Characterizing and therapeutically exploiting chromosome instability induced by USP22 deficiency in colorectal cancer

Abstract

Colorectal cancer (CRC) remains the second leading cause of cancer-associated deaths in Canada. To develop new treatments with enhanced efficacy, a greater understanding of the processes driving CRC pathogenesis is required. In this regard, chromosome instability (CIN) is an aberrant phenotype observed in ~ 85% of CRCs that is characterized by an increased rate of chromosome gains and losses. Although CIN contributes to CRC development and is associated with poorer outcomes, the genetic defects giving rise to CIN in tumors remain largely unknown. In mitosis, chromatin compaction is critical to ensure accurate chromosome segregation. Intriguingly, monoubiquitination of histone H2B (H2Bub1) impairs chromatin compaction in vitro, while H2Bub1 is rapidly depleted from chromosomes upon mitosis onset in vivo. This suggests that H2Bub1 removal in mitosis is required for accurate chromosome compaction and segregation. Accordingly, impaired H2Bub1 removal may disrupt mitotic fidelity and promote CIN. In interphase, USP22 is a major enzyme catalyzing H2Bub1 removal, which may also be responsible for H2Bub1 depletion in mitosis. In this thesis, quantitative imaging microscopy revealed that siRNA-based USP22 depletion impairs H2Bub1 removal and mitotic chromatin compaction in CRC cell line HCT116 and induces CIN in two karyotypically stable cell lines (Chapter 4). As USP22 is deleted in ~ 48% of CRCs, I employed CRISPR/Cas9 methods to generate homozygous and heterozygous USP22 knockout models in malignant and non-malignant colonic epithelial cell lines. Long term monitoring of these USP22-deficient models revealed dynamic CIN phenotypes relative to controls, which identifies reduced USP22 expression as a novel genetic determinant of CIN and indicates that USP22 deletion may promote CRC pathogenesis (Chapter 5). To discover drug targets to selectively kill USP22-depleted cells, I employed the USP22-deficient models in a screen of 239 DNA damage response genes that identified 86 putative USP22 synthetic lethal interactors. Validation assays for two top candidates revealed that the clinically approved drug sorafenib preferentially targets USP22-deficient cells relative to controls (Chapter 6). Collectively, these findings provide novel insight into the molecular origins of CIN and represent a first step towards the development of therapeutic strategies that effectively kill USP22-deficient CRCs.