Table 1.
Demographics of the dogs enrolled in the study.
Fig 1.
Diversity measures between the fearful dogs (case) and the dogs with normal behavior (control).
Alpha diversity was assessed by using Observed species ((A), p = 0.14), Chao1 ((B), p = 0.14) and Shannon ((C), p = 0.97), indices and the ANOVA test. Β diversity was also evaluated by using the unweighted ((D, p<0.52) and weighted ((E, p<0.2) UniFrac distance measures, applying the PERMANOVA test. No significant differences were identified by all the above-mentioned tests.
Fig 2.
Taxonomic assignment highlighted different microbial taxa in case and control groups.
A different bacterial composition (relative abundance, %) was highlighted at the phylum level (A) and confirmed by core microbiome analysis showing a different taxa profile in Case (B) and Control (C) groups. These differences in taxa composition were also identified at the genus level, as shown in panel (D).
Fig 3.
Significantly different taxa identified between case and control groups by EdgeR analysis (adjusted p-value <0.05).
Among the significantly different taxa we found that the Gammaproteobacteria class was less abundant in the cases respect to the controls (A). The Firmicutes-related taxa, Erysipelatoclostridiaceae family (B) and Peptostreptococcales Tissierellales order (C) were more abundant in the cases, when compared the controls. At genus level, the Firmicutes-belonging Dorea taxon was reduced in the case group (D).
Fig 4.
Comparative metabolome analysis of case (fearful dogs) versus controls.
(A) PLS-DA analysis was performed using normalized metabolite levels from cases (green) and controls (red); (B) The top-15 discriminant features identified with values of VIP scores >1.5 are reported. (C) Heatmap showing metabolite concentrations in each group and replicates. The intensity of the colored boxes represents the relative abundance of each molecule, whereas metabolite concentrations were normalized, log(10)-transformed, and auto-scaled. (D) Volcano plot analysis of metabolites significantly different in the comparison case vs control. The red and green dots represent the significant increased and decreased metabolites, respectively. Non-colored dots refer to all the molecules identified in the dataset whose relative abundance is not significantly different between groups.
Table 2.
Significant metabolites found differentially abundant in fearful dogs versus controls.
Fig 5.
Statistical analysis and bioinformatic pathway enrichment of single metabolites from the differential metabolome.
(A) The abundances of single metabolites were evaluated in case vs control by parametric or non-parametric t-test. The plots report the analyte concentrations (means ± SEM on the bar, dots are all the measurements in replicates). The significance of statistical analysis is referred to as: * p<0.05, ** p<0.01, **** p<0.0001). (B) Pathway analysis plot with details of the significant pathways enriched by the analysis. (C) Overview of MSEA terms enriched with the over representation analysis (ORA) of significant metabolites. Nonetheless, to assess whether faecal species and circulating metabolites varied with the same trends in fearful animals, we performed a correlation analysis using both types of features. Significant microbiota species were correlated with the global metabolome, highlighting significant correlations (Fig 6). Interestingly, bile acids (namely TDCA, GUDCA, GLCA) were positively correlated with the Peptostreptococcales Tissierellales (Fig 6A), whereas some amino acids (Val, Thr, Lys) were associated to the Dorea (Fig 6B). Remarkably, many of these molecules were found as significant from both volcano plot and VIP analyses.
Fig 6.
Correlation analysis of the differential microbiome with the global metabolome.
Significantly correlating metabolites were reported for (A) Peptostreptococcales T and (B) Dorea organisms. The correlation plots report the computed r coefficient and the statistical significance (p). TDCA: Taurodeoxycholic acid, GUDCA: Glycoursodeoxycholic acid, GLCA: Glycolithocholic acid, Val: Valine, Thr: Threonine, Lys: Lysine.