Abstract
Faithful and accurate transmission of the genetic information and maintenance of genomic integrity requires DNA duplication to proceed with remarkable fidelity. Excessive or incomplete DNA replication, that were not properly resolved by the cell, can result in genomic instability, developmental abnormalities, and cancer. Eukaryotic cells have evolved a multitude of regulatory mechanisms that work in concert to ensure a timely and accurate DNA replication and coordinate progression through S-phase. Failure to limit DNA replication to a single round per cell cycle results from erroneous origin re-firing leading to rereplication. Excessive rereplication is toxic to cells due to the accumulation of replication intermediates and stalled replication forks, and can lead to DNA breaks, chromosomal abnormalities, apoptosis, senescence and tumorigenesis. Amongst the regulatory mechanisms that guards against origin re-licensing is the degradation, during S-phase of the cycle, of key replicative factors via the ubiquitin proteasomal system.
Ubiquitin-mediated proteolysis is a highly regulated mechanism that controls the degradation of most cellular proteins with remarkable timing and specificity. CRL4CDT2 E3 ubiquitin ligase plays a critical role in preventing excessive origin re-licensing in the same S-phase through the degradation of the licensing factor CDT1, the histone mono-methyltransferase SET8, and the CDK inhibitor p21. Our work shows that the expression of CDT2, the CRL4CDT2 substrate adaptor, is elevated in melanoma and head and neck squamous cell carcinoma (HNSCC), and its elevated expression in melanoma correlates with poor patient outcome. Inactivation of CRL4CDT2, via CDT2 depletion or pharmacological inhibition using MLN4924 (Pevonedistat), an inhibitor of the activity of all cullin-based E3 ligases, suppresses melanoma proliferation through the induction of a p21 and SET8-dependent rereplication and senescence. We showed that MLN4924 suppresses melanoma tumor growth irrespective of the BRAF/NRAS mutational status. We have also shown that MLN4924-induced toxicity, both in vitro and in vivo, is mediated through failure to degrade p21 or SET8 during S-phase. We found that MLN4924 synergizes with the BRAF kinase inhibitor vemurafenib to suppress BRAF melanomas in vivo and is effective against vemurafenib-resistant melanomas. We have also shown that MLN4924 inhibits and radio-sensitizes HPV-negative HNSCC in vivo. Subsequent analysis demonstrates that rereplication is sufficient to confer radiation sensitivity in HNSCC.
My recent work with ionizing radiation (IR) identified rereplication as an underlying mechanism for inducing cytotoxicity in a subset of cancer cells of various epithelial origins. I have shown that DNA double-strand breaks (DSBs) are sufficient to induce rereplication in these cancer cells. The DSB-induced rereplication, or DIRR, correlates with IR-induced toxicity in melanoma cells, and is thus likely to impact the efficacy of radiotherapy in clinical settings. Mechanistically, we show that DIRR does not involve origin re-firing, and likely initiated by unshielded, hyper-resected broken DNA ends invading non-homologous sequences early in S-phase.
In summary, using various cancer model systems, my studies have shown how rereplication induction in cancer cells can exhibit anti-tumorigenic activities and demonstrate that it mediates the efficacy of new therapeutic agents (MLN4924), IR and other DSB-inducing chemotherapies.
