Abstract
Research suggests contact sports affect neurological health. This study used permutation-based mediation statistics to integrate measures of metabolomics, neuroinflammatory miRNAs, and virtual reality (VR)-based motor control to investigate multi-scale relationships across a season of collegiate American football. Fourteen significant mediations (six pre-season, eight across-season) were observed where metabolites always mediated the statistical relationship between miRNAs and VR-based motor control ( ppermSobelpSobelperm<math> <mrow><msubsup><mi>p</mi> <mrow><mi>S</mi> <mi>o</mi> <mi>b</mi> <mi>e</mi> <mi>l</mi></mrow> <mrow><mi>p</mi> <mi>e</mi> <mi>r</mi> <mi>m</mi></mrow> </msubsup> </mrow> </math> ≤≤<math><mrow><mo>≤</mo></mrow> </math> 0.05; total effect >><math><mrow><mo>></mo></mrow> </math> 50%), suggesting a hypothesis that metabolites sit in the statistical pathway between transcriptome and behavior. Three results further supported a model of chronic neuroinflammation, consistent with mitochondrial dysfunction: (1) Mediating metabolites were consistently medium-to-long chain fatty acids, (2) tricarboxylic acid cycle metabolites decreased across-season, and (3) accumulated head acceleration events statistically moderated pre-season metabolite levels to directionally model post-season metabolite levels. These preliminary findings implicate potential mitochondrial dysfunction and highlight probable peripheral blood biomarkers underlying repetitive head impacts in otherwise healthy collegiate football athletes.