Extended Data Fig. 7: PARP1 mutations promote USP15-mediated PARP1 stabilization.
(a) Bioinformatic analysis of basal-like breast cancer patients carrying mutations in USP15 and PARP1. (b) Analysis of the interactions between PARP1 and WT or mutant USP15. (c) Generation of MDA-MB-231 cell lines harboring endogenous mutations in PARP1 genes via CRISPR-Cas9 technology. (d) Co-IP analysis of the PARP1-USP15 interaction in the PARP1 PARP1+/+, PARP1+/E90K, PARP1+/S104R MDA-MB-231 cells. (e) Analysis of the K48-linked ubiquitination on PARP1 in PARP1+/+, PARP1+/E90K, PARP1+/S104R MDA-MB-231 cells. (f) Analysis of BER efficiency in PARP1 PARP1+/+, PARP1+/E90K, PARP1+/S104R MDA-MB-231 cells. n = 3 independent biological samples. (g) Analysis of genomic stability in PARP1+/+, PARP1+/E90K, PARP1+/S104R MDA-MB-231 cells with USP15 depleted in the absence or presence of 200 μM MMS using an alkaline comet assay. At least 50 cells per group were included for analysis. (h) Analysis of the enzymatic activity of PARP1 WT, E90K and S104R in vitro. (i) Analysis of DNA binding ability of PARP1 WT, E90K and S104R by EMSA assay. (j, k) Analysis of recruitment of WT, E90K and S104R PARP1 to DNA damage sites in U2OS cells following microirradiation. n = 8 for each group were quantified using Leica Las X. Representative images are shown in (k). (l-o) PARP1 E90K or S104R mutants significantly suppressed MMS-induced apoptosis in MDA-MB-231 cells. n = 3 independent biological samples. (p) Analysis of PARP1 trapping on chromatin in PARP1 WT and mutant cells. (q–s) Analysis of xenograft growth in vivo. The PARP1-KO HCC70 cells were infected with lentiviruses bearing vectors expressing PARP1 WT, E90K or S104R. These cell lines were further depleted USP15 with shRNA. Tumor volumes are shown in (r). Tumor weights at the experimental endpoint are shown in (S). A representative picture of the tumors is shown in (s). Data are expressed as mean ± s.e.m., n = 5 mice for each group. (t) Analysis of PARP1 and USP15 levels in different groups of HCC70 cells. (u–w) Analysis of xenograft weight. The PARP1-KO HCC70 cells were infected with lentiviruses bearing vectors expressing PARP1 WT, E90K or S104R. These cell lines were further chromosomally integrated with a control vector or a vector expressing USP15. Cells were engrafted into immunocompromised mice, and mice were treated with DMSO or olaparib. Data are expressed as mean ± s.e.m., n = 5 mice for each group. Data are expressed as mean ± s.d. (f, l, n). Data are expressed as mean ± s.e.m., each point represents a cell, the cells used for analysis in each experiment were from a single replicate (g, j). n.s., not significant. P values are indicated (f-g, j, l, n, r-s, u-w). Statistical significance was determined by Student’s t test (f, j, l, n), or one-way ANOVA test followed by Tukey’s multiple comparison (g, s, u-w), or two-way ANOVA test followed by Tukey’s multiple comparison (r). Experiments were repeated three times independently with similar results; data of one representative experiment are shown (b, d-f, h-i, l, n, p-q, t).
