The metabolomic profile of a high starch versus no starch diet in athletic horses

TL;DR

This study compares the plasma metabolomic profiles of athletic horses fed a high-starch diet versus a no-starch diet, revealing significant differences in metabolite concentrations.

Contribution

This is the first study to use metabolomics to investigate the effects of high-starch diets in horses.

Findings

Diet F increased concentrations of metabolites like 2-hydroxybutyrate and citrate, while diet FC increased glycine.

A PLS-DA analysis successfully discriminated between the two diets with good predictive power.

Diet F was associated with host-microbial co-metabolism-related metabolites.

Abstract

Feeding a high amount of starch-rich grains is common practice for performance horses even though the horse has evolved to eat a grass based, i.e. low starch diet. To our knowledge, there are no studies using metabolomics to investigate the effects of a high-starch diet in horses. In this study we investigated differences in the plasma metabolic profile of 6 Standardbred horses fed a no-starch, forage-only (F) diet or a high-starch forage-concentrate (FC) diet for 29 days, respectively in a cross-over design. Postprandial plasma samples were collected on the morning of day 25 of each dietary period. Metabolomics analysis of plasma using a targeted 1H NMR resulted in the quantification of 52 metabolites. Both a univariate and multivariate analysis of metabolites was performed. The univariate analysis found increased (p < 0.05) plasma concentrations of 2-hydroxybutyrate, citrate, dimethyl sulfone, hippurate, methionine, myo-inositol and proline in diet F and higher concentrations of glycine in diet FC. A PLS-DA analysis could discriminate between diets with good predictive power (Q2 (cum) = 0.745, p = 0.032 in CV-ANOVA). We conclude that diet F was strongest identified by metabolites originating from host-microbial co-metabolism and that the clear metabolomic profile discrimination between diets may have implications for health, performance and behaviour.

The online version contains supplementary material available at 10.1038/s41598-025-23422-z.

Figures