Mutations accumulate within somatic cells and have been proposed to contribute to aging. It is unclear what level of mutation burden may be required to consistently reduce cellular lifespan. Human cancers driven by a mutator phenotype represent an intriguing model to test this hypothesis, since they carry the highest mutation burdens of any human cell. However, it remains technically challenging to measure the replicative lifespan of individual mammalian cells. Here, we modeled the consequences of cancer-related mutator phenotypes on lifespan using yeast defective for mismatch repair (MMR) and/or leading strand (Polε) or lagging strand (Polδ) DNA polymerase proofreading. Only haploid mutator cells with significant lifetime mutation accumulation (MA) exhibited shorter lifespans. Diploid strains, derived by mating haploids of various genotypes, carried variable numbers of fixed mutations and a range of mutator phenotypes. Some diploid strains with fewer than two mutations per megabase displayed a 25% decrease in lifespan, suggesting that moderate numbers of random heterozygous mutations can increase mortality rate. As mutation rates and burdens climbed, lifespan steadily eroded. Strong diploid mutator phenotypes produced a form of genetic anticipation with regard to aging, where the longer a lineage persisted, the shorter lived cells became. Using MA lines, we established a relationship between mutation burden and lifespan, as well as population doubling time. Our observations define a threshold of random mutation burden that consistently decreases cellular longevity in diploid yeast cells. Many human cancers carry comparable mutation burdens, suggesting that while cancers appear immortal, individual cancer cells may suffer diminished lifespan due to accrued mutation burden.

Heterozygous mutations affecting DNA polymerase (Pol) exonuclease domains and homozygous inactivation of mismatch repair (MMR) each generate "mutator" phenotypes capable of driving tumorigenesis. Cancers with both defects exhibit an explosive increase in mutation burden that appears to reach a threshold, consistent with selection acting against further mutation accumulation. In haploid yeast, simultaneous defects in polymerase proofreading and MMR select for "antimutator" mutants that suppress the mutator phenotype. We report here that spontaneous polyploids also escape this "error-induced extinction" and routinely out-compete antimutators in evolved haploid cultures. We performed similar experiments to explore how diploid yeast adapt to the mutator phenotype. We first evolved cells with homozygous mutations affecting proofreading and MMR, which we anticipated would favor tetraploid emergence. While tetraploids arose with a low frequency, in most cultures, a single antimutator clone rose to prominence carrying biallelic mutations affecting the polymerase mutator alleles. Variation in mutation rate between subclones from the same culture suggests there exists continued selection pressure for additional antimutator alleles. We then evolved diploid yeast modeling MMR-deficient cancers with the most common heterozygous exonuclease domain mutation (POLE-P286R). Although these cells grew robustly, within 120 generations, all subclones carried truncating or nonsynonymous mutations in the POLE-P286R homologous allele (pol2-P301R) that suppressed the mutator phenotype as much as 100-fold. Independent adaptive events in the same culture were common. Our findings suggest that analogous tumor cell populations may adapt to the threat of extinction by polyclonal mutations that neutralize the POLE mutator allele and preserve intra-tumoral genetic diversity for future adaptation.

In yeast, the pol3-01,L612M double mutant allele, which causes defects in DNA polymerase delta (Pol δ) proofreading (pol3-01) and nucleotide selectivity (pol3-L612M), confers an “ultramutator” phenotype that rapidly drives extinction of haploid and diploid MMR-proficient cells. Here, we investigate antimutator mutations that encode amino acid substitutions in Pol δ that suppress this lethal phenotype. We find that most of the antimutator mutations individually suppress the pol3-01 and pol3-L612M mutator phenotypes. The locations of many of the amino acid substitutions in Pol δ resemble those of previously identified antimutator substitutions; however, two novel mutations encode substitutions (R674G and Q697R) of amino acids in the fingers domain that coordinate the incoming dNTP. These mutations are lethal without pol3-L612M and markedly change the mutation spectra produced by the pol3-01,L612M mutator allele, suggesting that they alter nucleotide selection to offset the pol3-L612M mutator phenotype. Consistent with this hypothesis, mutations and drug treatments that perturb dNTP pool levels disproportionately influence the viability of pol3-L612M,R674G and pol3-L612M,Q697R cells. Taken together, our findings suggest that mutation rate can evolve through genetic changes that alter the balance of dNTP binding and dissociation from DNA polymerases.

26DNA polymerase (Pol) proofreading and mismatch repair (MMR) defects produce 27 "mutator" phenotypes capable of driving tumorigenesis. "Ultra-mutated" cancers with defects in 28 both pathways exhibit an explosive increase in point mutation burden that appears to reach an 29 upper limit of ~250 mutations/Mb, which may be evidence for negative selection. Simultaneous 30 defects in proofreading and MMR drive error-induced extinction (EEX) of haploid yeast. Some 31 yeast mutants escape extinction through antimutator mutations that suppress the mutator 32 phenotype. We report here that other yeast mutants escape by spontaneous polyploidization. 33 We sought to compare these two adaptive strategies during the evolution of haploid and diploid 34 mutators. We first evolved 89 independently isolated haploid mutator strains with mutation rates 35 an order of magnitude below the lethal threshold. Polyploid cells overtook nearly every culture.36 In a similar manner, we evolved strong diploid mutators, but found only a single tetraploid 37 emerged after ~250 cellular generations. Antimutators dominated these late passage cultures 38 and were present after only 30-40 cellular generations. Whole genome sequencing of 39 independent isolates from the same culture allowed us to construct phylogenetic trees that 40 revealed that a single subclone swept through each culture carrying multiple antimutator 41 mutations affecting the mutator polymerase. Variation in mutation rate between subclones from 42 the same culture suggests some strains carry additional antimutator alleles in other genes. Our 43 findings in diploid mutator yeast suggest that mutator-driven cancer cells may also adapt to the 44 threat of extinction by accumulating multiple antimutator alleles that cooperatively reduce 45 mutation rate to tolerable levels. 46Author summary 47 "Mutator" tumor cells that cannot correct DNA replication errors exhibit an extremely high 48 mutation rate that accelerates their evolution. But this gamble puts them at risk for extinction 3 49 due to the accumulation of harmful mutations. Mutator yeast cells escape extinction through 50 "antimutator" mutations that lower mutation rate. We discovered that yeast cells also adapt by 51 duplicating their genome. To understand which adaptation may emerge in mutator-driven 52 tumors, we evolved haploid and diploid mutator yeast cells by long-term propagation. Haploid 53 mutators primarily adapted by genome duplication, thereby protecting themselves from 54 recessive lethal mutations. In contrast, diploid mutators evolved by acquiring multiple 55 antimutator mutations. Our results suggest that mutator-driven tumors may similarly accumulate 56 combinations of mutations that attenuate their high mutation rates.4 57 58The high fidelity of DNA replication and repair prevents mutations that might otherwise 59 lead to cancer. As first proposed more than 40 years ago, cells with defects in these pathways 60 exhibit an elevated mutation rate ("mutator phenotype") and an accelerated path to malignancy 61...

Mutations that compromise mismatch repair (MMR) or DNA polymerase ε or δ exonuclease domains produce mutator phenotypes capable of fueling cancer evolution. Here, we investigate how combined defects in these pathways expands genetic heterogeneity in cells of the budding yeast, Saccharomyces cerevisiae, using a single-cell resolution approach that tallies all mutations arising from individual divisions. The distribution of replication errors present in mother cells after the initial S-phase was broader than expected for a single uniform mutation rate across all cell divisions, consistent with volatility of the mutator phenotype. The number of mismatches that then segregated to the mother and daughter cells co-varied, suggesting that each division is governed by a different underlying genome-wide mutation rate. The distribution of mutations that individual cells inherit after the second S-phase is further broadened by the sequential actions of semiconservative replication and mitotic segregation of chromosomes. Modeling suggests that this asymmetric segregation may diversify mutation burden in mutator-driven tumors.