From Aldehyde-Induced DNA Damage to Pan-Cancer Biology: Linking Mutational Signatures to Their Biological Processes

Abstract

Cancer is mainly driven by the long-term accumulation of somatic mutations in normal cells. The cause of these mutations in different cancers can range from exogenous and endogenous chemicals to defective biological processes. The genomic analysis of mutational landscapes in different cancer types has revealed many distinct patterns of mutations collectively known as mutational signatures. The characterization of these mutational signatures is in progress; however, many signatures of unknown etiology remain. Although the International Agency for Research in Cancer (IARC) has classified both formaldehyde and acetaldehyde as human carcinogens, there is still minimal comprehension of their mutagenesis at the nucleotide level. We used a highly sensitive yeast genetic reporter system, which features the generation of single-stranded DNA (ssDNA), to demonstrate formaldehyde- and acetaldehyde-induced mutations. My former lab colleague, Reena Fabros, researched formaldehyde mutagenesis, while I focused on acetaldehyde, and computationally analyzed the mutational signature induced by formaldehyde and acetaldehyde. I assessed the cell viability and mutation frequency of acetaldehyde-induced yeast mutants, and isolated and sequenced their genomes. My findings revealed that the relative contribution of C->A, C->T, and T->C substitutions is higher in formaldehyde and acetaldehyde yeast compared to the untreated control yeast. Most importantly, the formaldehyde-induced mutational signature resembles a single base substitution SBS40 of unknown etiology from Catalogue of Somatic Mutations in Cancer (COSMIC) database. Moreover, acetaldehyde induced an excess of deletion events longer than five bases, while formaldehyde did not. Despite deciphering many of these mutational signatures from different model systems, a significant research gap remains for many mutational signatures with unknown or partially understood etiology. I used 52 curated cancer datasets of 19 different cancer types from the cancer genomics cBioPortal database. After reconstructing all the mutational signatures from mutation data, I used Gene Set Enrichment Analysis (GSEA) and Gene Ontology (GO) enrichment analysis to analyze the reconstructed mutational signatures and gene expression data and identify correlated biological processes. The common GO terms across male-only, female-only, and mixed sample datasets were found in most signatures. The results of many signatures are consistent with the known proposed etiology of the COSMIC signatures. For those signatures of unknown etiology, SBS8, SBS16, SBS28, and SBS41 are linked to DNA repair; SBS12, SBS19, SBS33, and SBS37 are associated with immune function; SBS17a/17b are linked to reactive oxygen species damage response; SBS5 with cell division; and SBS40 is linked to xenobiotic metabolism. The GO semantic similarity showed a range of values for different signatures. The correlation between age and mutation count varied for different mutational signatures. These findings could contribute to a better understanding of formaldehyde- and acetaldehyde-induced mutational signature and show how different mutational signatures vary not only by different related biological processes but also across different age groups.