BRD4 sustains melanoma proliferation and represents a new target for epigenetic therapy

Abstract

Metastatic melanoma remains a mostly incurable disease. Although newly approved targeted therapies are efficacious in a subset of patients, resistance and relapse rapidly ensue. Alternative therapeutic strategies to manipulate epigenetic regulators and disrupt the transcriptional program that maintains tumor cell identity are emerging. Bromodomain and extraterminal domain (BET) proteins are epigenome readers known to exert key roles at the interface between chromatin remodeling and transcriptional regulation. Here, we report that BRD4, a BET family member, is significantly upregulated in primary and metastatic melanoma tissues compared with melanocytes and nevi. Treatment with BET inhibitors impaired melanoma cell proliferation in vitro and tumor growth and metastatic behavior in vivo, effects that were mostly recapitulated by individual silencing of BRD4. RNA sequencing of BET inhibitor-treated cells followed by Gene Ontology analysis showed a striking impact on transcriptional programs controlling cell growth, proliferation, cell-cycle regulation, and differentiation. In particular, we found that, rapidly after BET displacement, key cell-cycle genes (SKP2, ERK1, and c-MYC) were downregulated concomitantly with the accumulation of cyclin-dependent kinase (CDK) inhibitors (p21 and p27), followed by cell-cycle arrest. Importantly, BET inhibitor efficacy was not influenced by BRAF or NRAS mutational status, opening the possibility of using these small-molecule compounds to treat patients for whom no effective targeted therapy exists. Collectively, our study reveals a critical role for BRD4 in melanoma tumor maintenance and renders it a legitimate and novel target for epigenetic therapy directed against the core transcriptional program of melanoma.

Conflict of interest statement

Disclosure of Potential Conflicts of Interest

M.-M. Zhou, G. Zhang, and M. Ohlmeyer have ownership interest (including patents) in patent on BrD ligands. No potential conflicts of interest were disclosed by the other authors.

©2013 AACR.

Figures

Figure 1

BRD2 and BRD4 are overexpressed in melanoma. A, heatmap representing the relative expression of BrD-containing proteins in melanocytes, primary melanoma cell lines and metastatic melanoma cell lines (GSE22301; ref. 25). B, microarray expression levels of BRD2 (top) and BRD4 (bottom) in nevi and melanoma human samples (GEOD3189; ref. 26). Line represents the median value. C and D, representative images of BRD4 staining of TMAs of human melanocytes, nevi, primary, and metastatic melanomas (C) followed by quantification of BRD4 signal (D). Bar, 100 μm. E and F, representative images of BRD2 staining of TMAs of human melanocytes, nevi, primary, and metastatic melanomas (E) followed by quantification of BRD2 signal (F). Bar, 100 μm. G, heatmap representing the BRD2 and BRD4 copy number in melanocytic and melanoma cell lines as extrapolated from a SNP array analysis (GSE22305; ref. 25). H, linear correlation of BRD2 (left) and BRD4 (right) mRNA expression levels with gene copy number calculated in G. n.s., not significant; a.u., arbitrary units. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 2

BRD or BET proteins inhibition attenuates proliferation in vitro. A and B, normalized proliferation curves of A375 (A) and SK-MEL-147 (B) treated with vehicle (DMSO), MS436 (10 μmol/L), or MS417 (10 μmol/L), measured by crystal violet staining. C and D, histograms representing the average percentage of A375 (C) or SK-MEL-147 (D) cells in G1, S, or G2–M phases after 72 hours of treatment with vehicle (DMSO), MS436 (10 μmol/L) or MS417 (10 μmol/L). E and F, macroscopic and microscopic images and quantification of colonies formed by A375 (E) or SK-MEL-147 (F) melanoma cell lines treated with vehicle (DMSO), MS436 (10 μmol/L), or MS417 (10 μmol/L). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 3

BET inhibition impairs melanoma tumor growth in vivo. A, average tumor volume of mice injected daily with either vehicle or MS417 (50 mg/kg; n = 10/treatment). B, macroscopic image of resected tumors at the conclusion of the experiment. Bar, 1 cm. C, average weight of resected tumors. D, percentage of mice bearing lung micrometastasis at the conclusion of the experiment. E, average number of micrometastasis per lung. F, representative microscopic H&E images of lungs. Metastatic focus is circled. Bar, 100 μm. *, P < 0.05; **, P < 0.01; ***, P < 0.001; n.s., not significant.

Figure 4

BRD4 knockdown is sufficient to recapitulate the antitumoral effects of BET inhibition in melanoma cells. A and B, mRNA levels of BET family genes in A375 (A) or SK-MEL-147 (B) melanoma cell lines treated with siRNA oligos (50 nmol/L). C and D, normalized proliferation curves of A375 (C) and SK-MEL-147 (D) treated with Mock, scr control, or siRNA oligos against BRD2, BRD3, and BRD4 (50 nmol/L), measured by crystal violet staining. E–H, relative BRD4 expression in A375 (E) or SK-MEL-147 (G) melanoma cells stably transduced with shRNA vectors against BRD4. Macroscopic images and quantification of clonal colonies formed by A375 (E and F) or SK-MEL-147 (G and H) melanoma cell lines transduced with shRNA vectors against BRD4. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5

BRD4 is essential for melanoma tumor maintenance in vivo. A, tumor volume of mice injected with either NSC- or shBRD4-transduced A375 cells (n = 6/group), measured for 18 days. B, macroscopic image of resected tumors at the conclusion of the experiment. Bar, 1 cm. C, average weight of resected tumors. D, representative microscopic images of tumor histologic sections stained for either BRD4 (top) or Ki67 (bottom) in NSC- or shBRD4 A375 tumors. E, percentage of mice containing ≤50 or >50 lung micrometastasis at the conclusion of the experiment. F, average number of lung metastasis per lung. NSC, nonsilencing control. Line in F represents the median value. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 6

BET inhibition impacts transcriptional programs that control cell proliferation and differentiation. A, top categories of mRNA species significantly enriched (P < 0.005 for all categories) in three melanoma cell lines (SK-MEL-5, SK-MEL-147, and A375) treated with MS436 (10 μmol/L) when compared with vehicle (DMSO)–treated cells. B, relative mRNA expression of candidate downstream effectors following vehicle/MS417 treatment (10 μmol/L) over a 24-hour time course. C, protein levels of candidate downstream effectors following vehicle/MS417 treatment (10 μmol/L) over a time course of 24 hours. D, protein levels p-ERK1/2 following vehicle/MS417 treatment (10 μmol/L) for 24, 48, and 72 hours. E, a correlation between BRD2 or BRD4 mRNA levels with either SKP2 (left) or ERK1 (right) mRNA levels (EGEOD7553; ref. 32). a.u., arbitrary units.

Figure 7

BRAF and NRAS mutation status does not influence sensitivity to BET inhibitors. A and B, sensitivity of primary melanocytes or melanoma cell lines to increasing concentrations of MS417 (A) or MS436 (B) determined by percentage of viability after 72 hours.