Introduction

The cytochrome P450 enzymes (CYP) play a central role in the metabolism of many therapeutic agents. Differences in the activities of these enzymes are thought to be responsible for individual variability in response and/or toxicity to numerous drugs. Among the CYP enzymes many isoforms exhibit genetic polymorphisms; examples include 1A1, 2C9, 2C19, 3A4, and 3A5. Different populations and different ethnicities differ greatly in the genotype and allele frequencies of many of the metabolizing enzymes including the CYP enzymes.

The CYP1A1 gene, located on chromosome 15q24.1. It comprises seven exons and six introns and encodes a 512–amino acid protein [1]. It plays a key role in phase I metabolism of polycyclic aromatic hydrocarbons to their ultimate DNA-binding forms [2]. A great part of the inter-individual and inter-ethnic differences in relation to xenobiotic effects is now attributable to genetic differences in their metabolism [35].

Three restriction fragment length polymorphic (RFLP) variants have received the most attention: m1 (MspI RFLP-CYP1A1*2A-(T6235C)-rs4646903); m2 (Bsrd1 RFLP-CYP1A1*2C-(A4889G)-rs1048943- Ile462Val); and m3 (MspI RFLP-CYP1A1*3-(T5639C)-rs4986883). The rare Val and M2 alleles of the CYP1A1 gene have been associated with increased individual cancer risk [6, 7]. Ethnic differences have been observed in the distribution of the CYP1A1 polymorphic genotypes [8] with the m1 and m2 polymorphisms being much more frequent in Asians compared to Caucasians and African-Americans [9, 10].

CYP2C9 and CYP2C19 enzymes constitute 50 %, and 16 %, respectively of the CYP2C subfamily [11]. The two genes are located on chromosome 10q24 [12]. CYP2C9 is a phase I drug-metabolizing CYP450 enzyme isoform [13], it is responsible for the catalyses of the oxidation and metabolic clearance of up to 15–20 % of clinically important drugs [14, 15] including phenytoin, warfarin, tolbutamide, losartan, and a large number of nonsteroidal anti-inflammatory-drugs [16, 17]. Changes in amino acids in this enzyme can affect its activity and produce individual variability in the concentration and/or dosage requirements of prototypic CYP2C9 substrates as warfarin and clopidogrel, respectively [18, 19].

The gene coding for the CYP2C9 enzyme is highly polymorphic, including functional variants of major pharmacogenetic importance. Thirteen variant alleles have been reported to date [13]; however, only two amino acids substitutions, the non-synonymous variants CYP2C9*2 (rs1799853-Arg144Cys) and CYP2C9*3 (rs1057910-Ile359Leu), are recognized in humans as main CYP2C9 variants [17, 20]. CYP2C9*2 and CYP2C9*3 have reduced catalytic activity compared with wild type (CYP2C9*1) [21, 22]. The frequency of CYP2C9 allelic variants has been reported to differ among Caucasian, African, and Asian populations. The allele frequencies of CY2C9*2 and CYP2C9*3 tend to be greater in Caucasian populations than in African-American and Asian populations [2024]. The CYP2C9*2 allelic variant was not found in East Asians including Chinese and Japanese [23, 24].

CYP2C19 metabolizes several therapeutically important drugs, namely omeprazole, lansoprozole, imipramine, clopidogrel and diazepam [25, 26]. Among different variants, two alleles with separate mutations have been associated with defective enzyme in Caucasian. The first is CYP2C19*2 (rs4244285-G681A/C), which has a mutation in exon 5 causing a splice site. The other variant allele is CYP2C19*3 (rs4986893-G636A) with a point mutation in exon 4 producing a premature stop codon. The most commonly mutated allele is CYP2C19*2 in Caucasian poor metabolizers (PMs) [27] whereas, CYP2C19*3, is rare among Caucasian subjects [28].

The CYP3A locus consists of four genes, CYP3A4, CYP3A5, CYP3A7, and CYP3A43, all of which reside in a 231-kb region of chromosome 7q21.1 [29]. In adults, CYP3A4 and CYP3A5 are predominant among the four known isoforms whether in liver or intestine [30]. All genes show high degree of polymorphic expression [31]. Genetic variation within the CYP3A genes may contribute to inter-individual variability in drug metabolism. It has been suggested up to 60 % of the variability in CYP3A4 activity may be because of a genetic component [32]. The most prevalent polymorphism in CYP3A4 is (CYP3A4*1B) (rs2740574) occurs in the 5 V flanking region of the gene, it involves an A > G transition at -293 position from the transcription start site [33]. Large ethnic differences have been reported in this single nucleotide polymorphism (SNP) [30, 33].

CYP3A5 is expressed in a polymorphic manner in 10–29 % of adult livers [34, 35]. Several polymorphic variants in CYP3A5 (chromosome 7q22.1) appear to have a functional effect on CYP3A5 activity. CYP3A5*3 (rs776746) polymorphism occurs in intron 3 of CYP3A5, creating a cryptic splice site leading to the inclusion of a novel exon, and ultimately a premature stop codon, which leads eventually to very low amount of CYP3A5 protein [35, 36]. The CYP3A5 allele 6986A > G (CYP3A5*3) is associated with reduced clearance of drugs such as simvastatin, lovastatin, midazolam, cyclosporine and tacrolimus [33, 37, 38].

The expression of CYP3A5 can be reduced by quite a few other alleles in addition to CYP3A*3. The CYP3A5*6 (rs10264272) is a 14690G > A synonymous mutation (Lys–Lys) that causes the formation of a splice variant mRNA. Exon 7 is deleted, resulting in a frame shift and a truncated protein [36]. The frequency of this SNP have been found to be 17 % in African–Americans and is absent in Asians [33].

The frequencies of the important allelic variants in the CYP1A1, CYP3A4, CYP3A5 CYP2C9 and CYP2C19, genes have been studied in many ethnicities, and the data, so far, show striking interethnic variation in the distribution of these variants. However, no information is available for the Jordanian population. In the present study, we report here the allele frequencies of CYP1A1 (*2A and *2C), CYP2C9 (*2 and *3), CYP2C19 (*2 and *3), CYP3A4*1B, and CYP3A5 (*3 and *6), in a sufficiently large sample of the Jordanian population, providing a basis for future clinical studies concerning variability in the response and/or toxicity to drugs known to be substrates for these enzymes and proteins. To our knowledge this is the first genotyping study of cytochrome P450 polymorphisms in the Jordanian population reported to date.

While native Jordanians are mostly descended from people of villagers and Bedouin descent originating in the Arabian Peninsula [39], ethnically, the Jordanians represent a mixed stock. Most of the population is Arab (approximately 98 %) with 1 % of the population is Armenian, and another 1 % is Circassian. There are also small Kurd, Druze, and Chechen minorities [40, 41].

Methods

Subjects

Apparently healthy unrelated Jordanian subjects that participated in different genotyping studies were selected from many hospitals (Jordan University hospital, Al-Bashair Hospital and Royal medical service). Each subject gave a sample of about 1 ml of blood after detailed explanation of the purpose of the study; a consent form was also obtained from each subject. Genomic DNA was isolated from the blood using Wizard Genomic® (DNA purification kit) (Promega Corporation, USA). The isolated DNA samples were prepared for genotyping.

Genotyping

Polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) that were used to identify different single nucleotide polymorphisms (SNPs) (CYP1A1 (m1 (*2A) and m2 (*2C)), CYP2C9 (*2 and *3), CYP2C19 (*2 and *3), CYP3A4*1B, CYP3A5 (*3 and *6), primers, restriction enzymes and the condition of PCR are summarized in Table 1. Findings of the PCR-RFLP were validated by:

Table 1 List of the primers, PCR conditions and restriction enzymes used in the identification of the designated SNPs
  • *Every time PCR reactions were done a negative control was run simultaneously. A negative control contains all PCR components except the DNA template.

  • *The sequence of the reverse, forward or both primers of (CYP2C9, CYP2C19, CYP3A4, and CYP3A5) differed from original nucleotide sequence at that region of the genomic DNA, to create an internal control site (ICS). The ICS serves to make sure that when the restriction enzyme dose not cut, it does so because of the sequence and not because it is not functioning properly. If the restriction enzyme is functional, it will always cut the PCR product at ICS irrespective of sequence of SNP.

  • *Around 15 % of all samples were repeated to confirm findings of the PCR-RFLP.

  • *The PCR-RFLP results (CYP3A4, and CYP3A5) were confirmed by direct DNA sequencing using BigDye Terminator Cycle Sequencing on 3730xl DNA sequencer (Macrogene® Co., Korea).

Statistical analysis

Allele and genotype frequencies for different alleles among Jordanian population were estimated from the results of the above PCR-RFLP test. This estimation was according to formulas reported previously [42]. Genotype and allele frequency were matched to expectation by Hardy-Weinberg Equilibrium, differences in allele frequencies between Jordanians and other ethnic populations were assessed using χ2 test [43]. A p value below 0.05 was considered statistically significant throughout the population comparisons. Table 2

Table 2 Demographic data for all subjects enrolled in the presents study

The minimum sample size needed to accomplish the objectives of the study was calculated based on the following equation [44]

$$ {\text{N}} = {\text{P}}*{\text{Q }}({\text{Z}}_{\alpha } + {\text{ Z}}_{\beta } )^{ 2} /\Updelta^{ 2} $$

where N is the sample size needed; P is the reported frequency of the allele at that SNP; Q is (1 − P); Zα is the critical value at type 1 error; Zβ is the critical value at type 2 error; and Δ is the acceptable margin of error.

Based on previously reported frequencies among Caucasians and assuming α = 5 % and β = 20 % and Δ = 4 to 5 %, then the minimum sample size needed for estimation of the less-frequent allele type of CYP1A1, CYP2C9, CYP2C19, CYP3A4 and CYP3A5 was 242, 218, 143, 156, and 207 respectively.

Results

CYP1A1

The frequencies of CYP1A1 alleles among 290 Jordanian subjects were as follows; CYP1A1*1 (76.4 %), CYP1A1*2A (16.6 %) and CYP1A1*2C (7.1 %). The distribution of CYP1A1 genotypes is summarized in Table 3.

Table 3 The observed and expected genotype distribution of the tested variants among the Jordanian subjects according to Hardy–Weinberg equilibrium

CYP2C9

The frequencies of CYP2C9 alleles among 263 Jordanian subjects were as follows; CYP2C9*1 (79.7 %), CYP2C9*2 (13.5 %), and CYP2C9*3(6.8 %). The distribution of CYP2C9 genotypes is summarized in Table 3.

CYP2C19

The frequencies of CYP2C19 alleles among 158 Jordanian subjects were as follows; CYP2C19*1 (87.65 %), CYP2C19*2 (12.35 %), and CYP2C19*3(0). The distribution of CYP2C19 genotypes is summarized in Table 3.

CYP3A4*1B

Among 173 subjects tested, 161 were homozygous for wild type allele (CYP3A4*1/*1, 93.1 %) and 12 individuals were heterozygous for variant allele (CYP3A4*1B/*1, 6.9 %). No homozygote was detected in the study samples. Thus, the allele frequency of the CYP3A4*1B variant allele among Jordanian population is 3.5 % (Table 3).

CYP3A5

The frequencies of CYP3A5 alleles among 229 Jordanian subjects were as follows; CYP3A5*1 (7.2 %), CYP3A5*3 (92.58 %), CYP3A5*6 (0.22 %). The distribution of CYP3A5 genotypes is summarized in Table 3.

All loci examined were in consistent with HWE, with the exception of one loci CYP3A5*3 (Table 3).

Discussion

CYP1A1

Ethnic differences in CYP1A1 polymorphisms have been previously reported [35, 45] and our results corroborate the ethnic difference in the allelic frequency of CYP1A1 between Caucasians and Africans (Table 4). The frequency of CYP1A1 mutations varies among different ethnic groups, with a high prevalence of m1 and m2 in Far East Asian populations and a low rate in Caucasians [4648]. The m2 is found always linked with m1 [46, 48]. CYP1A1 allele *2B, containing mutations m1 and m2, was identified to place the carrier at increased risk for lung cancer [46, 49, 50] and both (*2A and *2C) are also known to increase risk for lung cancer [6, 7]. In the current study, both polymorphisms *2A/*2C were found to be carried in 34 subjects (using RFLP). All examined loci of CYP1A1 were consistent with HWE.

Table 4 Comparison of allele frequencies of CYP1A1 reported from different ethnic populations

Our results show that the prevalence of CYP1A1*2A, and *2C were closest to the one reported for the Turkish population (Table 4). The observed genotype and allele type is quite different from Asians and Africans.

CYP2C9

The frequency of the CYP2C9*2 allele in the Jordanian population was 13.5 % which is in a range comparable with other Caucasian populations (Table 5); 8 % in American, 12.5 % in British [20], and 10.6 % in Turkish [51]. It was reported that the CYP2C9*2 allele occurs at a significantly lower frequency in the African-American population (1 %) [18]. In contrast, the CYP2C9*2 allele was reported to be absent or at least very rare in the East Asian populations [52, 53].

Table 5 Comparison of allele frequencies of CYP2C9 reported from different ethnic populations

The CYP2C9*3 allele occurred with a frequency of 6.8 % in the Jordanian subjects. This finding is similar to that of other Caucasian populations in which the frequency of the CYP2C9*3 allele was reported to be 6 % in American, 8.5 % in British and 10.0 % in Turkish (Table 5). Lower frequencies were reported for the CYP2C9*3 variant in East Asian populations; 1.1–2.1 % in Japanese [52] and Korean [53] and the lowest frequency of CYP2C9*3 was reported in African-Americans (0.5 %) [20].

Warfarin is an oral anti-coagulant with narrow therapeutic index. It has been recommended that the initial dose and dose adjustment should be based on the genetic polymorphism of CYP2C9 in addition to VKORC1 [54]. Patients with CYP2C9*2 and *3 variants have longer times to dose stabilization and are at higher risk of serious and life-threatening bleeding [55]. The high prevalence of CYP2C9*2and *3 alleles among Jordanians clearly indicates the benefits of genetic testing of CYP2C9 in warfarin dosing, allowing physicians to identify patients at high risk for warfarin hyper responsiveness. In addition, our data could be extrapolated to other drugs metabolized by the CYP2C9.

CYP2C19

The genetic polymorphism of CYP2C19 has been shown to have the most interethnic variation with PM frequency ranges from 2 to 7 % in Caucasians, to 14–25 % in Asians [51]. In the present study, we found that the incidence of CYP2C19*2 among Jordanians (12.3 %) was similar to that found in other Caucasian populations, in Europe, and in Saudi Arabia [56]. Higher frequencies of CYP2C19*2 were reported for Asian populations [56], Africans and African Americans [51] (Table 6)

Table 6 Comparison of allele frequencies of CYP2C19 reported from different ethnic populations

The presence of CYP2C19*3 in one Jordanian subject (0.0) was consistent with other Caucasian populations (Table 6). CYP2C19*3 has been regarded as an Asian mutation and accounted for the remaining alleles in Asian PMs [51].

The importance of genotyping CYP2C19 *2 and *3 has been highlighted by the United States Food and Drug Administration (FDA) which recommends that Clopidogrel therapy in populations with a high prevalence of those ‘‘risk’’ alleles be carefully administered [57]. Moreover, some investigators have shown associations between the CYP2C19 polymorphism and certain types of cancers (e.g., esophagus, stomach, lung and bladder cancers)[58, 59].

CYP3A4*1B

CYP3A4*1B allele frequency varies among different ethnic groups and has lower frequency among Caucasians (Table 7). Recently, CYP3A4*1B allele was found to be associated with lower CYP3A4 metabolism capacity. This lower capacity is explained by reduction of CYP3A4 expression due to presence of A-392G SNP which suggests that CYP3A4*1B diminishes the binding affinity of a transcription factor, resulting in lower transcriptional activity and therefore, lower enzyme activity [60].

Table 7 Comparison of allele frequencies of CYP3A4*1B reported from different ethnic populations

The current study shows an allelic frequency of this SNP of 3.5 % among Jordanian population. Another study had been conducted among Jordanian population and found an allelic frequency of this SNP with 11.1 % [61] which is consistent with that reported previously for Caucasians (Table 4). The CYP3A4*1B allelic frequency varies among different ethnic groups: 0 % in Chinese and Taiwanese; 9.3–11 % in Hispanic Americans [33]; 2–9.6 % in Caucasians [62]; and 35–67 % in African Americans [30, 36].

There is debatable association of this variant with prostate cancer [63], chemotherapy-related leukemia [64], breast or ovarian cancer [65]. The presence of the CYP3A4*1B allele was shown to increase risk for Small Cell Lung Cancer in women [66].

CYP3A5

As mentioned previously, and in contrast to CYP3A4, the CYP3A5 enzyme is known to be expressed in only a small percentage of Caucasian individuals (10–30 %) and this has been linked to a common transition in intron 3 and results in nonfunctional protein [36].

Due to the importance of this SNP (CYP3A5*3) as a determinant for the expression of CYP3A5 gene in Caucasians, the current study estimated the allelic frequency of this SNP. In the Jordanian population the allelic frequency of this SNP was 92.58 %, which is fully consistent with that for other Caucasian populations [30, 67]. The prevalence of this SNP was 89–91 % in Caucasian-Americans; 92–94 % in European-Caucasians; 60–66 % in Hispanics, 71–75 % in East Asian populations (Chinese, Japanese, Koreans); 55–65 % in South Asians (Malay and Indians); and 29–35 % in African-Americans [33] (Table 8).

Table 8 Comparison of allele frequencies of CYP3A5 reported from different ethnic populations

Contrary to examined HWE in all previously tested loci, we found a significant difference between the observed and the expected frequencies with regard to CYP3A5*3. The findings could be spurious. If only one sample was *1/*3 instead of *1/*1, the χ2 will change from 8.5 to 4.1 and the p value will change from 0.003 to 0.04. Alternatively, this evolution can be explained by genetic drift that occurs when the population size is limited and therefore by chance, certain alleles increase or decrease in frequency. Allele frequencies in small populations do not generally reflect those of larger populations since too small of a set of individuals cannot represent all of the alleles for the entire population [68, 69]. Only a single locus out of nine examined loci deviated from HWE.

CYP3*6 allele, which encodes for aberrantly spliced CYP3A5 enzyme was found in 0.22 % of Jordanian population resembling results found in other Caucasian populations [30, 67].

Conclusion

The current study has led to the determination of common allelic variants of a number of important metabolizing enzymes namely CYP1A1; CYP2C9; CYP2C19; CYP3A4; CYP3A5 in Jordanian population. Some of these variant alleles, to our knowledge, are being reported for the first time among the Jordanian population. The frequencies obtained are comparable to data previously reported in other populations of Caucasian origin but differ from that observed in African and Asian populations. Comparably, a mitochondrial genetics analysis revealed that Jordanians, carry genetic variants similar to other Caucasians [70]. This is of importance as it has been demonstrated in many studies that allele frequencies of the metabolic genes are not randomly distributed throughout the human population but follow diverse ethnic and/or geographic-specific patterns [4, 7173]. The findings of current study supported by other studies all over the globe provide merits to suggest the use of genetic polymorphisms of CYP450 SNPs as markers for ethnicity and ancestral origin.