In the context of enhancing the quality and fermentation efficiency of Baijiu, identifying sorghum alternatives represents an important research direction. This study demonstrated that rice can serve as a substitute for sorghum in sesame-flavor Baijiu brewing, yielding a typical product without altering the fermentation trend. The rice group yielded higher levels of lactic acid ethyl ester (3.20 g/L vs. 2.32 g/L) but lower acetic acid ethyl ester (2.82 g/L vs. 3.44 g/L). Moreover, it maintained higher starch and reducing sugar levels and achieved a higher final yield (222.25 kg vs. 166.88 kg). The rice group exhibited a greater fungal abundance, particularly Pichia kudriavzevii and Saccharomyces cerevisiae, contributing to ethanol production. During pit fermentation, the abundance of Lactobacillus acetotolerans was higher, while Acetilactobacillus jinshanensis and Limosilactobacillus pontis were lower in the rice group. L. acetotolerans possessed a complete starch degradation pathway, while A. jinshanensis and L. pontis lacked starch utilization but possessed glucose uptake protein. A. jinshanensis depended on phosphoketolase for growth, upregulating metabolic flux to acetic acid rather than lactic acid, its abundance likely drives differential production of corresponding ethyl esters. Actual fermentation complemented by kinetic simulations suggested potential changes in starch and D-glucose levels resulting from raw material substitution may not alter bacterial structure. The findings advance the understanding of using rice for Baijiu fermentation and inform the design of strategies using diversified raw materials.

Microbial communities play pivotal roles in the fermentation of cigar tobacco leaves. Although high-throughput sequencing technology has facilitated the exploration of these communities, a comprehensive understanding of their assembly mechanisms remains elusive. This review integrates the current knowledge regarding microbial sources, ecological dynamics, and evolutionary processes during cigar tobacco fermentation. We systematically assess the abiotic factors (temperature, humidity, nutrients) and biotic interactions (quorum sensing, metabolic coordination) which influence the microbial community. Moreover, we put forward strategies for synthetic community engineering and discuss the emerging applications of artificial intelligence in fermentation optimization. These insights deepen the understanding of microbial communities in cigar tobacco leaves, and provide new perspectives on regulating microbial communities to enhance the fermentation quality of cigar tobacco leaves.

Abstract Efficient xylose bioconversion is limited by the catalytic performance of xylulokinase (XylB), which phosphorylates d-xylulose to xylulose-5-phosphate. This study aimed to improve XylB from Lactococcus lactis N8 using a multi-dimensional semi-rational design strategy integrating sequence conservation, structure-based screening, mutagenesis, molecular docking, and molecular dynamics simulations. Asn301 was identified as a key functional hotspot, and Met259 was found to modulate the active-site hydrophobic environment. Combinatorial mutagenesis generated two superior variants, XylBM259H/N301G and XylBM259H/N301H, with more than 200% higher specific activity than the wild-type XylB (XylBWT) while retaining stable expression and environmental adaptability. Mechanistic analyses showed strengthened ATP binding, improved d-xylulose positioning, enhanced ligand affinity, and balanced local flexibility with structural compactness. These findings provide efficient XylB variants and a transferable strategy for engineering lignocellulose-related enzymes.

The flavor formation and quality stability of solid-state fermented foods are closely linked to their storage process, and Chinese baijiu Daqu serves as an ideal model for studying such microecological succession. Analysis of Daqu produced in spring, autumn, and winter showed that despite seasonal differences in initial communities, all exhibited highly convergent succession during storage. Bacillaceae increased steadily, while Lactobacillaceae and Saccharomycetaceae declined. This restructuring corresponded with directional metabolic shifts: metabolites that decreased were enriched in amino acid pathways linked to the TCA cycle. Although amino acid levels declined, the metabolic potential for their utilization increased, while the abundances of core carbon metabolites and several TCA-related intermediates were relatively maintained during storage. Comparative genomic analysis further demonstrated that Bacillaceae MAGs possess a complete TCA cycle and abundant secondary metabolite clusters, whereas Lactobacillaceae lack a full TCA cycle, reflecting contrasting ecological strategies—Bacillaceae as stress-tolerant/homeostasis-maintaining and Lactobacillaceae as rapid-growth/resource-acquiring. Random forest analysis using three core families explained variation in key fermentation functions (cross-validated R2 = 0.504-0.787), supporting an association between microbial succession and Daqu performance. This work suggests that Daqu storage represents a directionally structured ecological succession associated with environmental variation, providing new insight into storage-stage ecological dynamics in solid-state fermentation.

IntroductionEnriched microbial consortia have been proven to effectively regulate agricultural fermentation, yet their practical applications in the fermentation of cigar tobacco leaves are still limited.MethodsIn this study, three natural microbial communities collected from high-quality cigar tobacco leaves were continuously cultured in a medium made of ordinary cigar tobacco leaves to construct enriched stable consortia via the top-down approach. Then, these constructed microbiotas were inoculated into ordinary cigar tobacco leaves to test their fermentation performance.ResultsHigh-throughput sequencing results indicated that after three rounds of transfer culture, three simple and stable bacterial communities were formed, dominated by Corynebacterium, Georgia, Bacillus and Enterococcus. Fermentation with the enriched consortia significantly improved the aroma quality, aroma intensity, sweetness, and aroma richness of CTLs.DiscussionThis study presented a straightforward method for constructing efficient and stable microbiotas based on microbial interactions and the material properties of cigar tobacco leaves in practical production, simplifying the development of microbial consortia for CTL fermentation.

Temperature is a key factor driving microbial community succession and volatile flavor compound formation during the fermentation of cigar tobacco leaves (CTLs). This study systematically investigated microbial community dynamics, co-occurrence networks, and volatile flavor compound (VFCs) profiles of CTLs from Dominica and Yunnan under a 20–60 °C fermentation gradient. High-throughput sequencing identified Staphylococcus, Oceanobacillus, and thermophilic fungi as core microbes potentially associated with aroma formation. Dominica CTLs exhibiting higher microbial diversity than Yunnan CTLs. Dominica CTLs produced abundant esters, alcohols and ketones across different temperature stages, whereas Yunnan CTLs accumulated more pyrazines, indole and terpenoids at high temperatures (≥ 50 °C). Co-occurrence network analysis revealed temperature-driven shifts in microbial interactions: Dominica CTLs formed balanced networks with mixed positive/negative correlations at low temperatures, while Yunnan CTLs developed stable networks dominated by positive correlations at high temperatures. PERMANOVA indicated significant differences in microbial community structure among temperature gradients (R2 = 0.78, p < 0.001). Spearman correlation analysis suggested that core microbes (e.g., Staphylococcus) were significantly correlated with the accumulation of key VFCs (e.g., esters, alcohols). These findings propose a conceptual temperature–microbe–VFC interaction framework, providing theoretical support for optimizing CTL fermentation processes.

Daqu serves as the saccharifying and fermenting starter in Baijiu production, yet the effects of raw material variation on community succession and metabolite formation during fermentation remain unclear. In this study, amplicon sequencing and metabolomic profiling were applied to compare mung bean Daqu (MBQ), produced by substituting pea with mung bean, with traditional medium-temperature Daqu. Microbial succession patterns were largely conserved after mung bean substitution, whereas stage-specific differences were observed in community composition, environmental associations, and microbial co-occurrence patterns. MBQ was enriched in thermotolerant spore-forming bacteria, particularly Kroppenstedtia, and functional fungi, including Thermomyces, Saccharomycopsis, and Aspergillus. These microbial differences were associated with higher saccharification, liquefaction, and fermentation powers at different stages. Spearman correlation analysis revealed distinct microbe-metabolite associations in MBQ involving pyrazines and plant-derived secondary metabolites. MBQ showed higher accumulation of higher alcohols during the middle stage, particularly phenylethyl alcohol (1.05-23.02 mg/kg), exclusive detection of 2,5-dimethylpyrazine and 2-ethylpyrazine, and flavonoid-related non-volatile metabolite profiles. These findings provide new insights into how raw material variation is associated with microbial succession and metabolic features during Daqu fermentation.

Microbial generalists are pivotal for maintaining the stability of fermentation systems, yet their distribution across different Baijiu types remains poorly understood. This study identified generalists and specialists during the pit fermentation of strong-flavor, jian-flavor, and sesame-flavor Baijiu in Anhui Province, and further elucidated their genomic features. Results showed that bacterial communities in all three types are dominated by generalists, whereas fungal communities depend more on diverse specialists. Bacterial generalists were represented by OTUs classified as Acetilactobacillus, Lactobacillus, and Limosilactobacillus. Targeted removal of these generalists increased the robustness of time-series networks, as they correlated negatively with most other taxa and were strongly linked to physicochemical properties. The major species belonging to bacterial generalists included Acetilactobacillus jinshanensis, Lactobacillus acetotolerans, and Limosilactobacillus pontis. These generalists possessed specialized genomic features for niche dominance, characterized by: (i) a low-acquisition, high-growth life history strategy (A/Y < 1); (ii) a preference for sugar metabolism (SAP >0); (iii) a complete multi-layered defense system conferring tolerance to acid and ethanol; and (iv) a streamlined (< 2 Mb) and non-redundant (lacking the TCA cycle) genome that minimizes regulatory burden. This study provides a systematic analysis of generalists across distinct Baijiu types in Anhui Province, offering a theoretical framework for understanding the rules of microbial assembly in the brewing process.

Natural exogenous additives (EA) suitable for the tobacco fermentation need to be developed to enhance the fermentation quality and economic value of low-grade cigar tobacco leaves (CTLs). This study analyzed the impacts of three compound Chinese herbal medicine (CHM) on metabolites and microorganisms during CTLs fermentation. The results manifested that EA facilitated the degradation of total sugar, starch and protein, while enhancing the accumulation of reducing sugar in CTLs. Furthermore, EA raised contents of free amino acids (FAAs), while Asp, Glu, Ser and His were found to be key differential FAAs of CTLs. During fermentation, the total contents of volatile flavor components (VFCs) initially increased and then declined. Furthermore, EA contributed to more harmonious compositions of VFCs by promoting the formation of neophytadiene, ketones, esters and aldehydes, as well as facilitating nicotine degradation. According to variable importance in the projection (VIP) > 1 and odor activity value (OAV) > 1, 7 key differential VFCs were identified. EA enhanced positive microbial interactions and led to a more stable and coordinated symbiotic network. Linear discriminant analysis effect size (LEfSe) identified 9 genera as differentially dominant microorganisms in CTLs, which were closely associated with chemical compositions and key differential flavor metabolites. In addition, EA promoted cigar tobacco characteristics (CTCs) by altering bacterial alpha diversity and influencing the assembly of dominant microbial communities. Overall, this study offered theoretical insights into the innovative applications of CHM in CTLs fermentation, and presented new perspectives for enhancing CTLs quality and customizing flavor profiles.

Quorum sensing (QS) is a central system reflecting microbial collective behavior; however, its role in shaping functional microbial communities within complex solid-state fermentation matrices such as Daqu remains insufficiently understood. Here, we integrated amplicon sequencing, metagenomics, proteomics, and metabolomics to investigate autoinducer-2 (AI-2)-mediated quorum sensing dynamics during Daqu storage. Storage induced a directional succession of the microbial community, revealing two distinct ecological stages. The rapid adjustment stage (0-2 months) was characterized by strong homogeneous selection and rapid species turnover, whereas the slow stabilization stage (3-9 months) was dominated by gradual shifts in microbial relative abundances. Notably, the LuxS/AI-2 pathway, the only QS system detected during Daqu storage, declined rapidly and then stabilized, coinciding with the transition between the two ecological stages. During the early stage, the core QS protein LuxS was tightly associated with the dominant taxon Lactobacillaceae and the methyl donor S-adenosylmethionine, forming a synergistic functional module. In contrast, during the late stage, LuxS became decoupled from stress-tolerant taxa and showed weakened associations with resistance-related metabolic networks. This shift was accompanied by a metabolic transition, with carbon flux gradually redirected from active glycolysis toward the pentose phosphate pathway and amino acid biosynthesis during later stages. Collectively, these findings demonstrate that temporal modulation of the LuxS/AI-2 quorum sensing system represents a critical regulatory node reflecting the transition of the Daqu microbial community from cooperative growth to stress-resilient survival, ultimately shaping metabolic phenotypes and ecosystem functions during storage.

In order to deeply analyze the changing situation of nutrients during the fermentation of Dangshan pear juice by lactic acid bacteria, in this study, the ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) was used to analyze the metabolomics changes during the mixed fermentation of Dangshan pear juice by Lactobacillus plantarum and Lactobacillus acidophilus. The sensory evaluation results indicated that the fermented pear juice exhibited a better balanced sourness and sweetness, along with a more complex aroma profile. The principal component analysis (PCA) results indicated that all samples could be clearly classified into three categories. A total of 80 differential metabolites were identified through metabolomics analysis, mainly including amino acids, peptides, sugars and organic acids. The results by KEGG pathway enrichment analysis indicated that these differential metabolites were mainly involved in some key metabolic pathways such as phenylalanine, tyrosine and tryptophan metabolism, starch and sucrose metabolism, fructose and mannose metabolism, glutathione metabolism, arginine metabolism, as well as the tricarboxylic acid cycle and pyruvate metabolism. The research results showed that lactic acid bacteria fermentation could significantly change the composition of metabolites in pear juice, promoting the biodegradation of complex macromolecular substances, thereby improving its nutrition and functional properties. This study provides a theoretical basis and technical support for the industrial production of fermented beverages from Dangshan sweet pears.

Agricultural waste-derived biochemical fulvic acid (BFA) is a renewable organic resource with potential value as a bio-based plant biostimulant. However, its relatively inert aromatic skeleton and insufficient abundance of polar functional groups limit its high-value utilization. In this study, a fungal co-culture system composed of Coriolopsis trogii and Paraconiothyrium brasiliense was developed for the green bioactivation of BFA. The biotransformation process, structural evolution, and functional performance of the modified BFA were systematically evaluated. The co-culture system showed enhanced Lac and MnP activities and was associated with coordinated changes in amino acid metabolism, cofactor biosynthesis, nucleotide metabolism, and redox-related pathways. Multispectroscopic characterization showed that 12 d fermentation reduced aliphatic structures and enriched oxygen- and nitrogen-containing functional groups, fulvic acid-like fluorescent components, and O/N-associated proton environments. These changes suggested oxidative activation and molecular rearrangement of BFA during fungal co-culture fermentation. In wheat pot assays, purified and sterilized 12 d-BFA significantly increased plant height, dry weight, and total chlorophyll content by 36.04%, 42.10%, and 44.76%, respectively, compared with the control. SOD, POD, and CAT activities were increased by 47.18%, 123.22%, and 96.10%, respectively, while MDA content was decreased by 31.91% (p < 0.05). These findings demonstrate that co-culture fermentation effectively converts recalcitrant BFA into a functionalized derivative with robust biostimulant potential, offering an eco-friendly biotechnological avenue for agricultural waste valorization.

Pit mud serves as a habitat for functional bacteria involved in Baijiu fermentation. While pH is recognized as a key driver of bacterial succession in pit mud, the underlying pattern remains insufficiently characterized. This study systematically investigated the relationship between pH and bacterial communities in pit mud by integrating dataset analysis with pH gradient experiments. The results revealed strong correlations between pH and the key bacterial families, which occurred not only across the time dimension but also in relation to spatial heterogeneity or artificial disturbance. Along the pH gradient, heterogeneous selection may govern bacterial succession. As pH increased, the appearance of Oscillospiraceae signaled the onset of bacterial succession, while the near disappearance of Lactobacillaceae indicated enhanced stability. Six representative genera were further studied, revealing distinct genomic features that changed consistently with pH. During bacterial succession, the abundance of Lactobacillus declined, which predominantly carries genes associated with lactate and acetate metabolism. In contrast, the abundance of Caproicibacterium, which possesses a complete gene cluster for reverse β-oxidation, increased. The pH gradient experiments confirmed that pH influences bacterial assembly mainly through heterogeneous selection, which appeared to be facilitated both by the lactate-utilizing ability of Caproicibacterium and the pH-mediated suppression of Lactobacillus. These findings advance the theoretical understanding of pit mud aging and contribute to the optimization of Baijiu fermentation.

Microbial communities in fermentation pit mud play a key role in determining the quality of Chinese strong-flavor baijiu (CSFB). However, the ecological processes underlying pit mud maturation across spatial and temporal scales remain unclear. In this study, amplicon sequencing and metagenomic analyses were employed to investigate the taxonomic succession, community assembly, and metabolic functions of bacterial and archaeal communities during the transition from fresh pit mud (FPM) to new pit mud (NPM) and old pit mud (OPM). A pronounced depth-dependent succession pattern was observed, with 4 cm representing a critical ecological boundary separating distinct community structures and maturation trajectories. During surface-layer maturation, community assembly shifted from stochastic to deterministic processes, accompanied by homogeneous selection and increasing network complexity. In contrast, stochastic processes remained dominant throughout deep-layer maturation. Metagenomic analyses revealed a functional transition from lactate and acetate production, primarily associated with Lactobacillus in FPM and NPM, to butyrate and caproate production associated with Clostridium and Caproiciproducens in OPM. This functional transition was accompanied by enhanced amino acid metabolism, which was associated with the enrichment of Proteiniphilum and Aminobacterium. Notably, methanogen-mediated interspecies hydrogen transfer (IHT) emerged as a key ecological feature during pit mud maturation. In OPM, IHT networks primarily involving Methanobacterium and Methanosarcina linked methanogenesis with reverse β-oxidation through diverse hydrogen-transfer pathways, reinforcing metabolic interactions underlying caproate production. These findings provide new insights into the ecological mechanisms underlying pit mud maturation and offer a theoretical basis for the directed cultivation of high-quality pit mud in CSFB production.

This review comprehensively examines the structural architecture, catalytic mechanisms, and targeted molecular engineering of α-amylase (primarily the GH13 family), a pivotal biocatalyst in the food industry. We highlight diverse microbial sources of α-amylases and their cost-effective heterologous expression in well-characterized hosts like Bacillus subtilis and Escherichia coli. To overcome extreme operational bottlenecks—such as elevated temperatures and acidic environments—recent advances in protein engineering are critically evaluated. These strategies, including directed evolution, semi-rational design, and advanced immobilization on nanomaterials, synergistically enhance the enzyme’s thermostability, catalytic efficiency, and reusability. Furthermore, this paper synthesizes the state-of-the-art applications of engineered α-amylases across key food processing sectors, including baking, sugar refining, and brewing. By integrating structural biology with advanced material science, this review provides a targeted roadmap for developing next-generation, high-performance α-amylases to address current and future challenges in sustainable food processing.

Mung bean-flavor Baijiu (MBF) incorporates mung bean to develop a distinctive mung bean aroma (MBA), yet its key aroma compounds and formation mechanisms remain unclear. In this study, through aroma extract dilution analysis (AEDA), quantitative measurements, and aroma addition and omission, 22 key aroma compounds were identified that define the flavor profile of MBF. Among these, 2-isopropyl-3-methoxypyrazine, 1,1-diethoxy-3-methylbutane, 1-hexanol, furfural, furfuryl ethyl ether, and dimethyl trisulfide collectively shape the characteristic MBA. Furthermore, the integration of metabolic pathway analysis with amino acid and flavor profiling of the raw materials revealed that mung bean addition elevates the levels of linoleic acid, valine, leucine, and methionine, thereby promoting lipid oxidation, amino acid catabolism, and Maillard reactions that ultimately drive the formation of MBA compounds. These findings establish a molecular basis for MBA and advance the understanding of raw material driven regulation of characteristic aroma formation.

Agaricus bisporus is the most widely consumed edible mushroom and an important source of dietary protein and bioactive compounds. The casing layer plays a critical role in its cultivation, where microbial communities and metabolic activities strongly affect yield and quality, however, the stage-resolved research gap remains that it is still unclear which microbial and metabolic shifts in the casing layer are associated with primordia initiation and subsequent quality formation. In this study, the ecological and metabolic basis of mushroom quality formation was investigated by the combination of microbiome sequencing, non-targeted metabolomics, and functional genomics. Microbial communities exhibited stage-specific dynamics, with significant restructuring during primordia formation, when community cohesion and niche breadth reached their highest levels (0.68 +/- 0.06 and 4.48 +/- 1.05, respectively). A total of 1108 non-volatile metabolites were identified from metabolomic profiling. The adenosine and tryptophan exhibited significant changes and were enriched in energy and amino acid metabolism pathways. A representative strain, Pseudomonas putida AT130, was isolated from the genus Pseudomonas. The gene clusters related to phosphate solubilization, potassium mobilization, lignin degradation, and indole-3-acetic acid biosynthesis were revealed by genome analysis with the multifunctional activities being confirmed through in vitro assays. Pot experiments further showed that AT130 inoculation improved mushroom performance, increasing fruiting body yield by 123.55% and enhancing nutritional traits (protein and soluble sugars increased, whereas ash decreased) relative to the control. These findings linked casing-layer microbiota with mushroom quality and identified AT130 as a promising food-grade bioinoculant to enhance A. bisporus nutritional value and productivity.

In this study, an engineered Kluyveromyces marxianus strain with enhanced mannitol biosynthesis capacity was developed, enabling efficient conversion of molasses and Jerusalem artichoke tubers into mannitol via consolidated bioprocessing. An efficient mannitol-producing strain was constructed by screening of five mannitol synthesis-related genes, with mannitol dehydrogenase (CmMDH) from Candida magnoliae identified as the most effective enzyme. Mannitol production was further enhanced by increasing the copy number of CmMDH. Among the four tested NADPH synthesis-related genes, overexpression of KmUTR1 led to an additional increase in mannitol production. For the final engineered strain Y2CmMDH-KmUTR1, aeration enhancement and pH control substantially increased mannitol production. When employing fructose and glycerol as carbon sources, the maximum mannitol titer reached 175.43 ± 2.02 g/L in single-feed batch fermentation, with peak productivity of 5.26 ± 0.04 g/(L·h) achieved in two-feed batch fermentation. The Y2CmMDH-KmUTR1 strain produced 47.96 ± 0.65 g/L mannitol from non-sterilized molasses at 42 °C, which is higher than that reported in other studies, and produced 48.30 ± 1.18 g/L mannitol from non-sterilized Jerusalem artichoke tubers (JAT) at 42 °C through consolidated bioprocessing. To the best of our knowledge, this is the first successful demonstration of mannitol production at 42 °C and producing mannitol from JAT through consolidated bioprocessing. The developed strain exhibits significant industrial potential, particularly for valorizing agricultural waste molasses and Jerusalem artichoke tubers.

The quality, flavor, and stability of fermented foods depend on the microbial community. However, microbial dynamics are difficult to observe directly, leading to limited control over fermentation. High-throughput sequencing is a revolutionary tool for microbial characterization, among which DNA-based amplicon and metagenomic sequencing are core techniques. Nevertheless, the related data processing workflows in the context of fermented foods have not yet been systematically summarized, hindering the translation of research findings into fermentation practices. This review clarifies the applications of amplicon and metagenomic sequencing in fermented foods. For amplicon sequencing, the impacts of target regions, data preprocessing, and reference databases are addressed. For metagenomic sequencing, sequencing strategies, read-based and binning-based analytical methods, functional annotation, and species-specific databases are discussed. In addition, major strategies for downstream analysis of community data are summarized, including microbial diversity, co-occurrence networks, niche and community assembly, key environmental drivers, and machine learning-based prediction. Amplicon sequencing efficiently reveals microbial succession during fermentation but has limitations in functional annotation. Metagenomic sequencing is notable for functional annotation, enabling the linkage between microbial communities and metabolic potential alongside community characterization. Standardized data preprocessing and specific databases are critical for improving characterization. For community data, integrated analysis allows uncovering the driving factors of microbial succession, thereby helping to regulate fermentation. Notably, the compositional nature of the data must be considered and validated to avoid spurious associations. In summary, the exponential growth of sequencing data will propel the era of precision fermentation.