Metabolic, Mitochondrial, and Inflammatory Effects of Efavirenz, Emtricitabine, and Tenofovir Disoproxil Fumarate in Asymptomatic Antiretroviral-Naïve People with HIV

Abstract

This study aimed to comprehensively assess the metabolic, mitochondrial, and inflammatory effects of first-line efavirenz, emtricitabine, and tenofovir disoproxil fumarate (EFV/FTC/TDF) single-tablet regimen (STR) relative to untreated asymptomatic HIV infection. To this end, we analyzed 29 people with HIV (PWH) treated for at least one year with this regimen vs. 33 antiretroviral-naïve PWH. Excellent therapeutic activity was accompanied by significant alterations in metabolic parameters. The treatment group showed increased plasmatic levels of glucose, total cholesterol and its fractions (LDL and HDL), triglycerides, and hepatic enzymes (GGT, ALP); conversely, bilirubin levels (total and indirect fraction) decreased in the treated cohort. Mitochondrial performance was preserved overall and treatment administration even promoted the recovery of mitochondrial DNA (mtDNA) content depleted by the virus, although this was not accompanied by the recovery in some of their encoded proteins (since cytochrome c oxidase II was significantly decreased). Inflammatory profile (TNFα, IL-6), ameliorated after treatment in accordance with viral reduction and the recovery of TNFα levels correlated to mtDNA cell restoration. Thus, although this regimen causes subclinical metabolic alterations, its antiviral and anti-inflammatory properties may be associated with partial improvement in mitochondrial function.

Keywords: HIV; antiretroviral treatment; efavirenz; emtricitabine; inflammatory effects; metabolic profile; mitochondrial DNA; mitochondrial toxicity; tenofovir.

Figure 1

Metabolic parameters in Naïve HIV patients (n = 33) compared to patients treated with cART (combined antiretroviral therapy) based on TDF/FTC/EFV for more than one year (n = 29). (A) Plasma glucose values; (B) Total Cholesterol values; (C) LDL: LDL-Cholesterol values; (D) HDL: HDL-Cholesterol values; (E) TG: Triglycerides levels; (F) Total bilirubin levels; (G) Indirect bilirubin levels; (H) GGT: Gamma-glutamyl transferase values; (I) ALP: Alkaline phosphatase values; (J) LDH: Lactate Dehydrogenase values. Box and whiskers plots showing median, minimum, and maximum values. # p-value = (0.05–0.1), * p-value < 0.05, ** p-value < 0.01, *** p-value < 0.001.

Figure 2

Mitochondrial parameters in Naïve HIV patients (n = 33) compared to patients treated with cART (combined antiretroviral therapy) based on TDF/FTC/EFV for more than one year (n = 29). (A) PBMC-mtDNA: Peripheral blood mononuclear cells mitochondrial DNA; (B) Plasma-mtDNA: Plasma mitochondrial DNA levels; (C) CoQ: Coenzyme Q values; (D); VDAC: Voltage-Dependent Anion-selective Channel vs. β-actin ratio; (E) COX-II: Cytochrome c oxidase subunit II vs. β-actin ratio; (F) COX IV: Cytochrome c oxidase subunit IV vs. β-actin ratio; (G) COX-II/COX-IV ratio; (H) COX-II/VDAC ratio; (I) COX-IV/VDAC ratio. Box and whiskers plots showing median, minimum, and maximum values. # p-value = (0.05–0.1), * p-value < 0.05, ** p-value < 0.01. (J) Representative Western Blot bands for proteins quantification in PBMC from two Naïve HIV patients compared to two patients treated with TDF/EFV/EFV STR for more than one year: β-actin (47 kDa), VDAC (31 kDa), COX-II (25.6 kDa) and COX-VI (15 kDa).

Figure 3

Inflammatory and soluble mediators in Naïve HIV patients (n = 15) compared to patients treated with cART (combined antiretroviral therapy) based on TDF/FTC/EFV for more than one year (n = 10). (A) TNFα: Tumor Necrosis Factor α levels. (B) IL6: Interleukin 6 values; (C) IL8: Interleukin 8 values; (D) MCP-1: Monocyte Chemoattractant protein 1 levels; (E) NGF: Nerve Growth Factor values; (F) Leptin: Serum Leptin levels; (G) HGF: Hepatocyte Growth Factor levels; (H) FGF21: Fibroblast Growth Factor 21 levels. Box and whiskers plots showing median, minimum, and maximum values. # p-value = (0.05–0.1) * p-value < 0.05.

Figure 4

Significant correlations between virologic and metabolic parameters with mitochondrial biomarkers in both cohorts. Linear regression line (solid line) and 95% confidence band of the best-fit line (dotted line) are shown. (A) Correlation between mitochondrial DNA content in PBMCs (PBMC-mtDNA) vs. patient viral load; (B) Correlation between PBMC-mtDNA vs. total cholesterol values; (C) Correlation between PBMC-mtDNA vs. LDL values; (D) Correlation between PBMC-mtDNA vs. alkaline phosphatase (ALP) values; (E) Correlation between PBMC-mtDNA vs. Creatine Kinase values. * p-value < 0.05, ** p-value < 0.01.

Figure 5

Significant correlations between virologic and mitochondrial parameters with inflammatory biomarkers in both cohorts. Linear regression line (solid line) and 95% confidence band of the best-fit line (dotted line) are shown. (A) Correlation between TNFα vs. viral load; (B) Correlation between Tumor Necrosis Factor α (TNFα) vs. IL-8 values; (C) Correlation between TNFα vs. Monocyte Chemoattractant protein 1 (MCP-1) values; (D) Correlation between TNFα vs. mitochondrial DNA in PBMCs; (E) Correlation between IL-6 vs. mitochondrial DNA in plasma. * p-value < 0.05, ** p-value < 0.01, *** p-value < 0.0001.