HSCARG, a novel regulator of H2A ubiquitination by downregulating PRC1 ubiquitin E3 ligase activity, is essential for cell proliferation

Abstract

Histone H2A ubiquitination plays critical roles in transcriptional repression and deoxyribonucleic acid (DNA) damage response. More attention has been focused on ubiquitin E3 ligases of H2A, however, less is known about the negative regulators of H2A ubiquitination. Here we identified HSCARG as a new negative regulatory protein for H2A ubiquitination and found a possible link between regulator of H2A ubiquitination and cell cycle. Mechanistically, HSCARG interacts with polycomb repressive complex 1 (PRC1) and deubiquitinase USP7 and inhibits PRC1 ubiquitination in a USP7-dependent manner. As ubiquitination of PRC1 is critical for its E3 ligase activity toward H2A, HSCARG and USP7 are further shown to decrease the level of H2A ubiquitination. Moreover, we demonstrated that HSCARG is involved in DNA damage response through affecting the level of H2A ubiquitination and localization of RAP80 at lesion points. Knockout of HSCARG results in persistent activation of checkpoint signaling and leads to cell cycle arrest. This study unravels a novel mechanism for the regulation of H2A ubiquitination and elucidates how regulators of H2A ubiquitination affect cell cycle.

Figure 1.

HSCARG interacts with RING1 and inhibits RING1 ubiquitination. (A) Endogenous interaction of HSCARG with RING1. HEK293T cells were harvested 48 h after seeding, and 5% of the cell lysates were analyzed directly (input). The remaining cell lysates were immunoprecipitated with anti-RING1 antibody or control rabbit IgG, followed by western blot with anti-RING1 and anti-HSCARG antibodies. (B) Co-localization of RING1 and HSCARG in HeLa cells. Subcellular localization of endogenously expressed RING1 and HSCARG was visualized by immunofluorescence microscopy using anti-RING1 antibody (red), anti-HSCARG antibody (green) and DAPI (blue). Scale bar, 10 μm. (C) and (D) HSCARG inhibits endogenous RING1 ubiquitination. HEK293T cells were transfected with indicated plasmids. At 48 h after transfection, cells were lysed with denaturing buffer and incubated with Ni-NTA beads, and the ubiquitinated proteins were purified and subjected to western blot with the indicated antibodies.

Figure 2.

HSCARG depends on USP7 in inhibiting RING1 ubiquitination. (A) HSCARG, RING1 and USP7 form a complex. HEK293T cells were transiently transfected with Flag-USP7, HA-RING1 and Myc-HSCARG. Forty-eight hour later, Co-IP analysis was performed with anti-Flag/HA/Myc antibodies or control mouse IgG, followed by western blot with anti-Flag, anti-HA and anti-Myc antibodies. (B)–(E) HSCARG depends on USP7 in inhibiting RING1 ubiquitination. The level of endogenous RING1 ubiquitination was monitored by His-ubiquitin pull-down analysis in HEK293T cells (B) or HSCARG^−/−, USP7^−/− and wild-type HCT116 cells transfected with indicated plasmids (C)–(E).

Figure 3.

HSCARG inhibits PRC1-mediated H2A ubiquitination. (A) HSCARG suppresses H2A ubiquitination. HEK293T cells were transfected with HA-H2A, His-ubiquitin and Flag-HSCARG as indicated. Forty-eight hour later, cells were harvested and lysed in denaturing buffer, and H2A conjugates were isolated by immunoprecipitation using the anti-HA antibody and protein G. The bound complexes were analyzed by western blot with anti-His and anti-HA antibodies. (B) HSCARG inhibits endogenous H2A ubiquitination. HEK293T cells were transfected with His-ubiquitin and Flag-HSCARG as indicated. Forty-eight hour later, His-ubiquitin pull-down analysis was performed using indicated antibody. (C) HSCARG reduces RING2-catalyzed H2A ubiquitination. HEK293T cells were transfected with HA-H2A, His-ubiquitin, Flag-HSCARG and Flag-RING2 as indicated. Forty-eight hour later, the levels of ubiquitinated H2A were monitored by denaturing immunoprecipitation analysis as previously described followed by western blot with indicated antibodies.

Figure 4.

HSCARG depends on USP7 in inhibiting H2A ubiquitination. (A) USP7 inhibits H2A ubiquitination. HEK293T cells were transfected with HA-H2A, His-ubiquitin and Flag-USP7 as indicated. Forty-eight hour later, denaturing immunoprecipitation was performed using the anti-HA antibody and protein G. The bound complexes were analyzed by western blot with anti-His and anti-HA antibodies. (B) USP7 inhibits endogenous H2A ubiquitination. HEK293T cells transfected with His-ubiquitin and Flag-USP7 were subjected to His-ubiquitin pull-down assay using indicated antibody. (C)–(F) HSCARG depends on USP7 in inhibiting H2A ubiquitination. The level of endogenous H2A ubiquitination was monitored by His-ubiquitin pull-down analysis in HEK293T cells (C) or HSCARG^−/−, USP7^−/− and wild-type HCT116 cells transfected with indicated plasmids (D)–(F).

Figure 5.

HSCARG affects DNA damage response. (A) HSCARG inhibits IR-induced H2A ubiquitination. HCT116 cells were treated with IR (10 Gy) and incubated for 1 h and then the level of H2A ubiquitination was detected by His-ubiquitin pull-down analysis. (B) HSCARG affects subcellular location of endogenous RAP80 in response to DNA damage. HeLa cells were transfected with or without Flag-HSCARG. For the USP7-depleted group, to get efficient USP7-knockdown cells, HeLa cells were transfected with USP7 shRNA and selected with puromycin before used; the knockdown effect was examined in Supplementary Figure S4A. At 24 h after transfection, cells were treated with IR (10 Gy), and 1 h later, cells were stained using anti-RAP80 antibody (red), anti-γ-H2AX antibody (green) and DAPI (blue). The average number of RAP80 foci per cell was quantified in at least 300 cells per sample. Error bars represent the SD of five independent experiments. P value was determined by Student's t-test. **indicates P < 0.01. Scale bar, 10 μm. (C) HSCARG deletion increases the persistence of H2A ubiquitination in response to IR. Both wild-type and HSCARG^−/− HCT116 cells were treated with or without IR (1 Gy), cells were harvested at indicated time and the endogenous level of H2A ubiquitination was detected and quantified. The levels of uH2A were normalized against the total amount of uH2A and H2A. (D) RAP80 was trapped in HSCARG^−/− HeLa cells. The subcellular location of endogenous RAP80 was detected in both wild-type and HSCARG^−/− HeLa cells with indicated treatments. The average number of RAP80 foci per cell was quantified in at least 300 cells per sample. Error bars represent the SD of five independent experiments. P value was determined by Student's t-test. **indicates P < 0.01. *indicates P < 0.05. Scale bar, 10 μm. (E) HSCARG^−/− cells are more sensitive to IR. Clonogenic survival assay was performed (see the Materials and Methods section). Data are from three independent experiments.

Figure 6.

Deletion of HSCARG results in persistent activation of checkpoint signaling and reduces cell proliferation. (A) The phosphorylation level of CHK2 is increased in HSCARG^−/− cells. Wild-type and HSCARG^−/− HCT116 cells were either treated with IR (1 Gy) or were not treated with IR, and the levels of CHK2 and p-CHK2 were detected at indicated time and quantified. The levels of p-CHK2 were normalized against the total amount of p-CHK2 and CHK2. (B) The mRNA levels of ddb2, sesn1 and p53R2 are increased in HSCARG^−/− cells. The total RNA populations of wild-type and HSCARG^−/− HCT116 cells were extracted, after which qRT-PCR was performed to detect the mRNA levels of ddb2, gadd45, sesn1 and p53Rs. Values were reported as mean ± SD; P value was determined by Student's t-test. *indicates P < 0.05, **P < 0.01 and ***P < 0.001. Each data are from three independent experiments. (C) The protein and mRNA levels of p21 are up-regulated in HSCARG^−/− cells. The protein and mRNA levels of p21 were examined in wild-type HCT116 cells, HSCARG^−/− HCT116 cells and HSCARG^−/− HCT116 cells transfected with Flag-HSCARG. The levels of p53 and p-p53 were also detected in these cells. (D) Deletion of HSCARG leads to a decreased cell growth rate, which can be partially rescued by USP7. Living cells were counted by trypan blue staining at different time points after initial seeding with 5 × 10³ cells. (E) FACS analysis showed that HSCARG^−/− cells were arrested in the G1 phase. The percentages of cells in different stage of cell cycle were detected by FACS analysis. Values were reported as mean ± SD; P value was determined by Student's t-test. **indicates P < 0.01.