Frequency of C3435T single nucleotide MDR1 genetic polymorphism in an Asian population: phenotypic–genotypic correlates

Abstract

Aims

To investigate the frequency of the single nucleotide polymorphism C3435T in exon 26 of the MDR1 gene in Asians and to determine the functional significance of this SNP with the clinical pharmacokinetics of oral cyclosporin (Neoral) in 10 stable heart transplant patients.

Methods

The MDR1 C3435T polymorphism was investigated in 290 healthy Asian subjects (98 Chinese, 99 Malays and 93 Indians). We also compared the MDR1 polymorphism between the Asian population studied here and the published data on Africans and Caucasians. The clinical relevance of this SNP on oral bioavailability of a known P-gp substrate, cyclosporin, was assessed in 10 stable Chinese heart transplant patients.

Results

The homozygous TT genotype was observed in 32%, 28% and 43% of Chinese, Malays and Indians. The homozygous CC genotype was found in 25% of Chinese and Malays compared with 18% of Indians. The Indians had a lower frequency of the C allele [0.38 (0.31–0.45)] compared with the Chinese [0.46 (0.39–0.53)] and Malays [0.48 (0.42–0.55)]. Chi-squared test showed that the distribution of allele frequencies between the Malays and Indians differed significantly (P = 0.04). In this Asian population, the overall distribution of genotypes (CC, CT and TT) and allele frequencies were significantly different from those in Africans (P < 0.001). The results were also significant when the Chinese, Malays and Indians were compared separately with the African group (P < 0.001). Compared with the Caucasian data, the overall distribution of genotype and allele frequencies in the Asian population were also significantly different (P ≤ 0.05). However, when each Asian ethnic group was compared separately with the Caucasians, only the Indians were found to be significantly different (P ≤ 0.004). Genotypic–phenotypic correlations of this SNP were assessed in 10 stable Chinese heart transplant patients. The median AUC(0,4 h) was 11% lower in patients with CC genotype compared with subjects with TT genotype. However, the interpatient variability in AUC(0,4 h) was high in patients, especially in those with CC genotype.

Conclusions

The distribution of the SNP C3435T in exon 26 in the Chinese and Malay population was found to be similar to the Caucasians whereas the Indians were different. The Asian population also differed significantly from the African and Caucasian population in the distribution of the C3435T SNP. The low frequency of the T allele in the Indian population implies lower expression of P-gp and may have important therapeutic and prognostic implications for use of P-gp dependent drugs in individuals of Indian origin.

Introduction

P-glycoprotein (P-gp) is a 170-kD plasma glycoprotein that belongs to the superfamily of ATP-binding cassette (ABC) transporters [1]. It is composed of two homologous halves, each containing six transmembrane domains and an ATP/utilization domain, separated by a linker polypeptide. P-gp was discovered by Juliano & Ling [2] in multidrug-resistance (MDR) cancer cells after several reported observations that mammalian cancer cells, selected for resistance to a single cytotoxic agent, displayed crossresistance to a broad spectrum of structurally and functionally unrelated compounds. When expressed in tumour cells, P-gp results in MDR, by causing the active extrusion of a wide range of cancer chemotherapeutic drugs including the Vinca alkaloids, anthracyclines, colchicine, epipodophyllotoxins, actinomycin and the taxanes [3].

In addition to being highly expressed in tumour cells, P-gp is also expressed constitutively in healthy tissues and is found in high concentrations on the apical surfaces of epithelial cells in the liver (bile canaliculi) [4], kidney (proximal tubule), pancreas (pancreatic ductal cell), small intestine and colon (columnar mucosal cell) and adrenal gland [5, 6]. It is also expressed in the capillary endothelium of the brain and testes, suggesting that it may function to limit the entry of certain molecules into specific anatomical compartments [7]. The tissue distribution of P-gp leads to reasonable expectation that P-gp plays a role in defence against xenotoxins by excreting cytotoxic agents into bile, the gastrointestinal tract and urine as well as participating in blood–brain barrier function.

Expression of P-gp concentrations increase longitudinally along the intestine, with lowest concentrations in the stomach and highest concentrations in the colon [8]. P-gp concentrations also show significant interindividual variability, with two- to eight-fold variations found in small intestine biopsies from kidney transplant patients and healthy individuals [9]. The high interpatient variability in P-gp expression can cause substantial variation in both the rate and extent of absorption of orally administered drugs that are P-gp substrates. The complete loss of P-gp from the gut in mdr1a–/– knockout mice supports the important role for intestinal P-gp in the oral bioavailability of drugs which are P-gp substrates. For example, the oral bioavailability of paclitaxel is three-fold higher in mdr1a–/– mice compared with mdr1a+/+ mice [10].

In humans, there are two MDR genes, MDR1 and MDR2 (also known as MDR3). The MDR1 gene is located on the long arm of chromosome 7 and consists of a core promoter region and 28 exons. MDR1 encodes P-gp whereas MDR2 encodes a P-gp that is specific for phosphatidylcholine translocation in cells [11].

MDR1 is polymorphic and 15 mutations, mostly involving noncoding single nucleotide polymorphisms (SNP) in introns and wobble positions, have been identified in the gene. Recently, a functional SNP resulting in a C→T transition has been described in exon 26 (C3435T), where the homozygous T allele was shown to be associated with more than two-fold lower duodenal P-gp expression levels compared with CC samples [12]. The genotype frequency of the homozygous CC genotype was highest among the African population and lowest in the south-west Asian groups. Heterozygous individuals displayed an intermediate phenotype. Although ethnic variation as well as the population frequency of the C3435T transition has been observed in the Caucasian population, we evaluated the C3435T SNP in three ethnic groups: Chinese, Malays and Indians. We also determined the functional significance of the C3435T SNP with the clinical pharmacokinetics of oral cyclosporin (CsA) given as Neoral, a microemulsion formulation (CsA-ME), to 10 stable Chinese heart transplant patients.

Methods

Subjects and identification of MDR1 polymorphisms

All subjects gave written consent prior to study accrual and the protocol was approved by the ethics committee. Blood samples (5 ml) were collected in EDTA vacutainer tubes for DNA extraction and MDR1 genotyping from the three different ethnic groups, Chinese (n = 98), Malays (n = 99) and Indians (n = 93). Purified genomic DNA was extracted from peripheral blood leucocytes using the phenol–chloroform extraction method. A polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) was used for the detection of C3435T SNP. The sequences for the forward primer, MDR1F (5′-TGCTGGTCCTGAAGTTGATCTGTGAAC-3′) and reverse primer, MDR1R (5′-ACATTAGGCAGTGACT CGATGAAG GCA - 3′) were obtained from known sequences of exon 26 (Genbank: J05168 and AC 005068). The PCR assay was performed in a 20 µl reaction volume containing 100 ng genomic DNA, 0.25 µmol l⁻¹ of each primer, 10 × PCR buffer containing 10 mmol l⁻¹ Tris and 50 mmol l⁻¹ KCl, 1.5 mmol l⁻¹ MgCl₂, 200 µmol l⁻¹ each of DNTPs and 1 U Taq DNA polymerase. The PCR conditions consisted of a denaturaturation step at 94 °C for 1 min, annealing at 55 °C for 1 min and elongation at 72 °C for 2 min, followed by a final extension at 72 °C for 5 min. The amplification reaction product was purified using a PCR purification kit (QIAGEN, Valencia, California, USA) for direct purification of PCR amplified products. This was followed by digestion with the restriction enzyme MboI for 2 h at 37 °C. The digested products were then separated on 2% agarose gel.

Neoral pharmacokinetics in heart transplant subjects

Neoral pharmacokinetics were evaluated in 10 Chinese heart transplant patients. All patients gave written informed consent. The patients were stable clinically and had undergone cardiac transplantation at least 6 months before. The mean (± SD) duration of follow-up from the time of transplantation was 5.2 ± 4.4 years, mean age of patients was 47.9 ± 9.8 years and mean weight was 63.5 ± 16.1 kg. The mean serum creatinine value was 127.7 ± 38.6 µmol l⁻¹. The mean creatinine clearance was 58.1 ± 13.5 ml min⁻¹ and was assumed to be stable for the duration of the study. All other biochemical parameters were within normal range.

On the day of the study, all patients took the recommended dose of CsA-ME with a glass of water (200 ml) on an empty stomach. Whole blood samples (3 ml) for AUC analysis were obtained at the following times: predose (C₀), and 1, 2, 3, 4, 6, 12 h (C_1h, C₂h, C₃, C_4h, C_6h, C_12h, respectively) postdose during the first dosing interval. The blood samples were collected in plain vacutainer glass tubes containing EDTA as an anticoagulant. The samples were assayed with the commercially available cyclosporin whole-blood monoclonal antibody fluorescence polarization assay (FPIA; TDx; Abbott Laboratories, Chicago, IL, USA). The inter- and intra-day variability were less than 4% and the assay detection limit was 25 µg l⁻¹.

Pharmacokinetic parameters were determined by noncompartmental analysis using a nonlinear regression program (WinNonLin, version 2.1; Pharsight Inc, Montain View, CA, USA). The linear trapezoidal rule was used to calculate the area under the concentration-time curve from 0 to 4 h [AUC(0,4 h)].

Statistical analysis

The allelic frequencies were calculated based on the observed number of the two different alleles, C and T, derived from genotype data in each ethnic group. The interethnic comparison of observed and expected allele and genotype frequencies was done using the Hardy–Weinberg equation and chi-squared goodness-of-fit analysis. The R × C contingency table analysis was used to compare the interethnic differences in the distribution of alleles and genotypes, and the chi-squared test was used for comparisons of two different ethnic groups. We also compared results of this study with published results in Africans [13] and Caucasians (Germans) [12] using the chi-squared test; 95% confidence intervals were calculated for all observed allele frequencies. The correlation between genotypes and AUC(0,4 h) normalized by dose in Chinese heart transplant subjects was analysed using the Jonckheere–Terpstra test, a nonparametric statistical procedure that considers the quantitative relationship (number of defective alleles) between the levels of a variable. P < 0.05 was taken to be the minimum level of statistical significance. The statistical analysis was done using SAS programme version 6 (SAS Institute, Cary, MA, USA).

Results

The frequency of the homozygous TT variant was 43% in the Indian subjects compared with 18% of them carrying the homozygous CC variant. Almost 50% of the Chinese and Malay population were heterozygous carriers of the C/T variant compared with 39% of Indians (Fig. 1). There were no statistically significant differences observed in the distribution of C and T genotypes among the Chinese, Malays and Indians. A marginally significant difference in allele frequency was found between the Malay and Indian ethnic groups (P = 0.04, Table 1). The observed frequencies of the genotypes in all three ethnic groups of Asian origin were consistent with Hardy–Weinberg equilibrium (P > 0.5)

Figure 1.

Genotype frequencies for the C3435T MDR1 polymorphism in Chinese, Malay and Indian ethnic groups.

Table 1.

Interethnic differences in the observed genotype and allele frequencies of MDR1 polymorphism by ethnic group

		Population			P value
		Chinese (n = 98) (A)	Malays (n = 99) (B)	Indians (n = 93) (C)	A vs B	A vs C	B vs C
Genotype	C/C	0.24	0.25	0.18	0.874	0.243	0.096
	C/T	0.44	0.46	0.39
	T/T	0.32	0.28	0.43
Allele (95% CI)	Cs	0.46 (0.39, 0.53)	0.48 (0.42, 0.55)	0.38 (0.31, 0.45)	0.758	0.102	0.041
	Ts	0.54 (0.47, 0.61)	0.52 (0.45, 0.59)	0.62 (0.55, 0.69)

We also compared the differences in the distribution C and T genotypes and allele frequencies in the pooled Asian subjects (n = 290) with the published data on African subjects (Ghanaian, Kenyan and Sudanese; n = 337) [13] and Caucasians [German; n = 188][12]. The genotype frequency of the homozygous TT genotype ranged from 28 to 43% in Asians compared with 0–6% in the African group [13] and 24% in the Caucasian [12] (Table 1). The allele frequency of wild-type C allele ranged from 0.38 to 0.48 in the Asian subjects compared with 0.73–0.83 in the African and 0.52 in the Caucasian. The allele frequency of the mutant T allele ranged from 0.52 to 0.62 in the Asian subjects compared with 0.17–0.27 in the African [13] and 0.48 in the Caucasian [13]. Each Asian ethnic group was significantly different from the Africans (P < 0.001 in each case, Table 2, Fig. 2). Compared with the Caucasian population, only the Indians were found to be statistically different with respect to the distribution of genotype (P = 0.004, Table 2) and allele frequency (P = 0.002, Table 2).

Table 2.

Differences in C and T genotypes and allele frequencies between Asian, African and Caucasian populations measured by chi-squared test (P values)

		Chinese	Malay P value	Indian	Pooled Asian
Genotype	African	< 0.001	< 0.001	< 0.001	< 0.001
	Caucasian	0.371	0.699	0.004	0.053
Allele	African	< 0.001	< 0.001	< 0.001	< 0.001
	Caucasian	0.228	0.458	0.002	0.022

Figure 2.

Genotype frequencies for the C3435T MDR1 polymorphism in Asian, African and Causasian populations.

The correlation of MDR1 genotype in exon 26 and the dose-normalized systemic exposure of CsA AUC(0,4 h) in 10 Chinese heart transplant patients is depicted in Figure 3. Although the data show a trend towards higher systemic exposure in patients homozygous for TT compared with patients homozygous for CC genotype, the differences were not statistically significant because the interpatient variability in dose-normalized CsA AUC(0,4 h) was relatively large in patients with the homozygous CC genotype compared with those with CT or TT genotypes (CV 51% vs 16% vs 30%).

Figure 3.

Correlation of exon 26 SNP with Neoral AUC(0,4 h) normalized by dose in stable Chinese heart transplant patients (n = 10).

Discussion

The characterization of MDR1 gene and the utilization of pharmacogenetic testing for the identification of different MDR1 alleles in patients may provide a useful tool for optimizing therapy with drugs that are substrates of P-gp. The identification of the different allelic variants and their relationship to P-gp expression levels may also be useful in determining therapy outcome. So far the SNP C3435T at a wobble position in exon 26 has been correlated with intestinal P-gp expression levels and shown to influence the absorption of orally taken drugs that are P-gp substrates. Information gathered on the distribution of this C3435T polymorphism in populations of different ethnic origin may be essential in explaining the interindividual and interethnic differences in drug response and/or side-effects.

This study showed that the frequency distributions of the allelic variants were similar between the Chinese and Indian population and different between the Malays and Indians. The variation in genotype frequency and distribution of C and T alleles between the Indians and Malays was greater compared with Indians and Chinese (Table 1). Almost 30% of the Chinese and Malay individuals were homozygous carriers of the variant T allele compared with 40% of the Indians. The results obtained for the Chinese in this study are consistent with those published previously [13]. The frequency distribution of the homozygous C allele in Asians is similar to the Japanese [14].

Comparision of the pooled Asian data with Africans revealed the distribution of the SNP C3435T in the Asian population to be significantly different from the African population [13]. The frequency distribution of the genotypes and the C and T alleles were statistically different when the three ethnic groups in the Asian population (Chinese, Malays and Indians) were compared with the combination of three ethnic groups in the African population (Ghanaian, Kenyan and Sudanese) [13] (P < 0.001 in each case, Table 2). The Chinese and Malays were found to be similar to the Caucasian (German) population [12, 15] compared with the Indian subjects who were different from both the African and Caucasian population.

To evaluate the clinical consequences of the exon 26 SNP, we correlated csa AUC(0,4 h) with MDR1 genotypes in exon 26 in 10 stable Chinese heart transplant patients. CsA is a well-known P-gp substrate and its pharmacokinetic profile exhibits great inter- and intrapatient variability. Although the Neoral formulation of CsA exhibits improved pharmacokinetic characteristics over the corn-oil formulation of CsA (Sandimmune), its bioavailability is still highly variable especially during the first 4 h. In their study on renal transplant patients, Johnston et al. [16] showed that the greatest degree of inter- and intrapatient variability in CsA blood concentrations occurred during the first 4 h postdose, and this variabilty was minimal from 4 to 12 h postdose. Similar findings were also reported by Cooney et al.[17] in a separate study on cardiac transplant patients.

Comparision between AUC(0,4 h) and MDR1 genotypes showed a trend with Chinese patients homozygous for the exon 26 CC genotype having lower plasma AUC(0,4 h) values compared with subjects with TT genotypes. The low AUC(0,4 h) and high degree of interpatient variability in homozygous CC subjects is probably due to higher intestinal P-gp expression that affects Neoral absorption. Hoffmeyer et al.[12] showed that subjects carrying the CC genotype had lower steady-state maximum plasma concentrations of digoxin compared with those in TT subjects and this was due to higher duodenal expression of P-gp in CC subjects. An association was observed between the genotype status of the heart transplant patients and their CsA AUC(0,4 h), although we were limited by the number of patients with CT (n = 2) or TT (n = 2) genotypes. To date we only have a total of 10 Chinese heart transplant patients in Singapore.

Although we did not determine genetic polymorphisms associated with the CYP3A4 gene in this study, genetic variability in CYP3A4-mediated drug disposition may also account for the high degree of interpatient variability in the AUC(0,4 h) of Neoral. The expression of both CYP3A4 and MDR1 appear to be regulated independently [18] and depends on both the dose and duration of exposure of their substrates. Thus, in patients with CC genotype, P-gp and CYP3A4 may both cooperate to minimize exposure to Neoral. Recently, two allelic variants of CYP3A have been identified, one in the 5′-flanking region of CYP3A (CYP3A4*1b) [19] and the other that results in a Ser222Pro change (CYP3A4*2) [20]. Ethnic differences in the frequency of CYP3A4*1b are more common than CYP3A4*2; the latter was found in 2.7% of white subjects but was absent in black and Chinese subjects [20]. However, both allelic forms have been suggested not to contribute significantly to interindividual differences in the disposition of CsA in White subjects and African Americans [21].

The SNP C3435T is located at a noncoding nonpromoter position in the MDR1 gene and hence is unlikely to regulate the expression of MDR1. It is possible that this SNP may function in concert with as yet unidentified regions of the MDR1 gene that regulate expression. Synold et al.[22] recently showed that the orphan nuclear receptor, SXR (steroid and xenobiotic receptor) plays a central role in regulating CYP3A4 and MDR1 expression. They showed that paclitaxel but not docetaxel was able to activate SXR and caused increased expression of CYP3A4, CYP2C8 and MDR1 in intestine and hepatocytes, which in turn resulted in increased paclitaxel efflux from LS180 colon cancer cells. Thus, it would be interesting to investigate SXR activity and MDR1 genotypic status in subjects from different ethnic groups. Nevertheless, it seems that the C3435T SNP is important in defining an allele with respect to MDR1 expression and activity.

In conclusion, this study showed that the allelic frequency of C3435T SNP is different between the Malays and Indians. The Asians also differed significantly from the African population in the distribution of the C3435T SNP. Compared with Caucasians, only the Indians were found to be different. Knowledge of genetic variability and functional polymorphisms in ABC transporters between different ethnic groups are relevant pharmacological factors that can be used to understand variability in drug response.