The cadherin superfamily of molecules has been implicated in many cellular developmental processes including adhesion, proliferation and morphogenesis. However, the mechanisms by which cadherins affect neuronal development remain enigmatic, particularly with respect to axon guidance and fasciculation, and the formation and maintenance of synapses. The C. elegans cadherin superfamily is comprised of 12 loci, yet only the
hmr-1 locus is capable of encoding a "classical" cadherin, as defined by conserved regions in the cytoplasmic C- terminus responsible for binding to a set of cytoplasmic proteins, the catenins (1,2). In contrast, Drosophila possesses two distinct classical cadherin loci, encoding DE- and DN-cadherin. DE- cadherin is expressed in epithelial tissues, and is required for the maintenance of adherens junctions, whereas DN-cadherin is expressed in neuronal tissues, and is involved in fasciculation of axon bundles and pathfinding (3,4,5). The
hmr-1 locus has been shown previously to encode an epithelial cadherin, HMR-1A, which is essential for epidermal morphogenesis during embryonic development (2). However, we have demonstrated recently that the
hmr-1 locus can also encode a second isoform, HMR-1B, via an unusual mechanism involving alternative splicing and an alternative promoter (6). Recent information from the C. briggsae genome project suggests that this mechanism is also conserved in the C. briggsae
hmr-1 locus. In contrast to the mainly epithelial HMR-1A isoform, HMR-1B is expressed exclusively in the nervous system in a particular subset of neurons, and shows a striking degree of sequence similarity to DN-cadherin. We have selectively disrupted the functions of HMR-1B using a hmr- 1 null allele in conjunction with a HMR-1A rescuing construct, and a "snapback" RNAi construct designed to specifically target HMR-1B. Animals lacking HMR-1B function assayed by either of these approaches display defects in motorneuron pathfinding and ventral nerve cord fasciculation. We have also observed similar defects in animals carrying a null mutation in the
hmp-1 _-catenin gene that have been rescued for epidermal (but not neuronal) HMP-1 function. We are currently using this epidermally-rescued
hmp-1 background to identify molecules that act in concert with the cadherin-catenin complex to regulate nervous system development in C. elegans.