Saccharomyces cerevisiae-derived metabolites as putative regulators of feed intake in cattle

Dry matter intake (DMI) is a key factor influencing animal production efficiency. Although various microbial intervention strategies have been widely studied and applied in animal nutrition, our literature search indicated that systematic comparisons of their effects on DMI remain limited. This study employed network meta-analysis to comprehensively evaluate the effects of multiple microecological preparations on DMI in cattle, aiming to identify the highest-ranked intervention type and further explore its potential molecular mechanism using network pharmacology, Mendelian randomization (MR) and colocalization analyses based on human genetic data, bovine transcriptomic analysis, molecular docking, and exploratory two-step MR. The network meta-analysis revealed that Saccharomyces cerevisiae fermentation products were more effective in enhancing dry matter intake (DMI) than supplementation with live Saccharomyces cerevisiae alone, ranking first among all intervention strategies. Network pharmacology analysis identified 75 candidate target genes, and subsequent Mendelian randomization analysis based on human genetic data identified potential associations between 17 candidate genes and circulating ghrelin (GHRL) levels. Bovine colonic transcriptomic analysis using a dataset from calves supplemented with live Saccharomyces cerevisiae boulardii rather than SCFP showed an upward trend in HTR2B expression that was directionally consistent with the human MR result but did not reach statistical significance after FDR correction (logFC = 0.410, P = 0.032, FDR = 0.081). Molecular docking analysis yielded predicted docking energies of − 6.8, − 3.4, and − 6.0 kcal/mol for adenosine, choline, and thiamine with the HTR2B receptor, respectively. Colocalization analysis provided no evidence of a shared local genetic signal between HTR2B expression and circulating GHRL levels. Finally, exploratory two-step MR suggested a small opposing indirect effect through Mollicutes, corresponding to an estimated shielding proportion of 8.63%, although this signal showed limited robustness. This study, through a network meta-analysis, showed that Saccharomyces cerevisiae fermentation products ranked highest among 22 intervention strategies for increasing DMI in dairy cows and showed a larger estimated effect than live Saccharomyces cerevisiae supplementation. Further integrative analyses prioritized adenosine, choline, thiamine, and HTR2B as candidate components of a potential GHRL-related mechanism. However, because the MR evidence was derived from human genetic data, the proposed HTR2B –GHRL relationship should be considered a hypothesis-generating cross-species finding rather than a validated mechanism in cattle. Bovine experimental validation is required before these human MR estimates can be extrapolated to cattle.