Characterisation of CYP2C8, CYP2C9 and CYP2C19 polymorphisms in a Ghanaian population

Variant allele frequencies of many pharmacogenetically-relevant polymorphisms have been demonstrated to vary greatly between populations of different countries. However, some areas of the world especially indigenous African populations have scarcity information in the current pharmacogenetics research [1, 2] Cytochrome P450 2C (CYP2C) subfamily of enzymes form 18-30% of human CYPs and metabolises nearly 20% of all therapeutic drugs commonly prescribed in clinical practice [3]. CYP2C gene is made up of four isoforms, CYP2C8, CYP2C9, CYP2C18 and CYP2C19 which are located together on chromosome 10q24. This CYP2C subfamily of enzymes constitutes 15-20% of the CYP protein in the liver [3] and exhibit genetic polymorphisms leading to differences in activities of these enzymes. Genetic polymorphisms of this CYP2C subfamily of enzymes are thought to influence both efficacy of drugs and the likelihood of ADRs [4].

CYP2C8, CYP2C9, and CYP2C19 enzymes constitute 26%, 50%, and 16% respectively of the CYP2C subfamily [5]. They are polymorphically expressed with variable allele frequencies among different ethnic populations [2, 6, 7], Some of these CYP2C variant alleles have been associated with either an increased or decreased metabolism of several clinically important drugs [3, 5, 8, 9]. Some of these drugs have narrow therapeutic index and have been involved in the development of adverse side effects. Variant allele frequencies in different populations do have implications for adverse effects. Patients carrying either the homozygous or the heterozygous variant alleles of CYP2C9 are at risk of acute gastrointestinal bleeding when treated with NSAID that are substrates of CYP2C9 [10]. CYP2C9 polymorphisms have been associated with an increase risk of bleeding in patients treated with standard doses of warfarin [11] while phenytoin toxicity has also been reported in some patients [12]. The U.S. FDA has added genetic information to prescribing information on some of these drugs such as Warfarin, carbamazepine and codeine to assist prescribers safely achieve therapeutic doses http://www.fda.gov/Drugs/ScienceResearch/ResearchAreas/Pharmacogenetics/ucm083378.htm.

Our aim is to genotype pharmacologically-relevant CYP2C8, CYP2C9 and CYP2C19 variants in normal Ghanaian population to address the lack of data in indigenous African population. Allele frequencies of major Ghanaian tribes were also compared to determine the genetic variation within the country. Allele frequencies were also compared with previously reported frequencies in published literature to determine inter ethnic variation.

Intra-population variations were determined for five major tribes within the Ghanaian population were self-identified as Akan, Fanti, Ewe, Northern, and Ga tribes. Intra-population data was similar for all the SNPs analysed except for CYP2C8*2 variant which was significantly higher in the Northern tribe (p = 0.025). CYP2C9 *11 variant was detected in only the Ewe and Ga tribes at frequencies of 5% and 3% respectively. The Ghanaian population was found to a large extent to be genetically homogenous.

CYP2C8*2 variant detected at 17% in this study and has a reduced potential of metabolising paclitaxel and arachidonic acid. The CYP2C8*2 variant allele results in 15% reduction in activity towards paclitaxel metabolism in vitro compared to the wild-type allele (CYP2C8*1) [6] while the CYP2C8*3 variant allele showed 50% reduction in activity to paclitaxel metabolism [14]. CYP2C8*2 and CYP2C8*3 variant genotypes can lead to poor metaboliser (PM) phenotype which could potentially cause drug toxicity. Although CYP2C8*2 and CYP2C8*3 variant alleles have been associated with reduced enzymatic activity only the CYP2C8*2 variant was detected in the current study.

Five pharmacologically relevant CYP2C9 variants were surveyed in the current study and the results compared with other studies (Table 3). The CYP2C9*2 and CYP2C9*3 variant alleles were not found in this study population and this was consistent with an earlier report in a Beninese population [15]. This observation, however, was different from frequencies obtained within the African American and Ethiopian populations. The CYP2C9*2 and CYP2C9*3 variant alleles were reported at 3.3% and 2.3% respectively in African American populations [18] and at 4.3% and 2.3% in the Ethiopian population [24]. In contrast, CYP2C9*2 variant allele has been reported at 8-19% and the CYP2C9*3 variant allele at 3.3-17% among the Caucasian population [2]. The CYP2C9*2 variant allele was rarely seen in Asian populations while the CYP2C9*3 variant allele was prevalent at 1.1-6.8% within the Asian populations [2].

The CYP2C9*5 variant allele was not detected in the current study but has been reported at a low level in Beninese (2%), but slightly higher than the Tanzanian (0.8%) and African American (1%) populations [15, 18]. This variant allele is not reported among the Caucasian populations [2]. The CYP2C9*11 variant allele was detected at 2% frequency within this study population. This frequency was similar to data from Beninese (3%) and African American (2%) populations [15, 18]. This variant allele was rare among the Caucasian [15] and was not reported in the Asian population.

CYP2C9*2 and CYP2C9*3 polymorphisms have been shown to cause a decrease in enzymatic activity of 30% and 80% respectively [25, 26] and have been shown to have clinical implications for patients carrying these alleles [27]. The CYP2C9*3 variant allele has been reported to cause the largest reduction in metabolic capacity for many CYP2C9 substrates, followed by the CYP2C9*2 variant allele causing an intermediate reduction when compared to the wild-type allele (CYP2C9*1) [28]. Poor metabolisers, mostly CYP2C9*3/CYP2C9*3 carriers, may experience severe toxicity when metabolising CYP2C9 substrates with narrow therapeutic index such as warfarin and phenytoin. Conversely, these PMs may not have an adequate drug response and may experience therapeutic failure when taking a pro-drug such as losartan and cyclophosphamide, that requires bio-activation by CYP2C9 [21]. ADRs to CYP2C9 substrate drugs like warfarin and glipizide are more evident in African/Afro-Caribbean populations than Caucasians, however, CYP2C9*2 and CYP2C9*3 polymorphisms are not common in this ethnic group [29, 30].

CYP2C19*2 and CYP2C19*3 variant alleles are the most characterised alleles of the CYP2C19 gene [31, 32]. The CYP2C19*2 variant allele was the most common allele found in the current study as in other studies (Table 3). Significant inter-ethnic difference has been reported for CYP2C19*2 variant allele in the published literature [15, 33]. It occurred at a lower frequency of 6% within this study population compared to 13% in the Beninese [15] and Zimbabwean (13%) populations [34]. CYP2C19*2 variant allele has been reported at similar frequencies in Caucasian populations (9%-14%) [33, 35]. The prevalence of the CYP2C19*2 variant allele has been reported at significantly high frequencies within the African American (25%) and the Asian populations (23-30.5%) (p = 0.003; and p = 0.001, respectively) [2, 36]. CYP2C19*3 variant allele was not detected in this study which is consistent with findings within other African populations [15, 34]. Although the CYP2C19*3 variant has been rarely reported in Caucasian populations [15, 33], it has been found in Asian populations at a frequency of 10% [33, 37].

CYP2C19*2 variant detected in 6% of the study population were predicted to be PMs for CYP2C19 substrates and are potentially more prone to the possibility of suffering from adverse drug effects than extensive metabolisers (EMs) when taking therapeutic doses of drugs such as warfarin, diazepam and omeprazole. CYP2C8*5, CYP2C9*6 and major CYP2C18 gene variants were not analysed due to shortage of DNA sample.

Although the ultimate aim of pharmacogenetics is to lead to personalised treatment of individual patients based on their genetic profile, the identification of polymorphisms within populations can also be useful in improving the quality of healthcare in that specified population. Investigating the genetic variation within genes encoding drug metabolising enzymes (DMEs), drug transporters and drug targets within different populations is becoming increasing important because of the drug-drug interactions that results from enzyme inhibitions or inductions. This type of information is increasingly becoming useful for improving drug therapy and explaining inter-individual and inter-ethnic differences due to drug response [38, 39]. It is also being used to predict and explain ADRs which cause 7% of all hospital admissions and 4% withdrawal of new medications [40, 41]. It is becoming increasing important to derive data from different populations to build a database which can then be used in epidemiological investigations to better understand the genetic risk factors which affect many diseases and to be in a better position to predict them in the future.

The current study has led to the determination of allelic variants of CYP2C8, CYP2C9 and CYP2C19 in a Ghanaian population. Some of these variants alleles to our knowledge are being reported for the first time among the Ghanaian population. The frequencies obtained are comparable to data previously reported in other populations of African origin but differ from that observed in Caucasian and Asian populations. These results provide additional information on polymorphisms of this CYP2C subfamily of enzymes in an indigenous African population which is scarce in published literature.