Rice, supplemented with Dendrobium officinale, was subjected to cofermentation using Saccharomyces cerevisiae FBKL2.8022 (Sc) and Wickerhamomyces anomalus FBKL2.8023 (Wa). The alcohol content was determined with a biosensor, total sugars with the phenol–sulfuric acid method, reducing sugars with the DNS method, total acids and total phenols with the colorimetric method, and metabolites were analyzed using LC-MS/MS combined with multivariate statistics, while metabolic pathways were constructed using metaboAnalyst 5.0. It was found that the quality of rice wine was higher with the addition of D. officinale. A total of 127 major active substances, mainly phenols, flavonoids, terpenoids, alkaloids, and phenylpropanoids, were identified. Among them, 26 substances might have been mainly metabolized by the mixed-yeasts fermentation itself, and 10 substances might have originated either from D. officinale itself or from microbial metabolism on the newly supplemented substrate. In addition, significant differences in metabolite could be attributed to amino acid metabolic pathways, such as phenylalanine metabolism and alanine, aspartate, and glutamate metabolism. The characteristic microbial metabolism of D. officinale produces metabolites, which are α-dihydroartemisinin, alantolactone, neohesperidin dihydrochalcone, and occidentoside. This study showed that mixed-yeasts cofermentation and fermentation with D. officinale both could increase the content of active substances in rice wine and significantly improve the quality of rice wine. The results of this study provide a reference for the mixed fermentation of brewer’s yeast and non-yeast yeasts in rice wine brewing.

Dendrobium nobile Lindl. polysaccharide (DNP1) showed good anti-inflammatory activity in our previous study. In this study, the structural characterization of DNP1 and its mode of action on TLR4 were investigated. Structural characterization suggested that DNP1 was a linear glucomannan composed of (1 → 4)-β-Manp and (1 → 4)-β-Glcp residues, and the acetyl group was linked to the C-2 of Manp. The possible repeating structural units of DNP1 were [→4)-2-OAc-β-Manp-(1→]₃ →4)-β-Glcp-(1→. Surface plasmon resonance (SPR) binding test results showed that DNP1 did not bind directly to TLR4. The TLR4 and MD2 receptor blocking tests confirmed that DNP1 needs MD2 and TLR4 to participate in its anti-inflammatory effect. The binding energy of DNP1 to TLR4-MD2 was −7.9 kcal/mol, indicating that DNP1 could bind to the TLR4-MD2 complex stably. Therefore, it is concluded that DNP1 may play an immunomodulatory role by binding to the TLR4-MD2 complex and inhibiting the TLR4-MD2-mediated signaling pathway.