Differential subcutaneous adipose tissue gene expression patterns in a randomized clinical trial of efavirenz or lopinavir-ritonavir in antiretroviral-naive patients

Abstract

Gene expression studies of subcutaneous adipose tissue may help to better understand the mechanisms behind body fat changes in HIV-infected patients who initiate antiretroviral therapy (ART). Here, we evaluated early changes in adipose tissue gene expression and their relationship to fat changes in ART-naive HIV-infected patients randomly assigned to initiate therapy with emtricitabine/tenofovir plus efavirenz (EFV) or ritonavir-boosted lopinavir (LPV/r). Patients had abdominal subcutaneous adipose tissue biopsies at baseline and week 16 and dual-energy-X-ray absorptiometry at baseline and weeks 16 and 48. mRNA changes of 11 genes involved in adipogenesis, lipid and glucose metabolism, mitochondrial energy, and inflammation were assessed through reverse transcription-quantitative PCR (RT-qPCR). Additionally, correlations between gene expression changes and fat changes were evaluated. Fat increased preferentially in the trunk with EFV and in the limbs with LPV/r (P < 0.05). After 16 weeks of exposure to the drug regimen, transcripts of CEBP/A, ADIPOQ, GLUT4, LPL, and COXIV were significantly down-regulated in the EFV arm compared to the LPV/r arm (P < 0.05). Significant correlations were observed between LPL expression change and trunk fat change at week 16 in both arms and between CEBP/A or COXIV change and trunk fat change at the same time point only in the EFV arm and not in the LPV/r arm. When combined with emtricitabine/tenofovir as standard backbone therapy, EFV and LPV/r induced differential early expression of genes involved in adipogenesis and energy metabolism. Moreover, these mRNA expression changes correlated with trunk fat change in the EFV arm. (This was a substudy of a randomized clinical trial [LIPOTAR study] registered at ClinicalTrials.gov under identifier NCT00759070.).

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

Figures

FIG 1

Total body fat and limb and trunk fat changes from baseline (week 0) to week 16 (Δ16) and week 48 (Δ48). The bars represent means and standard errors of the mean (SEM) of all the patients included in each group (EFV, n = 10; LPV/r, n = 9). Significant differences in fat changes relative to baseline are indicated (*, P < 0.05; **, P < 0.01). No significant differences were observed in any comparison between the two treatments.

FIG 2

Transcript expression levels of the assessed genes at week 16 compared to baseline (week 0) in the EFV arm (n = 10) (A), the LPV/r arm (n = 9) (B), and the EFV arm versus the LPV/r arm (C). The data are expressed relative to the housekeeping gene (18S), and baseline transcript levels were set to 1 (dashed lines). The bars represent means and SEM of all the patients included in each group. Statistically significant differences between baseline and week 16 (*, P < 0.05; **, P < 0.01) and between EFV and LPV/r (#, P < 0.05; ##, P < 0.01) are indicated.

FIG 3

Spearman′s correlations between gene expression changes at week 16 and total body fat and limb and trunk fat changes at week 16 and week 48 (EFV, n = 10; LPV/r, n = 9). (A) The numbers represent Spearman′s rho (r) values. Statistically significant correlations between the two variables are shown in boldface. *, P < 0.05; **, P < 0.01. (B to E) Scatter plots of the significant correlations between trunk fat change at week 16 and CEBP/A change in the EFV arm (B), LPL change in the EFV arm (C), LPL change in the LPV/r arm (D), and COXIV change in the EFV arm (E).