Contemporary Antiretroviral Therapy Dysregulates Iron Transport and Augments Mitochondrial Dysfunction in HIV-Infected Human Microglia and Neural-Lineage Cells

Abstract

HIV-associated cognitive dysfunction during combination antiretroviral therapy (cART) involves mitochondrial dysfunction, but the impact of contemporary cART on chronic metabolic changes in the brain and in latent HIV infection is unclear. We interrogated mitochondrial function in a human microglia (hμglia) cell line harboring inducible HIV provirus and in SH-SY5Y cells after exposure to individual antiretroviral drugs or cART, using the MitoStress assay. cART-induced changes in protein expression, reactive oxygen species (ROS) production, mitochondrial DNA copy number, and cellular iron were also explored. Finally, we evaluated the ability of ROS scavengers or plasmid-mediated overexpression of the antioxidant iron-binding protein, Fth1, to reverse mitochondrial defects. Contemporary antiretroviral drugs, particularly bictegravir, depressed multiple facets of mitochondrial function by 20-30%, with the most pronounced effects in latently infected HIV+ hμglia and SH-SY5Y cells. Latently HIV-infected hμglia exhibited upregulated glycolysis. Increases in total and/or mitochondrial ROS, mitochondrial DNA copy number, and cellular iron accompanied mitochondrial defects in hμglia and SH-SY5Y cells. In SH-SY5Y cells, cART reduced mitochondrial iron-sulfur-cluster-containing supercomplex and subunit expression and increased Nox2 expression. Fth1 overexpression or pre-treatment with N-acetylcysteine prevented cART-induced mitochondrial dysfunction. Contemporary cART impairs mitochondrial bioenergetics in hμglia and SH-SY5Y cells, partly through cellular iron accumulation; some effects differ by HIV latency.

Keywords: HIV; antiretroviral drug; combination antiretroviral therapy; human microglia; iron; metabolic reprogramming; mitochondrial dysfunction; neural cell.

Figure 1

Effects of a 24-h incubation with individual antiretroviral drugs on the basal oxygen consumption rate (OCR) in: (a,d) uninfected hμglia, (b,e) latently infected HIV+ hμglia, and (c,f) activated HIV+ hμglia. (g,h), Effects of a 24-h incubation with drugs individually (g) or in combination (h) in SH-SY5Y neural-lineage cells. Median and interquartile ranges of OCR values and results of at least 3 independent experiments (each with a minimum of 6 technical replicates) are shown. Abbreviations: Reg (Regimen)1, dolutegravir, abacavir, lamivudine; Reg2, bictegravir, emtricitabine, tenofovir; 3TC, lamivudine; DTG, dolutegravir; BIC, bictegravir; TFV, tenofovir; FTC, emtricitabine. * p < 0.05; ** p < 0.005; *** p < 0.0001 (Mann–Whitney U test).

Figure 2

Effects of a 24-h incubation of HIV-uninfected hμglia with Regimen2 (Reg2), showing (a) the MitoStress curve (blue = vehicle-treated controls; red = drug-treated cells), (b) non-mitochondrial oxygen consumption rate (OCR), (c) basal respiration, (d) maximal respiration, (e) spare respiratory capacity; (f) ATP production, and (g) proton leak. Median OCR values and interquartile ranges in treated vs. control (UT) cells from at least six technical replicates and three separate experiments are shown. (p-values obtained by Mann–Whitney U test).

Figure 3

A 24-h incubation of latently infected HIV+ hμglia with antiretroviral Regimen2 (Reg2) significantly reduces all mitochondrial respiratory parameters, including non-mitochondrial oxygen consumption rate (OCR). (a) MitoStress curve (blue = vehicle-treated controls; red = drug-treated cells); (b) non-mitochondrial OCR; (c) basal respiration; (d) maximal respiration; (e) spare respiratory capacity; (f) ATP production; (g) proton leak. Median OCR values and interquartile ranges in treated vs. control (UT) cells from at least six technical replicates and three independent experiments are shown. (p-values obtained by Mann–Whitney U test).

Figure 4

In activated HIV+ hμglia, the antiretroviral drug Regimen2 (Reg2) reduces multiple components of mitochondrial respiratory function at 24 h. (a) MitoStress curve (blue = vehicle-treated controls; red = drug-treated cells); (b) non-mitochondrial OCR; (c) basal respiration; (d) maximal respiration; (e) proton leak; (f) ATP production; (g) spare respiratory capacity. Median OCR values and interquartile ranges in treated vs. control (UT) cells from at least six technical replicates and three independent experiments are shown. (p-values obtained by Mann–Whitney U test).

Figure 5

Effects of a 24-incubation with BIC on (a) mitochondrial respiratory function in the MitoStress test (blue = vehicle-treated controls; red = drug-treated cells), and (b) basal and maximal OCR, and (c) ATP production, measured in the ATP rate assay, in SH-SY5Y cells. Abbreviations: OCR, oxygen consumption rate; Veh, vehicle-treated controls; BIC, bictegravir-treated cells; Mito, mitochondrial ATP production; Glyco, glycolytic ATP production; ns, not statistically significant. ** p < 0.005; *** p < 0.0001.

Figure 6

Effects of Regimen1 and Regimen2 ARVs after 24 h on (a–d) total cellular and (e–h) mitochondrial ROS in (a,e) uninfected, (b,f) latently infected HIV+, and (c,g) activated HIV+ hμglia and in (d,h) SH-SY5Y cells. Abbreviations: UT, untreated/vehicle controls; Reg1, Regimen1; Reg2, Regimen2; BIC, Bictegravir. * p < 0.05; ** p < 0.005 (Mann–Whitney U test).

Figure 7

(a–l) Expression of NDUFS1, SDHB, Nox2, and UQCRFS1 in uninfected, latently infected HIV+(GFP−), and activated HIV+(GFP+) hμglia after exposure to Regimens1 and 2 for 24 h. (m) Immunoblot images of protein expression in hμglia and (n–q) SH-SY5Y cells. Abbreviations: D−, no drug (vehicle control); R1, Regimen1; R2, Regimen2. NDUFS1, SDHB, and UQCRFS1 are mitochondrial respiratory subunit proteins; Sod2, a mitochondrial ROS-scavenging enzyme; Nox2, NADPH oxidase 2, a major cytoplasmic oxidase; Tom20, a mitochondrial protein translocase; Lrfn2, a marker of neuronal/synaptic integrity; I + III₂ and III₂ + IV, mitochondrial supercomplexes. * p < 0.05; ** p < 0.005 (Mann–Whitney U test).

Figure 8

(a) Mitotracker green image of mitochondrial fragmentation in SH-SY5Y cells after 24 h of bictegravir (BIC) exposure, compared with vehicle-treated controls (Veh). (b) Relative mitochondrial DNA copy number in Regimen2-treated vs. vehicle (no drug) in SH-SY5Y cells. (c–f) Relative mtDNA copy number in hμglia: (d) HIV-uninfected (HIV−), (e) latently infected HIV+(GFP−), and (f) activated HIV+(GFP+) hμglia. Abbreviations: UT, vehicle-treated controls; Reg1, Regimen1; Reg2, Regimen2; mtDNA, mitochondrial DNA. * p < 0.05; ** p < 0.005 (Mann–Whitney U test).

Figure 9

(a–c) Quantification of total ferritin, (d–f) ferritin heavy-chain (Fth1), (g–i) transferrin receptor (TfR), and (j–l) ferroportin (Fpn) in uninfected, latently infected HIV+(GFP−), and activated HIV+(GFP+) hμglia, after 24-h exposure to vehicle (UT or D−) or ARVs in Regimen1 (Reg1, R1) or Regimen2 (Reg2, R2). (m) Immunoblot gel images showing iron-related protein expression in hμglia. Results of a minimum of 3 separate experiments are shown. * p < 0.05 (Mann–Whitney U test).

Figure 10

Iron dysregulation in hμglia and SH-SY5Y cells after 24 h of exposure to Regimen1 (Reg1) or Regimen2 (Reg2) ARVs, or bictegravir (BIC). (a) Changes in TfR, (b) Mitoferrin, and (c) Fth1 expression in SH-SY5Y cells after BIC. Changes in cellular iron in (d) uninfected, (e) latently infected HIV+(GFP−), and (f) activated HIV+(GFP+) hμglia and (g,h) SH-SY5Y cells. Fth1, ferritin heavy-chain 1, Fth1-trans, cells transfected with Fth1-expressing plasmid; TfR, transferrin receptor; EV, empty plasmid vector; Veh, Vehicle. * p < 0.05; *** p < 0.005 (Mann–Whitney U test).