DNA topoisomerase 1 represses HIV-1 promoter activity through its interaction with a guanine quadruplex present in the LTR sequence

Abstract

Background: Once integrated in the genome of infected cells, HIV-1 provirus is transcribed by the cellular transcription machinery. This process is regulated by both viral and cellular factors, which are necessary for an efficient viral replication as well as for the setting up of viral latency, leading to a repressed transcription of the integrated provirus.

Results: In this study, we examined the role of two parameters in HIV-1 LTR promoter activity. We identified DNA topoisomerase1 (TOP1) to be a potent repressor of this promoter and linked this repression to its catalytic domain. Additionally, we confirmed the folding of a Guanine quadruplex (G4) structure in the HIV-1 promoter and its repressive effect. We demonstrated a direct interaction between TOP1 and this G4 structure, providing evidence of a functional relationship between the two repressive elements. Mutations abolishing G4 folding affected TOP1/G4 interaction and hindered G4-dependent inhibition of TOP1 catalytic activity in vitro. As a result, HIV-1 promoter activity was reactivated in a native chromatin environment. Lastly, we noticed an enrichment of predicted G4 sequences in the promoter of TOP1-repressed cellular genes.

Conclusions: Our results demonstrate the formation of a TOP1/G4 complex on the HIV-1 LTR promoter and its repressive effect on the promoter activity. They reveal the existence of a new mechanism of TOP1/G4-dependent transcriptional repression conserved between viral and human genes. This mechanism contrasts with the known property of TOP1 as global transcriptional activator and offers new perspectives for anti-cancer and anti-viral strategies.

Keywords: DNA topoisomerases; Guanine quadruplex; HIV-1 LTR promoter; HIV-1 latency; HIV-1 transcription; Host-virus interaction; Transcriptional regulation.

Fig. 1

Top1 shRNA silencing induces a reactivation of HIV-1 LTR promoter activity in HeLa-LTR luciferase cells. HeLa LTR luciferase cells were transduced with pLKO.1 vector expressing shRNAs directed against Top1, Top2A or Top2B genes or a scrambled shRNA sequence (sh-), at ratios of 100–300 µL vector/2. 10⁶ cells. A The amount of TOP1, TOP2A and TOP2B proteins was evaluated 7 days post-transduction by SDS-PAGE and Western-Blot of cellular extracts. The Luciferase activity (B) or the levels of mRNA coding for the luciferase gene (C) were also quantified 7 days post-transduction (n = 5) at the highest transduction ratio

Fig. 2

Top1 shRNA silencing or CRISPR/Cas9 depletion induces a reactivation of HIV-1 LTR promoter activity in J-Lat cells. A–C J-Lat A1 and J-Lat 10.6 cells were transduced with pLKO.1 vectors expressing a shRNA directed against Top1, Top2A or Top2B genes or a scrambled shRNA sequence (sh-) (transduction ratio of 300 µL vector / 2 × 10⁶ cells). The % of GFP positive cells (n = 6 for J-Lat A1 and n = 3 for J-Lat 10.6) (A, B) or the level of mRNA coding for GFP (n = 3 for both cell lines) (C) were quantified 7 days post-transduction. D–F J-Lat A1 cells were transfected by a LentiCRISPRV2 plasmid expressing guide RNAs directed against Top1, Top2A or Top2B genes. The percentage of GFP-positive cells was measured after 10 days of puromycin selection (n = 9 for WT, n = 5 for Top1 bulk and n = 3 for TopA and Top2B bulk) (D). Three clones (K2, K18 and K30) were selected after transfection by a LentiCRISPRV2 targeting Top1 and analyzed for the % of GFP positive cells (n = 3) (E) or the level of mRNA coding for GFP (n = 3) (F)

Fig. 3

Catalytic active TOP1 restores the repression of HIV-1 LTR promoter in Top1 depleted J-Lat A1 cells. WT or two CRISPR TOP1 clones of J-Lat A1 cells were transduced by a pTRIP vector expressing WT or Y723F TOP1 or no protein (nc). A Seven days post-transduction, the presence of recombinant and endogenous TOP1 proteins was evaluated by SDS-PAGE and western blot directed against TOP1. B Concomitantly, the % of GFP-positive cells was measured by FACS and normalized to the % measured in WT cells, transfected by an empty vector (n = 4 for empty vector and n = 3 for expressed protein conditions)

Fig. 4

Characterization of G4 structures present in HIV-1 LTR. A HIV-1 promoter sequence and previously characterized G4s (LTR I to IV [50, 53, 55]) present along this sequence. B Oligonucleotides covering HIV-1 and c-myc G4s used in this study. Name, sequence and G4 folding scores (G4hunter and QGRS) of oligonucleotides covering PQSs present in HIV-1 LTR (LTR I, LTR II and LTR III, LTR III Short and LTR IV Short), mutated PQS LTR III (Mut1 to Mut6) and a PQS present in c-myc promoter. These oligonucleotides were used in the biophysical and structural studies presented below. C–F Biophysical and structural characterization of WT and mutated HIV-1 LTR III oligonucleotides G4 sequences. C Isothermal difference spectra with and without 100 mM KCl. D Thioflavin (ThT) fluorescence assay. E Circular dichroism (CD) spectra measured in 100 mM KCl. Oligonucleotides used in these studies are described in Additional file 5: Table S1

Fig. 5

TOP1 interacts with G4 LTR III and this interaction inhibits TOP1 DNA relaxation activity. A–C G4 DNA pull-down assays performed with biotinylated oligonucleotides covering PQS present in HIV-1 LTR or c-myc promoters (see Fig. 4B and Additional file 5: Table S1). These oligonucleotides, either folded into a G-quadruplex (G4 folded) or hybridized to their complementary strand (DS) were attached to streptavidin magnetic beads and incubated with recombinant human 6His-TOP1 enzyme (A) or Jurkat cellular extracts (B, C). After several washes with increased salinity buffers, retained TOP1 was quantified by SDS-PAGE and western blotting. All these experiments were repeated at least 3 times. D Kinetic analysis by SPR of LTR III DNAs binding to TOP1. G4 LTR III WT and Mut1 DNAs, prepared in the running buffer containing 50 mM potassium chloride, were injected at increasing concentrations (111, 333 and 1000 nM) over the protein immobilized by amine coupling. Six independent experiments were performed with DNA samples injected in duplicate. Red lines represent the recorded sensorgrams. E Effect of WT and mutated G4 LTR III on TOP1 catalytic activity. This activity was measured by a DNA relaxation assay performed with recombinant human TOP1 (70 nM), a supercoiled pBR322 plasmid (200 ng) and different concentrations of G4-folded LTR III oligonucleotides (2.2 to 9 μM, from right to left). At the end of the reaction, the Open Circular (OC), Relaxed (R) and Supercoiled (SC) forms of the plasmid were separated by electrophoretic migration on a 1% Agarose gel and stained with Ethidium Bromide

Fig. 6

TOP1-dependent repression of HIV-1 LTR promoter activity requires LTR III G4 folding and TOP1 interaction to this G4 structure. A Mutations disrupting G4 folding and TOP1 interaction reactivate HIV-1 promoter activity. The luciferase activities of HeLa LTR luciferase clones containing WT or mutated HIV-1 LTR upstream of the luciferase gene were measured and normalized for the activity of the WT clone (n = 3). B Scheme of the LTR HIV-1 promoter and GFP/Luciferase transcribed genes present in the J-Lat A1 and HeLa LTR Luciferase cell genomes. This scheme highlights the G4 region in the LTR (grey box) and the regions studied by TOP1 ChIP (Nuc0, Enhancer, Nuc1, GFP and Luc coding). C TOP1 occupancy along the HIV-1 promoter (Nuc1, Enhancer and Nuc0 regions) and transcribed gene (GFP region) present in the HIV-1 minigenome integrated in J-Lat A1 cells [63]. The occupancy was measured by ChIP assay (n = 3) performed in cells with Top1 gene previously silenced (shTop1) or control (sh-) (similar conditions as in Fig. 2). D TOP1 occupancy along the HIV-1 promoter (Nuc1, Enhancer and Nuc0 regions) and the transcribed gene (Luciferase coding region) measured by ChIP assay (n = 3) performed in HeLa LTR luciferase cells of WT, Mut1 or Mut4 LTR III sequence [62]

Fig. 7

G4 are enriched in the positive strand of TOP1 repressed cellular genes. A, B Boxplots of G4Hunter maximum scores (score threshold > 1.7, window size = 25 nts) in the positive (A) or negative (B) strands of the TSS-500 bp-TSS of genes that are significantly up-regulated (FC > 1.5) (up) or for a same number of genes (104) which RNA levels are not modified (ctrl) in shTop1 versus shCtrl HCT116 cells (transcriptomic data from [65]). Differences between the scores of the two groups were tested with a Wilcoxon rank sum test of which p-values are reported on the graphs. (C-D) Same analysis performed in the positive (C) or negative (D) strands of the promoter sequences of up-regulated and same number of control genes (123) in A549 cells (transcriptomic data from [14])