NF-kappaB p50 promotes HIV latency through HDAC recruitment and repression of transcriptional initiation

Abstract

Cells latently infected with HIV represent a currently insurmountable barrier to viral eradication in infected patients. Using the J-Lat human T-cell model of HIV latency, we have investigated the role of host factor binding to the kappaB enhancer elements of the HIV long terminal repeat (LTR) in the maintenance of viral latency. We show that NF-kappaB p50-HDAC1 complexes constitutively bind the latent HIV LTR and induce histone deacetylation and repressive changes in chromatin structure of the HIV LTR, changes that impair recruitment of RNA polymerase II and transcriptional initiation. Knockdown of p50 expression with specific small hairpin RNAs reduces HDAC1 binding to the latent HIV LTR and induces RNA polymerase II recruitment. Similarly, inhibition of histone deacetylase (HDAC) activity with trichostatin A promotes binding of RNA polymerase II to the latent HIV LTR. This bound polymerase complex, however, remains non-processive, generating only short viral transcripts. Synthesis of full-length viral transcripts can be rescued under these conditions by expression of Tat. The combination of HDAC inhibitors and Tat merits consideration as a new strategy for purging latent HIV proviruses from their cellular reservoirs.

Figure 1

p50/RelA complexes displace NF-κB1 p50 from the latent HIV promoter. (A) Schematic of the HIV genome and location of oligonucleotide primers. ChIP primers for analysis of the HIV LTR were designed to span the nuc-0 to nuc-1 region, including the duplicated κB enhancers, the TATA box, and the transcriptional initiation site. Primers for analysis of initiated HIV transcripts were directed against TAR, and primers for downstream analysis were targeted at HIV tat. (B) NF-κB1 p50 is constitutively recruited to the HIV-1 promoter, whereas RelA and RNA Pol II are inducibly recruited. Fixed chromatin extracts from J-Lat 6.3 cells treated with 20 ng/ml TNF-α for 30 min or left unstimulated were immunoprecipitated with the indicated antibodies. Samples were assessed for enrichment in HIV LTR DNA, downstream HIV DNA (HIV DS), nonspecific control DNA (β-actin US), or transcriptionally active control DNA (β-actin DS) by UV visualization of PCR products in an ethidium bromide-stained agarose gel. Specific enrichment was quantitated by real-time PCR; mean of three measurements is indicated beneath each band image. Data are representative of three independent experiments. (C) A RelA–NF-κB1 p50-containing complex is recruited to the activated HIV LTR. Fixed chromatin extracts from J-Lat 6.3 cells treated as in panel B were immunoprecipitated with the indicated antibodies, and enriched complexes were subjected to a second round of immunoprecipitation with the indicated antibodies. Samples were assessed for enrichment of HIV LTR DNA. Nonspecific enrichment in β-actin DNA was not detected, supporting the specificity of these immunoprecipitations (data not shown). Data are representative of three independent experiments.

Figure 2

NF-κB1 p50 inhibits basal transcription of HIV in latently infected T cells. (A) Transient reduction of NF-κB1 p50 expression with anti-p50 shRNA vector. J-Lat 6.3 cells were cotransfected with an shRNA vector directed against NF-κB1 p50 and a plasmid expressing the cell-surface H-2K^k marker to identify transfected cells. H-2K^k-expressing cells were sorted and lysed 72 h after transfection, and samples were assessed for NF-κB p50 by Western blot. Data are representative of three independent experiments. (B) Reduction of NF-κB1 p50 expression is associated with increased basal HIV expression. Cells were treated as in panel A, and H-2K^k-expressing cells were assessed for HIV-LTR-driven expression of GFP by FACS. Experiments were conducted in triplicate; error bars represent standard deviation. Data are representative of three independent experiments.

Figure 3

HDAC1 recruitment and loss from the latent HIV LTR RNA Pol II complexes. (A) HDAC1 is present on the latent HIV-1 promoter and is lost when T cells are activated. Fixed chromatin extracts from J-Lat 6.3 cells treated with 20 ng/ml TNF-α for 30 min or left untreated were immunoprecipitated with the indicated antibodies. Samples were analyzed for enrichment in HIV LTR DNA by UV visualization of PCR products on an ethidium bromide-stained agarose gel. Specific enrichment was quantitated by real-time PCR; mean of three measurements is indicated beneath each band image. (B) HDAC1 is excluded from RelA- or RNA Pol II-containing complexes on the HIV-1 LTR. Fixed chromatin extracts from J-Lat 6.3 cells treated as in panel A were immunoprecipitated with the indicated antibodies and enriched complexes were subjected to a second round of immunoprecipitation with the indicated antibodies. Data are representative of three independent experiments.

Figure 4

Recruitment of RNA Pol II to the HIV promoter is modulated by histone acetylation. (A) Both the activation of NF-κB and addition of the HDAC inhibitor TSA induce histone hyperacetylation and RNA Pol II recruitment to the latent HIV LTR. Fixed chromatin extracts from J-Lat 6.3 cells treated with 20 ng/ml TNF-α for 30 min, TSA for 4 h, or untreated cells were immunoprecipitated with the indicated antibodies. Samples were assessed for enrichment in HIV-LTR DNA by UV visualization of PCR products on an ethidium bromide-stained agarose gel. Specific enrichment was quantitated by real-time PCR; mean of three measurements is indicated beneath each band image. (B) TNF-α, but not TSA, induces a processive RNA Pol II complex. Fixed chromatin extracts prepared as in panel A were analyzed for enrichment in downstream HIV DNA (HIV DS). (C) TSA induces initiation but not elongation of HIV RNA transcripts. Total RNA was extracted from J-Lat 6.3 cells treated as in panel A, and initiated or elongated HIV RNA transcripts were quantitated by real-time RT–PCR. Bars represent the mean of triplicate samples; error bars represent standard deviation. Data are representative of three independent experiments.

Figure 5

NF-κB1 p50 inhibits basal HIV expression in latently infected T cells by excluding RNA Pol II. (A) Stable shRNA knockdown of NF-κB1 p50 in latently infected T cells. Parental J-Lat 6.3 cells or stably transfected clones expressing an shRNA vector directed against a scrambled sequence or NF-κB1 p50 were selected and lysates were analyzed for p50 or control β-tubulin expression. (B) Fixed chromatin extracts from J-Lat 6.3 cells stably transfected with scrambled control shRNA or anti-NF-κB1 p50 shRNA treated with 20 ng/ml TNF-α for 30 min or left untreated were immunoprecipitated with the indicated antibodies. Samples were analyzed for enrichment of HIV LTR DNA by UV visualization of PCR products on an ethidium bromide-stained agarose gel. Specific enrichment was quantitated by real-time PCR; mean of three measurements is indicated beneath each band image. (C) Cells stably transfected with control- or NF-κB1 p50-shRNA were treated with 100 nM TSA for 2 h or left untreated. Total RNA was extracted, and initiated or elongated HIV RNA transcripts were quantitated by real-time RT–PCR. Bars represent the mean of triplicate samples; error bars represent standard deviation. Data are representative of three independent experiments.

Figure 6

p50 and TSA-sensitive HDACs mediate desensitization of the latent HIV LTR to Tat. (A) TSA enhances Tat induction of latent HIV expression. J-Lat cells were transfected with control, Tat, or RelA expression vectors, pulse treated with TSA (400 nM) for 1 h or left untreated, and transfected cells were assessed for GFP expression (left panel). Lysates were prepared and probed for β-actin, RelA, and Tat expression as a control (right panel). Note the low basal sensitivity of J-Lat cells to Tat induction and the sensitization to Tat induced by TSA treatment. (B) TSA treatment does not alter inherent Tat transactivating potential. Jurkat cells were transfected with an HIV LTR firefly luciferase reporter vector and a control Renilla luciferase vector in conjunction with control, Tat, or RelA expression vectors, pulse treated with TSA (400 nM) for 1 h or left untreated, and relative increase in firefly luciferase activity was quantitated. (C) p50 shRNA enhances Tat induction of latent HIV expression. Scramble- or p50-shRNA stable cells were transfected with control, Tat, or RelA expression vectors, pulse treated with TSA (400 nM) for 1 h or left untreated, and transfected cells were assessed for GFP expression (left panel). Lysates were prepared and probed for β-actin, RelA, and Tat expression as a control (right panel). Note the sensitization to Tat expression in p50-shRNA cells and relative lack of additional TSA sensitivity.

Figure 7

Model of p50-mediated repression of basal HIV expression in latently infected T cells. (A) NF-κB1 p50 homodimers bound to the latent HIV-1 promoter recruit HDAC1, which deacetylates regional histones and compacts local histone structure, thereby inhibiting the binding of RNA Pol II. (B) TNF-α liberates p50/RelA heterodimers, which displace constitutively bound p50/p50 homodimers present on the HIV LTR, thereby removing HDAC1. The regional shift in favor of HAT activity promotes increased acetylation of surrounding histones, relaxation of chromatin, and increased accessibility to RNA Pol II. Recruitment of CTD kinases by RelA induces transcriptional elongation. (C) TSA treatment inhibits HDAC1-mediated deacetylation of regional histones, inducing local histone acetylation, chromatin relaxation, and increased RNA Pol II binding. Under these conditions, RNA Pol II is non-processive owing to the absence of phosphorylation of its CTD. (D) shRNA knockdown of p50 displaces the p50–HDAC1 complexes from the latent LTR, promoting local histone acetylation and increased recruitment of RNA Pol II. Similarly, the lack of CTD kinases recruited to the LTR in this context produces a non-processive RNA polymerase complex.