Modulation of feeding and its role in appetitive memory consolidation in Lymnaea stagnalis

Feeding in the pond snail (Lymnaea stagnalis) is heavily regulated by modulatory neurons throughout the brain. Following appetitive conditioning many of these identified neurons show changes in electrophysiological properties that contribute to the conditioned feeding response. The aim of this thesis was to further characterise modulatory feeding neurons and the neural plasticity that occurs within them during appetitive memory consolidation. Previous research had shown that the cerebral giant cell (CGC) is persistently depolarised 16-24 hours following appetitive conditioning and onwards. A Hodgkin-Huxley model of the CGC predicted this nonsynaptic plasticity to occur via an increase in the maximum conductance of the persistent sodium current (INaP), delayed rectifier current (IK) and high voltage activated calcium current (IHVA). To test this hypothesis these three modelled conductances were introduced into the membrane of the CGC of untrained snails, using dynamic clamp. Little effect on electrophysiological properties was initially seen due to an unexpectedly large cell surface area. To correct for this underestimation the response of the membrane to sine wave current injection was used to calculate the total cell surface area prior to dynamic clamp. When correctly scaled conductances were applied to the cell, there was still no significant change in the resting membrane potential (RMP). This suggests that the changes in ionic conductances causing CGC nonsynaptic plasticity during long term memory are more complex than previously realised and may involve other membrane conductances. A simpler attempt to depolarise the CGC with pattern clamp was able to cause significant RMP depolarisation, but also led to ectopic spike initiation in the proximal axon. Sub-optimal classical conditioning of Lymnaea with dilute sucrose (US) results in memory expression which is temporarily undetectable both at 30 minutes and 2 hours after training. The neural mechanisms involved during these memory lapses are unknown. Therefore, two key modulatory neurons, the CGC and the pleural buccal interneuron (PlB) were recorded during lapse and non-lapse time points after sub-optimal conditioning. These neurons were targeted as they can upregulate and downregulate the entire feeding system, and modulatory cells have previously been shown to play a role in memory consolidation. When conditioned preparations were compared to unconditioned preparations there was no significant difference in spontaneous firing frequency or CS-induced firing frequency in the CGC or PlB. This suggests that neither of these identified modulatory neurons play a role at 1 hr (non-lapse), 2 hr (lapse), and 3 hr (non-lapse) following sub-optimal appetitive conditioning. To better understand the neural mechanisms of appetitive memory storage it is also important to characterise the feeding network. Particularly little is known about the control of the modulatory feeding neurons which project to the buccal ganglia, where the feeding CPG and motoneurons are located. Therefore, a search was made for a neuron presynaptic to modulatory projection cells. Such a cell was identified in the parietal ganglion and was termed parietal dorsal 4 (PD4). Intracellular dye filling revealed an expansive morphology with a large axon projecting through the pleural ganglion, cerebral ganglion, and lastly the buccal ganglion. The axon was also seen to branch extensively and cross both the cerebral and buccal commissure. Furthermore, a peripheral projection was observed from a nerve of the visceral ganglion. This extensive morphology makes this neuron an ideal candidate to regulate control of the feeding system as a whole. When artificially activated by depolarisation PD4 caused monosynaptic excitation of both the ipsilateral and contralateral CGC. Delayed excitation was also observed on the ipsilateral CV1a and inhibition on the ipsilateral PlB. However, no direct connection was found between PD4 and identified dorsal buccal feeding neurons. Depolarisation of PD4 could occasionally cause feeding cycles to occur but could also inhibit spontaneous fictive feeding. This differential effect is likely due to polysynaptic depolarisation of the inhibitory CPG neuron N3t, which becomes dominant when there are high frequency action potentials in the CGC. In a semi-intact preparation PD4 did not respond to sucrose applied to the lips, demonstrating it is not involved in the unconditioned feeding response. PD4 was depolarised to threshold by extracellular stimulation of multiple nerves including the lip, tentacle, anal/intestinal and parietal nerves. Activation of PD4 also caused inhibition of the caudodorsal cells (CDCs) and excitation of the ring neuron (RN), suggesting that the neuron may be involved in controlling modulation of the feeding network during egg laying behaviour. In summary, this thesis examines important conceptual gaps in the understanding of appetitive memory consolidation in Lymnaea. It also provides the characterisation of an entirely new cell with widespread influence over the feeding system.