CYP2D6 genotype and phenotype relationship in South IndiansAT Naveen, T Prasanna, BL Farzana, S Rajan, C Adithan
Pharmacogenomics Laboratory, Department of Pharmacology, JIPMER, Pondicherry - 605 006, India
Background : Genotypes of the drug-metabolizing enzyme CYP2D6 influence plasma levels of 25% of commonlyprescribed drugs. This is the first study in India to investigate the genotype-phenotype relationship of CYP2D6. Aim : To study the influence of some CYP2D6 genotypes on the metabolism of its substrate dextromethorphanin healthy South Indian volunteers and to assess the contribution of the CYP2D6*10 and CYP2D6*4 alleles. Materials and Methods : Twenty-six subjects from a previous CYP2D6 genotyping study of healthy volunteerswere included for phenotyping in this study. Selected volunteers belonged to any one of three genotype groups:Group I - two normal activity alleles, Group II - one reduced activity allele and one normal activity allele andGroup III - one loss of function allele along with either a wild type or reduced activity allele. Volunteers werephenotyped for the CYP2D6 enzyme using dextromethorphan as probe drug. Concentrations of the parent drugand metabolite dextrorphan were estimated using high performance liquid chromatography. Metabolic ratioswere calculated as the ratio of parent drug to metabolite in 0-8h urine samples. Statistical Analysis : Metabolic ratios from each genotype group were compared using the Mann-Whitney testat 5% significance, to observe their difference between genotype groups. Results : The mean metabolic ratios±SD in Groups I, II and III were 0.0039±0.0031, 0.0032±0.0017 and0.0391±0.0331 respectively. The mean metabolic ratio of Group III was significantly higher when comparedwith Groups I or II. In heterozygous individuals, the *1 or *2 alleles compensated for the reduced enzymeactivity due to the *10 allele. However, if a heterozygous individual had a *4 allele, the reduced enzyme activitycould not be compensated by the *1 or *2 alleles. Conclusions : The CYP2D6 enzyme activity was found to be decreased in individuals carrying *4 or *5 alleles.The *1 or *2 allele could compensate for the reduced function due to *10 allele, but not for the loss of functiondue to *4 allele.
Keywords: CYP2D6, genotype-phenotype correlation, South Indians
Inter-individual variation in response to drugs within populations may be mainly attributed to genetic polymorphisms and variation in drug metabolism plays a major role. Studying polymorphic drug metabolism in a population identifies the proportion of individuals differing in ability to metabolize certain drugs and who therefore are likely to react differently or adversely to drugs. The drug metabolizing enzyme CYP2D6 is involved in the metabolism of many routinely prescribed drugs and therefore genetic polymorphism in this gene is very extensively studied. Phenotyping populations with respect to this enzyme using suitable probe drug substrates, has helped define distinct subgroups of individuals within each population, all differing in their metabolic capacity. Subgroups such as poor metabolizer, intermediate metabolizer, extensive metabolizer and ultra extensive metabolizer have been described in major populations by these studies. In our earlier phenotyping studies on South Indians using dextromethorphan as probe drug for the polymorphic drug-metabolizing enzyme CYP2D6 , the proportion of poor metabolizers among this population was identified. The frequencies in the four South Indian states of Kerala, Karnataka, Andhra Pradesh and Tamil Nadu were 4.8%, 4.0%, 1.8% and 3.6% respectively, contributing to 3.5% of CYP2D6 poor metabolizers among South Indians.,,
Genotype data seen alongside phenotype data, explains the mutations or the molecular genetics involved and also helps to relate the phenotype of an individual with his genotype, which may help to predict phenotype from genotype within a population. About 90 variant alleles have been reported in the CYP2D6 gene in world populations and around 15 key mutations in this gene cause loss of enzyme activity.
In our earlier CYP2D6 genotyping study of 447 healthy individuals, the frequency of CYP2D6*10 allele, which causes a decrease in enzyme function, was 10.2%. The frequencies of CYP2D6*4 and *5 alleles, which cause absence of enzyme activity, were 7.3 and 1.9% respectively. CYP2D6*2 was the most common variant allele (34.8%). The *3, *14 and *17 alleles were absent in South Indians. In the present study, we studied the contribution of the CYP2D6* 10 and *4 alleles towards the metabolism of dextromethorphan in South Indians.
Twenty-six healthy South Indian volunteers (20 males and 6 females, mean age ± SD = 23.1 ± 8.4 years) who participated in the previous genotyping study were included in the present study. They were not suffering from any chronic illnesses and were not on any medications including herbal remedies or food supplements. Subjects were nonsmokers and nonalcoholics. The study protocol was explained to all subjects in detail and written informed consent obtained. The institutional ethics committee approval was also obtained.
Each subject was genotyped for the CYP2D6*2, *3, *4, *5, *10, *14 and *17 mutations of the gene. PCR for amplifying the desired genomic sequences and PCR-RFLP for mutation detection were carried out based on the methods discussed earlier.,,,, The subjects were divided into three genotype groups viz: Group I: having two wild type (normal activity) alleles - CYP2D6*1/*1 or CYP2D6*1/*2 (n=11), Group II: having one wild type and one reduced function mutant allele - CYP2D6*1/*10 or CYP2D6*2/*10 (n=8) and Group III: having one loss of function allele, along with either a wild type or a reduced activity allele - CYP2D6*1/*4 , *2/*4 or CYP2D6*5/*10 (n=7).
The CYP2D6 phenotyping of the volunteers was done after oral administration of 5 ml of Lactuss LA (FDC limited, Aurangabad, India) syrup containing 30 mg of dextromethorphan, after emptying the bladder completely, as described previously. The samples were collected for a period of 8 hours, (overnight). Total urine output was recorded. An aliquot of 5 ml of urine was stored at -25°C until analysis by high performance liquid chromatography (HPLC).
The urinary levels of dextromethorphan and its metabolite dextrorphan were estimated by the HPLC method as described previously, with slight modification. In the procedure, the urine samples were thawed to room temperature, 0.5 ml was taken in a clean screw-capped glass tube, 150 ml of 10M HCl was added and vortexed for 5 seconds, centrifuged at 2000 rpm for 5 min. This was then incubated at 100°C for 90 min, cooled and 250 ml of saturated sodium carbonate solution was added four times, briefly vortexing after each addition. Six ml of extraction mixture (20:9:1 of diethylether: chloroform: propan-2-ol) was added and closed airtight. These were shaken for 15 min in an orbital shaker and centrifuged at 400 rpm for 20 min. The organic layer was transferred to a clean screw-capped glass tube containing 400 ml of 0.2N HCl. The tubes were kept in a shaker for 15 min and centrifuged at 4000 rpm for 20 min. Three hundred microlitres of the acid layer was separated out into 0.5 ml microfuge tubes and centrifuged at 10,000 rpm for 10 min. Tubes were kept open in a protected container till injection into the HPLC. Two hundred microlitres was injected into the HPLC system (Shimadzu LC-10AS pump, Thermo Hypersil-Keystone CPS Hypersil column, Shimadzu RF-530 fluorescence detector with excitation and emission wavelengths of 280 and 310 nm respectively). The inter-day and intra-day coefficients of variation for the method were less than 10% and 5% respectively.
For calculation of the metabolic ratio (MR), the following formula was used:
Concentration of dextromethorphan in 0-8h urine ¸
Concentration of O-demethylated metabolite (dextrorphan) in 0-8 h urine
Statistical analyses were done using GraphPad InStat software version 3.06 and data were expressed as mean±SD. Mann-Whitney test was used to test significant difference between groups. P value < 0.05 was considered as statistically significant.
Both dextromethorphan and its metabolite dextrorphan were detectable in urine samples of all volunteers. None of the subjects reported adverse events due to the probe drug. The mean MR of Group I and Group II was not significantly different. The MR was significantly higher in Group III subjects when compared to Groups I or II [Table - 1].
Comparison of the MR in various allelic combinations viz. * 1/*1, *1/*2, *1/*4, *2/*4, *1/*10, *2/*10 and *5/*10 , is given in [Table - 2]. There was no significant difference in the MR of CYP2D6 *1/*1 and *1/*2 . In heterozygous individuals having *10 allele, the MR was not significantly changed when compared to CYP2D6*1/*1 or CYP2D6*1/*2 .
In individuals having *4 or *5 alleles, the MR was higher, but the numbers were small to make any statistical inference [Table - 2]. Individual MR values of subjects in all the three genotype groups are shown in [Figure - 1].
The genotype and phenotype relation of CYP2D6 drug-metabolizing enzyme has not been studied in Indians. The results of the present study indicate that the metabolic capacity of individuals towards dextromethorphan differs with respect to their CYP2D6 genotypes. The increase in mean MR in Group III indicates a decrease in CYP2D6 enzyme activity, which can be explained by the presence of nonfunctional alleles (i.e, CYP2D6*4 and *5 ). In individuals carrying the CYP2D6*10 allele, the enzyme activity was found to be similar to that reported by studies in other Asian populations like the Japanese. The results of our study with respect to the CYP2D6*10 allele therefore are similar to that reported in other populations and are in agreement with other studies with respect to the *4 and *5 alleles.
In Group III, there is a larger reduction in the metabolism of dextromethorphan compared to that in Group II since the former group carries the *4 allele which is a nonfunctional allele of the CYP2D6 gene. Individuals having *1/*4 allelic combination may have a significant reduction in metabolism of CYP2D6 substrates such as tricyclic antidepressants, SSRIs, beta blockers etc., Drug metabolizer genotype when conveyed in the right manner to individuals, especially carriers of poor metabolizer genotype, would possibly prevent adverse events or therapy failure with CYP2D6 substrates.
Genotype groups compared in the light of phenotype data may give valuable clues to the uniqueness of a population with respect to drug metabolism. For example, the CYP2D6*17 allele, was found not only to cause a reduction in CYP2D6 enzyme activity, but also a change in substrate specificity in Africans. In another population of Gabonese, the CYP2D6*2 allele was found to be associated with lesser CYP2D6 enzyme activity when compared with other populations such as Caucasians. In the present study, the wide distribution of MR values in Group III may suggest the presence of other variants apart from those already detected, in these individuals.
In South Indians, both CYP2D6*4 and CYP2D6*5 alleles cause significant reduction in the activity of CYP2D6 enzyme. In heterozygous individuals, the presence of the *1 or *2 allele may compensate for the presence of the *10 mutated allele but not that of the *4 or *5 allele.
It would be interesting to compare the findings of this study with those done in other ethnic Indian populations. But to the best of our knowledge, this is the first study reporting genotype-phenotype relationship in Indians. Since about 25% of commonly prescribed drugs are metabolized by CYP2D6 , the findings of this study may be clinically important.
[Figure - 1][Table - 1], [Table - 2]