Cerebellar ataxias are a rare and incurable group of neurodegenerative disorders. Several ataxias, including SCA1, are inherited polyQ disorders caused by expansion of unstable CAG repeats. Improved understanding of the cellular mechanisms that drive the early stages of ataxia progression is therefore critical for identifying new therapeutic leads. However, interpretation can be complicated by adaptive structural changes if disease progression also occurs during the critical period of brain development. Here we take advantage of a conditional (doxycycline repressible), Purkinje neuron (PN)-specific mouse model of human ataxia, SCA1 (82Q expansion in the gene for ataxin-1) to determine specific changes in mature PN function during SCA1 progression distinct from wider structural disruption to PNs caused by 82Q during the critical period of cerebellar development.We used SCA1 mice at 6 and 12 weeks of age (82Q ON) and mice where 82Q was turned OFF by doxycycline (200mg/kg in feed) during gestation to age 6 weeks, but then turned ON by removal of doxycycline from 6-12 weeks (82Q OFF-82Q ON). In this way 82Q OFF-82Q ON mice received the same period of 82Q as the ataxic 6 week old 82Q ON mice but AFTER the critical period of cerebellar development. Motor performance and gait analysis revealed mild ataxia in 6 and 12 week old 82Q ON mice (P < 0.01, one and two-way ANOVAs) whereas 12 week old 82Q OFF-82Q ON mice behaved normally. PN electrical input resistance, action potential firing frequency, molecular layer thickness and climbing fibre extension were reduced in all 82Q ON mice (P < 0.0001, one-way ANOVAs) consistent with structural and functional disruption of the cerebellum, but all these parameters were normal in the pre-symptomatic 82Q OFF-82Q ON mice.In contrast, PNs from pre-symptomatic 82Q OFF-82Q ON mice and all ataxic 82Q ON mice exhibited abnormal, long-lasting parallel fibre (PF)-evoked mGluR1 mediated synaptic currents (P < 0.0001, two-way ANOVAs). This result suggests that enhanced mGluR1 signalling occurs before the onset of ataxia and independent of cerebellar structural disruption. mGluR1 activation can also mobilise PN calcium and we observed enhanced synaptic calcium responses in the PN outer dendrites (P < 0.05, one-way ANOVA) from 12 week 82Q ON PNs, but otherwise calcium handling was remarkably intact.To determine the functional significance of the enhanced mGluR1 synaptic current we administered a very low dose of the potent, negative allosteric modulator of mGluR1, JNJ 16259685 (0.03 mg/kg, sub cutaneous) to mice prior to an acute motor learning test. Remarkably, JNJ-treated 6 week and 12 week 82Q ON mice exhibited significantly improved performance compared with vehicle-treated 82Q ON mice (P < 0.0001, two way ANOVAs) whilst the performance of JNJ and vehicle-treated wild type mice was unaffected.Our results show that developing cerebellar PNs are more susceptible than mature PNs to structural and functional disruption caused by 82Q expansion in ataxin-1. We identified prolonged mGluR1 currents at the PF-PN synapse as an early cellular mechanism that drives SCA1 progression in this model, independent of complications from developmental structural disruption, and that may mark the beginning of PN degeneration. Furthermore, mild pharmacological moderation of mGluR1 partially restored motor learning in SCA1 mice and provides a potential new therapeutic lead for treating the early stages of ataxia.