Phosphorylation of ETS1 by Src family kinases prevents its recognition by the COP1 tumor suppressor

Abstract

Oncoproteins and tumor suppressors antagonistically converge on critical nodes governing neoplastic growth, invasion, and metastasis. We discovered that phosphorylation of the ETS1 and ETS2 transcriptional oncoproteins at specific serine or threonine residues creates binding sites for the COP1 tumor suppressor protein, which is an ubiquitin ligase component, leading to their destruction. In the case of ETS1, however, phosphorylation of a neighboring tyrosine residue by Src family kinases disrupts COP1 binding, thereby stabilizing ETS1. Src-dependent accumulation of ETS1 in breast cancer cells promotes anchorage-independent growth in vitro and tumor growth in vivo. These findings expand the list of potential COP1 substrates to include proteins whose COP1-binding sites are subject to regulatory phosphorylation and provide insights into transformation by Src family kinases.

Figure 1. ETS2 is a COP1 Substrate

(A) Silver stained gel of proteins recovered after sequential immunoprecipitation (anti-FLAG followed by anti-HA) of cell extracts prepared from HCT116 colorectal cancer cells (parental or producing FLAG-HA-COP1). Proteins captured with the anti-FLAG antibody were eluted with Flag peptide (FLAG eluate) and applied to an anti-HA column, followed by elution with HA peptide (HA eluate). Shown are proteins present in the FLAG eluate, anti-HA flow-through and HA eluate. (B) Proteins identified in immunopurified COP1 complex by mass spectrometry. (C) Immunoblot analysis of proteins recovered after initial anti-FLAG immunoprecipitation and peptide elution as in (A). WCL, whole cell lysate; *, IgG; S.E., short exposure; L.E., long exposure. (D) Immunoblot analysis of 293FT cells transiently transfected with plasmids encoding the indicated 3xHA-tagged ETS2 variants and, where indicated, DET1 and COP1 (wild-type or dominant-negative AE7). A plasmid encoding GFP was included as a transfection efficiency control. (E) Immunoblot analysis of 293FT cells expressing a DOX-inducible COP1 shRNA and the indicated HA-tagged ETS2 variants. Cells were treated with or without DOX (1 μg/mL) for 48 hours. (F) In vitro ubiquitylation of 35S-ETS2 (wild-type or VPAA1&2) in the presence of recombinant E1 and E2 (UbcH5a) and immunopurified COP1. (G) Anti-HA immunoblot analysis of ETS2-V5 (wild-type or VPAA1&2) that was immunoprecipitated under denaturing conditions from 293FT cells that were transiently transfected with plasmids encoding the indicated proteins. See also Figure S1.

Figure 2. Differential Regulation of ETS1 and ETS2 by COP1

(A) Schematics of ETS1 and ETS2. (B) Alignments of COP1 degrons in ETS1 and ETS2 (C-terminal site) across species. (C) In vitro ubiquitylation of ³⁵S-ETS1 (wild-type or VPAA) in the presence of recombinant E1 and E2 (UbcH5a) and immunopurified COP1. (D) Immunoblot analysis of anti-HA immunoprecipitates from 293FT cells producing the indicated HA-tagged ETS1 and ETS2 variants. Cells were treated with 10 μM MG132 6 hours before harvest. (E) Immunoblot analysis of 293FT cells expressing a DOX-inducible COP1 shRNA and the indicated 3xHA-tagged ETS1 variants. Cells were treated with or without DOX (1 μg/mL) for 48 hours.

Figure 3. Phosphorylation of ETS1 Governs its Recognition by COP1

(A) Immunoblot analysis of COP1 recovered from 293FT cell extracts after incubation with NeutraAvidin beads loaded with the indicated peptides. (B) Immunoblot analysis of anti-HA immunoprecipitates from 293FT cells producing the indicated 3xHA-tagged ETS1 variants. Cells were treated with 10 μM MG132 6 hours before harvest. (C) Immunoblot analysis of 293FT cells expressing the indicated 3xHA-ETS1 variants after the addition of 100 μg/mL cyclohexamide (CHX). Different exposure times were used to normalize the 0 minute band intensities for the different ETS1 variants. (D) Immunoblot analysis of 293FT cells expressing a DOX-inducible COP1 shRNA and the indicated 3xHA-tagged ETS1 variants. Cells were treated with or without DOX (1 μg/mL) for 48 hours. See also Figure S2.

Figure 4. Dasatinib Downregulates ETS1

(A and B) Immunoblot analysis of anti-ETS1 immunoprecipitates or whole cell extracts of MDA-MB-231 cells expressing a DOX-inducible COP1 shRNA treated with increased amounts of dasatinib for 8 hours with or without pretreatment of MG132 (10 μM) for 2 hours (A) or Dox (1μg/mL) for 48 hours (B). (C) ³⁵S Pulse-chase analysis of ETS1 recovered from MDA-MB-231 cells pretreated with dasatinib (10 μM) for 3 hours and/or ionomycin (2 μM) for 0.5 hours. MDA-MB-231 cells treated with DMSO were used as a control (CNTL). (D) Immunoblot analysis of anti-ETS1 immunoprecipitates or whole cell extracts of MDA-MB-231 cells at the indicated timepoints after addition of dasatinib (10 μM). Where indicated cells were pretreated with MG132 (10 μM) for 2 hours. (E) Immunoblot analysis of anti-HA immunoprecipitates or whole cell extracts of MDA-MB-231 cells producing the indicated 3xHA-tagged ETS1 variants pretreated with MG132 (10 μM) for 2 hours and, where indicated, with dasatinib (10 μM), for an additional 8 hours. (F) Immunoblot analysis of MDA-MB-231 cells expressing a DOX-inducible COP1 shRNA and the indicated HA-tagged ETS1 variants. Where indicated cells were pretreated with Dox (1μg/ml) for 48 hours and dasatinib (10 μM) for an additional 8 hours. Dotted line indicates where irrelevant lanes were removed from the blot image.. See also Figure S3.

Figure 5. Src and Yes Redundantly Control ETS1 Stability

(A and B) Immunoblot analysis of MDA-MB-231 cells expressing the indicated shRNAs. (C and D) Immunoblot analysis of MDA-MB-231 cells expressing the indicated Src (C) or Yes (D) variants and treated, where indicated, with dasatinib. (E and F) Immunoblot of the indicated breast cancer cell lines. BaA = Basal A subtype. BaB = Basal B subtype. Dasatinib (1μM) was added for 8 hours where indicated in (F). S.E., short exposure. L.E., long exposure. See also Figure S4.

Figure 6. Role of Src and ETS1 in Mammary Transformation

(A and B) Immunoblot analysis (A) and soft agar assay (B) of MCF10A cells expressing a v-Src-ER fusion protein and the indicated DOX-inducible shRNAs. Cells were grown in the presence of DOX (1 μg/mL). 4-OH-tamoxifen (TAM, 1μg/ml) was also added where indicated. Parental MCF10A were included in (A) as a control. Scale bars, 0.5 mm. (C, D and E) Immunoblot analysis (C and D) and soft agar assay (E) of MCF10A-v-Src-ER cells stably expressing control shRNA (shControl) or shRNA against ETS1 (shETS1-3) and indicated shRNA-resistant (res) ETS1 cDNAs. Cells were treated with vehicle or 4-OH-tamoxifen (1μg/ml) for 12 hours where indicated. Scale bars, 0.5 mm. Representative images shown in (B) and (E) are from three biological replicates. See also Figure S5.

Figure 7. ETS1 Promotes Tumor Growth in Human Triple Negative Breast Cancer

(A and B) Immunoblot analysis (A) and soft agar assay (B) of MDA-MB-231 cells stably expressing shRNAs against GFP or ETS1 (shRNA 2 or 3) and, where indicated, an mRNA encoding venus fluorescent protein (Venus) or an shRNA-resistant (for shRNA 3) mRNA encoding ETS1 (ETS1res). Representative images in (B) are from three biological replicates. (C, D) Representative bioluminescent images of mice orthotopically injected with MDA-MB-231 cells stably expressing firefly luciferase cDNA and the indicated shRNAs (C) and quantification of shEST1-3/shGFP signal ratios at the indicated time (D, n=5). Error bar represents SEM. (E and F) Representative appearance of tumors (E) and mean tumor mass (F) at necropsy. Error bar indicates SEM. (G and H) Representative images of breast cancer tumors with high (G) or low (H) levels of ETS1 and phospho-Tyr416 Src detected by immunohistochemistry. Red solid box in (a) and (b) are magnified in (c) and (d), respectively. Scale bars in (a) and (b) represent 100 μM; Scale bars in (c) and (d) represent 20 μM. (I) The classification of tumor cores based upon the level of ETS1 and phospho-Tyr416 Src. See also Figure S6 and Table S1.