Characterization of the preweaned Holstein calf fecal microbiota prior to, during, and following resolution of uncomplicated gastrointestinal disease

Abstract

Little is known about shifts in the fecal microbiome of dairy calves preceding and following the incidence of gastrointestinal disease. The objective of this cohort study was to describe the fecal microbiome of preweaned dairy calves before, during, and after gastrointestinal disease. A total of 111 Holstein dairy calves were enrolled on 2 dairies (D1 and D2) and followed until 5 weeks old. Health assessments were performed weekly and fecal samples were collected every other week. Of the 111 calves, 12 calves from D1 and 12 calves from D2 were retrospectively defined as healthy, and 7 calves from D1 and 11 calves from D2 were defined as diarrheic. Samples from these calves were sequenced targeting the 16S rRNA gene and compared based on health status within age groups and farms: healthy (0-1 week old) vs. pre-diarrheic (0-1 week old), healthy (2-3 weeks old) vs. diarrheic (2-3 weeks old), and healthy (4-5 weeks old) vs. post-diarrheic (4-5 weeks old) calves. Healthy and diarrheic samples clustered together based on age rather than health status on both farms. Based on linear discriminant analysis, a few species were identified to be differently enriched when comparing health status within age groups and farm. Among them, Bifidobacterium sp. was differently enriched in pre-diarrheic calves at D1 (0-1 week old) whereas healthy calves of the same age group and farm showed a higher abundance of Escherichia coli. Bifidobacterium sp. was identified as a biomarker of fecal samples from healthy calves (2-3 weeks old) on D1 when compared with diarrheic calves of the same age group and farm. Feces from diarrheic calves on D2 (2-3 weeks old) were characterized by taxa from Peptostreptococcus and Anaerovibrio genera whereas fecal samples of age-matched healthy calves were characterized by Collinsella aerofaciens and Bifidobacterium longum. After resolution of uncomplicated diarrhea (4-5 weeks old), Collinsella aerofaciens was more abundant in D2 calves whereas Bacteriodes uniformis was more abundant in D1 calves. Taken together, these findings suggest that the age of the preweaned calf is the major driver of changes to fecal microbiome composition and diversity even in the face of uncomplicated gastrointestinal disease.

Keywords: calf diarrhea; case definitions; gastrointestinal microbiota; gut health; microbiome.

Figure 1

Observed species index (A), Shannon index (B), and Simpson index (C) are plotted for samples grouped by age and health status. The median is depicted by the line inside the box whereas the lowest and highest values observed are represented by the whiskers.

Figure 2

Principal coordinate analysis (PCoA) based on Bray-Curtis compositional dissimilarity of the bacterial 16S rRNA gene sequence data for fecal samples collected from healthy calves on farms D1 (n = 12) and D2 (n = 12) at 0–1, 2–3, and 4–5 weeks old. The proportion of variance explained by each principal coordinate axis is expressed by the corresponding axis label. Points represent individual samples. Samples that are more similar to one another appear closer together. The ellipses represent a 95% confidence interval calculated based on a t-distribution.

Figure 3

Principal coordinate analysis (PCoA) based on Bray-Curtis compositional dissimilarity of the bacterial 16S rRNA gene sequence data for fecal samples collected from healthy (n = 12) and diarrheic (n = 7) calves on farm D1 (panel A), and healthy (n = 12) and diarrheic (n = 11) calves on farm D2 (panel B) at 0–1, 2–3, and 4–5 weeks old. The proportion of variance explained by each principal coordinate axis is expressed by the corresponding axis label. Points represent individual samples. Samples that are more similar to one another appear closer together. The ellipses represent a 95% confidence interval calculated based on a t-distribution.

Figure 4

Linear discriminant analysis (LDA) effect size (LEfSe) bar plot comparing D1 pre-diarrheic calves (n = 7; red bars) with D1 healthy calves (n = 12; blue bars) at 0–1 week old. The LDA score (log₁₀) indicates the effect size of each differentially abundant bacterial taxon (p < 0.05).

Figure 5

Linear discriminant analysis (LDA) effect size (LEfSe) bar plot comparing D1 diarrheic calves (n = 7) with D1 healthy calves (n = 12; blue bars) at 2–3 weeks old. The LDA score (log₁₀) indicates the effect size of each differentially abundant bacterial taxon (p < 0.05).

Figure 6

Linear discriminant analysis (LDA) effect size (LEfSe) bar plot comparing D1 post-diarrheic calves (n = 7; red bars) with D1 healthy calves (n = 12; blue bars) at 2–3 weeks old. The LDA score (log₁₀) indicates the effect size of each differentially abundant bacterial taxon (p < 0.05).

Figure 7

Linear discriminant analysis (LDA) effect size (LEfSe) bar plot comparing D2 diarrheic calves (n = 11; red bars) with D2 healthy calves (n = 12; blue bars) at 2–3 weeks old. The LDA score (log₁₀) indicates the effect size of each differentially abundant bacterial taxon (p < 0.05).

Figure 8

Linear discriminant analysis (LDA) effect size (LEfSe) bar plot comparing D2 post-diarrheic calves (n = 11; red bars) with D2 healthy calves (n = 12; blue bars) at 4–5 weeks old. The LDA score (log₁₀) indicates the effect size of each differentially abundant bacterial taxon (p < 0.05).