Abstract
Being able to learn is important for animals to survive in a variable environment. A behavior that is learned quickly provides a good model for studying learning. Conditioned taste aversion (CTA) is one such learning paradigm. During CTA an animal associates a novel taste (conditioned stimuli, CS) with gastric malaise (unconditioned stimuli, US, usually caused in the wild by a toxic food or in the laboratory by administration of Lithium Chloride), in a single learning trial, and subsequently avoids novel the taste (conditioned response, CR). The neural pathways underlying this behavior have been well studied, but much less is known about how the memory is stored in changes in neurons and how gene expression gives rise to those changes. Synaptic plasticity has often been proposed to be the main cellular mechanism that is used to store memories in the nervous system. In studies using in vitro preparations of the rodent hippocampal system and studies of classical fear conditioning, signaling pathways involving the NMDA (N-methyl-D-aspartate) receptor mediated Ca2+ currents and CREB (cAMP responding element binding protein) mediated gene expression have been extensively studied and are thought to be the main molecular mechanism for memory formation. However, little is known about the molecular and cellular mechanisms underlying CTA.We identified a group of genes expressed in the BLA (basolateral nucleus of the amyg- dala) that appears to participate in CTA learning. We also found that intrinsic plasticity of BLA neurons, which changes their ability to convert received current into action potentials, is important for CTA learning. This is interesting because synaptic plasticity is usually thought to be the most important cellular mechanism underlying learning and memory. There have been multiple reports of intrinsic plasticity occurring in different learning behavior paradigms.
In Chapter 2, I present results from experiments we did to identify the molecular and cellular mechanisms underlying CTA. Instead of focusing on a specific candidate gene, we did RNA sequencing in mice that underwent CTA and control training and examined the expression level difference. Using this unbiased systematic approach, we identified a group of genes in the BLA that changed expression levels during CTA. To verify the function of the genes, we used conditional knockout experiments and focused on two of the genes Stk11 (Serine-threonine kinase 11)and Fos. We found that expression of both genes in the BLA is essential for CTA learning. We further studied the cellular mechanisms underlying their function during CTA and found that both genes regulate intrinsic excitability of BLA neurons. We found that conditional knockout (cKO) of Stk11 or Fos in the BLA neurons increased their neuronal excitability. We also found that excitability decreased following learning. To confirm that the intrinsic excitability regulation is important for CTA learning, we did chemogenetics experiment and found that increasing BLA neuronal excitability to mimic the effects of Stk11 or Fos cKO, impaired memory formation of CTA. The findings from this project suggest that gene expression is essential for CTA memory formation and that many genes are important, including Stk11 and Fos gene that we specifically examined in our experiments. Intrinsic excitability has been shown to be an important cellular mech- anism for learning and memory. We also found that Stk11 is not important for retrieval of the aversive association but is required during learning. This may indicate that excitability changes enable learning but are not part of the long-term storage mechanism encoding the memory. In this project we identified a new signaling pathway important for aversive learning and found that a novel form of intrinsic plasticity was involved in this learning process.
In Chapter 3, we did experiments to identify candidate ion channels that could be involved in intrinsic excitability regulation of BLA neurons. Since there are many ion channels that could influence the neuronal excitability, we examined gene that are expressed in BLA neurons and that have been shown in previous studies to regulate neuronal excitability especially during learning and memory. The channels studied could be categorized into two groups, one affecting subthreshold properties and others affecting action potential firing properties. We found that voltage-dependent ion channels could significantly influence the BLA neuron firing rate due to their influence on the action potential themselves. Another type of ion channel that had a significant influence is HCN channels. Channels examined that showed little influence on excitability include M-channels, calcium-dependent potas- sium channels and TASK3 leak channels. Channels that were not examined but could be candidates for excitability regulation are various types of leak channel and inward rectifying potassium channels. These results provide useful information about ion channels and their influence on excitability in BLA neurons.
I also collaborated with researchers from Janelia Farm Research Institute on studying the gene expression profiles of different type of thalamus neurons. We found that the thalamic projection neurons are composed of three groups with different molecular profiles. This profile categorization could be applied to neurons in different sensory systems in the thala- mus. It was also found that these molecular profiles correlate well with the anatomical and electrophysiological properties of the neurons. These results are included in this thesis as appendix.