Metabolic disruption identified in the Huntington's disease transgenic sheep model

Abstract

Huntington's disease (HD) is a dominantly inherited, progressive neurodegenerative disorder caused by a CAG repeat expansion within exon 1 of HTT, encoding huntingtin. There are no therapies that can delay the progression of this devastating disease. One feature of HD that may play a critical role in its pathogenesis is metabolic disruption. Consequently, we undertook a comparative study of metabolites in our transgenic sheep model of HD (OVT73). This model does not display overt symptoms of HD but has circadian rhythm alterations and molecular changes characteristic of the early phase disease. Quantitative metabolite profiles were generated from the motor cortex, hippocampus, cerebellum and liver tissue of 5 year old transgenic sheep and matched controls by gas chromatography-mass spectrometry. Differentially abundant metabolites were evident in the cerebellum and liver. There was striking tissue-specificity, with predominantly amino acids affected in the transgenic cerebellum and fatty acids in the transgenic liver, which together may indicate a hyper-metabolic state. Furthermore, there were more strong pair-wise correlations of metabolite abundance in transgenic than in wild-type cerebellum and liver, suggesting altered metabolic constraints. Together these differences indicate a metabolic disruption in the sheep model of HD and could provide insight into the presymptomatic human disease.

Figure 1. Differential Abundance of Single Metabolites in OVT73 Cerebellum and Liver GC-MS metabolite analysis of cerebellum and liver in transgenic (n = 6) and wild-type (n = 6) sheep.

Metabolite abundance is expressed relative to the abundance of 2,3,3,3-d₄-alanine internal standard and back-transformed. (a) Phenylalanine and pyroglutamic acid abundance in cerebellum (b) Dodecanoic acid, myristic acid and nicotinic acid abundance in liver. False Discovery Rate adjusted P < 0.05*, Standard Least Squares. Error bars are the standard error of the mean back-transformed relative abundance.

Figure 2. Transgene-Altered Metabolite Correlations in the OVT73 Cerebellum Scatterplots display the correlation of the abundance of the 22 metabolite-pairs that had a transgene-altered correlation in the cerebellum (P < 0.05, Fisher r-to-z-transformation).

The Pearson correlation coefficient (r) for wild-type (top left) and OVT73 (top right) is shown for each metabolite pair. Metabolite abundances are normalised to 2,3,3,3-d₄-alanine internal standard, duplicate measurements for each sample averaged and expressed as log10-transformed arbitrary units. Wild-type (n = 6) is shown in blue, transgenic (n = 6) is shown in red. Ellipsoids are the 95% confidence interval for correlations. Triangles are rams, circles are ewes *Metabolites which were also differentially abundant (P < 0.05, Standard Least Squares).

Figure 3. Transgene-Altered Metabolite Correlations in the OVT73 Liver Scatterplots display the correlation of the abundance of the 6 metabolite-pairs that had a transgene-altered correlation in the liver (P < 0.05, Fisher r-to-z-transformation).

The Pearson correlation coefficient (r) for wild-type (top left) and OVT73 (top right) is shown for each metabolite pair. Metabolite abundances are normalised to 2,3,3,3-d₄-alanine internal standard, duplicate measurements for each sample averaged and expressed as log10-transformed arbitrary units. Wild-type (n = 6) is shown in blue, transgenic (n = 6) is shown in red. Ellipsoids are the 95% confidence interval for correlations. Triangles are rams, circles are ewes. *Metabolites which were also differentially abundant (P < 0.05, Standard Least Squares).