A chlorzoxazone-baclofen combination improves cerebellar impairment in spinocerebellar ataxia type 1

Background A combination of central muscle relaxants, chlorzoxazone and baclofen (chlorzoxazone-baclofen), has been proposed for treatment of cerebellar symptoms in human spinocerebellar ataxia (SCA). However, central muscle relaxants can worsen balance. The optimal dose for target engagement without toxicity remains unknown. Objectives Using the genetically precise Atxn1154Q/2Q model of SCA1, we determine the role of cerebellar dysfunction in motor impairment. We also aim to identify appropriate concentrations of chlorzoxazone-baclofen needed for target engagement without toxicity to plan for human clinical trials. Methods We use patch clamp electrophysiology in acute cerebellar slices and immunostaining to identify the specific ion channels targeted by chlorzoxazone-baclofen. Behavioral assays for coordination and grip strength are used to determine specificity of chlorzoxazone-baclofen for improving cerebellar dysfunction without off-target effects in Atxn1154Q/2Q mice. Results We identify irregular Purkinje neuron firing in association with reduced expression of the ion channels Kcnma1 and Cacna1g in Atxn1154Q/2Q mice. Using in vitro electrophysiology in brain slices, we identify concentrations of chlorzoxazone-baclofen that improve Purkinje neuron spike regularity without reducing firing frequency. At a disease stage in Atxn1154Q/2Q mice when motor impairment is due to cerebellar dysfunction, orally administered chlorzoxazone-baclofen improves motor performance without affecting muscle strength. Conclusion We identify a tight relationship between baclofen-chlorzoxazone concentrations needed to engage target, and levels above which cerebellar function will be compromised. We propose to use this information for a novel clinical trial design, using sequential dose escalation within each subject, to identify dose levels that are likely to improve ataxia symptoms while minimizing toxicity.


Introduction
The CAG triplet repeat-associated spinocerebellar ataxias (SCA) are a group of neurodegenerative diseases that result in progressive loss of motor function and early death [1].
While the six genetic causes of CAG triplet repeat-associated SCA are unrelated, recent work illustrates that shared underlying mechanisms of disease may contribute to neuronal dysfunction across these disorders [2][3][4][5]. SCA1, one of the most common and well-studied of the SCAs, is caused by a glutamine-encoding CAG triplet expansion in the ATXN1 gene.
Cerebellar Purkinje neurons are prominently affected in SCA1 and are believed to underlie motor impairment, while dysfunction of brainstem nuclei and other non-cerebellar structures correlates more closely with early death [6][7][8]. However, the relative contribution of these structures to motor impairment at various stages of disease remains unclear.
Recent work demonstrates that Purkinje neuron dysfunction is a central cause of motor impairment in rodent models of SCA. In murine models, abnormalities in Purkinje neuron membrane excitability often manifest as irregular, slow, or absent spiking. Interestingly, specific ion-channels have been associated with firing dysfunction across several of these models [2][3][4][5], suggesting shared mechanisms of disease and potentially overlapping therapeutic targets.
These channels include the large conductance calcium-activated potassium (BK) channel, along with calcium sources that can activate BK function [2]. In the transgenic ATXN1[82Q] mouse model of SCA1, targeting reduced BK channel function with the FDA-approved potassium channel activating compounds chlorzoxazone and baclofen (chlorzoxazone-baclofen) can restore Purkinje neuron spiking and improve motor dysfunction [9]. Due to Purkinje neuronspecific expression of the ATXN1[82Q] transgene in this model of SCA1 [10], the ability of chlorzoxazone-baclofen to improve cerebellar symptoms without worsening non-cerebellar symptoms is not currently known.
The benefit of chlorzoxazone-baclofen on Purkinje neuron spiking has previously been assessed at a single concentration in the transgenic ATXN1[82Q] model of SCA1 [9]. In these mice, Purkinje neurons undergo depolarization block due to profound loss of BK channel current early in disease, causing a majority of cells to cease firing at the onset of symptoms [4]. Since ATXN1[82Q] transgene expression is restricted to Purkinje neurons in this model, it is unclear whether chlorzoxazone-baclofen may have off-target effects outside of the cerebellum. It is therefore imperative to identify the concentration ranges of chlorzoxazone-baclofen that can optimally improve Purkinje neuron function while minimizing extra-cerebellar toxicity. This will enable the design of future clinical trials to assess the safety and efficacy of chlorzoxazonebaclofen in human SCA.
In the present work, we use a genetically precise knock-in mouse model of SCA1 to explore appropriate concentrations and target engagement of chlorzoxazone-baclofen. Our studies demonstrate that BK channels are the likely calcium-activated potassium channel target in SCA1. Importantly, chlorzoxazone-baclofen co-treatment improves motor function in SCA1 mice at an age when the motor phenotype is not confounded by changes in grip strength. In addition, chlorzoxazone-baclofen show no signs of toxicity in SCA1 mice and do not worsen deficits in grip strength at a later disease timepoint when motor weakness is present. These data indicate that chlorzoxazone-baclofen treatment specifically targets cerebellar dysfunction in SCA1, and supports the further development of chlorzoxazone-baclofen as a treatment for cerebellar symptoms in SCA1.

Animals
All animal studies were reviewed and approved by the University of Michigan Institutional Animal Care and Use Committee and were conducted in accordance with the United States Public Health Service's Policy on Humane Care and Use of Laboratory Animals. Atxn1 154Q/2Q mice, which express an expanded CAG triplet repeat in the endogenous Atxn1 locus [11], were maintained on a C57BL/6 background. Heterozygous Atxn1 154Q/2Q mice and wild-type littermate controls were used for all studies. All studies were performed at 14 weeks of age or 20 weeks of age. Sexes were balanced for all animal studies.
Patch-clamp recordings were performed at 33°C. Pre-warmed, carbogen-bubbled aCSF was perfused at a rate of 150 mL/hour for all recordings. Recordings were acquired using an Axopatch 200B amplifier, Digidata 1440A interface, and pClamp-10 software (Molecular Devices, San Jose, CA). Current clamp data were acquired at 100 kHz in the fast current-clamp mode with bridge balance compensation, and filtered at 2 kHz. Cells were included only if the series resistance did not exceed 15 MΩ at any point during the recording, and if the series resistance did not change by more than 20% during the course of the recording. All presented voltage clamp data have been corrected for a 10 mV liquid junction potential.

Pharmacology
For some recordings, aCSF contained the following reagents (as specified in the results section

Tissue Immunofluorescence and Microscopy
Brains from Atxn1 154Q/2Q mice and wild-type littermate controls were placed in 1% paraformaldehyde for 1 hour at room temperature, and were then transferred to a solution of 30% sucrose in phosphate-buffered saline (PBS) for 48 hours at 4°C. Brains were preserved in Sections were imaged on an Axioskop 2 plus microscope (Zeiss, White Plains, NY) and quantified using ImageJ by measuring mean pixel intensity within a box covering the molecular layer of lobule 5. Representative images were acquired on a C2+ confocal microscope (Nikon, Melville, NY) at 60x magnification.
Cell counts and molecular layer thickness measurements were performed on 20-week-old Atxn1 154Q/2Q mice and wild-type littermate controls after completion of behavioral studies. Tissue was stained for calbindin to mark cerebellar Purkinje neurons, as described above. A distance to 150 µM was drawn from the base of lobule 4, at which point a perpendicular line was drawn from the middle of the soma of the nearest Purkinje neuron to the tip of its dendritic arbor. This distance was recorded as the molecular layer thickness. For cell counts, a distance of 700 µM was measured from the base of lobule 4 and all stained Purkinje neuron somata were counted in that distance. All microscopy and image analyses were performed with the reviewer blind to genotype.

In Vivo Delivery of Chlorzoxazone-Baclofen
Chlorzoxazone and baclofen were prepared in drinking water as described previously [9]. For longitudinal drug delivery, baclofen (350 µmol/L) and chlorzoxazone (15 mmol/L) were dissolved in drinking water containing 0.05% β-(hydroxypropyl)-cyclodextrin, 40 µL/L NaOH, and 6% sucrose. Vehicle drinking water contained 0.05% β-(hydroxypropyl)-cyclodextrin, 40 µL/L NaOH, and 4% sucrose to ensure that an equal volume of water was consumed across groups. Mice were treated with drinking water beginning at 13 weeks of age. Water was provided ad libitum until the end of experimentation, at 20 weeks of age. Water bottles were changed twice weekly.

Behavior Assays
Motor performance was analyzed using a rotarod protocol as described previously [9]. Mice were handled 3 times before 8 weeks of age to acclimate them to the experimenter. Mice were then trained on the rotarod at 9 weeks of age, using an accelerating speed from 4-40 rpm for 3 consecutive days followed by one day at a constant speed of 24 rpm. At 10 weeks of age, mice were tested for 4 consecutive days at 24 rpm and then randomized into groups based on their baseline performance, keeping baseline consistent between groups within genotype. Drug or vehicle was then administered via water bottles for the duration of the experiment. When the mice reached 14 and 20 weeks of age, they were re-trained for one day and then tested for 4 consecutive days at 24 rpm.
The hanging wire task was performed as described previously [13]. Mice were placed upside down on a suspended wire of 3 millimeter diameter and held approximately 12 inches above a soft landing surface. The time for the mouse to lose grip of the wire was recorded, to a maximum time of 120 seconds. Mice were tested at 14 and 20 weeks of age, on the final day of the rotarod task.

Statistical Analysis
Data were compiled in Excel (Microsoft Corporation, Redmond, WA) and analyzed using Prism (GraphPad, San Diego, CA). Electrophysiology data were analyzed by using either paired or unpaired Student's t-test as appropriate. Analyses of grip strength and body weight were completed using a one-way analysis of variance (ANOVA) with a Holm-Sidak correction for multiple comparisons. Analyses of the AHP and rotarod behavioral assays were completed using a two-way repeated measures ANOVA with a Holm-Sidak correction for multiple comparisons. Statistical significance was determined with an α-level of 0.05 for all studies.

Purkinje neuron dysfunction is associated with motor impairment preceding neurodegeneration in Atxn1 154Q/2Q mice
Purkinje neuron dysfunction is an early feature of disease in many models of spinocerebellar ataxia (SCA) [3][4][5][13][14][15][16][17][18]. Changes in Purkinje neuron firing are directly relevant to disease, as improving Purkinje neuron firing also improves motor dysfunction in murine SCA models [5,9,13,[16][17][18]. Abnormalities in Purkinje neuron firing are particularly relevant in SCA1, as pharmacologic and genetic strategies to improve Purkinje cell intrinsic excitability improves not just motor dysfunction but also degeneration in murine models of disease [4,18,19]. Prior studies in SCA1 have focused primarily on the transgenic ATXN1[82Q] model of SCA1, where polyglutamine expanded ATXN1 expression is restricted to Purkinje cells. While the ATXN1[82Q] model of SCA1 accurately models cerebellar features of disease, the relevance of cerebellar dysfunction to the overall motor phenotype is unknown in the more genetically precise Atxn1 154Q/2Q model of SCA1, in which a polyglutamine-expanded Atxn1 transgene is knocked into the endogenous Atxn1 locus, and therefore widely expressed in the nervous system and elsewhere. In this model of SCA1 with a hyperexpanded polyglutamine repeat motor dysfunction and premature death have been attributed, at least in part to spinal cord and brainstem motor neuron degeneration [7,11]. We therefore sought to identify whether cerebellar dysfunction is present, and relevant to the motor phenotype in Atxn1 154Q/2Q mice.
We examined Purkinje neuron membrane excitability in Atxn1 154Q/2Q mice at 14 weeks of age, an age at which motor impairment is prominent [11]. No evidence of neurodegeneration is present in Atxn1 154Q/2Q mice even at 20 weeks of age (Supplementary Figure 1A Figure 1D) was noted after the addition of synaptic inhibitors, suggesting that altered Purkinje neuron spiking in Atxn1 154Q/2Q mice is intrinsically driven. When neurons were held at -80 mV and injected with depolarizing current, we found that Atxn1 154Q/2Q neurons require a significantly lower amplitude of injected current to elicit depolarization block than wildtype littermate control neurons ( Figure 1E). Importantly, these alterations in spiking were accompanied by no significant change in input resistance between Atxn1 154Q/2Q and wild-type littermate control neurons ( Figure 1F).
Since irregular Purkinje neuron spiking is strongly associated with deficits in the afterhyperpolarization (AHP) in various mouse models of ataxia [3][4][5]20], we examined differences in spike waveform between Atxn1 154Q/2Q Purkinje neurons and wild-type littermate controls. We found that the AHP in Atxn1 154Q/2Q mice was significantly depolarized when compared to wild-type littermate control cells ( Figure 1G-I). No other significant changes in spike waveform were noted in Atxn1 154Q/2Q Purkinje neurons when compared to wild-type littermate controls (Table 1). Together, these data indicate that a specific reduction in AHP amplitude accompanies irregular spiking and neuronal hyperexcitability in Atxn1 154Q/2Q mice at a stage of disease with prominent motor impairment.

A combination of chlorzoxazone and baclofen improves Atxn1 154Q/2Q Purkinje cell firing
The altered Purkinje neuron spiking observed in symptomatic Atxn1 154Q/2Q mice is similar to neuronal dysfunction seen in another mouse model of SCA1, the ATXN1[82Q] transgenic model [4]. The more profound loss of BK and Cav3.1 channel expression in ATXN1[82Q] mice results in a more prominent electrophysiologic phenotype, in which neurons undergo a complete loss of spiking [4] , which nevertheless represents a continuum with the irregular spiking observed in Atxn1 154Q/2Q Purkinje neurons [2] (Figure 1). The shared underlying dysfunction of BK channels and reduced AHP amplitude suggests that BK channels may be an important target in SCA1.
In transgenic ATXN1[82Q] mice, combined treatment with chlorzoxazone and baclofen targets potassium channel dysfunction to improve Purkinje neuron spiking and motor impairment at both early-and mid-symptomatic ages [9]. Chlorzoxazone is a canonical KCa channel activator that can target BK channels [26]. Baclofen, a GABAB receptor agonist, is a known activator of subthreshold-activated potassium channels in Purkinje neurons [27]. Subthreshold-activated potassium channels can compensate for loss of the AHP and facilitate repetitive spiking in Purkinje cells [3]. In ATXN1[82Q] mice, a combination of chlorzoxazone and baclofen (chlorzoxazone-baclofen) was necessary to achieve adequate KCa channel activation and restoration of function [9]. We wished to determine the necessity of a combination of these agents to improve the more modestly impaired firing in Atxn1 154Q/2Q mice. Since SCA1-KI mice exhibit a gradation of Purkinje cell firing abnormalities, this allows for investigation of dosedependence of chlorzoxazone-baclofen to correct irregular spiking. It is important to determine the optimal concentrations of chlorzoxazone-baclofen as excessive activation of K + channels results in reduction in Purkinje cell firing frequency, which is independently correlated with motor dysfunction (reviewed in [28]). and baclofen (400 nM) has the optimal effect of improving Purkinje neuron spiking irregularity without affecting firing frequency in acute cerebellar slices. Additionally, these data clearly indicate the necessity of a combination of chlorzoxazone and baclofen to achieve the effect of improving spike regularity while not affecting firing frequency. Atxn1 154Q/2Q mice have a hyperexpanded polyglutamine repeat. In human SCA1, large repeat expansions are associated with brainstem predominant disease and premature death in the second decade of life [29][30][31]. It is important to know whether cerebellar dysfunction contributes to motor impairment in Atxn1 154Q/2Q mice and accurately serves as a model for testing therapy for cerebellar dysfunction, the major source of disability and quality of life in human SCA1.
Atxn1 154Q/2Q mice are described to have motor neuron involvement [7], which may confound interpretation of the rotarod task in mice. We therefore evaluated grip strength, a measure of motor neuron function, using the hanging wire task. At 14 weeks of age, no impairments in grip strength were noted in Atxn1 154Q/2Q mice when compared to wild-type littermate controls ( Figure   4A). Atxn1 154Q/2Q mice were administered chlorzoxazone-baclofen in drinking water at a dose expected to achieve the intermediate concentrations (see Figure 4 of [9]) of chlorzoxazone (1-10 µM) and baclofen (100 nM-1 µM) in the brain. Importantly, chlorzoxazone-baclofen did not affect grip strength of 14 week-old Atxn1 154Q/2Q mice ( Figure 4A). The rotarod assay is commonly used to test balance and gait coordination, and reflects balance impairment when not confounded by impaired muscle strength [32]. At 14 weeks, Atxn1 154Q/2Q mice display significantly impaired motor function on the rotarod assay when compared to wild-type littermate controls ( Figure 4B). The absence of grip-strength impairment in the Atxn1 154Q/2Q mice provides more confidence that the rotarod impairment at this stage of disease is due to cerebellar dysfunction. Treatment with chlorzoxazone-baclofen significantly improved motor performance in 14 week-old Atxn1 154Q/2Q mice ( Figure 4B). At 20 weeks of age, a later stage of disease, Atxn1 154Q/2Q mice display a significant impairment in grip strength compared to wild-type littermate controls ( Figure 4C). Reassuringly, chlorzoxazone-baclofen, which are FDA approved skeletal muscle relaxants, did not worsen the grip strength deficits in Atxn1 154Q/2Q mice ( Figure 4C). At this stage of disease, it would be expected that rotarod performance of Atxn1 154Q/2Q mice would be further impaired, referable to the observed grip strength deficits. Consistent with this hypothesis, 20-week Atxn1 154Q/2Q mice displayed significantly impaired motor performance on the rotarod compared to wild-type littermate controls to a larger degree than was observed at 14 weeks of age. Chlorzoxazonebaclofen did not rescue the rotarod impairment in Atxn1 154Q/2Q mice at this stage of disease ( Figure 4D), but, importantly, also did not worsen motor performance in 20 week-old Atxn1 154Q/2Q mice, consistent with the lack of effect of chlorzoxazone-baclofen on grip strength at this age.
Since chlorzoxazone-baclofen is being considered for clinical use in patients with ataxia [9], we evaluated for potential toxicity of long-term use of this combination in Atxn1 154Q/2Q mice. As reported previously, Atxn1 154Q/2Q mice fail to gain weight compared to wild-type littermate controls, which is evident in both male and female mice at both 14 ( Figure 5A-B) and 20 ( Figure   5C-D) weeks of age [11]. Chlorzoxazone-baclofen did not accelerate this failure to gain weight.
A slightly lower baseline weight was observed in chlorzoxazone-baclofen treated wild-type mice ( Figure 5A), which was also noted in these same mice at 20 weeks of age ( Figure 5C). However, this difference in weight was not accelerated in chlorzoxazone-baclofen treated female wild-type mice when compared to vehicle treated controls at the two recorded timepoints ( Figure 5E), suggesting that these mice still gain weight normally and are not experiencing toxicity from chlorzoxazone-baclofen treatment. Together, these behavioral studies indicate that at 14 weeks, cerebellar dysfunction contributes meaningfully to motor impairment in Atxn1 154Q/2Q mice and can be ameliorated by a treatment strategy targeting cerebellar dysfunction. At a later disease stage, Atxn1 154Q/2Q mice display a motor phenotype that is driven by impairments of muscle strength. These findings reinforce the idea that chlorzoxazone-baclofen specifically targets motor impairment due to cerebellar dysfunction in SCA1.

Discussion
Chlorzoxazone-baclofen has been proposed for the treatment of cerebellar motor dysfunction in SCA [9]. However, the optimal dose of chlorzoxazone-baclofen for target engagement without off-target toxicity is not known. In the present study, we identify a dose range of chlorzoxazonebaclofen that improves Purkinje neuron spike regularity in the Atxn1 154Q/2Q model of SCA1 without suppressing firing frequency. Importantly, chlorzoxazone-baclofen improves cerebellar motor dysfunction in Atxn1 154Q/2Q mice at an age when motor impairment is cerebellar in origin, and does not negatively affect grip strength. At a later stage of disease, when motor impairment is worsened by impaired grip strength, chlorzoxazone-baclofen does not cause further impairment in motor function. Taken together, these studies suggest that chlorzoxazonebaclofen can be dosed to properly engage target in the cerebellum while simultaneously minimizing off-target toxicity. Human clinical trials of chlorzoxazone-baclofen will determine whether this drug combination may be effective to also treat cerebellar motor impairment in human SCA.
In the transgenic ATXN1[82Q] mouse model of SCA1, chlorzoxazone-baclofen improves cerebellar Purkinje neuron physiology and motor impairment [9]. In the present study, chlorzoxazone-baclofen similarly improves motor function in Atxn1 154Q/2Q mice at an age when motor dysfunction is cerebellar in origin. In ATXN1[82Q] mice, the benefit of chlorzoxazonebaclofen persists at later stage disease, despite prominent Purkinje neuron dendritic degeneration seen at this disease stage [9]. Chlorzoxazone-baclofen was, however, unable to sustain improved motor function in Atxn1 154Q/2Q mice in later stage disease. This discrepancy may be explained by the genetic differences in these two models. The transgenic ATXN1[82Q] model of SCA1 overexpresses mutant ATXN1 specifically in cerebellar Purkinje neurons under control of the Pcp2 promoter [10]. Conversely, Atxn1 154Q/2Q mice express glutamine-expanded Atxn1 at its endogenous locus, resulting in widespread expression throughout the central nervous system [11]. As a result, motor impairment in ATXN1[82Q] mice is cerebellar in origin, even in late stage disease, and largely reflects Purkinje neuron dysfunction. Spinal motor neuron involvement in Atxn1 154Q/2Q mice likely contributes to motor impairment in this model of SCA1 [7,11]. The current study suggests that motor dysfunction in mid-stage disease in Atxn1 154Q/2Q mice is driven by cerebellar dysfunction, as evidenced by the improvement in motor function with chlorzoxazone-baclofen, whereas later stage motor impairment is a result of motor neuron dysfunction that is not improved by chlorzoxazone-baclofen. In this regard, Atxn1 154Q/2Q mice likely recapitulate features of human SCA1, where morbidity is primarily cerebellar in origin in earlier stages of disease, whereas end stage disease is due to cranial motor neuron dysfunction.
Chlorzoxazone and baclofen are used clinically as muscle relaxants and are not recommended for simultaneous dosing in patients due to concerns about tolerability [33]. A major concern of using central muscle relaxants in SCA is the potential to worsen motor impairment [34].
However, the present data suggest that chlorzoxazone-baclofen can engage target in cerebellar Purkinje neurons while avoiding toxicity resulting from impaired muscle strength. An optimal effect of combined chlorzoxazone and baclofen treatment would be to improve the regularity of spiking in SCA1 Purkinje neurons without slowing firing frequency, as both irregular spiking and slow Purkinje neuron firing are associated with motor impairment in mouse models of ataxia [5, 14-17, 20, 35]. In the current study, we identify a tight relationship between drug concentrations needed to engage target, and levels above which cerebellar function will be compromised. In planning for a human clinical trial with this drug combination to improve cerebellar ataxia, it will be important to identify the specific orally administered dose of the combination of chlorzoxazone-baclofen that achieves appropriate brain levels. We propose that a scheme of sequential dose escalation within each clinical trial-subject would be the most effective way to identify doses of chlorzoxazone-baclofen that engage target optimally without compromising cerebellar function, by allowing for monitoring of plasma and potentially cerebrospinal fluid levels of chlorzoxazone-baclofen along with effects on cerebellar motor function. In the current study, a high cerebellar concentration of chlorzoxazone-baclofen (50 µM and 2 µM, respectively) impairs cerebellar function, while a low concentration of chlorzoxazone-baclofen (5 µM and 0.2 µM, respectively) fails to engage target. An intermediate concentration of chlorzoxazone-baclofen (10 µM and 0.4 µM, respectively) at the site of action in the cerebellum is associated with optimal improvements in cerebellar physiology. Brain concentrations of baclofen are ~5 fold lower than plasma concentrations in both humans and mice [36]. Our studies in mice suggest that brain and plasma concentrations of chlorzoxazone reach near equilibrium in mice following chronic administration [9]. We therefore propose, based on the in vitro data obtained in the current study, that a plasma concentration range between 8-40 µM chlorzoxazone and 1-4 µM baclofen will be associated with improvement in cerebellar function in human SCA. It is likely that the relationship between administered drug dose, plasma/CSF concentrations of chlorzoxazone-baclofen, and changes in cerebellar function with each drug dose will need to be closely monitored for individual trial participants in a potential efficacy study.
Interestingly, the ion channels that underlie altered Purkinje neuron physiology in Atxn1 154Q/2Q mice are also implicated in neuronal dysfunction in other mouse models of SCA. These channels, BK (encoded by Kcnma1) and CaV3.1 (encoded by Cacna1g), show reduced expression in another mouse model of SCA1 [2], several mouse models of SCA2 [2], and a mouse model of SCA7 [5]. BK channels rely upon intracellular calcium for activation, and recent evidence suggests that CaV3.1 may be an important calcium source for BK [2,5]. Despite SCA1, SCA2, and SCA7 resulting from polyglutamine expansions in distinct genes, dysfunction of a BK channel functional module appears to contribute to altered Purkinje neuron spiking and cerebellar motor impairment in all of these disorders [2]. As illustrated in the present study, and in the ATXN1[82Q] model of SCA1 [9], chlorzoxazone-baclofen improves Purkinje neuron physiology through engagement of potassium channel targets. Therefore, it is possible that chlorzoxazone-baclofen may have relevance for treating motor impairment beyond SCA1. This possibility has not been tested in rodent models, although a related genetic strategy of viral BK channel replacement improves Purkinje neuron physiology in SCA7 [5]. It is tempting to speculate that a shared mechanism of dysfunction exists across multiple etiologies of SCA, and that a treatment strategy such as chlorzoxazone-baclofen may therefore have wider relevance for targeting cerebellar motor impairment.