Transcriptional competence of the integrated HIV-1 provirus at the nuclear periphery

Abstract

Spatial distribution of genes within the nucleus contributes to transcriptional control, allowing optimal gene expression as well as constitutive or regulated gene repression. Human immunodeficiency virus type 1 (HIV-1) integrates into host chromatin to transcribe and replicate its genome. Lymphocytes harbouring a quiescent but inducible provirus are a challenge to viral eradication in infected patients undergoing antiviral therapy. Therefore, our understanding of the contribution of sub-nuclear positioning to viral transcription may also have far-reaching implications in the pathology of the infection. To gain an insight into the conformation of chromatin at the site of HIV-1 integration, we investigated lymphocytes carrying a single latent provirus. In the silenced state, the provirus was consistently found at the nuclear periphery, associated in trans with a pericentromeric region of chromosome 12 in a significant number of quiescent cells. After induction of the transcription, this association was lost, although the location of the transcribing provirus remained peripheral. These results, extended to several other cell clones, unveil a novel mechanism of transcriptional silencing involved in HIV-1 post-transcriptional latency and reinforce the notion that gene transcription may also occur at the nuclear periphery.

Figure 1

The J-lat A1 cellular model. (A) Diagram of the HIV-1 construct integrated in position ChXp21.1 of J-lat A1. The 5′-LTR is followed by the packaging signal, the major splice donor site (SD) and a portion of the gag gene fused to the Rev-responsive element (RRE). The Tat acceptor site SA7 precedes a cassette containing the Tat101 gene fused to a flag tag (f-Tat101), an internal ribosome entry site (IRES) and the GFP reporter. Sites of cleavage by HindIII (H) and NotI (N) are also indicated. The position of the primers used in the 4C protocol is indicated by black arrows above the bait indicated by a double grey arrow. Below, the pre-mRNA and spliced RNA are shown together with the dimensions of the fragment amplified by RT–PCR with the indicated primers. (B) Activation of the J-lat A1 cell line. Cytofluorimetric analysis of cells was conducted in non-induced cells (left panel) and after induction with TPA for 15 h (right panel). More than 85% of cells were activated as shown by the expression of the GFP reporter. (C) Assembly of Cyclin T1 with Tat on induction of J-lat A1 cells. To show that induction by TPA induces f-Tat101 expression and the formation of the Tat∷Cyclin T1 complex, cells were lysed and immunoprecipitated with a flag antibody. Top panel, western blot analysis probing with an anti-flag antibody: IgL immunoglobulin light chains. Bottom panel, western blot probing with an antibody against Cyclin T1. (D) Time course of provirus expression in J-lat A1 cells. RT–PCR analysis was conducted for the spliced RNA (exon, top panels), the pre-mRNA (intron, middle panels) and the β-actin control (bottom panels) before (left panels) and after induction with TPA (right panels). Samples were analyzed at the indicated time points.

Figure 2

4C analysis at the site of HIV integration in J-lat A1 cells. (A) Diagram of the 4C protocol. The bait (grey line) derived from the HIV-1 provirus (see Figure 1A) is cross-linked to an unknown genomic locus (black line). After digestion with HindIII, the reaction is diluted and ligated to generate both intra- and inter-molecular ligations. Intra-molecular ligations could be reduced by NotI treatment. After reversal of cross-linking, a nested PCR is performed with primers within the bait pointing outwards, as shown also in Figure 1A. (B) 4C amplified fragments from J-lat A1 cells. Products of the nested PCR amplification of cells either not induced or induced with TPA were resolved on agarose gels. Molecular weight marker (1 kb, M) and positions of the major product of intra-ligation (asterisk) are shown. (C) 4C amplified fragments from J-lat A1 cells after NotI digestion. To reduce intra-molecular ligation in the TPA-activated cells, the cross-linked material was treated with NotI before performing the nested PCR. Molecular weight marker (1 kb, M) and positions of the major product of intra-ligation (asterisk) are shown. (D) 3C analysis of the interaction of the provirus and Ch12q12. To confirm 4C data, a 3C analysis was performed by hemi-nested amplification using a primer within the Ch12q12 region (see diagram). Control amplification was performed with primers mapping within the provirus (see Figure 1A). (E) Loss of the interaction between the provirus and Ch12q12 on induction of transcription. 3C analysis of the Ch12q12/provirus interaction was performed both in induced (left) in and non-induced (right) cells. Two-fold serial dilutions of the template show that in non-induced cells there is more cross-linked material for the interaction.

Figure 3

Analysis of the interaction of Chr12q12 with ChXp21.1. (A) J-lat A1 cells were analyzed in FISH with a probe for Ch12q12 (BAC RP11-379CZ4, green) and a probe for ChXp21.1 (BAC RP11-77013, red). The position of the two loci varied from colocalizing (left panel) to adjacent at ⩽0.6 μm (central panel) to distal at >0.6 μm (right panel). (B) Quantitative analysis of the distribution of the distances between Ch12q12 and ChXp21.1 in non-induced J-lat A1 cells (n=135). The density is intended as the frequency of distances between the two loci that fall within a discrete interval divided for the interval amplitude. The data were fitted and showed a bi-Gaussian distribution (solid line with the two Gaussians shown as dotted lines) with a subset of nuclei where the two genomic loci were closely associated. The mean values of the two Gaussians are also indicated by vertical lines (0.6 and 2.8 μm, respectively). (C) As a control, the same analysis as in panel B was conducted for the distances between chromosomes 12. This analysis showed a Gaussian distribution (dotted line). (D) The distribution of minimal distances between chromosome X and 12 is compared between non-induced (dotted line, n=135) and induced cells (solid line, n=114). Histogram shows the distribution of the induced cells. (E) Quantitative analysis of the frequency of nuclei showing colocalization (dark grey), proximity at ⩽0.6 μm (light grey) or distance at >0.6 μm (black). Numeric values are indicated below. Decrease of the colocalization is significant (χ², P=0.016).

Figure 4

Interaction of the Chr12 centromere with ChXp21.1 and the provirus. (A) J-lat A1 cells were analyzed in FISH with a probe for Ch12 α-repeats (red) and BAC RP11-77013 (green) mapping at ChXp21.1. Similar to Figure 3, the position of the two loci varied from colocalizing (left panel) to adjacent at ⩽0.6 μm (middle panel) to distal at >0.6 μm (right panel). (B) Quantitative analysis performed as in Figure 3E (non-induced, n=105; induced, n=103). Decrease of the colocalization is significant (χ², P=0.016). (C) J-lat A1 cells were analyzed in FISH with a probe for Ch12alpha repeats (red) and the provirus (green). Variation in the position of the two loci is shown as in panel A. (D) Quantitative analysis (non-induced, n=104; induced, n=78). Decrease of the colocalization is significant (χ², P=0.04).

Figure 5

Localization of the J-lat A1 provirus to the nuclear periphery. (A) The sub-population of J-lat A1 non-induced cells carrying the interaction between the provirus and Ch12 centromeric α-repeats was analyzed for distance of the interaction from the nuclear periphery (n=25). The abscissa represents the radius of the cell where 0 is the periphery and 0.5 is the centre. Values are normalized for the nuclear diameter. The density is intended as the frequency of distances between the two loci that fall within a discrete interval divided for the interval amplitude. (B) J-lat A1 cells were also analyzed for the localization of the provirus with respect to the nuclear periphery in the non-induced state (n=60) (B) and on induction (n=60) (C). Analysis was conducted as in Figure 5A. Inset: example of provirus detection. (D) Box plot analysis of the distribution of absolute distances from the periphery of the subset of interactions of provirus and Ch12 and centromeric α-repeats (median=1 μm). In addition, the distance of the provirus from the periphery, independent of its association with chromosome 12, is shown both in the inactive (median=1.3 μm) and in active (median=1.7 μm) states. Status of induction with TPA is indicated below. The differences in the distribution of the distances to the periphery of the provirus in the inactive or active state were not statistically significant (K–S test, P=0.5). (E) Box plot analysis of the distribution of distances from the periphery of the integrated provirus in different cell clones. When two integrations are present, as for U1 and HOS_B3, each of them is shown. Distances are presented as absolute values (values normalized to the diameter of the cell are shown in Supplementary Figure 1D). Values that differ significantly from those obtained from J-lat A1 are indicated by an asterisk (K–S test, P<0.01).

Figure 6

Localization of the HIV-1 nascent RNA to the nuclear periphery in single living cells. (A) HOS_A4 cells were analyzed for the localization of the HIV-1 nascent RNA with respect to the nuclear periphery in Tat-activated cells (n=62). Analysis of distances was conducted as in Figure 5. (A) Inset: example of the RNA in situ hybridization with an intronic probe that shows localization of nascent HIV-1 RNA at the site of transcription close to the nuclear periphery. (B) Localization of transcription in single living cells. HOS_A4 cells expressing Tat and EYFP_MS2nls were monitored in time for the localization of nascent RNA. Distances from periphery for 13 cells (±s.d.) are plotted at 5 min intervals for 30 min. A single nucleus is also shown at the various time points to show the position of the transcribing locus (white arrowhead).

Figure 7

Schematic drawing that summarizes the concepts emerging from the experimental data. In J-lat A1 cells, the provirus is found integrated in the X chromosome and localized at the nuclear periphery. Although all cells harbour a transcriptionally silent provirus, this could be found spatially associated close to the centromere of Chromosome 12 in a fraction of the nuclei. We identify two states of the silenced provirus: one associated with pericentromeric heterochromatin of chromosome 12 at the nuclear periphery (off–off) and one not associated with chromosome 12 that may be poised for transcription (off). On activation, this interaction is lost and HIV RNA is transcribed, possibly within a transcription factory, without changing its localization to the nuclear periphery (on).