The efavirenz-induced increase in HDL-cholesterol is influenced by the multidrug resistance gene 1 C3435T polymorphism

Efavirenz treatment has been associated with increases in HDL-cholesterol concentrations, and the circulating levels of the drug have been related to the multidrug resistance gene 1 (MDR-1) C3435T polymorphism. The changes in the measured lipid parameters were evaluated in 59 HIV-infected patients initiating efavirenz-based treatment at baseline and at 12 months of follow-up. Efavirenz treatment increased HDL-cholesterol. The changes in concentrations appeared to be influenced by the MDR-1 gene polymorphism, in which CC > CT > TT.

The change in antiretroviral regimens, with non-analogue nucleotide reverse transcriptase inhibitors, such as efavirenz, replacing protease inhibitors (PI) also causes variation in lipid profiles and an increase in HDL ^[1].

The multidrug resistance gene 1 (MDR-1) codes for P-glycoprotein. P-glycoprotein is an energy-dependent transporter of multiple hydrophobic low molecular weight drugs that removes toxic substances from the cell and belongs to the ATP-binding cassette transporter family (ABCB1). All members of this family are closely related to lipoprotein metabolic pathways ^[2]. The MDR-1 C3435T gene polymorphism has been related to the efavirenz concentration in plasma and to the immune recovery of CD4 lymphocyte cell counts ^[3]; this issue, however, remains controversial ^[4]. Furthermore, this polymorphism has been correlated with P-glycoprotein protein levels and substrate uptake. Individuals homozygous for the T allele have lower P-glycoprotein expression compared with individuals homozygous for the C allele ^[5].

The aim of the present study was to evaluate the changes induced by efavirenz on the HDL-cholesterol concentration and the effect of the MDR-1 C3435T polymorphism.

We studied 59 HIV-infected patients who were treated with efavirenz over a 12-month observational period. All the patients showed an undetectable viral load and none presented with an opportunistic infection over the observational period. We collected clinical data and determined plasma total cholesterol, triglycerides, HDL-cholesterol, LDL-cholesterol, and the CD4 lymphocyte count at baseline and at 12 months of follow-up.

The serum viral load was measured using the Amplicor HIV-1 monitor assay (Roche Diagnostics, Branchburg, NJ, USA) and CD4 T-cell counts were measured by standard FACscan flow cytometry (Becton-Dickinson, Madrid, Spain). Total cholesterol, triglycerides and HDL-cholesterol were measured using standard methods and LDL-cholesterol was calculated by the Friedewald formula ^[6]. For genotyping the MDR-1 polymorphism, reactions were carried out using 5′-TGT TTT CAG CTG CTT GAT GG-3′ and 5′-GCA TGT ATG TTG GCC TCC TT-3′ as sense and antisense primers, respectively. The amplicon [193 base pairs (bp)] was digested with one unit of MboI, which results in fragments of 158 and 35 bp when a C is present at nucleotide 3435. The study was approved by the Ethics Committee of the Hospital Universitari de Sant Joan de Reus, and all patients provided full informed consent to participation. All statistical analyses were performed with the SPSS 11.0 program. Results are expressed as the mean and SD in parentheses.

The mean age of patients studied was 40.6 years (8.3) and 36 were men. Nine patients were naive for any antiretroviral treatment. The MDR-1 C3435T genotype was distributed as follows: 14 (23.7%) were CC; 31 (52.5%) were CT and 14 (23.7%) were TT. The CD4 lymphocyte count showed a significant increase from 374 (268) to 453 (282) cells/mm³ [95% confidence interval (CI) 33-124 cells/ml; P = 0.001] and this was not related to the MDR-1 C3435T genotypes. Total cholesterol was significantly increased by 20% (95% CI 0.48-1.29 mmol/l; P < 0.001), HDL-cholesterol presented a significant increase of 16.8% (95% CI 0.08-0.28 mmol/l; P < 0.001) and the LDL-cholesterol was also 22.6% higher (95% CI 0.26-0.91 mmol/l; P = 0.001). The increase in HDL-cholesterol in univariate analysis of variance showed a significant difference between genotypes (P = 0.024). The changes in HDL levels were related to the MDR-1 genotypes, in which CC > CT > TT. Analysis of variance for repeated measures of lipid profile constituents using age, sex, antiretroviral experience, triglycerides, total cholesterol and MDR-1 C3435T genotype as covariates revealed significant differences in HDL-cholesterol between the genotypes studied (P = 0.041). In a Wilcoxon test the patients with the TT genotype did not show a significant increase (0.07 mmol/l), whereas the CT genotype showed an increase of 11.8% [0.13 mmol/l (0.23); P = 0.002] and the CC genotype showed an increase of 36.5% [0.45 mmol/l (0.6); P = 0.015] (Fig. 1). Although serum triglyceride concentrations did not vary during the follow-up, we observed significant differences among genotype; the TT patients presenting with mild hypertriglyceridemia of 3.8 mmol/l (3.5) (P = 0.019), whereas the patients homozygous for the CC allele had lower serum triglyceride concentrations [1.3 mmol/l (0.29)].

Fig. 1

The treatment with efavirenz in HIV-infected patients resulted in a significant increase in HDL-cholesterol without an accompanying decrease in serum triglyceride concentrations and it was affected by the MDR-1 C3435T polymorphism. Interestingly, the increase in HDL-cholesterol was limited to patients with either the CC or the CT genotype, and those patients homozygous for the mutated allele (TT) did not show any increase during the one-year treatment with efavirenz.

This association could be related to drug circulating levels, but we hypothesize a direct relationship between MDR-1 and HDL. The efavirenz-based regimen itself induces the overexpression of MDR-1 ^[7], but the expression in vitro of its product, P-glycoprotein, is significantly lower for the homozygous mutated genotype (TT) than for the homozygous wild-type allele (CC). It is likely that our patients are reproducing this pattern of expression in vivo.

P-glycoprotein mediates the ATP-dependent relocation of cholesterol from the cytosolic leaflet to the exoplasmic leaflet of the plasma membrane ^[8], and the ATP-binding cassette family of proteins facilitates the cholesterol efflux between cell membranes and HDL. The pharmacological overexpression of P-glycoprotein induced by efavirenz in our patients may increase the reverse transport of cholesterol by inducing a significant increase in the HDL-cholesterol concentration. However, in our patients, the presence of the T allele acted as a barrier to this effect and have may completely counterbalanced the increase in HDL-cholesterol in HIV-infected patients treated with efavirenz if they were TT.

In conclusion, knowledge of the MDR-1 C3435T polymorphism and the physiological role of the multiple drug transporter system will enable more rational changes to be made to the treatment of HIV infection so as to reduce some of the observed metabolic derangements associated with the disease.

Sponsorship: This study was supported by the Instituto de Salud Carlos III, RCMN (C03/08), Madrid, Spain.