hbl-1 is critical for hypodermal cell fates in the larva (Fay, 1999; Abrahante, 2003; Lin, 2003). At least three miRNAs repress
hbl-1 at a critical time to permit the transition from L2 to L3 cell fates (Abbott, 2005). We sought to understand how
lin-28 and
lin-46, which are apparently just upstream of
hbl-1, control the same decision. First, we found that
lin-28 positively regulates
hbl-1 expression. In a strain lacking the three miRs,
hbl-1 is constitutive, whereas in a strain also lacking
lin-28, it is down-regulated. Therefore, either
lin-28 directly supports
hbl-1 or it represses a fourth repressor. In investigating mammalian Lin28, we found it associated with mRNPs, ribosomes and P-bodies, suggesting it is an mRNA binding protein (Moss, 2003; Balzer, 2007). Furthermore, it positively regulates the translation of Igf2 mRNA (Polesskaya, 2007). However, several recent studies have shown-remarkably-that mammalian Lin28 specifically binds to and blocks the processing of
let-7 family miRNA precursors (Viswanathan, 2008; Newman, 2008, Piskounova, 2008; Rybak, 2008, Heo, 2008). Using a yeast three-hybrid assay, we found that LIN-28 specifically binds precursors of four of seven
let-7-family miRNAs:
let-7, miR48, miR-84, miR-241, but not miR-793, -794 or -795. It does not bind pre-miRs of miR-1,
lin-4, or several other RNA sequences, including portions of the
hbl-1 3''-UTR. To determine whether
lin-28 affects the accumulation of miRNAs in vivo, we used TaqMan miRNA qRT-PCR assays. Using RNA from approximately 150 wildtype or
lin-28(0) larvae in the pre-molt lethargus at each stage, we found, to our surprise, no effect of
lin-28 on the level of miR-48 at any stage. By contrast, we saw large changes in
let-7 levels at both the L1 and L2 molts. While further assays are being conducted, our preliminary results suggest that
lin-28, indeed, specifically binds to and greatly affects
let-7 accumulation in C. elegans. We know, however, that blocking
let-7 can''t be the only thing
lin-28 is doing. LIN-46 resembles certain proteins involved in protein-protein interactions (Pepper, 2004). Genetic data suggest
lin-46 acts upstream of
hbl-1, working with or in parallel to the three miRs. We conducted a two-hybrid screen to identify LIN-46 interactors and found, to our surprise, LIN-46 specifically binds the product of another heterochronic gene, both in yeast and in vitro. LIN-46 interacts with a conserved C-terminal zinc-finger domain of this protein, which, when expressed in wildtype animals, produces a phenotype like that of
lin-46(0). Thus, circumstantial evidence suggests that LIN-46 does not act at the level of the miRNA regulation of
hbl-1, but rather just downstream of that.