Effects of Condensed Tannins on Bacterial and Fungal Core Microbiomes Involved in the Ensiling and Aerobic Spoilage of Purple Prairie Clover (Dalea purpurea Vent.) Silage

Effects of condensed tannins (CT) on rumen microbes have been well documented, whereas little information exists about their effects on the microbial communities involved in ensiling. The objective of this study was to determine the effects of CT on the composition and diversity of bacterial and fungal core microbiomes associated with ensiling and aerobic spoilage of purple prairie clover (PPC). Purple prairie clover (Dalea purpurea Vent.; 60 g CT/kg DM) was harvested at full flower and ensiled in PVC laboratory silos with and without polyethylene glycol (PEG) for 76 d. Silage was then subjected to aerobic exposure for 14 d. Bacterial and fungal core microbiomes in the silage and in the aerobically exposed silage were examined using real-time qPCR and high-throughput sequencing. Real-time qPCR analysis revealed that PPC ensiled without PEG exhibited less (P<0.01~0.001) gene copy numbers of total bacteria, Lactobacillus, yeasts and fungi than PEG treated silage. This trend was also observed for d-7 aerobically exposed silage with the exception of Lactobacillus which had greater (P<0.05) gene copy numbers for non-PEG than for PEG treated silage. Metagenome analyses generated a total of 4,273,668 bacterial sequences and 3,455,929 fungal sequences which were assigned to 225 bacterial and 142 fungal genera, respectively. Addition of PEG increased (P<0.001) the abundance of Lactobacillus and Pediococcus, but decreased (P<0.01) that of Lactococcus, Leuconostoc, Agrobacterium, Erwinia, Methylobacterium, Pseudomonas and Sphingomonas. The abundance of the fungal genera Colletotrichum, Xylogone, Galactomyces, Penicillium, Fusarium and Cryptococcus in silage were also increased (P<0.05) by PEG. Diversity measurements of bacterial and fungal communities indicated that addition of PEG decreased (P<0.01) the number of microbial core genome operational taxonomic units (OTUs), ACE, Chao 1 and Shannon indexes. It did not affect the diversity of fungal communities. The PEG treated silage had a higher (P<0.001) abundance of Pediococcus, but lower (P<0.001) Lactococcus and Leuconostoc than non-PEG silage after aerobic exposure. The PEG treated silage also had a greater (P<0.05~0.01) abundance of Colletotrichum, Penicillium and Fusarium at d 7, and greater (P<0.05~0.001) abundance of Candida, Colletotrichum, Wickerhamomyces, Penicillium and Pterula after aerobic exposure. The observed bacterial and fungal OTUs, ACE and Chao 1 were lower (P<0.05) for PEG treated than for non-PEG treated silage after aerobic exposure. The results indicated that CT decreased population of majority bacteria, fungus and yeast but increased bacterial diversity during ensiling and aerobic deterioration, but increased fungal diversity only after aerobic exposure.