Pioneering a Method to Elucidate Metabolic Networks of Unknown Diseases: a Proof of Concept for Personalized Metabolomic Medicine With Equine Osteochondritis Dissecans

Abstract

Osteochondritis dissecans (OCD), a painful osteoarthritic disorder affecting cartilage and joints of equids, causes significant financial losses in the equine industry. Nearly 40% of all horses are afflicted with OCD. While it is known to have multiple causal factors, the underlying cause of OCD is unknown. Little is known about pathogenesis and etiology of OCD, and the pathophysiology of OCD at the metabolic level is poorly understood. Thus, no surefire cure exists for OCD aside from surgery, which frequently presents complications and has a lengthy recovery period filled with obstacles, and unreliable therapeutic drugs that give no guarantee of a cure. If the pathophysiology of OCD were to be elucidated at the metabolic level, a cure could be formulated targeting those implicated metabolites. This study pioneered the use of a novel statistical workflow comprising STOCSY, PCA, PLS-DA, and O-PLS-DA to (1) identify novel biomarkers that are implicated in the metabolic network of a poorly understood disease such as OCD, (2) utilize those identified biomarkers to accurately and precisely diagnose patients with OCD, and (3) generate a/elucidate the metabolic network for OCD, detailing/listing all metabolites, proteins, and gene mutations that are implicated in the pathogenesis of OCD. In sum, the workflow presented in this study takes a completely unknown disease and elucidates it in ways that are highly critical and useful for patient diagnosis, therapeutic formulation, and clinical treatment. Successfully performing a proof of concept of this method using equine OCD, this study found the metabolic network of OCD to be more vast than ever previously thought in the metabolomic and clinical literature. The metabolites hyaluronic acid, erythrose, benzoquinoneacetic acid, homogentisic acid, maleylacetoacetic acid, chondroitin sulfate, heparan sulfate, and sebacic acid and the genes ACAN, HGD, CSPG4, HSPG2 were found to be implicated in the OCD metabolic network. Clinical recommendations for treating OCD are made based on these metabolites. Moreover, novel signals were identified and assigned on the 1-H NMR equine spectrum, including a novel addition to the Glyc signal cluster, titled “GlycC”, and a new creatinine signal. The result was a fully identified 1-H NMR equine spectrum with all significant, implicated unknown peaks assigned for the first time. Established diagnostic integrals and ratios in the literature (GlycA and SPC/GlycA, respectively) were not found to be useful in diagnosing equine OCD; however, the newly discovered GlycC proved to be highly accurate in predicting OCD disease condition. Specifically, the SPC/GlycC ratio proved to be highly accurate in diagnosing OCD, and is recommended for use in clinical settings as a diagnostic tool for OCD. The successes of this study are very promising for the future of personalized medicine.