Comparison of Fecal Microbiota of Horses Suffering from Atypical Myopathy and Healthy Co-Grazers

Abstract

Equine atypical myopathy (AM) is caused by hypoglycin A (HGA) and methylenecyclopropylglycine (MCPG) intoxication resulting from the ingestion of seeds or seedlings of some Acer tree species. Interestingly, not all horses pasturing in the same toxic environment develop signs of the disease. In other species, it has been shown that the intestinal microbiota has an impact on digestion, metabolism, immune stimulation and protection from disease. The objective of this study was to characterize and compare fecal microbiota of horses suffering from AM and healthy co-grazers. Furthermore, potential differences in fecal microbiota regarding the outcome of diseased animals were assessed. This prospective observational study included 59 horses with AM (29 survivors and 30 non-survivors) referred to three Belgian equine hospitals and 26 clinically healthy co-grazers simultaneously sharing contaminated pastures during spring and autumn outbreak periods. Fresh fecal samples (rectal or within 30 min of defecation) were obtained from all horses and bacterial taxonomy profiling obtained by 16S amplicon sequencing was used to identify differentially distributed bacterial taxa between AM-affected horses and healthy co-grazers. Fecal microbial diversity and evenness were significantly (p < 0.001) higher in AM-affected horses as compared with their non-affected co-grazers. The relative abundance of families Ruminococcaceae, Christensenellaceae and Akkermansiaceae were higher (p ≤ 0.001) whereas those of the Lachnospiraceae (p = 0.0053), Bacteroidales (p < 0.0001) and Clostridiales (p = 0.0402) were lower in horses with AM, especially in those with a poor prognosis. While significant shifts were observed, it is still unclear whether they result from the disease or might be involved in the onset of disease pathogenesis.

Keywords: Lachnospiraceae; MCPA-CoA; Ruminococcaceae; equine; fecal; hypoglycin A; intestinal; intoxication; microbiome; rhabdomyolysis.

Figure 1

Bacterial intrinsic diversity deduced from inverse Simpson index (A). Bacterial genus richness deduced from Chao1 index (B). Bacterial genus evenness deduced from Simpson index (C). Data are scatter dot plots at the genus level for individual horses in the three defined groups with the mean and standard deviation. Data with different superscript letters are significantly different at q < 0.05 (Kruskal–Wallis).

Figure 2

Nonmetric dimensional scaling with three axes of the three horses groups (co-grazers: CG, atypical myopathy survivors: AM-S and atypical myopathy non-survivors: AM-NS). Model stress is 0.092.

Figure 3

Changes in bacterial populations in the fecal content, assessed by 16S V1-V3 profiling. The bar chart depicts the relative abundance of the bacterial families (A) and bacterial genera (B) accounting for more than 1% of the total abundance in the feces. Co-grazers (CG) were compared to horses diagnosed with atypical myopathy (AM) with different outcomes (survivors, S, and non-survivors, NS).

Figure 4

Changes in bacterial family populations in the feces, assessed by 16S V1-V3 profiling and expressed as relative population abundance. Bar plot shows mean value with standard deviation of selected families whose relative abundance is significantly different between the groups (two-way ANOVA with Benjamini–Hochberg false discovery rate correction). Data with different superscript letters are significantly different at q < 0.05. Corresponding global p-Values and q-Values can be found in Table 1. CG: Co-grazers, AM-S: Atypical myopathy survivors, AM-NS: Atypical myopathy non-survivors.