TRANSLATING BIOLOGICAL MECHANISMS INTO COMPUTATIONAL MODELS OF THE REACTIVE GLIO- SIS.

AUZMENDI JA; L. MOFFATT; RAMOS AJ

Buenos Aires

Congreso; 2° Reunión Conjunta de Biociencias; 2017

We developed a Bayesian Computational Model (BCM) of the re-active gliósis (RA), a typical response of the glial cells after brain injury characterized by a change in the morphology of asctrocytes in a graded maner. The mechanism by it occurred is still unclear. Experimental data showed a controversy about the mechanism of this propagation, whether the initial spatial distribution of molecules associated to damage (DAMP) was determinant or if a secondary secretion of soluble mediators (SM) like interleukins was needed the microglia played a major role. The BCM is fed with experimental data from the GFAP immuno-histochemistry of the brain cortex with a pial disruption vessel and with the prior distribution of the model parameters. It returns their posterior distribution and the model evidence, which allows compa-ring alternative propagation mechanisms. The model divided the cortex in segments and in each segment contained a concentration of each kind of cell, a concentration of DAMPs and of SM. Each cell had a number of DAMP and SM-recep-tors and the molecules could be free or bound. The receptors acted as catalytic centers, each bound molecule would be degraded. We modeled the astrocyte morphological change as a Markovian Pro-cess that depended on the proportion of bounded receptors of each kind. The model uses partial differential equations for the diffusion of the molecules, Michaelis Menten the instantaneous proportion of bound molecules and the changes in the morphological states of the astrocytes and microglia. The soluble molecules where liberated by the cells depending on the state of the cell and the proportion of bound receptors. The BMC analysis indicated that there was strong evidence for models where the microglia drives the propagation of the gliosis.