Microbiome network traits in the rumen predict average daily gain in beef cattle under different backgrounding systems

Abstract

Background: Backgrounding (BKG), the stage between weaning and finishing, significantly impacts feedlot performance in beef cattle; however, the contributions of the rumen microbiome to this growth stage remain unexplored. A longitudinal study was designed to assess how BKG affects rumen bacterial communities and average daily gain (ADG) in beef cattle. At weaning, 38 calves were randomly assigned to three BKG systems for 55 days (d): a high roughage diet within a dry lot (DL, n = 13); annual cover crop within a strip plot (CC, n = 13); and perennial pasture vegetation within rotational paddocks (PP, n = 12), as before weaning. After BKG, all calves were placed in a feedlot for 142 d and finished with a high energy ration. Calves were weighed periodically from weaning to finishing to determine ADG. Rumen bacterial communities were profiled by collecting fluid samples via oral probe and sequencing the V4 region of the 16S rRNA bacterial gene, at weaning, during BKG and finishing.

Results: Rumen bacterial communities diverged drastically among calves once they were placed in each BKG system, including sharp decreases in alpha diversity for CC and DL calves only (P < 0.001). During BKG, DL calves showed a substantial increase of Proteobacteria (Succinivibrionaceae family) (P < 0.001), which also corresponded with greater ADG (P < 0.05). At the finishing stage, Proteobacteria bloomed for all calves, with no previous alpha or beta diversity differences being retained between groups. However, at finishing, PP calves showed a compensatory ADG, particularly greater than that in calves coming from DL BKG (P = 0.02). Microbiome network traits such as lower average shortest path length, and increased neighbor connectivity, degree, number and strength of bacterial interactions between rumen bacteria better predicted ADG during BKG and finishing than variation in specific taxonomic profiles.

Conclusions: Bacterial co-abundance interactions, as measured by network theory approaches, better predicted growth performance in beef cattle during BKG and finishing, than the abundance of specific taxa. These findings underscore the importance of early post weaning stages as potential targets for feeding interventions that can enhance metabolic interactions between rumen bacteria, to increase productive performance in beef cattle.

Keywords: Average daily gain; Backgrounding systems; Beef cattle; Rumen microbiome.

Fig. 1

Rumen fluid sampling protocol for 38 Angus and Angus x Simmental beef calves across five points during backgrounding (BKG) and finishing. After weaning, the calves were randomly allocated to three different BKG systems: (i) dry lot (DL, n = 13); (ii) cover crop (CC, n = 13) and (iii) a third group remained grazing on perennial pasture (PP, n = 12), as before weaning. Rumen fluid samples were collected at weaning (T1), twice during BKG (T2–T3), and twice at finishing (T4–T5), when calves were kept on a high energy feedlot diet (FLD). Details on the nutritional composition of BKG and FLD diets can be seen in Table 1

Fig. 2

Taxonomic composition at phylum level in the rumen microbiome of calves at weaning (T1), backgrounding (T2–T3), and finishing (T4–T5). a Barplot showing the relative abundance of phyla at weaning, backgrounding and finishing. b Boxplots showing relative abundances of Proteobacteria in calves within every BKG system, from weaning to finishing. c Boxplots showing relative abundances of Proteobacteria in calves within specific time points, BKG:T2–T3 and finishing: T4–T5. Dry lot (DL); cover crop (CC) and perennial pasture (PP)

Fig. 3

Rumen bacterial alpha diversity at weaning (T1), backgrounding (T2–T3), and finishing (T4–T5). a The number of different ASVs detected decreased significantly for calves moved to DL and CC BKG, but remained stable for calves on PP during this period. This number decreased again for all BKG groups at the finishing stage. b The patterns observed with the number of different ASVs as shown in (a) were replicated when measuring the Shannon index of diversity, which not only takes into account presence or absence of different ASVs, but also their abundance distribution. Dry lot (DL); cover crop (CC) and perennial pasture (PP). Asterisks show significant differences based on Kruskal–Wallis tests adjusted for multiple comparisons (**P < 0.001, ***P < 0.0001)

Fig. 4

Rumen bacterial composition and abundance of indicator taxa at weaning (T1), backgrounding (T2–T3), and finishing (T4–T5). a Principal coordinate analysis (PCoA, Bray–Curtis distances) showing bacterial compositional differences at weaning, BKG and finishing. ANOSIM and PERMANOVA test results, along with their statistics are displayed below each PCoA plot. b Cumulative abundance of indicator ASVs, distinguishing each BKG system, at weaning, BKG and finishing. The line plots show how the cumulative abundance of these indicator ASVs peaks during BKG (T2). The pies show the taxonomic affiliation, at the order level, of the indicator ASVs. Asterisks show significant differences in the abundance of indicator taxa between groups based on Kruskal–Wallis tests adjusted for multiple comparisons (**P < 0.001, ***P < 0.0001). Dry lot (DL); cover crop (CC) and perennial pasture (PP)

Fig. 5

Co-abundance networks of rumen bacterial taxa at weaning (T1), backgrounding (T2–T3), and finishing (T4–T5). a Network topology and attributes in the rumen microbiome of calves at weaning, BKG and finishing can be visualized. Nodes represent a given ASV and edges show the association (correlation) between two given bacterial taxa (nodes). Color key represents neighbor connectivity, which measures the average connectivity of all surrounding nodes in the network. Size of node represents the average degree of connectivity or number associations a given node has in the network. Shape represents the average shortest path length, a measure of how fast information can travel through a network. Differences in the variation of all these network attributes between each group at BKG and finishing can be observed in the box plots depicted in panel (b), where letters represent significant differences based on Wilcoxon Rank Sum tests (P < 0.05). Dry lot (DL); cover crop (CC) and perennial pasture (PP)