Faecal microbiota characterisation of horses using 16 rdna barcoded pyrosequencing, and carriage rate of clostridium difficile at hospital admission

Abstract

Background: The equine faecal microbiota is very complex and remains largely unknown, while interspecies interactions have an important contribution to animal health. Clostridium difficile has been identified as an important cause of diarrhoea in horses. This study provides further information on the nature of the bacterial communities present in horses developing an episode of diarrhoea. The prevalence of C. difficile in hospitalised horses at the time of admission is also reported.

Results: Bacterial diversity of the gut microbiota in diarrhoea is lower than that in non-diarrhoeic horses in terms of species richness (p-value <0.002) and in population evenness (p-value: 0.02). Statistical differences for Actinobacillus, Porphyromonas, RC9 group, Roseburia and Ruminococcaceae were revealed. Fusobacteria was found in horses with diarrhoea but not in any of the horses with non-diarrheic faeces. In contrast, Akkermansia was among the three predominant taxa in all of the horses studied. The overall prevalence of C. difficile in the total samples of hospitalised horses at admission was 3.7 % (5/134), with five different PCR-ribotypes identified, including PCR-ribotype 014. Two colonised horses displayed a decreased bacterial species richness compared to the remaining subjects studied, which shared the same Bacteroides genus. However, none of the positive animals had diarrhoea at the moment of sampling.

Conclusions: The abundance of some taxa in the faecal microbiota of diarrhoeic horses can be a result of microbiome dysbiosis, and therefore a cause of intestinal disease, or some of these taxa may act as equine enteric pathogens. Clostridium difficile colonisation seems to be transient in all of the horses studied, without overgrowth to trigger infection. A large proportion of the sequences were unclassified, showing the complexity of horses' faecal microbiota.

Fig. 1

Microbiota faecal composition at phylum level for horses with and without diarrhoea. Bar chart detailing the relative abundance of the 17 core phylotypes common to the two groups of horses (with and without diarrhea)

Fig. 2

Microbiota faecal composition at genus level (cumulated mean relative abundance >4 %) for horses with and without diarrhoea. Samples 13, 14, 15, 17, 23, 25, 26, 27, 28 and 29: horses with diarrhoea. Samples 11, 12, 16, 18, 19, 20, 21, 22, 24 and 30: horses without diarrhoea. The unclassified populations correspond to defined groups of the genus level for which a taxonomical classification assignation to the genus cannot be attributed. These populations are therefore labelled with the first defined superior hierarchical taxonomic level followed by “_unclassified” to prevent confusion

Fig. 3

Bacterial biodiversity, bacterial richness and bacterial evenness in C. difficile negative horses with and without diarrhoea. Box plot of richness, evenness and diversity values showed that the microbiota structure of each group (diarrhoeic and non diarrhoeic horses) is statistically different from each other. Whiskers represent minimum and maximum value. Bottom and top of the box are the first and the third quartile. The median is shown as a band inside the box

Fig. 4

Bacterial genus whose relative abundance was statistically different between the 2 groups. Result of a White test (p value <0.05). Box plot showing mean relative sequence abundance of Actinobacillus, Porphyromonas, RC9, Roseburia and Ruminococcaceae_unclassified in horses with and without diarrhea. The error bar indicates the diversity between samples in terms of proportions of sequences

Fig. 5

Microbiota faecal composition at genus level (cumulated mean relative abundance >4 %) of C. difficile culture-positive and -negative horses. Samples HF_01, 03, 04, 09 and 10: C. difficile positive horses detected by faeces culture. Samples HF_02, 05, 06, 07 and 08: C. difficile negative horses detected by faeces culture