Assessing the microbiota of recycled bedding sand on a Wisconsin dairy farm

Abstract

Background: Sand is often considered the preferred bedding material for dairy cows as it is thought to have lower bacterial counts than organic bedding materials and cows bedded on sand experience fewer cases of lameness and disease. Sand can also be efficiently recycled and reused, making it cost-effective. However, some studies have suggested that the residual organic material present in recycled sand can serve as a reservoir for commensal and pathogenic bacteria, although no studies have yet characterized the total bacterial community composition. Here we sought to characterize the bacterial community composition of a Wisconsin dairy farm bedding sand recycling system and its dynamics across several stages of the recycling process during both summer and winter using 16S rRNA gene amplicon sequencing.

Results: Bacterial community compositions of the sand recycling system differed by both seasons and stage. Summer samples had higher richness and distinct community compositions, relative to winter samples. In both summer and winter samples, the diversity of recycled sand decreased with time drying in the recycling room. Compositionally, summer sand 14 d post-recycling was enriched in operational taxonomic units (OTUs) belonging to the genera Acinetobacter and Pseudomonas, relative to freshly washed sand and sand from cow pens. In contrast, no OTUs were found to be enriched in winter sand. The sand recycling system contained an overall core microbiota of 141 OTUs representing 68.45% ± 10.33% SD of the total bacterial relative abundance at each sampled stage. The 4 most abundant genera in this core microbiota included Acinetobacter, Psychrobacter, Corynebacterium, and Pseudomonas. Acinetobacter was present in greater abundance in summer samples, whereas Psychrobacter and Corynebacterium had higher relative abundances in winter samples. Pseudomonas had consistent relative abundances across both seasons.

Conclusions: These findings highlight the potential of recycled bedding sand as a bacterial reservoir that warrants further study.

Keywords: 16S rRNA sequencing; Bovine; Dairy farm; Microbiota; Recycled bedding sand.

Fig. 1

Alpha diversity of summer and winter sampling locations. Boxplots of Chao’s Richness and Shannon’s Diversity Index values of samples from each of the summer (A) and winter (B) sampling locations. Differing letters indicate a significant difference between locations within each season (P < 0.05) using the Wilcoxon Rank Sum test for summer (A) and Tukey’s HSD for winter (B) locations. Pairwise comparisons of Chao’s Richness were not performed for summer locations since no overall differences were found (P > 0.05). Surface and 7-in samples are not differentiated as no differences were found between them across locations (P > 0.05)

Fig. 2

Differences of bacterial community compositions across summer and winter sampling locations. Non-metric multidimensional scaling (nMDS) plot of the Bray-Curtis dissimilarity for microbial communities of recycled sand samples colleted in (A) summer and (B) winter. Each dot represents the bacterial community of a single sample and are colored by location. Ellipses represent standard error around centriod of each location and are colored by location. Surface and 7-in samples are not differentiated as no differences were found between them across locations (P > 0.05). Stress: A = 0.07, B = 0.052

Fig. 3

The core microbiota of a bedding sand recycling system. A heat-tree of the core microbiota of a bedding sand recycling system. Node and line widths increase with the mean relative abundance of the core taxonomic groups across all samples. Colors indicate the log2 ratio of the median counts of taxanomic groups between summer and winter samples. Increasing magenta indicates increased abundance in summer samples, cyan for increased abundance in winter samples, and grey represents no seasonal difference. Log2 ratio of median counts were set to zero for taxnomic groups having non-significant differential abundance by season (P < 0.05), per FDR-corrected pairwise Wilcoxon rank sum tests of median abundance counts