DNA replication and responses to DNA damage

How DNA damages cause genome instability

We have investigated the origin of dsDNA breaks and genomic rearrangements induced by nucleotide damages, using UV light as a prototypic clastogenic agent. We found that unreplicated UV-DNA damages can be transferred through mitosis, only collapsing to dsDNA breaks during the S phase of the cell cycle following that of exposure. End joining of these broken chromosomes underlies the punctuate accumulation of genomic rearrangements. This mechanism had not been recognized before. We hypothesize that it underlies large-scale genomic rearrangements observed in cancers that are associated with chronic exposure to low levels of DNA-damaging agents and with replication stress. In support, prostate carcinomas with many genomic aberrations are enriched in defects in nucleotide excision repair and in dsDNA breaks repair by homologous recombination; pathways that protect the cell against DNA damages and against dsDNA breaks-induced rearrangements, respectively (Jansen et al., manuscript submitted).

Genetic interactions between TLS and MMR

We have previously found that the MMR protein complex Msh2/Msh6 suppresses the mutagenicity of UV light, while it simultaneously induces DNA damage responses ¹. This was unexpected since a priori MMR is not expected to correct misincorporations by TLS opposite UV-DNA damages. We have now shown that the suppression of UV-induced mutations extends to the other MMR protein complex, Mlh1/Pms2. However, in contrast to deficiency for Msh2/Msh6, deficiency for Mlh1/Pms2 does not lead to compromised DNA damage responses, revealing partial separation-of-function between both MMR complexes. In addition, by generating double-deficiencies in MMR and TLS proteins, we now provide strong genetic evidence that MMR corrects mutagenic TLS (Jansen et al., in preparation).

To further study the interaction between TLS and MMR, we investigated the involvement of MMR in polymerase zeta-dependent TLS of methylguanines. Briefly, we found that MMR corrects TLS polymerase zeta-dependent misincorporations not only opposite methylguanines, but also during replication of the undamaged template, following of bypass of methylguanines ^2.     

Having demonstrated that MMR can suppress nucleotide substitutions induced by mutagenic compounds, we investigated whether gastrointestinal cancer development exposure in Lynch syndrome might be associated with exposure to food mutagens. Using mice, heterozygous for the MMR gene Msh2, we found that a ubiquitous food mutagen induces loss of the wildtype Msh2 allele in cultured cells and intestinal crypts. Moreover, consistent with hour other results, the mutagen induced significantly more mutations in MMR-deficient crypts than in MMR-proficient crypts. This hypermutator phenotype, in conjunction with the spontaneous mutator phenotype of MMR-deficiency, provoked the rapid onset of intestinal tumors in MMR-deficient crypts, and these histopathologically resembled gastrointestinal carcinomas in Lynch syndrome patients. These data suggest that food mutagens may contribute to intestinal cancer development in Lynch syndrome at multiple levels, providing a rationale for dietary adjustments as a cancer-preventive strategy in Lynch syndrome (³; Rayner et al, in preparation).

Replication stress as a tumor promoter

The mutagenicity of TLS predicts it to be pro-carcinogenic. We previously found that a defect in the TLS gene Rev1 indeed caused the reduced mutagenicity of UV light, but exacerbates replication stress and DNA damage responses. Paradoxically, however, UV light-induced carcinogenesis of the mouse skin was enhanced in Rev1-defective mice. This unexpected phenotype was associated with the induction of severe epidermal hyperplasia by low-dose UV light ⁴. We have now shown that replication stress induces the expression and secretion of mitogenic factors that induce hyperproliferation and hyperplasia of the epidermis. We have identified the intracellular signaling pathway of this paracrine response to replication stress, as well as multiple drugs that inhibit this pathway. Indeed, these drugs prevent replication stress-induced epidermal hyperplasia and, importantly, suppress skin cancer development. We therefore infer that such drugs may be useful for the prevention and treatment of cancer (Tsaalbi-Shtylik et al., in preparation).

Diagnostic tools to assess pathogenicity of MMR genes variants.

We have developed, calibrated and validated the CIMRA assay, the first in vitro assay to classify VUS in the four MMR genes, identified in individuals suspected of Lynch syndrome ^5-7. This assay is endorsed as diagnostic test by the American College of Medical Genetics/American Society of Molecular Pathology and currently used to classify MMR gene variants in the Dutch population, in the framework of the nation-wide INVUSE consortium. In addition, the CIMRA assay is being implemented in the diagnostic routine of the Department of Clinical Genetics that will offer it worldwide.

As a complementary approach to the CIMRA assay, we are performing cell-based genetic screens to map pathogenic variants in the MMR genes, at a large scale. These will be listed in Prospective Classification Catalogs that can be employed to identify pathogenic gene variants by simple lookup.