Epigenetic regulation of HIV-1 latency by cytosine methylation

Abstract

Human immunodeficiency virus type 1 (HIV-1) persists in a latent state within resting CD4+ T cells of infected persons treated with highly active antiretroviral therapy (HAART). This reservoir must be eliminated for the clearance of infection. Using a cDNA library screen, we have identified methyl-CpG binding domain protein 2 (MBD2) as a regulator of HIV-1 latency. Two CpG islands flank the HIV-1 transcription start site and are methylated in latently infected Jurkat cells and primary CD4+ T cells. MBD2 and histone deacetylase 2 (HDAC2) are found at one of these CpG islands during latency. Inhibition of cytosine methylation with 5-aza-2'deoxycytidine (aza-CdR) abrogates recruitment of MBD2 and HDAC2. Furthermore, aza-CdR potently synergizes with the NF-kappaB activators prostratin or TNF-alpha to reactivate latent HIV-1. These observations confirm that cytosine methylation and MBD2 are epigenetic regulators of HIV-1 latency. Clearance of HIV-1 from infected persons may be enhanced by inclusion of DNA methylation inhibitors, such as aza-CdR, and NF-kappaB activators into current antiviral therapies.

Figure 1. A cDNA screen to identify novel regulators of HIV-1 latency.

(A) Flow cytometry of latent HIV-1 reactivation in indicated J-Lat cell lines after treatment with PBS or TNF-α. GFP-positive cells are indicated by gate and percentage. (B) Genome of retrovirus vector used in expression screen. (C) Flow cytometry of latent HIV-1 reactivation in J-Lat 6.3 cells after transduction with retrovirus expression vectors marked with the tomato fluorescent protein. Histograms indicate GFP expression after gating for tomato-positive cells.

Figure 2. MBD2 regulates HIV-1 latency and transcriptional repression.

(A) Latent HIV-1 reactivation in J-Lat 6.3 cells after transfection with the indicated expression plasmids and flow cytometry. Percent GFP-positive cells after gating for tomato-positive cells is shown. Experiment was performed in triplicate and error bars represent standard deviation. (B) Transcriptional activation of latent HIV-1 in J-Lat 6.3 after transfection with a siRNA that targets MBD2. Levels of MBD2 (left panel) or HIV-1 (right panel) mRNA were determined by reverse transcription and quantitative PCR and normalized to those after transfection with a non-targeting control siRNA. Data are representative of three different experiments. Error bars indicate standard deviation of qPCR results. (C) Flow cytometry of GFP expression in 293T cells that were mock transfected, transfected with methylated pEGFP-N1 (meGFP), or with unmethylated pEGFP-N1 (GFP). Gate indicates GFP-positive cells. (D) Flow cytometry of GFP expression in 293T cells cotransfected with methylated pEGFP-N1 (left panel) or unmethylated pEGFP-N1 (right panel) and an expression vector marked by the tomato fluorescent protein. Tomato-positive cells were gated to measure GFP expression in populations that received a control vector lacking an insert or one that encodes MBD2_1345–1947. Gates indicate GFP-positive cells.

Figure 3. HIV-1 encodes two CpG islands that are methylated during latency.

(A) Attributes of DNA surrounding HIV-1 transcript initiation site. Initiation: Transcription initiation site within HIV-1 LTR is indicated by arrow. CpG islands: Plot of GC content of DNA surrounding initiation site. Light grey areas indicate CpG islands 1 and 2. Vertical lines indicate CpGs. Nucleosome position: Locations of nucleosomes in HIV-1 promoter. Enhancers: Transcription factor binding sites. (B) Bisulfite-mediated methylcytosine mapping of latent HIV-1 in J-Lat cells. CpG islands 1 and 2 and J-Lat line are indicated. Each column represents one CpG position, with each circle in the column indicating either cytosine (open circles) or methylcytosine (filled circles) in an independently cloned DNA molecule. Asterisks indicate cytosines with statistically significant methylation.

Figure 4. Methylation-dependent recruitment of transcription factors and maintenance of HIV-1 latency.

(A) Bisulfite-mediated methylcytosine mapping of HIV-1 in J-Lat 6.3 cells treated with PBS (left panel) or aza-CdR (right panel). Asterisks indicate cytosines with a statistically significant lesser level of methylation in cells treated with aza-CdR. (B) Panel 1: Chromatin immunoprecipitation analysis of MBD2 recruitment to an unrelated DNA sequence between the β-actin and fascin-1 genes (unrelated), HIV-1 CpG island 1 (CpG1), and HIV-1 CpG island 2 (CpG2). Panels 2–4: Chromatin immunoprecipitation analysis of MBD2, HDAC2, or Sp1 recruitment to CpG island 2 in cells treated with PBS or aza-CdR. Data are representative of three independent experiments. Error bars indicate standard deviation of quantitative PCR results. (C) Flow cytometric analysis of latent HIV-1 reactivation in the indicated J-Lat clones after treatment with aza-CdR, TNF-α, or aza-CdR plus TNF-α. Histograms indicate GFP fluorescence. Gates indicate GFP-positive cells. (D) Bar graph representation of reactivation data. HIV-1 expression was normalized to control cells treated with PBS, and either GFP fluorescence (top panel) or percentage of GFP-positive cells (bottom panel) is displayed. Results are representative of three independent experiments. (E) Latent HIV-1 reactivation in J-Lat 6.3 treated with increasing doses of aza-CdR alone, in combination with TNF-α, or in combination with prostratin. Data points on the y-axis represent fluorescence at 0 uM Aza-CdR. GFP fluorescence was measured by flow cytometry and normalized to control cells treated with PBS. Error bars indicate standard deviation of three experiments. (F) Latent HIV-1 reactivation in the J-Lat cell line A2 treated with aza-CdR, TNF-α, or aza-CdR plus TNF-α. Gate indicates GFP-positive cells.

Figure 5. Cytosine methylation maintains HIV-1 latency in polyclonal Jurkat T cells.

(A) Flow cytometry of Jurkat T cells infected with HIV-1 R7/E⁻/GFP clone. Gates indicate GFP-positive (productively infected) and GFP-negative (latently infected and uninfected) cells. (B) Quantitative PCR to measure HIV-1 DNA in infected Jurkat cells. For polyclonal cell populations, days after infection are indicated. For J-Lat clones, cell line is indicated. Levels of HIV-1 DNA were normalized to cellular DNA. Y-axis indicates fold over uninfected Jurkat negative control. Experiment was performed in triplicate and error bars indicate standard deviation. (C) HIV-1 expression over time in a polyclonal population of latently infected and uninfected Jurkat T cells. GFP fluorescence was measured by flow cytometry. The time point at which cells were treated with aza-CdR plus TNF-α is indicated on the x-axis. (D) Bisulfite-mediated methylcytosine mapping of HIV-1 CpG islands in polyclonal Jurkat T cells that are latently infected (GFP-negative, upper panel) or productively infected (GFP-positive, lower panel). Asterisks indicate cytosines with a statistically significant greater level of methylation in the GFP-negative population. (E) Latent HIV-1 reactivation in a polyclonal population of latently infected and uninfected Jurkat T cells treated with aza-CdR, TNF-α, or aza-CdR plus TNF-α. HIV-1 expression was measured by flow cytometry for GFP, and the percentage of cells that express GFP is displayed. Error bars indicate standard deviation of three experiments.

Figure 6. Cytosine methylation is associated with HIV-1 latency in primary CD4+ T cells.

(A) Flow cytometric analysis of HIV-1 gene expression in CD4⁺ T cells productively infected under non-polarizing, Th1, or Th2 conditions. HIV-1 was detected by intracellular staining for p24^gag. Gates indicate gag-positive cells, and the percentage of positive cells is indicated. (B,C) Flow cytometric analysis of HIV-1 gene expression in (B) mock infected or (C) latently infected CD4⁺ T cells under non-polarizing conditions, either at the basal state or after reactivation with antibodies against CD3 and CD28. HIV-1 was detected by intracellular staining for p24^gag. Gates indicate gag-positive cells, and the percentage of gag-positive cells is indicated. (D) Bisulfite-mediated methylcytosine mapping of HIV-1 CpG island 2 in actively infected CD4⁺ T cells infected under non-polarizing, Th1, or Th2 conditions. (E) Bisulfite-mediated methylcytosine mapping of HIV-1 CpG island 2 in latently infected CD4⁺ T cells stimulated under non-polarizing conditions.

Figure 7. Molecular mechanisms of HIV-1 latency: Role of cytosine methylation.

The first 821 nucleotides of the HIV-1 genome are shown, with genome position indicated relative to transcriptional start site. The virus promoter is comprised of the U3, R, and U5 regions. CpG island 2 is indicated by vertical lines, and filled circles on top of lines indicate methyl groups. Arrow indicates HIV-1 transcriptional start site. (A) Transcriptional repression during HIV-1 latency. Sequence-specific factors recruit HDAC1. CpG island 2 is methylated and bound by MBD2, which mediates transcriptional repression by recruitment of NuRD. (B) NF-κB activation triggers loss of p50 homodimers and CBF-1 from the HIV-1 promoter, resulting in decreased HDAC recruitment and partial reactivation of latent virus. Increased RelA (p65) recruitment has also been reported. Effects upon Yin-Yang 1 recruitment are unknown. (C) Treatment with aza-CdR decreases methylation of HIV-1 CpG islands, leading to a loss of MBD2 from the HIV-1 promoter. Latent virus is partially reactivated by removal of the methylation block. (D) Potent, synergistic reactivation of latent virus when NF-κB activation is combined with aza-CdR treatment. This occurs via loss of NF-κB-responsive transcriptional repressors and MBD2 from the HIV-1 promoter.