Human immunodeficiency virus Tat impairs mitochondrial fission in neurons

Abstract

Human immunodeficiency virus-1 (HIV) infection of the central nervous system promotes neuronal injury that culminates in HIV-associated neurocognitive disorders. Viral proteins, including transactivator of transcription (Tat), have emerged as leading candidates to explain HIV-mediated neurotoxicity, though the mechanisms remain unclear. Tat transgenic mice or neurons exposed to Tat, which show neuronal loss, exhibit smaller mitochondria as compared to controls. To provide an experimental clue as to which mechanisms are used by Tat to promote changes in mitochondrial morphology, rat cortical neurons were exposed to Tat (100 nM) for various time points. Within 30 min, Tat caused a significant reduction in mitochondrial membrane potential, a process that is regulated by fusion and fission. To further assess whether Tat changes these processes, fission and fusion proteins dynamin-related protein 1 (Drp1) and mitofusin-2 (Mfn2), respectively, were measured. We found that Drp1 levels increased beginning at 2 h after Tat exposure while Mfn2 remained unchanged. Moreover, increased levels of an active form of Drp1 were found to be present following Tat exposure. Furthermore, Drp1 and calcineurin inhibitors prevented Tat-mediated effects on mitochondria size. These findings indicate that mitochondrial fission is likely the leading factor in Tat-mediated alterations to mitochondrial morphology. This disruption in mitochondria homeostasis may contribute to the instability of the organelle and ultimately neuronal cell death following Tat exposure.

Fig. 1. Mitochondrial diameter is decreased in neurons by Tat.

a WT and Tat-tg mice were treated for 2 weeks with doxycycline. Vibratome sections of their brains were analyzed for mitochondrial morphology by transmission electron microscopy. Arrowheads point at mitochondria of the brains of two separate mice, each group. Scale bar = 500 nm. b Average quantification of diameter of neuronal mitochondria in WT and Tat-tg mouse brains. *p < 0.05 by Student’s t-test; n = 16 (8 WT and 8 Tat-tg mice). c–f Cortical neurons were fixed with 4% paraformaldehyde + 4% sucrose (pH 7.4) following exposure to control media (c) or media containing Tat (100 nM) (e) for 4 h. Cells were then stained for MAP2 (red) and TOM20 (green) to label neuronal microtubules and mitochondria, respectively. Images were acquired by STORM as previously described. d, f Enlargements (×5) of squares in c and e, respectively

Fig. 2. Tat changes the subcellular distribution of mitochondria.

Cortical neurons were exposed to boiled Tat (control) or Tat (100 nM) for up to 4 h. Mitochondria were visualized by TOM20 (red) in neurons identified by MAP2 (green) and counterstained with DAPI (blue). All images were acquired at ×40 magnification, Zoom 2. Scale bar = 8 µm. Note that most of mitochondria in control neurons are localized in processes (white arrowheads), those in Tat-treated neurons are localized perinuclearly (yellow arrowhead)

Fig. 3. Tat elicits a time-dependent decrease in mitochondrial membrane potential.

Cortical neurons were exposed to boiled Tat (control), Tat (100 nM) for the indicated times, or to FCCP (0.5 µM) for 30 min. Cells were then loaded with mitochondrial membrane permeant and potential-dependent dye, TMRE (5 nM), for 30 min (see Materials and Methods). The fluorescent intensity was quantified using ImageJ and expressed in arbitrary units (AU). Data are the mean ± SEM (n = 20 coverslips each time point). *p < 0.001, **p < .0001 vs control (ANOVA and Tukey’s test). TMRE tetramethylrhodamine ethyl ester

Fig. 4. Tat decreases mitochondrial size without altering mitochondrial number.

Cortical neurons were exposed to boiled Tat (control) or Tat (100 nM) for the specified time points alone (b–d) or in combination with mdivi-1 (10 μM) (e–g). a Cells were then fixed and stained for MAP2 (green) and TOM20 (red). Quantification of mitochondrial area (b, e), perimeter (c, f), and number (d, g) was then done as described in Materials and Methods. Data are the mean ± SEM of 20 neurons per treatment, normalized to control. *p < 0.05, **p < 0.01 vs control (ANOVA and Tukey’s test)

Fig. 5. Tat promotes a time-dependent increase in Drp1, but not Mfn2 levels.

Drp1 and Mfns protein levels were determined by western blot analysis in control and Tat-treated neurons. a Representative western blot analysis of cortical neuronal lysates probed with a Drp1 antibody. Blots were reprobed with beta-actin antibody. b Semi-quantification of Drp1 levels was done by densitometric analysis of the 82 kDa immunoreactive band normalized by the beta-actin (42 kDa) immunoreactivity. c Representative western blot analysis of cortical neuronal lysates probed with a Mfn2 antibody. d Semi-quantification of Mfn2 levels was done by densitometric analysis of the 86 kDa immunoreactive band normalized by beta-actin immunoreacivity. Data are the mean ± SEM of three independent experiments, normalized to control. *p < 0.05 vs control (ANOVA and Tukey’s test)

Fig. 6. Tat decreases pDrp1 in cortical neurons in a time-dependent manner.

Cortical neurons were exposed to Tat for the indicated times and then levels of pDrp1 were determined by western blot analysis in cell lysates following immunoprecipitation. a Representative western blot of lysates immunoprecipitated with anti-Drp1 antibody. The blot was analyzed with a pSer antibody. b Semi-quantification of pDRp1 immunoreactive band was done by densitometric analysis. Data, expressed as mean ± SEM, represent the average of three independent experiments. *p < 0.001, **p < 0.0001 vs control (ANOVA and Tukey’s test)

Fig. 7. Tat decreases pDrp1 S637 puncta but not pDrp1 S616 puncta in a time-dependent manner.

Cortical neurons were exposed to Tat for the indicated times, fixed, and then stained for pDrp1 S637 (a) or pDrp1 S616 (b). Quantification of pDrp1 S616 and pDrp1 S616 puncta was done as described in Materials and Methods. Data, expressed as mean ± SEM, are normalized to control and represent an average of three independent experiments (n = 10 neurons each experiment). *p < 0.05, ***p < 0.001 vs control (ANOVA and Tukey’s test)

Fig. 8. Calcineurin mediates the effect of Tat.

a Cortical neurons were exposed to boiled Tat (control), Tat (100 nM), or to cyclosporin A (CsA, 10 μM) and Tat for the indicated times. Lysates were then collected and calcineurin activity was measured as described in Materials and Methods. Data, expressed as mean ± SEM, are normalized to control and represent an average of three independent experiments (n = 2 each time point each experiment). *p < 0.01, **p < 0.001 vs control (ANOVA and Tukey’s test). b Cortical neurons were exposed to Tat (100 nM), CsA (10 μM), or CsA + Tat for the specified time points. Cells were then fixed and stained for MAP2 and TOM20 to label neuronal microtubules and mitochondria, respectively, as described in Fig. 3. Quantification of mitochondrial area (b), perimeter (c), and number (d) was then done as described in Materials and Methods. Data are the mean ± SEM of 20 neurons, normalized to control. *p < 0.05, **p < 0.01, ***p < 0.001 vs control (ANOVA and Tukey’s test)