Mechanisms Underlying Trigger-Induced Motor Attacks in the Episodic Ataxia Type 2 Mouse Model Tottering

Abstract

Episodic ataxia type 2 (EA2) is a channelopathy caused by mutations in the CACNA1A gene that encodes for the pore subunit of P/Q type voltage gated Ca+2 channels. Patients carrying these mutations display baseline cerebellar ataxia and episodes of severe ataxia and dystonia. The episodes can last from a few hours to a couple of days, and are triggered by physical or emotional stress, or caffeine or alcohol consumption. The mechanisms by which the stressors induce the episodes are not known. In this thesis, using a well-established mouse model of EA2, tottering, we sought to delineate the mechanisms underlying trigger-induced motor attacks. Because cerebellar Purkinje cells (PCs) are known to be required for the expression of attacks in tottering mice, we focused our studies on the physiology of these cells.

In the past decade our lab has extensively studied the conductances regulating intrinsic pacemaking of PCs. Through this comprehensive work it was established that the only conductance that when blocked can cause high frequency burst firing of PCs is the small conductance Ca+2 activated K + channel (SK) (Walter et al 2006, Womack et al 2004, Womack & Khodakhah 2003). In PCs, SK channel activity is regulated by Ca+2 entering through P/Q type Ca+2 channels, and is required to set the duration of the interspike interval (Walter et al 2006. Womack et al 2004). Therefore, due to the loss of function of P/Q type Ca+2 channels in tottering mice; there is a diminished outward SK current that results in aberrant firing of PCs and manifests as baseline ataxia (Alvina & Khodakhah 2010, Walter et al 2006). In chapter II we show that during dystonic attacks in tottering mice the firing of PCs shifts from irregular tonic firing to high frequency burst firing, which is remarkably similar to what is observed in WT PCs in vitro when SK channels are blocked (Womack & Khodakhah 2003). Interestingly, we found that independent of the trigger used to induce attacks, the irregularity of tottering PCs increased to the same extent during motor attacks, suggesting the triggers share a common pathway by which they induce high frequency burst firing of tottering PCs. Based on this information we hypothesized that further reduction in SK channel activity, mediated by the triggers, is responsible for the transition from a baseline irregular firing to high frequency burst firing of tottering PCs during motor attacks. In chapter II we test this hypothesis using pharmacology and in vivo awake recordings from tottering PCs during attacks. We show that the erratic firing of PCs during attacks can be normalized with SK channel activators, suggesting that indeed reduction in SK channel activity is the cause for the bursting of PCs.

In chapter III and IV we dissect the mechanism/s by which the triggers induce attacks. Using electrophysiology and pharmacology we show in chapter III that stress requires activation of cerebellar α1 adrenergic receptors (α1-R) by norepinephrine (NE) to induce attacks. Given that NE has been shown to reduce SK channel activity via activation of α1-Rs and casein kinase II (CK2), we hypothesized that block of cerebellar CK2 would prevent stress-induced attacks. Using small hairpin RNA (shRNA) packaged into adeno associated viruses I knocked down CK2 in the cerebellum of tottering mice and found that it prevented stress-induced attacks. Because our electrophysiology data suggested that the triggers share a common mechanism by which they induce attacks, I tested whether CK2 knock down in the cerebellum also prevented caffeine and ethanol-induced attacks, and found that it did.

Given that CK2 knock down prevented attacks by all triggers, we asked whether caffeine and ethanol also require noradrenergic signaling to induce attacks. In chapter IV we show that, unlike stress, caffeine and ethanol do not require activation of α1 or β adrenergic receptors. My data rather suggests that caffeine requires activation of metabotropic glutamate receptors 1 (mGluRls) in the cerebellum of tottering mice to induce attacks. Surprisingly, we further found that stress-induced attacks also require cerebellar mGluRls. I wasn't able to pinpoint the mechanism by which ethanol induces erratic activity of PCs. In the discussion of this thesis I suggest a few potential mechanisms that could be tested.

Based on our findings, we suggest that stress, through activation of α1-Rs on tottering PCs, and caffeine, possibly through block of A1 adenosine receptors, indirectly activate mGluRls. This in turn results in a CK2-dependent phosphorylation of SK2-bound CaM that reduces the activity of SK2 channels, and shifts the tonic baseline firing of PCs to high frequency burst firing. The erratic activity of PCs ultimately leads to aberrant cerebellar output that manifests as attacks of dyskinesia.