Fig 1.
Inhibition of HIV-1 reactivation in UTX knocked down E4 cells.
(A) Representative FACS measuring d2EGFP expression in two UTX knocked down clones from E4 cells stimulated with TNF-α (10 ng/ml) overnight. (B) Quantification of HIV-1 reactivation in two clones from UTX knocked down E4 cells, UTX knocked down 2D10, and UTX knocked down 3C9 cells stimulated with indicated conditions. 3C9 cells harbor Nef+, Wt Tat HIV, while 2D10 cells are infected with H13L mutant Tat HIV-1. Note that H13L Tat is a functional Tat, as it has been shown that we can easily reactivate HIV-1 in 2D10 cells by many different stimuli. Error bars: SEM of 3 separate experiments. (C) d2EGFP expression in UTX knocked down E4 clones left untreated or stimulated with SAHA (2 μM) overnight. Kruskal-Wallis test was used for statistical calculation. (D) Western blot measuring the levels of UTX and H3K27me3 in E4 cells expressing scramble or UTX shRNAs.
Fig 2.
UTX functions as a transcription activator of HIV-1 transcription.
(A) ChIP assays measuring the enrichments of RNAP II, UTX, EZH2, and H3K27me3 at the Nuc-0, HIV-1 promoter, Nuc-1, and Nuc-2 regions of HIV-1 before or after TNF-α reactivation. Latent proviruses in E4 cells were reactivated by TNF-α (10 ng/ml) for an hour. Cells were treated with ethylene glycol bis(succinimidyl succinate) (EGS) (1.5 mM) for 30 minutes and subsequently crosslinked with formaldehyde (1%) for additional 10 minutes. (B) ChIP assays measuring the enrichment of RNAPII, UTX, H3K4me3, and H3K27me3 along HIV-1 genome in one UTX knocked down clone. Error bars: SEM of 3 separate quantitative real-time PCRs. Note that Nuc-0 and HIV-1 promoter primers amplify both 5’LTR and 3’LTR of HIV-1; Nuc-1 and Nuc-2 primers specifically bind to 5’LTR.
Fig 3.
Inhibition of UTX demethylase activity by GSK-J4 prevents HIV-1 reactivation in Jurkat T cells.
(A) Quantification of d2EGFP expression in E4 and 3C9 cells pretreated with increasing concentrations of GSK-J4 for 24 hours and stimulated with SAHA (2 μM) or TNF-α (10 ng/ml) overnight. Error bars: SEM of 5 separate experiments. One-way ANOVA, P <0.05, Bonferroni posttests * p < 0.05, ** p<0.01, *** p<0.001, n = 5. (B) ChIP assays measuring the enrichments of RNAP II, H3K4me3, and H3K27me3 at the 5’LTR of HIV-1 when latent proviruses were left unstimulated or reactivated for an hour by TNF-α (10 ng/ml) with or without the presence of GSK-J4 (5 μM). E4 cells were pretreated for 24 hours with GSK-J4 (5 μM) then further stimulated with TNF-α (10 ng/ml) for an hour. Error bars: SEM of 3 separate quantitative real-time PCRs.
Fig 4.
H3K4 methylation is crucial for HIV-1 reactivation mediated by H3K27 removal.
(A) Western blot measuring the indicated histone H3 methylation levels of cells expressing HA tagged wild type (wt) H3.3 or indicated mutants of H3.3. The upper bands (marked with *) from blots with antibodies against H3K27me3 and H3 are from the H3.3-HA variants. (B) Quantification of HIV-1 reactivation in E4 cells expressing wt H3.3 or indicated H3.3 mutants. Presented statistical significance was based on comparisons with H3.3 K27M variant. One-way ANOVA, n = 5 p<0.005, Bonferroni posttests, * p<0.05, ** p<0.01, *** p<0.001. Error bars: SEM of 5 different replications.
Fig 5.
Depletion of UTX elevates DNA methylation levels of latent HIV-1 in Jurkat T cells due to enhanced recruitment of DNMT3A to HIV-1 5’LTR.
(A) The map indicates the positions of CpG islands and primer binding sites along HIV-1 genome. (B) HIV-1 reactivation in UTX knocked down cells treated with 5-AZC. E4 cells were pretreated with 1 μM 5-AZC for 72 hours then left untreated or treated further with a combination of anti-CD3 (125 ng/ml) and anti-CD28 (500 ng/ml), 500 nM of SAHA, or 1 ng/ml of TNF-α overnight. HIV-1 reactivation in the cells was measured by FACS. Error bars: SEM of at least 3 separate replicates. T-test, n = 3, * p < 0.05. (C) MeDIP assay measuring the levels of methylated cytosine of HIV-1 in UTX knock down E4 cells or E4 cells treated with 5 μM of GSK-J4 for 48 hours. Error bars: SEM of 3 separate quantitative real-time PCRs. (D) Expression levels of DNMT3A in E4 cells transduced with lentiviruses harboring Dnmt3a-3xFlag and ChIP assays performed on cells treated with indicated conditions using Flag MS2 magnetic beads to pull down DNMT3A-3xFlag proteins. Cells were transduced and selected by puromycin (2 μg/ml) for 3 days. Before ChIP, cells were treated with GSK-J4 (5 μM) for 48 hours. Error bars: SEM of 3 separate quantitative real-time PCRs. (E) DNA demethylases TET1 and TET2 are involved in HIV-1 reactivation in Jurkat T cells. Reactivation of latent HIV-1 induced overnight by SAHA (1 μM) or TNF-α (10 ng/ml) in TET1 or TET2 knocked-out E4 cells. Error bars: SEM of 7 separate replicates. One-way ANOVA, p <0.05 Bonferroni posttests, n = 7 * p < 0.05 *** p<0.001.
Fig 6.
Inhibition of UTX by GSK-J4 promotes the silencing of HIV-1 in Th17 primary cells and elevates HIV-1 DNA methylation.
(A) Schematic of experimental design. (B) Relative levels of H3K27me3 compared to histone H3 levels quantified from Western blot on treated Th17 cells from 4 donors at Day 4. One-tailed, paired t-test, * p<0.05. (C) Silencing kinetics of HIV-1 in Th17 cells from 4 different donors in the presence of 0 μM (vehicle) or 1 μM of GSK-J4. Error bars: SEM of at least 3 separate biological replicates. Two-way ANOVA, Bonferroni posttests, * p<0.05 ** p<0.01. (D) MeDIP measuring the levels of methylated cytosine of HIV-1 at Day 4. Error bars: SEM of 4 different donors. One-tailed, paired t-test, n = 4, * p<0.05. (E) GSK-J4 inhibits the reactivation of latent HIV-1 by SAHA & IL15 or TCR stimulation in Th17 primary cells. Cells were pretreated for 24 hours with GSK-J4, then further stimulated overnight with SAHA & IL15 or Human T-Activator CD3/CD28 Dynabeads with the ratio of 25 μl of beads per 1 million cells. Quantification of HIV-1 reactivation under the indicated conditions. Note that only viable cells were chosen for HIV-1 reactivation measurement. Error bars: SEM of more than 3 separate replicates. One-way ANOVA, Bonferroni posttests n >3.
Fig 7.
GSK-J4 inhibits reactivation of latent HIV-1 in CD4 memory T cells isolated from well suppressed HIV-1 infected donors.
(A) Schematic diagrams describing the experimental designs. (B) Inhibition of latent HIV-1 reactivation in CD4+ memory T cells from HIV-1 infected patients on cARTs by UTX inhibitor. CD4+ memory T cells isolated from HIV-1 infected patients were treated for 3 days with indicated concentrations of GSK-J4. Cells were left untreated or further treated with human T-Activator CD3/CD28 Dynabeads overnight. Levels of spliced HIV-1 env mRNA were measured by EDITS assays. Levels of relative reactivation were normalized to the levels of spliced HIV-1 env mRNA induced by human T-Activator CD3/CD28 Dynabeads (presented as 100%) for each donor. (C) Expression levels of CD69 from treated CD4+ memory T cells measured by FACS using CD69 antibody. One-way ANOVA p < 0.005, Bonferroni posttests * p<0.05 ** p<0.01 *** p<0.001 were performed for both Fig 7B & 7C. (D) Intracellular H3K27me3 levels of treated CD4+ memory T cells measured by FACS using a H3K27me3 antibody. Cells were stained with Alexa Fluor 647-conjugated trimethyl histone H3 (Lys27) antibody (12158, Cell Signaling) and analyzed by FACS as described previously [22]. (E) Viability of cells by PI staining. After drug treatment, cells were stained with PI at the final concentration of 5 μg/ml for 5 minutes, then analyzed by FACS. Two-tailed paired t-test was performed for both Fig 7D & 7E, * p<0.05, ns: not statistically significant.
Fig 8.
Temporary induction of DNA methylation at 5’LTR of latent HIV-1 in CD4 memory T cells upon the inhibition of UTX by GSK-J4.
(A) The map of CpG dinucleotides (numbered from 1 to 9) relative to HIV-1 transcription start site along the 5’LTR CpG cluster. (B) The average DNA methylation (as %) of CpGs and the percentage DNA methylation at CpG island 1 of the 5’LTR CpG cluster from CD4+ memory T cells treated with the indicated concentration of GSK-J4. (C) The average DNA methylation of CpGs and the percentage DNA methylation at the CpG island 1 at the 5’LTR CpG cluster of HIV-1 from CD4+ memory T cells induced with GSK-J4 before or after GSK-J4 removal. Experiments were performed in single replicates with three different donors and in duplicates with one additional donor; One-tailed paired t-tests were performed on all experiments, * p<0.05, ** p<0.01, *** p<0.001, ns: not statistically significant. Note: different HIV-1 donors were utilized for this experiment than those utilized in Fig 8B.