Worm Breeder's Gazette 11(2): 44
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.
We are interested in understanding how cell-cell interactions specify cell fates during development. With the ultimate goal of using a 'reverse genetic' approach to determine their biological function, we have begun to molecularly clone C. elegans protein- tyrosine kinase genes. The receptor-type tyrosine kinases (RTKs) are especially interesting because several members have been implicated in developmental decisions which require intercellular communication in Drosophila and vertebrates. A degenerate oligonucleotide probe corresponding to an amino acid motif highly conserved among RTKs (WMAPES in subdomain VIII, see Hanks et al., 1988) was used to screen a C. elegans genomic library. As opposed to the consensus target chosen by Kamb et al. (1989), which was based on the sequences of cytoplasmic tyrosine kinases (which lack transmembrane and extracellular domains), our target favors the isolation of RTKs. Sequence analysis of over 5.2 kb of the genomic DNA clone GS#L5 and corresponding cDNAs has identified two tandem genes, ctk(L5a) and ctk( L5b), which appear to encode novel transmembrane tyrosine kinases. These two genes possess nearly identical exon/intron boundaries, suggesting that they arose by gene duplication. Both possess all the hallmarks of the tyrosine kinases, including the consensus ATP-binding domain and a potential autophosphorylation site. The homology to other tyrosine kinases is limited to the kinase catalytic domain. They also contain a potential signal peptide sequence and membrane- spanning region, and are therefore probably inserted into the plasma membrane. However, the putative extracellular domain of ctk(L5a) and ctk(L5b) (20-30 amino acids) is extremely short relative to other RTKs (>500 amino acids) (see Figure). One other gene encoding a transmembrane tyrosine kinase lacking an extracellular domain has been identified in mice (Ben-Neriah and Bauskin, 1988). [See Figure 1] Possibly these 'short RTKs' function by interacting (via their extracellular, membrane-spanning and/or juxtamembrane sequences) with another protein which provides an extracellular ligand binding activity. These amino-terminal sequences are not well conserved between the two ctk gene products, in contrast to the striking homology exhibited in the kinase domain and two other regions, the kinase insert and C-terminal domains. The latter domains have been postulated to be involved in substrate specificity and/or modulating catalytic activity. One intriguing possibility is that these two ctk proteins interact with distinct ligand-binding proteins but phosphorylate common substrates. Physical mapping of GS#L5 (by Alan Coulson and John Sulston) and of three deficiency breakpoints (by Caroline Shamu and Jane Mendel) indicate that ctk(L5a) and ctk(L5b) are located between let-23 and spe- 2 on LGII, We are physically mapping additional deficiencies in this region to further refine their genetic map position. Interestingly, fingerprint analysis indicates that another clone we isolated containing a tyrosine kinase gene overlaps GS#L5. Perhaps this region contains a cluster of tyrosine kinase genes.