Genotype-Phenotype Analysis of CYP2C19 in Healthy Saudi Individuals and its Potential Clinical Implication in Drug Therapy

Abstract

CYP2C19 is a cytochrome P450 enzyme, which is involved in the metabolism of some clinically important medications and is encoded by a highly polymorphic gene. There is no available data on the distribution of the CYP2C19 *4 and *17 mutant alleles in the Saudi Arabian population. The aim of the study was to determine different CYP2C19 mutant allele (*2, *4 and *17) frequencies in healthy Saudi subjects and to determine genotype frequencies for these mutations. The CYP2C19 genotypes were then classified into phenotypes. Result: In 201 adults of Saudi ethnicity, the allele frequencies were CYP2C19*1 (62.9%), *17 (25.7%), *2 (11.2%) and *4 (0.2%). The most prevalent genotype combinations were CYP2C19 *1/*1 (40.3%) and *1/*17 (30.4%). The distribution of CYP2C19 phenotypes was divided into extensive metabolizers (EM) 77.6%, intermediate metabolizers (IM) 14.9%, ultra-rapid metabolizers (UM) 7% and poor metabolizers (PM) 0.4%. This finding has important clinical implications for the use of CYP2C19 metabolized medications in the Saudi population and further studies are needed.

Introduction

CYP2C19 is a cytochrome P450 enzyme involved in the metabolism of some clinically important medications. ¹. There is an association between CYP P450-based genetic variation and the outcome of drug therapy, adverse drug reactions and therapeutic failures. The genes encoding for CYP2C19 are in polymorphic expression ², with 30 variant alleles for CYP2C19 found to date ³. The CYP2C19 allele frequencies and genotype distribution were derived by gene counting. The CYP2C19 genotypes were classified into four phenotypes: (1) extensive metabolizer (EM) carrying normal function alleles (CYP2C19*1/*1, *1/*17, *2/*17, *4/*17 ); (2) intermediate metabolizer (IM) carrying one loss-of-function allele (*1/*2, *1/*4); (3) poor metabolizer (PM) carrying two loss-of-function alleles (*2/*2, *2/*4, *4/*4) and (4) ultra rapid metabolizer (UM) for alleles (*17/*17).

The alleles *2, *3, and *4 are associated with decreased metabolism of the substrates (drugs), although the *4 allele is uncommon ^4,5,6. The CYP2C19 mutant alleles *4 and *17 have not been well studied in most populations. The frequency of polymorphic alleles shows distinct inter-ethnic variation.

Some examples of commonly prescribed drugs metabolized by CYP2C19P are as follows: the antiplatelet drug (clopidogrel), proton pump inhibitors (omeprazole, lansoprazole), anticonvulsants (phenytoin, diazepam), selective serotonin reuptake inhibitor (citalopram), and the tricyclic antidepressants (amitriptyline, clomipramine) ^7,8,9,10,11. Recent studies have shown that CYP2C19 polymorphisms have caused a diverse responsiveness to clopidogrel ^{12, 13}. The risk of cardiovascular events is increased in patients who are PM (carrying at least one CYP2C19*2 allele) despite patients receiving adequate doses of an antiplatelet agent, clopidogrel ^{14, 15}. In contrast, patients carrying the CYP2C19*17 *17 allele with UM phenotype had greater protection from clopidogrel treatment after acute myocardial infarction with extensive platelet activity ^{16, 17}. A reverse occurrence was noted in the treatment of peptic ulcer disease with PPIs. A greater acid suppression was seen in patients who were PM carrying at least one CYP2C19*2 allele whereas poor acid suppression was noted in UM patients carrying CYP2C19 *17 (*17 allele in patients treated with PPIs ^{18, 19}.)

To our knowledge, only a single study has been published on the prevalence of *2 and *3 mutations related to the genetic polymorphism of CYP2C19 in the Saudi population in 1997 ²⁰. No information is available on the genotyping of CYP2C19 mutants *4 and *17 alleles in this population.

The aim of the study was to determine different CYP2C19 mutant allele (*2, *4 and *17) frequencies in healthy Saudi subjects and to determine genotype frequencies for these mutations. The CYP2C19 genotypes were then classified into phenotypes. We also compared our result with other population genetic polymorphisms of CYP2C19. The study results should allow us in future to predict adverse effects and to optimize treatment of medications metabolized through CYP2C19 in our population.

Materials and Methods

Study population

The study included 201 adults of Saudi ethnic origin (100 male and 101 female) aged 18 to 65 years between 1 August 2011 and 7 August 2011. The subjects were recruited randomly from King Fahad Medical City Blood Donation Center, Riyadh, Saudi Arabia. Subjects with any types of medical illness, organ transplant, drug or alcohol addiction, as well as pregnant females were excluded from the study. A prospective cross sectional study design was followed.

The study was approved by the Institutional Review Board of the hospital; all subjects were informed, both verbally and in writing, about the experimental procedures, confidentiality, and the purpose of the study. Written informed consents were obtained from all participants prior to entering the study.

Genotyping of CYP2C19

A blood sample (3 mL) was drawn from each subject into an EDTA tube and DNA was extracted using the QIAGEN DNA Isolation Kit (Qiagen, Germany) according to the manufacturer's instructions. Validated TaqManâ Master Mix and TaqManâ genotyping assay (4324018 Applied Biosystems, USA) were used to discriminate CYP2C19 for the following single-nucleotide polymorphism (SNPs): G681A (rs4244285), A1G (rs28399504), and C-806T (rs12248560) to characterize *2, *4, and *17 alleles, respectively. Polymerase chain reaction amplification for all single nucleotide polymorphisms was performed in 25 μL reactions with 20 ng of template DNA, 1X Taqmanâ Universal Master Mix (Applied Biosystems, USA), 1X each primer and probe assay, and water qsp. Thermal cycling was initiated with a first denaturation step of 10 min at 95°C, followed by 50 cycles of denaturation at 92°C for 15 s and annealing at 60°C for 1.30 min. The allele detection process was performed for 1.30 min at 60°C on a Fast 7500 Real-Time PCR System (Applied Biosystems, USA) to determine the allelic discrimination.

Statistical analysis

SPSS statistical package version 19.0 was used for data analysis. Chi-square and Fisher's exact tests were used to compare the allele and genotype frequencies and descriptive analysis was used to compare allele frequencies between the Saudi population and published data of other populations. The Hardy-Weinberg equation was used for the assessment of predicted frequencies of genotypes. A p-value of < 0.05 was deemed to represent statistical significance and confidence intervals of 95% were determined.

Results

The allele frequencies of CYP2C19*1, *2, *4, and *17 and genotype frequencies in the Saudi Arabian population are shown in Table 1. CYP2C19*1 was the most frequently identified allele (253/402, 62.9%) and 81 (40.3%) subjects were homozygous for the CYP2C19*1 *1 allele. CYP2C19*17 was the most common identified variant allele (103/402, 25.7%) 14 (7.0%) subjects were homozygous for the CYP2C19*17 *17 and 14 (7.0%) subjects were heterozygous CYP2C19*2 *17 allele. CYP2C19*2 was the second most common variant allele in our population (45/402, 11.2%); 29 (14.5%) subjects were heterozygous for CYP2C19*1 *2, 14 (7%) subjects were heterozygous for CYP2C19 *2 *17, and only one was homozygous for CYP2C19 *2 *2. No subjects were homozygous for the CYP2C19*4 mutant allele and one subject was heterozygous for CYP2C19 *1 *4.

Table 1.

Frequencies of CYP2C19 alleles and genotypes in the Saudi population sample (n=201)

CYP2C19	Allele	Frequency	95% Confidence interval	Actual number	Expected number by Hardy-Weinberg law
Alleles	*1	0.629	0.585 - 0.674	253	N/A
	*17	0.257	0.216 - 0.299	103	N/A
	*2	0.112	0.082 - 0.147	45	N/A
	*4	0.002	0.000 - 0.007	1	N/A
	Total	1.00		402	N/A
Phenotypes	Genotypes
EM	*1*1	0.403	0.338 - 0.468	81 (%)	79.6131a (%)
	*1*17	0.304	0.239 - 0.363	61 (%)	64.8233a (%)
	*2*17	0.07	0.04 - 0.104	14 (%)	11.5298a (%)
	*4 *17			0 (%)	0.2562a (%)
	Total	0.776		156
IM	*1*2	0.145	0.100 - 0.194	29 (%)	28.3208a (%)
	*1*4	0.004	0.00 - 0.015	1 (%)	0.6293a (%)
	Total	0.149		30 (%)
UM	*17*17	0.07	0.04 - 0.104	14 (%)	13.1952a (%)
	Total	0.07		14
PM	*2*2	0.004	0.00 - 0.015	1 (%)	2.5186a (%)
	*2 *4			0 (%)	0.1119a (%)
	*4 *4			0 (%)	0.0012a (%)
	Total	0.004		1
	Overall Total	1.00		201	200.9994

Abbreviations: PM, Poor metabolizer; IM, Intermediate metabolizer; EM, Extensive metabolizer; UM, Ultra rapid metabolizer, N/A = not applicable

^a Represents no statistically significant difference between actual number and expected number by Hardy- Weinberg law in genotype distribution

Subjects were divided into 4 phenotypes for CYP2C19 polymorphism: EM, IM, UM, and PM. The distribution of CYP2C19 phenotypes were 77.6%, 14.9%, 7% and 0.5% for EM, IM, UM, and PM respectively. When comparing PM, IM, EM, and UM distribution of CYP2C19 phenotypes, we found a significant difference in phenotype distribution among four groups (p < 0.05). There were no significant associations between phenotypes and gender in our subjects (Table 2). The expected frequencies of CYP2C19 genotypes in the Saudi Arabian population were found not to deviate from the Hardy-Weinberg equilibrium. The majority of subjects included in the study were from the central region of Saudi Arabia (84%) followed by South region (8%), East and West regions (3%) each and North region (8%). The genotyping data for the Saudi Arabian sample was compared with previously studied populations to determine differences between populations (Table 3). The allele frequency of CYP2C19*2 in our study (11.2%) was not statistically different (p>0.05) from the previous study conducted by Goldstein JA et al on Saudi population ²⁰. Also the allele frequency of CYP2C19*2 in the Saudi Arabian population (11.2%) was not statistically different (p>0.05) from that of other European ethnic groups. For example, a similar distribution was found in Romanian (13.7%), Danish (15.0%), German (15.2%), Russian (11.4%), Italian (11.9%), and Portuguese ethnic groups (13.0%) ^{21, 22, 23, 24, 25, 26}. In addition, we did not find any significant differences (p >0.05) in the CYP2C19*2 allele frequency when comparing our population with other Middle East ethnicities such as Iranian (14% ), Israeli (15%), Turkish (12%), Lebanese (13%) and Egyptian people (10.9%) ^{29,30,31,32,33}. In East and South Asian groups (Chinese, Japanese, Koreans and Thai), the CYP2C19*2 allele frequencies (range: 24.9%-29.0%) were significantly higher than seen in the Saudi Arabian population (p<0.05) ^{34, 35, 36, 37, 38}. People of African origin (Tanzanians and African‑Americans) also had a significantly higher frequency of the CYP2C19*2 allele (average of 18%) compared to the Saudi group (11.2%; p< 0.05) ^{39, 40, 41, 42}. South American ethnic groups (Bolivians and Columbians) have a slightly lower CYP2C19*2 allele frequency (average of 8%) than our Saudi population, although there were statistically significant differences (p<0.05) ^{43, 44}.

Table 2.

Association between phenotypes and subject's average age and gender and phenotype distribution in the Saudi population sample (n=201)

Phenotypes	PM	IM	EM	UM	P-value
Phenotype distribution	1	30	156	14	<0.05
Average age (y)	62	41.2 ± 18.1	38.7 ± 14.9	38.4 ± 13.4	>0.05
Gender (M/F)	1 / 0	16 / 14	74 / 82	5 / 9	≥0.55

Abbreviations: PM, Poor metabolizer; IM, Intermediate metabolizer; EM, Extensive metabolizer; UM, Ultra rapid metabolizer

Table 3.

Comparison of allele frequencies between Saudi Arabian and other populations

Population	Sample size	CYP2C19 allele frequency, %			Reference
		*2	*4	*17
Saudi Arabian	201	11.2	0.2	25.7	Our study
Saudi Arabian	97	15	-	-	20
Romanian	200	13.7	0.25	-	21
Danish	276	15.0	-	20.1	22
German	237	15.2	-	25.5	23
Russian	290	11.4	-	-	24
Italian	360	11.9	-	-	25
Portuguese	153	13.0	-	-	26
Greece	283	-	-	19.6	27
Polish	125	11.6	-	27.2	28
Israeli	140	15.0	-	-	29
Iranian	200	14.0	-	-	30
Turkish	404	12.0	-	-	31
Lebanese	161	13.0	-	-	32
Egyptian	247	10.9	-	-	33
Chinese	384	24.9++	-	1.2++	34
Thai	774	29.0++	-	-	35
North Indian	300	26.2++	-	-	36
Japanese	265	27.9++	-	1.3++	37
Korean	271	28.4++	-	1.5++	38
Ethiopian	114	14.0	-	-	39
Tanzanian	251	18.0	-	-	40
African-American	236	18.2	-	-	41
African-American	114	-	-	21.0	42
Columbian	198	8.7	-	-	43
Bolivian	778	7.8	-	-	44

Represents either not tested or published data not available

⁺⁺ Represents statistically significant difference in comparison with current data (p <0.05)

There is very limited published data available from other populations on mutant alleles *4 and *17 making it difficult to compare our data with other ethnicities. We have no data from most of the other populations for the frequency of the CYP2C19*4 allele to compare except for the Romanians (0.25%), and they were comparable to our Saudi population (0.2%) ²¹.

The allele frequency of CYP2C19*17 in the Saudi Arabian population (25.7%) was not statistically different (p>0.05) from than that seen in European ethnic groups; a similar distribution was found in Danish (20.1%), German (25.5%), Greek (19.6%), and Polish (27.2%) populations ^{22, 23, 27, 28}. In East and South Asian groups (Chinese, Japanese, and Koreans), the CYP2C19*2 allele frequencies (range: 1.2% -1.5%) were significantly lower than seen in the Saudi Arabian population (p<0.05) ^{34, 37, 38}.

Discussion

The CYP2C19 mutant alleles *4 and *17 have not been studied well in the worldwide population. This study was a first attempt to determine the distribution of CYP2C19 mutant alleles *4 and *17 in Saudi Arabia, as there were no such data available on our population.

Our study result indicates that approximately 85% of the Saudi population (16.5 million), by extrapolation, would be considered as extensive metabolizers (EM +UM) for CYP2C19 drug substrates. Studies have shown that CYP2C19 genetic variation represents a key factor influencing the pharmacokinetics of the drugs that are substrates to CYP2C19. A profound acid suppression and high peptic ulcer healing rate was noted in Japanese and some Caucasians patients with CYP2C19 *2 *2 and*4 *4 mutations (PM phenotype) with PPI treatments ^{18, 45}. Other studies have shown that EMs metabolize PPIs at a rate that requires much higher doses than PMs (up to four times) to reach similar serum concentrations and effects ^{8, 9}. Based on current predictive model, one would extrapolate that treatment of peptic ulcer disease with conventional doses of PPIs in the Saudi population who are predominantly (77.6%) EM (*1/*1 genotype or subjects with*17 mutations) would produce a reduced acid-inhibitory effect. This may lead to therapeutic failure in our population however; further studies are needed to explore such effects.

We have described in the introduction some clinically important antiepileptic drugs, tricyclic antidepressants, selective serotonin reuptake inhibitors and benzodiazepines that are all substrates to CYP2C19 ^{11,12,13,14,15}. Our Saudi population with EM (77.6%) and UM (14%) theoretically will have significantly decreased drug exposure as the concentrations of these drugs may decrease by rapidly converting into inactive metabolites with a possible reduction of their therapeutic effects. Prior to initiating treatment with such drugs, CYP2C19 genotyping could be a reasonable approach, with respect to optimizing dosage adjustments, improving treatment efficacy, and optimizing treatment cost effectiveness. Also during the treatment with such drugs, therapeutic drug monitoring can improve treatment efficacy.

Clopidogrel is an inactive prodrug that requires hepatic bioactivation via CYP2C19 to exert its effect. This process is hindered in PM and decreases the production of the active metabolite ^{14, 15}. The risk of cardiovascular events is increased in patients who are PM (carrying at least one CYP2C19*2 allele) despite patients receiving adequate doses of an antiplatelet agent, clopidogrel ^{12, 13, 14, 15}. Since our population comprise of only 0.5% PM, they will be least affected by clopidogrel responsiveness, however additional studies are needed to recognize such effects in our population.

In conclusion, CYP2C19*1 was the most frequently identified allele (62.9%) and CYP2C19*17 was the most commonly identified variant allele (25.7%) in a sample of 201 ethnic Saudi Arabians. The predicted distributions of CYP2C19 phenotypes were 77.6% for EMs and only 0.5% for PMs. This finding has important clinical implications for the use of CYP2C19 metabolized medications in the Saudi population. Further studies are needed to explore the clinical effect of different drugs in relation to CYP2C19 polymorphic genotype distribution in the Saudi population. Our study result can also be used as a benchmark for future comparisons with other populations for the prevalence of CYP2C19 variant alleles.