Abstract

Abstract The action of the AID/APOBEC family of cytosine deaminases contributes to mutational clustering but fails to explain 50% of the clusters observed in cancer genomes. Stress-induced mutagenesis in bacteria occurs when double-strand breaks (DSB) happen in DNA in the context of additional cellular stress sufficient to initiate the SOS response. This results in mutational clustering driven by DinB where mutational abundance decays as a function of the distance from the DSB, but remains above background rates of mutagenesis up to 1 MB away. In humans, the orthologous genes to DinB have become specialized for translesion synthesis (TLS). The dysregulation of cell cycle and DNA repair that characterizes most tumors would logically increase the need for TLS in cancer. Therefore, we hypothesized that stress-induced mutagenesis in cancer would result in peaked clusters of SNVs driven by TLS. We used data from 764 cases with somatic mutation and structural variant calls from WGS from the ICGC database, release 19, and private de novo mutations derived from 130 trios in the 1000 Genomes project with WGS. SNV context was determined based both on the reference and the mutant allele, and called when only one mechanism (TLS, APOBEC, or AID) could be assigned, thereby underestimating the actual number of TLS events. We observed that as the total number of SNVs increases so does the number of clusters, the proportion of SNVs in clusters, and the average size of clusters in both de novo private mutations and somatic mutations in cancer. In normal, a median of 15.2% of the SNVs occurring in clusters occur in a sequence context indicative of TLS, compared to 0.53% and 1.6% for APOBEC and AID, respectively. In contrast, in cancer, TLS accounts for a median of 30.6% of SNVs in clusters. APOBEC and AID account for 2.3% and 2.6%, respectively. We devised a measure of cluster shape based on empirical cumulative distributions, termed the Stress-Introduced Heterogeneity (SItH) score. It ranges from -1 to 1 and quantifies how sharply peaked clusters are with more positive numbers indicative of a faster decay in mutational load over distance relative to the putative DSB. In normal, SItH scores range from 0.457 to 0.578 with a median of 0.504. In cancer, SItH scores ranged from 0.17 to 0.999 with a median of 0.46 and vary significantly by organ site (ANOVA, F=44.96, p<2.2x10-16). We assessed the contribution of TLS, APOBEC, and AID to SItH score using an additive, linear model of SItH scores derived solely from SNVs in that specific context and including organ site as a variable. We found that in both normal and cancer that TLS was the only context that contributed significantly to SITH score. Therefore, we conclude that the clustering seen in cancer is being driven primarily by TLS and these clusters have a shape characteristic of a process of stress-induced mutagenesis. Citation Format: Kimberly J. Bussey, Luis Cisneros. A signature of stress-induced mutagenesis in cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 507. doi:10.1158/1538-7445.AM2017-507

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call