Influence of CYP3A5 and MDR1 (ABCB1) Polymorphisms on the Pharmacokinetics of Tacrolimus in Renal Transplant Recipients

Abstract

Background. 

A body-weight-based dose of tacrolimus often results in marked individual diversity of blood drug concentration. Tacrolimus is a substrate for cytochrome P450 (CYP) 3A5 and p-glycoprotein encoded by CYP3A5 and MDR1 (ABCB1), respectively, having multiple single nucleotide polymorphisms. In this study, we genotyped CYP3A5 A6986G, MDR1 G2677(A/T), and C3435T polymorphisms and investigated the association between these polymorphisms and the pharmacokinetics of tacrolimus in renal transplant recipients.

Methods. 

Thirty consecutive recipients were enrolled in this study. The pharmacokinetics of tacrolimus was analyzed on day 28 after transplant, when the daily dose was adjusted to the target trough level of 10–15 ng/mL. The polymerase chain reaction-restriction fragment length polymorphism and direct sequence method were used for genotyping the CYP3A5 and MDR1 polymorphisms, respectively.

Results. 

The single tacrolimus dose per body weight was significantly higher in CYP3A5 *1 carriers than CYP3A5 *3/*3 carriers (0.143±0.050 vs. 0.078±0.031 mg/kg, P<0.001). The dose-adjusted trough level and the area under the concentration-time curve (AUC_0–12) were significantly lower in CYP3A5 *1 carriers than CYP3A5 *3/*3 carriers (0.040±0.014 vs. 0.057±0.024 ng/mL/mg/kg, P=0.015 and 0.583±0.162 vs. 0.899±0.319 ng·hr/mL/mg/kg, P=0.004), respectively. The MDR1 polymorphism was not associated with any pharmacokinetic parameters.

Conclusions. 

Kidney transplant recipients with the CYP3A5 *1 allele required a higher daily tacrolimus dose compared with those with the CYP3A5 *3/*3 genotype to maintain both the target trough level and AUC_0–12, suggesting that this polymorphism is useful for determining the appropriate dose of tacrolimus.

Tacrolimus, a widely used immunosuppressant, plays a key role in organ transplantation. However, marked individual diversity and a narrow therapeutic range of blood tacrolimus concentration are critical issues in the clinical setting. Since a low blood concentration of this drug is one of the factors responsible for acute rejection, while a high blood concentration induces adverse effects such as hypertension, hyperglycemia, and nephropathy, it is important to determine the appropriate tacrolimus dose, particularly in the early stages of transplantation. Tacrolimus is known to be a substrate of cytochrome P450 (CYP) 3A4, CYP3A5, and p-glycoprotein, which are encoded by CYP3A4, CYP3A5, and multidrug resistance 1 (MDR1), respectively (^1–3). The difference in expression level and the bioactivity of these proteins may explain the individual variations of the pharmacokinetics of tacrolimus.

The CYP3A5 gene located at 7p21 harbors an important single nucleotide polymorphism (A6986G) in intron 3, of which A and G alleles are designated as CYP3A5 *1 and CYP3A5 *3, respectively. The CYP3A5 *3 creates a cryptic acceptor splice site and transcribes the variant mRNA (SV1-CYP3A5) having an excess of 131bp between exon 3 and exon 4 (⁴). The protein translated from the SV1-CYP3A5 mRNA is truncated at amino acid 102 due to a premature stop codon and only a small amount of complete CYP3A5 protein is translated from the wild type CYP3A5 (wt-CYP3A5) mRNA (⁴). Thus, livers with CYP3A5 *1/*3 express 4 times higher and 10 times lower wt-CYP3A5 mRNA than livers with CYP3A5 *3/*3 and CYP3A5 *1/*1, respectively, (⁵). A higher CYP3A5 protein concentration in the liver was also reportedly associated with the CYP3A5 *1 allele (⁴). CYP3A5 is also the major enzyme for tacrolimus in the small intestine and its expression is believed to be responsible for the decreased tacrolimus bioavailability (⁶).

P-glycoprotein, an ATP-dependent membranous transporter, is co-expressed with CYP3A in the liver and intestines (⁷). P-glycoprotein plays an important role in drug disposition by limiting absorption from the intestine and promoting efflux into bile and urine (^7,8). MDR1, which is synonymous with ATP-binding cassette, sub-family B (MDR/TAP) member 1 (ABCB1), is located in 7q21 and more than 20 gene polymorphisms have so far been discovered (^9–11). The CC genotype of MDR1 C3435T polymorphism located in exon 26 was reportedly associated with a higher p-glycoprotein expression in the small intestine compared with the TT genotype, although the polymorphism causes no amino acid substitution (⁹). The MDR1 G2677(A/T) polymorphism located in exon 21 is found most frequently among MDR1 polymorphisms and induces amino acid substitution from Ala to Ser and Thr, respectively. It was reported that the G allele of G2677(A/T) polymorphism was associated with a higher expression level of p-glycoprotein in the placenta compared with other genotypes with a gene dosage effect (¹²).

Previous studies have mentioned the effect of gene polymorphisms on the blood concentration of tacrolimus, but no other pharmacokinetic parameters, except for the trough level, have been analyzed (^13–16). In this study, we investigated the association of CYP3A5 and MDR1 polymorphisms with the pharmacokinetics of tacrolimus in renal transplant recipients.

PATIENTS AND METHODS

Patients and Pharmacokinetic Analysis

Thirty consecutive recipients, including 14 males and 16 females, with a mean age of 41.1±11.5 (range: 20–66), who underwent renal transplantation in Akita University Hospital between January 2001 and March 2003, were enrolled in this study. All the recipients were Japanese and treated with a combination of immunosuppressants consisting of tacrolimus, mycophenorate mofetil (MMF), and steroids. An initial oral dose of tacrolimus (0.15 mg/kg) was administrated twice a day at 9 a.m. and 9 p.m. for 2 days before transplantation. A 24-hr continuous intravenous infusion of 0.05 mg/kg per day was given for 6 days and was then switched to oral administration to achieve a trough level of 15–20, 10–15, and 8–10 ng/mL for the first 2 weeks, for the next 6 weeks, and thereafter, respectively. An oral dose of 2,000 mg per day of MMF was also given twice a day starting 2 days before surgery and was decreased to 1,500 mg per day thereafter. An intravenous dose of 500 mg of methylprednisolone on the day of operation was gradually tapered to 40 mg by the 6th operative day and was then switched to an oral dose of 30 mg of prednisolone, which was tapered to 10 mg by the 28th operative day.

The pharmacokinetics of tacrolimus was assessed on day 28 after transplantation, when the target trough level was 10–15 ng/mL, by sampling the peripheral blood just before and 1, 2, 3, 6, 9, and 12 hr after morning oral administration. All patients received the same tacrolimus dosage for at least 1 week before this study. The concentration of tacrolimus in the blood samples was measured by a duplicated microparticle enzyme immunoassay (IMx Abbott Laboratories, Abbott Park, IL). The pharmacokinetic parameters for tacrolimus were calculated by a standard noncompartmental analytical method using WinNonlin version 3.1 (Pharsight Co., CA). The maximal tacrolimus concentration (C_max) and the time to C_max (t_max) for each subject were derived directly from the raw data. The area under the concentration-time curve (AUC_0–12) was calculated by using the linear trapezoidal rule from 0 to t_max and the logarithmic trapezoidal rule from t_max to 12 hr after the tacrolimus administration. The dose-adjusted trough level, C_max, and AUC_0–12 were calculated by dividing C0, C_max, and AUC_0–12 by both the single dose of tacrolimus and body weight, respectively.

Informed consent regarding the use of their DNA and clinical information was obtained from all of the subjects before participation in this study. The research protocol was approved by the Institutional Review Board at Akita University School of Medicine.

Genotyping of MDR1 and CYP3A5 Gene Polymorphisms

DNA was extracted from a peripheral blood sample using a QIAamp Blood kit (Qiagen, Hilden, Germany) and was stored at -4°C until analysis. To genotype the A6986G polymorphism in the CYP3A5 gene, the polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) method was used. Briefly, PCR was performed in a 20 μL aliquot containing 50 ng of genomic DNA, 50 pmol of each primer, 80 μM of each deoxynucleotide triphosphate, 0.6 units of Ampli-Taq Gold DNA polymerase, 1.2 mM MgCl₂, and 1 × reaction buffer (Applied Biosystems, Foster City, CA). The primers used were as follows: forward, 5′-ATGGAGAGTGGCATAGGAGATA-3′; reverse, 5′-TGTGGTCCAAACAGGGAAGAAATA-3′. PCR amplification conditions were 8 min of initial denaturation at 94°C, followed by 40 cycles of melting at 94°C for 30 sec, annealing at 55°C for 30 sec, and elongation at 72°C for 30 sec, followed by a final elongation for 10 min at 72°C. The PCR products were digested at 37°C overnight with 10 units of SspI (New England Biolabs, Inc., Beverly, MA). Digested products were separated on 2.5% agarose gel containing ethidium bromide. When the A allele (CYP3A5 *1 allele) was present, the 130 bp PCR fragment was divided into 107 bp and 23 bp fragments. The validity of the PCR- RFLP analysis was confirmed by the direct sequencing of several PCR products showing each genotype, using a Dye Terminator Sequencing Kit ver.1.0 (Applied Biosystems). Direct sequencing was carried out for genotyping MDR1 G2677(A/T) and C3435T polymorphisms. PCR was performed to amplify two polymorphic regions using the primer pairs and conditions described by Zheng (¹⁷). PCR products were incubated at 37°C overnight with exonuclease I (ExoI) and calf intestinal phosphatase and were sequenced using a Dye Terminator Sequencing Kit ver.1.0 (Applied Biosystems) on ABI prism310 (Applied Biosystems).

Statistical Analysis

The data were expressed as the mean ± SD and differences at P<0.05 were considered significant. The χ² test was used to examine the difference in categorical data and the Mann-Whitney U test was employed to determine the difference in continuous values between groups. The analysis was performed using SPSS 11.0 statistical software. To test the population homogeneity of the subjects, the genotype frequencies of the CYP3A5 and MDR1 polymorphisms were tested against Hardy-Weinberg equilibrium by the χ² test. The expected haplotype frequency was calculated using the EH program to examine the linkage disequilibrium between two polymorphisms (¹⁸).

RESULTS

Association of CYP3A5 Gene Polymorphism with Pharmacokinetics of Tacrolimus

The CYP3A5 *1/*1, *1/*3, and *3/*3 genotype was detected in 1 (3.3%), 12 (40.0%), and 17 (56.7%) of the 30 recipients, respectively, and the genotype distribution was in Hardy-Weinberg equilibrium (P=0.689). Since the CYP3A5 *1/*1 genotype was observed in only one case, the subjects were divided into two genotype groups; that is, CYP3A5 *1/*1+*1/*3 and CYP3A5 *3/*3 genotypes. The distribution of sex, age, and body weight was not significantly different between the two groups. The pharmacokinetic parameters of tacrolimus in each polymorphism group are shown in Table 1. The single dose of tacrolimus per body weight ranged from 0.09 to 0.58 mg/kg and was significantly higher in the CYP3A5 *1 allele carriers than the CYP3A5 *3/*3 carriers (0.143±0.050 vs. 0.078±0.031 mg/kg, P<0.001) (Fig. 1). Although the trough level (C0) was higher in the CYP3A5 *1 carriers (15.0±2.8 vs. 12.5±3.0 ng/mL, P=0.025, respectively), the dose-adjusted trough level and AUC_0–12 were significantly lower in the CYP3A5 *1 carriers than the CYP3A5 *3/*3 carriers, respectively (0.040±0.014 vs. 0.057±0.024 ng/mL/mg/kg, P=0.015 and 0.583±0.162 vs. 0.899±0.319 ng·hr/mL/mg/kg, P=0.004) (Fig. 2). There was no significant difference in the C_max, t_max, and the mean residence time (MRT) between the two groups, but the dose-adjusted C_max in the CYP3A5 *1 carriers was significantly lower than the CYP3 A5 *3/*3 carriers (0.066±0.020 vs. 0.103±0.048 ng/mL/mg/kg, P=0.017). The apparent oral clearance (CL/F) and the apparent volume of distribution (V_dss/F) were significantly higher in the CYP3A5 *1 carriers than the CYP3A5 *3/*3 carriers (0.644±0.226 vs. 0.467±0.176 L/hr/kg, P=0.025 and 11.3±5.6 vs. 7.3±5.9 L/hr/kg, P=0.039), while there was no significant difference in the half-life (t_1/2) between the two groups. The blood concentration-time curves of tacrolimus are shown in Figure 3. The mean blood level of tacrolimus was not significantly different between the two groups at any point of measurement except for the trough level (C0).

TABLE 1: Comparison of pharmacokinetic parameters of tacrolimus between groups of CYP3A5 and MDR1 polymorphism

FIGURE 1.: The association of the dose requirement of tacrolimus with genotypes of CYP3A5 and MDR1 polymorphisms. The open circle shows a single dose of tacrolimus per body weight (mg/kg), and the horizontal line indicates the mean dose per body weight (mg/kg) of each genotype group.

FIGURE 2.: Difference of the dose-adjusted AUC_0–12 between two genotype groups of CYP3A5 and MDR1 polymorphisms. The open circle shows the dose-adjusted AUC_0–12 (ng·hr/mL/mg/kg), and the horizontal line indicates the mean dose-adjusted AUC_0–12 of each genotype group.

FIGURE 3.: The blood concentration-time curve of tacrolimus in renal transplant recipients with CYP3A5*1/*1 or *1/*3 and CYP3A5*3/*3 on day 28 after transplantation when the target trough level was 10–15 ng/mL. The data show the mean ± SD in each group with the CYP3A5 polymorphism. Only the point of C0 (trough level) is significantly different between the two groups (P<0.05).

Association of MDR1 Gene Polymorphisms with Pharmacokinetics of Tacrolimus

For the MDR1 C3435T polymorphism, CC, CT, and TT genotypes were detected in 15 (50.0%), 11 (36.7%), and 4 (13.3%) of the 30 recipients, respectively, and the genotype distribution did not deviate from Hardy-Weinberg equilibrium (P=0.627). For the G2677(A/T) polymorphism, GG, GA, GT, AT, TT, and AA genotypes were detected in 8 (26.7%), 6 (20.0%), 8 (26.7%), 4 (13.3%), 4 (13.3%), and 0 (0.0%) of 30 recipients, respectively. The C allele and T allele of C3435T polymorphism were strongly linked to the G allele and T or A allele of G2677(A/T) polymorphism, respectively, (data not shown). Since each polymorphism was in linkage disequilibrium (P<0.001), the recipients were divided into two genotype groups, the CC and the CT+TT, according to the C3435T polymorphism. The distribution of sex, age, or body weight was not significantly different between the two groups. There was no significant difference between the CC and CT+TT genotypes in the single dose per body weight, the dose-adjusted trough level, C_max, and AUC_0–12 (Table 1, Figs. 1 and 2). No significant difference was observed in t_max, t_1/2, MRT, CL/F, V_dss/F, and blood concentrations at any point assessed between the two groups (data not shown).

DISCUSSION

Recent studies have revealed significant individual variations in CYP3A5 expression. The expression frequency of CYP3A5 detected by Western blot analyses in the adult liver ranged from 20% to 55% among different ethnic groups and the expression ratio of CYP3A5 to the total CYP3A also ranged from 6% to 99% in the liver and intestines (^4,19). Several studies (^{4,13,14,20,21}) have suggested that the polymorphic expression of CYP3A5, which may be attributed to the single nucleotide polymorphism, can explain the individual difference of the drug disposition. In this study, we observed more than a 1.5-fold difference of the body weight-adjusted daily tacrolimus dose in Japanese renal recipients. Pharmacokinetic analysis demonstrated that CYP3A5 *1 carriers required an increased dose of tacrolimus to achieve the optimal trough level and AUC_0–12 compared with CYP3A5 *3/*3 carriers. Thus, the CYP3A5 polymorphism appeared to have an impact on the blood concentration of tacrolimus in renal transplant recipients.

Previous reports demonstrated that renal transplant recipients with the CYP3A5 *1/*1 or CYP3A5 *1/*3 genotype had approximately two thirds of the dose-adjusted trough level compared with those with the CYP3A5 *3/*3 genotype (^13,14). However, there was no significant difference between the recipients with CYP3A5 *1/*1 and CYP3A5 *1/*3 genotype, indicating that only the recipients with the CYP3A5 *3/*3 genotype could be distinguished from the other genotypes. In this study, CYP3A5 *1 carriers also showed two thirds of the dose-adjusted trough level, C_max, and AUC_0–12 compared with CYP3A5 *3/*3 carriers and this result agrees with other reports. Our data also demonstrated that there was no significant difference in AUC_0–12 when targeting the same trough level in the two groups. These findings suggested that the tacrolimus dose should be optimized according to the CYP3A5 genotype of each recipient, but the same target trough level can be used as an index of drug exposure despite the different genotype. Interestingly, CYP3A5 *3/*3 carriers showed even a lower trough level although they tended to have a higher blood concentration of tacrolimus compared with CYP3A5 *1 carrier. This phenomenon may be attributed to a rapidly decreased dose of tacrolimus in CYP3A5 *3/*3 carriers, particularly in the early period after switching to oral administration.

Although the bioavailability of tacrolimus by intestinal absorption is known to vary, it has not been clarified how the CYP3A5 polymorphism causes a difference in tacrolimus concentration among individuals (²²). Our study showed that the dose-adjusted AUC_0–12 in CYP3A5 *1 carriers was significantly lower than in CYP3A5 *3/*3 carriers, whereas there was no significant difference in the t_1/2 and MRT between the two groups. This suggested that CYP3A5 affect the absorption, but not systemic elimination of tacrolimus. The metabolism of tacrolimus in the intestine contributes to extensive and variable first pass metabolism following the oral administration (²³), and tacrolimus absorption is reportedly regulated mainly by CYP3A-mediated metabolism in the jejunum but by p-glycoprotein efflux in the ileum (²⁴). Thus, CYP3A5 in the small intestine may play a key role in the mechanisms responsible for interindividual variability of blood tacrolimus concentration. However, microparticle enzyme immunoassay, which was employed to analyze the pharmacokinetics of tacrolimus in the present study, has cross-reactivity and measures substantial tacrolimus metabolites (²⁵). Actual parent drug and metabolites concentrations can be lower and higher, respectively, in CYP3A5*1 carriers compared with CYP3A5*3/*3 carriers. For a definitive evaluation of the pharmacokinetics, suitable measurement methods such as liquid chromatographic/mass spectrometry (LC/MS) and further analyses comparing the intravenous and oral administration for the respective genotypes are necessary.

There is a racial difference in the frequency of CYP3A5 polymorphism. Consistent with our result, the CYP3A5 *3/*3 genotype frequencies were reported to be 58.3%–60.5% in a Japanese population (^26,27). Macphee et al. (¹⁵) demonstrated that the frequency of the CYP3AP1 AA_-44 genotype, which is a pseudogene of CYP3A5 and is strongly linked to the CYP3A5 *3/*3 genotype, was 86% in Caucasian and 23% in African-American recipients, and that the median tacrolimus concentration adjusted by body weight and dose at 3 months after transplantation was 0.087 and 0.052 ng/ml/mg/kg in white and African American recipients, respectively, showing a significant difference. The distribution of tacrolimus concentration in Japanese recipients in this study was similar to that of the African American recipients reported by Macphee et al. (¹⁵). However, it should be noted that the CYP3A5 *3/*3 genotype frequency was higher in the Japanese, and the concentration was measured 1 month after transplantation in our study. Dietary habits or other polymorphisms may also be responsible for the metabolism of tacrolimus.

One research group has claimed a significant association of the MDR1 polymorphism with the trough level of tacrolimus (²⁸), but subsequent studies have failed to prove such an association (^14,15,20). Similar disagreements concerning the MDR1 polymorphism and the pharmacokinetics have been documented in digoxin (^9,29) and fexofenadine (^30,31). Hoffmeyer et al. (⁹) first reported that the CC genotype of the MDR1 C3435T polymorphism was associated with a high expression level of p-glycoprotein in the small intestine. Although they reported that CC genotype carriers showed a lower C_max of digoxin compared with TT genotype carriers, Sakaeda et al. (²⁹) reported the opposite result. The discrepancies between these studies are unclear but may be attributed to the diet and/or genetic background of the individual populations. In addition, some drugs, which are not substrates for p-glycoprotein, have been reported to be associated with the MDR1 polymorphism (³²). In a report that Mdr1 knockout mouse accelerated CYP3A4-mediated drug metabolism, Schruetz et al. (³³) suggested that decreased p-glycoprotein expression activates the enzymes involved in drug absorption or disposition. It is speculated that the interaction of p-glycoprotein with other transporters or enzymes is involved in this phenomenon.

Recently, Goto et al. (²¹) demonstrated that the MDR1 polymorphism is associated with CYP3A4 mRNA expression in the small intestine but not with MDR1 expression. Since MDR1 is located upstream of CYP3A4 on the complementary strand, unknown gene polymorphisms of CYP3A4, which are in linkage disequilibrium, may affect the expression of CYP3A4. Meanwhile, it has been reported that the CYP3A4 polymorphism in the 5′-untranslated region (A-392G) is associated with the trough level of tacrolimus (¹⁴). Although a number of polymorphisms have been identified in this gene, the frequency of any polymorphisms in CYP3A4 is quite low in Asian populations (³⁴). In this study, we failed to find an association between MDR1 gene polymorphisms and the pharmacokinetic parameters of tacrolimus. Taken together, both the MDR1 and CYP3A4 polymorphisms are considered to have only a minor effect, if any, on the pharmacokinetics of tacrolimus in a Japanese population.

CONCLUSION

Our findings suggest that pretransplant assessment of the CYP3A5 polymorphism could be a useful measurement to determine the initial oral dose of tacrolimus to minimize the risk of under- or over-immunosuppression in individual recipients especially in the early stage of renal transplantation. Moreover, the individual difference of blood tacrolimus concentration in the CYP3A5 polymorphism is likely to be caused by the difference of absorption but the systemic elimination, although more precise pharmacokinetic analyses are necessary.