(A) Activation of the Ty1 promoter by the NTP-depleting agent 6-azauracil eliminates the inhibitory effect of the 5′HIV-TR, both in a wild type and in a strain lacking TFIIS (dst1Δ). Experiment details as for Figure 1A. (B) The 5′HIV-TR inhibits GAL1-driven transcription when the promoter is weakly active. Cells transformed with plasmids pGAL1-HIV or pGAL1-HIVTARless were grown to mid-log phase in minimal medium with raffinose (0%) or with raffinose plus 0.02% galactose (0.02%) as carbon sources. (C) No effect of the 5′HIV-TR on mRNA levels is detected when an activated GAL1 promoter drives transcription, unless Tat and P-TEFb are present in the yeast cell. Cells transformed with plasmids pGAL1-HIV or pGAL1-HIVTARless, alone or together with p415GPD-CycT1, p414GPD-Cdk9 and p413GPD-Tat, were grown to mid-log phase as in B and incubated for 90 min in the presence of 2% galactose. (D) Tat and P-TEFb enhance Ty1-HIV expression in spt4Δ but they do not alter basal transcription in the wild type. Cells transformed with plasmids pTy1-HIV or pTy1-HIVTARless, alone or together with p415GPD-CycT1, p414GPD-Cdk9 and p413GPD-Tat, were grown to mid-log phase. In all cases, mRNA samples were taken and analyzed by Northern blot as in Figure 1A and the averages of three independent experiments are shown.

(A) shRNA-mediated depletion of SPT6 and CHD1. A population of HIV (LTR-Tat-IRES-GFP-LTR) latently-infected human cells was infected with Control, Spt6 (#1, target sequence CGCCTTGTACTGTGAATTTAT) or Chd1 (#3, GCAGTTGTGATGAAACAGAAT) shRNA expression lentiviruses (pLKO.1-Puro) and, 10 days after puromycin (2 mg/ml) selection, depletion of these factors was tested in Western blot with specific antibodies and tubulin as a loading control. (B) HIV reactivation in a heterogeneous population of HIV latently infected Jurkat cells, expressed as percentage of cells that become GFP-positive 10–15 days after infection with the indicated shRNA-expressing lentiviruses and continuous selection with puromycin. Values represent the mean±SD of three independent infection experiments. (C) Example of FACS analysis of GFP expression as a result of HIV promoter reactivation in a latently infected Jurkat clone after infection with a Spt6 specific shRNA-expressing vector or treatment with PMA (10 nM for 16 h) or TSA (400 nM for 16 h). The percentage of GFP-positive cells is indicated inside the corresponding gate. (D) HIV reactivation in individual cell clones. Four different clones containing single HIV latent integrations previously mapped were infected with the indicated shRNA-expressing lentiviruses, selected by puromycin treatment, and GFP expression was monitored by FACS over time. Values represent the mean±SD of percentage of GFP-positive cells of two independent experiments 10–15 days after infection.

(A) The 5′HIV-TR inhibits Ty1-HIV-1 expression and this inhibition partially depends on yDSIF. mRNA samples from the BY4741 wild-type yeast strains and from an isogenic spt4Δ strain, transformed with plasmids pTy1-HIV and pTy1-HIVTARless, were resolved in agarose gels and analyzed by Northern blotting. Quantification of the signals is shown, after normalizing with the levels of 25S rRNA. A typical result and the quantification of three independent experiments are shown. (B) ChIP analysis of RNApol II in the two indicated transcription units. ARG3 cells, isogenic to BY4741, except for the RPB1::cMyc allele, transformed with pTy1-HIV and pTy1-HIVTARless, were grown to mid-log phase. Cross-linked chromatin was immunoprecipitated with monoclonal anti-Myc antibody. PCR was conducted on two dilutions of whole cell extract (WCE) and two different amounts of immunoprecipitated DNA (IP; only the most diluted are shown). PCR primers flank segments located in the 5′, central and 3′ regions of each gene. Diagrams at the top indicate the position of the PCR amplicons relative to the HIV-1 transcription start site. A non-transcribed region adjacent to FUS1 was used as a control. A typical experiment for each gene is shown on the left and the averages of three independent experiments are quantified on the right. The HIV signals are marked with an asterisk. (C) The TAR RNA structure is not required for the repressive effect of the 5′HIV-TR on basal transcription. Northern analyses of Ty1-HIV, Ty1-HIVTARless and Ty1HIVTARmut in BY4741 were performed as in A. (D) Chromatin configuration of the 5′HIV-TR in yeast. Spheroplasts of formaldehyde-treated BY4741 cells containing pTy1-HIV and pTy1-HIVTARless were lysed and digested with MNase. After purification, DNA was quantified with real-time PCR as described in Materials and Methods, using the primers listed in Table S3. Signals were normalized against naked DNA by repeating the same procedure with control DNA, purified from formaldehyde-treated cells before digesting it with MNase. For comparison, the two profiles were represented as fractions of the signal obtained with amplicon 14 (+490/+537). Diagrams represent the estimated locations of positioned nucleosomes. Dashed ovals indicate possible alternative locations of nucleosomes in TyHIVTARless.

(A) The inhibitory role of the 5′HIV-TR is compromised in mutants affecting co-transcriptional chromatin reassembly. mRNA samples of the indicated mutants, transformed with pTy1-HIV and pTy1-HIVTARless, were analyzed by Northern blot, as in Figure 1A. Averages of three independent experiments (see Figure S5) were plotted. Dashed line represents all possible variations of mRNA levels that maintain the Ty1HIV/Ty1HIVTARless proportion of the wild type. The grey square covers all points with a level of Ty1HIV mRNA significantly higher than the wild type (more than two standard deviations), but with a similar level of Ty1HIVTARless mRNA that the wild type (within two standard deviations). (B) Representative Northern blots, comparing the mRNA levels of Ty1-HIV and Ty1-HIVTARless, in spt6-140 (FY137), spt16-197 (FY348), chd1Δ and their respective isogenic wild-type strains (FY120 and BY4741). (C) ChIP analysis of Spt6, Spt16 and Chd1. SJY25, DBY871 and DBY969 strains, transformed with pTy1-HIV, were grown to mid-log phase. Cross-linked chromatin was immunoprecipitated and quantified by real-time PCR as described in Materials and Methods. Averaged fold enrichments of three independent experiments, relative to an untranscribed telomeric region, are shown. (D) Chromatin configuration of the 5′HIV-TR in spt4Δ, spt6-140 and chd1Δ. Averages of three independent experiments are shown. Experimental details as in Figure 1D.

(A) Chromatin reassembly factors, recruited by the elongating form of RNApol II, stabilize nucleosomes on the transcribed region, keeping basal transcription at low rates. In the absence of chromatin reassembly factors, nucleosomal configuration of the transcribe region becomes instable, increasing basal transcription and eventually favoring promoter activation. (B) Chromatin reassembly factors may also contribute to the silencing of HIV by transcriptional interference. HIV proviruses integrated in highly expressed genes would remain untranscribed due to the repressive chromatin configuration established by chromatin reassembly factors during transcription elongation. In the absence of chromatin reassembly factors, nucleosomes repressing the 5′-LTR become instable, allowing transcription factors to activate the HIV promoter.