Project description:While oncogenes can potentially be inhibited with small molecules, the loss of tumor suppressors is more common and presents a conundrum for precision therapy because the proteins are no longer present. SMARCB1-mutant cancers epitomize this challenge because these highly lethal cancers are driven by inactivation of a single gene, a subunit of SWI/SNF chromatin remodeling complexes. To generate mechanistic insight into the consequences of SMARCB1 mutation and to seek vulnerabilities, we contributed 16 SMARCB1-mutant cell lines to a near-genomewide CRISPR screen as part of the Cancer Dependency Map1-3. Here we report that the little-studied gene DCAF5 (DDB1-CUL4 Associated Factor 5) is a specific dependency in SMARCB1-mutant cancers. We show that DCAF5 serves a quality control function for SWI/SNF complexes and in the absence of SMARCB1 DCAF5 causes degradation of incompletely assembled SWI/SNF complexes. Upon inhibition of DCAF5 SMARCB1-deficient SWI/SNF complexes re-accumulate, bind to target loci, and restore gene expression to levels sufficient to fully reverse the cancer state, including in a xenograft mouse model. Consequently, cancer results not from the loss of SMARCB1 function per se but rather from DCAF5-mediated degradation of SWI/SNF complexes. These data indicate that therapeutic targeting of DCAF5 may be sufficient to restore substantial SWI/SNF function and reverse cancer phenotypes caused by SMARCB1 loss.
Project description:While oncogenes can potentially be inhibited with small molecules, the loss of tumor suppressors is more common and presents a conundrum for precision therapy because the proteins are no longer present. SMARCB1-mutant cancers epitomize this challenge because these highly lethal cancers are driven by inactivation of a single gene, a subunit of SWI/SNF chromatin remodeling complexes. To generate mechanistic insight into the consequences of SMARCB1 mutation and to seek vulnerabilities, we contributed 16 SMARCB1-mutant cell lines to a near-genomewide CRISPR screen as part of the Cancer Dependency Map1-3. Here we report that the little-studied gene DCAF5 (DDB1-CUL4 Associated Factor 5) is a specific dependency in SMARCB1-mutant cancers. We show that DCAF5 serves a quality control function for SWI/SNF complexes and in the absence of SMARCB1 DCAF5 causes degradation of incompletely assembled SWI/SNF complexes. Upon inhibition of DCAF5 SMARCB1-deficient SWI/SNF complexes re-accumulate, bind to target loci, and restore gene expression to levels sufficient to fully reverse the cancer state, including in a xenograft mouse model. Consequently, cancer results not from the loss of SMARCB1 function per se but rather from DCAF5-mediated degradation of SWI/SNF complexes. These data indicate that therapeutic targeting of DCAF5 may be sufficient to restore substantial SWI/SNF function and reverse cancer phenotypes caused by SMARCB1 loss.
Project description:While oncogenes can potentially be inhibited with small molecules, the loss of tumor suppressors is more common and presents a conundrum for precision therapy because the proteins are no longer present. SMARCB1-mutant cancers epitomize this challenge because these highly lethal cancers are driven by inactivation of a single gene, a subunit of SWI/SNF chromatin remodeling complexes. To generate mechanistic insight into the consequences of SMARCB1 mutation and to seek vulnerabilities, we contributed 16 SMARCB1-mutant cell lines to a near-genomewide CRISPR screen as part of the Cancer Dependency Map1-3. Here we report that the little-studied gene DCAF5 (DDB1-CUL4 Associated Factor 5) is a specific dependency in SMARCB1-mutant cancers. We show that DCAF5 serves a quality control function for SWI/SNF complexes and in the absence of SMARCB1 DCAF5 causes degradation of incompletely assembled SWI/SNF complexes. Upon inhibition of DCAF5 SMARCB1-deficient SWI/SNF complexes re-accumulate, bind to target loci, and restore gene expression to levels sufficient to fully reverse the cancer state, including in a xenograft mouse model. Consequently, cancer results not from the loss of SMARCB1 function per se but rather from DCAF5-mediated degradation of SWI/SNF complexes. These data indicate that therapeutic targeting of DCAF5 may be sufficient to restore substantial SWI/SNF function and reverse cancer phenotypes caused by SMARCB1 loss.
Project description:Genes encoding subunits of SWI/SNF (BAF) chromatin remodeling complexes are mutated in >20% of cancers. SWI/SNF complexes exist in three distinct families that each contribute to regulation of transcription, although the functional interactions between the families are not well understood. Rhabdoid tumors constitute an informative model system as these highly aggressive cancers are driven by inactivation of a single SWI/SNF subunit, SMARCB1, which is present in two SWI/SNF families (cBAF and PBAF) but not in the third (GBAF/ncBAF). We and others have shown that BRD9, a therapeutically targetable member of ncBAF, is essential specifically in SMARCB1-deficient cancers, suggesting key functional relationships between SMARCB1-containing complexes and BRD9/ncBAF. However, the mechanistic underpinnings of these relationships are poorly understood. Here, we demonstrate that genomic binding of BRD9 is largely dependent upon SMARCB1 such that the absence of SMARCB1 results in significantly reduced BRD9 binding. At select sites, however, we show that SMARCB1-loss results in gain of BRD9 binding and BRD9-dependent accessibility. We find that this gain is associated with expression of genes promoting cell migration. Our results define relationships between SWI/SNF complex families, elucidate mechanisms by which SMARCB1 loss drives oncogenesis, and provide mechanistic insight into the synthetic-lethal relationship between SMARCB1 and BRD9.
Project description:Genes encoding subunits of SWI/SNF (BAF) chromatin remodeling complexes are mutated in >20% of cancers. SWI/SNF complexes exist in three distinct families that each contribute to regulation of transcription, although the functional interactions between the families are not well understood. Rhabdoid tumors constitute an informative model system as these highly aggressive cancers are driven by inactivation of a single SWI/SNF subunit, SMARCB1, which is present in two SWI/SNF families (cBAF and PBAF) but not in the third (GBAF/ncBAF). We and others have shown that BRD9, a therapeutically targetable member of ncBAF, is essential specifically in SMARCB1-deficient cancers, suggesting key functional relationships between SMARCB1-containing complexes and BRD9/ncBAF. However, the mechanistic underpinnings of these relationships are poorly understood. Here, we demonstrate that genomic binding of BRD9 is largely dependent upon SMARCB1 such that the absence of SMARCB1 results in significantly reduced BRD9 binding. At select sites, however, we show that SMARCB1-loss results in gain of BRD9 binding and BRD9-dependent accessibility. We find that this gain is associated with expression of genes promoting cell migration. Our results define relationships between SWI/SNF complex families, elucidate mechanisms by which SMARCB1 loss drives oncogenesis, and provide mechanistic insight into the synthetic-lethal relationship between SMARCB1 and BRD9.
Project description:Genes encoding subunits of SWI/SNF (BAF) chromatin remodeling complexes are mutated in >20% of cancers. SWI/SNF complexes exist in three distinct families that each contribute to regulation of transcription, although the functional interactions between the families are not well understood. Rhabdoid tumors constitute an informative model system as these highly aggressive cancers are driven by inactivation of a single SWI/SNF subunit, SMARCB1, which is present in two SWI/SNF families (cBAF and PBAF) but not in the third (GBAF/ncBAF). We and others have shown that BRD9, a therapeutically targetable member of ncBAF, is essential specifically in SMARCB1-deficient cancers, suggesting key functional relationships between SMARCB1-containing complexes and BRD9/ncBAF. However, the mechanistic underpinnings of these relationships are poorly understood. Here, we demonstrate that genomic binding of BRD9 is largely dependent upon SMARCB1 such that the absence of SMARCB1 results in significantly reduced BRD9 binding. At select sites, however, we show that SMARCB1-loss results in gain of BRD9 binding and BRD9-dependent accessibility. We find that this gain is associated with expression of genes promoting cell migration. Our results define relationships between SWI/SNF complex families, elucidate mechanisms by which SMARCB1 loss drives oncogenesis, and provide mechanistic insight into the synthetic-lethal relationship between SMARCB1 and BRD9.
Project description:Genes encoding subunits of SWI/SNF (BAF) chromatin remodeling complexes are mutated in >20% of cancers. SWI/SNF complexes exist in three distinct families that each contribute to regulation of transcription, although the functional interactions between the families are not well understood. Rhabdoid tumors constitute an informative model system as these highly aggressive cancers are driven by inactivation of a single SWI/SNF subunit, SMARCB1, which is present in two SWI/SNF families (cBAF and PBAF) but not in the third (GBAF/ncBAF). We and others have shown that BRD9, a therapeutically targetable member of ncBAF, is essential specifically in SMARCB1-deficient cancers, suggesting key functional relationships between SMARCB1-containing complexes and BRD9/ncBAF. However, the mechanistic underpinnings of these relationships are poorly understood. Here, we demonstrate that genomic binding of BRD9 is largely dependent upon SMARCB1 such that the absence of SMARCB1 results in significantly reduced BRD9 binding. At select sites, however, we show that SMARCB1-loss results in gain of BRD9 binding and BRD9-dependent accessibility. We find that this gain is associated with expression of genes promoting cell migration. Our results define relationships between SWI/SNF complex families, elucidate mechanisms by which SMARCB1 loss drives oncogenesis, and provide mechanistic insight into the synthetic-lethal relationship between SMARCB1 and BRD9.