Optogenetic investigation of cortical network dynamics in epilepsy

Abstract

Understanding the cortical network properties which determine the susceptibility of cortex to the onset of seizures remains a major goal of epilepsy research. The determinants of seizure risk in cortical networks are dynamic, showing dependency on intrinsic cortical activity and environmental influences. The failure to identify reliable electrographic indicators of imminent seizure onset suggests that the contributory factors may not be electrographically obvious. A strong candidate for such a property is the activity dependent disinhibition of the excitatory network which results from increases in intracellular chloride concentration. Chloride loading has been shown previously to occur during periods of intense neuronal activity, resulting from concomitant excitatory and inhibitory synaptic transmission. To explore how network dynamics evolve from a stable healthy state to one permissive for the onset and propagation of seizures, I used an optogenetic approach to selectively interrogate dynamic changes to excitatory transmission between the principal cells of the cortical circuit following an acute ictogenic challenge, both in vitro and in vivo. Using ultra-low frequency optogenetic stimulation genetically targeted to the pyramidal cells of neocortex, I demonstrate that epileptiform activity, which develops spontaneously following an acute chemoconvulsant challenge, can both be reduced and monitored, using an active probing strategy. Delivering continuous and focal optogenetic stimulations to superficial neocortex and regions of the hippocampal formation evokes glutamatergic responses in the LFP which can be used to assay dendritic excitability in the network. At ultralow frequencies, between 0.1-0.033 Hz, optogenetic stimulation markedly reduced the rate of evolution of epileptiform activity, when delivered to neocortex or hippocampal structures, in acutely prepared adult mouse brain slices bathed in 0Mg2+ perfusate. The response evoked by these test pulses undergoes an all-or-nothing transformation observable in the LFP which reliably telegraphed the onset of ictal activity in two models of epilepsy. Using electrophysiological tools and 2-photon calcium imaging of individual dendrites, I demonstrate that this phenomenon likely reflects a reduction in the threshold for dendritic spikes. Using an anatomically realistic computational model pyramidal cell I show that this effect is reproduced by modest positive shifts in the GABAergic reversal potential in distal pyramidal cell dendrites. Finally, I report preliminary data demonstrating a potential mechanism for the diurnal modulation of seizure risk. Diurnal periodicity in seizure susceptibility have been observed longitudinal recordings from both patients and chronically epileptic experimental animals. Using the optical chloride sensor ClopHensor I examine steady-state pyramidal cell chloride concentration over the diurnal period and show that periodicity in chloride homeostasis is consistent with the phase of diurnally modulated seizure risk. In this thesis I use a range of optical and electrophysiological tools to explore the contribution of dynamic chloride concentration in pyramidal cells in determining cortical susceptibility to seizures onset. Using two acute epilepsy models I demonstrate that an assayable increase in dendritic excitability precedes ictogenesis, and demonstrate a potential mechanism by which variation in [Cl-]i can give rise to this effect. I go on to show diurnal variation in [Cl-]i in cortical pyramidal cells, and link this to circadian modulation of susceptibility to chemoconvulsants, suggesting a functional mechanism for the dynamic seizure risk observed in epileptic patients.