Ataxia with loss of Purkinje cells in a mouse model for Refsum disease

Abstract

Refsum disease is caused by a deficiency of phytanoyl-CoA hydroxylase (PHYH), the first enzyme of the peroxisomal alpha-oxidation system, resulting in the accumulation of the branched-chain fatty acid phytanic acid. The main clinical symptoms are polyneuropathy, cerebellar ataxia, and retinitis pigmentosa. To study the pathogenesis of Refsum disease, we generated and characterized a Phyh knockout mouse. We studied the pathological effects of phytanic acid accumulation in Phyh(-/-) mice fed a diet supplemented with phytol, the precursor of phytanic acid. Phytanic acid accumulation caused a reduction in body weight, hepatic steatosis, and testicular atrophy with loss of spermatogonia. Phenotype assessment using the SHIRPA protocol and subsequent automated gait analysis using the CatWalk system revealed unsteady gait with strongly reduced paw print area for both fore- and hindpaws and reduced base of support for the hindpaws. Histochemical analyses in the CNS showed astrocytosis and up-regulation of calcium-binding proteins. In addition, a loss of Purkinje cells in the cerebellum was observed. No demyelination was present in the CNS. Motor nerve conduction velocity measurements revealed a peripheral neuropathy. Our results show that, in the mouse, high phytanic acid levels cause a peripheral neuropathy and ataxia with loss of Purkinje cells. These findings provide important insights in the pathophysiology of Refsum disease.

Fig. 1.

Generation of targeting vector and Phyh^−/− mice. (A) Strategy of gene targeting. The WT allele, targeting vector and the targeted allele are depicted. After homologous recombination, exons 4–7 and parts of introns 3 and 7 are deleted. The location of the 5′ probe used for Southern blot analysis is shown. (B) Southern blot analysis of genomic DNA isolated from ES cells, digested with EcoRV and BamHI, and probed with the 5′ probe yielded a 12.3-kb fragment for the WT allele (+/+) and a 10-kb fragment for the targeted allele (+/−). (C) Genotyping by PCR on genomic DNA. Amplification from the WT allele (+/+) resulted in a 201-bp fragment and amplification from the targeted allele (−/−) a fragment of 260 bp. (D) (Upper) PCR analysis of liver cDNA from WT and Phyh^−/− mice. No amplification of exons 1–8 could be detected in Phyh^−/− mice. (Lower) A control for cDNA amplification was carried out with primers for β-actin.

Fig. 2.

Phytanic acid levels in plasma and tissues of male WT and Phyh^−/− mice. Phytanic levels were determined in liver (white bars), kidney (gray bars), cerebellum (black bars), testis (checked bars), and plasma (Inset graph) from WT and Phyh^−/− mice on a control diet (C), a 0.1% phytol diet, and a 0.25% phytol diet. Data represent mean ± SD.

Fig. 3.

Phyh^−/− mice on a phytol diet develop hepatic lipidosis. H&E staining of livers from WT (A–C) and Phyh^−/− mice (D–F) on a control diet (A and D), on a 0.1% phytol diet (B and E), and on 0.25% phytol diet (C and F). In Phyh^−/− mice on a 0.25% phytol diet (F), steatosis is clearly detected by the large lipid vacuoles present throughout the liver parenchyma. On a 0.1% phytol diet (E), Phyh^−/− mice showed signs of microsteatosis (small lipid vacuoles within hepatocytes). (Scale bar: 100 μm.)

Fig. 4.

Loss of spermatogonia in phytol-fed Phyh^−/− mice. Immunohistochemical detection of calreticulin in testis of WT mice (A–C) and Phyh^−/− (D–F) mice on control diet (A and D), 0.1% phytol diet (B and E), and 0.25% phytol diet (C and F). Calreticulin is highly expressed in spermatogonia of WT mice on control diet (A). Loss of spermatogonia in Phyh^−/− mice on the 0.1% phytol (E) and 0.25% phytol (F) diets is demonstrated by decreased numbers of calreticulin-positive spermatogonia leaving gaps (arrows in E and F) in the epithelium of the seminiferous tubules. The slide is hematoxylin QS counterstained. (Scale bar: 50 μm.)

Fig. 5.

Automated gait analysis using CatWalk revealed ataxia in Phyh^−/− mice on a phytol diet. (A) Paw prints of the fore- and the hindpaws for representative runs by male WT and Phyh^−/− mice on different diets are shown. The paw print area of both the fore- and hindpaw of the Phyh^−/− mice on 0.25% phytol were markedly decreased. (B) Successive frames of the paw–floor contact area for fore- and hindpaws revealed a decreased contact area for male Phyh^−/− mice compared with WT mice on a 0.25% phytol diet. C, control diet; P, 0.25% Phytol diet. (C) Base of support for the hindpaws (white bars) was significantly decreased (*, P < 0.05, t test) in the Phyh^−/− mice on the phytol diets compared with the base of support of the WT mice on the same diets. The decrease was also significant for the Phyh^−/− mice before and after the 0.25% phytol diet (data not shown). There were no differences for base of support of the forepaws (black bars).

Fig. 6.

Loss of Purkinje cells in the cerebellum of Phyh^−/− mice fed phytol. Immunohistochemical detection of calbindin-D28K in the cerebellum of WT (A–C) and Phyh^−/− (D–F) mice fed a control diet (A and D), a 0.1% phytol diet (B and E), and a 0.25% phytol diet (C and F). Small areas lacking Purkinje cells are evident in the cerebellum of Phyh^−/− mice fed the 0.1% phytol diet (E). Widespread loss of Purkinje cells is observed in the cerebellum of Phyh^−/− mice fed the 0.25% phytol diet, with only a few surviving Purkinje cells (F). The slide is hematoxylin QS counterstained. (Scale bar: 100 μm.)