Genomic approaches to DNA repair and mutagenesis

Abstract

DNA damage is a constant threat to cells, causing cytotoxicity as well as inducing genetic alterations. The steady-state abundance of DNA lesions in a cell is minimized by a variety of DNA repair mechanisms, including DNA strand break repair, mismatch repair, nucleotide excision repair, base excision repair, and ribonucleotide excision repair. The efficiencies and mechanisms by which these pathways remove damage from chromosomes have been primarily characterized by investigating the processing of lesions at defined genomic loci, among bulk genomic DNA, on episomal DNA constructs, or using in vitro substrates. However, the structure of a chromosome is heterogeneous, consisting of heavily protein-bound heterochromatic regions, open regulatory regions, actively transcribed genes, and even areas of transient single stranded DNA. Consequently, DNA repair pathways function in a much more diverse set of chromosomal contexts than can be readily assessed using previous methods. Recent efforts to develop whole genome maps of DNA damage, repair processes, and even mutations promise to greatly expand our understanding of DNA repair and mutagenesis. Here we review the current efforts to utilize whole genome maps of DNA damage and mutation to understand how different chromosomal contexts affect DNA excision repair pathways.

Keywords: Excision repair; Mutagenesis; Mutation signature; Sequencing.

Fig. 1. Schematic showing experimental strategies for mapping DNA lesions throughout the genome

A. Immunoprecipitation followed by microarray analysis. B. Pre-digestion Excision-seq [37]. Cyclobutane pyrimidine dimers (CPD) are represented with a T=T symbol in brown, while 6-4 photoproducts (6-4PP) are in blue. Deoxyuracil lesions (dUracil) are in green. In the Excision-seq method, the damaged DNA is initially cleaved with uracil DNA glycosylase (UDG) and Endonuclease IV (Endo IV). This example shows the expected sequencing results for mapping dUracil lesions arising due to misincorporation during replication

Fig. 2. DNA damage and repair influence mutation signatures

Where DNA damage occurs as well as how it is processed and repaired determine the spectrum of mutations that results from exposure to a DNA damaging agent. For example, UV radiation occurs at specific dinucleotides: TT, TC, CC, and CT (not pictured). Nucleotide Excision Repair (NER) can remove these lesions, but in some contexts, such as when a lesion resides in dense chromatin, repair may be inefficient allowing subsequent processes to convert these lesions into mutations (red letters). Unrepaired T-T dimers are frequently bypassed in an error-free manner by DNA polymerase eta, while deamination of cytidines in dimers as well as error-prone bypass results in nearly all UV induced mutations occurring at TC and CC dinucleotides (bottom, all possible trinucleotide DNA sequences are shown with the mutated base capitalized, complementary sequences are combined).

Fig. 3. Profiling cancer mutations against genomic features

A. To identify genomic features that associate with elevated mutation densities, the genomes of multiple excised human tumors are sequenced by next-generation sequencing techniques and a list of somatically acquired mutations (red) is compiled by comparison to a patient’s germline genome. To call mutations using next-generation sequencing, the mutant allele must be supported by multiple reads in the sequencing data, resulting in the positions of only sub-clonal mutations being determined. B. In parallel, the locations of various genomic features (e.g. histone modifications (black), chromatin accessibility (light blue), replication timing (purple)) is determined by shot-gun sequencing DNA ends associated with immune-precipitated proteins or DNA modifications (i.e. ChIP-seq, Repli-seq) or DNA ends generated by nucleolytic cleavage (DNase-seq). Cell lines utilized for these experiments are generally chosen to closely resemble the tissue of origin of the tumor whose mutations are analyzed. C. Correlations between mutation position and any genomic feature is determined by either comparing mutation densities to sequencing tag densities in a given window of genomic space, or by stratifying regions of the genome by the tag densities of a chromosome feature, and subsequently determining the mutation density with each feature category.