NF-kappaB-repressing factor inhibits elongation of human immunodeficiency virus type 1 transcription by DRB sensitivity-inducing factor

Abstract

Human immunodeficiency virus type 1 (HIV-1) is able to establish a latent infection during which the integrated provirus remains transcriptionally silent. In response to specific stimuli, the HIV-1 long terminal repeat (LTR) is highly activated, enhancing both transcriptional initiation and elongation. Here, we have identified a specific binding sequence of the nuclear NF-kappaB-repressing factor (NRF) within the HIV-1 LTR. The aim of this work was to define the role of NRF in regulating the LTR. Our data show that the endogenous NRF is required for transcriptional activation of the HIV-1 LTR in stimulated cells. In unstimulated cells, however, NRF inhibits HIV-1 LTR activity at the level of transcription elongation. Binding of NRF to the LTR in unstimulated cells prevents recruitment of elongation factor DRB sensitivity-inducing factor and formation of processive elongation complexes by hyperphosphorylated RNA polymerase II. Our data suggest that NRF interrupts the regulatory coupling of LTR binding factors and transcription elongation events. This inhibitory mechanism might contribute to transcriptional quiescence of integrated HIV-1 provirus.

FIG. 1.

NRF binds to the negative regulatory domain in the HIV-1 LTR. (A) HIV-LTR and proximal binding sites for several transcription factors are illustrated at the top. The NRF binding sequences in the IFN-β promoter (NI) and HIV-1 LTR (NH) as well as a functional mutant of the NRF binding site (Nm) are shown for comparison. Numbers indicate positions of nucleotides with respect to the transcriptional start site of the HIV-1 LTR and IFN-β promoter, respectively (+1). The gap (asterisk) was introduced to obtain maximal homology. (B) Three femtomoles of radiolabeled NH oligonucleotide was incubated with 1 μg nuclear extract from HeLa cells where indicated. A 100-fold concentration of cold competitor or 1 μg of antibody against NRF (α-NRF) or p65 (α-p65) was added to the binding reaction mixtures as indicated. Protein-DNA complexes were analyzed by EMSA. NRF-specific complexes are indicated by arrowheads. The fronts of unbound oligonucleotides were run out of the gel to separate the upper complexes.

FIG. 2.

NRF binding site in the HIV-1 LTR inhibits the transcriptional activity of NF-κB. (A) Schematic diagrams of reporter plasmids used for monitoring the NRF and NF-κB interaction. The plasmids contained a minimal synthetic TATA box to direct CAT reporter gene transcription. NF- κB binding sites of the HIV-1 LTR (κB), NRF binding sites of HIV-1 LTR (NH) or IFN-β promoter (NI), and mutated NRF binding element (Nm) were inserted upstream of the TATA box in the indicated orientation. (B) HeLa cells were transfected with 5 μg of the indicated CAT reporter plasmids and 1 μg of a firefly luciferase control plasmid. After 48 h CAT and luciferase reporter activities were determined. CAT activities were normalized to luciferase activities. CAT activity in cells transfected with empty vector p0 was set to 1 in each experiment. The mean activity, expressed as relative CAT activity, is shown ± the standard error of the mean (SEM) from three independent experiments. (C) HeLa cells were cotransfected with 1 μg of reporter plasmid pκB-NH and either 1 μg of an empty control plasmid pMBC (white bars) or NRF-expressing plasmid (black bars). In addition, 1 μg of either p65-expressing vector alone or a combination of expression vectors encoding p50 and p65 (0.5 μg each) or p52 and p65 (0.5 μg each) was cotransfected. After 48 h CAT and luciferase reporter activities were determined. CAT activities were normalized to luciferase activities. CAT activity in cells cotransfected with pκB-NH was set to 1 for each single experiment. The mean activity, expressed as relative CAT activity, is shown ± the SEM from three independent experiments.

FIG. 3.

Endogeous NRF inhibits HIV-1 LTR transcriptional activity. HeLa-tTA cells were stably transfected with tetracycline-regulated plasmids containing none (pTBC) or a 300-bp fragment of the NRF cDNA in antisense orientation (pTBC-NRFantisense). Two stable cell lines were obtained and were designated TBC and TBC-NRFantisense. (A) TBC and TBC-NRFantisense cells were cultured in the presence (+) or absence (−) of 2 μg/ml tetracycline. After 30 h, endogenous NRF expression was determined by Western blotting using α-NRF antibody. The NRF-specific protein band is indicated. As control, p65 expression was determined using α-p65 antibody. (B) TBC and TBC-NRFantisense cells were cultured in the presence (white bars) or absence (black bars) of 2 μg/ml tetracycline. After 24 h cells were cotransfected with 1 μg of HIV-1 luciferase reporter plasmid (pHIVLTR-LUC) and 1 μg of CAT-expressing control plasmid. Where indicated, cells were incubated with 10 mM 2-AP. After 48 h cells were harvested and luciferase and CAT activities were determined. In each experiment, luciferase activities were normalized to CAT activities. Luciferase activity of TBC cells in the presence of 2 μg/ml tetracycline was set to 1 in each single experiment. The mean value of the relative luciferase activity is shown ± the standard error of the mean (SEM) from four independent experiments. (C) TBC-NRFantisense cells were incubated with tetracycline where indicated. After 24 h cells were incubated with 10 mM 2-AP, and 4 h later cells were stimulated with synthetic dsRNA where indicated. After 2 h PKR phosphorylation was monitored by direct Western blotting using α-P-PKR antibody. Phosphorylated PKR signal is indicated by an arrowhead. (D) TBC-NRFantisense cells were cultured in the presence (white bars) or absence (black bars) of 2 μg/ml tetracycline. After 24 h cells were cotransfected with 1 μg of HIV-1 luciferase reporter plasmids (pHIVLTR-LUC or pM-HIVLTR-LUC) and 1 μg of CAT-expressing control plasmid. After 48 h cells were harvested and luciferase and CAT activities were determined. Luciferase activities were normalized to CAT expression levels. In each single experiment, luciferase activity of TBC-NRFantisense cells transfected with pHIVLTR-LUC in the presence of 2 μg/ml tetracycline was set to 1. The mean value of the relative luciferase activity is shown ± the SEM from three independent experiments.

FIG. 4.

Endogenous NRF is required for activation of HIV-1. (A) On the right side, a schematic description shows the generation of pseudotyped HIV-1 particles in 293T cells (see Materials and Methods for details). TBC and TBC-NRFantisense cell lines (as outlined in the legend of Fig. 3) were cultured in the presence (white bars) or absence (black bars) of 2 μg/ml tetracycline. After 16 h cells were infected with pseudotyped HIV-1 virus (500 ng of viral p24 antigen) containing the luciferase reporter gene. Cells were harvested after 40 h, and luciferase activity was determined. Luciferase activity of TBC cells in the presence of 2 μg/ml tetracycline was set to 1 in each single experiment. The mean value of relative luciferase activity is shown ± the SEM from four independent experiments. (B) TBC and TBC-NRFantisense cells were cultured in the presence (white bars) or absence (black bars) of 2 μg/ml tetracycline. After 24 h cells were cotransfected with 1 μg of HIV-1 luciferase reporter plasmid (pHIVLTR-LUC) and 1 μg of CAT-expressing control plasmid. After 24 h cells were stimulated with 10 ng/ml PMA. Eighteen hours later cells were harvested and luciferase and CAT activities were measured. Luciferase activities were normalized to CAT expression. The mean value of normalized luciferase activity (RLU) of PMA-stimulated transfected cells is shown ± the standard error of the mean from four independent experiments.

FIG. 5.

NRF binding to LTR is a prerequisite for the effects on transcriptional activity of the HIV-1 LTR. HeLa cells (A) and Jurkat T cells (B) were cotransfected with 10 μg of either wild-type HIV-1 LTR reporter containing an intact NRF binding site (pHIVLTR-LUC) or a mutant HIV-1 LTR reporter lacking the NRF binding site (pM-HIVLTR-LUC) and 2.5 μg of a Renilla luciferase expression plasmid. DNA amounts were kept constant by adding empty expression vector pMBC in each transfection experiment. After 18 h cells were left untreated (white bars) or were stimulated with 10 ng/ml PMA in combination with 500 ng/ml ionomycin (black bars). Cells were lysed 24 h after transfection, and firefly and Renilla luciferase reporter activities were determined. Firefly luciferase activities were normalized to Renilla luciferase activities. Normalized luciferase activity of untreated cells transfected with wild-type LTR reporter (pHIVLTR-LUC) was set to 1 in each single experiment. The mean values of luciferase activities compared to the wild-type LTR reporter activity in untreated cells are shown ± the standard error of the mean from three independent experiments.

FIG. 6.

NRF regulates transcription initiation and elongation from the HIV-1 LTR. (A) The pHIVLTR-LUC plasmid and probes corresponding to initiated or elongated transcripts are illustrated. Lengths of protected regions are indicated. (B) HeLa cells were transfected either with pHIVLTR-LUC or with pM-HIVLTR-LUC as indicated. A Renilla luciferase expression plasmid (phRG-B; Promega) was cotransfected to estimate the transcriptional efficiency in each experiment. After 48 h cells were stimulated with 10 ng/ml PMA for the indicated times or left untreated. Then, cells were harvested and total RNA was purified. Twenty-micrograms aliquots of RNAs were hybridized to antisense initiation or elongation probes of transfected constructs. As control, 10 μg of total RNA was hybridized to transferrin receptor coding sequences. Hybridization products were subjected to the S1 mapping analysis and were analyzed by electrophoresis through 6% denaturing polyacrylamide gel electrophoresis. The autoradiography of protected transcripts is shown. (C) Before electrophoresis the total amounts of protected fragments were measured (in counts per minute). The number of transcripts per 20 μg of total RNA is shown ± the standard error of the mean from three independent experiments.

FIG. 7.

Recruitment of transcription factors to the HIV-1 LTR. HIV-LTR and proximal binding sites for several transcription factors are illustrated at the top. A ChIP assay was performed using soluble chromatin extract from control or PMA- and ionomycin-stimulated (PMA/Iono) HeLa and Jurkat T cells transfected with wild-type HIV-1 LTR containing intact NRF binding sites (pHIVLTR-LUC) or mutant HIV-1 LTR missing NRF binding sites (pM-HIVLTR-LUC), as indicated. In order to internally control the experiments for variations in transfection efficiency, a Renilla luciferase expression plasmid (phRG-B; Promega) was cotransfected. Cells were harvested, and an aliquot was subjected to a Renilla luciferase assay (data not shown). The extracts were precipitated using the indicated antibodies or left untreated as an internal quality control (input). The precipitated DNAs were used for PCR with primers spanning the core and LTR-proximal region as indicated at the top. Results are representative of four independent ChIP experiments.