Decoding time-varying calcium signals by the postsynaptic biochemical network: Computer simulations of molecular kinetics

Computer simulations and mathematical analyses were applied to a study of t he molecular signaling mechanisms that underlie post-synaptic responses to synaptic activation. In this report three models are described: a new detai led kinetic model examining possible relationships between spike frequency modulation and calmodulin-dependent kinase II (CaMKII) activity; a second p ost-synaptic biochemical network model involving CaMKII, calmodulin, calcin eurin, adenylate cyclase, phosphodiesterase, cAMP-dependent protein kinase, protein phosphatase 1, inhibitor 1, Ras protein, synaptic Ras-GTPase activ ating protein (p135 Syn-GAP), mitogen-activated protein kinase and other re gulatory factors; and a biophysical model which combines the post-synaptic biochemical network with known ionic mechanisms in dendritic spines. The ki netic model of CaMKII was first shown to replicate experimental evidence by Koninck and Schulman (Science 279 (1998) 227-230.) that CaMKII can decode the frequency of post-synaptic Ca2+ spikes induced by pre-synaptic activity into distinct amounts of kinase activity. The model was then used to sugge st that the functional role of CaMKII may extend beyond simple frequency de coding to allow differential responses to different temporal patterns of pr e-synaptic activation. Perturbation analyses of model suggests how CaMKII a ctivity reflects time-varying Ca2+ signals and also suggests that the behav ior of this entire biophysical pathway depends on the total amount of CaMKI I and other postsynaptic proteins. Our mathematical model provides a tool t o quantitatively estimate the role of protein synthesis at the spine head i n modulating synaptic function. (C) 1999 Elsevier Science B.V. All rights r eserved.