Metabolomics and the equine microbiome: does diet matter?

Horses have evolved to spend long hours grazing on poor-quality pastures. However, this forage diet cannot provide the energy required for competition in racing, jumping or hunting.

Often, horses are then given a high-concentrate feed – usually barley or oats, and always carbohydrate-based.

Starch is the major carbohydrate in plants and contains two main components, both polymers made up of glucose units. Initially, these are broken down in the gut by amylase and other enzymes to provide small amounts of free glucose, which are then absorbed across the gut wall into the bloodstream.

The feed mass then moves to the lower gut, where the gut microbiome – a highly complex collection of microorganisms, including bacteria, fungi and protozoa – converts fibre to short-chain fatty acids (SCFAs), such as acetate propionate and butyrate, which are the major source of energy in horses.

These two processes provide the power needed for speed and endurance.

However, horses have relatively low levels of amylase activity. Two genes are involved – salivary amylase, AMY1; and pancreatic amylase, AMY2. Both genes are polymorphic in horses, with variations in the genomic sequences that may affect the capacity to break down starch.

On a high-starch diet, the amylase activity is not sufficient to extract all the energy from the feed, so much is wasted. Then, the undigested carbohydrate and fibre is passed down into the hindgut, where it starts to ferment with enrichment of lactic acid bacteria.

D-Lactate is normally converted to SCFA by bacteria such as Veillonellaceae; however, with a starch overload, different types of bacteria become dominant and acidification of the gut contents occurs as D-lactate accumulates. The gut wall becomes damaged and is more permeable to the toxins generated by the new microbiome, leading to a systemic proinflammatory state.

Metabolomics and the microbiome

The microbiome has been recognised as one of the most complex, important and least understood areas of mammalian physiology. However, it has been difficult to study, since the components are dynamically evolving and interacting with each other.

Nevertheless, all the species produce a characteristic spectrum of metabolites, which can be analysed by “metabolomics”. This is, as the name suggests, the science of metabolites – and its great advantage is that it is able to pick up all existing biochemical pathways, rather than only those that are already known to be involved (Hunter and Waring, 2017).

This is particularly useful when considering the gut microbiome, which contains a large number of species, many of which are not yet characterised and cannot be cultured. However, even though they cannot be identified, their presence leaves a biochemical “fingerprint” that reflects their biochemical pathways.

Generally, metabolites in tissues or excreta are separated by some form of liquid chromatography, then analysed after further separation by mass spectrometry or nuclear magnetic resonance.

Using in-depth computer analysis, a “profile” can be obtained, which is produced by the complex range of species present; any changes can then be monitored. This technique is of critical importance when studying multifactorial physiological dysfunction or in identifying factors that had previously not been considered.

An example of this has been shown using a population of 45 horses and ponies (age range 4 to 23 years old) at Redwings Horse Sanctuary in Norfolk. A study was carried out from early May to mid-June 2018 – for inclusion, the animals had to be older than two years of age, with no physical evidence of disease, and resident at the sanctuary for more than one year. Faecal samples were collected and the volatile organic compounds (VOCs) were characterised by selected-ion flow-tube mass spectrometry, in conjunction with multivariate data analysis (Snalune et al, 2019).

Before being put out to pasture, the horses had a volatile faecal metabolome (VFM) dominated by organic acids, alcohols and ketones. However, marked changes were detected after six weeks – showing that grazing on the spring pasture dramatically altered the VFM.

Unlike humans, horses have a very diverse faecal microbiome – the number of unique operational taxonomic units with more than 97% nucleotide sequence identity may be up to 2,000 in humans, but, in horses, ranges from 2,500 to 10,500 (Proudman et al, 2015).

Many very low-abundance bacterial populations exist so that the equine microbiota is both rich and complex, and potentially interactive; this may predispose to instability and has been suggested as explaining the susceptibility of horses to digestive problems (Dougal et al, 2013).

The microbiome and disease states

Apart from infections caused by pathogens, a number of disease states have been linked with changes in the microbiome (gut dysbiosis). This is true of both animals and humans (Walton et al, 2016).

Equine colic is a frequent cause of emergency treatment in horses. Although many factors – both environmental and genetic – may predispose to the condition, excessive consumption of carbohydrate-rich diets increases the risk (Shirazi-Beechey, 2008).

Laminitis may, similarly, be triggered by starch overload and high intakes of fructans from new grass. It is associated with an abnormal intestinal microbiome (Biddle et al, 2013; Steelman et al, 2012), and overgrowth of Gram-positive and Gram-negative intestinal bacteria (Onishi et al, 2012).

The non-structural carbohydrate composition of pasture has both diurnal and seasonal variations, reflecting exposure to sunlight, and is highest in late spring; increases in fructan levels are known to occur and these have been linked with the subsequent development of laminitis.

None of the horses in the authors’ grazing study had colic or became lame (Snalune et al, 2019), possibly because the total daily forage intake was consumed over a long period, rather than concentrated into one or two feeds. Nevertheless, the results do suggest, as consumption of spring grass itself alters the equine gut microbiome, these changes may induce a window of susceptibility by destabilising the bacterial profile and, therefore, potentially dysregulating the enteroinsular axis (EIA). If this is the case, it may partly explain why the incidence of laminitis increases at this time of year.

Laminitis has been shown to be multifactorial. Prolonged exposure to hyperinsulinaemia can cause the condition and endocrinopathies have been identified in 90% of ponies with lameness due to laminitis (Patterson-Kane et al, 2018).

Hyperinsulinaemia has been considered a response to the insulin resistance (IR) often seen in equids, but now seems to be driven by a gastrointestinal aetiology. It is clear hormones from the proximal intestine – such as incretin – alter insulin regulation, and that their release partly depends on the gut microbiome and diet (de Laat et al, 2016).

In turn, hyperinsulinaemia and IR are principal components of equine metabolic syndrome (EMS) with obesity and lipid dysregulation. An “EIA” has been proposed to describe the synergy between the microbiome, and the endocrine and enteric nervous systems (de Laat et al, 2016).

Attempts to improve the microbiome

Given that the gut bacteria are clearly involved in equine chronic disease states, much interest has existed in attempts to manipulate their effects.

Using fistulated horses (Grimm et al, 2017), it was shown bacterial analyses of faecal samples could be used to represent hindgut microbial ecosystems and this method has been widely used.

Several groups have shown feeding starch-rich diets alters the equine microbiome (Grimm et al, 2017; Kristoffersen et al, 2016; Warzecha et al, 2017) and care in not overfeeding soluble carbohydrate remains a cornerstone of equine husbandry.

Digestive aids have been used, with varying amounts of success, to improve the digestibility of equine diets (Coverdale, 2016). Feeding lactic acid bacteria had some limited effects on nutrient digestibility (Swyers et al, 2008), as did live yeast (Saccharomyces cerevisiae) cultures (Jouany et al, 2008), while dietary supplementation with short-chain fructooligosaccharides improved insulin sensitivity in obese horses (Respondek et al, 2011).

The use of probiotics in horses has been reviewed (Schoster, 2018), which concluded few well-designed studies existed and the results were conflicting, so any beneficial effects could not be supported.

Generally, prebiotics and probiotics have given modest improvements at best.

Use of enzyme-rich malt supplements

An alternative approach to manipulation of the microbiome is to change it indirectly by modifying undigested food residues entering the lower bowel.

In a study with Thoroughbred racehorses in active race training, an enzyme-rich preparation of malted barley was used as a dietary supplement (Proudman et al, 2015). This provided extra amylase to increase digestion of starch and reduce starch entry into the colon. Faecal samples were collected and VOCs were analysed by thermal desorption/gas chromatography/mass spectrometry.

The VOC profiles changed significantly after supplementation, reflecting changes in the metabolic activity of the microbiota – and this was confirmed by using error-corrected 454 pyrosequencing data from 165 ribosomal RNA gene amplicons.

The authors concluded dietary intervention induced metabolic adaptation of existing bacterial communities, probably within a defined range, and the inter-horse responses were not identical. Such individual variation between Thoroughbreds of similar genetic background, where all were on almost identical diets, confirms the complexity of the problem.

This research, with a population of Thoroughbred equines on a controlled diet, was then extended to determine whether similar results would be obtained with a wider range of non-Thoroughbred horses on standard pasture. It was assumed this would be a better reflection of the general pleasure horse population in the UK and would provide information on horses that were not being fed high-carbohydrate diets.

A follow-on study with metabolomics and the same malt-based supplement was, therefore, undertaken in the aforementioned Redwings horses. Again, six weeks of supplementation altered the VOCs and improved nutrition, as shown by reduced excretion of SCFAs (Snalune et al, 2019).

Although alterations in the gut microbiome are certainly of academic interest, the real question is whether this translates into improved performance on the racetrack.

A leading racehorse trainer introduced the malt-based supplement into the feeding schedules on his yard. Increases in performance are notoriously difficult to measure, since, to reduce the risk of injury, training gallops are never at full speed and the going on the track varies with many factors. However, a very significant increase occurred in both the number of winners and prize money won by the yard compared before and after the introduction of the supplement.

It is of interest that the supplement not only provided better feed utilisation, as shown by the improved performances, but also benefited gut health, as shown by the horses’ stool samples.

Use of the malt-based preparation on horses on a high-concentrate diet was found to change the microbiome, reversing it to what would be expected on a diet higher in forage. A further advantage was that horses found the product very palatable and its use improved consumption of the forage/concentrates diet – even in “finicky feeders”.

Conclusions and future prospects

The importance of the microbiome in equine health means increasing attempts will be made to control its composition.

To date, enzyme supplementation seems to offer the most exciting prospects. Such a supplement derived from malt increases the palatability of a forage diet and improves digestion in equines. It also alters the gut microbiome, reducing the levels of more pathogenic species, which, in turn, will reduce inflammation.

These results have been obtained in both racehorses and non-Thoroughbreds, so have general application. Use of this malt preparation would provide the enzymes, enabling horses to gain more nutrition from both forage-based and high carbohydrate diets.

As systemic inflammatory processes seem to be involved in so much chronic equine disease, supplements of this type may, therefore, have the potential to improve horses’ health. This, with the possibility of a return to a more natural diet, is an exciting prospect.