Abstract
Fungi play critical roles in the production of flavors during pit fermentation. To date, however, knowledge regarding core fungi in JFB remains limited. This study aimed to identify core fungi and evaluate their fermentation performance. Based on criteria including abundance, distribution range, interaction strength, and flavor-producing capacity, Zygosaccharomyces, Aspergillus, Millerozyma, and unclassified Dipodascaceae were identified as core fungi. Dominant species A. niger m6 and Z. pseudobailii J71 exhibited high metabolic activities. Co-culture systems of Saccharomyces cerevisiae J73 with Z. pseudobailii (SZ 1:10) and with A. niger (SA 1:0.1) enhanced the production of alcohols, acids, and esters. Compared with D35, the contents of certain flavor compounds were significantly higher in the SA and SZ co-culture systems. Collectively, this study firstly provides comprehensive exploration and evaluation of core fungi in JFB, establishing a preliminary foundation for constructing synthetic microbial communities and elucidating microbial interaction mechanisms during the production of baijiu flavors.
Keywords: Jiang-flavor baijiu (JFB), Fermented grains (FG), Core fungi, High-throughput sequencing (HTS), Cultivation approach, Mixed fermentation
Graphical abstract
Highlights
-
•
Core fungi were identified and verified based on integrative analysis.
-
•
Zygosaccharomyces and Aspergillus were cultured core fungi.
-
•
High ratio (SZ1:10) and low ratio (SA = 1:0.1) enhanced alcohols, acids, esters production.
-
•
Contents of certain flavors in SA and SZ were higher than those in sample D35.
1. Introduction
As a typical representative of Chinese traditional fermented food, baijiu is produced through highly complex processing steps via spontaneous solid-state fermentation, using Daqu as the starter and sorghum as the raw material. Thousands of flavor substances including alcohols, aldehydes, acids, esters, pyrazines, furans, phenols, and sulfur compounds have been identified in baijiu, particularly in Jiang-flavor baijiu (JFB) (Liu & Sun, 2018). The fermentation process of JFB generally consists of two stages: stacking fermentation and pit fermentation. Study has indicated that the aforementioned flavor compounds are mainly produced during the pit fermentation stage rather than stacking fermentation stage (Liu & Sun, 2018). It is commonly accepted that most flavors in JFB are attributed to the microbial community and their interactions within the cellar composed of bacteria, yeasts, and molds. The stacking fermentation stage facilitates the production of several flavor precursors and enriches the yeast population, thereby preparing for the subsequent pit fermentation. Specifically, starch, protein, and other macromolecules in raw materials are initially hydrolyzed into low-molecular-weight compounds by bacteria and mold (Sakandar et al., 2020). This is followed by alcohol synthesis primarily driven by Saccharomyces cerevisiae. In the final step, aroma-producing yeasts and other fungi generate a wide array of flavor substances, such as alcohols, acids, and esters. Collectively, fungal communities played an important role in the synthesis of flavor substances (Li, Cheng, et al., 2022). Therefore, elucidating the fungal communities in fermented grains (FG) during pit fermentation is essential.
It is well-known that the production of flavors in fermented food, especially baijiu, is often susceptible to interference from fluctuations in the external environment due to its spontaneous fermentation process. As a result, it frequently causes inconsistent product quality and potential safety concerns (Jin et al., 2017). Additionally, the microbial compositions in Jiang-flavor baijiu (JFB) are significantly more complex, and their interaction mechanisms are more intricate compared to other types of baijiu (Wang, Yang, et al., 2019; Zuñiga et al., 2017). To address this intractable challenge, synthetic microbial communities (SynComs) have been proven to be an effective solution. As artificially designed microbial consortia, SynComs significantly simplify the complexity of natural microbial systems and further allow for the targeted design of microbial communities. Recent findings have demonstrated that SynComs are widely applied not only in environmental bioremediation (Ruan et al., 2024), agriculture (Zhou et al., 2022), biochemical synthesis (Jiang et al., 2023), human health (van Leeuwen et al., 2023), but also in fermented foods (Jin et al., 2024). For example, core microbes associated with the formation of key flavors in pepper and Huangjiu fermentation were identified and validated (Li et al., 2023; Peng et al., 2024). Kang et al. (2024) systematically summarized the core microbes and their interactions involved in the baijiu fermentation (Kang et al., 2024). In a recent study, a defined SynCom consisting of 5 bacterial and 4 fungal strains was established, which successfully replicated the fine flavor attributes of chocolate under controlled conditions, demonstrating the potential of SynComs as a promising alternative to spontaneous fermentation in the future (Gopaulchan et al., 2025). As evidenced by these findings, it is apparent that SynComs are frequently accompanied by core microbes. Only by accurately defining these core microbes can we effectively design SynComs and further investigate microbial interaction mechanisms.
Core microbes are generally defined by the following criteria: (1) dominant microorganisms; (2) ubiquitously distributed microorganisms; (3) microorganisms with strong flavor-producing capabilities; and (4) microorganisms engaged in intensive interactions. Initially, core microbes were often identified either as dominant taxa or as those with high flavor-producing yields (Yu et al., 2022; Zhang et al., 2023). Then, Wang and his colleagues firstly proposed a novel integrated standard for defining core microbes that incorporates all four criteria in the context of light-flavor baijiu (Wang, Wu, et al., 2019). Subsequently, the core microbiota in the FG of Luzhou-flavor baijiu was identified (Jiao et al., 2022). Nevertheless, to date, no studies have reported the identification of core microbes in Jiang-flavor baijiu (JFB) based on a comprehensive analysis incorporating all four criteria. In addition, a limitation of previous studies is that core microbes were often identified solely through bioinformatic analysis of high-throughput sequencing and flavor data, without experimental validation (Gutleben et al., 2018). Further research is still needed to gain deeper insights into the microbial interaction mechanisms, ecological roles, and fermentation performance of core microbes following their isolation and cultivation.
Therefore, the main objectives of the present study were to identify core fungal microorganisms based on an integrated analysis of abundance, distribution range, interaction intensity, and flavor compound synthesis capacity, and to further evaluate the fermentation performance of these core fungi after cultivation. Since the quality of Jiang-flavor baijiu (JFB) is highest during the 3rd to 5th fermentation rounds, FG from the 5th round were chosen to identify core fungi through high-throughput sequencing (HTS) combined with correlation analysis. After cultivation, the physiochemical properties, fermentation performances, and flavor profiles of the identified core fungi were assessed and compared under both pure and mixed fermentation conditions. Finally, the flavor synthesis capacity of the optimized co-culture systems was evaluated by comparison with FG collected at the end of pit fermentation.
2. Materials and methods
2.1. Sample collection
FG samples from the fifth round of fermentation were randomly collected from three different cellars at a renowned Jiang-flavor baijiu (JFB) company located in Moutai Town, Guizhou Province. Samples were taken from the surface, middle, and bottom layers of the FG using a five-point sampling technique. A total of 45 sub-samples were thoroughly mixed as one sample, which was immediately transported to the lab under low-temperature conditions and then stored at −20 °C for subsequent analysis of physicochemical parameters, flavor compounds, and HTS. FG samples collected on day 0, 5, 10, 21, and 35 of the cellar fermentation process were designated as D0, D5, D10, D21, and D35, respectively.
2.2. Detection of physiochemical parameters
The moisture, acidity, starch, reducing sugar, and ethanol contents in FG were measured according to the local standard for chemical analyzing solid-state fermented grains (DB34/T 2264–2014). Specifically, moisture was determined gravimetrically and acidity was analyzed using titration. Reducing sugar and starch contents were quantified by reverse titration with a standard glucose solution. Ethanol was measured with alcohol meters following distillation.
2.3. DNA extraction, library construction, and high-throughput sequencing
Genomic DNA was extracted from FG using the E.Z.N.A.® soil DNA Kit (Omega Bio-Tek, USA) according to the manufacturer's instruction. The fungal ITS region was amplified using genomic DNA as the template and the primers ITS1F and ITS2R. The PCR reaction system (20 μL) consisted of 4 μL of 5 × Fast Pfu buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of each primers (5 μM), 0.4 μL of Fast Pfu DNA Polymerase, 10 ng of DNA template, and ddH2O to a final volume of 20 μL. The PCR procedure was as follows: initial denaturation at 95 °C for 3 min, 27 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 45 s, followed by a final extension at 72 °C for 10 min. After verification, the target ITS amplicons were subjected to high-throughput sequencing on the Illumina MiSeq platform (Illumina, San Diego, USA). The raw reads after sequencing required pretreatment and then performed bioinformatic analysis using QIIME2. Specifically, paired-end reads were quality-filtered and denoised, followed by chimera removal with DADA2. High-quality amplicon sequence variants (ASVs) were prepared for the following analysis. The raw sequences data are deposited in the NCBI SRA database under accession number: PRJNA1220299.
2.4. Detection of volatile substance by HS-SPME-GC–MS
Volatile compounds in the FG of JFB were analyzed using headspace-solid phase microextraction coupled with gas chromatography–mass spectrometry (HS-SPME/GC–MS) (Agilent 6890 GC, USA). 2-octanol was used as an internal standard. Chromatographic column DB-WAX and carrier gas helium with a flow rate of 3.0 mL/min were used for GC–MS analysis. Mass spectra were acquired in electron impact mode at 70 eV, with the ion source temperature set to 230 °C. Data were collected in full scan mode across a mass range of 20–550 amu. Compound identification was carried out by comparing the mass spectra with those in the NIST11 reference library. Only compounds with match factors ≥70 % were considered for qualitative analysis.
2.5. Statistics analysis
Physiochemical parameters are expressed as the mean ± standard deviation (SD). Bioinformatic analysis of HTS data was performed on the Majorbio cloud platform. Alpha diversity indices were calculated, and non-metric multidimensional scaling (NMDS) based on the Bray-Curtis distance algorithm was used to assess sample similarity. One-way ANOVA was used to evaluate significant differences among groups. Furthermore, A heatmap was used to visualize the succession of flavor compounds during pit fermentation. Principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) were performed using SIMCA 4.1 software to identify key flavor compounds. Network analysis was conducted for the top 20 dominant fungal species using the Spearman correlation algorithm. Finally, a correlation heatmap between the top 20 dominant fungi and the key flavor compounds was generated based on Spearman correlations (|ρ| > 0.6 & p < 0.05).
2.6. Isolation and molecular identification of core fungi
Fungal strains in FG (samples D0-D35) throughout the whole pit fermentation process were primarily isolated with streak plate approach on both PDA medium and Bengal red medium. After cultivation at 28 °C for 3–6 days, these fungal colonies with obvious colonies and microscopic characters were chosen for molecular identification. Genomic DNA was extracted using a fungal DNA extraction kit (Solarbio Co., Ltd., Beijing). Polymerase chain reaction (PCR) amplification was performed with the extracted DNA and the universal fungal primers ITS1/ITS4. The PCR mixture (50 μL) consisted of 25 μL of 2 × EsTaq MasterMix, 2 μL of DNA template, 2 μL of each primers, and 19 μL of ddH2O. The PCR conditions were shown below: initial denaturation at 94 °C for 6 min; 30 cycles of denaturation at 94 °C for 45 s, annealing at 53 °C for 45 s, and extension at 72 °C for 2 min; followed by a final extension at 72 °C for 10 min. PCR products were checked by 1 % (m/v) agarose gel electrophoresis, and the target bands were sent to Sangon Biotech company for molecular identification. The obtained sequences were submitted to the NCBI database for homology alignment to determine taxonomic classification.
2.7. Detection of physiochemical characteristics and stress tolerance of isolated core fungi strains
The physiological characteristics and stress tolerance of the cultured core fungal strains were evaluated. For Aspergillus strains, biomass, total acids content, enzymatic activities (glycosylase, α-amylase, acid protease), and spore count were determined to reflect their physiological traits. Due to the difficulty of detecting biomass directly in the bran-based solid state fermentation, glucosamine content was used as an indicator of fungal biomass. Total acids were detected by titration, and mold spores were counted using a hemocytometer. For Zygosaccharomyces strains, physiological parameters such as biomass, pH, ethanol production, and reducing sugar content were detected. Biomass was assessed via plate colony counting, while reducing sugar and starch were determined by titration with a standard glucose solution. Ethanol concentration and pH were measured using alcohol meters and pH meters, respectively. Based on these results, selected Aspergillus and Zygosaccharomyces strains were further subjected to stress tolerance assays. For Aspergillus, mycelium dry weight was measured at different ethanol (2 %,4 %,6 %,8 %,10 %,12 %), temperatures (20 °C,24 °C, 28 °C, 32 °C, 36 °C, 40 °C), and pH (2.0,3.0,3.5,4.0,4.5,5.0,5.5,6.0) to assess tolerance capacity. Likewise, the tolerance of Zygosaccharomyces was determined by measuring OD600 of the liquid culture under different stresses: ethanol (2 %–16 % in 2 % increments), temperature (20–40 °C at graded intervals), pH (3.0–6.0), and glucose concentrations (10 %–60 %).
2.8. Evaluating the performance of the co-culture systems by comparison of flavor profiles between mixed fermentation and pure fermentation or fermented grains
Selected strains of Aspergillus and Zygosaccharomyces with excellent physiological characteristics were co-inoculated with Saccharomyces cerevisiae J73—a dominant alcohol-producing yeast—at ratios of 1: 0.1, 1: 0.2, 1: 1, 1: 5, and 1: 10. Following fermentation, parameters including CO2 weight loss, sugar utilization rate, ethanol content, and urea content were determined for each co-culture systems. CO2 weight loss was calculated as the weight of the fermentation system before fermentation-(the weight after fermentation + the weight of water evaporated from the control group). The sugar utilization rate was defined as (initial sugar - residual sugar) / initial sugar. Urea content was detected by Diacetyl monoxime method. Specifically, after heating in the acidic condition, urea condensed with diacetyl via the Fearon reaction to form diazine, whose concentration was quantified by measuring the OD₅₄₀ value using a microplate reader. The fermentation performance of mixed cultures was compared with that of pure cultures. Additionally, the volatile flavor compounds from both pure and mixed cultures at different inoculation ratios were analyzed using HS-SPME/GC–MS. Finally, the flavor profile of the optimized co-culture system was compared with that of sample D35 (fermented grains collected at the end of pit fermentation). All detection conditions were consistent with the methods described in Section 2.4.
3. Results
3.1. Detection of physicochemical parameters
Physicochemical parameters serve as important driving factors for microbial community succession, further affecting flavor compositions in liquor. Five key parameters moisture, acidity, starch, reducing sugar, and ethanol, were measured in the FG. The results showed that moisture, acidity, and ethanol levels increased throughout the 35-day fermentation period, whereas starch and reducing sugar contents decreased (Supplementary Fig. S1).
3.2. Analysis of fungal alpha diversity
The alpha diversity of fungal communities in the FG was assessed (Table S1). The Chao1 index revealed significant variations in fungal abundance over the fermentation period. Microbial richness gradually increased, peaking at D21, before declining. The Shannon and Simpson indices, which reflect species diversity and evenness, respectively, also showed dynamic trends: the Shannon index reached its highest value at D5 and subsequently exhibited considerable fluctuation. A similar trend was observed for the Simpson index, consistent with the pattern seen in the Shannon index (Table S1).
3.3. Abundance analysis of fungal composition of FG during the pit fermentation
To characterize the fungal composition of the FG at different time points, fungi with an abundance >1.00 % were selected for abundance analysis at the phylum level and genus levels (Fig. 1A–B). A total of 7 fungal phyla were identified, among which only 3 fungal phyla had an abundance exceeding 1.00 %, namely, Ascomycota, unclassified_k_Fungi, and Basidiomycota. Ascomycota was overwhelmingly dominant across all samples, accounting for 90.4 % to 98.5 % of the fungal community (Fig. 1A). At the genus level, 117 genera were detected. The top 20 genera, including Zygosaccharomyces, Aspergillus, Millerozyma, unclassified_f__Dipodascaceae, Saccharomyces, and Meyerozyma, were selected to illustrate the succession of the fungal community (Fig. 1B). The genus Aspergillus was present in all FG samples, with higher abundance observed during the initial and final stages compared to the middle stage. Another dominant genus, Zygosaccharomyces, was primarily detected in samples D0 and D5.
Fig. 1.
The dominating and widely distributed fungi in the FG during the pit fermentation. The abundance distribution of fungi in the 5 fermented grain samples at the phylum (A) and genus (B) level; Venn analysis of fungi genera of 5 fermented grain samples (C); NMDS plot of significant difference among 5 FG samples. The green, red, blue, yellow and black circle represent fermented grain samples designated as D0, D5, D10, D21 and D35, respectively. (D); One way-ANOVA analysis was performed to reveal the discrepancy in fungal genera which were dominating and widely distributed (E). “*”, “**” and “***” denoted p < 0.05, p < 0.01 and p < 0.001, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Furthermore, Venn analysis was performed to reveal fungal genera common to all FG samples, revealing 20 genera shared across all five fermentation time points (Fig. 1C). Therein, 13 genera including unclassified_f__Dipodascaceae, Zygosaccharomyces, Millerozyma, Aspergillus, Saccharomyces, Thermoascus, Issatchenkia, Meyerozyma, Monascus, were not only dominant but also widely distributed throughout the fermentation process.
NMDS analysis result showed clear separation among the five samples, indicating distinct fungal community compositions at each time point (Fig. 1D). One-way ANOVA was performed to further to assess the significance of these differences. The results showed significant differences (p < 0.05) in the abundance of 12 genera—including unclassified Dipodascaceae, Zygosaccharomyces, Aspergillus, and Saccharomyces—across the JFB fermentation stages, with the exception of Issatchenkia. Thereinto, unclassified Dipodascaceae, and Saccharomyces increased firstly and then decreased as fermentation progressed. Conversely, Aspergillus and unclassified Aspergillaceae decreased initially and subsequently increased. Zygosaccharomyces was most dominant in the initial phase and reduced throughout fermentation (Fig. 1E).
3.4. Flavor substance detection by HS-SPME/GC–MS
Volatile flavor compounds in five FG samples were analyzed using HS-SPME/GC–MS in the present study. A total of 38 flavor substances were determined, including 18 esters, 7 acids, 6 alcohols, 3 pyrazines, 2 phenols, and 2 other compounds. According to heatmap analysis, most flavor compounds gradually increased during the pit fermentation. Acids including acetic acid, butanoic acid, hexanoic acid, and ethyl esters such as ethyl caproate, ethyl butyrate, ethyl lactate, and ethyl heptanoate mainly accumulated in the later phase. However, compounds such as isobutyl alcohol and pyrazines (e.g., 2,3-dimethylpyrazine and 2,3,5,6-tetramethylpyrazine) were more abundant in the initial phase than in subsequent stages (Fig. 2A). Sample similarity analysis revealed that D0 and D5 clustered closely together, as did D10 and D21, whereas D35 was distinctly separated from the other samples, indicating substantial differences in flavor composition (Fig. 2B). Next, OPLS-DA modeling yielded R2 and Q2 values both above 0.5, confirming the reliability of the model (Fig. 2C). Ten flavor compounds with variable importance in projection (VIP) scores greater than 1—including phenylethanol, ethyl phenylacetate, ethyl lactate, and ethyl caproate—were identified as key flavor substances (Fig. 2D).
Fig. 2.
The succession of flavor substances during the fermentation process and establishment of key flavor substances by OPLS-DA analysis. The dynamic changes of 38 flavors during the pit fermentation were displayed according to their contents by heatmap analysis (Fig. 2A). Principal component analysis (PCA) was performed to reveal differences in flavors profile of FG (Fig. 2B). Samples D0, D5, D12, D27 and D43 represented samples collected on fermentation Day 0, Day 5, Day 12, Day 27 and Day 43, respectively. Values of R2 and Q2 usually suggested the reliability of the predicted model by OPLS-DA (Fig. 2C). The key flavors were determined according to the value of VIP after calculation (Fig. 2D).
3.5. Correlation analysis
3.5.1. Network analysis
To reveal core fungi, it was essential to locate those occupying critical nodes within the network and involved in flavor formation. Therefore, it was necessary to perform a correlation analysis between dominant fungal genera and key flavors. Network analysis showed that 9 genera including unclassified_f__Dipodascaceae, Zygosaccharomyces, Aspergillus, Millerozyma, Thermoascus, Thermomyces, and Monascus possessed no fewer than eight connections, indicating strong interactions with other dominating fungi. Accordingly, these nice genera were considered as highly interactive. Surprisingly, the genus Zygosaccharomyces exhibited negative correlations with eight other fungal genera (Fig. 3A).
Fig. 3.
Correlation analysis among physiochemical factors, dominating microbes and key flavors. Network analysis of the top 20 fungi based on the Spearman algorithm (coefficient > |0.7|, P value <0.05). Nodes in the network denoted dominating fungi. Green and red lines represented positive and negative correlations, respectively (A). Distance-based redundancy analysis (db-RDA) between the top 20 fungi and 5 physiochemical factors. Circles denoted samples' name. Arrows and inverted triangles meant the physiochemical factors and dominating fungi, respectively (B). Heatmap analysis between the top 20 dominating microbes and 10 key flavors based on Spearman algorithm (|ρ|>0.6 & p<0.05). Asterisks in diagram show significant difference. *: P ≤ 0.05, **: 0.001 < P ≤ 0.01, ***: P ≤ 0.001 (C). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3.5.2. db-RDA analysis of fungi and physiochemical factors
The db-RDA results indicated that starch and reducing sugar were positively correlated with samples D0, D5, and D10, whereas acidity, ethanol, and moisture were positively associated with samples D21 and D35 (Fig. 3B). According to Monte Carlo permutation tests, the influence of each physicochemical factor on fungal community structure, ranked by R2 values, was as follows: ethanol > reducing sugar > moisture > starch > acidity (Table S2). Starch and reducing sugar showed positive correlations with Aspergillus, unclassified Dipodascaceae, and Zygosaccharomyces. In contrast, the genus Thermoascus was positively correlated with ethanol, acidity, and moisture.
3.5.3. Correlation analysis of dominating fungi and key flavor substances
Spearman correlation analysis was performed between the top 20 fungal genera and the 10 key flavor substances. Meyerozyma and unclassified_f__Dipodascaceae showed significant positive correlations with all key flavors, particularly ethyl caproate and ethyl caprylate. Similarly, Kodamaea and Millerozyma were positively correlated with most key flavors. In contrast, Zygosaccharomyces and Aspergillus exhibited negative correlations with most selected flavor compounds (Fig. 3C). Regardless of positive or negative correlations, fungi strongly associated with key flavors—such as unclassified_f__Dipodascaceae, Zygosaccharomyces, Aspergillus, and Millerozyma—were identified as important flavor-producing taxa.
Based on a comprehensive evaluation of abundance, distribution breadth, interaction strength, and flavor-producing capacity, the genera Zygosaccharomyces, Aspergillus, unclassified Dipodascaceae, and Millerozyma were identified as core fungi.
3.6. Cultivation of core fungi
To further investigate the physiological and fermentation characteristics of the core microorganisms and validate their flavor synthesis capabilities, cultivation of identified core fungi was a prerequisite. A total of 124 fungal strains were isolated, with representative colonies were illustrated in Supplementary Fig. S2, Supplementary Fig. S3. Among these, 66 fungi (25 molds and 41 yeasts) were selected for taxonomic identification. Based on cultural characteristics, the molds were classified into 4 genera encompassing 9 species, including Aspergillus niger, A. flavus, A. oryzae, and Monascus purpureus. Eight yeast strains including Zygosaccharomyces pseudobailii, Z. bailii, and S. cerevisiae were grouped into 6 genera. Notably, despite repeated attempts, strains belonging to the core genera unclassified Dipodascaceae and Millerozyma were not successfully isolated. Given the high isolation frequency and cultivability of Aspergillus and Zygosaccharomyces, representative strains from these genera were chosen for further analysis.
3.7. Physiochemical and fermentation properties evaluation of genera Aspergillus and Zygosaccharomyces
The biomass, total acid content, enzymatic activities (glucoamylase, α-amylase, acid protease), and spore count were measured for the cultured Aspergillus strains. Additionally, the biomass, pH, ethanol content, and reducing sugar content were detected for the yeast strains Z. peudobaili J71 and Z. bailii J63. Among the Aspergillus strains, A. niger m6 exhibited a faster growth rate, higher saccharification and acid production capacity, and lower spore production compared to the other three strains (Supplementary Fig. S4). Similarly, Z. pseudobailii J71 displayed a higher growth rate, higher sugar consumption, and higher ethanol yield than Z. bailii J63 (Supplementary Fig. S5). Therefore, A. niger m6 and Z. pseudobailii J71 were selected for subsequent tolerance assays.
3.8. Stress tolerance evaluation of A. niger m6 and Z. pseudobailii J71
The tolerance of A. niger m6 to high temperature, ethanol, and acidity assessed by measuring mycelial dry weight as an indicator of viability. In addition, the tolerance of Z. pseudobailii J71 to high temperature, ethanol, acidity, and reducing sugar concentration were determined by optical density at the wavelength of 600 nm (OD₆₀₀). The results showed that A. niger m6 could tolerate high acidity (pH 2.0), elevated temperature (40 °C), and ethanol (10 %) (Supplementary Fig. S6). Meanwhile, Z. pseudobailii J71 tolerated low pH (3.0), high sugar concentration (60 %), and ethanol (16 %) despite weak heat-resistance (Supplementary Fig. S7).
3.9. Effects of A. niger m6 and Z. pseudobailii J71 on the fermentation performance of S. cerevisiae J73
Since S. cerevisiae is a dominating yeast responsible for ethanol production during pit fermentation, the effects of A. niger m6 and Z. pseudobailii J71 on the fermentation performance of S. cerevisiae J73 were further investigated. Parameters including CO₂ weight loss, sugar consumption, ethanol content, and urea content were measured (Table S3). A low inoculation ratio of A. niger m6 (S: A = 1: 0.1) enhanced the fermentation activity of S. cerevisiae J73, resulting in higher sugar utilization, increased ethanol yield, and reduced urea accumulation in the co-culture system. In contrast, a high inoculation ratio of Z. pseudobailii J71 (S: Z = 1: 10) improved sugar consumption and ethanol production by S. cerevisiae J73. Interestingly, urea content was also effectively reduced and maintained at a lower level under this condition.
3.10. Comparison of flavor profiles between mixed and pure cultures
3.10.1. Comparison of flavor profiles between S. cerevisiae J73 and co-culture with Z. pseudobailii J71
The influence of Z. pseudobailii J71 on the flavor compounds produced by S. cerevisiae J73 was investigated, and the results are summarized in Table 1. Compared with the pure culture of S. cerevisiae J73, a high inoculation ratio of Z. pseudobailii J71 (S: Z = 1: 10) reduced the diversity of acids and esters, although the variety of alcohols remained unaffected. However, the total contents of alcohols, acids, and esters were significantly higher in the co-culture system. Specifically, the high inoculation ratio of Z. pseudobailii enhanced the synthesis of higher alcohols, acids, and esters—such as isobutanol, 2,3-butanediol, acetic acid, and ethyl phenylacetate. Therefore, the co-culture system with a ratio of S: Z = 1: 10 was identified as the optimal treatment.
Table 1.
The diversity and contents of flavor substances in the mixed culture system Z. pseudobailii J71 and S. cerevisiae J73.
| No. | Volatile flavors | CAS No. | Inoculation ratio |
||||||
|---|---|---|---|---|---|---|---|---|---|
| S.cerevisiae J73 | Z.pseudobailii J71 | SZ a 1:0.1 | SZ 1:0.2 | SZ 1:1 | SZ 1:5 | SZ 1:10 | |||
| 1 | Isobutanol | 000078–83-1 | 206.19 | 132.18 | 289.21 | 132.16 | 146.82 | 85.42 | 1510.39 |
| 2 | Isopentanol | 000123–51-3 | 1413.18 | 2271.57 | 2733.68 | 2249.37 | 2269.72 | 2740.26 | 12,370.56 |
| 3 | Diisobutylcarbinol | 000108–82-7 | 10.10 | 8.61 | ND | 7.39 | 10.01 | 6.98 | ND b |
| 4 | 2,3-Butanediol | 000513–85-9 | 9.72 | 14.62 | 29.93 | 22.53 | 18.99 | 36.28 | 120.22 |
| 5 | Phenylethanol | 000060–12-8 | 6624.96 | 7562.21 | 15.91 | 1.58 | 67.83 | 7402.04 | 7551.95 |
| 6 | Liningool | 000078–70-6 | 8.07 | 10.06 | 11.03 | 6.59 | 6.90 | 8.35 | 9.23 |
| 7 | 3-Methylthio Propanol | 000505–10-2 | 4.23 | 6.52 | 10.77 | 5.56 | 5.94 | 5.68 | 5.61 |
| 8 | Citronellol | 000106–22-9 | 13.27 | 18.94 | 29.49 | 22.69 | 14.97 | 18.60 | 44.81 |
| 9 | Geraniol | 000106–24-1 | 13.60 | 12.42 | 14.37 | 8.34 | 9.79 | 7.55 | ND |
| 10 | Hexanol | 000111–27-3 | ND | 3.63 | 9.83 | 4.50 | 5.16 | ND | ND |
| 11 | 2,4-Pentanediol | 000625–69-4 | ND | 2.68 | ND | ND | ND | ND | 17.77 |
| 12 | 2-Heptanol | 000543–49-7 | ND | 4.28 | ND | ND | 2.39 | ND | ND |
| 13 | 2-Nonanol | 000628–99-9 | ND | 5.61 | ND | 3.26 | ND | ND | ND |
| 14 | Terpineol | 010482–56-1 | ND | 3.53 | ND | ND | ND | 3.38 | ND |
| 15 | Butanol | 000071–36-3 | ND | ND | ND | 3.19 | ND | ND | 38.43 |
| 16 | 2-Ethyl butanol | 000097–95-0 | ND | ND | 8.46 | 3.97 | 3.53 | ND | 49.64 |
| 17 | alpha- Terpineol | 000098–55-5 | ND | ND | 4.39 | 2.34 | 2.22 | ND | ND |
| 18 | 1- Phenyl-2-propanol | 000698–87-3 | ND | ND | 8.32 | ND | 3.55 | 7.14 | ND |
| Alcohols(18) | / | 8303.33 | 10,056.86 | 3165.40 | 2473.46 | 2567.81 | 10,321.67 | 21,718.61 | |
| 19 | Ethyl phenylacetate | 000101–97-3 | 50.16 | 33.53 | 62.78 | 39.39 | 30.21 | 29.18 | 130.26 |
| 20 | Phenethyl acetate | 000103–45-7 | 209.06 | 745.35 | 752.88 | 517.67 | 559.47 | 744.72 | 862.79 |
| 21 | Gamma-nonanolactone | 000104–61-0 | 71.80 | 25.68 | 77.27 | 34.56 | 25.13 | 32.77 | 816.81 |
| 22 | Isoamyl acetate | 000123–92-2 | 5.61 | 41.78 | 0.00 | 38.29 | 36.30 | 41.92 | ND |
| 23 | Ethyl caprate | 000110–38-3 | 19.53 | ND | ND | ND | ND | ND | 31.22 |
| 24 | Ethyl hexanoate | 000123–66-0 | 21.39 | ND | 9.38 | 88.73 | ND | ND | ND |
| 25 | Ethyl nonanoate | 000123–29-5 | 8.16 | ND | ND | 5.87 | ND | 8.41 | ND |
| 26 | Ethyl heptanoate | 000106–30-9 | 4.08 | ND | ND | ND | ND | ND | ND |
| 27 | Ethyl caprylate | 000106–32-1 | 28.45 | ND | ND | ND | ND | ND | ND |
| 28 | Monobenzyl phenylmalonate | 025774–02-1 | 47.53 | ND | ND | ND | ND | ND | ND |
| 29 | Phenethyl butyrate | 000103–52-6 | ND | 34.30 | 39.27 | 29.38 | ND | 44.01 | ND |
| 30 | Ethyl acetate | 000141–78-6 | ND | 27.22 | 42.81 | 25.75 | 24.26 | 28.86 | ND |
| 31 | Ethyl butanoate | 000105–54-4 | ND | 4.11 | 1.51 | 3.51 | 3.21 | 0.67 | ND |
| 32 | Isoamyl propionate | 000105–68-0 | ND | 4.26 | 5.55 | 3.53 | 3.09 | 5.10 | ND |
| 33 | Diethyl succinate | 000123–25-1 | ND | 6.46 | ND | ND | 3.88 | 6.48 | ND |
| 34 | Propionate - 2-phenethyl ester | 000122–70-3 | ND | 35.69 | ND | ND | ND | ND | ND |
| 35 | Ethyl palmitate | 000628–97-7 | ND | 13.27 | ND | ND | ND | ND | ND |
| 36 | Ethyl oleate | 000111–62-6 | ND | 11.07 | 7.88 | 5.44 | 2.96 | ND | ND |
| 37 | 2- Ethyl methylbutyrate | 007452–79-1 | ND | ND | 4.17 | 2.80 | ND | ND | ND |
| 38 | Ethyl isovalerate | 000108–64-5 | ND | ND | 1.86 | 1.03 | ND | ND | ND |
| 39 | Isoamyl formate | 000110–45-2 | ND | ND | 3799.89 | 1824.74 | ND | ND | ND |
| Esters(21) | / | 465.77 | 982.73 | 4805.26 | 2620.68 | 688.53 | 942.12 | 1841.08 | |
| 40 | Acetate | 000064–19-7 | 46.65 | 55.83 | 94.76 | 42.45 | 50.36 | 49.44 | 562.32 |
| 41 | Isobutyric acid | 000079–31-2 | 27.81 | 59.64 | 80.58 | 37.06 | 51.54 | 41.47 | 365.75 |
| 42 | caproic acid | 000142–62-1 | 29.45 | ND | ND | ND | ND | ND | ND |
| 43 | Octanoic acid | 000124–07-2 | 36.18 | ND | ND | ND | ND | ND | ND |
| 44 | Nonanoic acid | 000112–05-0 | 21.34 | ND | ND | ND | ND | ND | ND |
| 45 | Heptanoic acid | 000111–14-8 | 18.92 | ND | ND | ND | ND | ND | ND |
| 46 | Benzoic Acid | 000065–85-0 | 4.71 | ND | ND | ND | ND | ND | ND |
| 47 | Propanoic acid | 000079–09-4 | ND | 7.70 | 11.35 | ND | 5.57 | ND | 36.77 |
| 48 | 2-Methylbutyric acid | 000116–53-0 | ND | 63.93 | 78.99 | 39.42 | 56.44 | ND | 237.38 |
| 49 | Butyric acid | 000107–92-6 | ND | 14.34 | 22.58 | 10.67 | 12.79 | 7.26 | ND |
| 50 | Methylphenylacetic acid | 000644–36-0 | ND | 7.58 | 23.01 | ND | ND | ND | ND |
| Acids(11) | / | 185.07 | 209.02 | 311.27 | 129.60 | 176.70 | 98.17 | 1202.22 | |
| 51 | 3- hydroxyl −2- butanone | 000513–86-0 | 135.10 | 66.25 | 212.04 | 77.42 | 71.71 | 27.64 | 1474.35 |
| 52 | Damascenone | 023726–93-4 | 7.09 | ND | 8.81 | 9.92 | 5.10 | ND | 47.65 |
| 53 | 3-Phenyl −3- butene −2- ketone | 032123–84-5 | 185.07 | ND | ND | 106.14 | 140.60 | 78.61 | 2334.16 |
| 54 | Benzaldehyde | 000100–52-7 | 20.71 | 28.91 | 28.28 | 10.23 | 18.59 | 16.37 | 929.08 |
| 55 | Phenylacetaldehyde | 000122–78-1 | 110.58 | 7.51 | 107.89 | ND | ND | 4.89 | 52.83 |
| 56 | Aldehyde acetal | 000105–57-7 | 25.24 | ND | ND | ND | ND | ND | 173.97 |
| 57 | 3,5- Dimethylbenzaldehyde | 005779–95-3 | 15.67 | ND | 28.96 | 17.59 | ND | 47.57 | 167.73 |
| 58 | 2,4- Dimethylbenzaldehyde | 015764–16-6 | 0.00 | 23.32 | ND | ND | 12.88 | ND | ND |
| Aldehyde ketone(8) | / | 499.47 | 126.00 | 385.98 | 221.30 | 248.87 | 175.08 | 5179.76 | |
| 59 | 4-Ethenyl −2- methoxyphenol | 007786–61-0 | 17.19 | 20.22 | 29.40 | 14.52 | 24.01 | 27.14 | 59.78 |
| 60 | 2,4- Di-tert-butylphenol | 000096–76-4 | 164.60 | 329.57 | 258.97 | ND | 86.45 | ND | ND |
| 62 | 2,3- Dihydrobenzofuran | 000496–16-2 | 23.09 | 30.68 | 44.82 | 21.70 | 28.66 | 29.53 | 85.50 |
| Others(3) | 204.88 | 380.47 | 333.19 | 36.21 | 139.12 | 56.68 | 145.27 | ||
a: SZ denoted the mixed fermentation with S. cerevisiae J73 and Z. pseudobailii J71, b: No detection.
3.10.2. Comparison of flavor profiles between S. cerevisiae J73 and co-culture with A. niger m6
Likewise, the flavor profiles of pure and co-culture systems containing A. niger m6 and S. cerevisiae J73 were analyzed. Overall, the contents of alcohols, esters, and acids were higher in all S. cerevisiae/A. niger co-cultures compared to the pure S. cerevisiae J73 culture. For example, the co-culture system with a ratio of S: A = 1: 0.1 showed increased levels of alcohols (e.g., isobutanol and 2,3-butanediol), esters (e.g., ethyl phenylacetate and phenethyl acetate), and acids (e.g., acetic acid and phenylacetic acid). However, the contents of these flavor compounds gradually decreased as the inoculation proportion of A. niger m6 increased. Given the production capacity of flavors, the co-culture system with a ratio of S: A = 1: 0.1 was confirmed as an optimized treatment (Table 2).
Table 2.
The diversity and contents of flavor substances in the mixed culture system A. niger m6 and S. cerevisiae J73(μg·L−1).
| No. | Volatile flavors | Chemical CAS No. | Inoculation ratio |
||||||
|---|---|---|---|---|---|---|---|---|---|
| S.cerevisiae J73 | A. niger m6 | SA a 1:0.1 | SA 1:0.2 | SA 1:1 | SA 1:5 | SA 1:10 | |||
| 1 | Isobutanol | 000078–83-1 | 206.19 | 492.58 | 520.60 | 153.45 | 397.63 | 293.24 | 312.85 |
| 2 | Isopentanol | 000123–51-3 | 1413.18 | 4038.74 | 3127.05 | 2381.24 | 2193.76 | 2131.73 | 1489.58 |
| 3 | Diisobutylcarbinol | 000108–82-7 | 10.10 | 15.68 | 17.63 | 11.58 | 7.72 | 6.16 | ND b |
| 4 | 2,3-Butanediol | 000513–85-9 | 9.72 | 88.75 | 79.57 | 69.04 | 54.57 | ND | 101.59 |
| 5 | Phenylethanol | 000060–12-8 | 66.96 | 39.67 | 29.28 | 22.67 | 18.31 | 18.01 | 12.55 |
| 6 | Liningool | 000078–70-6 | 8.07 | ND | 5.07 | 5.53 | 5.01 | 3.79 | 2.96 |
| 7 | 3-Methylthio Propanol | 000505–10-2 | 4.23 | 12.53 | 10.70 | ND | ND | ND | ND |
| 8 | Citronellol | 000106–22-9 | 13.27 | ND | ND | 20.54 | 25.81 | 23.08 | 10.39 |
| 9 | Geraniol | 000106–24-1 | 13.60 | ND | ND | 17.49 | ND | 10.36 | 12.78 |
| 10 | Butanol | 000071–36-3 | ND | 10.41 | 9.06 | 9.11 | 8.79 | 9.21 | 9.91 |
| 11 | 2- ethyl butanol | 000097–95-0 | ND | 21.34 | 21.48 | 17.26 | 7.23 | 4.69 | ND |
| 12 | Hexanol | 000111–27-3 | ND | 4.95 | ND | ND | ND | ND | ND |
| 13 | alpha-Terpineol | 000098–55-5 | ND | 8.24 | ND | ND | ND | ND | ND |
| 14 | 1- Phenyl-2-propanol | 000698–87-3 | ND | 9.04 | 5.60 | 6.29 | 3.09 | ND | ND |
| Alcohols (14) | / | 1745.33 | 4741.93 | 3826.04 | 2714.20 | 2721.92 | 2500.27 | 1952.61 | |
| 15 | Ethyl caprate | 000110–38-3 | 19.53 | 6.41 | 6.61 | 8.22 | ND | ND | 0.00 |
| 16 | Ethyl phenylacetate | 000101–97-3 | 50.16 | 97.58 | 79.55 | 55.07 | 58.78 | 52.17 | 64.70 |
| 17 | Phenethyl acetate | 000103–45-7 | 209.06 | 473.09 | 376.94 | 306.79 | 266.18 | 164.07 | 82.73 |
| 18 | gamma-Nonanolactone | 000104–61-0 | 71.80 | 315.53 | 230.69 | 43.81 | 67.54 | 55.57 | 17.61 |
| 19 | Isoamyl acetate | 000123–92-2 | 5.61 | ND | ND | ND | ND | ND | ND |
| 20 | Ethyl hexanoate | 000123–66-0 | 21.39 | ND | ND | ND | ND | ND | ND |
| 21 | Ethyl heptanoate | 000106–30-9 | 4.08 | ND | ND | ND | ND | ND | ND |
| 22 | Ethyl caprylate | 000106–32-1 | 28.45 | ND | ND | ND | ND | 4.91 | 9.72 |
| 23 | Ethyl nonanoate | 000123–29-5 | 8.16 | ND | 9.60 | ND | 7.38 | 10.72 | 10.50 |
| 24 | Monobenzyl phenylmalonate | 025774–02-1 | 47.53 | ND | ND | ND | ND | ND | ND |
| 25 | Phenethyl butyrate | 000103–52-6 | ND | 24.14 | 18.76 | 16.52 | 13.37 | 8.48 | 5.09 |
| Esters (11) | / | 465.77 | 916.74 | 722.15 | 430.41 | 413.24 | 295.93 | 190.35 | |
| 26 | Acetate | 000064–19-7 | 46.65 | 199.46 | 101.59 | 62.66 | 79.10 | 52.52 | 44.44 |
| 27 | Isobutyric acid | 000079–31-2 | 27.81 | 86.69 | 65.99 | 50.65 | 44.07 | 58.13 | 56.99 |
| 28 | Caproic acid | 000142–62-1 | 29.45 | 19.59 | 29.40 | 11.69 | ND | ND | ND |
| 29 | Octanoic acid | 000124–07-2 | 36.18 | 18.62 | 15.65 | 7.86 | 5.65 | ND | ND |
| 30 | Nonanoic acid | 000112–05-0 | 21.34 | 29.32 | 28.63 | ND | 7.55 | ND | ND |
| 31 | Heptanoic acid | 000111–14-8 | 18.92 | ND | ND | ND | ND | ND | ND |
| 32 | Benzoic Acid | 000065–85-0 | 4.71 | ND | ND | 6.65 | ND | ND | ND |
| 33 | Butyl boric acid | 004426–47-5 | ND | 348.03 | ND | ND | ND | ND | ND |
| 34 | Propanoic acid | 000079–09-4 | ND | 11.62 | ND | ND | ND | ND | ND |
| 35 | 2-methylbutyric acid | 000116–53-0 | ND | 69.50 | 54.03 | ND | ND | ND | 60.62 |
| 36 | Phenylacetic acid | 000103–82-2 | ND | 29.56 | 10.87 | ND | ND | ND | ND |
| 37 | Homovanillic acid | 000306–08-1 | ND | 13.45 | 10.00 | ND | ND | ND | ND |
| Acids (12) | / | 185.07 | 825.84 | 316.16 | 139.52 | 136.37 | 110.65 | 162.05 | |
| 38 | Aldehyde acetal | 000105–57-7 | 25.24 | 55.67 | 46.60 | 27.05 | ND | ND | ND |
| 39 | 3- hydroxyl −2- butanone | 000513–86-0 | 135.10 | 324.99 | 273.95 | 221.59 | 230.25 | 12.22 | 88.68 |
| 40 | Benzaldehyde | 000100–52-7 | 20.71 | 199.86 | 154.97 | 58.57 | 14.19 | 18.20 | 9.72 |
| 41 | Phenylacetaldehyde | 000122–78-1 | 110.58 | 35.05 | 12,994.41 | 18.40 | 14.74 | ND | ND |
| 42 | 3,5- Dimethylbenzaldehyde | 005779–95-3 | 15.67 | 87.95 | ND | 100.70 | ND | ND | ND |
| 43 | Damascenone | 023726–93-4 | 7.09 | 30.10 | 24.61 | 11.53 | ND | ND | ND |
| 44 | 3-Phenyl −3- butene −2- ketone | 032123–84-5 | 185.07 | ND | 727.77 | 443.85 | 255.58 | ND | ND |
| Aldehyde ketone (7) | / | 499.47 | 733.62 | 14,222.32 | 881.69 | 514.76 | 30.41 | 98.40 | |
| 45 | 4-Ethenyl −2- methoxyphenol | 007786–61-0 | 17.19 | 26.50 | 20.73 | 25.40 | 19.30 | 26.35 | 12.63 |
| 46 | 2,4- Di-tert-butylphenol | 000096–76-4 | 164.60 | 416.26 | 313.25 | 214.73 | ND | 154.00 | ND |
| 47 | 2,3- Dihydrobenzofuran | 000496–16-2 | 23.09 | 33.62 | 31.93 | 41.02 | 30.58 | 36.75 | 26.08 |
| 48 | 2- Tert-butyl −4- hydroxyanisole | 000088–32-4 | 4.56 | ND | ND | ND | 2.97 | ND | ND |
| Others (4) | / | 209.44 | 476.38 | 365.91 | 281.15 | 52.85 | 217.11 | 38.71 | |
a: SA denotes mix fermentation with S. cerevisiae J73 and A. niger m6, b: No detection.
3.11. Comparison of volatile flavors between FG and the optimized core fungi co-culture system
To evaluate the flavor-producing capacity of the optimized co-culture system, the volatile flavor profiles of the fermented grains (D35) were compared with those of the optimized combinations (SA and SZ). A total of 33 and 27 flavor compounds were detected in the SA and SZ systems, respectively, which was significantly lower than the 69 compounds identified in the fermented grains (D35). Fifteen flavor compounds were common to both D35 and SA, while 11 flavors were shared between D35 and SZ (Fig. 4A and B). Interestingly, several flavors particularly phenylacetaldehyde, phenethyl acetate, isobutanol, 2,3-butanediol, and acetoin，were more abundant in the SA system compared to D35. Similarly, the SZ system exhibited markedly higher contents of phenylethanol, phenethyl acetate, acetic acid, isobutanol, and acetoin relative to D35 (Fig. 4C and D). These results indicated that although the co-culture systems produced fewer types of flavor compounds, both SA and SZ significantly enhanced the production of esters and higher alcohols.
Fig. 4.
Comparison of volatile flavors between the FG sample D35 and the optimized co-culture system of core fungi. Venn analysis of volatile flavors between D35 and SA (A) as well as D35 and SZ (B). Heatmap analysis of shared volatile flavors between D35 and SA (C) along with D35 and SZ (D). D35 denoted the FG sample collected at day 35, the end of cellar fermentation. SA and SZ represented Saccharomyces cerevisiae/Aspergillus niger and Saccharomyces cerevisiae/Zygosaccharomycespseudobailii co-culture system, respectively.
4. Discussion
4.1. Identification of core microbes based on abundance, distribution, interaction strength, and flavor-producing capacity
Integrated analysis of HTS and flavors data revealed the core fungal community consisted of four genera: Zygosaccharomyces, Aspergillus, unclassified Dipodascaceae, and Millerozyma (Fig. 1, Fig. 3). For core microbes, we argued that the definition of core microbes somewhat ambiguous and continues to evolve rather than being fixed. Different researchers often adopt varying definitions: For instance, earlier study frequently designated dominant microorganisms as core microbes (Cai et al., 2021), while others emphasized metabolic activity (Wang et al., 2016). Afterward, criteria expanded to include a combination of two or three traits such as abundance, distribution, interaction strength, and flavor production (Jiao et al., 2022). A novel framework incorporating all four criteria was firstly proposed to identify core microbes in light-flavor baijiu, marking the inaugural application of this integrated approach within the baijiu research field (S. Wang et al., 2019). This definition has since gained broader adoption. Similarly, core microbes in Xiaoqu light-aroma baijiu were recently identified based on abundance, frequency, interaction intensity, and flavor-producing capacity, leading to the construction of a synthetic microbial community (Wang et al., 2024). In strong-flavor baijiu, core microbiota within fermented grains were also proposed, though isolation of these microbes and exploration of their interactions remained lack (Jiao et al., 2022). As one of the three classical flavor-types baijiu, JFB exhibits a more abundant and complex microbial ecosystem within FG compared with light- and strong-flavor baijiu. To our knowledge, this study is the first to identify core fungi in JFB using this integrated four-dimensional criterion. Furthermore, the identified core fungi were subsequently isolated, and their physiological characteristics, fermentation performance, and flavor profiles were systematically evaluated. We argue that this multi-trait standard offers a more accurate and comprehensive approach for identifying core microbes. It also facilitates the exclusion of non-core microorganisms, thereby streamlining the design and implementation of artificial synthetic microbial communities (Nikoloudaki et al., 2024).
4.2. Successions of core fungi influenced by physicochemical properties and microbial interactions
Following cultivation, fungal strains identified as Aspergillus and Zygosaccharomyces were successfully isolated. In the case of Aspergillus, its abundance decreased initially and then increased during pit fermentation, showing a positive correlation with starch and reducing sugar levels (Fig. 1, Fig. 3). Aspergillus is mainly originated from Daqu and serves as a key producer of amylase and protease, facilitating the biodegradation of raw materials (Xia et al., 2022). This enables Aspergillus to grow rapidly and dominate in the early fermentation stage. Subsequently, its abundance declined as fermentation progressed, but intriguingly increased again toward the end. This pattern can be attributed to the fact that Aspergillus, as a typically aerobic mold, thrives in the aerobic environment present in the FG during the initial phase. The growth of Aspergillus was inhibited due to the consumption of oxygen. However, the production of spores allowed Aspergillus to persist under anaerobic conditions. Furthermore, with the accumulation of acid and ethanol in the later fermentation stage, many acid−/ethanol-sensitive microorganisms dramatically decreased. Consequently, the proportion of Aspergillus increased due to spores formation and its tolerance to high acidity and ethanol. We speculate that the isolated strains—such as A. niger, A. flavus, and A. oryzae—may have originated either from active cells in the early stages (D0 and D5) or from resuscitated spores in the later phases. Correlation analysis revealed that Aspergillus and unclassified Aspergillaceae were positively associated with flavor compounds such as ethyl phenylacetate, ethyl palmitate, acetic acid, and butyric acid, indicating a potential role in the synthesis of various esters and acids. This aligns with previous reports on the contribution of Aspergillus to ester formation (Wang et al., 2023; Xu et al., 2021).
Zygosaccharomyces, another isolated core fungus, was the predominant yeast in the early stages of pit fermentation (D0 and D5; Fig. 1). Both Z. bailii and S. cerevisiae became enriched during stacking fermentation and kept dominant until liquor fermentation. However, Z. bailii was isolated from FG at the end of pit fermentation due to its high stress tolerance capacity (Wei et al., 2023). Zygosaccharomyces showed a positive correlation with starch and reducing sugar. Starch and reducing sugar were most abundant at the beginning of cellar fermentation, providing ample nutrients to support the growth of Zygosaccharomyces. Surprisingly, network analysis among dominant fungi revealed that Zygosaccharomyces exhibited negative interactions with several genera—including Aspergillus, Thermoascus, Thermomyces, Monascus, Meyerozyma, and Issatchenkia—as indicated by green connecting lines. This suggests that Zygosaccharomyces possibly competes with these fungi (Mannazzu et al., 2019). For instance, Zygosaccharomyces may inhibit Aspergillus in sample D0, and its abundance decreased as pit fermentation progressed. This competitive interaction contributed to microbial community succession, allowing other fungi to gradually become dominant (Zhang et al., 2022).
4.3. Stress tolerance mechanism of A. niger and Z. pseudobailii
The presence of A. niger in the FG during the final stage of pit fermentation can be attributed to its notable tolerance to both acid and ethanol stress, as demonstrated in this study (Fig. 1, Supplementary Fig. S6). Regarding acid tolerance mechanism, Seekles et al. (2022) reported that the transcription factors SdrA, WarA, and WarB in A. niger are essential for resistance to sorbic, benzoic, cinnamic, propionic, and acetic acids. Moreover, A. niger ES4 also develop an additional mechanism under ethanol stress, involving the upregulation of glycerolipid metabolism and the utilization of ethanol as a sole carbon source to effectively counteract ethanol-induced stress (Vinayavekhin et al., 2020). Similarly, Z. pseudobailii J71 exhibited tolerance to high sugar, acid, and ethanol stresses simultaneously. The Zygosaccharomyces bailii species complex comprises three recognized species: Z. bailii, Z. parabailii, and Z. pseudobailii. Notably, Z. pseudobailii MT15, initially identified as Z. bailii MT15 and isolated from the FG of JFB, was later designated as an interspecies hybrid of Z. bailii (Braun-Galleani et al., 2018), indicating a close phylogenetic relationship between them. In response to lactic acid stress, Z. parabailii modulates oxidative stress response, iron homeostasis, and cell wall and plasma membrane integrity (Ortiz-Merino et al., 2018). More recently, Wei et al. (2023) demonstrated that the amino acid serine enhances lactic acid stress tolerance and ethanol production in Z. bailii by improving cell wall and membrane integrity, increasing intracellular ATP levels, maintaining cell morphology, and reducing cytoplasmic leakage (Wei et al., 2023).
4.4. Effects of A. niger and Z. pseudobailii on the fermentation performance and flavors synthesis of S. cerevisiae
When co-cultured with S. cerevisiae, A. niger enhanced CO2 weight loss, sugar consumption, and ethanol production across all inoculation ratios compared to pure S. cerevisiae fermentation. These results may be attributed to the high abundance and functional role of A. niger in the early stage of fermentation (D0, D5; Fig. 1). Specifically, A. niger produces amylase, glucoamylase, and protease (Supplementary Fig. S5), supplying glucose and amino acids that support ethanol synthesis by S. cerevisiae. Furthermore, a low inoculation ratio of A. niger caused a reduction in urea levels, thereby inhibiting the formation of ethyl carbamate, which was not in agreement with a previous finding reported by Liu (Liu et al., 2021).
In the case of Z. pseudobailii, a high inoculation ratio enhanced ethanol metabolic activity while simultaneously inhibiting urea accumulation and subsequent ethyl carbamate formation. Saccharomyces cerevisiae was the main producer of ethyl carbamate (Wang et al., 2021). Microbial interactions play a critical role in the reduction of ethyl carbamate (Du et al., 2018; Fang et al., 2019). For instance, both lactic acid bacteria and non-conventional yeasts including Z. bailii C-7 isolated from JFB weakened the production of ethyl carbamate (Du et al., 2018). Therefore, the application of Z. pseudobailii contributing to the inhibition of harmful substances, especially ethyl carbamate deserves to be considered.
In mixed fermentation of Z. pseudobailii and S. cerevisiae, both the diversity and content of flavor compounds were significantly influenced by the inoculation ratio, with either positive or negative effects. As previously reported (Li, Xu, et al., 2022), mixed fermentations with Z. bailii from Baijiu and S. cerevisiae significantly reduced higher alcohols, fatty acids, and ethyl esters while increased acetate esters in winemaking compared with pure S. cerevisiae fermentation.
Intriguingly, heatmap analysis indicated negative correlations between Zygosaccharomyces / Aspergillus and key flavor compounds (Fig. 3). However, in co-culture systems with S. cerevisiae, all S. cerevisiae / A. niger combinations yielded higher contents of alcohols, esters, and acids than pure S. cerevisiae J73. Similarly, the co-culture treatment SZ (1:10) showed markedly higher total contents of these flavor groups despite a reduction in the diversity of acids and esters. In other words, both Z. pseudobailii and A. niger can produce or enhance the production of flavors (Table 1, Table 2). This apparent contradiction between correlation analysis and experimental flavor profiles may be explained by species-specific metabolic differences and variations in flavor composition. Firstly, the key flavors identified were primarily esters, which differed in composition from those produced in co-culture. Secondly, both Zygosaccharomyces and Aspergillus encompass numerous species, and correlations at the genus level may not reflect the behavior of individual species or strains.
4.5. Evaluation of flavors-producing capacity in optimized co-culture systems compared with FG
When comparing the flavor profiles between the actual fermented grain (FG) sample (D35) and the co-culture systems (SA and SZ), the diversity of flavor compounds was significantly higher in the FG sample than in optimized co-culture systems SA and SZ. However, surprisingly, the contents of several specific flavor compounds—such as phenylethanol, phenylacetaldehyde, phenethyl acetate, isobutanol, 2,3-butanediol, and acetoin—were higher in co-culture systems SA and SZ than in D35 (Fig. 4). This discrepancy can be attributed to the fact that FG contains a highly diverse microbial community comprising bacteria, actinomycetes, yeasts, and molds. The microbial community succession and their interactions play a crucial role in shaping the overall flavor profile (Melkonian et al., 2023). In contrast, the elevated levels of the aforementioned esters and higher alcohols in the SA and SZ systems likely result from the optimized inoculation ratios and reduced competition among microorganisms (Zuñiga et al., 2017).
In summary, Zygosaccharomyces, Aspergillus, Millerozyma, and unclassified Dipodascaceae were identified as core fungal genera in JFB based on an integrated analysis of abundance, distribution, interaction strength, and flavor-producing capacity. The physiological characteristics, fermentation performance, and flavor profiles of the isolated core species Z. pseudobailii and A. niger were systematically investigated. The flavor outcomes of the optimized co-culture systems were further validated against those of natural FG. Though the absence of core bacteria possibly influenced the final performance of the synthetic microbial communities, this study provides comprehensive insight into core fungal community of JFB, establishing a preliminary foundation for constructing more complex co-culture systems involving both bacteria and fungi. Such systems could ultimately enable more controlled and efficient fermentation processes. Future work will focus on integrating core bacteria with these fungal strains to explore their interactions and further optimize flavor production.
CRediT authorship contribution statement
Xiangyong Zeng: Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition, Formal analysis, Conceptualization. Yunman Zou: Writing – original draft, Investigation, Formal analysis. Dong Li: Writing – review & editing, Data curation. Chaoyang Wei: Writing – review & editing, Funding acquisition. Dounan Li: Writing – review & editing, Visualization, Funding acquisition. Yan Yan: Writing – review & editing. Qiang Lin: Writing – review & editing, Visualization, Funding acquisition.
Fundings
This work was supported by National Natural Science Foundation of China (32260580, 32460572), Liquor Making Biological Technology and Application of key laboratory of Sichuan Province (NJ2024-06), Natural Science Foundation of Sichuan Province (2025ZNSFSC0292), Guizhou Provincial Science and Technology Projects (ZK [2022] 114, ZK [2023] 088), Guizhou Key Laboratory of New Quality Processing and Storage of Ecological Specialty Food (No. ZSYS[2025]023).
Declaration of competing interest
I have nothing to declare.
Acknowledgments
It was grateful that the fermented grains samples were provided by a famous Moutai-flavor liquor company, Maotai town, Guizhou province. The bioinformatic analysis was performed based on Majorbio platform (https://www.majorbio.com/web/www/index).
Glossary
- FG
Fermented grains
- HTS
High throughput sequencing
- HS-SPME/GC-MS
Headspace-solid phase microextraction/gas chromatography–mass spectrometry
- JFB
Jiang-flavor Baijiu
- OPLS-DA
orthogonal partial least squares discriminant analysis
- PCA
Principal component analysis
- RDA
redundancy analysis
- SynComs
Synthetic microbial communities
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2025.103067.
Contributor Information
Xiangyong Zeng, Email: xyzeng1@gzu.edu.cn.
Qiang Lin, Email: linqiang@cib.ac.cn.
Appendix A. Supplementary data
Physiochemical factors of 5 fermented grain samples, including starch, reducing sugar, moisture, ethanol, and acidity. Samples D0, D5, D12, D27, and D43 represented samples collected on fermentation Day 0, Day 5, Day 12, Day 27, and Day 43, respectively.
The colonies morphology of isolated yeast strains from 5 fermented grain samples based on the traditional methods.
The colonies morphology of isolated mold strains from 5 fermented grain samples based on the traditional methods.
The physiochemical characters of strains from the cultured core fungi Aspergillus. Content of glucosamine (A), spore count (B), content of total acid (C), and enzymatic activities (D) were chosen to reflect the physiochemical characters. Shape triangle, circle, square, and diamond in the panel A-C represented strains A. niger m6, A. fumigatus m14, A. flavus m10, and A. oryzae m3, respectively. The black, gray, and light blue in the bar chart denoted glucoamylase, α-amylase, and acid protease, respectively.
The physiochemical characters of strains from the cultured core fungi Zygosaccharomyces. The biomass (A), reduced sugar (B), and content of ethanol (C) were chosen to reflect the physiochemical characters. Shape square and triangle in the panel A-C represented strains Z. pseudobailii J71 and Z. bailii J63, respectively.
Stress tolerance of optimized strain A. niger m6. The dry weight of mycelium was calculated under the different temperatures (20 °C–40 °C) (A), pH value (2.0–6.0) (B), and ethanol concentrations (2 %–12 %) (C).
Stress tolerance of optimized strain Z. pseudobailii J71. The OD values of cell suspension at the wavelength of 600 nm were calculated under the different temperatures (20 °C–40 °C) (A), pH value (2.0–6.0) (B), and ethanol concentrations (2 %–12 %) (C).
Supplementary material
Data availability
Data will be made available on request.
References
- Braun-Galleani S., Ortiz-Merino R.A., Wu Q., Xu Y., Wolfe K.H. Zygosaccharomyces pseudobailii, another yeast interspecies hybrid that regained fertility by damaging one of its MAT loci. FEMS Yeast Research. 2018;18:7. doi: 10.1093/femsyr/foy079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cai W., Wang Y., Ni H., Liu Z., Liu J., Zhong J., Hou Q., Shan C., Yang X., Guo Z. Diversity of microbiota, microbial functions, and flavor in different types of low-temperature Daqu. Foodservice Research International. 2021;150(Pt A) doi: 10.1016/j.foodres.2021.110734. [DOI] [PubMed] [Google Scholar]
- Du H., Song Z., Xu Y. Ethyl carbamate formation regulated by lactic acid bacteria and nonconventional yeasts in solid-state fermentation of Chinese Moutai-flavor liquor. Journal of Agricultural and Food Chemistry. 2018;66(1):387–392. doi: 10.1021/acs.jafc.7b05034. [DOI] [PubMed] [Google Scholar]
- Fang R., Zhou W., Chen Q. Ethyl carbamate regulation and genomic expression of Saccharomyces cerevisiae during mixed-culture yellow rice wine fermentation with Lactobacillus sp. Food Chemistry. 2019;292:90–97. doi: 10.1016/j.foodchem.2019.04.014. [DOI] [PubMed] [Google Scholar]
- Gopaulchan D., Moore C., Ali N., Sukha D., Florez Gonzalez S.L., Herrera Rocha F.E.…Castrillo G. A defined microbial community reproduces attributes of fine flavour chocolate fermentation. Nature Microbiology. 2025;10:2130–2152. doi: 10.1038/s41564-025-02077-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gutleben J., Chaib De Mares M., van Elsas J.D., Smidt H., Overmann J., Sipkema D. The multi-omics promise in context: From sequence to microbial isolate. Critical Reviews in Microbiology. 2018;44(2):212–229. doi: 10.1080/1040841x.2017.1332003. [DOI] [PubMed] [Google Scholar]
- Jiang Y., Wu R., Zhang W., Xin F., Jiang M. Construction of stable microbial consortia for effective biochemical synthesis. Trends in Biotechnology. 2023;41(11):1430–1441. doi: 10.1016/j.tibtech.2023.05.008. [DOI] [PubMed] [Google Scholar]
- Jiao W., Xie F., Gao L., Du L., Wei Y., Zhou J., He G. Identification of core microbiota in the fermented grains of a Chinese strong-flavor liquor from Sichuan. LWT. 2022;158 doi: 10.1016/j.lwt.2022.113140. [DOI] [Google Scholar]
- Jin G., Zhu Y., Xu Y. Mystery behind Chinese liquor fermentation. Trends in Food Science and Technology. 2017;63:18–28. doi: 10.1016/j.tifs.2017.02.016. [DOI] [Google Scholar]
- Jin R., Song J., Liu C., Lin R., Liang D., Aweya J.J.…Yang S. Synthetic microbial communities: Novel strategies to enhance the quality of traditional fermented foods. Comprehensive Reviews in Food Science and Food Safety. 2024;23(4) doi: 10.1111/1541-4337.13388. [DOI] [PubMed] [Google Scholar]
- Kang J., Huang X., Li R., Zhang Y., Chen X.X., Han B.Z. Deciphering the core microbes and their interactions in spontaneous baijiu fermentation: A comprehensive review. Food Research International. 2024;188 doi: 10.1016/j.foodres.2024.114497. [DOI] [PubMed] [Google Scholar]
- van Leeuwen P.T., Brul S., Zhang J., Wortel M.T. Synthetic microbial communities (SynComs) of the human gut: Design, assembly, and applications. FEMS Microbiology Review. 2023;47(2) doi: 10.1093/femsre/fuad012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li M., Xu X., Bi S., Pan X., Lao F., Wu J. Identification and validation of core microbes associated with key aroma formation in fermented pepper paste (Capsicum annuumL.). Food Research. International. 2023;163 doi: 10.1016/j.foodres.2022.112194. [DOI] [PubMed] [Google Scholar]
- Li R.R., Xu M., Zheng J., Liu Y.J., Sun C.H., Wang H., Guo X.W., Xiao D.G., Wu X.L., Chen Y.F. Application potential of baijiu non-saccharomyces yeast in winemaking through sequential fermentation with Saccharomyces cerevisiae. Frontiers in Microbiology. 2022;13 doi: 10.3389/fmicb.2022.902597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li Y., Cheng Y., Wang H., Hu X., Wang L., Huang Y. Diverse structure and characteristics of the fungal community during the different rounds of Jiang-flavoured baijiu production in Moutai town. LWT. 2022;161 doi: 10.1016/j.lwt.2022.113313. [DOI] [Google Scholar]
- Liu H., Sun B. Effect of fermentation processing on the flavor of baijiu. Journal of Agricultural and Food Chemistry. 2018;66(22):5425–5432. doi: 10.1021/acs.jafc.8b00692. [DOI] [PubMed] [Google Scholar]
- Liu X., Bai W., Zhao W., Qian M., Dong H. Correlation analysis of microbial communities and precursor substances of ethyl carbamate (EC) during soy sauce fermentation. LWT. 2021;152 doi: 10.1016/j.lwt.2021.112288. [DOI] [Google Scholar]
- Mannazzu I., Domizio P., Carboni G., Zara S., Zara G., Comitini F.…Ciani M. Yeast killer toxins: From ecological significance to application. Critical Reviews in Biotechnology. 2019;39(5):603–617. doi: 10.1080/07388551.2019.1601679. [DOI] [PubMed] [Google Scholar]
- Melkonian C., Zorrilla F., Kjærbølling I., Blasche S., Machado D., Junge M., Sørensen K.I., Andersen L.T., Patil K.R., Zeidan A.A. Microbial interactions shape cheese flavour formation. Nature Communications. 2023;14(1):8348. doi: 10.1038/s41467-023-41059-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nikoloudaki O., Aheto F., Di Cagno R., Gobbetti M. Synthetic microbial communities: A gateway to understanding resistance, resilience, and functionality in spontaneously fermented food microbiomes. Food Research International. 2024;192 doi: 10.1016/j.foodres.2024.114780. [DOI] [PubMed] [Google Scholar]
- Ortiz-Merino R.A., Kuanyshev N., Byrne K.P., Varela J.A., Morrissey J.P., Porro D., Wolfe K.H., Branduardi P. Transcriptional response to lactic acid stress in the hybrid yeast Zygosaccharomyces parabailii. Applied and Environmental Microbiology. 2018;84:5. doi: 10.1128/aem.02294-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Peng Q., Zheng H., Zhou H., Chen J., Xu Y., Wang Z., Xie G. Elucidating core microbiota in yellow wine (Huangjiu) through flavor-oriented synthesis and construction of microbial communities. Food Research International. 2024;197(Pt 1) doi: 10.1016/j.foodres.2024.115139. [DOI] [PubMed] [Google Scholar]
- Ruan Z., Chen K., Cao W., Meng L., Yang B., Xu M.…Xu X. Engineering natural microbiomes toward enhanced bioremediation by microbiome modeling. Nature Communications. 2024;15(1):4694. doi: 10.1038/s41467-024-49098-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sakandar H.A., Hussain R., Farid Khan Q., Zhang H. Functional microbiota in Chinese traditional baijiu and Mijiu Qu (starters): A review. Foodservice Research International. 2020;138 doi: 10.1016/j.foodres.2020.109830. [DOI] [PubMed] [Google Scholar]
- Seekles S.J., van Dam J., Arentshorst M., Ram A.F.J. Natural variation and the role of Zn(2)Cys(6) transcription factors SdrA, WarA and WarB in sorbic acid resistance of Aspergillus niger. Microorganisms. 2022;10:2. doi: 10.3390/microorganisms10020221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vinayavekhin N., Kongchai W., Piapukiew J., Chavasiri W. Aspergillus niger upregulated glycerolipid metabolism and ethanol utilization pathway under ethanol stress. Microbiologyopen. 2020;9(1) doi: 10.1002/mbo3.948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang C., Wang M., Zhang M.P. Ethyl carbamate in Chinese liquor (baijiu): Presence, analysis, formation, and control. Applied Microbiology and Biotechnology. 2021;105(11):4383–4395. doi: 10.1007/s00253-021-11348-1. [DOI] [PubMed] [Google Scholar]
- Wang H., Wang Y., Ruan Y., Ma D., Wang H., Yang S.…Chen Y. Core microbes identification and synthetic microbiota construction for the production of Xiaoqu light-aroma Baijiu. Food Research International. 2024;183 doi: 10.1016/j.foodres.2024.114196. [DOI] [PubMed] [Google Scholar]
- Wang M.Y., Yang J.G., Zhao Q.S., Zhang K.Z., Su C. Research progress on flavor compounds and microorganisms of Maotai flavor Baijiu. Journal of Food Science. 2019;84(1):6–18. doi: 10.1111/1750-3841.14409. [DOI] [PubMed] [Google Scholar]
- Wang S., Wu Q., Nie Y., Wu J., Xu Y. Construction of synthetic microbiota for reproducible flavor compound metabolism in Chinese light-aroma-type liquor produced by solid-state fermentation. Applied and Environmental Microbiology. 2019;85(10) doi: 10.1128/AEM.03090-18. e03090-03018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang Y., Quan S., Zhao Y., Xia Y., Zhang R., Ran M., Wu Z., Zhang W. The active synergetic microbiota with Aspergillus as the core dominates the metabolic network of ester synthesis in medium-high temperature Daqu. Food Microbiology. 2023;115 doi: 10.1016/j.fm.2023.104336. [DOI] [PubMed] [Google Scholar]
- Wang Z., Lu Z., Shi J., Xu Z. Exploring flavour-producing core microbiota in multispecies solid-state fermentation of traditional Chinese vinegar. Scientific Reports. 2016;6(1) doi: 10.1038/srep26818. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wei J., Nie Y., Du H., Xu Y. Serine improves lactic acid stress tolerance and ethanol production in Zygosaccharomyces bailii in baijiu fermentation. Journal of Agricultural and Food Chemistry. 2023;71(50):20295–20303. doi: 10.1021/acs.jafc.3c06246. [DOI] [PubMed] [Google Scholar]
- Xia Y., Zhu M., Du Y., Wu Z., Gomi K., Zhang W. Metaproteomics reveals protein composition of multiple saccharifying enzymes in nongxiangxing daqu and jiangxiangxing daqu under different thermophilic temperatures. International Journal of Food Science and Technology. 2022;57(8):5102–5113. doi: 10.1111/ijfs.15818. [DOI] [Google Scholar]
- Xu Y., Huang H., Lu H., Wu M., Lin M., Zhang C., Zhao Z., Li W., Zhang C., Li X., Sun B. Characterization of an Aspergillus niger for efficient fatty acid ethyl ester synthesis in aqueous phase and the molecular mechanism. Frontiers in Microbiology. 2021;12 doi: 10.3389/fmicb.2021.820380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yu S., Song J., Hu T., Wang J., Liu X., Zheng Y., Shi L., Wan S., Wang M. Unraveling the core functional bacteria and their succession throughout three fermentation stages of broad bean paste with chili. Food Science and Human Wellness. 2022;11(4):874–885. doi: 10.1016/j.fshw.2022.03.011. [DOI] [Google Scholar]
- Zhang H., Tan Y., Wei J., Du H., Xu Y. Fungal interactions strengthen the diversity-functioning relationship of solid-state fermentation systems. mSystems. 2022;7(4) doi: 10.1128/msystems.00401-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang L., Xiong S., Du T., Xu Y., Madjirebaye P., Huang G., Guan Q., Xiong T. Effect of microbiota succession on the dynamics of characteristic flavors and physicochemical properties during the soy sauce fermentation. Food Bioscience. 2023;54 doi: 10.1016/j.fbio.2023.102883. [DOI] [Google Scholar]
- Zhou X., Wang J., Liu F., Liang J., Zhao P., Tsui C.K.M., Cai L. Cross-kingdom synthetic microbiota supports tomato suppression of Fusarium wilt disease. Nature Communications. 2022;13(1):7890. doi: 10.1038/s41467-022-35452-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zuñiga C., Zaramela L., Zengler K. Elucidation of complexity and prediction of interactions in microbial communities. Microbial Biotechnology. 2017;10(6):1500–1522. doi: 10.1111/1751-7915.12855. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Physiochemical factors of 5 fermented grain samples, including starch, reducing sugar, moisture, ethanol, and acidity. Samples D0, D5, D12, D27, and D43 represented samples collected on fermentation Day 0, Day 5, Day 12, Day 27, and Day 43, respectively.
The colonies morphology of isolated yeast strains from 5 fermented grain samples based on the traditional methods.
The colonies morphology of isolated mold strains from 5 fermented grain samples based on the traditional methods.
The physiochemical characters of strains from the cultured core fungi Aspergillus. Content of glucosamine (A), spore count (B), content of total acid (C), and enzymatic activities (D) were chosen to reflect the physiochemical characters. Shape triangle, circle, square, and diamond in the panel A-C represented strains A. niger m6, A. fumigatus m14, A. flavus m10, and A. oryzae m3, respectively. The black, gray, and light blue in the bar chart denoted glucoamylase, α-amylase, and acid protease, respectively.
The physiochemical characters of strains from the cultured core fungi Zygosaccharomyces. The biomass (A), reduced sugar (B), and content of ethanol (C) were chosen to reflect the physiochemical characters. Shape square and triangle in the panel A-C represented strains Z. pseudobailii J71 and Z. bailii J63, respectively.
Stress tolerance of optimized strain A. niger m6. The dry weight of mycelium was calculated under the different temperatures (20 °C–40 °C) (A), pH value (2.0–6.0) (B), and ethanol concentrations (2 %–12 %) (C).
Stress tolerance of optimized strain Z. pseudobailii J71. The OD values of cell suspension at the wavelength of 600 nm were calculated under the different temperatures (20 °C–40 °C) (A), pH value (2.0–6.0) (B), and ethanol concentrations (2 %–12 %) (C).
Supplementary material
Data Availability Statement
Data will be made available on request.