Worm Breeder's Gazette 11(2): 45
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.
The lin-12 and glp-1 gene products are composed in part of repeats of an amino acid sequence motif that resembles epidermal growth factor. The lin-12 product contains 13 copies of the motif, and the glp-1 product contains 13 copies (Greenwald, 1985; Yochem et al., 1988; Yochem and Greenwald, 1989; Austin and Kimble, 1989; Priess and Fire, personal communication). We have attempted to isolate additional genes encoding proteins with EGF-like repeats. We reasoned that the products of such genes might be critical for development. For example, like lin-12 and glp-1, these products might be integral membrane proteins that mediate cell-cell interactions. Another possibility is that such products might be components of the extracellular matrix. By low-stringency hybridization with probes derived from both lin-12 and glp-1, we have been able to isolate several genomic clones that encode products with amino acid sequences that bear at least some resemblance to the EGF motif. The product predicted for one clone shares with EGF a characteristic array of six cysteine residues, although the other amino acids conserved in EGF-like repeats are absent. Clusters of tandem repeats of the array are separated by proline-rich stretches of amino acids. There are at least 23 copies of the array, and the product appears to lack a membrane-spanning domain. Based on partial DNA sequences, a second clone encodes a protein with tandem repeats of a motif based on 8 cysteine residues that closely resembles the EGF-like regions of the mammalian B1 and B2 chains of the extracellular matrix protein laminin. The predicted amino acid sequence is distinct from that of the B2 chain-like product of the unc-6 gene (Ishii et al., CSH 1989), and the clone has been positioned to a different chromosome than unc-6 (Sulston and Coulson, personal communication). The product of this clone may be the A or B1 chain of laminin or a novel molecule related in structure. A third clone encodes a product with a remarkable resemblance to the membrane bound LDL receptor of mammals and, to a lesser extent, the membrane-bound EGF precursor of mammals. The product has all of the major features of the LDL receptor including a ligand binding region composed of multiple copies of an acidic, non-EGF-like cysteine repeat. As in the LDL receptor, this region is followed by two tandem copies of the EGF-like motif, a stretch of about 275 amino acids, and then a single copy of the EGF motif. A similar region of two EGF-like repeats, a stretch of 275 amino acids, and a single EGF-like copy is also present in the EGF precursor, but the EGF precursor contains additional EGF-like repeats (one of which is EGF in precursor form) and lacks the LDL ligand binding region. The C. elegans product is expected to be much larger than the mammalian LDL receptor because the overall structure discussed above is repeated at least four times in tandem with some variability in the number of amino acids between repeat units. Only the COOH-terminal most repeat unit is followed by a potential membrane spanning domain. One obvious possibility is that this protein is a direct counterpart of the mammalian LDL receptor and mediates a general or cell-specific endocytosis of cholesterol in the worm. Alternatively, it could function in the endocytosis of yolk proteins that are synthesized in the gut and accumulate in oocytes. We have also considered the possibility that it is a membrane-bound signalling molecule or precursor of a secreted signal. The product could, of course, have a combination of functions. The gene has been placed to the right of unc-13 on LGI (Sulston and Coulson, personal communication).