Administration of probiotic lactic acid bacteria to modulate fecal microbiome in feedlot cattle

Abstract

Modulation of animal gut microbiota is a prominent function of probiotics to improve the health and performance of livestock. In this study, a large-scale survey to evaluate the effect of lactic acid bacteria probiotics on shaping the fecal bacterial community structure of feedlot cattle during three experimental periods of the fattening cycle (163 days) was performed. A commercial feedlot located in northwestern Argentina was enrolled with cattle fed mixed rations (forage and increasing grain diet) and a convenience-experimental design was conducted. A pen (n = 21 animals) was assigned to each experimental group that received probiotics during three different periods. Groups of n = 7 animals were sampled at 40, 104 and 163 days and these samples were then pooled to one, thus giving a total of 34 samples that were subjected to high-throughput sequencing. The microbial diversity of fecal samples was significantly affected (p < 0.05) by the administration period compared with probiotic group supplementation. Even though, the three experimental periods of probiotic administration induced changes in the relative abundance of the most representative bacterial communities, the fecal microbiome of samples was dominated by the Firmicutes (72-98%) and Actinobacteria (0.8-27%) phyla, while a lower abundance of Bacteroidetes (0.08-4.2%) was present. Probiotics were able to modulate the fecal microbiota with a convergence of Clostridiaceae, Lachnospiraceae, Ruminococcaceae and Bifidobacteriaceae associated with health and growth benefits as core microbiome members. Metabolic functional prediction comparing three experimental administration periods (40, 104 and 163 days) showed an enrichment of metabolic pathways related to complex plant-derived polysaccharide digestion as well as amino acids and derivatives during the first 40 days of probiotic supplementation. Genomic-based knowledge on the benefits of autochthonous probiotics on cattle gastrointestinal tract (GIT) microbiota composition and functions will contribute to their selection as antibiotic alternatives for commercial feedlot.

Figure 1

Experimental design: probiotic experimental groups: A: Lactobacillus acidophilus CRL 2074; B: Limosilactobacillus fermentum CRL 2085; C: Limosilactobacillus mucosae CRL 2069; D: CRL 2085 + CRL 2069; E: CRL 2074 + CRL 2085 + CRL 2069; Control group (n = 21) included in a separate pen. Sampling time in days: 0 (T0), 40 (T1), 104 (T2) and 163 (T3). Experimetal periods of probiotics administration E-40, E-104 and E163. Sampling scheme: animals (n = 7) received probiotics up to 40 days and then removed (to separate pen) experimental period E-40; animals (n = 7) received probiotics up to 104 days and then removed, E-104 and, animals (n = 7) administered with probiotics for 163 days E-163. This scheme was replicated for the A, B, C, D and E experimental groups.

Figure 2

Phylum- (a) and order-level (b) distribution of the microbiome in probiotics (A: Lactobacillus acidophilus CRL 2074; B: Limosilactobacillus fermentum CRL 2085; C: Limosilactobacillus mucosae CRL 2069; D: CRL 2085 + CRL 2069; E: CRL 2074 + CRL 2085 + CRL 2069-treated and untreated (control) fecal samples of feedlot cattle at different sampling times (T0: 0 day; T1: 40 days; T2: 104 days; T3: 163 days). Each color represents a phylum and an order.

Figure 3

Core microbiome at the family level of feedlot cattle fecal samples displayed by a heatmap generated by pipeline Microbiome Analyst (http://microbiomeanalyst.ca/faces/home.xhtml). Color shading indicates the prevalence of each bacterial family among samples for each abundance threshold. As we increased the detection threshold, the prevalence decreased.

Figure 4

Principal coordinate analysis (PCoA) based on Bray Curtis ß-diversity. The plot illustrates the distances between bacterial communities in all individual fecal samples (A: Lactobacillus acidophilus CRL 2074; B: Limosilactobacillus fermentum CRL 2085; C: Limosilactobacillus mucosae CRL 2069; D: CRL 2085 + CRL 2069; E: CRL 2074 + CRL 2085 + CRL 2069 treated and untreated (control) at different sampling times (T0, T1, T2 and T3).

Figure 5

Genus-level distribution (a) of the microbiome in probiotic-treated (A: Lactobacillus acidophilus CRL 2074; B: Limosilactobacillus fermentum CRL 2085; C: Limosilactobacillus mucosae CRL 2069; D: CRL 2085 + CRL 2069; E: CRL 2074 + CRL 2085 + CRL 2069) and untreated (control) fecal samples of feedlot cattle analyzed at 163 days (T3) after different administration periods. Each color represents a genus. (b) Dendrogram representing hierarchical clustering distances based on Bray–Curtis dissimilarity indices calculated at the genus level. For each probiotic group and control sample, T3 was considered.

Figure 6

Heatmap showing the species profile in probiotic-treated (A: Lactobacillus acidophilus CRL 2074; B: Limosilactobacillus fermentum CRL 2085; C: Limosilactobacillus mucosae CRL 2069; D: CRL 2085 + CRL 2069; E: CRL 2074 + CRL 2085 + CRL 2069) and untreated (control) feedlot fecal samples at different experimental administration periods (E-40, E-104 and E-163). For each probiotic group, samples at T3 were considered, while for controls T0 and T3, samples were analyzed. The figure was generated using the heatmap web server http://www.heatmapper.ca/expression/.

Figure 7

Plot from LEfSe analysis indicating enriched bacterial families associated either with positive (pink) or negative (light blue) correlations in cattle fecal samples as affected by (a) experimental administration periods and (b) probiotic groups.

Figure 8

Differential PICRUSt2 predicted metacyclic pathways. Metabolic pathway comparison using STAMP between probiotic administration periods: (a) E-40 vs. E-104; (b) E-40 vs. E-163 and (c) E-104 vs. E-163. E- 40; E-104; E-163. The q-values are based on Welsh's t test and corrected with Benjamini–Hochberg FDR.

Figure 8

Differential PICRUSt2 predicted metacyclic pathways. Metabolic pathway comparison using STAMP between probiotic administration periods: (a) E-40 vs. E-104; (b) E-40 vs. E-163 and (c) E-104 vs. E-163. E- 40; E-104; E-163. The q-values are based on Welsh's t test and corrected with Benjamini–Hochberg FDR.