The germline is an excellent system for investigating translational control as it is a major mechanism for regulating gene expression. GLD-1 is a germline specific, maxi-KH motif containing RNA binding protein that controls multiple aspects of C. elegans germ cell development, suggesting that it regulates multiple mRNAs. To understand how GLD-1 regulates mRNAs to control germ cell development, we have previously identified multiple in vivo mRNA targets of GLD-1 that were co-immunoprecipitated (IP) with GLD-1 from cytosol extracts and recovered following subtractive hybridizationa, b. These target mRNAs are preferentially expressed in the germline and several have essential functions in oocyte differentiation and early embryogenesis. GLD-1 is a translational repressor, based on the analysis of mRNA and protein levels of five mRNA targets (
rme-2,
puf-5,
cep-1,
oma-1, and
oma-2) in wild-type and
gld-1 null germlines. However, it has been clear that we do not have the complete list of GLD-1 mRNA targets. Thus, we used microarray analysis to identify additional targets that are co-IPed with GLD-1. We identified 129 genes that are enriched more than two fold (p<0.05) in the GLD-1 IP over the control IP. Essentially all of our previously identified targets are recovered in the new screen. To determine the in vivo binding specificity of GLD-1 or GLD-1 containing complexes, we have used GLD-1 from wild-type cytosol extracts to identify 12 GLD-1 binding regions in six targets thus far. GLD-1 can bind the 5UTR, 3UTR or in the coding region depend on the target. Using the sequences of 13 GLD-1 binding regions (12 in the six targets and the
tra-2 3UTR) and the corresponding, orthologous C. briggsae regions, we found 3 potential GLD-1 binding motifs with PhyloCon, a program that identifies regulatory motifs among co-regulated orthologous gene setsc. One of the motifs is essentially identical to the hexanucleotide consensus that was identified in the
tra-2 3UTR by Ryder et al, 2004d. Interestingly, the
tra-2 3UTR has only this motif four times. Most other binding regions have two different motifs and three regions do not contain the hexanucleotide consensus. Biochemical analysis of the motifs indicates that two motifs, including the hexanucleotide consensus, are important for GLD-1 binding and synergistic interactions between motifs likely occur. Therefore, the data from the IP/microarray analysis and the binding studies suggest that, in addition to the hexanucleotide consensus, GLD-1/GLD-1 containing complexes likely utilize other sequences to achieve specificity. a, b Lee and Schedl, Genes Dev, 2001 & 2004; c, Wang and Stormo Bioinformatics, 2003; d, Ryder et al. Nat Struct Mol Biol, 2004