Investigating the Homeostatic Regulation of Kenyon Cells in Drosophila

How do Kenyon cells behave consistently with such a high variation of inputs? In Drosophila, stimulus-specificity of associative memories requires sparse coding in the neurons that are storing the memory. Effective sparse coding requires that the neurons have approximately equal probabilities of firing across all stimuli. Otherwise, some cells will be disproportionately active or silent, and thereby be less informative about stimulus identity. I study the problem of how distributed sparse coding is maintained in Kenyon cells (KCs), which are the third-order olfactory neurons. I hypothesised that KCs homeostatically adjust their intrinsic properties to ensure even activity across the population of KCs. I investigated this hypothesis using a combination of two-photon imaging and genetic manipulation of ion channel expression. Sodium and potassium channels are responsible for the depolarisation and repolarisation of the membrane, respectively. I conducted experiments to investigate the impact of artificially manipulating the expression of sodium and potassium channels on KCs' activity. Additionally, I examined whether this manipulation could potentially regulate the expression of other sodium and potassium channels through homeostatic mechanisms. My results indicate that constitutively overexpressing NaChBac, an exogenous voltage-gated sodium channel, in KCs paradoxically decreases their activity. The neurons’ activity was measured by amplitude of odour responses with calcium imaging. The developmental expression and two days of acute NaChBac expression in adults produced a significant decrease in odour responses. However, 4 days of acute expression in adults increased the odour responses of specific KCs. When NaChBac was expressed in adults for 8 days, odour responses were not significantly different from control. Thus, only specific acute expression of NaChBac causes an increase in excitability. When investigating whether the KCs had homeostatic mechanisms that counteracted NaChBac’s effect, I found that the levels of other ion channels had adjusted. Constitutively expressing NaChBac decreased expression of para, the endogenous voltage-gated sodium channel responsible for generating action potentials. It may be that NaChBac expression decreases Para levels to prevent overexcitation. Furthermore, NaChBac expression also increases endogenous levels of the potassium channel, Shaker. As NaChBac causes abnormally long action potentials due to its slow kinetics, while Shaker is normally responsible for membrane repolarisation at the end of an action potential, it may be that the increased Shaker is an attempt to compensate for the prolonged depolarisation caused by NaChBac. In contrast, knocking down or disrupting a multitude of endogenous potassium channels in KCs appeared to have very little effect on KC activity. This finding is unlike what it is typically found in most past literature that focuses on disrupting potassium channels. The project revealed a paradoxical response of KCs when they were artificially excited by expressing NaChBac and disrupting their potassium channels. Additionally, the project discovered that KCs can modulate their ion channel expression levels in response to NaChBac expression.