Neal Waxham, M.D.
Department of Neurobiology and Anatomy
University of Texas Medical School
Houston, Texas
November 11, 2002

Constraints on Neuronal Signaling through the Calmodulin and CaM-kinase II Pathways Based on Three- Dimensional Structural Analysis and Live Cell Imaging

Neuronal plasticity is governed in part by well-orchestrated intracellular signaling pathways. The orchestration involves the where and when individual molecules become activated. Ca2+/calmodulin-dependent protein kinase II (CaM-kinase II) is the most abundant protein kinase in nerve cells and its role in regulating neuronal plasticity is well documented.

Inhibiting the kinase through either genetic knockout strategies or pharmacological treatments negatively impacts one of the widest- studied models of neuronal plasticity, long-term potentiation. A unique feature of CaM-kinase II is that the molecule catalyzes the covalent modification of itself through an autophosphorylation event and this autophosphorylation impacts the functionality of the enzyme. Each CaM-kinase II molecule can therefore exist in different stable states adding significant, but fascinating, complexity to the enzyme population within neurons.

CaM-kinase II is a complex of multiple subunits that forms an oligomeric structure and it is the interactions between subunits of a holoenzyme that is proposed to underlie the autophosphorylation mechanism. In addition, the multimeric structure of the kinase provides numerous opportunities for protein-protein interactions either between CaM-kinase II and other proteins, such as the NMDA receptor, or possibly between CaM-kinase II holoenzymes. The hypothesis is put forth that the combined molecular features of CaM- kinase II subserve a Ca2+/calmodulin regulated scaffolding function leading to the additional idea that CaM-kinase II is a structural protein.

The talk addressed several of these issues. First, work describing the solution of the three-dimensional structure of the kinase at a resolution of approximately 27 angstroms was presented. CaM-kinase II is a dodecameric (12-subunit) molecule that has dimensions of ~20 nm x 20 nm. The 12 subunits are arranged in a tail-to-tail fashion producing a molecule with a dense central core and catalytic domains extending away from the core on short stalks. A comparison of the three-dimensional structure of each CaM-kinase II gene family member (a, b, g and d) reveals significant homology in overall architecture. Thus the differences in the biology ascribed to these different isoforms cannot be due to differences in their overall architecture. The distances between and rigidity of the catalytic domains suggested by these three-dimensional reconstructions leaves open the question, from a structural viewpoint, how two subunits within one holoenzyme can be brought into contact with each other to autophosohorylate.

With the three-dimensional structure in hand it is now possible to paint CaM-kinase II into electron micrographs of synaptic contacts as a means of providing a visual framework to constrain thinking about the molecules associated with the synaptic architecture. The second half of the talk addressed the mechanism of CaM-kinase II self-ssociation in living cells. The multimeric nature of the holoenzymes provides ample opportunities for interactions that lead to macromolecular assemblies of CaM-kinase II molecules. Live cell imaging indicates that enzyme clusters assemble and disassemble on the minute time scale in a stimulus- dependent manner. Binding of the activator Ca2+/CaM is essential for the expression of this property as is the holoenzyme structure. One possibility is that self-assembly of CaM-kinase II holoenzymes may provide synaptic contacts with a previously unidentified Ca2+/CaM-regulated structural element to help stabilize synaptic contacts. Another consequence of self- association would be to limit the diffusion of CaM-kinase II away from its point of activation, tagging that spot (i.e., a particular synapse) with activated CaM-kinase II molecules.

Overall, the hope of the presentation is to provide new data from CaM- kinase II structure-function studies that will stimulate novel hypotheses about the role of this enzyme in the regulation of neuronal function. Incorporating temporal and spatial changes in the kinase's localization and providing a spatial framework for possibly relating the enzyme's function at synaptic contacts is a further goal.