Rat (n = 3) 5 m g h 33.2 t 14.3 51/49 Dog (n = 1) 1 mgkg 13.6 46/54

TABLE 2. Enantiomeric composition of reduced dolasetron excreted in rat, dog, and human urine (M h) after oral administration of dolasetron or rac-reduced dolasetron

Drug administered

~~ ~

% Dose excreted Enantiomeric composition (+I-) Species Dose as reduced dolasetron of reduced dolasetron

~~ ~

Dolase tron Rat (n = 3) 5 m g M 6.3 t 1.3 Dog (n = 1) 5 mgkz 15.2 Human (n = 2) 100 mg 17.1 t 2.5

91/9 85/15 9515

rac-Reduced dolasetron Rat (n = 2) 5 m g k 3.4 t 1.0 58/42 Dog (n = 1) 2 mgkg 9.3 47/53

whereas the (+)-(R)-enantiomer formed around four times as much 5’-hydroxy reduced dolasetron as the (-)-(3-enanti- omer. Although these results demonstrate some stereoselec- tivity for the oxidation of the enantiomeric alcohols, these differences are probably too small to contribute to the enanti- omeric ratio (+/-I of 9 0 10 found in urine, especially as these are relatively minor pathways. Work is in progress to study the steroselective oxidation of the enantiomeric alcohols in rats, dog, and man.

In the present study, stereoselective reduction of dolas- etron was seen in all species studied, with the (+)-(R)-enanti- omer of reduced dolasetron predominant in urine. Some spe- cies differences were seen in the enantiomeric composition found in urine, after iv administration of dolasetron. The dog appeared to show the least stereoselectivity for the (+)-(R)- enantiomer, followed by the rat, and man showed the highest stereoselectivity. Interestingly, stereoselectivity appeared to increase in both rat and dog, after oral administration of dola- setron. This could indicate a possible first-pass effect for stereoselective reduction in these two species. These results are unlikely to be due to stereoselective absorption, distribu- tion, metabolism or excretion of enantiomers, as iv or oral administration of rac-reduced dolasetron to rat and dog lead to the recovery, in urine, of the same enantiomeric composition as the dose administered. Unfortunately, rac-reduced dolas- etron could not be administered to human volunteers as drug safety studies were carried out on dolasetron. However, do- lasetron was reduced, when incubated with human whole blood in vitro, and this reduction was also stereoselective for the (+)-(R)-enantiomer. These results provide further evi- dence that stereoselective reduction is the major metabolic pathway that leads to the enantiomeric composition of re- duced dolasetron found in human urine.

Fig. 5. Prelog’s rule for product stereoselectivity of enzymatic ketone reduc- tion.

Stereoselective reduction of a carbonyl function has also been reported for the hypolipidaemic drug fenofibrate. ’” Urine obtained from rat, guinea pig, and dog, after oral admin- istration of the drug, showed over 95% stereoselectivity for the (-)-isomer, whereas, unlike dolasetron, in man there was no evidence of stereoselective reduction. In contrast, the reduction of haloperidol in human tissues was found to be extremely stereoselective for the (- )-6)-enantiomer. l 1 The formation of the (S)-enantiomer was predicted, using Prelog’s rule. Prelog, who studied the reduction of ketones by bacte- riological systems, formulated a rule to predict the configura- tion of the resulting alcohol. Prelog’s rule states that if a ketone is projected in a plane with the largest or bulkiest group at the left, the resulting alcohol will predominantly have the configuration with the hydroxyl group above the plane of the paper (Fig. 5). The validity of Prelog’s rule has been confirmed in vitro and in vivo by Prelusky et al., l3 who studied the reduction of several arylalkyl ketones in rabbits and rats. If dolasetron is oriented in a plane as shown in Figure 5, then the bulkiest group is on the left and the formation of the (R)-enantiomer would be predicted from Prelog’s rule.

The relative pharmacological activities of the (+)-(R)- and (- )-(3-enantiomer of reduced dolasetron have been stud- ied. l4 5-HT,, binding assays showed that the (+ )-(R)-enanti-

348 DOW AND BERG

omer was approximately four times more potent than the (-)-(S)-enantiomer. Thus, fortuitously, carbonyl reductase predominantly forms the more active enantiomer.

In conclusion, the stereoselective reduction of the prochiral carbonyl group of dolasetron to give a chiral secondary alco- hol, reduced dolasetron, was observed in rat, dog, and man. The reduction was enantioselective for the (+ )-(R)-enanti- omer, although the dog showed lower stereoselectivity, espe- cially after iv administration. An approximate enantiomeric ratio (+/-) of 9010 was found in rat and human urine. The contribution of further metabolism to this enantiomeric ratio was considered small as preliminary studies showed that oxi- dation of the enantiomeric alcohols by human liver mi- crosomes demonstrated only minor stereoselectivity. Fur- ther evidence for the role of stereoselective reduction in man was obtained from in vitro studies, where dolasetron was incubated with human whole blood. The enantiomeric compo- sition of reduced dolasetron formed in human whole blood was the same as that found in human urine after administration of dolasetron. Enantioselectivity was not due to differences in the absorption, distribution, metabolism, or excretion of enantiomers, as iv or oral administration of racemic reduced dolasetron to rats and dogs lead to the recovery, in urine, of essentially the same enantiomeric composition as the dose administered.

ACKNOWLEDGMENTS The authors thank Muriel Rohfritsch-Chiesa for the prepa-

ration of the manuscript.

LITERATURE CITED 1. Gittos, M. W., Fatmi, M. Potent 5-HT,, antagonists incorporating a novel

bridged pseudopelletierine ring system. Actual Chm. Ther. 1 6 187-189, 1989.

2. Miller, R.C., Galvan, M., Gittos. M.W., van Giersbergen, P.L.M., Moser, P.C., Fozard, J.R. Pharmacological properties of dolasetron. a potent and selective antagonist at 5-HT:, receptors. Drug Dev. Res. 28:87-93, 1993.

3. Aapro, M.S. 5-HT,, receptor antagonists. An overview of their present status and future potential in cancer therapy-induced emesis. Drugs 4 2 551-568, 1991.

4. Boxenbaum, H., Gillespie, T., Heck, K., Hahne, W. Human dolasetron pharmacokinetics: 1. Disposition following single-dose intravenous admin- istration to normal male subjects. Biopharm. Drug Dispos. 13:693-701, 1992.

5. Boxenbaum. H., Gillespie. T., Heck, K., Hahne, W. Human dolasetron pharmacokinetics: 11. Absorption and disposition following single-dose oral administration to normal male subjects. Biopharm. Drug Dispos. 14: 131-141, 1993.

6. Dale, J. A., Dull, D.L., Mosher, H.S. a-Methoxy-a-trifluoromethylphe- nylacetic acid, a versatile reagent for the determination of enantiomeric composition of alcohols and amines. J. Org. Chem. 34254S2549, 1969.

7. Dulery, B.D. Personal communication. 8. Felsted. R.L., Bachur, N.R. Mammalian carbonyl reductases. Drug

Metab. Rev. 11: 1-60, 1980. 9. Sanwdld, P., Huebert, N. D., Haegele, K. D. Simultaneous measurement

of the major metabolites of dolasetron mesilate in human urine using solid-phase extraction and high-performance liquid chromatography. J. Chromatogr. 661: 101-107, 1994.

10. Weil, A., Caldwell, J., Guichard, J.-P., Picot, G. Species differences in the chirality of the carbonyl reduction of [14C1 fenofibrate in laboratory animals and humans. Chirality 1: 197-201, 1989.

11. Eyles. D. W., Pond, S.M. Stereospecific reduction of haloperidol in human tissues. Biochem. Pharmacol. 448674371, 1992.

12. Prelog, V. Specification of the stereospecificity of some oxidoreductases by diamant lattice sections. Pure Appl. Chem. 9 119-130, 1964.

13. Prelusky, D.B., Coutts, R.T., Pasutto. F.M. Stereospecific metabolic reduction of ketones. J. Pharm. Sci. 71:139&1393, 1982.

14. Bigaud, M., Elands, J., Kastner, P.R., Bohnke, R.A., Emmert, L.W., Galvan, M. Pharmacology of the human metabolites of dolasetron, an antiemetic 5-HT,, receptor antagonist. Drug Dev. Res. 34:289-296, 1995.