Different lactic acid bacteria and their combinations regulated the fermentation process of ensiled alfalfa: ensiling characteristics, dynamics of bacterial community and their functional shifts

Abstract

The objectives of this study were to investigate the adaptation and competition of Lactobacillus plantarum, Pediococcus pentosaceus and Enterococcus faecalis inoculated in alfalfa silage alone or in combination on the fermentation quality, dynamics of bacterial community, and their functional shifts using single-molecule real-time (SMRT) sequencing technology. Before ensiling, alfalfa was inoculated with L. plantarum (Lp), P. pentosaceus (Pp), E. faecalis (Ef) or their combinations (LpPp, LpEf, LpPpEf) and sampled at 1, 3, 7, 14 and 60 days. After 60-days fermentation, the Lp-, Pp- and LpPp-inoculated silages had lower pH but greater concentrations of lactic acid were observed in Pp, LpEf and LpPpEf-inoculated silages. The inoculants altered the keystone taxa and the bacterial community dynamics in different manners, where L. plantarum, Weissella cibaria and L. pentosaceus dominated the bacterial communities after 14 days-fermentation in all treatments. The silages with better fermentation quality had simplified bacterial correlation structures. Moreover, different inoculants dramatically changed the carbohydrate, amino acid, energy, nucleotide and vitamin metabolism of bacterial communities during ensiling. Results of the current study indicate that effect of different inoculants on alfalfa silage fermentation was implemented by modulating the succession of bacterial community, their interactions and metabolic pathways as well during ensiling.

Fig. 1

Microbial community dissimilarities and diversities in alfalfa silage during ensiling. C, Control (samples without inoculants); Lp, samples inoculated with L. plantarum; Pp, samples inoculated with P. pentosaceus; Ef, samples inoculated with E. faecalis; LpPp, samples inoculated with L. plantarum and P. pentosaceus; LpEf, samples inoculated with L. plantarum and E. faecalis; LpPpEf, samples inoculated with L. plantarum, P. pentosaceus and E. faecalis. Arabic number indicating days of ensiling. A. The community dissimilarities in different inoculant treatments and fermentation time, calculated via weighted UniFrac distances, with coordinates calculated using principal coordinates analysis (PCoA). B. The variations in community alpha‐diversities (Shannon index). C. Relative abundance of alfalfa silage bacterial genera across different inoculant treatments and fermentation time. D. Relative abundance of alfalfa silage bacterial species across different inoculant treatments and fermentation time.

Fig. 2

Interaction networks of the alfalfa silage microbiota. 16S rRNA gene‐based correlation network of the alfalfa silage microbiota is calculated from the bacteria with a relative abundance greater than 0.2%. Node size is scaled based on the overall abundance of each taxon in the microbiota. Edge width is proportional to the strength of association between each metabolite‐phylotype pair (as measured by the correlation), red edge indicates positive correlations and green edge indicates negative corrections. A. Control (samples without inoculants), (B) samples inoculated with L. plantarum, (C) samples inoculated with P. pentosaceus, (D) samples inoculated with E. faecalis, (E) samples inoculated with L. plantarum and P. pentosaceus, (F) samples inoculated with L. plantarum and E. faecalis, (G) samples inoculated with L. plantarum, P. pentosaceus, and E. faecalis.

Fig. 3

Correlation analysis of the bacterial communities with fermentation characteristics. C, Control (samples without inoculants); Lp, samples inoculated with L. plantarum; Pp, samples inoculated with P. pentosaceus; Ef, Samples inoculated with E. faecalis; LpPp, samples inoculated with L. plantarum and P. pentosaceus; LpEf, samples inoculated with L. plantarum and E. faecalis; LpPpEf, samples inoculated with L. plantarum, P. pentosaceus and E. faecalis. A. Redundancy analysis (RDA) plot showing the correlations between fermentation characteristics and bacterial community composition. The canonical axes are labelled with the percentage of total variance explained (%). Arrow lengths indicate the variance explained by fermentation characteristics. Different inoculant treatments at different fermentation times are presented as individual data points. Arabic numbers indicate days of ensiling. B. Association analysis between bacterial species and fermentation characteristics. Fermentation characteristics are displayed horizontally and the bacterial community information is displayed vertically. The corresponding value of the middle heat map is the Spearman correlation coefficient r, which ranges between − 1 and 1; r < 0 indicates a negative correlation (red), r > 0 indicates a positive correction (blue), and ‘*’ and ‘**’ represent P < 0.05 and P < 0.01, respectively.

Fig. 4

Microbial alterations contribute to functional shifts after fermentation with different inoculants and their combinations. C, Control (samples without inoculants); Lp, samples inoculated with L. plantarum; Pp, samples inoculated with P. pentosaceus; Ef, samples inoculated with E. faecalis; LpPp, samples inoculated with L. plantarum and P. pentosaceus; LpEf, samples inoculated with L. plantarum and E. faecalis; LpPpEf, samples inoculated with L. plantarum, P. pentosaceus, and E. faecalis. Summary of significant functional shifts predicted using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt). For each KEGG pathway, the second level of the predicted functional shift is shown with respect to the fermentation processes and inoculant treatments. ^a‐eIndicates significant differences between inoculant treatments with the same silage period at P < 0.05.