Larval development of C. elegans occurs in four stages that are punctuated by molts. Each molt is accompanied by a peak in expression of a broad spectrum of genes, referred to as molting cycle genes. Expression of these genes peaks at different times but with the same period as the molts. Such periodic gene expression suggests the existence of a timing mechanism capable of precise temporal regulation. The gene
lin-42 is thought to be implicated in such a mechanism as its deletion causes severe developmental defects with extended larval stage duration and asynchronous molts. The goal of the current research is to understand the mechanism of temporal control by
lin-42. In particular we are interested in how
lin-42 controls oscillatory expression of the molting cycle genes. To address this, we examine the dynamics of molting cycle genes in
lin-42(
ok2385) mutant animals. For this, we use a newly-developed fluorescence time-lapse microscopy technique, which allows us to follow gene expression at a single-animal level over the entire course of development. Initially we focused on
mlt-10, a gene that encodes for one of the cuticle components, whose expression exhibits peak ~ one hour prior to ecdysis. We found that even though in
lin-42(
ok2385) animals time of the
mlt-10 peaks was delayed, their duration was same as in wild-type. Moreover, we found that the L2 peak in
mlt-10 expression was absent in 30% of animals that reached the L2/L3 molt. Interestingly, presence of peak was not correlated with initiation of subsequent ecdysis and occurred in all-or-none fashion: L2 peak was either fully absent or exhibited wild-type amplitude. We use mathematical modelling to show that these phenotypes are consistent with regulation by a mixed incoherent feed-forward loop. Currently we are extending this analysis to other molting cycle genes that peak at different phases as well as other
lin-42 alleles.