We seek to answer two questions: how do neurons become connected during development, and why do they become disconnected during neurodegenerative disease? Over the past year, we have made three significant advances in our studies of the development of neural connectivity. First, we have used genome-wide expression profiling to identify downstream targets for the transcription factor Lola, an axon guidance regulator that is required for all Notch-dependent axon patterning decisions characterized to date. Second, we have found that the mechanisms and principles we have been studying in axon growth and guidance also control dendrite arborization. Third, we have developed FRET-based biosensors to report the in vivo activity of two major outputs of the Abl signaling pathway, Abl kinase and Rac GTPase. We will briefly describe these, in turn. The gene lola encodes more than 20 isoforms of a zinc finger transcription factor by alternative splicing. These forms heterodimerize to form probably well over a hundred protein species with distinct DNA-binding specificities. Our previous studies showed that lola is required for Notch-dependent axon guidance decisions, including longitudinal axon growth and midline crossing regulation in the CNS and development of the ISNb motonerve in the PNS. We have now performed genome-wide expression profiling of lola mutant embryos using microarrays. Gene ontology analysis verified regulation of a number of Notch pathway genes downstream of lola. It also revealed regulation of a variety of genes associated with axon growth, cell migration, signal transduction and cytoskeletal structure. In addition to genes known to act in Notch-dependent guidance processes, a substantial number of these genes were either anonymous ORFs without demonstrated function in any system, or were genes known to act in other guidance processes but not previously shown to interact with Notch. These experiments therefore significantly expand the window of genes available for analysis in our efforts to understand Notch-dependent neural wiring. Of particular interest was our discovery that a key aspect of lola-dependent axon growth is to suppress expression of the conserved actin nucleation factor, Spire. This observation accords well with evidence we published last year that emphasized the importance of balancing different kinds of actin structures in the growth cone to achieve effective axon growth and faithful axon guidance. We wanted to test whether the mechanisms we have described in axons also apply to dendritic development. Two lines of experiment from the past year support that hypothesis, but with interesting differences. First, we examined the role of Abl signaling components, particularly Rac GTPase and the Rac GEF, Trio, in dendritic arborization. We found that increased or decreased function of these genes alters dendritic arborization in vivo, and that these manipulations interact genetically with mutations of Abl itself, as for axons. The nature of the interaction is evidently different in dendrites, however: mutations that give similar phenotypes and interact synergistically in axons were found to give opposite effects in dendrites and interact antagonistically. The mechanistic basis for this difference is not yet clear. We also examined the role of Lola in dendrites, and its interaction with Spire (in collaboration with D. van Meyel, McGill Univ). Again, we found dendritic phenotypes consistent with what we had demonstrated previously in axons, and as in axons, we found genetic and molecular evidence that a key role of Lola in dendrogenesis is to suppress Spire function (ms submitted). These data support the idea that the molecules and principles governing dendritic arborization are closely akin, but not identical, to those acting in axons. The next phase of our studies will require in vivo monitoring of signal transduction in growth cones to correlate genetic inputs with cytoskeletal dynamics. To this end we have developed FRET-based in vivo biosensors that report the two key outputs of the Abl signaling pathway. One reports Rac activity using a modified form of the Raichu-Rac reporter first developed for mammalian cells. This reporter is identical to one developed and validated independently for flies by D. Montell. Our evidence, in concordance with hers, shows that FRET activity is increased by stimulation of Rac activity and suppressed by expression of a dominant-negative Rac. The other is a reporter for Abl tyrosine kinase activity, and was originally developed for mammalian Abl. We have verified that we can detect the Abl FRET signal in fly cells, that it is stimulated by expression of activated Abl and suppressed in an Abl mutant, and also that it is suppressed by administration of the Abl kinase inhibitor Gleevec. This gives us the reagents we need to detect the crosstalk between Abl and Trio directly, and thereby to validate or falsify the hypothesis that these act as parallel outputs of the Abl signaling network. It also gives us the reagents to correlate Abl kinase activity and Rac GTPase activity with specific aspects of growth cone morphology and motility in vivo, which is a major goal of our overall research program.
Showing the most recent 10 out of 13 publications
