Human immunodeficiency virus-restricted replication in astrocytes and the ability of gamma interferon to modulate this restriction are regulated by a downstream effector of the Wnt signaling pathway

Abstract

Astrocyte dysregulation correlates with the severity and the rate of human immunodeficiency virus (HIV)-associated dementia (HAD) progression, highlighting a pivotal role for astrocytes in HIV neuropathogenesis. Yet, astrocytes limit HIV, indicating that they possess an intrinsic molecular mechanism to restrict HIV replication. We previously established that this restriction can be partly overcome by priming astrocytes with gamma interferon (IFN-gamma), which is elevated in the cerebral spinal fluid of HAD patients. We evaluated the mechanism of restrictive HIV replication in astrocytes and how IFN-gamma priming modulates this restriction. We demonstrate that the downstream effector of Wnt signaling, T-cell factor 4 (TCF-4), is part of a transcriptional complex that is immunoprecipitated with HIV TAR-containing region in untreated astrocytes but not in IFN-gamma-treated cells. Blocking TCF-4 activity with a dominant-negative mutant enhanced HIV replication by threefold in both the astrocytoma cell line U87MG and primary fetal astrocytes. Using a TCF-4 reporter plasmid, we directly demonstrate that Wnt signaling is active in human astrocytes and is markedly reduced by IFN-gamma treatment. Collectively, these data implicate TCF-4 in repressing HIV replication and the ability of IFN-gamma to regulate this restriction by inhibiting TCF-4. Given that TCF-4 is the downstream effector of Wnt signaling, harnessing Wnt signaling as an intrinsic molecular mechanism to limit HIV replication may emerge as a powerful tool to regulate HIV replication within and outside of the brain.

FIG. 1.

Comparative analysis of early and late HIV reverse transcription (RT) between infected untreated and infected IFN-γ-treated astrocytes. U87MG cells were left untreated or treated with IFN-γ prior to HIV infection as described in Materials and Methods. Total DNA was isolated and amplified for either early HIV reverse transcription (reverse transcription initiation) using primer pair R and U5 at 24 h postinfection or late HIV transcription using primer pair R and 5NC at 96 h postinfection. (A) Comparison of amplified HIV DNA between untreated and IFN-γ-treated cultures. (B) All of the controls for real-time PCR including GAPDH amplification from HIV-positive (pos.) and -negative (neg.) cultures. Data shown are representative of at least two experiments. In Fig. 1, 2, and 3, Rn stands for reading normalized, which equals the SyberGreen value divided by the ROX reference dye value.

FIG. 2.

Chromatin structures of TAR and LTR sequences are competent for transcriptional activities. U87MG cells were HIV infected or left uninfected. At 96 h, ChIP was performed by immunoprecipitating DNA associated with acetylated histones by using acetylated histone H2B antibody or no antibody. The immunoprecipitated DNA was subsequently amplified for TAR sequences between genomic locations +1 and +153 (A) and LTR sequences between genomic locations −460 and −206 (B). Conventional ChIP assay controls were included and shown in this figure, such as an “input control,” referring to amplification of the DNA before the IP step, and a “no antibody control,” referring to amplification of DNA after the IP step but without the addition of an acetylated histone H2B antibody. Data are representative of at least three experiments. Ac, acetylated; Ab and ab, antibody; pos., positive; neg., negative.

FIG. 3.

TCF-4 is immunoprecipitated with HIV TAR-containing region in untreated but not IFN-γ-treated cultures. U87MG cells were left untreated or primed with IFN-γ, as described previously (7), and infected with HIV, and ChIP was performed using TCF-4 antibody for IP. (A) Comparison of untreated and IFN-γ-treated TAR-containing HIV DNA immunoprecipitated with TCF-4. (B) Additional controls of uninfected cultures, input DNA, and no-antibody controls. pos., positive; neg., negative; Ab, antibody.

FIG. 4.

(A and B) Transfection efficiencies of U87MG cells (A) and HFA (B). The cells were transfected with GFP (pMaxGFP) plasmid using the Nucleofector system (Amaxa). Expression of GFP was evaluated by flow cytometry at day 3 posttransfection. The black histogram represents an isotype control, and the gray histogram represents GFP expression. TCF-4 inhibition abrogates restriction of HIV replication in untreated/non-IFN-γ-primed cells. (C and D) U87MG cells (C) or primary HFA (D) were transfected with the TCF-4 dominant-negative (DN) mutant or GFP construct, keeping the DNA amount constant between the two cultures. The cells were then infected with HIV. Data represent mean HIV p24 levels (pg/ml) ± standard errors of the means. HIV p24 level was measured by conventional ELISA at day 7 postinfection. Asterisks in panels C and D indicate P = 0.008 and P = 0.002, respectively, calculated using the Wilcoxon rank sum test between dominant-negative TCF-4- and GFP-transfected cells. Cultures infected with HIV without GFP transfection had p24 values similar to those of HIV-infected, GFP-transfected cultures, as we previously reported (7).

FIG. 5.

Impact of IFN-γ on Wnt signaling. U87MG cells were left untreated or treated with IFN-γ for 24 h and then transfected with either TOPflash (containing four native TCF/LEF binding sites) and Renilla constructs or GFP and Renilla constructs. Transfected cells were then cultured with or without their initial treatment (with or without IFN-γ). Luciferase relative light activity (relative light units) was measured 24 h later using a luminometer and normalized to Renilla activity. Data are based on at least three experiments and are presented as the increase (n-fold) in luciferase relative light units over Renilla ± standard deviation. The asterisk shows P < 0.0001 between untreated (first column) and IFN-γ-treated (second column) cultures using the Bonferroni multiple comparison test.