CRISPR/Cas9 has revolutionized our ability to engineer genomes and conduct genome-wide screens in human cells. Whereas some cell types are amenable to genome engineering, genomes of human pluripotent stem cells (hPSCs) have been difficult to engineer, with reduced efficiencies relative to tumour cell lines or mouse embryonic stem cells. Here, using hPSC lines with stable integration of Cas9 or transient delivery of Cas9-ribonucleoproteins (RNPs), we achieved an average insertion or deletion (indel) efficiency greater than 80%. This high efficiency of indel generation revealed that double-strand breaks (DSBs) induced by Cas9 are toxic and kill most hPSCs. In previous studies, the toxicity of Cas9 in hPSCs was less apparent because of low transfection efficiency and subsequently low DSB induction. The toxic response to DSBs was P53/TP53-dependent, such that the efficiency of precise genome engineering in hPSCs with a wild-type P53 gene was severely reduced. Our results indicate that Cas9 toxicity creates an obstacle to the high-throughput use of CRISPR/Cas9 for genome engineering and screening in hPSCs. Moreover, as hPSCs can acquire P53 mutations, cell replacement therapies using CRISPR/Cas9-enginereed hPSCs should proceed with caution, and such engineered hPSCs should be monitored for P53 function.

Elucidation of the mutational landscape of human cancer has progressed rapidly and been accompanied by the development of therapeutics targeting mutant oncogenes. However, a comprehensive mapping of cancer dependencies has lagged behind and the discovery of therapeutic targets for counteracting tumor suppressor gene loss is needed. To identify vulnerabilities relevant to specific cancer subtypes, we conducted a large-scale RNAi screen in which viability effects of mRNA knockdown were assessed for 7,837 genes using an average of 20 shRNAs per gene in 398 cancer cell lines. We describe findings of this screen, outlining the classes of cancer dependency genes and their relationships to genetic, expression, and lineage features. In addition, we describe robust gene-interaction networks recapitulating both protein complexes and functional cooperation among complexes and pathways. This dataset along with a web portal is provided to the community to assist in the discovery and translation of new therapeutic approaches for cancer.

We show in this study that CRISPR-based screens have a significantly lower false-negative rate compared with RNAi-based screens, but have specific liabilities particularly in the interrogation of regions of genome amplification. Therefore, this study provides critical insights for applying CRISPR-based screens toward the systematic identification of new cancer targets. Cancer Discov; 6(8); 900-13. ©2016 AACR.See related commentary by Sheel and Xue, p. 824See related article by Aguirre et al., p. 914This article is highlighted in the In This Issue feature, p. 803.

Defects in epigenetic regulation play a fundamental role in the development of cancer, and epigenetic regulators have recently emerged as promising therapeutic candidates. We therefore set out to systematically interrogate epigenetic cancer dependencies by screening an epigenome-focused deep-coverage design shRNA (DECODER) library across 58 cancer cell lines. This screen identified BRM/SMARCA2, a DNA-dependent ATPase of the mammalian SWI/ SNF (mSWI/SNF) chromatin remodeling complex, as being essential for the growth of tumor cells that harbor loss of function mutations in BRG1/SMARCA4. Depletion of BRM in BRG1-deficient cancer cells leads to a cell cycle arrest, induction of senescence, and increased levels of global H3K9me3. We further demonstrate the selective dependency of BRG1-mutant tumors on BRM in vivo. Genetic alterations of the mSWI/SNF chromatin remodeling complexes are the most frequent among chromatin regulators in cancers, with BRG1/SMARCA4 mutations occurring in ∼10-15% of lung adenocarcinomas. Our findings position BRM as an attractive therapeutic target for BRG1 mutated cancers. Because BRG1 and BRM function as mutually exclusive catalytic subunits of the mSWI/SNF complex, we propose that such synthetic lethality may be explained by paralog insufficiency, in which loss of one family member unveils critical dependence on paralogous subunits. This concept of "cancerselective paralog dependency" may provide a more general strategy for targeting other tumor suppressor lesions/complexes with paralogous subunits.