Spinocerebellar ataxia type 6 knockin mice develop a progressive neuronal dysfunction with age-dependent accumulation of mutant CaV2.1 channels

Abstract

Spinocerebellar ataxia type 6 (SCA6) is a neurodegenerative disorder caused by CAG repeat expansions within the voltage-gated calcium (Ca(V)) 2.1 channel gene. It remains controversial whether the mutation exerts neurotoxicity by changing the function of Ca(V)2.1 channel or through a gain-of-function mechanism associated with accumulation of the expanded polyglutamine protein. We generated three strains of knockin (KI) mice carrying normal, expanded, or hyperexpanded CAG repeat tracts in the Cacna1a locus. The mice expressing hyperexpanded polyglutamine (Sca6(84Q)) developed progressive motor impairment and aggregation of mutant Ca(V)2.1 channels. Electrophysiological analysis of cerebellar Purkinje cells revealed similar Ca(2+) channel current density among the three KI models. Neither voltage sensitivity of activation nor inactivation was altered in the Sca6(84Q) neurons, suggesting that expanded CAG repeat per se does not affect the intrinsic electrophysiological properties of the channels. The pathogenesis of SCA6 is apparently linked to an age-dependent process accompanied by accumulation of mutant Ca(V)2.1 channels.

Fig. 1.

Generation of Sca6 KI mice. (A) Targeting scheme represents the targeting construct, the endogenous Sca6 allele, and the predicted structure of the mutant Sca6 allele generated by a homologous recombination and a Cre-mediated excision event. (B) Germ-line transmission of a Sca6^30Q allele. Southern blot analysis of HindIII-digested tail DNA revealed 6.8-kb WT and 3.8-kb mutant bands in Sca6^30Q/30Q mice with the internal probe shown in A. (C) Western blot analysis performed on cytoplasmic fraction of cerebellar lysates from 6-month-old Sca6^14Q/14Q, Sca6^30Q/30Q, and Sca6^84Q/84Q KI mice and a 6-month-old WT mouse, all blotted against CT-2 (Upper) and AKT (Lower, used as a loading control). Arrows indicate multiple CT-2 positive IRs detected with WT cerebellar lysates. Mutant Ca_V2.1 shows less mobility because of the polyglutamine stretch and its flanking human-specific sequence. (D) Detection of aggregated Ca_V2.1 channel in Sca6^84Q cerebella. Shown are immunoblots of cerebellar extracts from 2- (2m) or 15-month-old (15m) Sca6^14Q/14Q (14Q/14Q), Sca6^300Q/30Q, Sca6^84Q/+, Sca6^840Q/84Q KI, and WT mice. Arrow indicates CT-2 IRs detected in the stacking part of the gels. Molecular masses are indicated at the right of each panel in kDa.

Fig. 2.

Analysis of the Sca6 KI mice on the accelerating Rotarod apparatus. (A and B) Performance of 3- (A) and 7-month-old (B) Sca6^84Q84Q mice of 129/SvEv background. Sca6^84Q/84Q showed impaired motor performance (P = 0.035) at 7 months of age. (C) Performance of 19-month-old heterozygous Sca6^84Q/+ mice of C57BL/6J-129/SvEv background. (D) Performance of 19-month-old heterozygous Sca6^30Q/+ mice of C57BL/6J-129/SvEv background. Mice were trained in four trials per day (T1–T4) for four days (D1–D4). Error bars indicate SEM.

Fig. 3.

Whole-cell Purkinje Ba²⁺ currents are blocked by Aga-IVA and reduced in Sca6 neurons. (A) Time course of peak current for a WT (○) or an Sca6^84Q84Q (■) neuron. Aga-IVA was added to the bath when indicated, to a final bath concentration of ≈500 nM. (B) Individual traces from the two cells in A, before and after bath application of Aga-IVA. (C–F) Each group consists of 4–19 neurons. (C) Current density was plotted against voltage for WT (○) and Sca6 ^84Q/84Q (■) neurons. (D) Same data as in A, normalized to peak inward current. (E) Summary of whole-cell current density at −15 mV. *, P < 0.005 vs. WT (by two-tailed Student's unpaired t test). (F) Summary of whole-cell capacitance. P > 0.05 vs. WT for all three Sca6 populations. 14Q/14Q, 30Q/30Q, and 84Q/84Q represent Sca6^14Q/14Q, Sca6^30Q/30Q, and Sca6^84Q/84Q, respectively.

Fig. 4.

Sca6 mutations do not affect activation or inactivation. (A) Protocol used to measure activation (G–V) curves. Shown are currents from a WT PC. The cell was held at −80 mV and depolarized for 60 ms to incremental voltages from −80 through +20 mV. Peak tail current amplitude was normalized and plotted against voltage to generate the curves in C. (Scale bars: vertical, 50 pA/pF; horizontal, 3 ms.) (B) Protocol used to measure steady-state inactivation for the same cell shown in A. The cell was held at −80 mV, and a 3-s conditioning pulse was applied in 10-mV increments from −120 through +20 mV followed by a test pulse to −20 mV. Peak current amplitude during the test pulse was normalized and plotted against conditioning pulse voltage to generate the curves in C. (C) Summary of activation and inactivation curves for WT (○) and Sca6^84Q84Q (■) PCs. Solid lines are Boltzmann fits of the data. For activation, WT and Sca6^84Q/84Q V_0.5 are −34.0 ± 0.9 and −33.1 ± 0.5 mV, respectively (P >0.4, by two-tailed Student's unpaired t test); slope factors (k) are 4.8 ± 0.2 and 4.8 ± 0.1, respectively (P > 0.9). For inactivation, WT and Sca6^84Q/84Q V_0.5 are −41.8 ± 1.0 and −41.6 ± 0.8 mV, respectively (P > 0.8); slope factors (k) are 10.0 ± 0.7 and 9.3 ± 0.5, respectively (P > 0.4). n = 10–13 neurons.

Fig. 5.

NI formation in Sca6^84Q84Q Purkinje neurons. (A) Immunohistochemistry for A6RPT-polyQ revealed NI formation in Sca6^84Q/84Q PCs (22-month-old). (Inset) Individual PC at higher magnification. Note that these NIs were seen mainly in the cytoplasm of these neurons. Similar NI-harboring PCs were scattered in the cerebellum. (B and C) In contrast, only weak immunoreactivities were detected in the cerebellum of a 22-month-old Sca6^14Q/14Q (B) or an 8-month-old Sca6^84Q/84Q mouse (C). (D) The immunoreactivity was entirely blocked in the cerebellum of the 22-month-old Sca6^84Q/84Q mouse when A6RPT-polyQ was preincubated with its polypeptide antigens. (Magnification of original photographs, ×400.) Parallel staining undertaken with control human specimen also revealed similar NI formation in SCA6 PCs (data not shown).