Fig 2: MATR3 depletion impairs full reactivation of HIV-1 from latency. (A) Efficiency of MATR3 depletion in J-lat 8.4 cells. MATR3 depletion was obtained with shMATR3 as described above (Fig. 1A). Three days after puromycin selection, MATR3-depleted cells were harvested for immunoblotting. β-Actin is the protein loading control. (B) MATR3 depletion leads to reduced HIV-1 Gag expression. J-lat 8.4 cells were transduced with shMATR3. Three days after puromycin selection, MATR3-depleted cells were stimulated with TNF-α (30 ng/ml), and 48 h postinduction, samples were collected for immunoblotting. β-Actin is the protein loading control. (C) MATR3 depletion leads to the decrease in the extracellular release of HIV-1 Gag. MATR3-depleted J-lat cells generated as described above were stimulated with TNF-α, and supernatant was harvested 48 h postinfection. Virion production in the supernatant was quantified by HIV-1 p24 ELISA. Results from three separate experiments are shown as mean values ± SD. (D) HIV-1 LTR-driven transcription is not affected by MATR3 knockdown. MATR3-depleted J-lat 8.4 cells treated as described above were analyzed for GFP expression by cytofluorimetry. The results of three independent experiments are shown as mean values ± SD. (E) Expression of MATR3 and other cofactors in PBLs. Freshly isolated lymphocytes from peripheral blood from buffy coat were stimulated with PHA/IL-2 for 3 days. Cell lysates were blotted with the indicated antibodies. Loading control is the blotting membrane stained with Ponceau S. (F) Expression of MATR3 in primary CD4+ T lymphocytes. Primary CD4+ T cells were purified by positive selection with magnetic beads and stimulated with PHA/IL-2. Three and 6 days postinduction, cell lysates were blotted with the indicated antibodies using both α-tubulin and vimentin as loading controls.

Fig 3: Ectopic MATR3 promotes HIV-1 replication. (A) Overexpression of MATR3 in Jurkat cells. Flag-tagged MATR3 (f-MATR3) was overexpressed in Jurkat cells following transduction with a lentivector. Immunoblot with the indicated antibodies was performed following blasticidin selection. β-Actin is the protein loading control. (B) Overexpression of MATR3 leads to an increase of Gag expression. Jurkat cells were transduced with f-MATR3 or with a control lentivector. Five days after blasticidin selection, cells were infected with replication-competent HIV-1 pNL-4.3 virus and harvested 48 h postinfection for immunoblotting. β-Actin is the protein loading control. (C) MATR3 RNA is increased both in the nucleus (NF) and in the cytoplasm (CF). Nucleocytoplasmic fractionation of Jurkat cells was performed as described in Materials and Methods. Quantification of MATR3 mRNA normalized to GAPDH expression is presented as fold changes compared to mock-transduced cells. Average values of duplicate independent experiments are shown, with standard deviations (SD) and P values as described in Materials and Methods. (D) HIV-1 unspliced RNAs decrease in the nucleus when MATR3 is overexpressed. Quantitative analysis of RNA levels was performed as described above. (E) HIV-1 multiply spliced RNAs are not affected when MATR3 is overexpressed. Quantitative analysis of RNA levels was performed as described above. (F) Ectopic f-MATR3 expression in PBLs. Lentivectors for the expression of f-MATR3 were used to transduce PBLs. Cells were then lysed and blotted for MATR3 and β-actin as loading control. (G) Ectopic f-MATR3 expression enhances HIV-1 replication in PBLs. PBLs expressing f-MATR3 were infected with replication-competent HIV-1 NL4.3 virus. The supernatant was harvested every 2 days to monitor HIV replication. Virion production in the supernatant was quantified by HIV-1 p24 ELISA. The results of two independent experiments are shown as mean values ± SD.

Fig 4: MATR3 depletion inhibits HIV-1 replication. (A) Long-term depletion of MATR3 in Jurkat cells. Western blot analysis of MATR3 after transduction at different time points. β-Actin is the protein loading control. (B) Impaired HIV-1 replication in MATR3-depleted Jurkat cells. Jurkat cells depleted for MATR3 as described were infected with replication-competent HIV-1 pNL-4.3 virus, and supernatant was harvested after every 2 days to monitor HIV replication. Virion production in the supernatant was quantified by HIV-1 p24 ELISA (shCTRL cells are shown only until peak p24, then the signal decreased due to cytotoxic effects of infection). The results of two independent experiments are shown as mean values ± SD. (C) MATR3 knockdown in peripheral blood lymphocytes. Primary human peripheral blood lymphocytes (PBLs) were transduced with shMATR3. Four days posttransduction, MATR3-depleted cells were harvested for immunoblotting against MATR3 or Gag as indicated. β-Actin is the protein loading control. (D) MATR3 depletion inhibits HIV-1 replication in primary human peripheral blood lymphocytes. MATR3-depleted PBLs, generated as described above, were infected with replication-competent HIV-1 pNL-4.3 virus. The supernatant was harvested every 2 days to monitor HIV replication. Virion production in the supernatant was quantified by HIV-1 p24 ELISA. The results of two independent experiments from different donors are shown as mean values ± SD.

Fig 5: Rescue of MATR3 from J-lat-depleted cells shows dependence for HIV-1 release. (A) Design of a mutant MATR3 (f-MATR3R) resistant to shMATR3 depletion. The target sequence of shMATR3_905 was mutated in 3 positions without changing the amino acid composition of the protein. Inserted mutations were T552C (F147), A555G (R148), and A558G (R149). (B) Depleted endogenous MATR3 is rescued by ectopic f-MATR3R. J-lat 6.3 cells were transduced with shMATR3_905. Three days after puromycin selection (1 µg/ml), MATR3-depleted cells were transduced either with f-MATR3 or with the mutant f-MATR3R, which is the only one able to rescue MATR3 protein levels. Following incubation for 3 days in blasticidin (10 µg/ml), cells were processed for immune staining. (C) MATR3 is required for SAHA reactivation from latency in the J-lat model. J-lat 6.3 cells were left untreated or transduced as above (B) and then treated with SAHA at the concentrations indicated for 24 h. Extracellular p24 in the supernatant was quantified by HIV-1 p24 ELISA. Results from three independent experiments are shown as mean values ± SD. (D) Depleted endogenous MATR3 is rescued by ectopic f-MATR3R. J-lat 8.4 cells were transduced with shMATR3_905. Three days after puromycin selection, MATR3-depleted cells were transduced with f-MATR3R. Following incubation for 3 days, cells were processed for immune staining. (E) MATR3 is required for SAHA reactivation from latency in the J-lat model. J-lat 8.4 cells were transduced as above (D) and then treated with SAHA for 24 h. Extracellular p24 in the supernatant was quantified by HIV-1 p24 ELISA. Average results from two independent experiments are shown.