Phenotyping S-mephenyto P450 2C19) Zimbabwe

and genotyping of Iin hydroxylase (cytochrome in a Shona population of

The S-mephenytoin hydroxylase has recently been identied as cytochrome I?450 2C19 (CYP2C19). This enzyme metabolixes mephenytoin, diaxepam, omeprazole, and citalopram and has been shown to be polymorphically distributed One clinical implication of CYP2ClPdependent drug metabolism for per- sons who reside in tropical regions is in the use of the antimalarial drug chloroguanide hydrochloride, which is apparently biotransformed to its active metabolite by this isozyme. In this investigation we studied mephenytoin metabolism in 103 black Zimbabwean Shona subjects. Four were identified as poor metabohzers (4%). This prevalence is comparable to that in white subjects (2% to 5%) but lower than the 15% to 20% incidence of poor metabol&rs among Oriental subjects. Of the subjects pheno- typed, 84 were genotyped for the &A mutation in exon 5 of cYP2C19, which creates a cryptic splice site, causing the production of a nonfunctional protein. Three of the four poor metabolizers were ho- mozygous for this mutation, whereas the fourth one was heterozygous. The GA mutation has been shown to predict the incidence more than 60% of poor metabohzers among white subjects and Japanese subjects, and in the current investigation we also obtained a similar relationship in the black population. (CLINPHARMA~~LTHER 1995;57:656-61.)

Collen Masimirembwa, DPhil, Leif Bert&son, l?hD, Inger Johansson, MD, PhD, Julia A. Hasler, PhD, and Magnus Ingelman-Sundberg, PhD, BScM Harare, Zimbabwe, and Stockholm and Hz&ding, Sweden

Polymorphism in drug metabolism has become a subject of intensive research.’ The polymorphic ex- pression of drug-metabolizing enzymes is known to be one of the factors responsible for interethnic and inter-

From the Department of Medical Biochemistry and Biophysics, Karolinska institutet, Stockholm; the Department of Biochemis- try, University of Zimbabwe, Harare; and the Department of Medical Laboratory Sciences and Technology, Huddinge Univer- sity Hospital, Ijuddinge.

Supported by grants from the Bank of Sweden Tercentenary Foun- dation, the International Program in Chemical Sciences at Upp- sala University, the University of Zimbabwe Research Board, the Swedish Agency for Research Cooperation with Developing Countries (Swe-93-223), and the Swedish Medical Research Council. Dr. Masimirembwa is a recipient of a Swedish Institute Scholarship.

Received for publication Sept. 13, 1994; accepted Jan. 16, 1995. Reprint requests: Magnus Ingelman-Sundberg, PbD, Department of

Medical Biochemistry and Biophysics, Karolinska institutet, S-171 77 Stockholm, Sweden.

Copyright 0 1995 by Mosby-Year Book, Inc. OGQO-9236/95/$3.00 + 0 13/l/63425

656

individual variability in drug disposition, and hence in pharmacologic and in toxicologic responses.* The poor-metabolizer and extensive-metabolizer pheno- types have also been implicated in disease susceptibil- ities. Cytochromes P450 are a group of drug-metabo- lizing enzymes of which some are polymorphically expressed with varying prevalence of poor metaboliz- ers and extensive metabolizers in different popula- tions.4 Of such CYP45Os, CYP2D6, which metabo- lizes many clinically important drugs, has been extensively studied.* The molecular basis of the poor- metabolizer phenotype in both white subjects5 and Oriental subjects6 has been elucidated at the molecular level and includes two mutant alleles (CYP2D6A and CYP2D6B) among white subjects and an allele with a deletion of the functional gene (CYP2DtiD) in both ethnic groups. In addition, some individuals have been found to have duplicated or amplified active CYP2D genes and are therefore ultrarapid metaboliz- ers .7 The pharmacologic or toxicologic consequences of the extensive-metabolizer and poor-metabolizer

CLINICAL PHARMACOLOGY &THERAPEUTICS VOLUME 57, NUMBER 6 Masimirembwa et al. 657

phenotypes involves the lack of a pharmacologic ef- fect: in ultrarapid metabolizers because of subthera- peutic plasma concentrations or in poor metabolizers because of side effects caused by too high concentra- tions of the drug.2

The S-mephenytoin hydroxylase is a polymorphi- tally expressed cytochrome P450 that has been re- cently been identified as CYP2C19.* This isozyme is responsible for the metabolism of some clinically used drugs, for example, omeprazole, chloroguanide, and diazeparn.’ The prevalence of poor metabolizers in white subjects is 2% to 5%, whereas the frequency of poor metabolizers is much higher (15% to 20%) among Oriental subjects, with use of mephenytoin as a probe drug.‘0”1 The in vivo S-mephenytoin pheno- types can be based on the “hydroxylation index” val- ues12 or they can be based on the SIR urinary ratio,13 whereas a poor metabolizer is defined by a ratio greater than 0.9. Since the discovery of this polymor- phism, efforts to establish the molecular basis of the poor-metabolizer status have been difficult, but re- cently the major defect in poor metabolizers has been reported for white subjects and Oriental subjects.14 A single base pair (bp) mutation, G+A, in exon 5 of CYP2Cl9 creates an aberrant splice site, which alters the reading frame to produce a protein lacking the heme-binding region and therefore a nonfunctional product. This defect apparently accounts for more than 60% of poor metabolizers among white and Jap- anese populations. l4

In black subjects, apparently only two studies of the S-mephenytoin hydroxylase polymorphism have been performed. In a limited sample of 27 black Ameri- cans, 18% were found to be poor metabolizers” and, in a study among Nigerians,16 no apparent bimodal distribution of the mephenytoin hydroxylation indexes was observed. Since black subjects are evolutionarily distinct from Oriental subjects and white subjects,17 it is essential to establish the status of this enzyme in these populations. Knowledge of the prevalence of CYP2C19 poor metabolizers in relation to antipara- sitic drugs that are substrates of this enzyme, such as chloroguanide,‘* may become important in therapeutic programs for malaria. l9 This study in a Shona popula- tion of Zimbabwe is a contribution to research into the prevalence of the S-mephenytoin hydroxylase poor- metabolizer phenotype and the genetic basis of this status in a black African population.

MATERIAL AND METHODS Subjects. One hundred three unrelated Zimbabwe-

ans of the Shona tribe were recruited from the students

and staff of the University of Zimbabwe. Zimbabwe has two major ethnic groups, the Shona (80%) and the Ndebele (20%). The subjects were interviewed by a physician for drug and disease history and were con- sidered to be in good health. They participated in the study after giving oral and written informed consent. If any of the participants were to take medication a week before and during the study, they had to consult the study doctor, who was always on standby. None of the participating subjects took drugs during the course of this study other than the ones administered for the experiment. This study was approved by the Medical Research Council of Zimbabwe (Harare, Zimbabwe) and the ethical committee at Karolinska institutet (Stockholm, Sweden).

In vivo phenotyping. All 103 subjects participated in the phenotyping study. After fasting overnight, sub- jects emptied their bladders in the morning and each took a 100 mg tablet of racemic mephenytoin (Mesan- toin, Sandoz Pharmaceuticals Corp., East Hanover, N.J.). Urine was then collected for 8 hours after oral drug administration, the total volume was recorded, and 10 ml aliquots were stored at -20” C. The sam- ples were transported frozen from Zimbabwe to Swe- den and were kept at -20” C until analyzed at Hud- dinge University Hospital (Huddinge, Sweden).

The mephenytoin SIR ratio in urine was determined by chiral gas chromatography according to Wedlund et al. ,20 as modified by Sanz et a1.21 In urine samples in which R-mephenytoin, but not S-mephenytoin, could be detected, the SIR ratio was defined as the lower level of detection, 0.05. The S-mephenytoin poor-metabolizer phenotype associated with an SIR ra- tio greater than 0.9 was confirmed by the acidification procedure according to Tybring and Bertilsson,22 and each subject was assigned a poor-metabolizer pheno- type in case the acidification procedure did not change the urinary SIR ratio.

Genotyping. Ten milliliters of venous blood was obtained from 84 of the 103 subjects (19 subjects refused to give their blood for personal reasons) and deoxyribonucleic acid (DNA) was isolated from pe- ripheral leukocytes with a guanidium isothiocyanate method. The genotyping procedure was performed ac- cording to de Morais et a1.14 and involved two stages: (1) a polymerase chain reaction (PCR) amplification of the CYP2C19 intron Wexon 5 fragment of 169 bp and (2) digestion of PCR products with Sma I. The PCR incubation consisted of 1 to 1.5 ~1 (200 ng) genomic DNA, 2.5 ~1 (X10) PCR buffer, 11.5 p,l wa- ter, 4 ~1 (1.25 mmol/L deoxyribonucleoside triphos- phate (dNTP) mix, 0.625 ~1 of 10 PmollL forward

658 Masimirenzbwaetal. CLlNKAL. PHARMACOLOGY & THFXAl’EUTICS

JUNE 1995

169

PM EM

4 24 81 76 41 20 16 14 47 55 Subject number

Fig. 1. Polymerase chain reaction (PCR) analysis of the exon 5 mutation in the CYP2C19 gene. Shown are the results obtained with use of genomic deoxyribonucleic acid (DNA) from four CYP2C19 poor metabolizers (PM) and some extensive metabolizers (EM) in the Shona popula- tion. Upper panel shows the PCR amplification of the intron 4lexon 5 fragment. Lower panel shows the cleavage patterns of the PCR products with Sma I. PCR products from DNA of subjects with wild-type alleles will be cleaved by the restriction enzyme, whereas the DNA from individ- uals with the detrimental mutation lack the Sma I site and no fragments will be formed.

primer (S-AATTACAACCAGAGCTGGC-3’), 0.625 ~1 of 10 p,mol/L reverse primer (5’-TAT- CAC’MTCCATAAAAGCAAG-3’), 0.125 p1 of 5 units/@ Taq polymerase, and 5 l.~l of 7.5 mmol/L magnesium chloride. After an initial denaturation at 94” C for 90 seconds, 35 cycles of 60 seconds at 94" C, 90 seconds at 52” C, 90 seconds at 72“ C, and a final extension period of 7 minutes at 72” C were done. Eight microliters of each sample were analyzed on a 3% agarose gel stained with ethidium bromide. Twelve microliters of the PCR products were digested with Sma I in the PCR buffer without purification. The incubation for the digestion was: 12 pl of PCR product, 2 ~1 (X 10) restriction enzyme buffer for Sma I, 6 l.~l water and 3 units Sma I restriction en- zyme. The incubation was done over night at 25” C. The digested PCR products were analyzed on 3% aga- rose gels stained with ethidium bromide. In the wild- type allele, the PCR product has a Smu I site and two

fragments of 120 and 49 bp are then generated. In the mutant allele, the Sma I site is abolished and conse- quently no digestion occurs. Results from analysis of subjects who are heterozygous for this mutation will thus show three fragments, 169, 120, and 49 bp. Rep- resentative results are shown in Fig. 1.

RESULTS Phenotyping. Phenotyping of 103 Shona Zimba-

bweans was performed with use of racemic mepheny- toin. The urinary SIR ratio of mephenytoin was used to segregate poor metabolizers from extensive metab- olizers, and the frequencies of distribution obtained are shown in Fig. 2. Visual inspection of Fig. 2 shows a timodal distribution of the S/R ratios with the fol- lowing approximate groups: 0.05 to 0.28, 0.28 to 0.8, and 0.9 to 1.00. Three subjects in whom only the R-mephenytoin and not the S-mephenytoin was de- tected were given the S/R ratio of 0.05, which corre-

CLINICAL PHARMACOLOGY &THERAPEUTICS VOLUME 57, NUMBER 6 Masimirenzbwa et al. 659

sponds to the lower level of detection. F ive subjects had SIR ratios greater than 0.9, but after acid hydroly- sis, one of the subjects (241) showed an increase in SIR ratio from 0.92 to 1.52, a behavior that is associ- ated with extensive metabolizers. We therefore as- signed this subject the extensive-metabolizer status. The prevalence of poor metabolizers of S-mepheny- toin in the Shona population was 3.88% (4%).

Genotyping. The results of the genotype analysis are summarized in Table I. Among the poor metabo- lizers, three were homozygous for the mutation and one was heterozygous. In the extensive-metabolizer group, 15 subjects who were heterozygous for the exon 5 mutation were identified, whereas no individ- ual carried two defective alleles. The heterozygotes had S/R ratios (0.362 ? 0.189) that were higher than those for homozygotes (0.252 + 0.173), a difference that was statistically significant at p < 0.039 with use of the Mann-Whitney test. There was an overrepresen- tation of the heterozygotes in the group with SIR ratios 0.28 to 0.80. O f the 84 genotyped subjects, 15 were heterozygous extensive metabolizers and 10 of these had phenotypes within the SIR ratio range of 0.28 to 0.80. The allele frequency of the G+A mutation in exon 5 was 13% among the 84 subjects genotyped and, as shown in F ig. 1, this mutation therefore ac- counts for seven of the eight defective alleles among the four poor metabolizers. This indicates that the exon 5 mutation is the ma jor cause for the poor- metabolizer genotype in the black African population studied.

DISCUSSION The results from this study indicate that the fre-

quency of poor metabolizers of S-mephenytoin among black Shona Z imbabweans is 4%. These results are comparable to the 2% to 5% frequency of poor metab- olizers among white subjects but in contrast to the 15% to 20% frequency of poor metabolizers in Orien- tal subjects,*,” the 18% frequency of poor metaboliz- ers in black Americans,15 and the lack of a bimodal distribution in black Nigerians. l6 The SIR f requency distribution in the Shona subjects also show an anti- mode that separates poor metabolizers and extensive metabolizers similar to that in white subjects and Ori- ental subjects. The genotyping results supported the phenotype data in that the mutant allele, causing an aberrant splice site, was homozygously distributed in the three of the four poor metabolizers. The allele fre- quency of this mutation was 13% among the subjects genotyped. This mo lecular basis of the poor-metabo- lizer status is similar to that reported for Japanese sub-

0 .2 .4 .6 .8 1 1.2

SIR Mephenytoin ratio

F ig. 2. Phenotype analysis of the S-mephenytoin hydroxy- lase in a Shona Z imbabwean population. The urinary SIR mephenytoin ratio is shown from urinary analysis of 102 subjects. The value from one individual who had a ratio of 1.52 after hydrolysis has been excluded (see Results sec- tion).

jects and white subjects in which the homozygous dis- tribution of this mutation also accounts for more than 60% of the poor metabolizers.

These results, when compared to the findings among black Americans and black Nigerians, may in- dicate a heterogeneity with respect to the CYP2C19 gene of black populations. In light of these findings, it may therefore be important to specify the ethnic group of the black population studied. It may be that this would facilitate interpretations of results of studies performed in different black African populations.23

One potential clinical implication of the poor me- tabolizers in the Z imbabwean Shona population and other black African populations living in the tropics is in relation to chloroguanide metabolism. This antima- larial drug has been shown to be biotransformed to the active metabolite cycloguanil by the polymorphic CYP2C19. l8 Cycloguanil acts by inhibiting the plas- mod ial enzyme, dihydrofolate reductase, which is es- sential for parasite viability.” It is possible that the contribution of CYP2C19 to the activation is quantita- tively important and that the lack of CYP2C19-depen- dent activation will impair the antimalarial effect of this drug. The relative inability to form the active me- tabolite may result in prophylactic failures with this

660 Masirnirdwa et al. CLINICAL PHARMACOLOGY & THERAPEUTICS

JUNE 1995

Table I. Genotype analysis of 84 Shona Zimbabweans with respect to the exon mutation in the CYP2C19 gene that causes aberrant splicing of the gene

Genotype* No. of subjects Extensive metabolizers (n) Poor metabolizers (n) SIR metabolic ratio?

wtlwt 65 65 0 0.25 -+ 0.17 wtlm 16 15 1 0.36 k 0.19$ m/m 3 0 3 1.008 + 0.0435 TOTAL 84 80 4 -

The analysis was carried out by amplification of exon 5, followed by treatment with the restriction enzyme Sma I, according to de Morais et al. ,I4 on genomic DNA isolated from individuals who were previously phenotyped for S-mephenytoin.

*wt, Wild-type; m, mutant. The frequency of the mutant allele is 13%. TMean f SD of the S/R ratios. ‘&Significantly higher than the mean S/R ratio of the wt/wt, p < 0.039 with use of the Mann-Whitney test. BMean S/R ratio for poor metabolizers, including the wt/m poor metabolizer, significantly different from the wt/wt genotype, p < 0.0093 and p < 0.0069, respec-

tively.

drug and that subinhibitory concentrations of cy- cloguanil may promote the development of resistance of the parasite to the drug.19,18 Recently, however, CYP3A4 has also been implicated in the activation of chloroguanide. 24 Proof of a definitive relationship be- tween the S-mephenytoin polymorphism and the effect on chloroguanide metabolism has to await further clin- ical investigations.

Of the poor metabolizers found in this study, 75% were homozygous for the exon 5 mutation, which is similar to the percentage observed in white subjects and Oriental subjects. l4 This implies that the exon 5 mutation is relatively common in these populations, which diverged a long time ago.17 It appears that the human CYP2C locus on chromosome 10 is much more stable than the other well-established polymorphic lo- cus, namely, the CYP2D locus on chromosome 22. The instability of the CYP2D locus may be attribut- able to the close location to the immunoglobulin genes.25 The situation seems to be completely differ- ent for the CYP2C19 gene: the major defective gene appears to have been formed more than 100,000 years ago, and it appears not to have been the subject of clinically significant modifications.

We are indebted to Ingegerd Bertling and Olof Ekberg for valuable help in the analyses.

References 1. Daly AK, Cholerton S, Gregory W, Idle JR. Metabolic

polymorphisms. Pharmacol Ther 1993;57:129-60. 2. Kalow W, Bert&son L. Interethnic factors affecting

drug response. Adv Drug Res 1994;25:1-53. 3. Roots I, Brocketmoller J, Drakoulis N, Loddenkemper

R. Mutant genes of cytochrome P450IID6, glutathione S-transferase class mu, and arylamine N-acetyltrans- ferase in lung cancer patients. Clin Investig 1992;70: 307-19.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Meyer UA, Zanger M, Grant D, Blum M. Genetic polymorphism of drug metabolism. Adv Drug Res 1990;19:197-241. Kagimoto M, Heim M, Kagimoto K, Zeugin T, Meyer UA. Multiple mutations of the human cytochrome P450IID6 gene (CYP2D6) in poor metabolisers of de- brisoquine. J Biol Chem 1990;265:17209-14. Johansson I, Yue QY, Dahl ML, et al. Genetic analy- sis of the interethnic difference between Chinese and Caucasians in the polymorphic metabolism of debriso- quine and codeine. Eur J Clin Pharmacol 1991;40: 553-6. Johansson I, Lundqvist E, Bertilsson L, Dahl ML, Sjoqvist F, Ingelman-Sundberg M. Inherited amplifica- tion of an active gene in the cytochrome P450 2D locus as a cause of ultrarapid metabolism of debrisoquine. Proc Nat1 Acad Sci 1993;90:11825-9. Wrighton SA, Stevens JC, Becker GW, Vanden- Branden M. Isolation and characterization of human liver cytochrome P4502C19: correlation between 2C19 and S-mephenytoin 4’-hydroxylation. Arch Biochem Biophys 1993;306:240-5. Wilkinson GR, Guengerich FP, Branch RA. Genetic polymorphism of S-mepheytoin hydroxylation. Pharma- co1 Ther 1989;43:53-76. Nakamura K, Goto F, Ray A, et al. Interethnic differences in genetic polymorphism of debrisoquine and mephenytoin hydroxylation between Japanese and Caucasian populations. CLIN PHARMACOL TIER 1985; 38:402-g. Alvan G, Bechtel P, Iselius L, Gundert-Remy U. Hy- droxylation polymorphism of debrisoquine and mephen- ytoin in European populations. Eur J Clin Pharmacol 1990;39:533-7. Kupfer A, Preisig R. Pharmacogenetics of mepheny- toin: a new drug hydroxylation polymorphism in man. Eur J Clin Pharmacol 1984;26:753-9. Wedlund PJ, Aslanian WS, McAllister CB, Wilkinson GR, Branch RA. Mephenytoin hydroxylation deficiency in Caucasians: frequency of a new oxidative drug me-

cLINIcALlwARMA COLOGY&THERAWLTICS VOLUME57,NUMBER6 Masimirembwa et al. 661

tabolism polymorphism. CLIN PHAFWACOL THER 1984;36: 773-80.

14. de Morais SMF, Wilkinson GR, Blaisdell .I, Nakamura K, Meyer UA, Goldstein JA. The major genetic defect responsible for the polymorphis of S-mephenytoin me- tabolism in humans. J Biol Chem 1994;269:15419-22.

15. Pollock BG, Perel JM, Kirshner M, Altieri LP, Yeager AL, Regnolds CF. S-Mephenytoin 4-hydroxylation in older Americans. Eur J Clin Pharmacol 1991;40: 609-11.

16. Iyun AO, Tucker GT, Woods HF, Lennard MS. The 4-hydroxylation of S-mephenytoin in Nigerians: a popu- lation study. In: Ingelman-Sundberg M, Gustafsson JA, Orrenius S, eds. Eighth International Symposium: Mi- crosomes and Drug Oxidations. Stockholm: Karolinska institutet, 1990.

17. Cavalli-Sforza LL, Piazza A, Menozzi P, Mountain J. Reconstruction of human evolution: bringing together genetic, archeological and lingustic data. Proc Nat1 Acad Sci USA 1988;85:6002-6.

18. Ward SA, Helsby NA, Skjelbo E, Bmsen K, Gram LF, Breckenridge AM. The activation of the biguanide anti- malarial proguanil co-segregates with the mephenytoin oxidation polymorphism-a panel study. Br J Clin Pharrnacol 1991;31:689-92.

19. Manson-Bahr PEC, Bell DR. Manson’s tropical dis- eases. 19th ed. London: Bailliere Tindall, 1987.

20. Wedlund PJ, Sweetman BJ, McAllister CB, Branch RA, Wilkinson GR. Direct enantiomer resolution of mephenytoin and its N-demethylated metabolite in plasma and blood using chiral capillary gas chromatog- raphy. J Chromatogr 1984;307:121-7.

21. Sanz EJ, Villen T, Alm C, Bertilsson L. S-Mepheny- toin hydroxylation phenotypes in a Swedish population determined after coadministration with debrisoquin. CLIN PHARMACOL THER 1989;45:495-9.

22. Tybring G, Bertilsson L. A methodological investiga- tion on the estimation of the S-mephenytoin hydroxyla- tion phenotype using the urinary SIR ratio. Pharmacoge- netics 1992;2:241-3.

23. Lennard MS, Iyun AD, Jackson PR, Tucker GT, Woods HF. Evidence for a dissociation in the control of sparteine, debrisoquine and metoprolol metabolism in Nigerians. Pharmacogenetics 1992;2:89-92.

24. Birkett DJ, Rees D, Andersson T, Gonzalez FJ, Miners JO, Veronese ME. In vitro proguanil activation to cy- cloguanil by human liver microsomes is mediated by CYP3A isoforms as well as by S-mephenytoin hydrox- ylase. Br J Clin Pharmacol 1994;37:413-20.

25. Kimura S, Umeno M, Skoda RC, Meyer UA, Gonzalez FJ. The human debrisoquine 4-hydroxylase (CYP2D) locus: sequence and identification of the polymorphic gene, a related gene, and a pseudogene. Am J Hum Genet 1989;45:889-904.

Availability of JOURNAL Back Issues As a service to our subscribers, copies of back issues of CLNCAL Pm COLOGY & THERAPEUTICS for the

preceding 5 years are maintained and are available for purchase from the publisher, Mosby-Year Book, Inc., at a cost of $11.00 per issue. The following quantity discounts are available: 25% off quantities of 12 to 23, and 33% off quantities of 24 or more. Please write to Mosby-Year Book, Inc., Subscription Services, 11830 West- line Industrial Dr., St. Louis, MO 63146-3318, or call (800)453-4351 or (314)453-4351 for information on availability of particular issues. If unavailable from the publisher, photocopies of complete issues are available from UMI, 300 N. Zceb Rd., Ann Arbor, MI 48106 (313)761-4700.