We are interested in both the mechanisms of neuronal plasticity and its impairment in neurodegeneration. One facet of plasticity is the regulation of mRNA translocation and translation in dendrites. RNAs are not uniformly distributed in neurons, and a subset of mRNAs that extend into dendrites appear to position their translation products strategically to implement the morphological changes associated with activity-related changes in synapses. Emerging work has attempted to link the translational regulation of these dendritic mRNAs to synaptic activity. The population of RNAs which segregate to the dendrite create a specialized locale possibly capable of implementing activity-related structural changes, including dendritic spine morphogenesis associated with enduring long-term potentiation. In dendrites, mRNAs are present as granules. We have observed translocation of RNA granules in neurons along microtubules and have engineered a nucleic acid-peptide complex capable of directly visualizing in living neurons the translocation of a 5' untranslated RNA sequence complexed to green fluoresecent protein.

    Although synaptic activation can induce translation, how activation is coupled to translation of specific mRNAs is poorly understood. In contrast to local translation control, the delivery of new mRNAs in granules to the dendrite is a slower means of altering local protein composition. Using RNA sedimentation techniques we have isolated and characterized the RNA granule as a macromolecular control site where specific mRNAs are held in translational arrest until stimulated. Available projects are directed at identification of components within these granules, characterization of their interaction with microtubules, identification of specific proteins that interact with highly segregated neuronal RNAs, and the direct visualization of RNA transport. RNAs that sediment with a somewhat lower mass than granules are polysomes. In this fraction are a remarkable variety of microRNAs potentially capable of regulating translation locally.

    The basis for experience-dependent modification of neural circuitry involves changes in both the efficacy of existing synapses and the patterning of new anatomical connections such as the elaboration of motile filopodial, new synapses and spines. We cloned a protein called delta-catenin which is shared by both adherens junctions and synapses, two structures with an intertwined function and evolutionary history. Spine formation and filopodial elaboration must involve adhesive changes for neurite outgrowths to penetrate the neuropil and delimit portions of the newly extended membrane for a synapse of a defined composition. In the adherens junction, delta-catenin binds to classical cadherin as a neuronal specific Arm-repeat family member. Among the Arm-repeat family members, a sub-family that includes delta-catenin and the prototypical member, p120ctn, is distinguished by the presence of ten Arm-repeats and binding to the juxtamembrane region of the classical cadherins. Through its PDZ binding domain, delta-catenin is linked to the PDZ domains of several synaptic proteins. Delta-catenin is ideally positioned to bridge and coordinate activity-related changes in the synapse with changes in adhesion of the post-synaptic membrane.

    In parallel to this work are studies directed at the underlying cellular mechanisms by which plasticity is lost in the course of neurodegeneration. The microtubule-associated protein tau is the focus of these studies and the projects focus principally on the changes that neurons undergo as tau becomes vulnerable to the formation of aggregates and inclusions. Most recently we have completed a screen for compounds that are likely to inhibit the phosphorylation of tau, the likely first step in its transition toward the formation of intra-cellular aggregates.