Contribution of Nucleoside-Analogue Reverse Transcriptase Inhibitor Therapy to Lipoatrophy from the Population to the Cellular Level

Abstract

It has been proposed that the contribution of nucleoside-analogue reverse transcriptase inhibitor (NRTI) therapy to subcutaneous fat wasting involves adipose tissue-specific mitochondrial DNA toxicity. We have investigated the relationships between NRTI therapy, adipocyte mitochondrial DNA content, evidence of toxicity in adipose tissue and fat wasting in Caucasian male Western Australian HIV Cohort study participants. Longitudinal mixed effects analysis of fat wasting was undertaken in individuals receiving initial stavudine- or zidovudine-containing highly active antiretroviral measurements). Adipocyte mitochondrial DNA (mtDNA) depletion was also assessed according to current NRTI therapy in 92 subcutaneous fat biopsies from 69 HIV-positive individuals and seven healthy controls, and results were correlated with fat wasting among a subset of patients with biopsy data receiving initial stavudine- or zidovudine-containing HAART (n = 22, 103 DEXA measurements). Confocal microscopy was performed in 22 obtained after initiating/switching NRTI therapy. Stavudine therapy was associated with more severe adipocyte mitochondrial DNA depletion (P < 0.001) and fat wasting over time (P = 0.002) compared with zidovudine therapy in independent analyses. Among patients with concurrent biopsy and longitudinal DEXA data, fat wasting was associated with duration of NRTI therapy (P = 0.001) and adipocyte mtDNA copies/cell (P = 0.01). In this analysis, the significant association between choice of stavudine versus zidovudine and fat wasting (P = 0.03) was lost after adjustment for the effect of mtDNA depletion (P = 0.13). Confocal analysis provided direct evidence of a relationship between severity of adipose tissue toxicity and mitochondrial DNA depletion. No significant effects of HIV protease inhibitor therapy were detected in these analyses. Severity of subcutaneous fat wasting is primarily determined by choice of NRTI therapy (stavudine versus zidovudine) and by duration of exposure to the relevant NRTI. At the cellular level, evidence is provided that this effect manifests through NRTI-induced mitochondrial DNA depletion.