Three sets of findings stand out as being particularly significant. (1) In one series of experiments it was shown that axons from cerebellar cells (90 percent granule cells) in which Src has been knocked out extend axons about 1/2 as fast as wild type cells on L1 coated nitrocellulose. Cerebellar neurons from Fyn or Yes knockout mice have normal growth rates on L1. Conversely, NCAM dependent growth of Fyn- cerebellar cells grown on NCAM expressing fibroblasts is impaired, while growth of Src- and Yes- cells is unaffected. These findings suggest that normal L1 mediated outgrowth specifically requires Src function, while NCAM mediated outgrowth specifically requires Fyn- function. A second set of findings which amplify these results is the demonstration by immunoprecipitation and western blotting of complex formation between L1 and Src from a presumptive growth cone preparation, and between NCAM and Fyn from postnatal cerebellum and other sources. The demonstration of physical contact between these CAMS and specific kinases lends weight to the hypothesis that the kinases may be activated by Cam function. (3) Perhaps most interesting of all is the finding that olfactory epithelial axons in mice doubly mutant for Src- and Fyn- are less fasciculated and more scattered in their trajectories than axons in either single mutant alone or in wild type. These results support the hypothesis that Src and Fyn are required for normal pathfinding, but can partially compensate for one another. These accomplishments lead the field in rigor and in significance. The proposal has four specific aims. Work described in the first specific aim is a continuation of the current studies characterizing the association between Src, Fyn, L1 , and NCAM. Thus far there is good evidence that the kinases associate with their respective CAMS, but little information about whether this association is modulated by Cam-Cam binding, or whether such binding would activate the kinases. One objective is to determine if homophilic binding (or 'triggering') of the CAMS recruits the kinases into a complex, and if they thereby become activated. The basic strategy is similar to the work already accomplished, except now physical association will be compared in cells that are exposed to an appropriate Cam ligand to those that are not, and added to the assay will be a determination of kinase phosphorylation. A second objective is to identify the regions of Src and Fyn that mediate binding to CAMS. The strategy proposed is to use co-transfection of various kinase constructs along with the appropriate Cam in neuroblastoma cells, and then assay for physical associations by immunoprecipitation. A final objective is to further explore the possibility that L1 associates with Fyn by binding in cis to NCAM which in turn is bound to Fyn. The second specific aim proposes to examine in greater detail the possibility that Fak associates with CAMS and is activated when they bind their ligands. Again, the basic strategy is to use immunoprecipitation to characterize who is bound to whom, and antibodies against phosphotyrosine to determine if Fac is activated. An additional set of experiments is proposed to determine what other cytoskeletal components are organized by Cam-Cam binding using immunoprecipitation of Cam treated Cos cells cotransfected with NCAM and Fyn. In further experiments, beads coated with CAMS would be assayed for their ability to cluster Fak, Fyn, Tensin, and others.
In specific aim 3 it is proposed to examine the axonal trajectories of retinal ganglion cell axons in mice with L1, Fyn, or Src knockouts, or combinations of these knockouts. First, the overall structures of the retina, chiasm, and superior colliculus would be checked for gross defects. Retinal axons would be labeled with DiI and traced through these structures to ascertain if any defects in the trajectories are present.
In specific aim 4 it is proposed to further examine the expression pattern of a novel receptor kinase (REK, most closely related to Tyro) that is expressed in retinal ganglion cell axons and in the cerebellum. A second objective is to determine if REK can function as a homophilic adhesion molecule, as preliminary experiments suggest. It is also hoped that REK kinase activity can be activated by homophilic binding.
Showing the most recent 10 out of 39 publications
