HIV-tat alters Connexin43 expression and trafficking in human astrocytes: role in NeuroAIDS

Abstract

Background: HIV-associated neurocognitive disorders (HAND) are a major complication in at least half of the infected population despite effective antiretroviral treatment and immune reconstitution. HIV-associated CNS damage is not correlated with active viral replication but instead is associated with mechanisms that regulate inflammation and neuronal compromise. Our data indicate that one of these mechanisms is mediated by gap junction channels and/or hemichannels. Normally, gap junction channels shutdown under inflammatory conditions, including viral diseases. However, HIV infection upregulates Connexin43 (Cx43) expression and maintains gap junctional communication by unknown mechanism(s).

Methods: Human primary astrocytes were exposed to several HIV proteins as well as to HIV, and expression and function of Connexin43- and Connexin30-containing channels were determined by western blot, immunofluorescence, microinjection of a fluorescent tracer and chromatin immunoprecipitation (ChIP).

Results: Here, we demonstrate that HIV infection increases Cx43 expression in vivo. HIV-tat, the transactivator of the virus, and no other HIV proteins tested, increases Cx43 expression and maintains functional gap junctional communication in human astrocytes. Cx43 upregulation is mediated by binding of the HIV-tat protein to the Cx43 promoter, but not to the Cx30 promoter, resulting in increased Cx43 messenger RNA (mRNA) and protein as well as gap junctional communication.

Conclusions: We propose that HIV-tat contributes to the spread of intracellular toxic signals generated in a few HIV-infected cells into surrounding uninfected cells by upregulating gap junctional communication. In the current antiretroviral era, where HIV replication is often completely suppressed, viral factors such as HIV-tat are still produced and released from infected cells. Thus, blocking the effects of HIV-tat could result in new strategies to reduce the damaging consequences of HIV infection of the CNS.

Fig. 1

Glial expression of Cx43 is upregulated in HIV-infected individuals. Cx43 and Cx30 glial expression was evaluated using human brain tissue sections obtained from uninfected and HIV-infected individuals with mild and HIVE. Brain sections were evaluated by immunohistochemistry and confocal microscopy. Astrocyte expression of Cx43 or Cx30 (FITC green) was evaluated using glial fibrillary acid protein (GFAP, an astrocyte marker, red staining) and HIV infection by staining with HIV-p24 antibodies. In uninfected tissue sections, Cx43 and Cx30 localized in astrocytes (uninfected row). In contrast, in tissues obtained from HIV-infected individuals, Cx43 was highly upregulated, while Cx30 was downregulated. DAPI staining was used in counter staining. Arrows represent colocalization of GFAP, Cx43, and HIV-p24. Thus, HIV infection increases Cx43 expression, but not Cx30, in astrocytes

Fig. 2

HIV-tat increased mRNA and Cx43 protein expression as well as gap junctional communication in human primary astrocytes. Human primary astrocytes were treated with recombinant HIV-tat protein (100 ng/ml), and expression and function of Cx43-containing channels were analyzed. a Staining for the nucleus (DAPI, blue staining, in the insets in the left column), Cx43 (Alexa 488, green staining), and GFAP (an astrocyte marker, Cy3, red staining) in untreated (control) and HIV-tat-treated conditions (HIV-tat) after 24 h of treatment. The last panel represents the merge of all colors. Bar 75 μm. Arrows denote gap junction plaques. b qRT-PCR for Cx43 and GAPDH mRNA using untreated (control) and HIV-tat-treated cultures of astrocytes at different time points (0, 6, 12, 24, and 48 h). No significant differences in mRNA Cx43 expression were detected in control cells (white bars). HIV-tat treatment increased Cx43 mRNA expression in a time-dependent manner (*p ≤ 0.001, n = 4, black bars). c Western blot analysis of Cx43 protein expression in control and HIV-tat-treated human astrocytes for 6, 12, 24, and 48 h. As a loading control, tubulin was used (tub). As a positive control for Cx43 and its phosphorylation, mouse astrocytes were used (+). d Dye coupling experiments using Lucifer Yellow (LY) showed that in both control and HIV-tat-treated cultures, dye spread into neighboring cells was 100 % (images not shown). However, HIV-tat increased the numbers of coupled astrocytes for each microinjection (*p ≤ 0.003, n = 4), indicating increased gap junctional communication

Fig. 3

HIV-tat did not increase Cx30 and mRNA protein expression and did not contribute to the enhanced gap junctional communication induced by HIV-tat in human primary astrocytes. Human primary astrocytes were treated with recombinant HIV-tat protein (100 ng/ml), and expression and function of Cx30-containing channels were analyzed. a Labeling for nuclei (DAPI, blue staining), Cx30 (Alexa 488, green staining), and GFAP (Cy3, red staining) in untreated conditions (control) and after 24 h HIV-tat treatment. The last panel represents the merge of all colors. Bar 75 μm. b qRT-PCR for Cx30 and GAPDH using untreated and HIV-tat-treated (24 h) cultures of astrocytes. No significant differences in mRNA Cx30 expression were detected in control cells (white bars). HIV-tat treatment decreased Cx30 mRNA expression in a time-dependent manner (^# p ≤ 0.001, n = 4, black bars). c Western blot analysis of Cx30 protein expression in control and HIV-tat-treated human astrocytes. Tubulin (tub) was used as a loading control. As a Cx30 positive control, mouse brain was used (+). HIV-tat decreased Cx30 expression in a time-dependent manner (Fig. 2c). d To examine the contribution of Cx30 to the increased gap junctional communication induced by HIV-tat, we reduced further the expression of Cx30 using siRNA. In this condition, we reduced Cx30 by at least 80 %; however, no changes in dye coupling were observed, suggesting that Cx43 mediate most of the communication (n = 4, p ≤ 0.005)

Fig. 4

HIV-tat binds to the Cx43 promoter, but not to the Cx30 promoter, in human primary astrocytes. Human primary astrocytes were transfected with pcDNA3.1+/tat101-flag, and binding of this HIV protein to the Cx43 and Cx30 promoter was assayed by ChIP. a Representative PCR curves of Cx43 promoter DNA amplification after 24 h of HIV-tat-FLAG transfection and subsequent ChIP. In the plot, amplification of input, tat-FLAG, irrelevant IgG, and non-transfected are shown. b Compilation of four ChIP experiments. Primary astrocytes were transfected with HIV-tat-FLAG to examine its binding to the Cx43 promoter using ChIP. Negative controls for the ChIP using IgG, or non-transfected astrocytes (data not shown) did not show significant amplification. Only HIV-tat-FLAG amplify and increase the binding at least 15-fold as compared to the input amplification (B, ChIP Cx30). ChIP analysis for Cx30 like that for Cx43 showed no HIV-tat-FLAG binding to the Cx30 promoter (n = 4, p = 0.004)

Fig. 5

Schematic representation of our working model. We believe HIV infects a small population of astrocytes, inducing the expression of HIV-tat and subsequent upregulation of expression of Cx43. This upregulation of Cx43 expression results in the maintenance of gap junctional communication and opening of Cx43 hemichannels on the surface of the astrocytes. Both Cx43-containing channels, GJ and hemichannels, enable toxic intracellular signals (probably IP₃ and calcium related) to spread into uninfected neighboring cells resulting in apoptosis. In contrast, HIV-infected astrocytes survive apoptosis, generating CNS reservoirs