A Proton NMR Metabolomic Investigation of an Emerging Genetic Disease

Resistance to thyroid hormone (RTHα) due to mutations in the thyroid hormone receptor alpha (TRα1), which is encoded by the THRA gene, is a recently discovered genetic disease. Patients present a high variability in clinical features (skeletal dysplasia, growth retardation, intellectual disability, etc...), and the absence of reliable biochemical markers make the diagnosis of this disease difficult. Since its first description in 2012, 25 cases of RTHα have been reported worldwide, corresponding to 14 different mutations of TRα1. In this context, a 1H nuclear magnetic resonance (NMR) metabolomic study was initiated, among a range of other phenotyping studies (including skeletal, blood and heart phenotype, neurodevelopment), to evaluate the impact of different mutations of TRα1 on the metabolism, and pos- sibly identify biomarkers contributing to RTHα diagnostic in humans. Considering that the mouse Thra and human THRA genes display extensive sequence similarities, mouse lines with Thra mutations are highly relevant animal models for RTHα. We used CRISPR/Cas9 genome editing to introduce 5 different germline mutations in the mouse Thra gene: 4 frameshift and 1 missense mutation closely modeling the mutations found in RTHα patients. Urine and plasma samples from adult mice carrying these Thra mutations were analyzed by NMR with associated wild-type controls. Multi- variate statistical analysis (OPLS-DA models) shows that samples collected from specific groups of mice carrying frameshift mutations can be discriminated from control samples collected from wild-type lit- termates, while mutations associated with minor phenotypic changes do not discriminate from controls. Our results reveal the presence of metabotype changes induced by Thra mutations, which provides a proof-of-principle that NMR metabolomics can be used to diagnose RTHα.