We propose to develop mathematical network models of the mammalian neocortex in order to elucidate the genesis of event-related magnetic fields (ERFs) and electrical potentials (ERPs). We have previously provided a fundamental account of the genesis of ERF and ERP for the guinea pig hippocampus using a mathematical model based on RD Traub's 1991 model. We are proposing to extend this work to the neocortex in order to eventually help interpret magnetoencephalography (MEG) and electroencephalography (EEG) signals from the human brain. In this R03 application, we will limit our project to four specific aims that address a set of initial problems in developing mathematical models of the neocortex for interpretation of MEG and EEG.
Aim1 : We will first develop single-cell compartment models of the principal neurons of the neocortex based on Mainen's model. Our models will consist of the pyramidal cells in layers II/III and layer V, the spiny stellate cells in layer IIV and the inhibitory neurons (aspiny stellate cells) in layer III, based on real anatomical data and updated channel kinetics and distributions. The transmembrane potentials and cell properties (e.g. IV relation) will be 'computed and compared with published experimental data to validate the models.
Aim 2 : We will compute the intracellular current distribution in each cell type to infer the contributions of cells to MEG and EEG signal. The net current dipole moment in each cell will be computed at distances far from the cells in order to estimate the MEG and EEG signals per single neurons. We expect that the MEG and EEG signal will be primarily due to the pyramidal cells as conventionally assumed. Our new contribution will consist of quantitative estimates of the current dipole moments for realistically shaped cell models.
Aim 3 : We will determine the contributions of the distal and proximal apical dendrites, the perisomal region and basal dendrites to the ERF and ERP since they can be readily assessed with compartment models. This will be done for pyramidal cells with repetitive firing and bursting mode of firing. We predict that the currents in the perisomal region will dominate MEG and EEG signals for cells with repetitive firing and the currents in the distal as well as proximal trunk area of the apical dendrites to dominate for bursting cells, while the basal dendrites will have relatively small contributions due to their geometry.
Aim 4 : Once we develop and test single cell models in year 1, we will connect all cell types in a network, compute the ERF and ERP that would be produced by thalamocortical inputs into layer IV, and compare them with experimental waveforms.
| Murakami, Shingo; Okada, Yoshio (2006) Contributions of principal neocortical neurons to magnetoencephalography and electroencephalography signals. J Physiol 575:925-36 |
