The long-term goal of our group is to understand the quantitative mechanisms by which the cell signaling system controls the neuronal function. I employ mathematical and computational approaches to understand the synaptic functions at the molecular level. Specifically, I am interested in how post-synaptic Ca2+ signals activate calmodulin-dependent signaling systems and how downstream targets such as CaM Kinase II and calcineurin (PP2B) induce long-term synaptic plasticity in CA1 pyramidal neurons. A serious challenge in the understanding of the nature of the Ca2+ signal is the inadequate spatial and temporal resolution of currently available experimental approaches to measure Ca2+. Computer simulations are presently the only tractable means of investigating this critical aspect of synaptic physiology. Over the past five years, we have developed and extensively validated an efficient algorithm that accurately simulates molecular diffusion and chemical reaction in the synapse at the single molecule level. I also developed mathematical theories of chemical reaction network of postsynaptic spine. These mathematical and computational approaches are built upon and are tested against experimental data.
As mentioned earlier, we have established a series of simulation algorithms for chemical reaction and molecular diffusion inside the cell. The user-friendly simulation environment (Cellular Dynamics Simulator: CDS) (Initializer and Visualizer of the simulator) shown in Fig. 1 (Panel A and B) allows us to retrieve and implement experimental data to create a model spine.
Fig. 1A is a snapshot of the Intializer with a model synapse loaded (pre- and post- synaptic compartment). The right and bottom panels show a 2D drawing of the simulation space from a top view, side view and front view. The frame at the top left displays the compartments and elements in the model and their properties. Although we model exclusively the post-synaptic compartment in this project, for illustration purposes, we show two compartments (pre- and post- synaptic) facing each other with the empty volume between representing the cleft. Elements distributed throughout these compartments in this example include static obstacles, CaM, Ca2+/calmodulin-dependent protein kinase II, Ca2+ pumps, vesicles, and glutamate receptors. These elements quickly crowd the viewing area and this illustrates the need for the Initializer to have algorithms for automatic distribution of elements.
Fig. 1B is the Visualizer window showing a rendering of the simulation described in Fig. 1A. Meshes may be rendered opaque or with a variable degree of transparency and can be solid or wireframe. Colors can be assigned by surface, by state and or surface state. Colors can be adjusted by element, by group or by individual instances. Properties applied to an element are automatically applied to all instances of that element. A variety of playback controls allow precise rates of animation. The view can be panned, rotated and zoomed by using the three mouse buttons.
This user-friendly Initializer/Visualizer environment will allow us to construct models in an automated fashion and visualize and confirm the model environment before starting simulations or even run a short test simulation. For example, Fig. 1C shows the capability of the simulator to import the experimentally obtained geometry of a 3D reconstruction of a section of dendrite (the left hand side) and modify these structures (e.g., making multiple copies, connecting them to other artificially created objects). See our CDS website (http://nba.uth.tmc.edu/cds/) for more details.
Kubota, Y., and Bower, J.M. (2001) Transient versus Asymptotic Dynamics of CaM Kinase II: Possible Roles of Phosphatase Journal of Computational Neuroscience, 11: 263-279.
Bradshaw, J.M., Kubota, Y., Meyer, T., and Schulman, H. (2003) An Ultrasensitive CaMKII-PP1 Switch Facilitates Specificity in Postsynaptic Signaling. Proc. Natl. Acad. Sci. 100: 10512-10517.
Kubota, Y., Gaertner, T.R., Putkey. J.A., and Waxham. N. (2005) A novel Monte Carlo simulation for Molecular Interactions and Diffusion in Postsynaptic Spines Neurocomputing, 65: 595-602.
Sanabria, H., Kubota, Y., and Waxham, MN. (2007) Multiple diffusion mechanisms due to nanostructuring in crowded environments. Biophys. J. 92: 313-322.
Kubota, Y. Putkey, J.A., Waxham, M.N. (2007) Neurogranin controls the spatiotemporal pattern of postsynaptic Ca2+/CaM signaling. Biophys. J. 93: 1-12.
Kubota, Y. Putkey, J.A., Shouval H.Z., Waxham, M.N. (2008) IQ-Motif Proteins Influence Intracellular Free Ca2+ in Hippocampal Neurons through Their Interactions with Calmodulin. J. Neurophysiology 99: 264-276, 2008.
Search PubMed for additional articles.