Biological interactions of CYP2C19 genotypes with CYP3A4*18, CYP3A5*3, and MDR1-3435 in living donor liver transplantation recipients

Background Polymorphisms in CYP2C19 are related to the metabolic oxidation of drugs to varying degrees. The CYP3A4*18, CYP3A5*3, and MDR1-3435 variant alleles are very important, particularly in tacrolimus metabolism in organ transplant rejection. Aim The aim of this study is o explore possible interactions among different CYP2C19 genotypes, namely, between homozygous extensive metabolizers (HomEM), heterozygous extensive metabolizers (HetEM), and poor metabolizers (PM), and the CYP3A4*18, CYP3A5*3, and MDR1-3435 variants in living donors and patients who received a living donor liver transplant (LDLT). Methods This prospective study enrolled 133 living donors and 133 corresponding recipients. On the basis of the HomEM, HetEM, and PM CYP2C19 genotypes, the distributions of CYP3A4*18 (exon 10; T878C), CYP3A5*3 (intron 3; A6986G), and MDR1-3435 (exon 26; C3435T) genotypes were analyzed for single nucleotide polymorphisms among donors and recipients. Results Among 102 HomEM genotypes, including 56 donors and 46 recipients, 91.2% of individuals harbored the T/T genotype of CYP3A4*18; 53.9% possessed G/G, and 34.3% had A/G genotypes of CYP3A5*3; and 38.2% had C/C and 50.0% had C/T genotypes at MDR1-3435. Among 130 HetEM genotypes, including 58 donors and 72 recipients, 97.7% of individuals possessed T/T genotype at CYP3A4*18; 50.0% harbored G/G and 41.5% had A/G genotypes at CYP3A5*3; and 40.0% had C/C and 49.2% had C/T genotypes at MDR1-3435. In 34 PMs, including 19 donors and 15 recipients, 88.2% had T/T genotypes at CYP3A4*18; 41.2% had G/G and 58.8% had A/G genotypes at CYP3A5*3; and 47.1% possessed C/C and 47.1% had C/T genotypes at MDR1-3435. On the basis of the CYP2C19 genotypes, no statistically significant distribution of genotypes were observed between donors and recipients for all genotypes of CYP3A4*18, CYP3A5*3, and MDR1-3435 (P >0.05). Conclusions In conclusion, the CYP2C19 genotypes do not affect the expression of CYP3A4*18, CYP3A5*3, or MDR1-3435 variants, which are independently distributed among donors and recipients during LDLT.


Introduction
Cytochrome P450 in the liver is one of the key enzyme complexes in the primary drug-metabolizing system in humans. Recipients of living-donor liver transplantation (LDLT) exhibit interesting biological distributions of CYP2C19 genotypes that differ between the recipient's tissue and the newly grafted tissue [1]. Acute rejection or abnormal postoperative liver function after LDLT could result from complications in cytochrome P450 function [2,3]. We have recently reported a homogenous phenomenon observed in CYP2C19 genotypes [4]. Antirejection agents such as tacrolimus usually target CYP3A4, CYP3A5, and MDR-1, which are the major metabolic isoenzymes of cytochrome P450. In the present study, we aimed to investigate the CYP2C19 genotypes, which have been classified as homozygous extensive metabolizers (HomEM), heterozygous extensive metabolizers (HetEM), and poor metabolizers (PM), and to identify any interactions between the CYP3A4*18, CYP3A5*3, and MDR1-3435 variants by characterizing differences between the genotype distribution of healthy living donors and patients with liver disease who received LDLT.
Genotyping of CYP3A4, CYP3A5, and MDR1-3435 Polymerase chain reaction/ligase detection reaction assay (PCR/LDR) was employed for genotyping the CYP3A4*18B and CYP3A5*3 SNPs. The PCR conditions consisted of a denaturation step at 95°C for 15 min, followed by 35 cycles of 94°C for 30 s, 65°C for 1 min, and 72°C for 1 min, followed by a final extension step at 72°C for 7 min. The specific amplified fragments were used in an LDR assay to identify the mutations associated with CYP3A4*18B and CYP3A5*3. The LDR assay was performed as follows: 10 μL of the reaction mix contained 1 μL of 1× ligase reaction buffer (New England Biolabs, USA), 1 μL of probes (12.5 pmol/μL each), 0.05 μL (2 U) of thermostable Taq DNA ligase (New England Biolabs), and 1 μL of PCR product. The ligation reaction was performed with a GeneAmp PCR System 9600 (Perkin Elmer, USA) as follows: 15 min at 95°C, followed by 35 cycles of 30 s at 94°C and 2 min at 60°C. The products were separated by agarose gel electrophoresis and analyzed with an ABI PRISM 377 DNA sequencer [5]. Genotyping was performed using an independent external contractor (Biowing Applied Biotechnology Co. Ltd., China). Genomic DNA was isolated from whole blood using the UltraPure™ Genomic DNA Isolation Kit (Shanghai SBS Genetech Technology Co., China). PCR-RFLP was performed to genotype exon 26 (C3435T) variant alleles in the MDR1 gene, with slight modifications. The PCR conditions consisted of a denaturation step at 95°C for 5 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 54°C to 59°C for 50 s, and elongation at 72°C for 1 min, followed by a final extension at 72°C for 10 min. PCR products were digested with DpnII (C3435T) and analyzed by electrophoretic separation on agarose gels, followed by direct visualization over an ultraviolet transilluminator after ethidium bromide staining [6]. This work was supported by a grant from Chang Gung Memorial Hospital (CMRPG8A0631 to K-WC) of Taiwan, which also granted ethical approval to our study. This study was approved by the Institutional Review Board (IRB), and informed consent was obtained from participants or from a parent or guardian in case of minor participants.

Statistical analyses
Statistical analyses were performed using SPSS software (version 12.0; SPSS, Chicago, IL, USA). Parameters between HomEM or HetEM and PMs in donors and recipients were compared using the X 2 test, Fisher's exact test, and Student's t-test. P values less than 0.05 were considered statistically significant. On the day of LDLT, 107 adult recipients received tacrolimus and 26 pediatric recipients received cyclosporine A (CsA). The serum concentrations of tacrolimus and CsA were 2.51 ± 2.73 ng/mL and 283.89 ± 308.93 ng/mL, respectively, on D1 and 6.17 ± 9.58 ng/mL and 1058.30 ± 582.37 ng/mL, respectively, on D30; these results show that the serum concentrations of both tacrolimus and cyA were significantly higher on D30 than on D1 after LDLT (P <0.001; Table 2).
There were 34 PMs in this cohort, including 19 donors and 15 recipients. The 2 wild-type CYP3A4*18 genotypes were distributed as follows: 3.3% (16/30) of individuals with the T/T genotype were donors and 46.7% (14/30) were recipients, 75.0% (3/4) of individuals with the T/C genotype were donors and 25.0% (1/4) were recipients. Only 2 of the possible CYP3A5*3 genotypes were There was no statistically significant difference between the haplotypes of CYP3A4*18 (T/T and T/C), CYP3A5*3 (G/G, A/G, and A/A), and MDR1-3435 (C/C, C/T, and T/T) and also between the different CYP2C19 genotypes (HomEM, HetEM, and PM) between healthy donors and recipients with end stage liver disease. There were independent isoenzymes and no correlation of genetic interaction between CYP2C19 and CYP3A4*18, CYP3A5*3, or MDR1-3435 not only, but also the variant haplotypes or genotypes.

Discussion
From our previous studies, we know that CYP2C19 expresses three genotypes with different drug metabolization capacities [1,3]. In this study, we attempted to investigate possible genetic interactions between CYP2C19 and CYP3A4*18, CYP3A5*3, or MDR1-3435 in detail. In the present study, we focused on the expression of genetic polymorphisms in the CYP3A4*18, CYP3A5*3, and MDR1-3435 genotypes. No significant differences were found in the distributions between CYP3A4*18 (exon 10; T878C), CYP3A5*3 (intron 3; A6986G), or MDR1-3435 (exon 26; C3435T) genotypes on the basis of different CYP2C19 genotypes between healthy liver donors and patients with poor liver function who received LDLT. Although all these proteins are important isoenzymes of cytochrome P450 in drug metabolism in the liver, only the differences in CYP2C19 genotypes have been documented [1]. CYP3A4*18, CYP3A5*3, and MDR1-3435 polymorphisms do not appear to have a functional effect following LDLT in either the donor with normal liver function or the recipient with ESLD. A recent study showed that the CYP2C19 genotype, unlike MDR1 and IL-1B genotypes, had an impact on the efficacy of Helicobacter pylori eradication in peptic ulcer patients treated with pantoprazole in triple therapy administrations [7]. In the present study, individuals receiving LDLT who had different CYP2C19 genotypes (HomEM [41 and 43.1%], HetEM [57.7 and 50%], and PM [31.2 and 62.5%]) did not show different distributions of CYP3A4*18, CYP3A5*3, and MDR1-3435 genotypes (C/C and C/T). This observation was consistent in both donors and recipients.
After LDLT, the serum levels of immunosuppressive agents were significantly lower on D1 than on D30, whereas the results of all liver functional tests (ALT, AST, T-Bil, PT(INR), and Alb) were significantly higher on D1 than on D30. These results led us to hypothesize that the immunosuppressive agents affected the stability of metabolic enzymes in the cytochrome P450 system. Previous published reports describe CYP3A as the most abundant enzyme of the P450 subfamily in the human liver and intestine, accounting for 30% of the total P450 in the human liver, and metabolizing approximately 50% of currently used clinical drugs [8][9][10]. The impacts of different CYP2C19, CYP3A4*18, CYP3A5*3, and MDR1-3435 genotypes on LDLT have been outlined in a flow chart presented in Figure 1. In our previous study, the homogenous phenomenon was attributed to the different CYP2C19 genotypes (HomEM, HetEM, and PM) between an LDLT donor and recipient owing to the uniqueness of the human liver [4]. The metabolism of the proton pump inhibitor (PPI) was dependent on the CYP2C19 genotypes in the cytochrome P450 system, primarily in the liver. HomEM genotypes were found to better metabolize some drugs than did the HetEM and PM genotypes during LDLT [1]. If the donor possessed a CYP2C19 PM genotype, the recipient assumed a PM genotype (rather than Figure 1 Flow chart of the possible relationship between CYP2C19 genotypes and CYP3A4*18, CYP3A5*3, and MDR1-3435 genotypes. [4]. the original CYP2C19 HomEM genotype) because of the homogenous phenomenon [4]. In contrast, the CYP3A4*18 and CYP3A5*3 genetic polymorphisms have two origins: the liver and intestine [8,9]. In the postnatal human liver, CYP3A4 and CYP3A5 are the two major CYP3A enzymes, which have overlapping substrate specificities [10]. When the liver function worsens, the drug metabolic function of the intestine may compensate, leading to unaltered drug metabolism, such as for tacrolimus (Figure 1). Depending on the liver function, the CYP2C19 genotypes HomEM, HetEM, and PM were more likely to present abnormal postoperative liver function and graft pathology [3]; however, this was not observed in the present study for the CYP3A4*18, CYP3A5*3, and MDR1-3435 polymorphisms in LDLT. As reported previously, CYP2C19 genotype expression is not demonstrated well by western blotting [11]. CYP3A5 protein expression is highly variable in the human liver, in particular, because of the high frequency of a SNP CYP3A5*3 A6986G in an intron [12]. The CYP3A4*18B SNP in intron 10 was first discovered by direct sequencing in a Japanese population. It was speculated that this variant was associated with increased CYP3A4 activity [5], and this speculation was extended by exploring cyclosporin A (CsA) metabolism in healthy Chinese subjects [6]. A number of SNPs have been identified in the MDR1 gene by large-scale sequencing. For example, our study probed the C3435T variant in exon 26 [13,14]. The PCR-RFLP method to detect the MDR1 3435C/T polymorphism has also been widely used, as was used recently in our study [7]. Although there is no evidence suggesting that intestinal expression of CYP3A4, CYP3A5, and MDR-1 play an important role, human cytochrome P450 enzymes have been expressed in Escherichia coli. Simplified bacterial systems can explain the possibility of intestinal activation of these enzymes [15]. A previous study has shown that the intestinal mucosa contains prominent forms of cytochrome P450, which are similar to liver cytochrome P450p in their structure, function, and some regulatory characteristics [16].
Therefore, the results of the present study do not impel us to change strategies and to combine certain donors with certain recipients. According to our previous study, evidence for graft rejection occurred within a month after LDLT [2]. In other words, the cytochrome P450 system stabilizes with time, up to 1 month after LDLT. The increase in serum level of immunosuppressive agents is followed by clinical liver functions leading to stability. Whether these studies can be performed in cadaveric liver transplantation cases is presently unclear.
In our cohort, 68.4% of individuals receiving LDLT had underlying chronic viral hepatitis-related ESLD, including 39.9% of individuals who had HCC. There was no difference in CYP3A4*18, CYP3A5*3, and MDR1-3435 genotypes between the donors and recipients. In addition to the etiology of the underlying disease, age also did not influence the distribution of CYP3A4*18, CYP3A5*3, and MDR1-3435 genotypes. No remarkable difference was observed between donors and recipients, as well as between pediatric and adult recipients.
From our data, the haplotypes of the CYP3A4*18, CYP3A5*3 or MDR1-3435 do not seem to correlate with tacrolimus metabolism in these recipients, but the variant stability of these enzymes are significantly different on D1 and D30 after LDLT. Drug levels were lower on D1 and higher on D30, but there were no correlations to the haplotypes of CYP3A4*18, CYP3A5*3, or MDR1-3435 and/or different genotypes of CYP2C19 HomEM, HetEM, and PM, except for their stability.
In conclusion, the CYP2C19 genotypes, HomEM, HetEM, and PM, do not affect the expression of CYP3A4*18, CYP3A5*3, and MDR1-3435 polymorphisms. These polymorphisms were independently distributed among donors and recipients, as well as healthy and diseased livers, because the source may be located outside the liver during LDLT.