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

NJIT Mathematical Biology Seminar

Note special date and time:
Friday, September 29, 2006, 1:30pm
Cullimore Hall 611
New Jersey Institute of Technology

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

Modeling Projection Neuron and Neuromodulatory Effects on a Rhythmic Neural Network.

Nickolas Kintos

Thesis Defense

Department of Mathematical Sciences
NJIT

Projection neurons shape the activity of many neural networks. In particular, neuromodulatory substances, which are often released by projection neurons, alter the cellular and/or synaptic properties within a target network. However, neural networks in turn influence projection neuron input via synaptic feedback. This dissertation uses mathematical and biophysically-realistic modeling to investigate these issues in the gastric mill (chewing) motor network of the crab, Cancer borealis. The projection neuron MCN1 elicits a gastric mill rhythm in which the LG neuron and INT1 burst in anti-phase due to their reciprocal inhibition. However, bath application of the neuromodulator PK elicits a similar gastric mill rhythm in the absence of MCN1 participation; yet, the mechanism that underlies the PK-elicited rhythm is unknown. This dissertation develops a 2-dimensional model that is used to propose three potential mechanisms by which PK can elicit a similar gastric mill rhythm. The network dynamics of the MCN1-elicited and PK-elicited rhythms are also compared using geometrical properties in the phase plane. Next, the two gastric mill rhythms are compared using a more biophysically-realistic model. Presynaptic inhibition of MCN1 is necessary for coordinating network activity during the MCN1-elicited rhythm. In contrast, the PK-elicited rhythm is shown to be coordinated by a synapse that is not functional during the MCN1-elicited rhythm.

Next, the gastric mill rhythm that is elicited by two coactive projection neurons (MCN1 and CPN2) is studied. A 2-dimensional model is used to compare the network dynamics of the MCN1-elicited and MCN1/CPN2-elicited gastric mill rhythms via geometrical properties in the phase plane. While the MCN1-elicited rhythm requires the presence of reciprocal inhibition between INT1 and the LG neuron, the MCN1/CPN2-elicited rhythm persists in the absence of this reciprocal inhibition, due to an inhibitory feedback synapse from INT1 to CPN2 that changes the locus of coordination in the gastric mill rhythm. Next, the effect of a second feedback synapse, from the AB neuron to MCN1, is shown to change the motor pattern of the MCN1- and MCN1/CPN2-elicited rhythms. Finally, a third MCN1/CPN2-elicited rhythm is studied where the AB to MCN1 feedback synapse only affects the LG burst phase of the rhythm