The humanised CYP2C19 transgenic mouse exhibits cerebellar atrophy and movement impairment reminiscent of ataxia

Abstract

Aims: CYP2C19 transgenic mouse expresses the human CYP2C19 gene in the liver and developing brain, and it exhibits altered neurodevelopment associated with impairments in emotionality and locomotion. Because the validation of new animal models is essential for the understanding of the aetiology and pathophysiology of movement disorders, the objective was to characterise motoric phenotype in CYP2C19 transgenic mice and to investigate its validity as a new animal model of ataxia.

Methods: The rotarod, paw-print and beam-walking tests were utilised to characterise the motoric phenotype. The volumes of 20 brain regions in CYP2C19 transgenic and wild-type mice were quantified by 9.4T gadolinium-enhanced post-mortem structural neuroimaging. Antioxidative enzymatic activity was quantified biochemically. Dopaminergic alterations were characterised by chromatographic quantification of concentrations of dopamine and its metabolites and by subsequent immunohistochemical analyses. The beam-walking test was repeated after the treatment with dopamine receptor antagonists ecopipam and raclopride.

Results: CYP2C19 transgenic mice exhibit abnormal, unilateral ataxia-like gait, clasping reflex and 5.6-fold more paw-slips in the beam-walking test; the motoric phenotype was more pronounced in youth. Transgenic mice exhibited a profound reduction of 12% in cerebellar volume and a moderate reduction of 4% in hippocampal volume; both regions exhibited an increased antioxidative enzyme activity. CYP2C19 mice were hyperdopaminergic; however, the motoric impairment was not ameliorated by dopamine receptor antagonists, and there was no alteration in the number of midbrain dopaminergic neurons in CYP2C19 mice.

Conclusions: Humanised CYP2C19 transgenic mice exhibit altered gait and functional motoric impairments; this phenotype is likely caused by an aberrant cerebellar development.

Keywords: animal models; cerebellar ataxia; cerebellum; cytochrome P-450 Cyp2c19; movement disorders; neuroimaging; transgenic mice.

FIGURE 1

Motoric phenotype in CYP2C19 transgenic mice. CYP2C19 transgenic mice exhibited significantly slower weight gain during development (F _1,51 = 6.29, p = 0.015) until postnatal day 31 when this difference stops being significant (A). CYP2C19 transgenic mice experience abnormal hind paw elevation in youth (2.1‐fold [95% CI: 2.0, 2.2], F _1,51 = 415, p < 0.0001) and in adulthood (1.9‐fold [95% CI: 1.8, 2.1], F _1,28 = 129, p < 0.0001) (B) and clasping reflex (C). No abnormalities were found in CYP2C19 transgenic mice in any of the parameters measured in the footprint test (D) and rotarod test (F _1,97 = 0.008, p = 0.93) (E). CYP2C19 transgenic mice performed worse than wild‐types in the beam‐walking test with longer (1.14‐fold, [95% CI: 1.06, 1.22], F _1,83 = 10.9, p = 0.0014) beam‐crossing time and a greater number of paw slips (CYP2C19 transgenic: 1.7 [interquartile range (IQR): 1.0, 2.3] vs wild‐type 0.33 [IQR: 0.0, 0.92], U = 298, p < 0.0001) (F). Data presented as means with whiskers/area as 95% CI or as Tukey box‐plot. *Statistical significance p < 0.05.

FIGURE 2

Structural changes in brains of CYP2C19 transgenic mice measured with 9.4T magnetic resonance imaging (MRI) neuroimaging. Volumetric differences between CYP2C19 transgenic mice and wild‐types analysed with voxel‐based morphometry are presented as (A) 3D rendering and (B) coronal cross‐sections. Panel (C) presents magnitudes of unadjusted volumetric changes in CYP2C19 transgenic mice in 20 segmented regions of interest: cerebellar volume (−11.8% [95% CI: −14.7%, −9.0%], F _1,57 = 69.1; p < 0.0001, q < 0.0001), hippocampus (−4.2% [95% CI: −6.1%, −2.3%], F _1,57 = 18.4, p = 0.0001, q = 0.0027), pons (−3.9% [95% CI: −5.6%, −2.1%], F _1,57 = 19.6, p < 0.0001, q = 0.0017), epithalamus (−4.6% [95% CI: −7.4%, −1.9%], F _1,57 = 12.5, p = 0.0008, q = 0.030) and inferior colliculus (−4.2% [95% CI: −6.4%, −1.9%], F _1,57 = 14.1, p = 0.0004, q = 0.015). Bars and whiskers represent means and 95% CI.

FIGURE 3

Antioxidant enzyme activity in three brain regions. Compared with wild‐types, CYP2C19 transgenic mice exhibited (A) increased superoxide dismutase (SOD) activity in the cerebellum (1.14‐fold [95% CI: 1.06, 1.23], F _1,60 = 7.0, p = 0.0010, q = 0.023), (B) increased hippocampal activity of SOD (1.3‐fold, [95% CI: 1.18, 1.47], F _1,59 = 18.7, p < 0.0001, q = 0.0013) and glutathione reductase (GR, 1.2‐fold [95% CI: 1.13, 1.35], F _1,58 = 17.5, p < 0.0001, q = 0.0021) and (C) increased activity of glutathione peroxidase (GPx) in the whole hemisphere (1.38‐fold [95% CI: 1.16, 1.59], F _1,57 = 11.9, p = 0.0011, q = 0.023). Graphic representations of the brain were adopted from the reference atlas [27]. Data presented as means with whiskers as 95% CI or as Tukey box‐plot. Bars and whiskers represent means and 95% CI. *Statistical significance p < 0.05.

FIGURE 4

CYP2C19 mice exhibit hyperdopaminergism in adulthood but not during development. There was an initial statistical trend towards 21% (t ₂₀ = −1.95, p = 0.065) increased dopamine concentrations in the cerebella of CYP2C19 mice, which was no longer present after p‐value correction (q = 0.45). Also, homovanillic acid (HVA) level was 46% ([95% CI: 21%, 71%], t ₂₁ = −3.52, p = 0.0020, q = 0.016) increased in CYP2C19 mice compared with controls (A). Hippocampal levels of dopamine, 3,4‐dihydroxyphenylacetic acid (DOPAC) and HVA were not different between test groups (B). CYP2C19 transgenic mice were hyperdopaminergic with slightly increased (1.15‐fold [95% CI: 1.08, 1.23]; t ₂₂ = −4.26, p = 0.0003, q = 0.0029) dopamine concentrations in the hemisphere compared to wild‐types (C). Graphic representations of the brain [27] and embryo [28] were adopted from reference atlases. Bars and whiskers represent means and 95% CI. *Statistical significance p < 0.05.

FIGURE 5

The dopaminergic system is not associated with motoric impairment in CYP2C19 transgenic mice. Dopaminergic antagonists failed to ameliorate motoric impairment in CYP2C19 transgenic mice in the beam‐walking test, and they in fact slightly prolonged beam walking time (Ecopipam: 1.2‐fold [95% CI: 1.17, 1.30], F _1,76 = 13.9, p < 0.0001; raclopride: 1.2‐fold [95% CI: 1.12, 1.25], F _1,76 = 13.9, p < 0.0001) in them (A, B). Anti‐tyrosine hydroxylase (TH) immunohistochemistry (C) revealed no significant changes in the numbers of midbrain TH+ neurons (D) after false discovery rate (FDR) correction. Data presented as means with whiskers as 95% CI or as Tukey box‐plot. *Statistical significance p < 0.05.