Skip to main content
NCBI home page
Food Chemistry: X logoLink to Food Chemistry: X
. 2026 Sep 16;39:104484. doi: 10.1016/j.fochx.2026.104484

Microecological succession and quality enhancement of fermented tomato sour soup by synthetic functional microbiota

Shuting Zhong a,1, Chuanbo Zhang a,⁎,1, Anqin Zhu a, Tingyan Xu a, Liangliang Xiao b, Yunsheng Wang c, Rui Wang a, Yuping Deng a
PMCID: PMC13628686  PMID: 42825030

Abstract

Tomato sour soup, a traditional fermented food with health benefits, often exhibits inconsistent quality during natural fermentation. Three microbial consortia (yeasts, Eurotium, and their combination) were engineered to enhance fermentation. Physicochemical parameters, antioxidant activity, microbial communities, and metabolomic profiles were monitored throughout fermentation. Eurotium inoculation produced the strongest effects, including the highest total acid (21.10 g/kg), total phenolics (75.41 mg/g), and total flavonoids (29.41 mg/g), the lowest nitrite accumulation, and superior DPPH/ABTS radical scavenging activities. Multi-omics integration revealed that Eurotium reshaped microbial succession, activated phenylpropanoid and flavonoid biosynthesis pathways, and drove the biotransformation of flavonoids into more potent antioxidant derivatives, including quercetin and phloretin. These findings demonstrate that Eurotium is a promising microbial strain for improving the quality and antioxidant properties of tomato sour soup.

Keywords: Tomato sour soup, Functional microbiota, Antioxidant activity, Microbial diversity, Metabolomics

Graphical abstract

Unlabelled Image

Open in a new tab

Highlights

  • •

    Eurotium consortium enhanced acidification and nitrite degradation.

  • •

    Eurotium reshaped microbial community structure and metabolic functions.

  • •

    Eurotium activated phenylpropanoid and flavonoid biosynthesis pathways.

  • •

    Eurotium promoted flavonoid biotransformation into potent antioxidant derivatives.

1. Introduction

Tomatoes possess a distinctive flavor profile and are rich in bioactive antioxidant compounds, including lycopene, β-carotene, γ-aminobutyric acid (GABA), phenolics, and flavonoids (Kang et al., 2022). Tomato sour soup is a traditional fermented dish from the Qiandongnan Miao and Dong Autonomous Prefecture in Guizhou Province, China. It is produced by anaerobic fermentation of tomatoes supplemented with red chili, ginger, glutinous rice, and other ingredients (Zheng, Deng, et al., 2024). Beyond serving as a fundamental seasoning in traditional dishes such as sour fish and sour beef soup, tomato sour soup is associated with promising health benefits. Studies have indicated its potential to reduce fat accumulation, ameliorate hyperlipidemia, and prevent hyperuricemia and gout (Song et al., 2024; Yang et al., 2021). During tomato sour soup fermentation, microbial succession and metabolism drive the transformation and enrichment of bioactive components while simultaneously shaping the unique flavor characteristics of the product. Lactic acid bacteria (LAB), as the dominant microorganisms in this fermentation system, play a critical role in acidification and flavor development. They promote lactic acid accumulation and lower system pH, synergizing with acid-tolerant microorganisms to facilitate the synthesis and accumulation of organic acids, esters, and terpenoids, thereby significantly enhancing the system's aromatic characteristics and overall flavor (Feixia et al., 2024).

Traditional sour soup processing relies on spontaneous fermentation by environmental microorganisms, which often results in inconsistent and uncontrollable product quality because of environmental variability. Inoculation with selected starters offers a promising strategy to address this limitation. For instance, inoculation of red sour soup with Lactobacillus buchneri shortened the production cycle while maintaining flavor characteristics comparable to those of naturally fermented samples (Xiong et al., 2020). Co-inoculation of tomato sour soup with Lentilactobacillus buchneri and Pediococcus ethanolidurans not only promoted rapid accumulation of organic acids but also increased the levels of characteristic aroma compounds with high relative odor activity values (ROAVs) (Liu, Yang, Liu, et al., 2025). Inoculation of tomato sour soup with Lentilactobacillus buchneri, Pediococcus ethanolidurans, and Kazachstania bulderi enhanced fermentation, increased organic acid and total flavonoid levels, promoted the production of key flavor substances such as acetoin and methyl butyrate, and enhanced its fruity and creamy aroma characteristics (Liu, Hu, et al., 2023).

However, research on starter-enhanced tomato sour soup fermentation has largely focused on LAB and yeasts, with little attention given to other functional microbiota. Acid-tolerant genera such as Eurotium are commonly found in traditional fermentation ecosystems and exhibit robust metabolic activity. For example, Debaryomyces and Eurotium contribute to ham ripening and show positive correlations with flavor compounds (Li et al., 2024). Co-fermentation of Eurotium cristatum with Aspergillus niger in dark tea enhances floral and sweet notes (Chen et al., 2021), while co-fermentation with Bacillus promotes the production of bioactive secondary metabolites and inhibits Staphylococcus aureus (Wang, Chen, et al., 2023). As a sole starter, E. cristatum increased free polyphenols 4.98-fold during solid-state fermentation of soybean via hydrolase secretion (Shi & Shi & Hu, 2024), and its liquid fermentation of tree peony petals reduced astringency while enhancing antioxidant activity (Ren et al., 2026). Collectively, these findings highlight the broad metabolic versatility of Eurotium and suggest its potential as a functional starter for tomato sour soup fermentation.

To achieve directional enhancement of tomato sour soup fermentation, this study engineered three functional microbial consortia for inoculation. The SG group was inoculated with the yeast consortium, the EG group was inoculated with the Eurotium consortium, and the MG group was inoculated with both consortia, with an uninoculated UG group serving as the control. To systematically evaluate the effects of different microbial consortia on fermentation performance and antioxidant properties, we monitored physicochemical parameters (pH, total acidity, amino acid nitrogen, and nitrite), bioactive components (total phenolics, total flavonoids, and DPPH/ABTS radical scavenging activities), microbial community shifts (via high-throughput sequencing), and global metabolic profiles (via UPLC-MS/MS-based untargeted metabolomics) throughout the fermentation process. By integrating these multi-dimensional datasets, we analyzed the bio-perturbation effects induced by different inoculations and investigated the correlations between microbial succession, metabolic changes, and antioxidant activity. This work provides observational evidence for Eurotium-mediated tomato sour soup fermentation and offers a reference for starter culture selection.

2. Materials and methods

2.1. Materials

All strains were isolated from traditional Chinese fermented products. Eurotium strains include Eurotium amstelodami Q4 and Eurotium cristatum G12; yeast strains include Saccharomyces cerevisiae W1, Saccharomycopsis fibuligera Y4, Debaryomyces hansenii Y2, and Wickerhamomyces anomalus Z1. E. cristatum G12 was obtained from dark tea (Liubao, Guangxi), whereas the others were derived from high-temperature Daqu used in Jiang-flavor Baijiu production (Renhuai, Guizhou). All strains were molecularly identified and preserved in the research laboratory of microbial resources and industrial application at Guizhou Normal University. Fermentation experiments were conducted at Wuyanghong Company (Qiandongnan, Guizhou).

Spore and cell suspensions were prepared as described by Xu et al. (2025). Eurotium was cultured on PDA at 30 °C for 7 days, and yeasts were cultured on YPD at 30 °C for 1 day. Cultures were harvested with sterile water, shaken at 200 rpm for 10 min, and adjusted to 2 × 107 CFU/mL using a hemocytometer.

2.2. Enhanced fermentation of tomato sour soup

Cleaned tomatoes were crushed and transferred to traditional fermentation jars, then fermented at 24 °C for 18 days. The control group (UG) received no starter culture, whereas the EG group was inoculated with 0.5% (w/w) Eurotium consortium (E. amstelodami Q4 and E. cristatum G12, 1:1, v/v), the SG group with 0.5% (w/w) yeast consortium (S. cerevisiae W1, S. fibuligera Y4, D. hansenii Y2, and W. anomalus Z1, 1:1:1:1, v/v), and the MG group with 0.25% (w/w) Eurotium consortium plus 0.25% (w/w) yeast consortium (total: 0.5%, w/w). Physicochemical parameters were monitored on days 1, 3, 5, 7, 9, 12, 15, and 18. Antioxidant activity, microbial diversity, and metabolites were assessed on days 1, 3, 9, and 18. Samples were stored at 4 °C and − 80 °C.

2.3. Determination of physicochemical properties

The pH of tomato sour soup was measured with a pH meter. Total acid content was determined by potentiometric titration to an endpoint of pH 8.2, in accordance with GB 12456–2021 (National Food Safety Standard for Determination of Total Acid in Foods) and referenced against T/GZSX 025–2018 (Guizhou Sour Soup Seasoning Standard). Amino acid nitrogen content was determined according to GB 5009.235–2016 (National Food Safety Standard for Determination of Amino Acid Nitrogen in Foods). Nitrite content was determined following GB 5009.33–2016 (National Food Safety Standard for Determination of Nitrite and Nitrate in Foods).

2.4. Determination of TPC, TFC, and radical scavenging

Total phenolic content (TPC) was determined using the Folin-Ciocalteu colorimetric method with minor modifications. Briefly, samples were centrifuged at 8000 rpm for 10 min at 4 °C, and 125 μL of the supernatant was mixed with 125 μL of Folin-Ciocalteu reagent, then incubated in the dark for 2 min. Subsequently, 500 μL of 7.5% Na2CO3 solution and 250 μL of sterile water were added, vortexed, and incubated in a 40 °C water bath in the dark for 30 min. After incubation, 250 μL of the reaction mixture was transferred to a 96-well plate, and the absorbance was measured at 760 nm. Gallic acid was used as the standard to construct a calibration curve (y = 15.096× + 0.2191, R2 = 0.9999).

Total flavonoid content (TFC) was determined by the aluminum nitrate (Al(NO3)3) colorimetric method with minor modifications. Samples were centrifuged at 8000 rpm for 10 min at 4 °C, and 50 μL of the supernatant was mixed successively with 20 μL of 5% NaNO2 solution (incubated for 6 min), 20 μL of 10% Al(NO3)3 solution (incubated for 6 min), and 200 μL of 10% NaOH solution (incubated for 15 min), with vortexing after each addition and absorbance measured at 510 nm. Rutin was used as the standard to construct a calibration curve (y = 0.7766× + 0.0438, R2 = 0.9999).

DPPH and ABTS radical scavenging activities were determined using modified methods described by Segoviano-León et al. (2025). Samples were centrifuged at 8000 rpm for 10 min at 4 °C, and the supernatants were diluted 10-fold with 80% ethanol. For the DPPH assay, 100 μL of diluted sample was mixed with 100 μL of 0.2 mmol/L DPPH solution, vortexed, and incubated in the dark at room temperature for 30 min, followed by absorbance measurement at 517 nm. For the ABTS assay, 50 μL of diluted sample was mixed with 195 μL of ABTS working solution, vortexed, and incubated in the dark at room temperature for 6 min, and measured at 734 nm. For both assays, the absorbance of the sample group was recorded as A. The control group (B) was prepared by replacing the radical solution with 80% ethanol, while the blank group (C) was prepared by replacing the sample solution with 80% ethanol. The radical scavenging rate was calculated as follows:

Scavenging rate%=1−A−B÷C×100% (1)

2.5. Illumina sequencing of bacterial and fungal communities

Genomic DNA was extracted from tomato sour soup samples using the E.Z.N.A. Soil DNA Kit. DNA concentration and purity were measured with a NanoDrop 2000 spectrophotometer, and DNA integrity was verified by 1% agarose gel electrophoresis. For bacterial community analysis, the V5-V7 region of 16S rRNA was amplified with primers 799F (5’-AACMGGATTAGATACCCKG-3′) and 1193R (5’-ACGTCATCCCCACCTTCC-3′); fungal ITS1-ITS2 was amplified with ITS1F (5’-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2R (5’-GCTGCGTTCTTCATCGATGC-3′). The 20-μL PCR mixture contained 0.8 μL of each primer, 10 μL of 2 × Pro Taq Master Mix, and 1 μL of template DNA. The PCR protocol for bacterial amplification consisted of initial denaturation at 95 °C for 3 min, followed by 27 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s, with a final extension at 72 °C for 10 min. Fungal PCR was performed under the same conditions but with 35 cycles. Amplified products were separated on 2% agarose gels, purified using a PCR Clean-Up Kit, and quantified with a Qubit 4.0 fluorometer. Purified amplicons were prepared using the NEXTFLEX Rapid DNA-Seq Kit and sequenced on the Illumina NextSeq 2000 platform. Raw reads were quality-filtered with fastp (v0.19.6) and merged with FLASH (v1.2.7). Denoising was performed using the DADA2 algorithm to generate ASV representative sequences and abundance tables. Taxonomic classification of bacterial 16S and fungal ITS sequences was performed against SILVA 138.2 and UNITE 10.0, respectively, using the classify-sklearn (Naive Bayes) method with a 0.7 confidence threshold. The raw sequencing data have been deposited in the NCBI SRA under accession numbers PRJNA1398018 (bacteria) and PRJNA1398041 (fungi).

2.6. Untargeted metabolomics analysis

Approximately 200 ± 5 mg of each sample was weighed and mixed with 800 μL of extraction solution (methanol: water = 4:1, v/v) containing four internal standards, including 0.02 mg/mL L-2-chlorophenylalanine. The mixture was transferred to grinding-bead tubes, homogenized at −10 °C and 50 Hz for 6 min, ultrasonically extracted at 5 °C and 40 kHz for 30 min, and then kept at −20 °C for 30 min. After centrifugation at 13,000 rpm and 4 °C for 15 min, the supernatant was collected for LC-MS analysis. Quality control samples were prepared by pooling equal volumes of each sample and inserted into the analytical sequence to monitor system stability.

LC-MS analysis was performed using a Thermo Fisher Scientific UHPLC system coupled to an Orbitrap Exploris 240 mass spectrometer. Chromatographic separation was achieved on an ACQUITY UPLC HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm particle size, Waters, USA). The mobile phase consisted of solvent A (95% water +5% acetonitrile with 0.1% formic acid) and solvent B (47.5% acetonitrile +47.5% isopropanol +5% water with 0.1% formic acid). The injection volume was 3 μL, the flow rate was 0.40 mL/min, and the column temperature was maintained at 40 °C. MS spectrometry data were acquired in both ion modes over the mass range of 70–1050 m/z. The ion source parameters were set as follows: sheath gas at 60 arb., auxiliary gas at 20 arb., ion transfer tube temperature at 350 °C, spray voltages at +3400 V and −3000 V, and stepped collision energies at 20–40-60 V. Raw data were processed in Progenesis QI 3.0 (Waters Corporation, Milford, USA) for baseline filtering, peak detection, alignment, and integration, generating a matrix of retention times, m/z, and intensities. Feature peaks were identified by matching MS and MS/MS spectra against the Human Metabolome Database (HMDB, http://www.hmdb.ca/) and the Kyoto Encyclopedia of Genes and Genomes (KEGG, https://www.kegg.jp/), with a mass tolerance of <10 ppm and MS/MS fragmentation scores.

2.7. Statistical analysis

All experiments were conducted in triplicate, and the results were expressed as mean ± SD. Statistical analyses were performed in SPSS 26.0, and figures were generated with GraphPad Prism 9.5.0. Co-occurrence network analyses of microorganisms and metabolites were carried out in RStudio (v4.4.1) using relevant packages, and the resulting networks were visualized with Gephi (v0.10.1).

3. Results

3.1. Production process and physicochemical changes of tomato sour soup

The production process of enhanced fermented tomato sour soup is illustrated in Fig. 1, and the changes in physicochemical parameters and antioxidant activity during fermentation are presented in Fig. 2. As shown in Fig. 2A and B, the pH and total acid content of the UG, EG, and MG groups followed similar trends throughout fermentation. Specifically, pH decreased rapidly during the first three days, reaching final values below 3.6 in the three groups. Among the four groups, the EG group exhibited the highest total acid content (21.10 ± 0.18 g/kg) at the end of fermentation. Nitrite levels in all groups increased to a peak on day 5 and then declined progressively (Fig. 2C). Notably, the peak nitrite concentrations in the SG, EG, and MG groups were significantly lower than those in the UG group. Among the inoculated groups, the EG group maintained the lowest nitrite levels throughout the fermentation period, indicating the most pronounced inhibitory effect on nitrite accumulation. For amino acid nitrogen, levels decreased in all groups during the first 7 days of fermentation (Fig. 2D). On day 7, the MG group showed significantly lower values than the other groups; thereafter, a gradual increase was observed in both the EG and MG groups.

Fig. 1.

Fig. 1

Open in a new tab

Schematic diagram of the enhanced fermentation process for tomato sour soup.

Fig. 2.

Fig. 2

Open in a new tab

Dynamic changes in physicochemical parameters and antioxidant activity during tomato sour soup fermentation. pH (A), total acid (B), amino acid nitrogen (C), nitrite (D), total phenolic content (TPC) (E), total flavonoid content (TFC) (F), DPPH radical scavenging activity (G), ABTS radical scavenging activity (H). Significant differences were observed between sample groups denoted by different letters (P < 0.05).

As shown in Fig. 2E and F, TPC and TFC increased progressively in all four groups throughout fermentation. At the end of fermentation, the EG group exhibited the highest TPC (75.41 ± 1.75 mg/g), followed by the SG (69.79 ± 1.04 mg/g) and MG (69.25 ± 0.37 mg/g) groups, both of which were significantly higher than that of the UG group (64.48 ± 0.86 mg/g). A similar trend was observed for TFC, with the EG (29.41 ± 0.26 mg/g) and MG (29.50 ± 1.68 mg/g) groups showing the highest levels at the end of fermentation, followed by the SG group (25.55 ± 0.68 mg/g); all inoculated groups exhibited significantly higher TFC than the UG group (23.49 ± 1.12 mg/g). Regarding antioxidant activity, the DPPH radical scavenging rates of the SG, EG, and MG groups increased substantially from day 3 onward. They remained significantly higher than those of the UG group throughout the remainder of fermentation (P < 0.05) (Fig. 2G). As for ABTS radical scavenging activity, the MG group exhibited a significant 20% increase on day 3, whereas the UG group showed a decrease at the same time point (Fig. 2H). By the end of fermentation, the ABTS radical scavenging rates of all three inoculated groups were significantly higher than those of the UG group.

3.2. Microbial alpha diversity during tomato sour soup fermentation

High-throughput sequencing of tomato sour soup samples from the four groups yielded 4,563,057 high-quality bacterial sequences and 5,992,570 high-quality fungal sequences, which were clustered into 5791 bacterial ASVs and 2808 fungal ASVs, respectively (Supplementary Tables S1 and S2). Coverage values (99.99%–100%) and rarefaction curves (Supplementary Fig. S1) confirmed that the sequencing depth was sufficient to comprehensively capture the microbial diversity across all samples.

Alpha diversity analysis (Supplementary Fig. S2) revealed distinct dynamics between fungal and bacterial communities across the four groups. For fungi, the Chao index of the EG group increased on day 3 and then declined, whereas the SG and MG groups showed a significant decrease on day 3 and subsequently stabilized; at this time point, the UG and EG groups exhibited significantly higher Chao values than the SG and MG groups (P < 0.05). The Shannon index of the EG group increased on day 3, dropped sharply on day 9, and rebounded on day 18, whereas the UG, SG, and MG groups decreased on day 3 and remained relatively stable. Notably, the UG group exhibited the highest Shannon value on day 1 (P < 0.05), whereas the MG group showed the highest value on day 18 (P < 0.05). For bacteria, the Chao index decreased continuously across all groups with no significant intergroup differences (P > 0.05). The Shannon index of the UG group declined on day 3 and then stabilized, whereas the SG, EG, and MG groups decreased on day 3, increased on day 9, and then declined again on day 18.

3.3. Microbial community composition and dynamic succession

PCoA and ANOSIM analysis revealed pronounced disparities in fungal community structure among the four groups on day 1, persisting through day 3 (Fig. 3A). By day 9, the fungal assemblages of the UG and EG groups had converged, whereas the SG and MG groups showed significant differentiation, with further divergence observed on day 18. For the bacterial community (Fig. 3B), the SG group did not induce marked structural changes on day 1, whereas both EG and MG elicited substantial community reorganization. On day 3, the bacterial successional directions of EG and MG aligned, in contrast to the significant divergence between UG and SG. By days 9 and 18, significant differences in bacterial community structure were evident among all four groups.

Fig. 3.

Fig. 3

Open in a new tab

Changes in microbial community during tomato sour soup fermentation. Principal co-ordinates analysis of fungal (A) and bacterial (B) communities; relative abundance of fungal (C) and bacterial (D) communities at the genus level.

Fig. 3C and D illustrate the successional dynamics of microbial communities at the genus level. A total of 237 fungal and 208 bacterial genera were identified; those with relative abundances above 0.1% were retained for analysis. For the fungal community (Fig. 3C), Kazachstania proliferated in the UG and EG groups on day 3 and dominated on days 9 and 18; Saccharomyces surged in the SG group on day 3 and remained above 70% through day 18; and Wickerhamomyces established dominance in the MG group on day 3, exceeding 50% relative abundance throughout days 3 to 18. Aspergillus declined in the EG and MG groups on day 3, falling below 1% by day 9. Debaryomyces and Saccharomycopsis declined sharply in the SG group on day 3 but gradually in the MG group, while Pichia and Hanseniaspora were detected only at early time points. For the bacterial community (Fig. 3D), Lentilactobacillus was abundant across all groups on day 1, decreased on day 3, and rebounded during days 9 to 18, reaching 72.27% in the EG group on day 18. Weissella increased across all groups on day 3, while Lactiplantibacillus increased in the UG and EG groups on day 3 and in the SG and MG groups on day 9. Secundilactobacillus declined on day 3 but recovered in the SG, UG, and EG groups during days 9 to 18. Levilactobacillus showed group-specific patterns, whereas Pantoea, Staphylococcus, and Bacillus were detected only on day 1.

LEfSe analysis identified 5 fungal and 10 bacterial genera with significant differences among the four groups (LDA > 3.5, P < 0.05; Fig. 4). For the fungal community, the UG group was characterized by Cladosporium and Pichia on day 1; the EG group by Aspergillus on day 1 and Kazachstania on day 9; and the MG group by Wickerhamomyces on day 3. For the bacterial community, the UG group was distinguished by Lactococcus on day 3 and Secundilactobacillus on day 9; the SG group by Weissella on day 3 and Levilactobacillus on day 9; the EG group by Perweissella, Agrilactobacillus, and Pediococcus on day 1, Lacticaseibacillus on day 3, and Lentilactobacillus on day 18; and the MG group by Lactiplantibacillus on day 9.

Fig. 4.

Fig. 4

Open in a new tab

LEfSe analysis identified significantly different microbial genera during tomato sour soup fermentation. LDA scores of fungal (A) and bacterial (B) genera (LDA > 3.5, P < 0.05).

3.4. Co-occurrence patterns of microbial communities

Co-occurrence networks based on Spearman's correlations (|r| > 0.7, P < 0.05) revealed distinct microbial interaction patterns across the four groups (Fig. 5). In the UG group, positive correlations occurred between Levilactobacillus and Pichia and among Weissella, Hanseniaspora, and Lactiplantibacillus, whereas negative correlations linked Lentilactobacillus with both Lactiplantibacillus and Levilactobacillus, and Pichia with both Levilactobacillus and Kazachstania. In the SG group, Pichia correlated positively with Lentilactobacillus, while Saccharomyces and Kazachstania each showed strong positive associations with both Levilactobacillus and Lactiplantibacillus. In the EG network, strong positive correlations were observed among fungal genera, including between Aspergillus and Pichia, and between Saccharomyces and Hanseniaspora, as well as between Lentilactobacillus and Kazachstania. In the MG group, Aspergillus correlated positively with Secundilactobacillus, Saccharomyces with Lactiplantibacillus, and Levilactobacillus with both Debaryomyces and Saccharomycopsis.

Fig. 5.

Fig. 5

Open in a new tab

Co-occurrence patterns of microbial communities during tomato sour soup fermentation. Each node represents an individual ASV, with its size proportional to the relative abundances. Edges indicate significant correlations (|r| > 0.7, P < 0.05), with red and green lines representing positive and negative correlations, respectively. UG (A), SG (B), EG (C), MG (D). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.5. Metabolite classification and dynamic panorama during tomato sour soup fermentation

Untargeted metabolomics based on LC-MS/MS was employed to profile metabolite changes during the enhanced fermentation of tomato sour soup. A total of 4118 metabolites were identified (2051 in positive mode and 2067 in negative mode), of which 2687 were annotated against the HMDB database and 1139 were mapped to 326 KEGG pathways. PLS-DA revealed distinct metabolic clustering patterns (Fig. 6A), with PC1 and PC2 explaining 42.1% and 12.4% of the total variance, respectively (R = 0.9638, P = 0.001). On day 1, the metabolic profiles of UG and SG were similar, while EG and MG exhibited comparable patterns; by day 3, clear separation among the four groups was observed. The robustness of the PLS-DA model was validated by permutation testing, in which R2 exceeded Q2 (R2 = 0.1552, Q2 = −0.3382).

Fig. 6.

Fig. 6

Open in a new tab

PLS-DA score plots with permutation tests (A) and KEGG enrichment scatter plots of differential metabolites (B) during tomato sour soup fermentation.

HMDB classification grouped the identified metabolites into 17 superclasses (Supplementary Fig. S3), with organic acids and derivatives (22.55%) being the most abundant. Cluster analysis (Supplementary Fig. S4) revealed distinct temporal accumulation patterns. On day 1, trigonelline, citric acid, trehalose, adenosine, and naringenin were upregulated. Succinic acid, GABA, 3-hydroxybutyric acid, and glyceraldehyde were upregulated on day 9 and peaked by day 18. L-Phenylalanine, D-(+)-tryptophan, and trans-3-indoleacrylic acid were consistently upregulated in the EG group throughout fermentation but downregulated in the SG group. L-Isoleucine was upregulated on day 18 across the SG, EG, and MG groups, while D-sorbitol was upregulated on day 9 in all groups but only in the EG and UG groups by day 18.

To identify differentially abundant metabolites (DAMs) during tomato sour soup fermentation, OPLS-DA with stringent criteria (VIP > 1, P < 0.05, FC > 1.2 or < 0.833) was applied, and results were presented in volcano plots (Supplementary Fig. S5). Across all fermentation phases, the MG group exhibited the highest number of DAMs relative to the UG group, followed by the EG and SG groups. KEGG enrichment analysis (Fig. 6B) revealed that riboflavin metabolism and bile secretion were commonly enriched pathways across all three treatment groups. Additionally, the EG and MG groups showed significant enrichment of multiple phenylpropanoid and flavonoid pathways, including phenylpropanoid biosynthesis, flavonoid biosynthesis, flavone and flavonol biosynthesis, and flavonoid degradation.

3.6. Differential metabolite profiling

Among the 281 DAMs identified within the organic acids and derivatives superclass, carboxylic acids and derivatives were the predominant subclass (Fig. 7A). L-glutathione and its derivatives were consistently upregulated in the UG and SG groups but downregulated in the MG group throughout fermentation. Oxiglutatione was downregulated in the UG and SG groups during days 1–3, but upregulated across all groups during days 9–18. 2′-deoxymugineic acid was significantly upregulated in all groups between days 9 and 18. O-Acetyl-l-serine and homomethionine were upregulated across all groups during days 1–3. Gly-Phe was downregulated in the EG and SG groups on day 3, and in the UG and MG groups on day 9. Ile-Leu and Tyr-Pro exhibited upregulation trends in the MG and EG groups during days 1–3; during days 9–18, they were significantly upregulated in the SG group but markedly downregulated in the UG group.

Fig. 7.

Fig. 7

Open in a new tab

Clustering analysis of differential metabolites during tomato sour soup fermentation. Organic acids and derivatives (A), and phenylpropanoids and polyketides (B). Red and blue colors indicate relatively high and low abundance, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Within the phenylpropanoids and polyketides superclass, 201 DAMs were identified, with flavonoids being the most abundant subclass (Fig. 7B). Chalconaringenin was upregulated in the SG, EG, and MG groups during days 1–3. Vitamin P was upregulated across all four fermentation phases in the SG and UG groups. In contrast, in the EG and MG groups, upregulation was observed only on day 1. Chrysoeriol, kaempferol, and luteolin were upregulated throughout all four phases in the EG and MG groups but downregulated in the UG and SG groups. Phloretin was significantly upregulated in the SG, EG, and MG groups during days 3–9. Quercetin and ethyl caffeate were significantly upregulated in the same three groups between days 9 and 18.

3.7. Correlation analysis of differential metabolites with physicochemical indicators and core microbiota

A correlation analysis was performed on the top 20 differential metabolites from organic acids and derivatives, phenylpropanoids, and polyketides, together with the physicochemical parameters including pH, total acidity, TPC, TFC, and DPPH and ABTS radical scavenging activities (Supplementary Fig. S6). Meanwhile, the top 10 fungal and bacterial genera by relative abundance were selected to construct correlation network diagrams using the screening criteria |R| > 0.3 and P < 0.05 (Fig. 8).

Fig. 8.

Fig. 8

Open in a new tab

Network analysis illustrating correlations between microorganisms and differential metabolites in tomato sour soup. Edges represent significant correlations (|r| > 0.3, P < 0.05), with red and green indicating positive and negative correlations, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

As shown in Supplementary Fig. S6, chalconaringenin and vitamin P showed significant negative correlations with DPPH and ABTS radical scavenging activities, as well as with TPC and TFC. In contrast, their transformation products, quercetin and ethyl caffeate, showed strong positive correlations with these antioxidant indices, and luteolin also showed positive correlations.

In the microbial–metabolite correlation network (Fig. 8), Aspergillus showed significant positive correlations with luteolin, chrysoeriol, kaempferol, chalconaringenin, and Gly-Phe. Kazachstania correlated positively with 2′-deoxymugineic acid, quercetin, and luteolin, but negatively with vitamin P. Pichia correlated positively with vitamin P, chalconaringenin, Gly-Phe, and O-Acetyl-l-serine. Saccharomycopsis was positively correlated with chalconaringenin, and Saccharomyces with ethyl caffeate. Among bacterial genera, Lentilactobacillus showed significant positive correlations with ethyl caffeate, 2′-deoxymugineic acid, quercetin, oxiglutatione, and luteolin. Lactiplantibacillus correlated positively with phloretin, ethyl caffeate, 2′-deoxymugineic acid, quercetin, and luteolin, but negatively with vitamin P. Secundilactobacillus correlated positively with oxiglutatione and L-glutathione. Weissella showed a positive correlation with Tyr-Pro and phloretin.

4. Discussion

In this study, specific functional microbial consortia were inoculated into the tomato sour soup fermentation system, revealing their targeted regulatory effects on the fermentation process. During the initial fermentation phase, pH decreased rapidly across all groups, indicating that LAB and other microorganisms promptly utilized sugars to produce acids, thereby rapidly acidifying the system. This early acidification not only contributes to the characteristic sour taste of the product but also effectively inhibits the growth of most spoilage and pathogenic microorganisms (Xiong et al., 2020). Throughout fermentation, pH steadily declined, accompanied by a progressive accumulation of total acids (Luo et al., 2020). By the end of fermentation, the pH values of the UG, EG, and MG groups remained below 3.6, meeting the requirements of the Guizhou sour soup local standard (T/GZSX 025-2018).

Nitrite content peaked on day 5 and then declined steadily across all groups, consistent with the characteristic nitrite metabolism observed in vegetable fermentation, and this decline may be attributed to nitrite decomposition under acidic conditions and its degradation by LAB and other microorganisms (Huang et al., 2024). Among all groups, the EG group exhibited the lowest nitrite levels throughout fermentation. Although the precise mechanism remains unclear, this reduction is tentatively attributed to synergistic effects of chemical decomposition, potential enzymatic activities, and microbial interactions. Further investigation is required to elucidate the underlying pathways and to confirm the safety benefit of Eurotium inoculation in tomato sour soup fermentation.

Amino acid nitrogen content decreased in all groups during the early fermentation phase, likely reflecting the preferential utilization of free amino acids as a nitrogen source by growing microorganisms (Liu, Li, et al., 2023). The significantly lower amino acid nitrogen content in the MG group compared with the other three groups may reflect enhanced conversion of amino acid nitrogen to free amino acids, facilitated by the co-fermentation system comprising Eurotium, yeasts, and LAB. During the mid-to-late fermentation stages, amino acid nitrogen content increased in the EG and MG groups, a change primarily attributable to the inoculated Eurotium spp. This observation is further supported by the findings of Ryu et al. (2021), who demonstrated that the relative abundance of Aspergillus before fermentation correlates with increased post-fermentation amino acid nitrogen concentration. Given that the umami intensity of fermented foods is significantly positively correlated with amino acid nitrogen content (Cai et al., 2024), the increased amino acid nitrogen levels in the EG and MG groups may imply enhanced flavor quality. These dynamics illustrate the gradual conversion of proteins during fermentation, where strain-specific proteolytic activities differentially influence final nutritional and sensory attributes.

Regarding the fungal community, inoculation with functional microbial consortia directly shaped the dominant genera and their competitive dynamics within the fermentation system. The EG and UG groups reached similar successional endpoints, both dominated by Kazachstania as the absolute predominant genus. This indicates that in the absence of exogenous yeast competition, indigenous yeast can dominate the fermentation process, corroborating the observations of Liu, Yang, Liu, et al. (2025). Kazachstania has been shown to significantly enhance acetate esters, particularly phenylethyl acetate and isoamyl acetate, in red wines, thereby enriching aromatic complexity when used as a wine fermentation starter (Lin et al., 2022). The SG and MG groups were dominated by Saccharomyces and Wickerhamomyces, respectively. Co-fermentation with Saccharomyces and Wickerhamomyces has been reported to enhance the sensory quality of wine by increasing the levels of total phenols, tannins, and various aroma compounds, contributing to a more complex flavor profile and higher sensory scores (Wang, Yan, et al., 2023). Similarly, co-fermentation of Sichuan Paocai with Wickerhamomyces and Lactiplantibacillus increased levels of characteristic aroma compounds, including ethyl acetate, phenylethyl alcohol, and isobutanol, which impart fruity and floral notes and significantly improve the product's flavor profile (Zhang et al., 2023). Notably, the relative abundance of Wickerhamomyces was higher in the MG group than in the SG group, suggesting that the composition of the synthetic consortia differentially influenced yeast community structure. This observation was further supported by significant differences in the abundance of genera such as Debaryomyces and Saccharomycopsis across treatment groups, implying that the inoculated consortia exerted complex cascading effects, either inhibitory or promotive, on other microorganisms within the fermentation system. In the EG and MG groups, Eurotium played a central role primarily during the early and middle fermentation stages, with its activity and functional expression most pronounced during these phases. Zhou et al. (2019) demonstrated that fermenting Angelica dahurica with Eurotium cristatum substantially improved its antioxidant properties. Mechanistically, E. cristatum secretes tannase, polyphenol oxidase, cellulase, and α-amylase, driving phenolic biotransformation into diverse derivatives (Fan et al., 2025), and generates fungal-flowery aroma compounds via the phenylalanine pathway (Shen et al., 2023). Safety assessments confirm that E. cristatum does not produce aflatoxin B1, patulin, deoxynivalenol, or zearalenone (Efimenko et al., 2021).

Regarding the bacterial community, all groups ultimately converged on a succession pattern dominated by LAB, with Lentilactobacillus as the core genus. Lentilactobacillus exhibited robust growth and metabolic activity in the fermentation system, rapidly producing acids and reducing nitrite levels, thereby enhancing product safety (Liu, Zhang, Wu, et al., 2025). Its metabolic profile showed significant positive correlations with the synthesis of predominant flavor compounds, including ethyl hexanoate, acetaldehyde, and geraniol, further indicating its important role in regulating the final flavor quality (Duran et al., 2022). The peak abundance of Lactiplantibacillus occurred during the early fermentation stage in the UG and EG groups, but was delayed until the mid-fermentation stage in the SG and MG groups, indicating that the fungal inoculants in the treatment groups significantly influenced the timing and dynamics of this lactic acid bacterial genus. Weissella, as a typical transient genus, was most abundant during the early fermentation stage, a succession pattern consistent with the findings of Zhang, Hu, et al. (2025). Furthermore, opportunistic pathogens were rapidly eliminated in all groups upon initiation of fermentation, demonstrating that the tomato sour soup fermentation system inherently possesses reliable microbial safety.

Microbial co-occurrence network analysis revealed distinct community assembly strategies among treatment groups from an ecological interaction perspective. The UG group preserved the native microbial equilibrium, reflecting long-term coexistence. In contrast, functional inoculation triggered profound ecological restructuring and functional enhancement. The SG group exhibited a distinctive yeast-centric hub network, with Saccharomyces and Wickerhamomyces serving as central nodes connected to multiple LAB. In the EG and MG groups, the inoculated Eurotium drove the community toward a highly modular structure, with the MG group exhibiting stronger functional specialization in microbial interactions. The strong positive correlation between Aspergillus and Secundilactobacillus may mediate nitrogen decomposition and organic acid synthesis. Eurotium cristatum has been reported not only to promote the growth of LAB but also to effectively modulate the levels of organic acids, free amino acids, and catechins (Gao et al., 2025). Synergistic metabolism between Saccharomyces and Lactiplantibacillus enables coordinated regulation of organic acids and enhanced ester production (Hu et al., 2023). Multiple positive correlations between Levilactobacillus and the two yeast genera, Debaryomyces and Saccharomycopsis, further suggest the presence of diverse mutualistic relationships within the complex ecosystem of the MG group, which may contribute to enhancing bioactive compound content and improving antioxidant activity (Vurro et al., 2025).

Dynamic changes in core metabolites revealed temporal succession patterns, microbiota-specific regulation, and shared metabolic responses during tomato sour soup fermentation. The significant upregulation of metabolites such as trigonelline, citric acid, and naringenin during the early fermentation stage reflects the capacity of microorganisms to drive metabolic processes by promoting substrate degradation and transforming functional components. Citric acid accumulation directly contributes to the acidity of tomato sour soup (Zheng, Jiang, et al., 2024), while naringenin, a pivotal flavonoid intermediate, exhibits potent antioxidant and anti-inflammatory effects (Sakat et al., 2022). In the late fermentation stage, succinic acid, GABA, and other metabolites reached their peak levels, coinciding with the dominance of abundant organic acids and metabolic intermediates in the metabolic profile. GABA can be catabolized by GABA transaminase to succinic semialdehyde, which is subsequently converted to succinic acid and enters the tricarboxylic acid cycle (Sun et al., 2022). Succinic acid not only increases the acidity of the fermentation broth but also contributes to ester formation, thereby enhancing the flavor profile (Chidi et al., 2018). Regarding amino acid metabolism, L-isoleucine was universally upregulated in the three inoculated groups during the late fermentation stage, contributing to enhanced umami and sweetness (Liu, Zhang, Su, et al., 2025). L-Phenylalanine was consistently upregulated across all four stages in the EG group, facilitating its reversible transamination to phenylpyruvate, which is then sequentially converted by endogenous decarboxylase and reductase enzymes into the rose-scented compound, 2-phenylethanol (Hirasawa et al., 2025). Trinh et al. (2010) demonstrated that the co-fermentation of L-isoleucine and L-phenylalanine effectively promotes the production of aromatic ester-alcohol compounds, including active amyl alcohol and 2-phenylethyl esters.

Analysis of DAMs revealed that distinct functional microbial communities were associated with different metabolic profiles, particularly in the dynamic transformations of organic acids and derivatives, as well as phenylpropanoids and polyketides. This pattern was particularly This pattern was especially pronounced for organic acids and derivatives. The sustained upregulation of L-glutathione and its derivatives in the UG and SG groups, coupled with the downregulation of oxiglutatione across all groups during the mid-to-late fermentation stages, indicates enhanced oxidative stress throughout the fermentation process. L-Glutathione is also involved in the biosynthesis of various non-volatile flavor precursors (Binati et al., 2021). Notably, the bitter peptide Gly-Phe was significantly downregulated during the late fermentation stage (Maehashi et al., 2008). In contrast, the kokumi and umami peptide Ile-Leu was markedly upregulated in the inoculated fermentation groups (Wu et al., 2025). This differential accumulation reflects the impact of functional microbial consortia on protein degradation pathways, thereby modulating the flavor characteristics of tomato sour soup.

The differential accumulation patterns of flavonoids provided direct evidence of quality modulation by functional microbial communities. The sustained upregulation of chrysoeriol, kaempferol, and luteolin in the EG and MG groups, in contrast to their downregulation in the UG and SG groups, aligned with the pathway enrichment analysis, which revealed a specific activation of flavonoid biosynthesis in the EG and MG groups. This concordance confirms the functional role of Eurotium sp. in promoting flavonoid transformation and accumulation in the tomato-based substrate, consistent with previous studies (Guo et al., 2022). To substantiate the contribution of microbial inoculation to antioxidant enhancement, we analyzed correlations between flavonoid metabolites and antioxidant indices. In the inoculated fermentation groups, the precursors chalconaringenin and rutin were generally upregulated early, whereas their downstream products, including phloretin, quercetin, and ethyl caffeate, accumulated in the middle and late stages, indicating that functional microbial communities systematically promote polyphenol biotransformation (Liu et al., 2021). Notably, these precursors showed negative correlations with antioxidant indices, whereas their transformation products showed positive correlations. This contrasting pattern is consistent with active precursor-to-product conversion during fermentation, in which the continuous transformation of precursors into more potent antioxidant derivatives drives the overall increase in antioxidant activity. This pattern was further supported by the fact that rutin can be converted into its aglycone quercetin via enzymatic hydrolysis (Cheng et al., 2025), and chalconaringenin can be reduced to phloretin by chalcone double-bond reductase (Yauk et al., 2024), both of which are known to possess stronger antioxidant activity than their precursors (Nakhate et al., 2022; Song et al., 2020).

During the fermentation of tomato sour soup, microbial co-metabolism exhibited a significant synergistic trend, accompanied by the accumulation of various bioactive substances. Flavor-contributing compounds such as Gly-Phe and Tyr-Pro showed positive correlations with genera including Saccharomyces and Weissella, indicating that the development of the complex flavor profile relies on the combined action of fungi and bacteria (Zhang, Xiong, et al., 2025). Positive correlations of Pichia with O-Acetyl-l-serine and of Secundilactobacillus with L-glutathione and oxiglutatione suggest a potential role for the microbiota in modulating redox balance in the acidic fermentation environment (Hansen, 2018). Regarding polyphenol metabolism, Aspergillus showed strong positive correlations with multiple aglycones, including luteolin, chrysoeriol, kaempferol, and chalconaringenin. Similarly, fermentation by Eurotium cristatum has been reported to modulate total phenolic and flavonoid contents in dark tea infusion, suggesting the involvement of Eurotium in polyphenol metabolism (Huang et al., 2023). Lactiplantibacillus, Lentilactobacillus, and Kazachstania were positively correlated with quercetin, phloretin, and ethyl caffeate, but negatively correlated with rutin, implying their involvement in polyphenol conversion. This aligns with Ankutse et al. (2025), who observed similar changes in rutin and quercetin during Lactiplantibacillus fermentation of pineapple juice. Together, these findings suggest a sequential biotransformation process in which Eurotium initiates the deglycosylation of flavonoid glycosides, while LAB and yeasts further modify the released aglycones, collectively contributing to the antioxidant activity of tomato sour soup (Chen et al., 2022; Gu et al., 2019). Nevertheless, we acknowledge that the precise role of each microbial group in this bioconversion process requires experimental validation using our specific strains. Overall, the fermentation of tomato sour soup is not merely an acid-producing process driven solely by LAB, but rather a complex microecosystem involving coordinated metabolic interactions between fungi and bacteria.

This study linked specific microorganisms to key bioactive compound production, offering insights into microbially driven metabolism during tomato sour soup fermentation. Nevertheless, the mixed-strain approach does not allow us to separate the metabolic contributions of individual strains, an issue that single-strain controls could address in future work. The proposed Eurotium-mediated glycoside hydrolysis and biotransformation pathways require direct validation via enzymatic or genetic experiments. Similarly, the increases in antioxidant phenolics and flavonoids associated with Eurotium spp. are promising but limited to in vitro measurements, and in vivo studies are needed to assess their physiological relevance. The correlation between microbial succession and nitrite reduction also needs further mechanistic investigation. To gain a fuller picture of flavor, future studies will combine GC–MS profiling, electronic sensory analysis, and quantitative descriptive testing to validate the flavor-related predictions derived from the current metabolomic data. Lastly, optimization of the inoculation ratio and other parameters remains important for maximizing the functional performance of this fermentation system.

5. Conclusions

Inoculation with functional microbial consortia comprising yeasts and Eurotium spp. significantly enhanced acidification efficiency, nitrite degradation, and amino acid nitrogen conversion in the tomato sour soup fermentation system. Meanwhile, it reshaped microbial community structure and metabolic functions, promoting the synthesis of bioactive components. Therefore, targeted inoculation with functional microbial consortia represents an effective strategy for optimizing fermentation processes and improving product quality and safety. Future work will further evaluate the physiological effects of fortified tomato sour soup using animal models, including its impact on gut microbiota, oxidative stress, and lipid metabolism. Overall, this study provides a theoretical foundation and data support for constructing functionally defined microbial consortia and improving the nutritional quality of tomato sour soup through microbial community modulation.

Declaration of generative AI use

The authors did not use generative AI or AI-assisted technologies in the preparation, writing, or editing of this manuscript. All intellectual contributions, data interpretation, and final approval were performed solely by the authors.

CRediT authorship contribution statement

Shuting Zhong: Writing – original draft, Visualization, Validation, Methodology, Formal analysis, Data curation. Chuanbo Zhang: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Anqin Zhu: Software, Methodology, Formal analysis. Tingyan Xu: Writing – original draft, Data curation. Liangliang Xiao: Validation, Methodology. Yunsheng Wang: Software, Formal analysis. Rui Wang: Visualization, Data curation. Yuping Deng: Resources, Investigation.

Submission declaration

The authors confirm that this work is original and is not under consideration for publication elsewhere. The publication of this article has been approved by all co-authors. If accepted, it will not be published elsewhere in the same form, in English or in any other language, including electronically, without the written consent of the copyright holder.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the science and technology plan support project of Qiandongnan Prefecture, Guizhou Province, China [No. (2024)0003]; the science and technology achievement transformation project of Guizhou Province, China [No. (2023)020] and the National Natural Science Foundation of China (Nos. 31860438 and 81760688).

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2026.104484.

Appendix A. Supplementary data

Supplementary material

mmc1.docx (47.8MB, docx)

Data availability

Data will be made available on request.

References

  1. Ankutse P., Musumba F.A., Bilal M., Han Y., Duan Y., Li F.…Adase E. Modulation of phytochemical profiles and bioactive constituents in Ananas comosus via Lactiplantibacillus-mediated fermentation. Food Bioscience. 2025;66 doi: 10.1016/j.fbio.2025.106257. [DOI] [Google Scholar]
  2. Binati R.L., Lemos Junior W.J.F., Torriani S. Contribution of non-Saccharomyces yeasts to increase glutathione concentration in wine. Australian Journal of Grape and Wine Research. 2021;27(3):290–294. doi: 10.1111/ajgw.12473. [DOI] [Google Scholar]
  3. Cai T., Hai N., Guo P., Feng Z., Zhang Y., Wang J.…Ding L. Characteristics of umami taste of soy sauce using electronic tongue, amino acid analyzer, and MALDI−TOF MS. Foods. 2024;13(14) doi: 10.3390/foods13142242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chen Q., Zhang M., Chen M., Li M., Zhang H., Song P., An T., Yue P., Gao X. Influence of Eurotium cristatum and Aspergillus niger individual and collaborative inoculation on volatile profile in liquid-state fermentation of instant dark teas. Food Chemistry. 2021;350 doi: 10.1016/j.foodchem.2021.129234. [DOI] [PubMed] [Google Scholar]
  5. Chen Y., Wang Y., Zhu A., Zhang L., Zhang X., Zhang J., Zhang C. UPLC-TOF/MS-based metabolomics reveals the chemical changes and in vitro biological effects in fermentation of white ginseng by four probiotics. Frontiers in Microbiology. 2022;13 doi: 10.3389/fmicb.2022.1022200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cheng S., Zheng Y., Li H., Jiang Y., Zhang Y. Metabolic engineering of blueberry juice by Lactiplantibacillus plantarum JM065: Fermentation optimization. Food Chemistry. 2025;496 doi: 10.1016/j.foodchem.2025.146669. [DOI] [PubMed] [Google Scholar]
  7. Chidi B.S., Bauer F.F., Rossouw D. Organic acid metabolism and the impact of fermentation practices on wine acidity: A review. South African Journal of Enology and Viticulture. 2018;39(2) doi: 10.21548/39-2-3172. [DOI] [Google Scholar]
  8. Duran F.E., Özdemir N., Güneşer O., Kök-Taş T. Prominent strains of kefir grains in the formation of volatile compound profile in milk medium; the role of lactobacillus kefiranofaciens subsp. kefiranofaciens, Lentilactobacillus kefiri and Lentilactobacillus parakefiri. European Food Research and Technology. 2022;248(4):975–989. doi: 10.1007/s00217-021-03936-2. [DOI] [Google Scholar]
  9. Efimenko T.A., Shanenko E.F., Mukhamedzhanova T.G., Efremenkova O.V., Nikolayev Y.A., Bilanenko E.N.…Tsaltas D. Eurotium cristatum postfermentation of fireweed and apple tree leaf herbal teas. International Journal of Food Science. 2021;2021:1–14. doi: 10.1155/2021/6691428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fan Z., Qing Q., Yin J., He C., Hu Y., Chen Y., Ren Y., Zhu M., Liu Z., Peng X., Xiao Y. Metabolites of epigallocatechin gallate and changes in antioxidant activity through biotransformation with Eurotium cristatum during liquid-state fermentation. Food Chemistry: X. 2025;29 doi: 10.1016/j.fochx.2025.102618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Feixia D., Ya L., Dafei L., Dingjiang Z., Guiping H., Zeliang W., Lirong J. Kaili red sour soup: Correlations in composition/microbial metabolism and flavor profile during post-fermentation. Food Chemistry. 2024;435 doi: 10.1016/j.foodchem.2023.137602. [DOI] [PubMed] [Google Scholar]
  12. Gao Y., Zhuo R., Luo H., Ho C.-T., Chen Y., Zhou H.…Xiao Y. Characterizing and decoding the effects of co-culture fermentation on dark tea infusion flavor development: Elucidating microbial succession and metabolic dynamics mediated by Eurotium cristatum and water kefir. Food Chemistry: X. 2025;30 doi: 10.1016/j.fochx.2025.102876. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gu Q., Duan G., Yu X. Bioconversion of flavonoid glycosides from Hippophae rhamnoides leaves into flavonoid aglycones by Eurotium amstelodami. Microorganisms. 2019;7(5) doi: 10.3390/microorganisms7050122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Guo X., Chen F., Liu J., Shao Y., Wang X., Zhou Y. Genome mining and analysis of PKS genes in Eurotium cristatum E1 isolated from fuzhuan brick tea. Journal of Fungi. 2022;8(2) doi: 10.3390/jof8020193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hansen E.B. Redox reactions in food fermentations. Current Opinion in Food Science. 2018;19:98–103. doi: 10.1016/j.cofs.2018.03.004. [DOI] [Google Scholar]
  16. Hirasawa T., Satoh Y., Koma D. Production of aromatic amino acids and their derivatives by Escherichia coli and Corynebacterium glutamicum. World Journal of Microbiology and Biotechnology. 2025;41(2) doi: 10.1007/s11274-025-04264-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hu L., Chen X., Lin R., Xu T., Xiong D., Li L., Zhao Z. Quality improvement in apple ciders during simultaneous co-fermentation through triple mixed-cultures of Saccharomyces cerevisiae, Pichia kudriavzevii, and Lactiplantibacillus plantarum. Foods. 2023;12(3) doi: 10.3390/foods12030655. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Huang Y., Chen R., Chen Y., Ho C.-T., Hou A., Zhang X., Zhu M., Zhang C., Wang Y., Liu Z., Xiao Y. Dynamics changes in volatile profile, non-volatile metabolites and antioxidant activities of dark tea infusion during submerged fermentation with Eurotium cristatum. Food Bioscience. 2023;55 doi: 10.1016/j.fbio.2023.102966. [DOI] [Google Scholar]
  19. Huang Y.-Y., Huang F., Huang J.-Y., Liu Y., Li M.-N., Yao Q.-B., Zeng X.-A., Chen B.-R. Repercussions of Lactiplantibacillus plantarum HYY-DB9 and Levilactobacillus brevis GIM 1.773 on the degradation of nitrite, volatile compounds, and sensory assessment of traditional Chinese paocai. International Journal of Food Science and Technology. 2024;59(9):6522–6532. doi: 10.1111/ijfs.17400. [DOI] [Google Scholar]
  20. Kang D.-M., Kwon J.-M., Jeong W.-J., Jung Y.J., Kang K.K., Ahn M.-J. Antioxidant constituents and activities of the pulp with skin of Korean tomato cultivars. Molecules. 2022;27(24) doi: 10.3390/molecules27248741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li H., Zhao K., Zhang J., Chen L., Huang Q., Ma X., Ge S., Tang H. PacBio sequencing combined GC–IMS approach to investigate the flavor formation of Jinhua ham prepared using Taihu black pigs. Lwt. 2024;213 doi: 10.1016/j.lwt.2024.117024. [DOI] [Google Scholar]
  22. Lin M.M.-H., Boss P.K., Walker M.E., Sumby K.M., Jiranek V. Influence of Kazachstania spp. on the chemical and sensory profile of red wines. International Journal of Food Microbiology. 2022;362 doi: 10.1016/j.ijfoodmicro.2021.109496. [DOI] [PubMed] [Google Scholar]
  23. Liu H., Zhang M., Su M., Zeng W., Wang S., Du J., Zhou H., Yang X., Zhang X., Li X., Ye Z. Multidimensional analysis of the flavor characteristics of yellow peach at different ripening stages: Chemical composition profiling and sensory evaluation. Food Chemistry. 2025;471 doi: 10.1016/j.foodchem.2025.142772. [DOI] [PubMed] [Google Scholar]
  24. Liu L., Zhang C., Zhang H., Qu G., Li C., Liu L. Biotransformation of polyphenols in apple pomace fermented by β-glucosidase-producing lactobacillus rhamnosus L08. Foods. 2021;10(6) doi: 10.3390/foods10061343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu N., Hu Y., Qin L., Bao A., Qin W., Miao S. Flavor and quality characteristics of Guizhou red sour soup prepared by different artificially fortified fermentation methods. Lwt. 2023;186 doi: 10.1016/j.lwt.2023.115247. [DOI] [Google Scholar]
  26. Liu N., Li X., Hu Y., Qin L., Bao A., Qin W., Miao S. Effects of Lentilactobacillus buchneri and Kazachstania bulderi on the quality and flavor of Guizhou fermented red sour soup. Foods. 2023;12(20) doi: 10.3390/foods12203753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Liu N., Yang Q., Liu Z., Hou Y., Qin L., Bao A., Qin W., Miao S. Regulation of the quality, flavour, and microbial community of industrially fermented split red sour soup by Lentilactobacillus buchneri and Pediococcus ethanolidurans. Food Research International. 2025;203 doi: 10.1016/j.foodres.2025.115850. [DOI] [PubMed] [Google Scholar]
  28. Liu N., Zhang B., Wu M., Hou Y., Qin L., Bao A., Qin W., Miao S. Effects of different fermentation methods on the microflora composition and flavor quality of mixed red sour soup. Lwt. 2025;222 doi: 10.1016/j.lwt.2025.117653. [DOI] [Google Scholar]
  29. Luo Y., Liu Y., Ren T., Wang B., Peng Y., Zeng S., Su Y. Sichuan paocai fermented by mixed-starter culture of lactic acid bacteria. Food Science & Nutrition. 2020;8(10):5402–5409. doi: 10.1002/fsn3.1833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Maehashi K., Matano M., Wang H., Vo L.A., Yamamoto Y., Huang L. Bitter peptides activate hTAS2Rs, the human bitter receptors. Biochemical and Biophysical Research Communications. 2008;365(4):851–855. doi: 10.1016/j.bbrc.2007.11.070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nakhate K.T., Badwaik H., Choudhary R., Sakure K., Agrawal Y.O., Sharma C.…Goyal S.N. Therapeutic potential and pharmaceutical development of a multitargeted flavonoid Phloretin. Nutrients. 2022;14(17) doi: 10.3390/nu14173638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ren X., Xiong H., Liao X., Abduallah A.M.A., Su J., Wang X.…Zhang X. The physiochemical characteristics and metabolic components of Paeonia ostii petal tea produced through Eurotium cristatum-mediated liquid fermentation. Food Chemistry: X. 2026;33 doi: 10.1016/j.fochx.2025.103436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Ryu J.-A., Kim E., Yang S.-M., Lee S., Yoon S.-R., Jang K.-S., Kim H.-Y. High-throughput sequencing of the microbial community associated with the physicochemical properties of meju (dried fermented soybean) and doenjang (traditional Korean fermented soybean paste) Lwt. 2021;146 doi: 10.1016/j.lwt.2021.111473. [DOI] [Google Scholar]
  34. Sakat M.S., Kılıç K., Sahin A., Ozmen H.K., Yıldırım S., Kiziltunc A.…Saglam Y.S. The protective efficacy of quercetin and Naringenin against radiation-related submandibular gland injury in female rats: A histopathological, immunohistochemical, and biochemical study. Archives of Oral Biology. 2022;142 doi: 10.1016/j.archoralbio.2022.105510. [DOI] [PubMed] [Google Scholar]
  35. Segoviano-León J.P., Valdez-Morales M., Leyva-Morales J.B., Soto-Montoya D.G., Izaguirre-Díaz de León F., Perea-Domínguez X.P. Metabolites profile, DPPH and ABTS scavenging and myoglobin protection ratio of aqueous infusion and methanolic extracts of Mexican Lippia alba. Food Chemistry. 2025;467 doi: 10.1016/j.foodchem.2024.142259. [DOI] [PubMed] [Google Scholar]
  36. Shen S., Zhang J., Sun H., Zu Z., Fu J., Fan R.…Gao X. Sensomics-assisted characterization of fungal-flowery aroma components in fermented tea using Eurotium cristatum. Journal of Agricultural and Food Chemistry. 2023;71(48):18963–18972. doi: 10.1021/acs.jafc.3c05273. [DOI] [PubMed] [Google Scholar]
  37. Shi S.-s., Hu T. Effects of Eurotium cristatum on soybean (Glycine max L.) polyphenols and the inhibitory ability of soybean polyphenols on acetylcholinesterase under different conditions. Food Chemistry: X. 2024;23 doi: 10.1016/j.fochx.2024.101526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Song X., Wang Y., Gao L. Mechanism of antioxidant properties of quercetin and quercetin-DNA complex. Journal of Molecular Modeling. 2020;26(6) doi: 10.1007/s00894-020-04356-x. [DOI] [PubMed] [Google Scholar]
  39. Song X., Zhou Q., Li X., Huang Q., Zhou Y. Spontaneous fermentation tunes purine metabolism of tomato by modifying bacterical diversity: Based on metabonomics and microbiomics. Lwt. 2024;191 doi: 10.1016/j.lwt.2023.115486. [DOI] [Google Scholar]
  40. Sun X., Wang J., Li C., Zheng M., Zhang Q., Xiang W., Tang J. The use of γ-aminobutyric acid-producing Saccharomyces cerevisiae SC125 for functional fermented beverage production from apple juice. Foods. 2022;11(9) doi: 10.3390/foods11091202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Trinh T.T.T., Woon W.Y.B., Yu P.C., Liu S.-Q. Effect of L-isoleucine and L-phenylalanine addition on aroma compound formation during longan juice fermentation by a co-culture of Saccharomyces cerevisiae and Williopsis saturnus. South African Journal of Enology and Viticulture. 2010;31 doi: 10.21548/31-2-1408. [DOI] [Google Scholar]
  42. Vurro F., Anelli P., Zocchi D.M., De Bellis P., Pieroni A., Pasqualone A. Bessarabian wild hop sourdough: Microbial characterization and effect on the physicochemical properties and flavor of the bread. International Journal of Gastronomy and Food Science. 2025;42 doi: 10.1016/j.ijgfs.2025.101377. [DOI] [Google Scholar]
  43. Wang J., Yan J., Zhang W., Zhang Y., Dong Z., Luo H.…Su J. Comparison of potential Wickerhamomyces anomalus to improve the quality of cabernet sauvignon wines by mixed fermentation with Saccharomyces cerevisiae. Lwt. 2023;173 doi: 10.1016/j.lwt.2022.114285. [DOI] [Google Scholar]
  44. Wang Y., Chen Y., Xin J., Chen X., Xu T., He J., Pan Z., Zhang C. Metabolomic profiles of the liquid state fermentation in co-culture of Eurotium amstelodami and Bacillus licheniformis. Frontiers in Microbiology. 2023;14 doi: 10.3389/fmicb.2023.1080743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wu J., Lin X., Cui C., Li J. Comparative analysis of kokumi effects in Ile-Leu, Ile-Ile, Leu-Leu, and Leu-Ile: Sensory, simulation, and structural insights. Food Chemistry. 2025;474 doi: 10.1016/j.foodchem.2025.143004. [DOI] [PubMed] [Google Scholar]
  46. Xiong K., Han F., Wang Z., Du M., Chen Y., Tang Y., Wang Z. Screening of dominant strains in red sour soup from Miao nationality and the optimization of inoculating fermentation conditions. Food Science & Nutrition. 2020;9(1):261–271. doi: 10.1002/fsn3.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Xu T., Zhu A., Zhong S., Wang Y., Wang C., Zhang J., Chen Y., Wang R., Zhang C. Physicochemical properties and bioturbation analysis of high-temperature Daqu by functional microflora. Frontiers in Microbiology. 2025;16 doi: 10.3389/fmicb.2025.1601675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yang H., Xie J., Wang N., Zhou Q., Lu Y., Qu Z., Wang H. Effects of Miao sour soup on hyperlipidemia in high-fat diet-induced obese rats via the AMPK signaling pathway. Food Science & Nutrition. 2021;9(8):4266–4277. doi: 10.1002/fsn3.2394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Yauk Y.-K., Dare A.P., Cooney J.M., Wang Y., Hamiaux C., McGhie T.K.…Atkinson R.G. Naringenin chalcone carbon double-bond reductases mediate dihydrochalcone biosynthesis in apple leaves. Plant Physiology. 2024;196(4):2768–2783. doi: 10.1093/plphys/kiae515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Zhang J., He Y., Yin L., Hu R., Yang J., Zhou J., Cheng T., Liu H., Zhao X. Isolation of aroma-producing Wickerhamomyces anomalus yeast and analysis of its typical flavoring metabolites. Foods. 2023;12(15) doi: 10.3390/foods12152934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhang P., Hu M., Meng W., Wang J., Yuan Z., Wang F., Li S. Effect of Lactiplantibacillus plantarum MX14 and Pichia membranifaciens PY33 co-fermentation on quality characteristics of red sour soup. Food Chemistry. 2025;492 doi: 10.1016/j.foodchem.2025.145473. [DOI] [PubMed] [Google Scholar]
  52. Zhang X., Xiong K., Ye S., Du M., Wang Z. Effects of different raw materials on bacterial community and flavor compounds in fermented red sour soup. Food Bioscience. 2025;65 doi: 10.1016/j.fbio.2025.106034. [DOI] [Google Scholar]
  53. Zheng H., Jiang J., Huang C., Wang X., Hu P. Effect of sugar content on characteristic flavour formation of tomato sour soup fermented by Lacticaseibacillus casei H1 based on non-targeted metabolomics analysis. Food Chemistry: X. 2024;21 doi: 10.1016/j.fochx.2024.101116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zheng T., Deng Z., Tian M., Tang Q., Hu Z., Wang G., Zeng H. UPLC-MS/MS and HS-SPME-GC–MS reveal the flavor profiles of two geographical indications woody vegetables: Staphylea bumalda and Staphylea holocarpa. Food Chemistry: X. 2024;24 doi: 10.1016/j.fochx.2024.101811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zhou S.-D., Xu X., Lin Y.-F., Xia H.-Y., Huang L., Dong M.-S. On-line screening and identification of free radical scavenging compounds in Angelica dahurica fermented with Eurotium cristatum using an HPLC-PDA-triple-TOF-MS/MS-ABTS system. Food Chemistry. 2019;272:670–678. doi: 10.1016/j.foodchem.2018.07.173. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material

mmc1.docx (47.8MB, docx)

Data Availability Statement

Data will be made available on request.

Articles from Food Chemistry: X are provided here courtesy of Elsevier

RESOURCES