Exploring the Microbiome of Fermented Soy Products: Implications for Gut Health in China

doi:10.21203/rs.3.rs-4982604/v1

Download PDF

Research Article

Exploring the Microbiome of Fermented Soy Products: Implications for Gut Health in China

https://doi.org/10.21203/rs.3.rs-4982604/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Fermented foods have a long history in China, and they continue to be widely consumed today. Fermented foods have recently been reported as a pivotal approach to restoring gut microbial diversity and are recommended by the International Scientific Association for Probiotics and Prebiotics for inclusion in dietary guidelines. However, there are potential safety concerns associated with fermented foods, such as the transfer of antibiotic resistance genes to the human gut. This underscores the need for a deeper understanding of the microbial communities in fermented foods and additional data to facilitate health risk assessments.

Results: In this study, we employed shotgun metagenomic analysis to investigate the microbiota of three commonly consumed fermented soy products in China and compared them with the gut microbiota of the Chinese population. Our findings revealed significant differences in both the microbial composition and functions among these three fermented soy products. Intriguingly, network analysis revealed an antagonistic interaction between beneficial species Bacillales and Lactobacillales, and potentially harmful species Enterobacterales. In examining the Chinese gut microbiota, we identified a high prevalence of potentially harmful bacteria from the Enterobacterales order, which were also found in significant amounts in fermented foods. Using genome-level and strain-level analyses, we hypothesize that fermented foods may serve as a source of harmful bacteria, such as Klebsiella pneumoniae and Klebsiella quasipneumoniae, for gut microbiota. Horizontal gene transfer analysis highlighted the potential transfer of numerous antibiotic resistance genes from fermented foods microbes to those in the human gut microbiome.

Conclusions: While there is substantial evidence supporting the potential health benefits of consuming fermented foods, our research highlights important safety concerns. Notably, consuming fermented foods could increase exposure to pathogenic microorganisms and increase the risk of antibiotic resistance gene transmission. This accentuates the need for enhanced microbial monitoring and quality control measures for fermented foods.

Fermented Soy Products

Chinese Population

Gut Microbiota

Antibiotic Resistance Genes

China’s rich tradition of fermented foods remains a fundamental part of the Chinese diet today. Recently, fermented foods have garnered interest amongst consumers due to their health benefits. This is also due to the growing body of research highlighting the close relationship between gut microbiota and human health.[1–3] Fermented foods offer a way to replenish the “missing microbes” resulting from industrialized or Westernized diets.[4] Accumulating evidence showed that fermented foods can exert direct or indirect influence on gut microbiota composition and activity, leading to noticeable impacts on human health.[5, 6] In 2019, the International Scientific Association for Probiotics and Prebiotics (ISAPP) suggested that dietary recommendations include fermented foods, citing their significant content of live or potentially health-promoting microorganisms.[7] Microorganisms in fermented foods can survive gastric transit and reach the colon.[8, 9] A notable study comparing fermented foods with the human gut microbiota found a significant presence of food-associated lactic acid bacteria in the fecal metagenome.[10] These microorganisms, upon entering the gastrointestinal tract, can establish short-term colonies, synthesize bioactive compounds that inhibit enteric pathogens,[7] and mediate epithelial regulatory effects.[11] Long-term consumption of fermented foods can reinforce these dynamic interactions, further highlighting their relevance to human health.

Despite the documented health benefits associated with microorganisms in fermented foods, significant safety concerns remain. Fermented foods can harbor foodborne pathogens, some of which may be introduced during the production process.[12, 13] Furthermore, fermented foods have historically served as a prominent pathway for the transmission of antibiotic resistance genes to consumers.[14, 15] Isolated from fermented foods, bacteria with mobile antibiotic resistance genes can transfer these genes to human commensal and pathogenic microbiota through horizontal gene transfer.[16] Research indicated that dietary interventions involving fermented foods can significantly increase antibiotic resistance within the gut microbiota of most subjects.[14] The public health risks of using fermented food as interventions are likely underestimated in vulnerable populations, such as those with gastrointestinal symptoms or compromised immune function. These considerations highlight the need for a comprehensive assessment of both the benefits and potential risks associated with the consumption of fermented foods as well as call for increased vigilance in monitoring and ensuring compliance with food safety standards.

In China, traditional fermented foods include a diverse range of products, with a primary focus on fermented soy products such as fermented tofu (fu ru, FR), soybean paste (dou jiang, DJ), and fermented black beans (dou chi, DC), all of which are central to Chinese cuisine. Fermentation of soybean produces numerous functional and bioactive compounds, offering health benefits such as antioxidative, anti-inflammatory, as well as blood sugar- and lipid-lowering effects.[17–19] Most studies that investigated the microbiota associated with these fermented foods focused on the fermentation microorganisms. However, there is limited research on their potential impact on gut health. A comprehensive review of fermented foods highlighted the safety concerns associated with these products in China.[20] However, the majority of studies on these fermented foods still rely on 16S rRNA gene amplicon sequencing, which typically provides insights only at the genus level and offers limited information on the functional potential and gene expression of bacteria, archaea, or eukaryotic taxa. On the other hand, shotgun metagenomics provides information at the species or strain level and allows for the assembly of microbial genomes, thereby offering a deeper understanding of microbial composition and functional potential in food matrices. This approach may help illuminate the complex interplay between fermented soy products and human gut health, paving the way for more targeted and informative research.

In this study, we investigated three commonly consumed fermented soy products in China using shotgun metagenomics to explore their microbial and functional compositions. By comparing the microbiota and functionality of these three fermented foods, we aim to better understand the microbial characteristics of each product. Furthermore, by comparing these food-derived microorganisms with the gut microbiota of the Chinese population, our study has two main objectives: first, to explore the relationship between the microorganisms in these fermented foods and the gut microbiota of the Chinese population, and to investigate the potential transfer of antibiotic resistance genes between the fermented food microbiota and the human gut. Our research provides in-depth insights into the microbiota of fermented soy products, highlighting potential public health risks associated with these fermented foods.

Microbial composition in three fermented soy products

A total of 93 fermented soy products (DC, DJ and FR) were purchased from online and offline supermarkets across 20 provinces (Table S1). All samples were subjected to shotgun metagenomic sequencing, generating a total of 203.4 Gb data, with an average of 12,691,369 reads per sample.

Compared to DC and FR, DJ samples exhibited a notably higher abundance of eukaryotic organisms (Fig. 1A). To minimize false positives in Kraken2 classification, only species with assigned reads greater than 500 in at least 10 samples per group were retained for further analysis. Notably, DJ samples contained a diverse range of 10 distinct eukaryotic species, whereas DC and FR samples contained only 3 and 1 identified species, respectively (Figure S1A). All three fermented foods shared a single eukaryotic species, Aspergillus oryzae, which was highly prevalent in each (Figure S1B).

In regard to bacterial composition, the dominant phyla found in DC, FR, and DJ samples were Firmicutes (80.9%, 48.8%, 46.2%, respectively) and Proteobacteira (9%, 41.1%, 25.3%, respectively) (Fig. 1A). In DC, the dominant order was Bacillales (66%). On the other hand, Lactobacillales was the most dominant order in FR and DJ (39.9% and 30.1%, respectively), and was also the second most abundant order in DC (14.4%) (Fig. 1A). Notably, Lactobacillales encompassed a diverse group of lactic acid bacteria that play crucial roles in food production, fermentation, and development of probiotic products. As highlighted in Figure S1C, the primary genera in DC were Bacillus (50.3%), Caldibacillus (6.54%), Staphylococcus (5.8%), Tetragenococcus (5.31%), and Weissella (2.46%). In DJ samples, the dominant genera were Staphylococcus (9.38%), Pantoea (7.75%), Tetragenococcus (6.60%), Weissella (5.49%), and Leuconostoc (3.56%). Primary genera of FR samples included Tetragenococcus (15.4%), Enterobacter (9.19%), Lactococcus (8.04%), Pseudomonas (7.22%), and Leuconostoc (5.33%). FR had the highest proportion of identified species (76.2% of total bacterial species) among the three types of fermented foods, in which 52.9% can only be exclusively found in FR samples (Figure S1A). Prevalent bacterial species, present in over 90% of samples in each group, include Bacillus, Enterococcus, Lactococcus, Leuconostoc, Staphylococcus, Tetragenococcus, and Weissella. All of these genera fall within the lactic acid bacteria category and are classified under the class Bacilli, with the majority belonging to the order Lactobacillales.

Alpha diversity analysis showed that the DC group exhibited the least species diversity (Shannon index) among the three types of fermented soy products (Fig. 1B). This observation is justifiable considering the predominantly solid-state nature of DC samples. To investigate the microbial community structures among samples, we performed non-metric multidimensional scaling (NMDS) based on Bray-Curtis dissimilarity. The NMDS plot showed significant differences in microbial communities among the three fermented foods (Fig. 1C), in which PRMANOVA analysis indicated a 19% contribution to the total variability (R² = 0.19, p = 0.001). Notably, an even greater contribution to the dissimilarity was attributed to geographic origin (R² = 0.26, p = 0.001). These results indicated that the composition and structure of fermented soy products were dependent on both food type and geographic origin. A plausible hypothesis is that close proximity in regions leads to the adoption of similar fermentation processes. To validate this theory, the Mantel test was used to examine the correlation between geographical distance and Bray-Curtis dissimilarity of the microbiome. The results of the Mantel test, whether applied to the entire sample set (r = 0.092, p = 0.0045) or subgroup samples, supported this relationship (Fig. 1D). This indicated a strong positive correlation between the geographic distance and Bray-Curtis dissimilarity matrices (Figure S1D).

Linear discriminant analysis Effect Size (LEfSe) was used to analyze the microbial taxa in fermented foods. The results highlighted that Firmicutes and Bacteroidetes were the dominant microbial taxa in DC and FR samples respectively, and a multitude of other phyla were enriched in DJ samples (Fig. 1E). Notably, there were 56, 89, and 115 species enriched in DC, DJ, and FR samples respectively (Linear Discriminant Analysis, LDA > 2). The most enriched species in the three fermented foods were fermentation starters (DC: Bacillus subtilis (LDA = 5.07); DJ: Aspergillus oryzae (LDA = 4.88); FR: Tetragenococcus halophilus (LDA = 4.81). B.subtilis was identified as a product component in several DC products. Apart from B.subtilis, DC-enriched Bacillus species such as B.velegensis, B.amyloliquefaciens, and B.licheniformis were also commonly used in fermented soy products. All Bacillus species are able to produce bioactive compounds, which are known to exert health benefits.[21, 22] FR samples were enriched with several lactic acid bacteria, including Lactococcus lactis, Ligilactobacillus acidipiscis, Leuconostoc citreum, Leuconostoc lactis, Leuconostoc mesenteroides (Table S2).

On the other hand, a notable presence of opportunistic pathogenic species mostly from the Proteobacteria phylum and the Enterobacteriaceae family was observed in each group. This includes Klebsiella pneumoniae, Klebsiella aerogenes, Klebsiella oxytoca, Klebsiella michiganensis and Enterobacter hormaechei enriched in FR, as well as Enterobacter cancerogenus enriched in DJ. Using a read count of > 1000 as cut-off, we also detected several foodborne pathogens including Bacillus cereus, Clostridium botulinum, Clostridium perfringens, Cronobacter sakazakii, Escherichia coli, Listeria monocytogenes, Salmonella enterica, Staphylococcus aureus, Vibrio anguillarum and Yersinia enterocolitica in our fermented food samples (Fig. 1F). Amongst these, Salmonella enterica was notably present in FR (LDA = 2.63).

Functional landscapes of microbiome in fermented soy products

To better understand the functions of microbiome in fermented soy products, sequenced reads were assembled into contigs and the genes predicted from these contigs were dereplicated. This resulted in the Soybean Fermented Food non-redundant Gene Catalogue (SFFGC, see Methods) which contained a total of 2,359,387 genes. The functional annotations of these genes were accomplished using a multiple of available databases, including Clusters of Orthologous Genes (COGs), Kyoto Encyclopedia of Genes and Genomes (KEGG), Enzyme Commission Categories (ECs), Comprehensive Antibiotic Resistance Database (CARD), and Carbohydrate Active Enzymes (CAZys).

Our analysis revealed that 84.4% of the SFFGC genes could be successfully annotated using at least one of the aforementioned databases (Fig. 2A). Among these genes, only 8.54% originated from DJ samples, while over half (53.4%) were derived from FR, and 38.1% were from DC. A total of 864 KEGG pathways, 734 KEGG modules, 11930 KEGG Orthology (KO) groups, and 327 CAZy gene families were annotated to the SFFGC genes. For each functional component, the abundance was calculated based on the collective abundance of SFFGC genes associated with that element. Additionally, the abundance of 562 MetaCyc pathways were determined using HUMAnN3. NMDS plot of Bray-Curtis dissimilarity based on KO abundances revealed significant microbial function differences among the three types of fermented foods (Fig. 2B).

Using a LDA score of > 2 as threshold, the LEfSe results showed discernible differences in the abundance of genetic elements among the three groups. Notably, these differences encompassed 7.4% of KO genes, 79.8% of CAZy gene families, 61.7% of KEGG modules, 21.6% of KEGG pathways, and 21.5% of MetaCyc pathways. Interestingly, there was a notably higher count of DC-enriched KO genes, FR-enriched antibiotic genes, and DJ-enriched CAZy gene families (Fig. 2C). We also observed that most reads (66.7% on average) of MetaCyc pathways enriched in DJ were assigned to “UNMAPPED” by HUMAnN3. In contrast, 40% of reads in DC and 35.4% in FR were unassigned, indicating a significant proportion of unknown microbial functions in fermented foods. A significant increase in AA and GT counts were observed in DJ-enriched CAZy gene families (Fig. 2D). GTs are known for facilitating the synthesis of glycosidic linkages by transferring sugar moieties from phospho-activated sugar donors to various acceptors, including both saccharide and non-saccharide molecules. This enzyme family is crucial in the biosynthesis of disaccharides, polysaccharides, and oligosaccharides.

In regards to the resistome of the fermented soy products, a total of 2406 distinct antibiotic resistance genes were identified and annotated with 734 terms from the Antibiotic Resistance Ontology (ARO). Notably, the resistance-nodulation-cell division (RND) antibiotic efflux pump resistance genes were the most prominent (26.6% (640 genes)), followed by the major facilitator superfamily (MFS) antibiotic efflux pump genes (17.6% (424 genes)). 373 of the 734 ARO terms were present in all three fermented soy products, each with non-zero abundances in at least 10% of samples (Fig. 2E). Additionally, LEfSe analysis (LDA score > 2) revealed that 54.4% (399) of these terms varied in abundance across the different food types (Fig. 2C).

Antagonistic activity of Lactobacillales or Bacillales against Enterobacterales

To uncover the inter-relationships among these species, co-abundance microbial species networks were constructed for each food category. Only species present in at least 5 samples in each group were included in the microbial network inference using SparCC algorithm. For species networks, we identified a total of 458 correlations in DC, 371 in DJ, and 401 in FR (FDR < 0.05) (DC in Fig. 3A, DJ and FR in Figure S2). After clustering, we observed that species exhibited a tendency to form “bacteria clique” based on their taxonomic order affiliations (Fig. 3A, 3B).

Analysis revealed a notable competition between Lactobacillales and Enterobacterales across all three categories of fermented foods. In each group of fermented soy products, species belonging to the same taxonomic order demonstrated positive correlations, whilst those from distinct orders-such as Bacillales and Enterobacterales, or Enterobacterales and Lactobacillales-displayed negative correlations. An exception existed in the relationship between Bacillales and Lactobacillales, which exhibited predominantly positive correlations in DC and DJ but displayed negative correlations in FR (Fig. 3C). In the constructed networks, species with a degree exceeding the 80th percentile were designated as keystone species. Notably, in the context of DJ and FR networks, approximately 50% of the identified keystone species originated from the Enterobacterales order (Fig. 3D). This observation highlighted the significant role of Enterobacterales species in the microbial ecosystem of fermented foods.

Recovery of fermented food bacteria genomes from metagenomes

To better understand the distinctions among the three fermented soy products at the genomic level, we used the binning method in metaWRAP on assembled contigs to generate metagenomic assembled genomes (MAGs). Following this pipeline, we recovered a total of 707 high-quality MAGs, with completeness greater than 70% (average: 89.08%) and contamination less than 10% (average: 2.5%) (Figure S3A).

Median size of 707 MAGs was 2.31 megabases (MB) (2.08–3.75 MB), the N50 values ranging from 1.7 kilobases to 2.8 Mb. 363 MAGs (51.3% of the total dataset) were > 90% complete and < 5% contaminated; 10 MAGs even reached 100% completeness and thus referred to as near-complete genomes. Among these MAGs, 36%, 12%, and 52% were from DC, DJ and FR, respectively. The 707 MAGs were dereplicated at a 99% average nucleotide identity (ANI), resulting in 304 dereplicated MAGs, which were used to construct the phylogenetic tree (Fig. 4).

The 707 MAGs were spread over 2 archaea and 705 bacterial genomes. The majority of bacterial genomes belonged to the Phylum Firmicutes (533, 76%), Proteobacteria (87, 12%), Actinobacteriota (50, 7%) and Bacteroidota (30, 4%). A substantial number of MAGs (n = 648 or 91.6%) were successfully allocated to 184 known species, and the remaining 59 MAGs (8.4%) were currently unclassified (Figure S3B).

We estimated ANI against 666 reconstructed food MAGs from Pasolli et al,[10] using a 95% ANI cut-off defined as intraspecies genomes. The 666 food MAGs were reconstructed from 303 metagenomes of different fermented food types from 12 datasets, which were comprised mostly of fermented foods in the Western culture. Only 69 (22.7%) of our dereplicated MAGs matched 275 (41.3%) of the 666 food MAGs.

We used metaProbiotics to predict whether MAGs could potentially act as probiotics.[23] This method employed innovative biological sequence representation by converting 8-mers of DNA sequences into word vectors using natural language processing (NLP) techniques, and built prediction models using random forests. 138 of the 304 dereplicated MAGs were predicted as potential probiotics (Fig. 4). The majority of predicted probiotics were Actinobacteriota, the ratio of predicted probiotics to non-probiotics were 31:6, followed by Firmicutes (88:86), Proteobacteria (12:59), and Bacteroidota (6:15). These results suggested that fermented foods may be a source of potential probiotics.

Among the 2406 antibiotic resistance genes (ARGs) annotated, a total of 721 were detected in our set of 209 MAGs, and 342 of these genes (47.43%) were classified under the Enterobacterales order. Notably, the five most prevalent species that possessed ARGs were Klebsiella pneumoniae (32 instances), Enterobacter hormaechei_A (31 instances), Acinetobacter baumannii (28 instances), Enterobacter cloacae (27 instances), and Enterobacter kobei (26 instances). It is worth noting that ARGgene families within the Enterobacterales species included resistance-nodulation-cell division antibiotic efflux pump and major facilitator superfamily antibiotic efflux pump (Figure S4). These findings collectively suggest that Enterobacterales species serve as the principal reservoir of antimicrobial resistance in fermented soy products.

Comparative analysis indicated Klebsiella strains in gut of Chinese individuals may originate from fermented foods

To investigate the prevalence of fermented food bacteria within the gut microbiome of healthy Chinese individuals, we accessed the taxonomic profile of 689 individuals from 8 publicly available Chinese metagenomic datasets curated in CurateMetagenomicData.[24] We used MetaPhlan3 for taxonomic profiling of our metagenomic data derived from fermented foods to ensure comparability with data in CurateMetagenomicData. Only species present in more than 50% of samples in each food group were considered, resulting in a total of 67 species included in the analysis.

The results showed that the most prevalent species were from the order Enterobacterales, including three Klebsiella species: Klebsiella pneumoniae (62.4%), Klebsiella variicola (50.5%) and Klebsiella quasipneumoniae (42.5%). Additionally, the Enterobacter cloacae complex (16.9%) and Citrobacter freundii (12.04%) were also significant (Fig. 5A).

The most common lactic acid bacteria were Weissella confusa (7.55%) and Lactococcus lactis (4.35%). Notably, Lactococcus lactis in the Chinese population was observed to be lower than the global average of 7.5%.[10] To determine potential associations between Enterobacterales species and lactic acid bacteria in the gut microbiome, samples containing both K. pneumoniae and either W. confusa or L. lactis were selected for analysis. Findings indicated that the co-occurrence of Weissella confusa/Lactococcus lactis and Klebsiella pneumoniae was found to be mutually exclusive (validated through a chi-squared test, both p-values < 2.2e-16). This observation was consistent with the network analysis results mentioned above, revealing an antagonistic relationship between Enterobacterales and lactic acid bacteria.

We conducted a genomic level investigation to gain a deeper understanding of the complex relationship between bacterial species present in fermented foods and those in the gut of Chinese individuals. 24417 gut MAGs of Chinese individuals were retrieved from 154,723 human MAGs by Pasolli et al.[25] Using fastANI (95% ANI intra-species cutoff), we identified 203 hits from the 303 food-derived MAGs in this study. Remarkably, MAGs annotated with K. pneumoniae showed the highest number of matches (Fig. 5B). This supported the idea that K. pneumoniae found in both fermented foods and the gut belonged to the same species. Additionally, we observed that potentially harmful species were more prevalent than putative beneficial species in the gut metagenome-assembled genomes. Examples include Proteus mirabilis, a prevalent pathogen that can cause a range of diseases,[26] and Morganella morganii, a colorectal cancer-associated species enriched in the gut microbiota of inflammatory bowel disease (IBD) and colorectal cancer (CRC) patients.[27]

In order to analyze the relationship between fermented food and gut microbial species at the strain level, StrainPhlan3[28] was used to construct the phylogenetic relationships of common species in both fermented foods and gut samples of Chinese individuals. Only two phylogenetic trees (Klebsiella quasipneumoniae and Klebsiella pneumoniae)—had terminal branches representing lineages found in both food and gut samples (Fig. 5C). Notably, the phylogenetic tree related to K. pneumoniae showed a distinct branch specific to gut samples. However, it is noteworthy that several gut samples were scattered among the branch predominantly populated by fermented foods samples.

Potential Horizontal Gene Transfer of Antibiotic Resistance Genes from Fermented Foods to the Human Gut Microbiome

A comprehensive analysis was conducted to evaluate the risk of ARGs in fermented foods being transmitted to the human gut microbiota. A BLAST search was conducted on our set of 2,406 ARG protein sequences against the extensive collection of 216,849 ARGs from the human microbiome, as curated by Lee et al.[29]

Our analysis revealed 898 matches between ARGs in fermented foods and 4,518 analogous ARGs in the human gut microbiome. Of these, 4,219 hits (93.38% of the total) originated from fecal metagenomes. The hit matches spanned 140 ARG families, representing 32.4% of human gut ARG families. Given that Pasolli et al.'s Species-level genome bins (SGBs) were also used by Lee et al.,[10] we extended our analysis to explore potential routes for ARG transfer. Specifically, we linked the ARGs identified in fermented foods to Chinese SGB genomes from Pasolli et al.'s dataset to uncover potential pathways for ARG transfer. Interestingly, it was found that ARGs from fermented foods that matched Chinese SGB genomes were primarily associated with Enterobacteriaceae bacteria in the gut. The ARGs in K. pneumoniae from fermented foods showed striking similarities to those found in the gut, much like the similarities observed in Proteus mirabilis and Morganella morganii. Furthermore, the notable presence of ARGs in Enterobacter hormaechei highlighted the increase of gene transfer activity, as shown by gene transfer to various species in the gut microbiota (Fig. 6). Collectively, these findings strongly support the idea that ARGs in fermented foods can be horizontally transfer to intestinal microorganisms, thereby suggesting complex cross-microbial interactions.

Resistome risk scores were computed using MetaCompare2,[30] which generated two resistance risk assessments: ecological resistome risk (ERR) and human health resistome risk (HHRR). ERR scores account for various pathogens and ARGs, reflecting the potential mobility of ARGs within an environment. In contrast, HHRR specifically focused on pathogens and high-risk ARGs associated with antibiotic-resistant infections in humans (defined as Rank I ARGs by an omics-based framework). Our results indicated that FR exhibited the highest ERR, followed by DC. Meanwhile, DC exhibited a relatively higher HHRR compared to the other groups (Figure S5). Five samples with the highest HHRR in DC were further examined and it was revealed that 3 of them originated from natto in Japan. These results highlighted the diverse antibiotic resistance risks associated with different fermented foods.

Sample information

A total of 93 fermented soy products were collected from markets, including 42 DouChi ( fermented soybean), 33 FuRu ( fermented bean curd), and 18 DouJiang ( fermented soybean sauce) samples. The selection was based on region, sales volume, raw materials, and flavor to ensure representativeness of commonly consumed foods in China. Samples were purchased from both online and offline supermarkets. Homogenized samples were shipped frozen to the laboratory and stored at -80°C until assayed.

Library preparation, metagenomic sequencing, and bioinformatic pipeline

DNA was extracted using the PowerSoil DNA Isolation Kit (MoBio Laboratories) according to the manufacturer’s instructions. DNA concentration was measured using Qubit Fluorometric Quantitation (DS DNA High-Sensitivity Kit, ThermoFisher Cat. No. Q32851). DNA sequencing libraries were prepared using the Nextera XT DNA Library Prep Kit (Illumina Cat. No. FC-131-1096). Libraries were prepared with the NexteraXT DNA Library Preparation Kit (Illumina) and sequenced on the illumina NovaSeq X.

Raw paired-end reads were subjected to quality control using Trimmomatic (v0.39),[31] resulting in high-quality clean reads. Each sample's clean reads were then assembled into contigs using MEGAHIT (v1.1.3) with default parameters.[32] Next, gene prediction was performed on the contigs using Prodigal.[33] Redundant genes were removed using CD-HIT,[34] resulting in a SFFGC. To annotate the SFFGC, eggNOG was used by aligning the genes to the eggNOG database with eggnog-mapper (v2.1.3).[35] ARGs were annotated by aligning the genes against the CARD using RGI (v5.1.1).[36] The assignment of CAZys was performed using run_dbcan.py against the dbCAN HMM,[37] DIAMOND,[38] and Hotpep databases[39] with default settings. The final CAZyme annotations were determined by the best hits from all three databases. To estimate the abundance of each gene in each sample, BWA[40] was used to map the clean reads to the non-redundant gene catalog. The read count of each gene was normalized into relative abundance using reads/kilobase/million mapped reads (RPKM). To calculate the abundances of KO, KEGG pathway, eggNOG Orthology, ARO, and CAZy, the abundance of all genes annotated to each category was summed. Taxonomic profiling was performed on the clean reads using Kraken2[41] with the k2PlusPF database. To reduce noise, low-abundance filters were applied, retaining only taxonomic features with a read count greater than 500 and present in at least four samples. HUMAnN3 was also used for metabolic pathway functional profiling and reconstruction.[42]

Metagenomic assembled genomes (MAGs)

The MAGs were generated by MetaWRAP pipeline.[43] In detail, the contigs were clustered into metagenomic bins using metaWRAP binning module. The bins were then refined using the bin_refinement module (parameters: -c 70 -x 10 options, retain bins with completeness > 70% and contamination ˂10%). These refined bins were then dereplicated using dRep v3.2.2[44] with -str 100 -strW 0 as parameters. The taxonomy of the dereplicated MAGs was classified using GTDB-Tk v2.0.0 with the GTDB r207. Phylogenetic relationships among the bacterial MAGs were inferred using PhyloPhlAn (v.0.99).[45] Phylogeny was built using the 400 universal PhyloPhlAn markers with the following options “--diversity high --accurate”. The configuration file was customized to use Diamond v2.0.5 for mapping, MAFFT v7.505 for multiple sequence alignment, trimAl version v1.4.rev15 for trimming, FastTree for the first tree generation and RAxML v8.2.12 for the final tree generation. The tree was viewed and annotated using iTOL.[46]

Network construction and clustering

Co-abundance networks were generated separately for DC, DJ and FR microbiomes. Species read count data from Kraken2 was subjected to FastSpar,[47] which is an efficient and parallelizable implementation of the SparCC algorithm for rapid network inference. Microbial community networks, defined as clusters of species, were detected using the Louvain algorithm with the ‘cluster_louvain’ function in the igraph package.[48] Networks were visualized using R package ggraph, with species as nodes and correlations between species as edges.

Comparison of bacterial species between fermented food and Chinese gut microbiome

Sample information from 689 healthy Chinese individuals was retrieved from 8 publicly available Chinese metagenomic datasets in CurateMetagenomicData.[24] As taxonomic abundances were calculated using MetaPhlAn3[49] in CuratedMetagenomicData, we also used MetaPhlAn3 to generate taxonomic profiles for our fermented food samples to ensure comparability. Only species present in more than 50% of samples in each group were considered. To further compare at the genome level, we retrieved 24417 Chinese gut MAGs from 154,723 human MAGs from Pasolli et al.[10] Intra-species genomes were determined by ANI with a 95% cut-off using fastANI.[50] Strain-level analysis was performed using StrainPhlAn3[51] with the default parameters. Strain-level profiling using StrainPhlAn3 was performed on 82 species of the fermented food microbiome with sufficient coverage. We randomly selected 100 Chinese gut metagenomic samples for StrainPhlAn3 analysis.

The ARG proteins were aligned to the 216,849 ARGs retrieved from human microbiome in Lee et al[29] using the ‘blastp’ function of Diamond v2.0.15.[52] The BLAST hits were filtered by a maximum e-value 1e-20, a minimum identity 80%, and a minimum length 100. Using R package ggsankey, a Sankey plot was used to illustrate the potential horizontal gene transfer of ARGs from fermented food microbial species to gut species of Chinese.

Statistical analysis

Alpha diversity indices were calculated using vegan R package.[53] The Shannon index was used for alpha diversity evaluation based on species-level relative abundance. Statistical comparisons of pairwise groups were performed by Wilcoxon rank sum test. The Benjamini-Hochberg method was applied to adjust p values for multiple tests. To estimate community dissimilarities, Bray-Curtis distances were calculated using phyloseq[54] and PERMANOVA tests were implemented to evaluate microbiota compositional differences among groups using adonis2 function from the vegan package in R.

Differences between the three groups for both taxa and various functions were analyzed using LEfSe[55] with default settings. LEfSe uses the non-parametric Kruskal-Wallis test to assess significance among groups, followed by pairwise tests between 2 groups using the unpaired Wilcoxon test. Finally, LDA was used to estimate the effect size of each differentially abundant feature.

The ISAPP has recommended including fermented foods as a distinct category in dietary guidelines.[7] There is substantial evidence supporting the benefits of fermented foods,[56–59] especially in recent years, studies on the microbiome have also highlighted their positive impact on gut health.[56, 60–64] However, more research on fermented food microbiome is needed to strengthen the existing evidence. Concurrently, it is vital to critically assess the safety of microorganisms present in fermented foods and their potential impact on the human intestinal flora. Our study investigated the microbiomes of three traditional Chinese fermented soy products using metagenomics approaches. We also examined the potential interactions between microbial species of fermented foods and gut microbiome of Chinese population to better understand the possible health benefits and risk. This study aimed to provide insights that could influence the inclusion of fermented foods in the dietary guidelines for the Chinese population.

Our findings showed a significant difference in microbial composition among the three fermented foods. Notably, a substantial correlation was observed between microbial communities and geographical origins, indicating that fermented foods from nearby regions tend to share similar microbial compositions. This phenomenon can be attributed to the similar fermentation processes adopted in neighboring regions for the same type of fermented foods. Additionally, the non-random distribution of microorganisms was notably influenced by geographical and climatic factors[20, 65]. This study revealed that DJ samples showed a significant abundance of eukaryotic organisms. FR samples showcased the highest proportion of identified bacterial species, which can be attributed to the specific fermenting microorganisms and the fermentation processes used. These observed microbial compositions align with previously reported research outcomes.[18, 19, 66, 67] Despite identifying numerous beneficial bacteria in fermented soy products including lactic acid bacteria, our investigation also revealed the presence of opportunistic pathogens. Notably, certain species within the Enterobacteriaceae family, such as Klebsiella spp, as well as other well-recognized foodborne pathogens such as Listeria monocytogenes, were detected. These findings serve as a significant reminder of the safety considerations associated with consuming fermented foods.

By constructing co-abundance microbial species networks for each food category, we gained insights into the complex inter-species relationships within microbial ecosystems. In our investigation, we identified distinct “bacterial cliques” within their respective orders including beneficial cliques such as Bacillales and Lactobacillales, and harmful cliques such as Enterobacterales. Studies have indicated that, compared to just either species in isolation, coexistence of opportunistic pathogens can lead to more severe infections as exemplified by Acinetobacter baumannii and Klebsiella pneumoniae.[68] Similarly, our results suggested a potential interaction between these two species in DC and FR. Furthermore, our analysis highlighted an antagonistic relationship between these beneficial and harmful cliques. The substantial antagonistic activity of lactic acid bacteria against Enterobacteriaceae species, which are known for their pronounced multi-antibiotic resistance, has been consistently reported in the literature.[69, 70] Our findings align with and reinforce this observed phenomenon. This antagonistic relationship is also observed in other ecosystems, such as the human gut, where Lactobacillus plays a pivotal role in limiting the colonization of multidrug-resistant Enterobacteriaceae.[71] In vitro and animal studies also indicated that Lactobacillus spp. has an inhibitory effect against Klebsiella pneumoniae.[70, 72] These findings suggest that fermented foods could be a potential intervention strategy against harmful gut bacteria,[73] provided that the fermented products are free from pathogenic bacteria themselves.

A total of 707 microbial genomes was successfully recovered from the fermented foods using MAGs. Nearly 10% of the MAGs could not be assigned to specific species, underscoring the potential of fermented foods as a rich source for discovering novel microbial species. Our collection of MAGs not only enriches but also expands the existing repository of microbial genomes from fermented food. By exploring microbial functions, we identified substantial differences in the microbial functional profiles of the three fermented soy products. Notably, we observed a significant enrichment of ARGs in FR, a finding that is consistent with earlier literature.[74] Annotation of ARGs revealed that nearly half of the MAGs belonged to the Enterobacterales order. These results strongly suggest that Enterobacterales play a pivotal role in carrying ARGs in FR. This underscores the need for implementing more stringent measures to improve food safety protocols.

When conducting a comprehensive comparative analysis, we discovered that Klebsiella species was the predominant fermented food microorganism in the gut microbiota of Chinese individuals. Klebsiella pneumoniae, in particular, exhibited a notably high prevalence, exceeding 55%, in the Chinese population.[75] Our investigation at the genomic and strain levels suggests that fermented foods might be a source of Klebsiella species including Klebsiella pneumoniae and Klebsiella quasipneumoniae, in the gut environment of Chinese individuals, which is a compelling discovery that warrants additional research. The lower prevalence of lactic acid bacteria, such as Weissella confusa and Lactococcus lactis, in the gut environment of the Chinese population may be attributed to the antagonistic effects of Klebsiella bacteria. This interaction may contribute to the lower-than-average prevalence of these beneficial lactic acid bacteria in the Chinese gut.[10] Synergistic interactions among Klebsiella species may enhance their ability to colonize the gut.[73] An increased relative abundance of K. pneumoniae was linked to a higher risk of bacteremia, nosocomial transmission, and prolonged colonization.[76]

The widespread use of antibiotics significantly contributed to the rapid rise in antibiotic resistance (AR) and the perturbation of the host's gut microbiota, which is linked to modern diseases such as type 2 diabetes, cardiovascular diseases and cancers.[77–79] Fermented foods are increasingly recognized as a modern approach to restore imbalanced gut microbiota and mitigate associated health challenges.[5, 64] Although this study did not provide any direct evidence, it contributes to the understanding that fermented foods can significantly influence the gut antibiotic resistome. Our findings align with a recent study that used fermented food as intervention to highlight the significant effect of traditionally fermented foods on modulating host intestinal antibiotic resistance.[80]

Nevertheless, this study has some limitations. First, this study focused on only three prevalent fermented soy products, which were chosen for their representativeness. It is important to note that China has a diverse range of fermented foods. Future investigations should include a wider range of fermented food types. Furthermore, the use of metagenomic sequencing prevented us from accurately assessing viable bacterial counts in the fermented food samples. This limitation is particularly relevant as many fermented products are sterilized before reaching the market. Lastly, the number of gut microbiome samples from the Chinese population was limited and exhibited regional biases. Although these samples offer insights, they may not fully represent the entirety of gut microbiome in the Chinese population. Additionally, the indirect comparison of fermented food and gut microbes in this study requires validation through experimental approaches.

Despite ISAPP's recommendation to include fermented foods in national dietary guidelines, comprehensive evaluation of the microbiological composition and safety of such foods remains lacking. This study systematically examined the microbial composition and functions of three fermented soy products, compared them to the gut microbiome, and reached the following conclusion: Although fermented foods contain numerous beneficial microorganisms, safety concerns cannot be overlooked. These foods often contain opportunistic and foodborne pathogens, which may persist in the gut microbiota and contribute to the gut antibiotic resistome. While we endorse ISAPP's stance on promoting fermented foods in dietary guidelines, it is crucial to conduct a rigorous and systematic exploration of microbiological safety risks across various fermented foods prior to public recommendation.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and material

Raw data of our metagenomic sequencing has been deposited in the NCBI with accession code PROJECT ID: PRJNA1152330 (https://dataview.ncbi.nlm.nih.gov/object/PRJNA1152330).

Competing interests

The authors declare that they have no competing interests.

Funding

This study was supported by Beijing Municipal Science and Technology Commission (Z191100008619006), the Chinese Association for Student Nutrition & Health Promotion-Mead Johnson Nutritionals (China) Joint Fund (Grant No. CASNHP-MJN2023-04) and Major Research Project on Philosophy and Social Sciences of the Ministry of Education (21JZD039); National Key R&D Program Project of the Ministry of Science and Technology (2021YFC2600501).

Authors' contributions

YY and BZ designed, organized, and implemented the entire experiment. The manuscript was prepared by XX, YL, BL, MZ and JZ. JY, GH and MM were responsible for sample management and data analysis. All authors read and approved the final manuscript.

Acknowledgements

Not applicable.

de Vos WM, Tilg H, Van Hul M, Cani PD. Gut microbiome and health: mechanistic insights. Gut. 2022;71:1020–32.
Lee MH, Nuccio S-P, Mohanty I, Hagey LR, Dorrestein PC, Chu H, et al. How bile acids and the microbiota interact to shape host immunity. Nat Rev Immunol. 2024;
Ludgate ME, Masetti G, Soares P. The relationship between the gut microbiota and thyroid disorders. Nat Rev Endocrinol. 2024;
Wastyk HC, Fragiadakis GK, Perelman D, Dahan D, Merrill BD, Yu FB, et al. Gut-microbiota-targeted diets modulate human immune status. Cell. 2021;184:4137-4153.e14.
Taylor BC, Lejzerowicz F, Poirel M, Shaffer JP, Jiang L, Aksenov A, et al. Consumption of Fermented Foods Is Associated with Systematic Differences in the Gut Microbiome and Metabolome. mSystems. 2020;5:e00901-19.
Leeuwendaal NK, Stanton C, O’Toole PW, Beresford TP. Fermented Foods, Health and the Gut Microbiome. Nutrients. 2022;14:1527.
Marco ML, Sanders ME, Gänzle M, Arrieta MC, Cotter PD, De Vuyst L, et al. The International Scientific Association for Probiotics and Prebiotics (ISAPP) consensus statement on fermented foods. Nat Rev Gastroenterol Hepatol. 2021;18:196–208.
Milani C, Duranti S, Napoli S, Alessandri G, Mancabelli L, Anzalone R, et al. Colonization of the human gut by bovine bacteria present in Parmesan cheese. Nat Commun. 2019;10:1286.
Oozeer R, Leplingard A, Mater DDG, Mogenet A, Michelin R, Seksek I, et al. Survival of Lactobacillus casei in the human digestive tract after consumption of fermented milk. Appl Environ Microbiol. 2006;72:5615–7.
Pasolli E, De Filippis F, Mauriello IE, Cumbo F, Walsh AM, Leech J, et al. Large-scale genome-wide analysis links lactic acid bacteria from food with the gut microbiome. Nat Commun. 2020;11:2610.
Lebeer S, Vanderleyden J, De Keersmaecker SCJ. Host interactions of probiotic bacterial surface molecules: comparison with commensals and pathogens. Nat Rev Microbiol. 2010;8:171–84.
Liu L, Chen X, Hao L, Zhang G, Jin Z, Li C, et al. Traditional fermented soybean products: processing, flavor formation, nutritional and biological activities. Crit Rev Food Sci Nutr. 2022;62:1971–89.
Nout MJR. Fermented foods and food safety. Food Res Int. 1994;27:291–8.
Li Y, Fu S, Klein MS, Wang H. High prevalence of antibiotic resistance in traditionally fermented foods as a critical risk factor for host gut antibiotic resistome [Internet]. bioRxiv; 2023 [cited 2024 Jul 19]. p. 2023.04.21.537834. Available from: https://www.biorxiv.org/content/10.1101/2023.04.21.537834v2
Wang HH, Manuzon M, Lehman M, Wan K, Luo H, Wittum TE, et al. Food commensal microbes as a potentially important avenue in transmitting antibiotic resistance genes. FEMS Microbiol Lett. 2006;254:226–31.
Vinayamohan PG, Viju LS, Joseph D, Venkitanarayanan K. Fermented Foods as a Potential Vehicle of Antimicrobial-Resistant Bacteria and Genes. Fermentation. 2023;9:688.
do Prado FG, Pagnoncelli MGB, de Melo Pereira GV, Karp SG, Soccol CR. Fermented Soy Products and Their Potential Health Benefits: A Review. Microorganisms. 2022;10:1606.
Chen T, Jiang S, Xiong S, Wang M, Zhu D, Wei H. Application of denaturing gradient gel electrophoresis to microbial diversity analysis in Chinese Douchi. J Sci Food Agric. 2012;92:2171–6.
Ren F, Yan D-H, Liu Y, Wang C, Guo C. Bacterial and fungal communities of traditional fermented Chinese soybean paste (Doujiang) and their properties. Food Sci Nutr. 2021;9:5457–66.
Xu M, Su S, Zhang Z, Jiang S, Zhang J, Xu Y, et al. Two sides of the same coin: Meta-analysis uncovered the potential benefits and risks of traditional fermented foods at a large geographical scale. Front Microbiol. 2022;13:1045096.
Gopikrishna T, Suresh Kumar HK, Perumal K, Elangovan E. Impact of Bacillus in fermented soybean foods on human health. Ann Microbiol. 2021;71:30.
Ghelardi E, Abreu Y Abreu AT, Marzet CB, Álvarez Calatayud G, Perez M, Moschione Castro AP. Current Progress and Future Perspectives on the Use of Bacillus clausii. Microorganisms. 2022;10:1246.
Wu S, Feng T, Tang W, Qi C, Gao J, He X, et al. metaProbiotics: a tool for mining probiotic from metagenomic binning data based on a language model. Brief Bioinform. 2024;25:bbae085.
Pasolli E, Schiffer L, Manghi P, Renson A, Obenchain V, Truong DT, et al. Accessible, curated metagenomic data through ExperimentHub. Nat Methods. 2017;14:1023–4.
Pasolli E, Asnicar F, Manara S, Zolfo M, Karcher N, Armanini F, et al. Extensive Unexplored Human Microbiome Diversity Revealed by Over 150,000 Genomes from Metagenomes Spanning Age, Geography, and Lifestyle. Cell. 2019;176:649-662.e20.
Hao X, Cen X, He M, Wen Y, Zhang H. Isolation, biological and whole genome characteristics of a Proteus mirabilis bacteriophage strain. BMC Microbiol. 2023;23:215.
Cao Y, Oh J, Xue M, Huh WJ, Wang J, Gonzalez-Hernandez JA, et al. Commensal microbiota from patients with inflammatory bowel disease produce genotoxic metabolites. Science. 2022;378:eabm3233.
Truong DT, Tett A, Pasolli E, Huttenhower C, Segata N. Microbial strain-level population structure and genetic diversity from metagenomes. Genome Res. 2017;27:626–38.
Lee K, Raguideau S, Sirén K, Asnicar F, Cumbo F, Hildebrand F, et al. Population-level impacts of antibiotic usage on the human gut microbiome. Nat Commun. 2023;14:1191.
Rumi MA, Oh M, Davis BC, Juvekar A, Brown CL, Vikesland PJ, et al. MetaCompare 2.0: Differential ranking of ecological and human health resistome risks [Internet]. bioRxiv; 2024 [cited 2024 Jul 4]. p. 2024.01.17.576132. Available from: https://www.biorxiv.org/content/10.1101/2024.01.17.576132v1
Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinforma Oxf Engl. 2014;30:2114–20.
Li D, Liu C-M, Luo R, Sadakane K, Lam T-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinforma Oxf Engl. 2015;31:1674–6.
Hyatt D, Chen G-L, Locascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics. 2010;11:119.
Li W, Godzik A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinforma Oxf Engl. 2006;22:1658–9.
Cantalapiedra CP, Hernández-Plaza A, Letunic I, Bork P, Huerta-Cepas J. eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. Mol Biol Evol. 2021;38:5825–9.
Alcock BP, Raphenya AR, Lau TTY, Tsang KK, Bouchard M, Edalatmand A, et al. CARD 2020: antibiotic resistome surveillance with the comprehensive antibiotic resistance database. Nucleic Acids Res. 2020;48:D517–25.
Yin Y, Mao X, Yang J, Chen X, Mao F, Xu Y. dbCAN: a web resource for automated carbohydrate-active enzyme annotation. Nucleic Acids Res. 2012;40:W445-451.
Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. 2015;12:59–60.
Busk PK, Pilgaard B, Lezyk MJ, Meyer AS, Lange L. Homology to peptide pattern for annotation of carbohydrate-active enzymes and prediction of function. BMC Bioinformatics. 2017;18:214.
Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinforma Oxf Engl. 2009;25:1754–60.
Wood DE, Salzberg SL. Kraken: ultrafast metagenomic sequence classification using exact alignments. Genome Biol. 2014;15:R46.
Beghini F, McIver LJ, Blanco-Míguez A, Dubois L, Asnicar F, Maharjan S, et al. Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. eLife. 2021;10:e65088.
Uritskiy GV, DiRuggiero J, Taylor J. MetaWRAP-a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome. 2018;6:158.
Olm MR, Brown CT, Brooks B, Banfield JF. dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. ISME J. 2017;11:2864–8.
Segata N, Börnigen D, Morgan XC, Huttenhower C. PhyloPhlAn is a new method for improved phylogenetic and taxonomic placement of microbes. Nat Commun. 2013;4:2304.
Letunic I, Bork P. Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool. Nucleic Acids Res. 2024;52:W78–82.
Watts SC, Ritchie SC, Inouye M, Holt KE. FastSpar: rapid and scalable correlation estimation for compositional data. Bioinforma Oxf Engl. 2019;35:1064–6.
Seth S, Mallik S, Bhadra T, Zhao Z. Dimensionality Reduction and Louvain Agglomerative Hierarchical Clustering for Cluster-Specified Frequent Biomarker Discovery in Single-Cell Sequencing Data. Front Genet. 2022;13:828479.
Truong DT, Franzosa EA, Tickle TL, Scholz M, Weingart G, Pasolli E, et al. MetaPhlAn2 for enhanced metagenomic taxonomic profiling. Nat Methods. 2015;12:902–3.
Hernández-Salmerón JE, Moreno-Hagelsieb G. FastANI, Mash and Dashing equally differentiate between Klebsiella species. PeerJ. 2022;10:e13784.
Scholz M, Ward DV, Pasolli E, Tolio T, Zolfo M, Asnicar F, et al. Strain-level microbial epidemiology and population genomics from shotgun metagenomics. Nat Methods. 2016;13:435–8.
Buchfink B, Reuter K, Drost H-G. Sensitive protein alignments at tree-of-life scale using DIAMOND. Nat Methods. 2021;18:366–8.
Dixon P. VEGAN, a package of R functions for community ecology. J Veg Sci. 2003;14:927–30.
McMurdie PJ, Holmes S. Shiny-phyloseq: Web application for interactive microbiome analysis with provenance tracking. Bioinforma Oxf Engl. 2015;31:282–3.
Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12:R60.
Hill C, Tancredi DJ, Cifelli CJ, Slavin JL, Gahche J, Marco ML, et al. Positive Health Outcomes Associated with Live Microbe Intake from Foods, Including Fermented Foods, Assessed using the NHANES Database. J Nutr. 2023;153:1143–9.
Zhang X-F, Qi Y, Zhang Y-P, Deng J-L, Chen X-L, Li R-N, et al. Fermented foods and metabolic outcomes in diabetes and prediabetes: A systematic review and meta-analysis of randomized controlled trials. Crit Rev Food Sci Nutr. 2023;1–18.
Jalili M, Nazari M, Magkos F. Fermented Foods in the Management of Obesity: Mechanisms of Action and Future Challenges. Int J Mol Sci. 2023;24:2665.
Jahn LJ, Rekdal VM, Sommer MOA. Microbial foods for improving human and planetary health. Cell. 2023;186:469–78.
Marco ML, Hill C, Hutkins R, Slavin J, Tancredi DJ, Merenstein D, et al. Should There Be a Recommended Daily Intake of Microbes? J Nutr. 2020;150:3061–7.
Leech J, Cabrera-Rubio R, Walsh AM, Macori G, Walsh CJ, Barton W, et al. Fermented-Food Metagenomics Reveals Substrate-Associated Differences in Taxonomy and Health-Associated and Antibiotic Resistance Determinants. mSystems. 2020;5:e00522-20.
Sato T, Yamada Y, Ohtani Y, Mitsui N, Murasawa H, Araki S. Production of menaquinone (vitamin K2)-7 by Bacillus subtilis. J Biosci Bioeng. 2001;91:16–20.
Gänzle MG. Food fermentations for improved digestibility of plant foods – an essential ex situ digestion step in agricultural societies? Curr Opin Food Sci. 2020;32:124–32.
Wastyk HC, Fragiadakis GK, Perelman D, Dahan D, Merrill BD, Yu FB, et al. Gut-microbiota-targeted diets modulate human immune status. Cell. 2021;184:4137-4153.e14.
Yang Y, Shi Y, Kerfahi D, Ogwu MC, Adams JM. Elevation-related climate trends dominate fungal co-occurrence patterns on Mt. Norikura, Japan. Cold Spring Harb Lab [Internet]. 2021 [cited 2024 Jul 4]; Available from: http://www.xueshufan.com/publication/3122293494
Wang Y, Xiang F, Zhang Z, Hou Q, Guo Z. Characterization of bacterial community and flavor differences of different types of Douchi. Food Sci Nutr. 2021;9:3460–9.
Li X, He Y, Yang W, Mu D, Zhang M, Dai Y, et al. Comparative analysis of the microbial community and nutritional quality of sufu. Food Sci Nutr. 2021;9:4117–26.
Semenec L, Cain AK, Dawson CJ, Liu Q, Dinh H, Lott H, et al. Cross-protection and cross-feeding between Klebsiella pneumoniae and Acinetobacter baumannii promotes their co-existence. Nat Commun. 2023;14:702.
Fedorova TV, Vasina DV, Begunova AV, Rozhkova IV, Raskoshnaya TA, Gabrielyan NI. Antagonistic Activity of Lactic Acid Bacteria Lactobacillus spp. against Clinical Isolates of Klebsiella pneumoniae. Appl Biochem Microbiol. 2018;54:277–87.
Tang H-J, Chen C-C, Lu Y-C, Huang H-L, Chen H-J, Chuang Y-C, et al. The effect of Lactobacillus with prebiotics on KPC-2-producing Klebsiella pneumoniae. Front Microbiol. 2022;13:1050247.
Djukovic A, Garzón MJ, Canlet C, Cabral V, Lalaoui R, García-Garcerá M, et al. Lactobacillus supports Clostridiales to restrict gut colonization by multidrug-resistant Enterobacteriaceae. Nat Commun. 2022;13:5617.
Ermolenko E, Gromova L, Borschev Y, Voeikova A, Karaseva A, Ermolenko K, et al. Influence of different probiotic lactic Acid bacteria on microbiota and metabolism of rats with dysbiosis. Biosci Microbiota Food Health. 2013;32:41–9.
Mäklin T, Thorpe HA, Pöntinen AK, Gladstone RA, Shao Y, Pesonen M, et al. Strong pathogen competition in neonatal gut colonisation. Nat Commun. 2022;13:7417.
Tan G, Hu M, Li X, Pan Z, Li M, Li L, et al. Metagenomics reveals the diversity and taxonomy of antibiotic resistance genes in sufu bacterial communities. Food Control. 2021;121:107641.
Lin Y-T, Siu LK, Lin J-C, Chen T-L, Tseng C-P, Yeh K-M, et al. Seroepidemiology of Klebsiella pneumoniae colonizing the intestinal tract of healthy Chinese and overseas Chinese adults in Asian countries. BMC Microbiol. 2012;12:13.
Shimasaki T, Seekatz A, Bassis C, Rhee Y, Yelin RD, Fogg L, et al. Increased Relative Abundance of Klebsiella pneumoniae Carbapenemase-producing Klebsiella pneumoniae Within the Gut Microbiota Is Associated With Risk of Bloodstream Infection in Long-term Acute Care Hospital Patients. Clin Infect Dis Off Publ Infect Dis Soc Am. 2019;68:2053–9.
Wu H, Tremaroli V, Schmidt C, Lundqvist A, Olsson LM, Krämer M, et al. The Gut Microbiota in Prediabetes and Diabetes: A Population-Based Cross-Sectional Study. Cell Metab. 2020;32:379-390.e3.
Tang WHW, Kitai T, Hazen SL. Gut Microbiota in Cardiovascular Health and Disease. Circ Res. 2017;120:1183–96.
Sepich-Poore GD, Zitvogel L, Straussman R, Hasty J, Wargo JA, Knight R. The microbiome and human cancer. Science. 2021;371:eabc4552.
Li Y, Fu S, Klein MS, Wang H. Traditionally fermented foods still as a critical avenue impacting host gut antibiotic resistome [Internet]. bioRxiv; 2023 [cited 2024 Jul 4]. p. 2023.04.21.537834.

No competing interests reported.

Additionalfile1.docx
Additional file1: Supplementary figures. Figure S1. (A) Venn diagram of eukaryotic (top) and bacterial (bottom) species. (B) Prevalence of eukaryotic species. (C) Relative abundance at Genus level. (D) Scatter plot between the microbiome distance and geographical distance in the three-type fermented soy products. Figure S2. SparCC network plot of co-abundance and co-exclusion correlations between species in DJ and FR; species were clustered into distinct groups using cluster_louvain. Nodes represent species involved in either significant co-abundance (red edges) or co-exclusion (blue edges) relationships; the magnitude of correlation expressed as the intensity of the respective edge colors. The color of each node indicates the Order of the species. Figure S3. Quality metrics of MAGs recovered from the fermented food bacterial community. (A) Scatter plot illustrating the distribution of the 707 recovered MAGs based on their completeness and contamination levels. (B) MAG count at genus and species level. Figure S4. Antibiotic resistance genes (ARGs) distribution in Enterobacterales species. Figure S5. Comparison of Ecological Resistance Group Risk (ERR) and Human Health Resistance Group Risk (HRR).
Additionalfile2.xlsx
Additional file2: Supplementary tables. Table S1 Group and location information of fermented soybean products. Table S2 Species enriched in fermented soybean foods.

Download PDF

Version 1

posted

You are reading this latest preprint version

Exploring the Microbiome of Fermented Soy Products: Implications for Gut Health in China

Status:

Version 1

Abstract

Figures

Introduction

Results

Microbial composition in three fermented soy products

Functional landscapes of microbiome in fermented soy products

Antagonistic activity of Lactobacillales or Bacillales against Enterobacterales

Recovery of fermented food bacteria genomes from metagenomes

Comparative analysis indicated Klebsiella strains in gut of Chinese individuals may originate from fermented foods

Materials and Methods

Sample information

Library preparation, metagenomic sequencing, and bioinformatic pipeline

Metagenomic assembled genomes (MAGs)

Network construction and clustering

Comparison of bacterial species between fermented food and Chinese gut microbiome

Statistical analysis

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1