p53-Independent regulation of p21Waf1/Cip1 expression and senescence by PRMT6

Abstract

p21 is a potent cyclin-dependent kinase inhibitor that plays a role in promoting G1 cell cycle arrest and cellular senescence. Consistent with this role, p21 is a downstream target of several tumour suppressors and oncogenes, and it is downregulated in the majority of tumours, including breast cancer. Here, we report that protein arginine methyltransferase 6 (PRMT6), a type I PRMT known to act as a transcriptional cofactor, directly represses the p21 promoter. PRMT6 knock-down (KD) results in a p21 derepression in breast cancer cells, which is p53-independent, and leads to cell cycle arrest, cellular senescence and reduced growth in soft agar assays and in severe combined immunodeficiency (SCID) mice for all the cancer lines examined. We finally show that bypassing the p21-mediated arrest rescues PRMT6 KD cells from senescence, and it restores their ability to grow on soft agar. We conclude that PRMT6 acts as an oncogene in breast cancer cells, promoting growth and preventing senescence, making it an attractive target for cancer therapy.

Figure 1.

PRMT6 is overexpressed in breast tumours. (A) PRMT6 is overexpressed in tumours as compared with the normal tissue based on the analysis of 37 matched control cases. On the right, representative stainings from the TMA are shown as follows: normal breast epithelium (left) and breast carcinoma (right). (B) PRMT6 overexpression is positively correlated (P = 0.019) with tumour stage. Analysis is based on 166 breast cancer samples. (C) PRMT6 mRNA expression in a panel of breast cancer cell lines. (D) PRMT6 protein levels in a panel of breast cancer cell lines.

Figure 2.

Loss of PRMT6 induces cell cycle arrest and inhibition of tumour growth in vitro and in vivo. (A) Two independent shRNA constructs against PRMT6 (sh-1 and sh-2) efficiently KD the protein levels along with the associated H3R2me2a arginine methylation mark as compared with the control shRNA (sh-c). The experiment is performed in MCF7 (left panel) and MDA-MB-231 (right panel) cells. (B) Both shRNA reduce proliferation as compared with the control in both the cells lines. (C) PRMT6 KD induces morphological changes typical of senescent cells, appearing flat and (D) staining for SABG. (E) Orthotopic injections of MCF7 (left panel) and MDA-MB-231 (right panel) cells depleted of PRMT6 (sh-2, black line) in SCID mice mammary fat pad completely inhibits tumour formation as compared with the mice injected with control cells (sh-c, grey line). Number of females injected for each condition are indicated on the side of each graph.

Figure 3.

PRMT6 directly represses the p21 promoter. qPCR analysis of cyclin inhibitors (A) p16 and (B) p21 in PRMT6 depleted (sh-1 and sh-2) or control (sh-c) MCF7 and MDA-MB-231 cells. (C) Immunoblot analysis of p21 protein levels in PRMT6 depleted (sh-1 and sh-2) or control (sh-c) MCF7 and MDA-MB-231 cells. (D) Chromatin immunoprecipitation in MCF7 cells using the antibody indicated on each panel. Data are shown as the % of input DNA for PRMT6 and as % of input DNA normalized to total H3 (% input/H3) for the histone PTMs. The p21 promoter in MCF7 cells is enriched for inactive- (H3R2me2a and H3K27me3) and depleted for active- (H3K4me3 and H3acetyl) chromatin marks in control cells as compared with PRMT6 depleted ones (sh-2, grey lines). Experiment was repeated four times, and a representative plot is shown.

Figure 4.

Bypassing p21 expression rescues PRMT6 depletion phenotype. (A) p21 KD (sh-p21) rescues the morphological changes (flat cell phenotype) induced by PRMT6 (sh-2) KD alone in MCF7 (left panel) and MDA-MB-231 (right panel). (B) Senescent cells are scored as assessed by SABG staining. p21 KD (sh-p21) is able to rescue PRMT6 KD induced senescence in both the cell lines. Error bars represent the variations in the number of cells staining positive for SABG in three independent experiments. P-values of the statistical significance are indicated.

Figure 5.

Depletion of p21 rescues the anchorage independent growth of PRMT6 depleted cells on soft agar. (A) Soft agar assay in MCF7 (left panel) and MDA-MB-231 (right panel), using PRMT6 KD (sh-2), p21 KD (sh-p21) or double PRMT6 and p21 KD (sh-p21/sh-2). (B) Quantification of the number of colonies in different conditions, error bars represents the variation in three independent experiments, P-values of the statistical significance are indicated.