Caveolin-1 suppresses human immunodeficiency virus-1 replication by inhibiting acetylation of NF-κB

Abstract

Caveolin-1 is an integral membrane protein primarily responsible for the formation of membrane structures known as caveolae. Caveolae are specialized lipid rafts involved in protein trafficking, cholesterol homeostasis, and a number of signaling functions. It has been demonstrated that caveolin-1 suppresses HIV-1 protein expression. We found that co-transfecting cells with HIV-1 and caveolin-1 constructs, results in a marked decrease in the level of HIV-1 transcription relative to cells transfected with HIV-1 DNA alone. Correspondingly, reduction of endogenous caveolin-1 expression by siRNA-mediated silencing resulted in an enhancement of HIV-1 replication. Further, we observed a loss of caveolin-mediated suppression of HIV-1 transcription in promoter studies with reporters containing mutations in the NF-κB binding site. Our analysis of the posttranslational modification status of the p65 subunit of NF-κB demonstrates hypoacetylation of p65 in the presence of caveolin-1. Since hypoacetylated p65 has been shown to inhibit transcription, we conclude that caveolin-1 inhibits HIV-1 transcription through a NF-κB-dependent mechanism.

Figure 1. Overexpression of caveolin-1 suppresses HIV-1 protein expression and virus release in 293T cells

Here, 293T cells were co-transfected with HIV-1 molecular clone pYU-2 and caveolin-1expression construct or vector control. Cells and supernatants were harvested 48 hours post-transfection. A. HIV-1 Gag protein expression was analyzed by Western blot analysis with an anti-Gag mAb (upper panels). We confirmed expression of caveolin-1 by probing with a polyclonal antibody (middle panels) and used β-tubulin as a loading control (bottom panel). B. Virus particle production was assessed by performing HIV-1 p24 ELISA using supernatants from transfected cells. Graph represents 3 independent experiments done in replicates with S.E.M., p value< 0.05. C. 293T cells were first transfected with one of 2 independent caveolin-1 specific or control siRNAs, then 48 hours later transfected with pYU-2. Protein expression was analyzed after an additional 24 hours.

Figure 2. Caveolin-1 overexpression in 293T cells does not cause global changes in free cholesterol content

293T cells transfected with caveolin-1 or empty vector DNA were stained with fluorescent cholesterol-binding reagent filipin at 48 hours post –transfection to assess total free cholesterol in the cells. A. As a positive control for the assay, untransfected cells were treated with conditions that stimulated cholesterol efflux (5uM TO901317 for 48 hours)or a cholesterol sequestering agent (1% Beta-Cyclodextrin for 1 hour), prior to filipin staining as described in Methods. Cellular free cholesterol content was determined by flow cytometry and graphed in a histogram as indicated. Dotted lines are control cells that were not stained. B. Caveolin-1 transfected or control cells were stained with filipin to determine the cellular cholesterol levels of transfected cells. Data for filipin-stained cell populations are plotted as histograms.

Figure 3. Caveolin-1 suppresses HIV-1 promoter activity

A. 293T cells were transfected with caveolin-1-RFP expression construct (filled histogram) or RFP control vector (open histogram), along with HIV-1-LTR GFP reporter. HIV-1 LTR GFP signal was measured by flow cytometry 48 hours post transfection and data was plotted in a histogram, where MFI of samples are indicated. B. 293T cells were co-transfected with caveolin-1 (stripped) or vector control (black) and either Cytomegalovirus (CMV) - or HIV-1 LTR-promoter driven Gag-GFP plasmids. Cells were harvested 48 hours post-transfection and GFP-MFI signal from the individual promoters was analyzed by flow cytometry. Graphs represent 3 independent experiments done in replicates with S.E.M., p value< 0.05.

Figure 4. Caveolin-1 suppresses both tat-independent and dependent HIV-1 LTR transcription

Cells were transfected with caveolin-1DNA and transcriptional activity from HIV-1 reporters was measured by flow cytometry. A. 293T cells were co-transfected with caveolin-1 (stripped) or empty vector control (black) along with HIV-1 LTR reporter and −/+HIV-1 Tat. HIV-1-LTR-GFP fluorescence was measured 24 hours post-transfection by flow cytometry. Displayed is the average of 3 independent experiments. B. THP-1 monocytic cells were transfected with vector or caveolin along with HIV-1 LTR reporter and treated with 50 ng/ml of TNF-α for the last 2hours of cell culture before harvest. The GFP signal from the HIV-1-LTR reporter was measured by flow cytometry. Caveolin-transfected samples were normalized to control vector transfected counterparts. TNF-α treated samples were normalized to 1.C. THP-1 cells were co-transfected with caveolin-1 expression construct or empty vector control along with RelA expression construct and HIV-1 LTR GFP reporter. All graphs represent at least 2 independent experiments done in replicates with S.E.M., p value< 0.05

Figure 5. Caveolin-1-dependentsuppression of HIV-1 transcription is mediated by NF-κB

A. 293T cells were transfected with caveolin-1-RFP or control RFP expression vector with a NF-κB-GFP reporter construct then treated with 50 ng/mL of TNF-α for last 14 hours of the experiment. Cells were visualized by fluorescence microscopy (top) 24 hours post-transfection and collected and analyzed for NF-κB activity (bottom) by flow cytometry. Relative expression (MFI-GFP) is graphed for RFP-caveolin-1 expression construct (stripped bars) or RFP control vector (black bars) transfected samples. B. Cells were transfected as in (A) with HIV-1-LTR GFP reporters (wild-type, wtLTR or mutated NF-κB binding site, nkLTR) along with RFP-caveolin-1 expression construct (stripped bars) or RFP control vector (black bars) and 24 hours post-transfection reporter activity was analyzed by flow cytometry. C. 293T cells were transfected with RFP-caveolin-1 (stripped bars) or RFP control vector (black bars) expression construct with HIV-1-LTR GFP reporter, then treated with IκB kinase inhibitor peptide, SN50 (50 μg/mL) for 18 hours. HIV-1-LTR reporter activity was analyzed by flow cytometry. All data shown represents 3 independent experiments. Graph represents three independent experiments done in replicates with S.E.M. p value < 0.05.

Figure 6. Caveolin-1 overexpression decreases total acetylation of p65

293T cells were transfected with caveolin-1 and myc-tagged RelA expression constructs as indicated. A. Total cells lysates were immunoprecipitated using either anti-myc or anti-acetyl-lysine antibodies and analyzed by and Western blot analysis. Blots were probed with monoclonal anti-p65 antibodies to determine efficacy of pulldown. B. Electrophoretic mobility shift assays (EMSA) performed with increasing volumes of nuclear extracts (0.5-4uL) from caveolin-1 and RelA transfected 293T cells. DNA binding was determined using biotinylated-HIV-1-LTR oligonucleotides and chemiluminescent detection. Replica nuclear extract samples were analyzed for p65 expression by Western blot (lower panel).

Figure 7. Model of caveolin-1 mediated suppression of HIV-1 transcription via hypoacetylation of NF-κB p65 subunit

Expression of caveolin results in hypoacetylation of p65 resulting in increased binding to DNA while making the protein transcriptionally inert.