Characterisation of the faecal metabolome and microbiome of Thoroughbred racehorses.

REASONS FOR PERFORMING STUDY: The intestinal bacterial community of the horse is a key determinant of intestinal and whole body health. Understanding the bacterial community structure and function is an important foundation for studies of intestinal health and disease. OBJECTIVES: To describe the faecal bacterial community and volatile organic compounds (VOCs) of the faecal metabolome of healthy Thoroughbred racehorses and to characterise responses to dietary supplementation with amylase-rich malt extract. STUDY DESIGN: Intervention study. METHODS: Faecal samples were collected noninvasively before and 6 weeks after supplementation in 8 privately owned Thoroughbred racehorses in active race training. Faecal metabolome was characterised using thermal desorption-gas chromatography-mass spectrometry (TD-GC-MS), with spectral analysis performed using AMDIS and compared against the NIST database. Taxonomic description of the faecal microbiota was achieved using error-corrected 454 pyrosequencing data from 16S rRNA gene amplicons. RESULTS: The faecal metabolome of our study population was dominated by organic acids, alcohols and ketones. We identified 81 different VOCs only 28 of which were present in >50% of samples indicating functional diversity. Faecal VOC profiles differed between first and second sampling point, some VOCs being significantly reduced post supplementation, consistent with a marked response to dietary amylase-rich malt extract. Faecal microbiota was characterised as highly diverse; samples demonstrated verifiable diversity in the range 1200-3000 operational taxonomic units (OTUs) per individual. The methods used also describe high levels of infrequent, low abundance OTUs. Faecal microbial community structure was found to be different following dietary supplementation. Differences in several low abundance bacterial taxa were detected and also some evidence of interhorse variation in response. CONCLUSIONS: The volatile faecal metabolome of Thoroughbred racehorses is dominated by organic acids, alcohols and ketones; this study demonstrates that dietary supplementation with amylase-rich malt extract may significantly alter the profile of VOCs. The faecal microbiome is highly diverse, dominated by Firmicutes and Bacteroidetes. Small but significant changes in microbial community structure were detected following dietary supplementation. This study describes the faecal metabolome and microbiome of healthy Thoroughbred racehorses against which future studies of disease and dietary intervention can be benchmarked.

Dominant components of the Thoroughbred metabolome characterised by (1) H-nuclear magnetic resonance spectroscopy: A metabolite atlas of common biofluids.

REASONS FOR PERFORMING STUDY: Metabonomics is emerging as a powerful tool for disease screening and investigating mammalian metabolism. This study aims to create a metabolic framework by producing a preliminary reference guide for the normal equine metabolic milieu. OBJECTIVES: To metabolically profile plasma, urine and faecal water from healthy racehorses using high resolution (1) H-nuclear magnetic resonance (NMR) spectroscopy and to provide a list of dominant metabolites present in each biofluid for the benefit of future research in this area. STUDY DESIGN: This study was performed using 7 Thoroughbreds in race training at a single time point. Urine and faecal samples were collected noninvasively and plasma was obtained from samples taken for routine clinical chemistry purposes. METHODS: Biofluids were analysed using (1) H-NMR spectroscopy. Metabolite assignment was achieved via a range of one- and 2-dimensional experiments. RESULTS: A total of 102 metabolites were assigned across the 3 biological matrices. A core metabonome of 14 metabolites was ubiquitous across all biofluids. All biological matrices provided a unique window on different aspects of systematic metabolism. Urine was the most populated metabolite matrix with 65 identified metabolites, 39 of which were unique to this biological compartment. A number of these were related to gut microbial host cometabolism. Faecal samples were the most metabolically variable between animals; acetate was responsible for the majority (28%) of this variation. Short-chain fatty acids were the predominant features identified within this biofluid by (1) H-NMR spectroscopy. CONCLUSIONS: Metabonomics provides a platform for investigating complex and dynamic interactions between the host and its consortium of gut microbes and has the potential to uncover markers for health and disease in a variety of biofluids. Inherent variation in faecal extracts along with the relative abundance of microbial-mammalian metabolites in urine and invasive nature of plasma sampling, infers that urine is the most appropriate biofluid for the purposes of metabonomic analysis.