A novel early onset spinocerebellar ataxia 13 BAC mouse model with cerebellar atrophy, tremor, and ataxic gait

Abstract

Spinocerebellar ataxia 13 (SCA13) is an autosomal dominant neurological disorder caused by mutations in KCNC3. Our previous studies revealed that KCNC3 (Potassium Voltage-Gated Channel Subfamily C Member 3) mutation R423H results in an early-onset form of SCA13. Previous biological models of SCA13 include zebrafish and Drosophila but no mammalian systems. More recently, mouse models with Kcnc3 mutations presented behavioral abnormalities but without obvious pathological changes in the cerebellum, a hallmark of patients with SCA13. Here, we present a novel transgenic mouse model by bacterial artificial chromosome (BAC) recombineering to express the full-length mouse Kcnc3 expressing the R424H mutation. This BAC-R424H mice exhibited behavioral and pathological changes mimicking the clinical phenotype of the disease. The BAC-R424H mice (homologous to R423H in human) developed early onset clinical symptoms with aberrant gait, tremor, and cerebellar atrophy. Histopathological analysis of the cerebellum in BAC-R424H mice showed progressive Purkinje cell loss and thinning of the molecular cell layer. Additionally, Purkinje cells of BAC-R424H mice showed significantly lower spontaneous firing frequency with a corresponding increase in inter-spike interval compared to that of wild-type mice. Our SCA13 transgenic mice recapitulate both neuropathological and behavioral changes manifested in human SCA13 R423H patients and provide an advantageous approach to understanding the role of voltage-gated potassium channel in cerebellar morphogenesis and function. This mammalian in vivo model will lead to further understanding of the R423H allelic form of SCA13 from the molecular to the behavioral level and serve as a platform for testing potential therapeutic compounds.

Keywords: KCNC3; Purkinje cells; R424H; cerebellar atrophy; spinocerebellar ataxia 13.

Fig. 1.

Generation of BAC-R424H mice. (A) A plasmid restriction map of BAC clone containing an inserted 224,674 bp mouse genomic fragment encoding the Kcnc3 locus. (a) The order of the genes encoded in RP-23-34H20 BAC in an approximate scale. (b) Kcnc3 intron exon structure. (c) Intron1/Exon2 fragment highlighting the location of the NEO cassette insertion in intron 1 and the L423L/R424H mutagenesis site. (B) PCR genotyping confirmed BAC-R424H mice, carriers displayed 631bp band. (C) Data of qPCR for Kcnc3 and Calbindin from wild-type mice and BAC-R424H mice. Note: ** P<0.01, n=5 per group.

Fig. 2.

Neurological function impairment of BAC-R424H mice. (A, B) Tremor analysis in wild-type control mice and BAC-R424H mice: (A, B) Periodograms describe the relative power of frequencies present in data collected from fore paw tremors of wild-type (n=5, trials=2) and BAC-R424H mice (n=5, trials=2). The powerful spike of frequencies is difference between the wild-type mice (A, normal power spectrum), and the BAC-R424H mice (B, an abnormal power spike in frequency at 46.35 ± 4 Hz). (C–F) Spatiotemporal gait analysis in wild-type control mice and BAC-R424H mice: (C) Velocities of locomotion, (D) Residual stride length (cm), (E) Residual step widths (cm) of fore, and (F) Residual step widths (cm) of hind limbs. (G–I) Rotarod testing in wild-type control mice and BAC-R424H mice: Coordination on a rotating rod was evaluated for mice at ages of 3 months (G, n=5 per group), 6 months (H, n=6 per group) and 9 months (I, n=7 per group). Note: One-way ANOVA Tukey multiple comparison test was used to analyze the statistical differences about the latency to fall between wild-type mice and BAC-R424H mice. **P<0.01, *P<0.05.

Fig. 3.

Cerebellar aggressive atrophy in BAC-R424H mice. Representative 7T MRI images of wild-type and BAC-R424H mice at ages of 1 month (A, B), 3 months (C, D), and 8 months (E, F). (G) The volumes of whole mice brains at 1 month (n=5–6), 3 months (n=5–6) and 8 months (n=4). (H) The volumes of mice cerebellum: 28% decrease at 1 month, 40% at 3 months, and 50% at 8 months compared to the wild-type mice. Note: Graphs represent median, interquartile range from Mann-Whitney statistical analysis for each group. **P<0.01, *P<0.05.

Fig. 4.

Cerebellar atrophy in BAC-R424H mice. (A–C) Representative images of Cresyl violet staining and analysis of molecular layer (ML) thickness in mice cerebellum. Two measurements were made for each lobe of the cerebellum, as indicated by the red arrows (A, B), and analysis of molecular layer thickness (C) of mice at 1 month (n=4 per group, wild-type vs BAC-R424H: ~30% reduction), 3 months (n=4 per group, wild-type vs BAC-R424H: ~50% reduction), and 8 months (n=3 per group, wild-type vs BAC-R424H: 55% reduction). (D–F) Representative images of Calbindin positive Purkinje cells in mice cerebellum at 1 month and 3 months. Calbindin positive Purkinje cells (white arrow) in cerebellum of wild-type mice (D) and BAC-R424H mice (E), quantitative analysis data (F, n=4 per group). **P<0.01, *P<0.05.

Fig. 5.

Abnormal spontaneous firing of Purkinje cells of BAC-R424H mice. Sagittal cerebellar slices were prepared from 1-month-old mice. Cell-attached recordings were acquired with a Multiclamp 700B amplifier in voltage-clamp mode by visually identifying Purkinje cell somata in anterior lobule III using an upright microscope. On-line data acquisition and off-line data analysis were performed using Clampfit. (A) Representative trace from wild-type mice cerebellar slices, (B) Representative trace from BAC-R424H mice cerebellar slices. (C) Spontaneous firing frequency of Purkinje cells: wild-type (~75 Hz) vs. BAC-R424H (~30 Hz), (D) Co-efficient of variation (CV) of the Inter-spike intervals (ISI) of Purkinje cells. Data was presented as mean ± SEM and analyzed using Mann Whitney U-test. wild-type, 4–5 slices per mouse, n=4, and BAC-R424H: 4–5 slices per mouse, n=4. **P<0.01.