......................................................................

NJIT Mathematical Biology Seminar

Tuesday, March 25, 2008, 4:00pm
Cullimore Hall 611
New Jersey Institute of Technology

......................................................................


Investigating How Feedback to a Descending Projection Neuron Influences Rhythmic Pattern Generation in the Target Network: A Modeling Study

Nickolas Kintos

Department of Mathematics, Fordham University


Abstract

Neural networks are often studied assuming a feed-forward architecture in which descending inputs from projection neurons initiate, terminate, or modify the output of a target network. Yet, in most such systems, feedback from the target network to descending inputs is pervasive, but its role is not well understood. We developed a mathematical model to understand how rhythmic feedback to a descending projection neuron influences the output of the gastric mill (chewing) motor circuit in the stomatogastric ganglion (STG) of the crab, C. borealis.

In the biological system, stimulating the projection neuron MCN1 elicits a gastric mill rhythm (GMR) in vitro by activating the half-center between the reciprocally inhibitory neurons LG and INT1 in the STG. In contrast, co-stimulation of MCN1 and the projection neuron CPN2 elicits a distinct GMR. The STG terminals of CPN2 excite the LG neuron, while the CPN2 soma is inhibited by a feedback synapse from INT1. Previous experiments indicated that the MCN1/CPN2-GMR (but not the MCN1-GMR) persists without the inhibitory synapse from INT1 to LG; and thus, without INT1-LG reciprocal inhibition.

Using a reduced mathematical model in parallel with a more biophysically-realistic model, we elucidate the mechanisms that enable the MCN1/CPN2-GMR (but not the MCN1-GMR) to persist without INT1-LG reciprocal inhibition. First, we show that both GMRs are driven by modulatory input from MCN1 to the LG neuron. However, the MCN1/CPN2-GMR exhibits a slower frequency due to CPN2 actions on LG. Next, we show that INT1 feedback to CPN2 allows for the MCN1/CPN2-GMR to persist without the INT1 to LG inhibitory synapse. In contrast, the MCN1-GMR, which does not involve feedback to CPN2, is disrupted without the INT1 to LG synapse. Interestingly, we found that although INT1-LG reciprocal inhibition is not required to elicit the MCN1/CPN2-GMR, it is required for the frequency of this GMR to be regulated by a previously documented local input from the pyloric circuit. We conclude that feedback to projection neurons can move the locus of pattern generation from a half-center oscillator to an excitation-feedback circuit, which in turn can alter inter-circuit interactions.

This is joint work with Farzan Nadim and Amitabha K. Bose (NJIT), and Michael P. Nusbaum (U Penn School of Medicine).




Last Modified: Nov 28, 2007
Horacio G. Rotstein
h o r a c i o @ n j i t . e d u
Last modified: Wed Mar 5 10:07:41 EST 2008