Mammalian SWI/SNF complexes in cancer: emerging therapeutic opportunities

Abstract

Mammalian SWI/SNF (BAF) chromatin remodeling complexes orchestrate a diverse set of chromatin alterations which impact transcriptional output. Recent whole-exome sequencing efforts have revealed that the genes encoding subunits of mSWI/SNF complexes are mutated in over 20% of cancers, spanning a wide range of tissue types. The majority of mutations result in loss of subunit protein expression, implicating mSWI/SNF subunits as tumor suppressors. mSWI/SNF-deficient cancers remain a therapeutic challenge, owing to a lack of potent and selective agents which target complexes or unique pathway dependencies generated by mSWI/SNF subunit perturbations. Here, we review the current landscape of mechanistic insights and emerging therapeutic opportunities for human malignancies driven by mSWI/SNF complex perturbation.

Figure 1. mSWI/SNF chromatin remodeling complexes in human cancer

Mammalian SWI/SNF complexes are separated into two separate forms: BRG1/BRM associated factor (BAF or SWI/SNF-A) and polybromo-associated BAF (PBAF or SWI/SNF-B). BAF and PBAF share numerous subunits, including both ATPases BRG1 and BRM (depicted in red). BAF and PBAF differ from one another by incorporation of key peripheral subunits (depicted in green). Mutations in the genes encoding mSWI/SNF complex subunits are present in over 20% of human cancers, with specific subunits mutated in specific malignancies, as shown [–6].

Figure 2. Genetic perturbation to genes encoding mSWI/SNF subunits gives rise to cancer-specific vulnerabilities

(A) Loss of BAF or PBAF subunits results in subunit(s)-deficient complexes resulting in transcriptional deregulation across distinct oncogenic pathways. Inhibition of resulting synthetic lethal dependencies represents a viable therapeutic avenue, using existing agents where available. (B) Rank gene list depicting specific vulnerabilities in ARID1A-deficient cells from an RNAi screen from [30]. ARID1B (black dot) is the top vulnerability in ARID1A-deficient cells, followed by PIK3 pathway genes (green dots). EZH2 (red dot) does not score as a top dependency in the dataset reported in [30]. Genes are ranked by RNMI (Ranked Normalized Mutual Information) score.

Figure 3. mSWI/SNF subunits contain several bromodomains with existing, potent small molecule probes

(A) Schematic of five mSWI/SNF subunits (PBRM1, SMARCA2, SMARCA4, BRD7, and BRD9) depicting the prevalence of bromodomains across proteins of the chromatin remodeling complex. In green, are bromodomains inhibited by PFI-3 [92] and Compound 17 [91]; in blue are bromodomains inhibited by LP99 [106], I-BRD9 [107], Compound 28 [108], BI-7273(1) and BI-9564(2) [109]. (B) Dissociation constants or Kd (nM) of each small molecule to their respective targets are tabulated. Values with (*) were calculated based on BROMOscan inhibitor binding platform. Remaining values were determined by isothermal titration calorimetry (ITC). (C) Chemical structures of bromodomains inhibitors targeting PBRM1, SMARCA2, SMARCA4, BRD7 and BRD9.

Figure 4. The SS18-SSX fusion protein hallmark to synovial sarcoma renders mSWI/SNF complexes oncogenic

Synovial sarcoma is driven by the t(X;18) chromosomal translocation which results in the fusion of the first 379 amino acids (aa) of SS18 to the last 78 aa of SSX1, 2, or 4. The ensuing SS18-SSX fusion protein dominantly replaces wild-type SS18 in BAF complexes and results in the inability of BAF47 (SMARCB1 or hSNF5 or INI1) to properly assemble. The oncogenic residual complex is targeted to new genomic loci leading to oncogenic gene expression and proliferation.