Temporal Changes in Fecal Unabsorbed Carbohydrates Relative to Perturbations in Gut Microbiome of Neonatal Calves: Emerging of Diarrhea Induced by Extended-Spectrum β-lactamase-Producing Enteroaggregative Escherichia coli

Abstract

Early gut microbiota development and colonization are crucial for the long-term health and performance of ruminants. However, cognition among these microbiota is still vague, particularly among the neonatal dairy calves. Here, extended-spectrum β-lactamase-producing enteroaggregative E. coli (ESBL-EAEC)-induced temporal changes in diversity, stability, and composition of gut microbiota were investigated among the neonatal female calves, with the view of discerning potential biomarkers of this arising diarrhea cases in local pastures. Nearly, 116 newborn calves were enrolled in this time period study during their first 2 weeks of life, and a total of 40 selected fecal samples from corresponding calves were used in this study. The results revealed that differentiated gut microbiome and metabolome discerned from neonatal calves were accompanied by bacterial infections over time. Commensal organisms like Butyricicoccus, Faecalibacterium, Ruminococcus, Collinsella, and Coriobacterium, as key microbial markers, mainly distinguish "healthy" and "diarrheic" gut microbiome. Random forest machine learning algorithm indicated that enriched fecal carbohydrates, including rhamnose and N-acetyl-D-glucosamine, and abundant short-chain fatty acids (SCFAs) existed in healthy ones. In addition, Spearman correlation results suggested that the presence of Butyricicoccus, Faecalibacterium, Collinsella, and Coriobacterium, key commensal bacteria of healthy calves, is positively related to high production of unabsorbed carbohydrates, SCFAs, and other prebiotics, and negatively correlated to increased concentrations of lactic acid, hippuric acid, and α-linolenic acid. Our data suggested that ESBL-EAEC-induced diarrhea in female calves could be forecasted by alterations in the gut microbiome and markedly changed unabsorbed carbohydrates in feces during early lives, which might be conducive to conduct early interventions to ameliorate clinical symptoms of diarrhea induced by the rising prevalence of ESBL-EAEC.

Keywords: enteroaggregative E. coli; extended-spectrum β-lactamase producing E. coli; gut microbiome; metabolome; neonatal dairy calves; unabsorbed carbohydrates.

Figure 1

Gut microbiota assembly of neonatal calves post-ESBL-EAEC infection. (A) The relative abundance of top 20 bacterial genera in calf feces. (B) Top 50 bacteria of fecal samples presented using cluster heatmap analysis in diarrheal (D) or healthy (H) calves. In the graph of species clustering, the default species are UPGMA clustered according to the Pearson correlation coefficient matrix of their constituent data and arranged according to the clustering results. The red color block indicates that the abundance of the genus in this group is higher than in the other groups, while the blue color block indicates that the abundance of the genus in this group is lower than in the other groups. The corresponding relationship between the color gradient and the value is shown in the gradient color block.

Figure 2

Gut microbiota diversity of neonatal calves post-ESBL-EAEC infection. (A) The α-diversity of different groups by Chao1 and Simpson indices. Data were presented as mean ± SEM values. P-values were determined using the nonparametric Kruskal–Wallis test. (B) Principal coordinate analysis (PCoA) of fecal bacteria was performed based on the weighted UniFrac distance matrix. The statistical tests were accomplished using PERMANOVA, with 999 permutations. The enriched gut microbiota taxa were shown by LEfSe [linear discriminant analysis (LDA) coupled with effect size measurements] of H_1 vs D_1 (C), H_2 vs D_2 (D), H_1 vs H_2 (E), and D_1 vs D_2 (F).

Figure 3

Differentiated genera were displayed using a random forest supervised machine learning algorithm between H_1 vs D_1 (A) and H_2 vs D_2 (B). The respective name of the bacterial genus is shown on the left. The relative abundance of genera was clustered using a UPGMA dendrogram and showed in a heatmap. The color indicates the relative abundance of the genus in the group of samples, and the corresponding relationship between the color gradient and the value is shown in the gradient color block. The genus variation is shown using Z-Score. The top 29 genera in the fecal samples were included and the rank values are shown using the Importance Index.

Figure 4

Alterations in the fecal metabolome profiles of neonatal calves post-ESBL-EAEC infection. (A) The classifications of total metabolome compounds in H_1, H_2, D_1, and D_2 groups. The total number of significantly changed metabolites in this class is indicated and the corresponding proportions are shown in parentheses. (B) Partial least squares discriminant analysis (PLS-DA) was used here to cluster the fecal metabolome profiles of calves. Metabolome profile for the H_1, H_2, D_1, or D_2 groups is shown in the same color, respectively. Data were presented as mean ± SEM values. P-values were acquired using the nonparametric Kruskal–Wallis test. (C) KEGG pathway enrichment analysis was associated with dramatically changed metabolites. The respective name of the KEGG pathway is shown on the left, and the corresponding P-value is shown on the right with a gradient color. P-values were acquired following two-side Fisher's exact tests with Benjamini-Hochberg correction for multiple testing. (D) Differentiated metabolites were displayed using a random forest supervised machine learning algorithms among H_1, H_2, D_1, and D_2 groups. The respective name of the metabolite is shown on the left. The top 10 metabolites in fecal samples are shown in different colors, and the rank values are shown as MeanDecreaseGini.

Figure 5

The alterations in fecal metabolome profiles of neonatal calves post-ESBL-EAEC infection. The concentrations of fecal rhamnose (A), N-acetyl-D-glucosamine (B), xylose (C), acetic acid (D), butyric acid (E), isovaleric acid (F), lactic acid (G), hippuric acid (H), and α-linolenic acid (I) are displayed as box and dot plots. Data were presented as mean ± SEM values. P-values were determined using the nonparametric Kruskal–Wallis test. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 6

The Spearman correlation between differentiated gut microbial taxa and fecal metabolites in H_1 vs H_2 vs D_1 vs D_2 groups. Red squares indicate a positive correlation, and blue squares indicate a negative correlation. The intensity of the color was proportional to the strength of the Spearman correlation. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.