Doxorubicin enhances Snail/LSD1-mediated PTEN suppression in a PARP1-dependent manner

Abstract

The transcription factor Snail not only functions as a master regulator of epithelial-mesenchymal transition (EMT), but also mediates cell proliferation and survival. While previous studies have showed that Snail protects tumor cells from apoptosis through transcriptional repression of PTEN, the specific mechanism remains unclear. In this study, we demonstrated that Snail cooperates with LSD1 to repress PTEN in a PARP1-dependent manner. Upon doxorubicin treatment, Snail becomes tightly associated with PARP1 through its pADPr-binding motif and is subject to poly(ADP-ribosyl)ation. This modification can enhance Snail-LSD1 interaction and promote the recruitment of LSD1 to PTEN promoter, where LSD1 removes methylation on histone H3 lysine 4 for transcription repression. Furthermore, treatment of tumor cells with PARP1 inhibitor AZD2281 can compromise doxorubicin-induced PTEN suppression and enhance the inhibitory effect of doxorubicin. Together, we proposed a tentative drug-resistant mechanism through which tumor cells defend themselves against DNA damage-induced apoptosis. PARP1 inhibitors in combination with DNA damaging reagents might represent a promising treatment strategy targeting tumors with over-activated Snail and LSD1.

Keywords: LSD1; PARP1; PTEN; Snail; poly(ADP-ribosyl)ation.

Figure 1. Doxorubicin enhances PARP1–Snail interaction. (A) The Snail complex was isolated from the stable HEK293 cells expressing Snail. The complexes were separated on SDS-PAGE and visualized by silver staining. LSD1 and PARP1 were identified by mass spectrometry. (B) Flag-tagged PARP1 and HA-tagged Snail were co-expressed in HEK293 cells. Cells were treated with 1 µM of Doxorubicin (DOX) for 6 h before harvesting. After immunoprecipitation of PARP1, bound Snail was examined by western blotting. (C) MDA-MB157 and HCT116 cells were treated with Doxorubicin as described above. After immunoprecipitation of PARP1 endogenous PARP1, bound endogenous Snail was examined by western blotting.

Figure 2. PARP1 positively regulates Snail–SD1 interaction. (A) Flag-tagged LSD1 and HA-tagged Snail were co-expressed in HEK293 cells. After immunoprecipitation of LSD1, bound Snail was examined by western blotting. For comparison, cells were either co-expressed with Flag-tagged PARP1 or treated with 1 µM of doxorubicin 6 h before harvesting cells. (B) Endogenous LSD1 was immunoprecipitated from MDA-MB157 and HCT116 cells and bound endogenous Snail was examined by western blotting. For comparison, cells were treated with doxorubicin (1 µM), AZD2281 (2 µM), or transfected with PARP1 siRNA.

Figure 3. Snail contains a potential pADPr-binding motif and is subject to poly(ADP-ribosyl)ation. (A) Sequence alignment of Snail with previously established pADPr-binding motif. The conserved residues were highlighted with red color. (B) Flag-tagged PARP1 was co-expressed with HA-tagged wild-type (WT) or mutant (R151A/K152A) Snail in HEK293 cells. Cell were treated with or without doxorubicin. After immunoprecipitation of PARP1, the bound Snail was examined. (C) Flag-tagged LSD1 was co-expressed with HA-tagged WT or mutant Snail. After immunoprecipitation of LSD1, the bound Snail was examined. (D) Flag-tagged LSD1 was co-expressed with HA-tagged Snail. Cell were treated with 10 µM of gallotannin (GN) for 6 h. After immunoprecipitation of PARP1, bound Snail was examined. (E) WT or mutant Snail was expressed in HEK293 cells and treated with 10 mg/ml of cycloheximide (CHX) for different time intervals. The level of Snail was analyzed by western blotting. Densitometry results were statistically analyzed and plotted (bottom panel, mean ± SD from 3 separate experiments). A representative blot is shown in the top panel. (F) HKE293 cells stably expressing Snail-HA were treated with doxorubicin and AZD2281. After Snail was immunoprecipitated, poly(ADP-ribosyl)ation of Snail was analyzed by western blotting using antibody against pADPr.

Figure 4. The enzymatic activity of PARP1 is required for Snail–LSD1 binding to PTEN promoter. (A) MDA-MB157 and HCT116 cells were treated with doxorubicin or AZD2281, or transfected with PARP1 siRNA. The association of endogenous Snail and LSD1 with the PTEN promoter was analyzed by ChIP assay. Methylation of H3K4 on the PTEN promoter was also analyzed by ChIP assay using antibody against H3K4me2. (B) The ChIP samples were analyzed by quantitative real-time PCR (mean ± SD from 3 separate experiments).

Figure 5. AZD2281 enhances the killing effect of doxorubicin on cancer cells. (A) MDA-MB157 and HCT116 cells were treated with AZD2281, doxorubicin, expression of PTEN, Akt, and phosphorylated Akt (Akt-P) was examined by western blotting. (B) The proliferation of cells in (A) was analyzed by MTT assays (mean ± SD from 3 separate experiments). (C) A proposed model illustrating that Snail recruits LSD1 to PTEN promoter in a PARP1-dependent manner. Under DNA damage condition, Snail becomes tightly associated with PARP1 and is subject to PARP1-mediated poly(ADP-ribosyl)ation, which promotes the interaction of Snail with LSD1, resulting in the recruitment of LSD1 at the PTEN promoter for transcription repression. Inhibition of PARP1 facilitates the growth-suppressive effect of doxorubicin by restoring PTEN expression.