Copy number aberrations drive kinase rewiring, leading to genetic vulnerabilities in cancer.

Danish Memon,Martin L. Miller,Clive S. D’Santos,Pedro Beltrao,Michael B. Gill,Evangelia K. Papachristou,David Ochoa

doi:10.1016/j.celrep.2021.109155

Danish Memon, Martin L. Miller + Show 5 more

Open Access

PDF Available

https://doi.org/10.1016/j.celrep.2021.109155

Copy DOI

Export

Save

Cite

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

SummarySomatic DNA copy number variations (CNVs) are prevalent in cancer and can drive cancer progression, albeit with often uncharacterized roles in altering cell signaling states. Here, we integrate genomic and proteomic data for 5,598 tumor samples to identify CNVs leading to aberrant signal transduction. The resulting associations recapitulate known kinase-substrate relationships, and further network analysis prioritizes likely causal genes. Of the 303 significant associations we identify from the pan-tumor analysis, 43% are replicated in cancer cell lines, including 44 robust gene-phosphosite associations identified across multiple tumor types. Several predicted regulators of hippo signaling are experimentally validated. Using RNAi, CRISPR, and drug screening data, we find evidence of kinase addiction in cancer cell lines, identifying inhibitors for targeting of kinase-dependent cell lines. We propose copy number status of genes as a useful predictor of differential impact of kinase inhibition, a strategy that may be of use in the future for anticancer therapies.

Highlights

Kinases are druggable proteins that are key targets for cancer treatment, because they are highly prone to acquiring somatic mutations and acting as oncogenes (Gross et al, 2015)
From the integration of large-scale pan-cancer genomic, transcriptomic, and reverse-phase protein arrays (RPPAs) phosphoproteomic datasets, we identified genes likely to drive changes in kinase signaling
To focus on changes in stoichiometry, we restricted our analysis to 37 phosphosites that had matched protein abundances in the RPPA datasets for normalization purposes (STAR Methods)

Summary

Introduction

Kinases are druggable proteins that are key targets for cancer treatment, because they are highly prone to acquiring somatic mutations and acting as oncogenes (Gross et al, 2015). Several studies have proposed that cancer cells can become dependent on or addicted to changes in kinase signaling (Sawyers, 2004). Cancer genomes are characterized by a large number and variety of mutations, including somatic point mutations and copy number aberrations. These mutations may have many direct and indirect effects that are likely to rewire signaling pathways, giving cancer cells a growth advantage. The impact of copy number aberrations on kinase signaling activity are often unknown, besides some well-characterized cases of copy number variation (CNV) in signaling genes such as PTEN (Chalhoub and Baker, 2009) and ERBB2, which encodes for the HER2 protein (Moasser, 2007). Identifying downstream effects of copy number changes is complicated, because they encompass large segments of genomes with many genes; it is difficult to identify the likely causal gene within the CNV region

Results

Discussion

Conclusion