Some biological systems give rise to highly reproducible phenotypes despite variation in environmental conditions. In such robust systems, phenotypic variation may occur within the developing system yet be buffered at the level of its output. This buffering allows the accumulation of "cryptic" mutations affecting the system without changing its end product. Cryptic genetic variation affecting the activity of the vulval signaling network was previously uncovered among C. elegans wild isolates (Milloz et al., 2008). One approach was to sensitize the system by introducing into different wild backgrounds a mutation that affects vulva cell fate induction. For example, the
let-23(
sy1)/egfr allele causes stronger vulva induction defects in the N2 background than in the AB1 background. To identify the underlying molecular variation, we used a quantitative genetic approach. We built 60 recombinant inbred lines between strains carrying
let-23(
sy1) in the N2 and AB1 backgrounds. Quantitative trait locus analyses detected one major-effect locus on chromosome I. Fine-mapping followed by functional analyses indicated that the causative polymorphism was a non-synonymous SNP in
nath-10, the homolog of the human N-acetyltransferase 10 gene whose function is largely unknown in C. elegans. The identified polymorphism modifies an otherwise conserved amino acid in the putative acetyltransferase domain. The N2 allele of
nath-10 behaves as a hypomorph. By genotyping other wild isolates and LSJ1, a lab relative of N2, we found that the
nath-10(N2) allele likely appeared during the early lab culture of N2. In addition to its cryptic role in the vulva system, the
nath-10 polymorphism also affects in a non-cryptic manner two life history traits with potentially opposite effects on fitness: the
nath-10(N2) allele confers a 10% larger brood size and an older age at maturity than the AB1 allele. Both effects can be explained through a modulation in the timing of the sperm/oocyte switch in hermaphrodites. Indeed,
nath-10(N2) animals produce more sperm and present a delay in oogenesis onset. These results show that cryptic genetic variation does not necessarily accumulate neutrally during the evolution of robust systems because of pleiotropic effects. Together with previous studies (McGrath et al., 2009; Weber et al., 2010), our results also indicate that the laboratory environment, which differs in several aspects from natural habitats, has driven fast phenotypic evolution of the N2 lab strain. We are currently performing competition assays to determine whether the N2 allele of
nath-10 confers a greater adaptive value in lab conditions.