MDR1 Polymorphisms and Modulation of ARV Drug Induced Hepatotoxicity

Antiretroviral (ARV) regimen is used to manage the progression of HIV infection. MDR1 Polymorphism is associated with the treatment outcome. Ethnic disparities have been observed in the distribution of MDR1 genotypes. Hence we analyzed the association of MDR1 polymorphism with the modulation of ARV associated hepatotoxicity.MDR1 polymorphisms (1236 C/T, 3435 C/T) was genotyped in a total 165 HIV-infected individuals (34 ARV drug induced hepatotoxicity were labeled as cases, 131 those without hepatotoxicity were controls) and 155 healthy individuals by utilization of PCR-RFLP. N= number of subjects, (%) = frequency of subjects, OR and 95% CIs were derived from logistic regression model comparing the homozygous wild-type genotype/allele (CC genotype and C allele for MDR1 1236 C/T and 3435 C/T) with other genotypes.


Conclusion
Haplotype TC was associated with hepatotoxicity severity. MDR1 1236TT and 3435CT genotypes in presence of nevirapine and who consume alcohol showed higher risk for acquisition and severity of ARV associated hepatotoxicity.
1.0 Background Antiretroviral (ARV) therapy is being extensively utilized for the management of HIV patients. The death rate among individuals with HIV is constantly increasing. Long-term e cacy and toxicity are the signi cant issues associated with ARV regimen, to worry about. Hepatotoxicity is an adverse outcome of ARV drugs in HIV-infected individuals [1,2]. A higher occurrence of hepatotoxicity was seen with the utilization of nevirapine-based regimen than the efavirenz-based regimen [3]. The occurrence of ARV drug induced hepatotoxicity because of nevirapine was shown to be 3.19% [4]. In another investigation, the occurrence of grade 3 or 4 hepatotoxicity was 10.8% in the patients treated with efavirenz and 8.9% in patients treated with nevirapine (5). ABCB1 is one of the universal adenosine triphosphate (ATP)-binding cassette (ABC) genes, responsible for cell homeostasis [6][7][8].
ABCB1 is situated on chromosome 7q21. It is a part of the MDR subfamily [8]. ABCB1 is expressed in a few tissues including epithelial cells of the blood-brain barrier [9,10] and transports numerous drugs [11]. P-glycoprotein (P-gp), a transmembrane transporter protein, is encoded by the ABCB1/MDR1 gene. P-gp is an ATP dependent e ux system that transports substances including drugs from intracellular to the extracellular matrix [12][13][14]. The variation in P-gp expression may vary its function, thus can affect the transport of drugs including nevirapine (NVP) [15]. The absorption and penetration of efavirenz (EFV) and NVP have evidently shown to be in uenced by the P-gp expression [16]. Chelule et al., (2003) have reported the prevalence of MDR1 3435CC genotype to be 85.9% in Africans, 41.70% in Indians, and 35.7% in the whites of Kwazulu-Natal and South Africa, respectively. In the African population, the presence of the MDR1 3435CC genotype was found to be associated with an overexpression of P-glycoprotein, whereas, the patients with TT genotype demonstrated a lower expression of P-gp [17,18]. Studies have reported signi cantly increased CD4 cell counts among HIV patients with MDR1 3435TT genotype [19,20]. Haas et al., (2006) proposed that MDR1 3435C/T polymorphism was not linked to efavirenz exposure [19]. Salem et al., (2014) suggested that MDR1 3435C/T polymorphism had no impact on efavirenz clearance [21]. Zhu et al., (2013) proposed that polymorphisms of MDR1 3435T/C and 2677T/G were linked to the response of nevirapine treatment (P=0.031, P=0.001) and could help to predict the drug response in HIV patients [22]. MDR1 3435TT genotype among individuals treated with EFV or nel navir correlated with a more elevated level of CD4+ count than the CT/CC genotypes [23]. Leschziner et al., (2007) showed that MDR1 3435 TT genotype were linked with higher adverse outcomes of 3TC (Lamivudine) and NVP treatment than EFV treatment [24]. Ritchie et al. (2006) indicated a signi cantly reduced risk of hepatotoxicity with NNRTI-containing regimens in the presence of MDR1 3435C/T polymorphism [25]. The link between polymorphism of MDR1 3435C/T and nevirapine induced hepatotoxicity has also been documented by studies [25,26].
Few studies have shown an association between MDR1 polymorphism and adverse outcomes of ARV drugs whereas, other studies suggest no correlation. The results were not conclusive. Moreover, the association between MDR1 polymorphism and ARV drug induced hepatotoxicity has not been reported from India. Thus, we have analyzed the association between MDR1 (1236C/T and 3435C/T) polymorphisms and modulation of ARV drug induced hepatotoxicity.

Study design and population
The study was carried out at the National AIDS Research Institute (NARI), Pune. The institute harbors an ARV therapy clinic, approved in 2005, by the National AIDS Control Organization (NACO) and provides free HIV testing and antiretroviral drugs under full NACO ARV therapy roll out programme. This works in collaboration with the Pune Municipal Corporation (PMC). Thus the patients are from the PMC area. A Page 4/29 case control study was conducted. The study population consisted of a cohort of patients who tested HIV positive and were on ARV therapy from August 2014 to September 2017, in the NARI clinic. Patients start ARV therapy based on the WHO criteria for the ARV therapy initiation in the adults and adolescents. All the individuals in this study were on rst line ARV regimen, NNRTIs (Non-nucleoside reverse transcriptase inhibitors) during the period of study. A subclassi cation of the study population was made on the basis of liver function test (LFT) and this subset of population was de ned as the HIV-infected individuals with ARV drug induced hepatotoxicity were labeled as cases. The male patients with SGOT>93.8 U/ml, Alkaline phosphatase >550.8 U/ml, total bilirubin >3. 22mg /ml and SGPT>229.5 U/ml and female patients with SGOT>163.2 U/ml, Alkaline phosphatase >550.8 U/ml, total bilirubin >3. 22mg /ml and SGPT>173.4 U/ml were classi ed as case group possessing hepatotoxicity. Exclusion criteria: (1) patients with Tuberculosis, Hepatitis B and C infections (2) patients with immune reconstitution syndrome and untreated opportunistic infections (3) patients on other known hepatotoxic medications for the case group showing hepatotoxicity.
At the same time HIV-infected individuals those without hepatotoxicity were labeled as a control group was age matched and recruited. Control population for the HIV positive group, tested negative for TB, Hepatitis B and C and also was not from a similar family as the study cases. The HIV infected males and females with, SGOT<32 U/ml, Alkaline phosphatase <108 U/ml, total bilirubin <1. 24mg /ml, and SGPT<34 U/ml. For the purpose of analysis the independent variables were categorized as: variables related to HIV (CD4+ count to de ne the stage of HIV infection and NNRTI regimen for drug induced hepatotoxic status), variables related to habits and lifestyle (drinking and tobacco usage) and physiological variables like the status of the liver enzymes.
Clinical information was noted through the reviews of case records, questionnaires, and personal interviews. The ethical endorsement was taken from the Ethics Committee, National AIDS Research Institute, Pune, India (Reference number: August 28, 2013, EC/NARI/Genetic Susceptibility/13-14/146) and written consents were taken from every single quali ed subject.
The stages of HIV infection were de ned on the basis of CD4+ cell counts of the patients at the time of recruitment. Fluorescence-activated cell sorting (FACS) analysis was utilized to estimate the CD4+ count. HBsAg and hepatitis C testing was completed by ELISA with the Ortho HCV ELISA test system.
With regard to the ARV therapy, Efavirenz and Nevirapine were the antiretrovirals administered. A questionnaire was utilized to record the usage of tobacco and alcohol.

Extraction of DNA
The collection of 2ml peripheral blood was done from all subjects and put at -70 0 C until DNA extraction.
The DNA extraction was done from blood leukocytes pellets utilizing the QIAamp DNA Mini Kit according to the kit manual.

Genotyping
Restriction fragment length polymorphism (RFLP) analysis was utilized to genotype the MDR1 (1236 C/T and 3435 C/T) polymorphism. Primers to amplify the MDR1 (C1236T and C3435T) polymorphism were utilized as designated by the previous report [27]. PCR was performed in a volume of 20µl. The PCR conditions for ampli cation of MDR1 C1236T and C3435T polymorphisms were used as described previously [28]. A thermocycler was utilized to complete all the reactions. PCR products were visualized by ethidium bromide staining. The PCR products of MDR1 C1236T and C3435T polymorphism was digested by utilization of restriction enzymes HaeIII and MboI at 37°C for 16 hours separately. 10% polyacrylamide gel along with molecular weight markers was utilized to the genotype of MDR1 C1236T and C3435T polymorphisms. The sequences and location of SNPs were employed for genotyping of MDR1 C1236T and C3535T polymorphisms. Genotypes for MDR1 C1236T were: 93bp and 87bp for CC, 87, 58, and 35bp for TT, and 93, 87, 58, and 35bp for CT; for MDR1 C3435T: 130 and 76bp for CC, 206bp for TT and 206, 130, and 76bp for CT. Re-genotyping of 20% of the samples were done to check the disparities in genotyping by addi1tional staff. The errors in genotyping were cross-veri ed by DNA sequencing of 10% of the samples.

Statistical examination
The variable, age was communicated as mean ± standard deviation (SD). Hardy-Weinberg equilibrium was examined by utilization of the Chi-Square goodness of t test in healthy individuals. Fisher's exact test was utilized to analyze the genotype frequencies between groups. Logistic regression was utilized to compute the odds ratios (ORs) and 95% con dence interval (CI). SPSS (SPSS Inc) programming form 17.0 was utilized for statistical examination and the two-sided value was taken as a test of statistical signi cance. A p-value under ≤0.05 was considered for signi cance. SNPStats online software was utilized to compare the frequency of haplotypes among groups (26). Linkage disequilibrium (LD) was analyzed between both the loci by computing the relative LD value (D') as D' = Dij/ Dmax (28). The Dij value was compared among various groups by the comparison of con dence intervals.

Results
Out of who tested HIV positive and were on ARV therapy from August 2014 to September 2017, in the NARI clinic. A total of 165 HIV positive individuals, on ARV therapy who visited NARI clinic from August 2014 to September 2017 were considered for the study. Out of these, 34 individuals showed ARV drug induced hepatotoxicity and constituted the rst case group of HIV positive individuals. The remaining 131 individuals didn't show hepatotoxicity (by LFT) and formed the second study group of HIV positive individuals on ARV therapy without ARV drug induced hepatotoxicity. Control population consisted of 155 healthy individuals. The demographic pro les of the participants are outlined in Table 1. N, total number of subject participants; 0, Alcohol and tobacco status was not reported The mean age ± SD of HIV infected individuals those without hepatotoxicity, with ARV drug induced hepatotoxicity and healthy individuals were 40.27 ± 2.45, 37.24 ± 3.29, and 37.25 ± 6.30 years. Each of the study population and control group has been further characterized on the basis of the NNRTI regimen (Efavirenz or Navirapine), alcohol and tobacco usage (users or non-users) and the CD4+ status (to de ne the stage of HIV infection). These parameters were used to categorize the study populations and the control group and the incidence of MDR1 polymorphisms were analyzed in all the categories as shown in the Tables 6-9. N= number of subjects, (%) = frequency of subjects, OR and 95% CIs were derived from logistic regression model comparing the homozygous wild-type genotype/allele (CC genotype and C allele for MDR1 1236 C/T and 3435 C/T) with other genotypes.

MDR1 polymorphisms and stages of HIV
The incidence of MDR1 polymorphism among individuals in different stages of HIV infection and the healthy controls was also studied as outlined in

Gene-environment interaction
The distribution of MDR1 polymorphisms among ARV drug induced hepatotoxicity cases and those without hepatotoxicity and the healthy control group was analyzed by categorizing them on the basis of tobacco and alcohol consumption and NNRTI regimen as shown in Tables 6-9. The occurrence of polymorphisms of MDR1 (1236C/T and 3435C/T) was not different among the individual consuming tobacco in both the study populations and the healthy control group. MDR1 1236TT genotype was overrepresented among the tobacco consumers than the non-consumers in the ARV drug induced hepatotoxicity group (28.6% versus 14.8%) ( Table-6). Also its occurrence was higher among the alcohol consumers than the non-consumers in the ARV drug induced hepatotoxicity group (28.6% versus 14.8%, OR=1.50, 95% CI: 0. 13 -17.35, P=0.88

Discussion
We analyzed the association between MDR1 polymorphism and ARV drug induced hepatotoxicity. MDR1 encodes for the ATP-dependent membrane e ux transporter (14). P-glycoproteins are substrate for the genetic variant that impact patient drugs. The occurrence of MDR1 polymorphism changes from population to population [17]. We analyzed the MDR1 genotypes and haplotypes among the individuals with ARV drug induced hepatotoxicity and those without hepatotoxicity.
We found that the occurrence of MDR1 3435C/T polymorphism in our healthy control population is identical to the investigations done in European, North Indian, Turkish, and Asian populations [30][31][32][33][34][35] and contrasted with the similar studies done in the Chinese, Iranian and Thailand populations [27,33,36].
Also, the genotypic dispersal of MDR1 1236C/T polymorphism in our healthy individuals was almost alike to the similar investigations in the North Indian population [35]. However, it contrasted from the Mexican, Chinese and South African populations [18,31,38,39]. We have done a genotype-phenotype analysis and found that the individuals with MDR11236TT genotype were at increased risk for severity of hepatotoxicity (OR=1.37, P=0.57). However, due to the small number of ARV drug induced hepatotoxicity cases, the risk could not reach statistical signi cance. The low phenotypic expression was linked with polymorphisms in P-gp. People with 3435TT genotype were found to have lower levels of P-gp than CC and CT genotypes. Also, MDR1 3435C/T polymorphism was associated with the reduced risk of NNRTIinduced liver toxicity [25].
We also studied the gene-gene interactions to understand the synergistic impact of MDR1 polymorphism on ARV drug induced hepatotoxicity. The gene-gene interactions are known to have greater effects on gene expression than a single gene [40]. Likewise, we analyzed the association between the MDR1 genotypes and the stage of HIV infection. In our investigation, the incidence of MDR1 genotypes did not vary signi cantly among individuals of various stages of HIV infection as well as the healthy population. MDR1 1236CT, 1236TT, and 3435CT genotypes have been shown to be correlated with the HIV disease progression, haven't been found to regulate the susceptibility to HIV-1 infection [41]. Additionally, the patients with 3435 TT genotype had an increase in the CD4+ counts, following a treatment for half an year [23].
An analysis of the gene-environment interaction helps to know its impact on the disease etiology [42,43].
We did a case-only analysis to study the gene-environment interaction. We did not rather go for a casecontrol analysis in light of the fact that a case-control investigation, requires a coordination of cases with the control population [44]. HIV infected individuals who are naïve to ARV therapy and consume alcohol have been found to have a reduction in the CD4+ cell count [45]. Also, in the HIV infected women who consume tobacco, a diminished response to ARV therapy has been observed [46]. In our study, in ARV drug induced hepatotoxicity cases, who consume alcohol, MDR11236TT genotype exposed a risk for severe hepatotoxicity (OR=1.50, P=0.88). Among the individuals those without hepatotoxicity and used alcohol, the incidence of 3435CT genotype posed a higher risk of acquisition of ARV drug induced hepatotoxicity (OR=2.47, P=0.12) because of combined effect of gene and environment. Among the ARV drug induced hepatotoxicity cases on nevirapine, presence of MDR1 1236TT genotype was likely to be associated with the higher risk of severe hepatotoxicity (OR= 2.11, P=0.55). Whereas among individuals those without hepatotoxicity on nevirapine, MDR1 1236CT, 1236TT genotypes exposed a higher risk of acquisition of ARV drug induced hepatotoxicity (OR=1.66, P=0.45; OR=1.96, P=0.55). In individuals with ARV drug induced hepatotoxicity on nevirapine who consume, MDR1 1236TT genotype exposed a higher susceptibility to severe hepatotoxicity (OR=2.21, P=0.55) because of combined effect of gene and environment.
In the individuals those without hepatotoxicity, who consume alcohol, MDR1 3435CT genotype showed a susceptibility to acquisition of ARV drug induced hepatotoxicity (OR=2.04, P=0.23). This suggests that individuals with MDR1 1236TT and 3435CT genotypes with or without ARV drug induced hepatotoxicity may have a combined effect on acquisition and severity of hepatotoxicity. Also, individuals on nevirapine, possessing 3435T allele has a reduced risk of hepatotoxicity [26]. Individuals with MDR1 1236T and 1235T alleles have been associated with a diminished plasma NNRTI concentration, in uencing the virological response to HAART [21]. Haas et al., (2005) have not found any signi cant association between ABCB1 variations and plasma EFV concentrations [19].
The fact that this work can just assess the association and does not indicate causation, is one of the limiting points of the study. Also, the present investigation was planned to constitute a 1:4 proportion of cases versus controls but this couldn't be accomplished and we recruited them in 1:3 proportion which may be su cient.

Conclusions
MDR1 haplotypes have an impact on the severity of ARV drug induced hepatotoxicity. In the individuals on nevirapine who consume alcohol, MDR1 1236TT and 3435CT genotypes, had a combined effect on acquisition and severity of hepatotoxicity.
MDR1 is associated with drug clearance. As MDR1 expression in uences the response to NVP and EFV regimen. Therefore, further association studies between MDR1 polymorphism and plasma drug concentration would be done with a larger sample size. In addition, the association of polymorphisms of other drug transporter genes with plasma drug levels is required to comprehend the effect of genetic variants on treatment effect.

Declarations
Ethical Approval and Consent to participate: Study is approved from Institutional Ethics committee (IEC): NARI/EC/ICF version 1.0, dated 28 August 2013. Consent was taken from all the participants Consent for publication: Written informed consent was obtained from the patient for publication of this report. A copy of the written consent is available for review by the Editor-in-Chief of this journal." Availability of supporting data:The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests in this section.