Daniel E. Gottschling

There is a striking link between increasing age and the incidence of cancer in humans. Yet there is no satisfactory mechanistic explanation for this correlation. One of the hallmarks of cancer, genomic instability, is observed in all types of organisms including the yeast Saccharomyces cerevisiae. We recently discovered that as yeast cells enter the middle-to-late time of their replicative lifespan, they switch to a state of high genomic instability that persists until death. The genomic instability is manifested as a dramatic increase in loss of heterozygosity (LOH) in the progeny of older cells. The age-induced LOH is qualitatively different than in young cells; LOH proceeds by reciprocal recombination in young cells, while in old cells it proceeds primarily via the non-reciprocal pathway of break-induced replication. Furthermore, there is a striking bias for the mother to remain wild type, and the daughter to exhibit LOH.

In considering these and other recent results, we are working under the hypothesis that accumulation of damaged protein, or other molecules, in aging cells results in the loss of function of gene products critical for maintaining genome integrity. We are actively developing methods to determine the identity of these proteins and how they become damaged. In addition, we have used genetic approaches that are unique to yeast to uncover all the pathways by which LOH proceeds. We are currently focusing on those mutants that mimic the age-induced LOH events. The combination of these approaches will allow us to develop a molecular understanding of the relationship between age and genetic instability.

Our working hypothesis for age-induced genomic instability has led us to consider whether there are pathways that normally monitor and eliminate damaged proteins in the nucleus. The accumulation of defective or damaged proteins can have catastrophic effects on cell function and viability. To guard against this, cells have protein quality control pathways that detect aberrant proteins and then either try to repair them, by refolding them into their native structure, or eliminate them through degradation. In eukaryotes, protein quality control degradation pathways have been characterized in the cytoplasm, secretory pathway, and mitochondria, all sites of protein synthesis. In the nucleus however, where there is little or no protein synthesis, nuclear proteins are isolated from the known protein quality control degradation machineries. Given the importance of nuclear protein function, we were interested in finding protein quality control degradation pathways that operate in the nucleus.

Studying the budding yeast S. cerevisiae, we recently discovered the first protein quality control degradation pathway of the nucleus. It is defined by San1p, an ubiquitin-protein ligase. We found that San1p targets an assortment of mutant nuclear proteins for ubiquitination and subsequent destruction by the proteasome. As expected for a dedicated nuclear protein quality control pathway, San1p function requires its nuclear localization. We speculate that San1p-mediated degradation may act as the last line of defense against the deleterious accumulation of aberrant proteins in the nucleus. We are now taking advantage of biochemical and genetic approaches unique to yeast to develop a molecular understanding by which this pathway detects and destroys aberrant nuclear proteins.