Fine Network Structure of Mossy Fiber Sprouting Is a Major Determinant of Dentate Hyperexcitability
Abstract number :
BS.24
Submission category :
Translational Research-Basic Mechanisms
Year :
2006
Submission ID :
6132
Source :
www.aesnet.org
Presentation date :
12/1/2006 12:00:00 AM
Published date :
Nov 30, 2006, 06:00 AM
Authors :
Robert J. Morgan, and Ivan Soltesz
Following many insults, the mammalian dentate gyrus undergoes dramatic structural rearrangement. Structural changes include the loss of hilar interneurons and mossy cells as well as the sprouting of recurrent connections between granule cells. The functional relevance of these changes has been widely debated, but much evidence points to an increase in recurrent excitation that can predispose the neuronal microcircuit to seizure-like activity. Although the phenomenon of mossy fiber (granule cell axon) sprouting has been clearly defined, the principles of organization of the resulting recurrent granule cell network are not understood. Here we demonstrate the functional effects of varied organizational principles on a model dentate gyrus network[apos]s predisposition to hyperactivity., The model network contained biophysically realistic, multicompartmental single cell models of both excitatory and inhibitory neurons connected by weighted synapses, as published previously (Santhakumar et al., 2005). Four cell types were included: granule cells (50000), mossy cells (1500), dendritically projecting HIPP cells (600), and somatically projecting, fast spiking basket cells (500), for a total of 52600 single cells. Connectivity between cell types was determined probabalistically based on the anatomically observed axonal extents of the various cell types. Simulations were carried out with 50% hilar interneuron and mossy cell loss as well as 50% of the maximal sprouting of mossy fibers. Structural changes were introduced into the sprouted granule cell network by altering the connectivity rules while maintaining the same number of total connections throughout all simulations., We tested various structural organizational principles for their effects on network hyperexcitability. First, we implemented a Hebbian-like connectivity rule, where the connection probability of two granule cells increased in proportion to the number of shared presynaptic cells. This had no effect on the observed hyperexcitability of the network. Similarly, overrepresentation of numerous different three neuron motifs (connection patterns between three neurons in the network, such as feedforward or feedback loops) had no effect on network hyperexcitability. However, when we introduced a small number ([lt]5%) of highly interconnected granule cells (with approximately 5x the average number of connections) to serve as hubs, the network displayed significantly increased levels of self-propagating activity., These data indicate that the presence of a small population of highly interconnected granule cell hubs could promote fast, efficient, and recurrent propagation of activity through the dentate gyrus circuit, thereby creating a pathological entity that is predisposed to seizing., (Supported by NIH (NS35915 to IS) and UCI MSTP (RJM).)
Translational Research