DETERMINATION OF HIDDEN PARAMETERS OF SIGNAL TRANSDUCTION IN SYNAPSES

Maryna Hliatsevich1, Pavel Bulai1, Taras Pitlik1

1 Department of Biophysics, Physics Faculty, Belarusian State University, Belarus

Mathematical modelling of signal transduction between neuronal cells in the brain is one of the most challenging problems in the modern neuroscience. The characteristics of this signal transduction (synaptic transmission, ST) determine the work of the neural networks and the brain as a whole. The importance of complete understanding ST has led to development of a variety of different models describing this process and each model has its own limitations and area of practical application.

Previously we have reported the deterministic model of ST which represents the system of the ordinary differential equations [1]. Each equation describes the content changes for the following biochemical participants: gates of ionic channels, calcium ions, calcium sensor – synaptotagmin protein, different pools of synaptic vesicles, proteins of exocytosis machinery, neurotransmitter – glutamate, states of neurotransmitter receptors. The Cauchy problem for the given system is solved numerically and this model can simulate the synaptic current provided by the channels of ionotropic neurotransmitter receptor.

The synaptic current can be regarded as a principal signal in ST. However experimentally ST is usually investigated with the help of microelectrode technique. This technique implies the electrical stimulation of neuronal ensemble in the definite area (CA3) of hippocampal slice. The stimulation signal undergoes ST from neuronal ensemble in CA3 to another one, placed in CA1. Synaptic currents generate action potentials on the dendrite membrane of neurons in CA1 area. The summed signal from the neuronal ensemble is registered as field excitatory postsynaptic potential (fEPSP) and one may choose conditions of stimulation when fEPSP is substantially determined by synaptic currents.

We chose the described method and conditions for experimental verification of our suggested model. To make verification procedure relevant we have taken into account the mechanism of fEPSP formation and constructed the equivalent scheme of connection between neuron and microelectrode. The synaptic current calculated in previously described model of ST was regarded as source of current in the elaborated scheme. Such approach enabled us to get good coincidence between forms of measured and simulated fEPSP.

Suggested model of ST includes many parameters that determine the result of model simulation. These parameters are rate constants of biochemical reactions, rates of binding and unbinding of agonist from receptors, surface densities of regarded proteins and so on. The majority of parameters' values can be found from the experimental measurements, but part of parameters can not be measured directly (hidden parameters). Furthermore, experimental determination of some parameters is technically complicated and does not give accurate result. Our model can suggest another way of parameters' values establishment: the exact approximation of experimentally measured fEPSP form by results of model simulation can be achieved if only the model parameters are adequate. Therefore we aimed was to find parameters' values that give the best coincidence between experimental and modelling results.

The standard deviation of simulated signal from experimentally registered one was regarded as a measure of coincidence and we searched for parameters' values that give the minimum of this objective function. The parameters' values determined with the help of such approach are listed in the Table 1.

Table 1. Determined values of model's parameters.
parameterdetermined value
the bouton surface to its volume ratio12 µm-1
apparent maximal conductance of single calcium channel8 pS
forward rate constant for active SNARE complex formation1.1·10-3 ms-1
forward rate constant for synaptic vesicles transition from docked to primed pool5.0·10-6 ms-1
time constant for inactivation of synaptic vesicles3.3 ms
rate constant of glutamate clearance from synaptic cleft45 ms-1
rate constant of AMPAr's channel opening1.7 mM-1·ms-1
rate constant of AMPAr's channel closing0.45 ms-1

At present the main goal of our investigation is comparison of simulated effects of ST inhibitors and experimental ST inhibition. This approach is supposed to make possible the establishment or updating more biochemical parameters of ST.

Suggested model provides ways of finding the set of parameters with the help of only one experimental technique and can be useful for the approbation of neuroactive drug candidate.

Acknowledgments: This work was supported by Belarusian Republican Foundation for Fundamental Research.


[1] M.A. Hliatsevich, P.M. Bulai, T.N. Pitlik et al., Design of Deterministic Model of Signal Transduction between Neuronal Cells. Mathematical Modelling and Analysis 20, 76-93, (2015).