Prostate cancer-associated SPOP mutations confer resistance to BET inhibitors through stabilization of BRD4

Abstract

The bromodomain and extraterminal (BET) family of proteins comprises four members-BRD2, BRD3, BRD4 and the testis-specific isoform BRDT-that largely function as transcriptional coactivators and play critical roles in various cellular processes, including the cell cycle, apoptosis, migration and invasion. BET proteins enhance the oncogenic functions of major cancer drivers by elevating the expression of these drivers, such as c-Myc in leukemia, or by promoting the transcriptional activities of oncogenic factors, such as AR and ERG in prostate cancer. Pathologically, BET proteins are frequently overexpressed and are clinically linked to various types of human cancer; they are therefore being pursued as attractive therapeutic targets for selective inhibition in patients with cancer. To this end, a number of bromodomain inhibitors, including JQ1 and I-BET, have been developed and have shown promising outcomes in early clinical trials. Although resistance to BET inhibitors has been documented in preclinical models, the molecular mechanisms underlying acquired resistance are largely unknown. Here we report that cullin-3^SPOP earmarks BET proteins, including BRD2, BRD3 and BRD4, for ubiquitination-mediated degradation. Pathologically, prostate cancer-associated SPOP mutants fail to interact with and promote the degradation of BET proteins, leading to their elevated abundance in SPOP-mutant prostate cancer. As a result, prostate cancer cell lines and organoids derived from individuals harboring SPOP mutations are more resistant to BET-inhibitor-induced cell growth arrest and apoptosis. Therefore, our results elucidate the tumor-suppressor role of SPOP in prostate cancer in which it acts as a negative regulator of BET protein stability and also provide a molecular mechanism for resistance to BET inhibitors in individuals with prostate cancer bearing SPOP mutations.

Figure 1. The Cullin 3^SPOP E3 ubiquitin ligase negatively regulates the stability of BET proteins

a. Immunoblot (IB) analysis of whole cell lysates (WCL) derived from 22Rv1 cells. Where indicated, MG132 or MLN4924 was added for 10 hours before harvesting the cells. b. IB analysis of WCL and immunoprecipitates (IP) derived from 293 cells transfected with Flag-BRD4 and various Myc-tagged Cullin constructs. 30 hours post-transfection, cells were treated with 10 μM MG132 for 10 hours before harvesting. c. IB analysis of WCL derived from 22Rv1 cells infected with the indicated lentiviral shRNAs. Infected cells were selected with 1 μg/ml puromycin for 72 hours to eliminate non-infected cells before harvesting. d. IB analysis of WCL and IP derived from 293 cells transfected with HA-BRD4 and Flag-tagged BTB domain-containing protein constructs. 30 hours post-transfection, cells were treated with 10 μM MG132 for 10 hours before harvesting. EV, empty vector. e. IB analysis of WCL derived from 22Rv1 cells transfected with increasing doses (0.5–3 μg) of Flag-SPOP. f. IB analysis of WCL derived from C4-2 cells with SPOP knockout by the CRISPR technology. Parental C4-2 cells are used as the control. g. IB analysis of WCL derived from 22Rv1 cells infected with the indicated lentiviral shRNAs. Infected cells were selected with 1 μg/ml puromycin for 72 hours to eliminate non-infected cells before harvesting. h. IB analysis of WCL derived from SPOP^+/+ and SPOP^−/− knockout mouse embryonic fibroblasts (MEFs). i. IB analysis of WCL and IP derived from HeLa cells transfected with plasmids expressing the indicated proteins. 30 hours after transfection, cells were treated with the proteasome inhibitor MG132 (20 μM) for 6 hours before collecting. j. SPOP knockout cells (sgSPOP) as well as parental C4-2 cells (Con) were treated with 100 μg/ml cycloheximide (CHX) for the indicated time period before harvesting. Equal amounts of WCL were immunoblotted with the indicated antibodies. k. Quantification of the band intensities in (j). BRD4 bands were normalized to vinculin, then normalized to the t = 0 time point. l. The growth curve of 22Rv1 cell lines with depletion of SPOP and/or BRD4. shScr, Scramble. *p < 0.05, t-test. m. Colony formation assays of 22Rv1 cell lines with depletion of SPOP and/or BRD4. *p < 0.05, t-test. n. Soft agar assays of C4-2 cell lines with depletion of SPOP and/or BRD4. shScr, Scramble. *p < 0.05; **p < 0.01, t-test. o. IB analysis of WCL derived from 22Rv1 cells infected with the indicated lentiviral shRNAs. shScr, Scramble. Infected cells were selected with 1 μg/ml puromycin for 72 hours to eliminate non-infected cells before harvesting. p. Representative images of migrated 22Rv1 cells infected with indicated lentiviral shRNAs. shScr, Scramble. q. Quantification of migrated cells in (p). Data was shown as mean ± SD for three independent experiments. **p < 0.01, t-test.

Figure 2. Prostate cancer-associated SPOP mutants promote prostate tumorigenesis largely by elevating protein levels of BET proteins

a. A schematic illustration of SPOP domains and prostate cancer-associated mutations. b. Immunoblot (IB) analysis of whole cell lysates (WCL) and immunoprecipitates (IP) derived from 293 cells transfected with Flag-BRD4 and HA-SPOP-WT or deletion of MATH domain-SPOP constructs. 30 hours post-transfection, cells were treated with 10 μM MG132 for 10 hours before harvesting. EV, empty vector. c. IB analysis of WCL derived from 293 cells transfected with indicated plasmids. EV, empty vector. d. IB analysis of WCL and IP derived from 293 cells transfected with Flag-SPOP-WT or prostate cancer (PrCa)-associated SPOP mutants. Cells were treated with 10 μM MG132 for 10 hours before harvesting. EV, empty vector. e. IB analysis of WCL derived from LNCaP cells stably expressing Flag-SPOP-WT or PrCa-associated SPOP mutants. EV, empty vector. f. 293 cells transfected with Flag-BRD4 together with the indicated HA-SPOP expressing plasmids. 30 hours post-transfection, cells were treated with 100 μg/ml cycloheximide (CHX) for the indicated time period before harvesting. WCL were subjected to IB analysis. EV, empty vector. g. In vivo ubiquitination assays of WCL and IP derived from HeLa cells transfected with plasmids expressing indicated proteins. 30 hours post-transfection, cells were treated with the proteasome inhibitor MG132 (20 μM) for 6 hours before cell collection. h. Colony formation assays of C4-2 cell lines stably expressing SPOP-WT or PrCa-associated SPOP mutants. **p < 0.01, t-test. i. C4-2 cell lines stably expressing PrCa-associated SPOP mutants were infected with shScr (Scramble) or shBRD4 lentivirus. Infected cells were selected with 1 μg/ml puromycin for 72 hours to eliminate non-infected cells and used for colony formation assays. **p < 0.01, t-test. j–l. Parental or BRD4 knockout C4-2 cells stably expressing SPOP-F102C were injected into nude mice (n=10 for each group). The in vivo tumor growth was monitored for indicated time period (j). Tumors were dissected and weighed (k–l). **p <0.01, t-test. m. Representative images of primary prostate cancer patient samples stained for BRD4 by immunohistochemistry. Scale bar, 100 μm. n. Statistical analysis of BRD4 expression in primary prostate cancer patient samples harboring SPOP-WT or SPOP-mutations. p value, Chi-squared test.

Figure 3. SPOP promotes ubiquitination and subsequent destruction of BET proteins in a degron-dependent manner

a. A schematic illustration of BRD4 domains and truncation of BRD4 constructs used in this study. b. Immunoblot (IB) analysis of whole cell lysates (WCL) and GST pull down products. EV, empty vector. c. IB analysis of WCL derived from 293 cells transfected with indicated Xpress-BRD4 truncated constructs different doses of Flag-SPOP expressing construct. d. Sequence alignment of BET proteins with the SPOP binding motif (degron) in known SPOP substrates. e. IB analysis of WCL and immunoprecipitates (IP) derived from 22Rv1 prostate cancer cells transfected with HA-WT-BRD4, ΔD-BRD4 constructs. 30 hours post-transfection, cells were treated with 10 μM MG132 for 10 hours before harvesting. EV, empty vector. f. IB analysis of WCL derived from 293 cells transfected with indicated HA-BRD4 constructs with different doses of indicated Flag-SPOP expressing constructs. g. 293 cells were transfected with indicated HA-BRD4 and Flag-SPOP expressing plasmids. 30 hours post-transfection, cells were treated with 100 μg/ml cycloheximide (CHX) for the indicated time period before harvesting. WCL were subjected to IB analysis. h. Quantification of the band intensities in (g). BRD4 bands were normalized to Vinculin, then normalized to the t = 0 time point. i. In vivo ubiquitination assays of WCL and IP derived from HeLa cells transfected with plasmids expressing the indicated proteins. 30 hours after transfection, cells were treated with the proteasome inhibitor MG132 (20 μM) for 6 hours before cell collection. j. The Cul 3^SPOP E3 ligase complex promotes BRD4 ubiquitination in vitro. Affinity-purified SPOP complexes were incubated with recombinant GST-BRD4 proteins, purified E1, E2 and ubiquitin. The ubiquitination reaction products were subjected to IB analysis with the anti-BRD4 antibody. LE, Longer Exposure; SE, Shorter Exposure. k. IB analysis of WCL derived from 22Rv1 cells transfected with indicated constructs. EV, empty vector. l. The growth curve of cells ectopically expressing SPOP and/or BRD4. EV, empty vector. *p < 0.05, t-test. m–n. Representative images of migrated 22Rv1 cells transfected with indicated constructs in migration assay (m) and quantification of migrated cells (n). EV, empty vector. Data was shown as mean ± SD for three independent experiments. **p < 0.01, t-test; ns, non-significant.

Figure 4. BRD4 protein abundance largely determines the resistance of SPOP-deficient prostate cancer cells to BET inhibitors

a. The growth curve of C4-2 cells with CRISPR-mediated SPOP knockout (sgSPOP) or BRD4 knockout (sgBRD4) that were treated with JQ1 for the indicated time period. Parental C4-2 cells were used as control. **p < 0.01, t-test. b. Cell viability of C4-2 cells depleted for SPOP and/or BRD4 treated with indicated concentration of JQ1 for 6 days. shScr, Scramble. **p < 0.01, t-test. c. Cell viability of C4-2 cells stably expressing SPOP-WT or PrCa-derived SPOP mutants that were treated with indicated concentration of JQ1. EV, empty vector. **p < 0.01, t-test. d–f. 22Rv1 cells stably expressing SPOP-WT or SPOP-W131G mutant were injected into nude mice (n=10 for each group) and grown until tumors reached a size of approximately 100 mm³. Xenografted mice were randomized and then received (n=7 for each group) vehicle, 50 mg/kg JQ1, 5 days a week. In vivo tumor growth was monitored for indicated time period (d). Tumors were dissected and weighed (e, f). **p <0.01, t-test. ns, non-significant. g. The growth curves of parental as well as CRISPR-mediated BRD4 knockout (sgBRD4) of C4-2 cells stably expressing the SPOP-W131G mutant that were treated with JQ1 or vehicle. (10% cyclodextran in autoclaved water). **p < 0.01, t-test. h. Colony formation of parental as well as CRISPR-mediated BRD4 knockout (sgBRD4) of C4-2 cells with stably expressing SPOP mutants that were treated with or without JQ1. * p <0.05, **p < 0.01, t-test. i–j. Immunoblot (IB) analysis of C4-2 cells depleting SPOP and/or BRD4 (i) or ectopically expressing SPOP-WT or the SPOP-W131G mutant (j), treated with indicated doses of JQ1 for 24 hours before harvesting. shScr, Scramble; EV, empty vector. k. IB analysis of whole cell lysates (WCL) derived from human prostate organoid MSK-PCax. MSK-PCa15 is the SPOP-W131R mutant organoid and MSK-PCa2, MSK-PCa11 and MSK-PCa16 are SPOP-WT organoids. l–m. Cell viability of human prostate MSK-PCax organoids in 2-D (l) and 3-D (m) culture conditions that were treated with indicated concentration of JQ1 for 4 days. **p < 0.01, *** p < 0.001, t-test. n. IB analysis of WCL derived from human prostate MSK-PCa15 (SPOP-W131R) organoids infected with lentiviral shBRD4. Infected cells were selected with 0.5 μg/ml puromycin for 48 hours to eliminate non-infected cells before harvesting. shScr, Scramble. o. Cell viability of human prostate organoids MSK-PCa15 depleted for BRD4 in 2-D culture conditions treated with indicated concentrations of JQ1 for 72 hours. shScr, Scramble. **p < 0.01, t-test. q. A schematic illustration of the proposed mechanism through which SPOP mutations lead to BET inhibitor (BETi) resistance in the prostate cancer setting. TFs, transcription factors.