Differences in Faecal Microbiome Taxonomy, Diversity and Functional Potential in a Bovine Cohort Experimentally Challenged with Mycobacterium avium subsp. paratuberculosis (MAP)

Abstract

Mycobacterium avium subspecies paratuberculosis (MAP) is the causative agent of Johne's disease in ruminants, a chronic enteritis which results in emaciation and eventual loss of the animal. Recent advances in metagenomics have allowed a more in-depth study of complex microbiomes, including that of gastrointestinal tracts, and have the potential to provide insights into consequences of the exposure of an animal to MAP or other pathogens. This study aimed to investigate taxonomic diversity and compositional changes of the faecal microbiome of cattle experimentally challenged with MAP compared to an unexposed control group. Faecal swab samples were collected from a total of 55 animals [exposed group (n = 35) and a control group (n = 20)], across three time points (months 3, 6 and 9 post-inoculation). The composition and functional potential of the faecal microbiota differed across time and between the groups (p < 0.05), with the primary differences, from both a taxonomic and functional perspective, occurring at 3 months post inoculation. These included significant differences in the relative abundance of the genera Methanobrevibacter and Bifidobacterium and also of 11 other species (4 at a higher relative abundance in the exposed group and 7 at a higher relative abundance in the control group). Correlations were made between microbiome data and immunopathology measurements and it was noted that changes in the microbial composition correlated with miRNA-155, miR-146b and IFN-ɣ. In summary, this study illustrates the impact of exposure to MAP on the ruminant faecal microbiome with a number of species that may have relevance in veterinary medicine for tracking exposure to MAP.

Keywords: Johne’s disease; MAP; experimental model; immune function; microbiome.

Figure 1

Phylum level analysis revealed an increase in the relative abundance of members of Euryarchaeaota and Firmicutes in the exposed group with a decreased relative abundance of Actinobacteria and Bacteriodetes in the exposed group, with the greatest differences observed in month 3.

Figure 2

The microbiome was dominated by a combination of four species; Methanobrevibacter unclassified, Bifidobacterium pseudolongum, Butyrivibrio unclassified and Peptostreptococcaceae unclassified.

Figure 3

Discriminatory species were identified using linear discriminant analysis (LDA) effect size (LEfSE), which allows for the identification of species which explain differences between groups. Abundances of Bifidobacterium species decreased over time, particularly B. pseudolongum, B. angulatum and B. adolescentis.

Figure 4

Diversity of the microbiome of both the exposed and control animals across time. Alpha diversity within subjects using Shannon, Simpson and Observed species measures of species-level output from MetaPhlAn2. The control group shows a higher within animal diversity; however, this was non-significant.

Figure 5

Bray-Curtis beta diversity among animals in both the exposed and control group, across time. Month 3 shows the most variation between the groups. The microbiome becomes more stable as animals age or potentially as a result of MAP leaving the system as can be observed at the month 6 and 9 time points. MDS1 and MDS2 represent the two dimensional space in which the points are arranged.

Figure 6

Functional potential of the microbiome at level 1 subsystem from the SUPER-FOCUS output. SUPER-FOCUS uses the SEED database, which is a subsystem database, assigning function at four different levels. Subsystems are divided into hierarchies. Number s on the x axis refers to the linear discriminant analysis (LDA) score. The linear discriminant analysis (LDA) effect size was determined using LEfSe. LEfSe uses the two-tailed non-parametric Kruskal–Wallis test to examine the significance of differences of species and functional potential in the two groups. A set of pairwise tests among the two groups was performed using the Wilcoxon test. LDA was performed to estimate the effect size of each differently abundant species. The functional potential of each of the groups was thought to be significantly different if their differences had a p-value < 0.05 and an LDA score (log10) > 2.