Structural and Functional Insights Into Mammalian SWI/SNF Complex Misassembly in Oncogenesis

Abstract

Mammalian SWI/SNF (mSWI/SNF) complexes are approximately 1-1.5-MDa ATP-dependent chromatin remodelers responsible for nucleosome repositioning, hierarchical chromatin organization, and recruitment of ancillary transcriptional machinery to promoters and enhancers. The mSWI/SNF complex family is comprised of three types of chromatin remodeling complexes that are compositionally and functionally distinct including the BAF complex, PBAF complex, and the recently discovered ncBAF complex. The BAF and PBAF complexes are composed of up to fifteen subunits, many of which have spatially and temporally restricted paralogues that are thought to function in targeting specificity through combinatorial assembly mechanisms. The ncBAF complex is a minimized complex containing several core components found in both BAF and PBAF complexes, suggesting a similar chromatin remodeling functionality. Nevertheless, functional assays have yet to be conducted. Strikingly, mutations in genes encoding for mSWI/SNF subunits are found in greater than 20% of cancers. Given the compositional complexity of each of the mSWI/SNF complexes, many of these mutations have been postulated to perturb their assembly pathway, producing residual complexes with aberrant biological functions. The oncogenic potential of BAF and PBAF complex misassembly is perhaps most apparent in SMARCB1 deficient tumor types, including malignant rhabdoid tumor (MRT), atypical teratoid/rhabdoid tumor (AT/RT), and a subset of epithelioid sarcomas (EpS). Herein, I demonstrate that the bi-allelic inactivation of SMARCB1 in these tumor types results in destabilization of both BAF and PBAF complexes, and the formation of a novel residual PBAF-like complex through alteration of PBRM1 stoichiometry. Although my collaborators have noted that elevated expression of PBAF specific subunits in AT/RT patients is correlated with poor prognosis, biochemical efforts taken to identify residual PBAF-like complexes as a lethal target in this disease have yielded conflicting results. However, during these studies I unexpectedly determined that ncBAF, the only mSWI/SNF complex that does not integrate SMARCB1 in normal conditions, to have an important maintenance function in these tumor types. Nevertheless, I show that PBAF complexes play a role in the up-regulation of genes involved in mTOR signaling, and also in the down-regulation of receptor tyrosine kinase degraders during oncogenic transformation mediated by SMARCB1 loss. Throughout these studies, I have also demonstrated that PBAF subunit PBRM1 likely exists in a homodimeric state in wild type PBAF complexes. I have also identified BRD7 as an anchor for PBRM1 binding to the PBAF complex core, an event that could potentially give insight into disease mechanisms that include bi-allelic loss of BRD7 as well as PBRM1, which includes a wide spectrum of human cancers.