Genome wide interrogation of somatic mutational signatures of oxidative stress and driver mutations in mouse and human cancers

Purpose: Over 40% of human cancers are attributable to modifiable risk factors such as obesity, diet, smoking, ionizing radiation and inflammatory diseases, which are thought to cause chronic or excessive exposures to oxidative stress. Whole genome sequencing (WGS) of thousands of human tumors have revealed “mutational signatures” that provide a molecular footprint of cancer origins. Whether such signatures exist for oxidative stress is not clear though several signatures have been implicated as being related to reactive oxygen species (ROS).

Methodology: Using a compendium of WGS data from >300 mouse tumors and in silico analysis of WGS from thousands of human tumors, we explored the extent to which ROS mutational signatures (SBS.ROS) contribute to the etiology of solid tumors.

Results: Mutations attributable to ROS were broadly found in 25% of all mouse tumors induced by inflammation driven models (diet, obesity, ROS generating carcinogens, wounding). Interestingly, they are the most prominently observed in tumors from mice that are exposed to carcinogens in utero, suggesting that the developmental age of mutagen exposure may impact the repair of ROS generated mutations. In contrast, only 10.7% of mouse tumors induced by environmental carcinogens bear SBS.ROS, suggesting that exogenous carcinogens act through mutagenic mechanisms rather than through inflammatory or promotional mechanisms. However, one compound, anthraquinone, was strongly associated with SBS.ROS in liver tumors. In addition, we analyzed 21 mouse mammary tumors in Trp53 deficient mice exposed to either gamma or high-LET total body exposure the levels of SBS.ROS was preferentially enriched in tumours induced by gamma radiation, as compared to those induced by high LET radiation (p < 0.032, Fisher’s Exact test).
Finally, we compared the observed driver mutations with those previously published in these large scale human studies, demonstrating that the driver mutation genes identified from 316 mouse tumors recapitulated >50% of established human cancer driver genes. In addition, we developed a novel approach to driver mutation signature analysis which allowed us to computationally assess the probability a given driver mutation is caused by a specific environmental or endogenous process. We found that while environmental carcinogens causes the bulk of all driver mutations observed, we observed a strong bias in the contribution of SBS.ROS mutations to driver mutations in genes such as Trp53 and Tert, which occur typically later during the process of carcinogenesis. Finally, we showed that there is a similarly strong prevalence of SBS.ROS in human malignancies associated with inflammatory risk factors, e.g. over 58% of human colorectal cancers bear SBS.ROS.

Conclusions: Taken together, we have analyzed the largest compendium of WGS data from over 300 mouse tumors, showing that while exogenous promotional factors do not increase mutation burden or induce novel mutational patterns, they have a major rate-limiting role in determining cancer risk.

References: