Defining genes that govern longevity in Caenorhabditis elegans.

We previously identified five regions on the chromosomal map of Caenorhabditis elegans, containing genes that help specify life span in this species, by comparing the genotypes of young and long-lived progeny from a cross between strains Bristol-N2 and Bergerac-BO [Ebert et al. (1993): Genetics 135:1003-1010]. Analyses of additional crosses, and of putative polymorphisms for the implicated genes, are necessary to clarify the roles of naturally occurring polymorphic alleles in determining longevity. We therefore carried out a second multigenerational cross, between strains Bristol-N2 and DH424 (both nonmutators at 20 degrees C), to create a different heterogeneous recombinant-inbred population. We again found strong evidence implicating multiple genes, which differ between the parental strains, in the determination of life span. Increased variance of survival, for F2 and homozygous F25 worms relative to F1 hybrids, is consistent with such alleles assorting randomly in the cross progeny. Moreover, chromosome mapping data corroborate the polygenic nature of this quantitative trait. Genotypes of young and very long-lived adult worms from a synchronous F15 population were determined by polymerase chain reaction, to identify the parental strain of origin for each of 10 polymorphic loci. Two regions, on chromosomes II and IV, each contain at least one gene with allelic differences in associated longevity. A recombinant-inbred Bergerac-BO x Bristol-N2 population, derived from the earlier cross between those strains, was exposed to an acute toxic level of hydrogen peroxide. Genotyping of H2O2-resistant worms implicated at least one of the five chromosomal regions previously identified in the same cross progeny as harboring a longevity-determining gene. Superoxide dismutase and catalase levels, determined for the three parental strains as they aged, confirm the existence of polymorphisms in the corresponding genes (or their regulatory mechanisms) inferred from the chromosome-II mapping data, and are consistent with the hypothesis that increased longevity is conferred by high levels of these enzymes late in life.