Search for: All records

Transcriptional addiction is recognized as a valid therapeutic target in cancer, whereby the dependency of cancer cells on oncogenic transcriptional regulators may be pharmacologically exploited. However, a comprehensive understanding of the key factors within the transcriptional machinery that might afford a useful therapeutic window remains elusive. Herein, we present a cross-omics investigation into the functional specialization of the transcriptional cyclin dependent kinases (tCDKs) through analysis of high-content genetic dependency, gene expression, patient survival, and drug response datasets. This analysis revealed specialization among tCDKs in terms of contributions to cancer cell fitness, clinical prognosis, and interaction with oncogenic signaling pathways. CDK7 and CDK9 stand out as the most relevant targets, albeit through distinct mechanisms of oncogenicity and context-dependent contributions to cancer survival and drug sensitivity. Genetic ablation of CDK9, but not CDK7, mimics the effect on cell viability the loss of key components of the transcriptional machinery. Pathway analysis of genetic co-dependency and drug sensitivity data show CDK7 and CDK9 have distinct relationships with major oncogenic signatures, including MYC and E2F targets, oxidative phosphorylation, and the unfolded protein response. Altogether, these results inform the improved design of therapeutic strategies targeting tCDKs in cancer.

Cellular adaptation to hypoxia is a hallmark of cancer, but the relative contribution of hypoxia-inducible factors (HIFs) versus other oxygen sensors to tumorigenesis is unclear. We employ a multi-omics pipeline including measurements of nascent RNA to characterize transcriptional changes upon acute hypoxia. We identify an immediate early transcriptional response that is strongly dependent on HIF1A and the kinase activity of its cofactor CDK8, includes indirect repression of MYC targets, and is highly conserved across cancer types. HIF1A drives this acute response via conserved high-occupancy enhancers. Genetic screen data indicates that, in normoxia, HIF1A displays strong cell-autonomous tumor suppressive effects through a gene module mediating mTOR inhibition. Conversely, in advanced malignancies, expression of a module of HIF1A targets involved in collagen remodeling is associated with poor prognosis across diverse cancer types. In this work, we provide a valuable resource for investigating context-dependent roles of HIF1A and its targets in cancer biology.