Worm Breeder's Gazette 11(3): 43
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Experiments with a mec-3 lacZ fusion suggest that mec-3 is expressed in ten cells: the six touch receptor neurons, the PVD cells, and the FLP cells. (Way & Chalfie, Genes and Development 3: 18231833, 1989). In mec-3 mutants, touch cell differentiation appears to be blocked and the PVD cells no longer function. To investigate how this putative specification gene acts, we have extended the molecular analysis of the mec-3 product. The mec-3 transcript had previously been defined by a partial cDNA that contained four exons from the genomic sequence (Way & Chalfie, Cell 54:5-16,1988). This sequence encoded a homeodomain and an acid- rich region. Examination of the genomic sequence (Way & Chalfie, WBG 10:3, and Freyd et al., WBG 11:1) led to predictions of a cysteine- rich region at the N-terminus of the protein product. We have isolated seven mec-3 cDNAs from among 3.5 million plaques from a 1-2kb size-selected library from wild-type animals (the library was made by Chris Martin in our lab). All have poly A tails and are approximately 1.26kb in length. There is some variation at the 5' ends of the cDNAs and at the site of poly A addition. All seven cDNAs have the same ORF which encodes a polypeptide of 321 amino acids. We believe that the entire coding sequence is contained in the cDNAs since the first ATG codon corresponds to the eukaryotic translation initiation site consensus (A at -3 position, G at +4 position; Kozak, M. N.A.R., 15:8125,1987) and is preceded, in frame, by a stop codon. The cDNAs confirm many of the predicted exons, but differ from both predictions at the first two exons. Nonetheless, the transcript appears to be generated from 8 exons and does have the extended LIM repeat region (cysteine-rich domain) predicted by Gwen Freyd et al ( Freyd et al., WGB 11:1) and found in ceh-14 (Burglin et al ,WBG 11:2). So far we have seen no evidence of alternate splicing. Therefore, the use of alternately spliced transcripts in the different cells expressing mec-3 seems unlikely. To identify conserved elements within the coding sequence and in putative regulatory sequences in the mec-3 gene, we have isolated and sequenced 3.8kb of genomic DNA from the mec-3 gene of C. briggsae. This DNA contains about 1300bp 5' of the putative start site of transcription (as estimated from the C. elegans cDNAs) and DNA for all but last exon (the one encoding the acid-rich region). Both genes are highly homologous. In the LIM domain, 107 of 120 amino acids are identical; the greatest divergence (a string of 5 amino acids) occurs at the junction between the two LIM repeats. The homeodomain is also highly conserved: 57 out of 60 amino acids are identical. The similarity between the two proteins drops off between these two regions, including one region (corresponding to a C. elegans exon) in which only three of 30 amino acids are identical. This divergence and the finding by Burglin et al (WBG, 11:2) that the region between the LIM and homeo domains is small in the ceh-14 suggests that this linking region may not be functionally important. Upstream of the putative start for transcription, we have found four regions (ranging from 26bp to 42bp in length) that are conserved in both species and are thus potential cis-regulatory regions for this gene. No striking homologies are seen in other regions except at the putative beginning of transcription The possible binding sites for unc- 86 predicted by Finney et al. (Cell 55:757, 1988) in the introns of the mec-3 are not conserved. In preliminary experiments,we have tested whether any of these regions bind the mec-3 protein produced in E. coli (using the Studier system; J. Mol. Biol. 189:113-130, 1986) since the mec-3 lac Z studies suggested that mec-3 may display autoregulation. We find that the mec-3 protein does bind to DNA from one of these upstream regions in gel shift experiments. This result demonstrates that the mec-3 is a DNA binding protein and may regulate its own expression by binding to its own promoter. Future experiments will be directed towards further delineating these binding regions and to test their functional importance.