Poster Presentation 29th Lorne Cancer Conference 2017

Recurrent promoter mutations in melanoma are defined by an extended context-specific mutational signature 

Kerryn Elliott 1 , Johan Fredriksson 1 , Stefan Filges 2 , Anders Ståhlberg 2 , Erik Larsson 1
  1. Department of Medical Biochemistry and Cell Biology, Institute of Biomedicine, Sahlgrenska Academy, Gothenburg University, Gothenburg, Västra Götaland, Sweden
  2. Sahlgrenska Cancer Centre, Institute for Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Västra Götaland, Sweden

Sequencing of whole tumour genomes holds the promise of revealing functional somatic
 regulatory mutations, such as those described for the TERT promoter. Recurrent 
promoter mutations have been identified in many additional genes and appear to be 
particularly common in melanoma, however, obtaining convincing functional data which influences gene expression has been elusive. We have identified recurrent promoter mutations in melanoma patients using whole genome sequencing data from TCGA from 38 metastatic skin cutaneous melanoma tumours together with matching RNA-seq and copy number data. These promoter mutations have a distinct 5’-NCTTCCGGN-3’ signature. In active, but not inactive, promoters, mutation frequencies for the cytosine at the 5’ end of the signature are considerably higher than expected based on the UV trinucleotide mutation signature, and the rate of mutation at these sites increases with total mutation load. This argues against positive selection, as had been suggested as this motif corresponds to the binding site for ETS transcription factors. Further to the bioinformatic analysis, ultrasenstive SiMS-seq shows that low dose UV is sufficient to induce these mutations at low frequencies, and preliminary results using biochemical methods show that these sites display increased cyclopyrimidine dimer (CPD) formation. This may indicate that ETS transcription factor binding alters the local DNA structure leading to increased availability for CPD formation.

This finding has implications for the interpretation of somatic mutations in regulatory regions, and underscores the importance of genomic context to accurately describe mutational signatures in cancer.