Neural correlates of Pavlovian conditioning in components of the neural network supporting ciliary locomotion in Hermissenda

Abstract

Pavlovian conditioning in Hermissenda consists of pairing light, the conditioned stimulus (CS) with activation of statocyst hair cells, the unconditioned stimulus (US). Conditioning produces CS-elicited foot shortening and inhibition of light-elicited locomotion, the two conditioned responses (CRs). Conditioning correlates have been identified in the primary sensory neurons (photoreceptors) of the CS pathway, interneurons that receive monosynaptic input from identified photoreceptors, and putative pedal motor neurons. While cellular mechanisms of acquisition produced by the synaptic interaction between the CS and US pathways are well-documented, little is known about the mechanisms responsible for the generation or expression of the CR. Here we show that in conditioned animals light reduced tonic firing of ciliary activating pedal neurons (VP1) below their pre-CS baseline levels. In contrast, pseudorandom controls expressed a significant increase in CS-elicited tonic firing of VP1 as compared to pre-CS baseline activity. Identified interneurons in the visual pathway that have established polysynaptic connections with VP1 were examined in conditioned animals and pseudorandom controls. Depolarization of identified type Ie interneurons with extrinsic current elicited a significant increase in IPSPs recorded in VP1 pedal neurons of conditioned animals as compared with pseudorandom controls. Conditioning also enhanced intrinsic excitability of type Ie interneurons of conditioned animals as compared to pseudorandom controls. Light evoked a modest increase in IPSP frequency in VP1 of conditioned preparations and a significant decrease in IPSP frequency in VP1 of pseudorandom controls. Our results show that a combination of synaptic facilitation and intrinsic enhanced excitability in identified components of the CS pathway may explain light-elicited inhibition of locomotion in conditioned animals.

Figure 1

Mean suppression ratios ±SEM for conditioned (n = 50) and pseudorandom controls (n = 50). Conditioning produced statistically significant suppression of light-elicited locomotion compared with pseudorandom controls (*P < 0.001).

Figure 2

Components of the CS pathway involved in visually mediated ciliary locomotion in Hermissenda. Only connections with a single photoreceptor are shown. The diagram shows a monosynaptic connection between a lateral B-photoreceptor and type I_e and I_i interneurons. The dashed lines between the type I_e and I_i and type III_i interneurons indicate a polysynaptic pathway with unidentified interneurons. The monosynaptic connections between type III_i interneurons and VP1 have been previously established (Crow and Tian 2003). Filled triangles denoted inhibitory synapses, open excitatory.

Figure 3

Examples of light-elicited changes in spike activity for conditioned animals and pseudorandom controls. (A) Light-elicited decrease in the spike activity recorded in pedal neuron VP1 from a conditioned animal. (B) Light-elicited increase in the spike activity recorded in VP1 from a pseudorandom control. (C) Group data depicting the mean difference in VP1 spike activity evoked in 5 min of light and a 5-min period in the dark immediately preceding light onset collected from conditioned animals ((n = 10) and pseudorandom controls (n = 9; *P < .05).

Figure 4

Examples of depolarizing current elicited IPSPs in VP1 for conditioned animals and pseudorandom controls. (A1)An example from a conditioned animal of spikes evoked bya2sec0.3 nA depolarizing current pulse applied to an identified type I_e interneuron. (A2) An increase in the number of IPSPs relative to baseline recorded in VP1 elicited by depolarization of I_e.(B1)An example from a pseudorandom control animal of spikes evoked by a 2 sec 0.3 nA depolarizing current pulse applied to an identified type I_e interneuron. (B2) IPSPs in VP1 evoked by depolarization of I_e from a pseudorandom control. (C) Group data depicting the mean percent increase from baseline in VP1 IPSPs elicited by depolarization of identified type I_e interneurons from conditioned animals (n = 12) and pseudorandom controls (n = 9; *P < .05).

Figure 5

Conditioning produces enhanced excitability of identified type I_e interneurons as compared to pseudorandom controls. (A1) An example of the CS-elicited depolarization and increase in spike frequency recorded in a type I_e interneuron from a conditioned animal and a pseudorandom control (B1). (A2–A4) Excitability was assessed in the type I_e interneuron shown in (A1) from a conditioned animal with 2-sec depolarizing extrinsic current pulses of increasing intensity (0.1 nA–0.3 nA). (B2–B4) Excitability of the type I_e interneuron shown in (B1) from a pseudorandom control assessed with 2-sec depolarizing extrinsic current pulse of increasing intensity (0.1 nA–0.3 nA). (C) Group data showing the mean evoked spikes recorded in type I_e interneurons from conditioned animals (n = 10) and pseudorandom controls (n = 12) for the three different depolarizing current levels (*P<.05).

Figure 6

Light-elicited changes in VP1 IPSP frequency in conditioned animals and pseudorandom controls. (A) Example of a light-elicited increase in VP1 IPSPs relative to baseline from a conditioned preparation. (B) Example of a light-elicited decrease in VP1 IPSPs relative to baseline from a pseudorandom control preparation. (C) Group data depicting the mean percent change from baseline during light-evoked (1 min) IPSP frequency for conditioned animals (n = 12) and pseudorandom controls (n = 11) (*P < .001).