Abdominal ganglion

Cellular and molecular mechanisms contributing to classical conditioning.
Many theories of learning have proposed that plasticity at specific synapses in the CNS is critical for memory storage, and a number of mechanisms of synaptic plasticity have been identified that might contribute to learning and memory. However, in most cases it has been very difficult to test the causal relationship between these mechanisms and learning, primarily because of the immense complexity of the mammalian CNS. For this reason, invertebrate preparations such as Aplysia are advantageous. The Aplysia gill- and siphon-withdrawal reflex undergoes classical conditioning with many of the behavioral features of conditioning in mammals, suggesting that conditioning in Aplysia and mammals may share common mechanisms. Furthermore, monosynaptic connections between sensory neurons and motor neurons that contribute to the withdrawal reflex exhibit a cellular analog of conditioning with temporal parameters similar to the behavioral conditioning. However, these studies of synaptic changes in the isolated nervous system have not been able to address the contribution of the plasticity to behavior. For these reasons, I recently developed a simplified preparation for studying the siphon-withdrawal reflex of Aplysia, with which it is relatively easy to record the activity of identified neurons and their synaptic connections simultaneously with behavioral learning.

Neuronal Mechanism of Classical Conditioning in Aplysia: a) activity-dependent presynaptic facilitation; b) retrograde signaling and AMPA receptor insertion

Facilitatory intemeurons (FIN) release several transmitters, including serotonin (5-HT), that bind to receptors coupled to adenylyl cyclase on the sensory neuron, stimulating production of cAMP, activation of cAMP-dependent protein kinase (PKA), phosphorylation and closure of K+ channels, broadening of subsequent action potentials, increased Ca2+ influx, and increased transmitter release. Spike activity in the sensory neuron just before the serotonin causes an influx of Ca2+ that "primes" the cyclase, leading to enhanced activation of the cAMP cascade.

The spike activity also causes release of glutamate, which binds with AMPA and NMDA-type receptors on the motor neuron. Depolarization of the motor neuron relieves the Mg2+ block of the NMDA receptor channels, allowing the glutamate to stimulate Ca2+ influx. The Ca2+ may have postsynaptic actions, but it also appears to stimulate production of a retrograde messenger that interacts with the cAMP cascade in the sensory neuron.

I showed that monosynaptic connections from LE siphon sensory neurons to LFS siphon motor neurons make a substantial contribution to the reflex in this preparation (Antonov et al., 1999). I then assessed the contribution of various cellular mechanisms to classical conditioning of the reflex and found that it is due in part to associative activity-dependent plasticity at synapses from the LE sensory neurons to LFS motor neurons (Antonov et al., 2001). I next investigated the possible contribution to the conditioning of two associative cellular mechanisms: activity-dependent neuromodulation involving enhanced activation of the PKA pathway in the LE neuron, or Hebbian long-term potentiation (LTP) involving enhanced Ca++ elevation in the LFS neuron. My results provide the most direct evidence to date that each of these mechanisms contributes to behavioral learning, and also suggest that the two mechanisms are not independent but rather interact through retrograde signaling (Antonov et al., 2003a).
I now propose to extend these studies in several ways.

First, I will perform additional experiments to test the roles of these synaptic mechanisms in behavior. In the experiments I have performed so far, injecting the Ca++ chelator BAPTA into a single LFS neuron or the PKA inhibitor PKAi into a single LE neuron has blocked plasticity of the monosynaptic PSP but has not blocked behavioral learning, presumably because the behavioral stimuli activate several LE and LFS neurons in parallel. I will therefore test whether injecting two or three LE or LFS neurons in the same preparation reduces behavioral conditioning.
Fluorescence image LE sensory neuron and LFS motor neuron

Second, I will examine pre- and postsynaptic molecular mechanisms of the plasticity during conditioning. Recent studies of LTP in mammalian systems indicate that it involves insertion of AMPA-type glutamate receptors into the postsynaptic membrane. I have found that the monosynaptic PSP in Aplysia also has an AMPA component, and I will test whether conditioning involves insertion of new AMPA receptors. In addition, I will examine retrograde signaling and possible presynaptic mechanisms. My colleagues and I have found that bathing the abdominal ganglion in the nitric oxide (NO) synthase inhibitor nitroarginine blocks behavioral conditioning, suggesting that NO could be a retrograde messenger (Antonov et al., 2003b). I will test this idea at the cellular level, and I will also examine possible presynaptic targets of NO including PKG and ryanodine-sensitive Ca2+ stores. These studies should begin to elucidate coordinate pre- and postsynaptic mechanisms of plasticity that contribute to behavioral learning in Aplysia

Structural changes during the conditioning

The early phase of synaptic plasticity in many systems is accompanied by the rapid redistribution of synaptic proteins. Facilitation by serotonin in Aplysia and synaptic activity in hippocampal neurons are both accompanied by a decrease in the number of clusters or puncta of the synaptic vesicle associated protein synapsin, which is thought to reflect vesicle mobilization. On the other hand, the onset of long-term potentiation in hippocampal neurons is accompanied by a rapid increase in the number of puncta of synapsin and another vesicle associated protein, synaptophysin, as well as the GluR1 subunit of postsynaptic AMPA receptors and sites where the pre- and postsynaptic proteins co-localize (Antonova et al., 2001). These changes are thought to reflect either the rapid formation of new functional synapses or an early step in that process.
I have begun to perform experiments to examine the possible redistribution of synaptic proteins during conditioning, first focusing on the presynaptic protein synaptophysin. My colleagues and I have cloned a DNA construct for a synaptophysin-GFP (Green Fluorescent Protein) fusion protein into the Aplysia expression vector pNEX3 and microinjected the purified plasmid DNA into the sensory neuron. Expression of the fluorescent protein in living neurons can be visualized one day later on a 2-photon microscope.
Expression of the fluorescent protein in living neurons and examine the distribution of the fluorescence before and after the conditioning
Structural changes in the synaptophysin-GFP fluorescent puncta in the processes of an LE sensory neuron in the abdominal ganglion following paired training.
These results are consistent with the idea that postsynaptic AMPA receptor insertion could contribute to facilitation of the PSP during conditioning.
To investigate whether such changes occur during the long-term retention of conditioning in the siphon-withdrawal preparation, I will express GFP, which labels the entire neuron, in either the sensory neuron or motor neuron and look for structural changes at various times after training. If such changes occur, I will test whether they are blocked by inhibiting protein or RNA synthesis during training.
Home || Contact || Research Interests: Learning and Memory, Neurotransmitter secretion ||