Skip to main content
NCBI home page
Frontiers in Cellular and Infection Microbiology logoLink to Frontiers in Cellular and Infection Microbiology
. 2026 Jun 15;16:1828012. doi: 10.3389/fcimb.2026.1828012

Metagenomic analysis of human feces reveals gut microbiome role in colorectal cancer

Zehui Gu 1,†, Qi Tan 1,†, Dong Mao 2, Yun Zhang 3, Yadi Wang 4, Dongning He 5, Suxian Chen 1,*
PMCID: PMC13310921  PMID: 42376319

Abstract

Background

This study aimed to identify the microbiota and specific genes that are closely associated with colorectal cancer (CRC) through metagenomic sequencing and integrative multi-omics analysis.

Methods

Fecal samples were collected from 11 healthy volunteers and 20 patients with CRC. Genomic DNA was extracted for metagenomic analysis and high-throughput sequencing. Compositional differences and correlations of the gut microbiome were compared based on species and functional diversity.

Results

The overall species composition included 1,980 species, with 1,707 species identified in the CRC group and 1,525 in the healthy control group. Alpha diversity was significantly lower in the CRC group than in the healthy control group (p = 0.014). Beta diversity analysis revealed significant differences between the two groups (stress = 0.1308, p = 0.021). Based on LEfSe analysis, Shigella, Porphyromonas, Proteus, Bacteroides, Alistipes, Fusobacterium, and Escherichia were more abundant in patients with CRC, whereas Eubacterium, Clostridium, Dialister, Faecalibacterium, Blautia, Coprococcus, Dorea, Subdoligranulum, Megamonas, Roseburia, and Prevotella were significantly more abundant in the healthy control group (p < 0.05).

Conclusion

A multidimensional microbial diagnostic model, incorporating Shigella, Porphyromonas, Proteus, Bacteroides, Fusobacterium, Escherichia, Eubacterium, Clostridium, Dialister, Faecalibacterium, Blautia, Coprococcus, Dorea, Subdoligranulum, Megamonas, Roseburia, and Prevotella, suggests the potential to enhance early CRC screening performance. Furthermore, LptA, tnaA, envC, and argB may represent promising candidates for novel therapeutic targets, warranting further investigation.

Keywords: colorectal cancer, fecal biomarkers, human gut microbiome, metagenomics, early screening

1. Background

Colorectal cancer (CRC) is the third most diagnosed cancer globally and the second leading cause of cancer-related mortality, accounting for approximately 10% of all cancer cases worldwide (Capparelli et al., 2023; Sadri et al., 2024). Colonoscopy is the most effective diagnostic method; however, it requires extensive bowel preparation, involves a painful procedure, and carries risks such as iatrogenic colon perforation and mortality (Adebisi et al., 2024), leading to low acceptance among patients. Alternatively, the fecal occult blood test (FOBT; Hemoccult® Test) is a widely used non-invasive CRC screening method. However, its sensitivity and specificity are limited, rendering it less effective in promptly identifying precancerous lesions. Therefore, there is an urgent need to develop more accurate and non-invasive screening methods.

The human gut is a dynamic and constantly changing ecosystem hosting trillions of microbes with essential roles in digestion, immune response, and metabolic functions, all closely linked to health and disease. Certain bacteria, such as Bifidobacterium and Lactobacillus, can be more effective in treating diseases than expensive drugs (Shah et al., 2024), which often have side effects. Disruption of the gut microbiome can contribute to cancer progression through epithelial-mesenchymal transition, angiogenesis, and metastasis to distant organs (Sevcikova et al., 2024). This highlights the intricate relationship between the gut microbiome and the occurrence and development of CRC, offering substantial potential for advancements in diagnosis and treatment.

Microbial dysbiosis is associated with various diseases, including breast, lung, liver, and pancreatic cancers (Kaźmierczak-Siedlecka et al., 2021; Esposito et al., 2022; Chen et al., 2024; Hong et al., 2024). The microbiome influences cancer development by promoting inflammation, regulating the immune system, and producing carcinogenic compounds (Saraswat and Goel, 2024). Among dysbiotic microbes, Faecalibacterium and Proteus are potentially linked to the gut microbiome. Faecalibacterium, one of the primary butyrate-producing bacteria in the gut, exhibits anti-inflammatory and microbiota-modulating characteristics, making it effective for treating inflammatory bowel disease, Crohn’s disease, and CRC (Song et al., 2024). As a probiotic, Faecalibacterium significantly reduces the frequency and formation of abnormal crypt foci in azoxymethane-induced colon cancer in rats, demonstrating its anti-tumor and anti-proliferative effects (Dikeocha et al., 2022). Furthermore, a higher relative abundance of Faecalibacterium is associated with a healthy gut microbiome (Antonetti et al., 2024).

Proteus is an opportunistic pathogen that can cause gastrointestinal, urinary tract, and diabetic foot infections (Phiri et al., 2024). Proteus is significantly enriched in the feces of patients with stage III–IV CRC and may contribute to CRC development (Liu et al., 2023). Genes associated with gut microbiota also play a crucial role in CRC development. The Wnt/β-catenin, MAPK/ERK, PI3K/AKT/mTOR, TGF-β, Notch, and DNA mismatch repair pathways are key in CRC pathogenesis (Fu et al., 2024; Li et al., 2024; Monge et al., 2024; Su et al., 2024; Wang et al., 2024; Nakayama et al., 2025). Traditional methods of addressing microbial dysbiosis, such as probiotics (Xu et al., 2024), aim to regulate the gut microbiome, improve gut health, and positively affect CRC. However, these approaches often fail to provide long-term benefits, raising the question of whether regulating microbiome-related genes can address the root causes of microbial dysbiosis. Therefore, it is essential to identify the genes associated with dysbiotic microbiota.

Existing literature lacks studies that systematically examine patients with CRC, differential microbiota, and differential genes as an integrated system. This study was based on CRC screening of the Chinese population and employed fecal metagenomic analysis and high-throughput sequencing to evaluate the differences and connections between beneficial and pathogenic microbes, genes, and their correlations in samples from patients with CRC and healthy controls. The aim was to identify multiple differential microbes and genes and to analyze the correlations thoroughly between microbe-microbe and microbe-gene interactions to provide new clinical biomarkers and more precise reference strategies for diagnosing and treating CRC.

2. Materials and methods

2.1. Materials

2.1.1. Selection of study subjects

This study included patients who underwent endoscopic diagnosis and treatment at the Third Affiliated Hospital of Jinzhou Medical University, China. Based on pathological diagnostic results, patients with CRC were selected as the experimental group, whereas healthy volunteers were recruited as the control group. All participants met the inclusion and exclusion criteria outlined in Table 1, and their fecal samples were collected. Each participant signed an informed consent form, and the study was approved by the Ethics Committee of the Third Affiliated Hospital of Jinzhou Medical University. The ethics number is KX2022036.

Table 1.

Inclusion and exclusion criteria.

Grouping of study subjects
colorectal cancer patient group healthy control group
Inclusion criteria

  • - Age: 40–75 years old

  • - BMI: 18.5–30 kg/m2

  • - No history of other tumors

  • - Histopathological diagnosis

  • - Underwent radical resection surgery

  • - Age: 40–75 years old

  • - BMI: 18.5–30 kg/m2
Exclusion criteria

  • - BMI > 30 kg/m2

  • - Non-first visit cases

  • - Pregnancy

  • - Lactose intolerance

  • - Immunodeficiency

  • - Other gastrointestinal diseases (such as inflammatory bowel disease)

  • - Had taken antibiotics within 3 months before surgery

  • - Infection in other parts of the body

  • - Had taken probiotics or prebiotics within 2 weeks before surgery

  • - Received chemotherapy or radiotherapy before surgery

 BMI > 30 kg/m2

  • - Pregnancy

  • - Lactose intolerance

  • - Immunodeficiency

  • - Other gastrointestinal diseases (such as inflammatory bowel disease)

  • - Had taken antibiotics within 3 months before study

  • - Infection in other parts of the body

  • - Had taken probiotics or prebiotics within 2 weeks before study

  • - Tumors in other parts of the body
Open in a new tab

*lactose intolerance introduces a confounding factor, it makes it difficult to distinguish whether the observed outcomes are caused by the experimental intervention or by the abnormal fermentation of lactose in the gut.

2.1.2. Main reagents and instruments

The main instruments and reagents used were an Illumina HiSeq 2500 High-Throughput Sequencing System (Illumina, USA) and a Qiagen QIAmp DNA Stool Mini Kit (Qiagen, Germany).

2.2. Methods

2.2.1. Collection of fecal samples and DNA quantity detection

Thirty-one participants underwent standard colonoscopy at the Third Affiliated Hospital of Jinzhou Medical University, including 20 patients with CRC and 11 healthy individuals. All participants with CRC had colonic lesions at the time of fecal collection. Fecal samples were collected and stored at −20 °C within 4 hours and at −80 °C within 24 hours for long-term preservation. DNA extraction was performed using the Qiagen QIAmp DNA Stool Mini Kit, according to the manufacturer’s instructions.

2.2.2. Metagenomic analysis of gut microbiome

Raw data of 10 Gb per sample was adopted. After total DNA extraction and quality inspection of intestinal microbiota, qualified samples were sequenced on the Illumina HiSeq 2500 high-throughput sequencing platform with paired-end 150 bp sequencing mode. The obtained raw reads were subjected to quality control. FastQC was used to evaluate the quality of raw data, including the distribution of base quality (Q20/Q30), GC content and adapter contamination. Subsequently, data filtering was performed: Illumina sequencing adapter sequences were removed; bases with quality values below Q20 at both ends of reads were trimmed; sliding window scanning was applied to truncate sequences when the average quality fell below the threshold; short sequences shorter than 50 bp after trimming were discarded. High-quality clean reads were obtained for subsequent analysis. Although fecal samples are predominantly composed of bacteria, they may still contain a small amount of DNA shed from human epithelial cells. Host read removal was conducted to acquire purer microbial data. Bowtie2 was used to align qualified clean reads against the human reference genome. Reads mapped to the human genome were regarded as host contamination and eliminated, while unmapped reads were retained as non-host data for subsequent assembly and taxonomic annotation. Metagenomic assembly of clean reads was performed using MEGAHIT, and contigs shorter than 300 bp were filtered out. QUAST was used to evaluate the assembly results. MetaGeneMark (http://exon.gatech.edu/meta_gmhmmp.cgi, Version 3.26) was applied to identify coding regions in the genome. MMseqs2 (https://github.com/soedinglab/mmseqs2, Version 12-113e3) was used to remove redundant sequences. A non-redundant gene catalogue was constructed with a 95% similarity threshold and a 90% coverage threshold. Functional annotation and taxonomic analysis of the non-redundant gene catalogue were performed across multiple databases, and the composition and abundance of species in samples were statistically analyzed. All analyzes were completed based on BMKCloud (http://www.biocloud.net/). Multiple hypothesis testing correction was conducted to control the false positive rate in inter-group differential analyzes (differential species, differential genes and differential pathways). The Benjamini-Hochberg method was used for P-value correction in multiple comparisons, and differences with FDR<0.05 were considered statistically significant. During this phase, samples were excluded due to insufficient DNA purity, inadequate sequencing depth (insufficient valid reads), or poor reproducibility. Ultimately, 22 samples were retained for subsequent metagenomic analysis.

Species composition analysis was performed to determine the relative abundance of dominant species between groups. Alpha diversity analysis was performed to assess the species or functional richness and diversity between the two groups. Beta diversity analysis, including principal component analysis (PCA), principal coordinate analysis (PCoA), and non-metric multidimensional scaling (NMDS), was used to examine differences in species or functional dimensions between the two groups. LEfSe analysis was performed to identify biomarkers in species between the two groups, with a linear discriminant analysis score threshold of 3.0. Random forest analysis was used to rank important species features between the two groups. Additionally, the non-redundant gene set sequences were aligned with the Kyoto Encyclopedia of Genes and Genomes (KEGG) and the Evolutionary genealogy of genes: Non-supervised Orthologous Groups (eggNOG) databases for functional annotation and beta diversity analysis to identify differential functional annotations between the two groups. The corresponding tools from the Carbohydrate-active enzymes (CAZy) and the comprehensive Antibiotic Research Database (CARD) were used to compare and analyze differences in carbohydrate-active enzymes and antibiotic resistance, respectively.

2.3. Statistical analysis

Statistical analysis was performed using SPSS version 26.0 software. Categorical data were presented as frequencies (percentages). Group comparisons were performed using the T-test or Fisher’s exact probability test, and independent sample comparisons were performed using the two-tailed Wilcoxon rank-sum test (Mann–Whitney U-test). The alpha diversity of the bacterial communities was calculated using the R vegan package and represented using Simpson and Shannon indices. Similarity analysis between intestinal tissues and fecal samples was performed using the binary Jaccard and Bray–Curtis dissimilarity models. Species analysis was performed using analysis of similarities(ANOSIM) and permutational multivariate analysis of variance (PERMANOVA) functions with 1,000 permutations. Statistical significance was set at p < 0.05.

3. Results

3.1. Overall comparison of fecal microbiota between healthy individuals and patients with CRC

We assessed the gut microbiome in fecal samples from 22 subjects, consisting of 1,980 operational taxonomic units (OTUs), with 1,707 OTUs in the CRC group and 1,525 OTUs in the healthy control group, most of which were shared between the groups (Figure 1a). Using Shannon and Simpson indices to measure microbial richness and diversity levels, Alpha diversity analysis revealed that the CRC group had significantly lower Shannon and Simpson diversity indices than the control group (p = 0.014 and p = 0.0099, respectively; Figures 1b, c). This primarily highlights the core characteristics of microbial structural disorder and ecological imbalance. Beta diversity between the two groups was examined by mapping the distance differences between the two groups using PCA and PCoA, which showed significant clustering between the groups (R2 = 0.21, p = 0.001; R2 = 0.07, p = 0.021, Figures 1d, e). NMDS was used to compare beta diversity between the CRC and control groups, which revealed significant differences between the two groups in the NMDS1 and NMDS2 dimensions (stress = 0.1308, p = 0.021, Figure 1f). Furthermore, ANOSIM and PERMANOVA revealed significant differences between the fecal samples of the two groups (R = 0.345, p = 0.001; R2 = 0.258, p = 0.001, Figures 1g, h). These findings suggest that during CRC progression, microbial alterations are not merely local or subtle fluctuations. Instead, the microbiota undergoes a multidimensional transformation—a systemic ‘collapse’ followed by ‘remodeling’—This process presents a possibility for clinical diagnosis that warrants further exploration.

Figure 1.

Panel a shows a bar chart of set intersections and sizes for groups H and C. Panel b and c present box plots comparing Shannon and Simpson diversity indices, respectively, with statistically significant differences between groups. Panel d and e display PCA plots indicating group separation, color-coded by group. Panel f provides an NMDS plot with stress value. Panels g and h illustrate box plots for Bray-Curtis distance among, within H, and within C groups, with R-squared and p-values noted.

Open in a new tab

Overview of gut microbiome profiles in healthy control (H) and CRC (C) groups. (a) UpSet plot of species composition, with a side bar graph showing the number of species included in each group. The connections between dots represent overlaps, similar to a traditional Venn diagram. (b) Shannon diversity index difference analysis plots showed that the healthy control group (H) index was significantly higher than that of the CRC group (p = 0.014). (c) Simpson diversity index difference analysis plots showed that the healthy control group (H) index was significantly higher than that of the CRC group (p = 0.0099). (d) The species abundance PCA plot shows points representing individual samples, with different colors indicating different groups. The x-axis represents the first principal component and its contribution to the sample differences, whereas the y-axis represents the second principal component and its contribution to the sample differences (R2 = 0.21, p = 0.001). (e) The PCoA plot of species abundance shows the x- and y-axes as the two eigenvalues that contribute the most to the sample differences, which are expressed as percentages to indicate the degree of their influence (R2 = 0.07, p = 0.021). (f) The NMDS analysis plot of species shows differences in the overall fecal microbiota composition between the healthy control (H) and CRC groups (stress = 0.1308, p = 0.021). (g) Boxplot of inter-sample distances for species (ANOSIM analysis) indicates greater differences between groups than within groups (R = 0.345, p = 0.001). (h) The boxplot of inter-sample distances for species (PERMANOVA) provides an R2 value, which indicates the extent to which the different groupings explain the sample differences (R2 = 0.258, p = 0.001).

3.2. Fecal composition analysis in healthy individuals and patients with CRC

Taxonomic composition analysis of the fecal samples from the two groups revealed that the collected sequences were primarily composed of six phyla: Bacteroidetes, Firmicutes, Proteobacteria, Actinobacteria, Uroviricota, and Chlamydiae (Figure 2a). At the genus level, based on the compositional makeup of the samples in the two groups, the top 30 bacterial genera were more abundant (Figure 2b). A significance test for the top-ranked bacterial genera revealed significant differences between the groups, with Faecalibacterium, Prevotella, Roseburia, Eubacterium, Megamonas, and Proteus showing differences in patients with CRC compared to those in the healthy control group (p < 0.05, Figures 2c, d). High-dimensional comparisons using LEfSe indicated that patients with CRC had higher expression levels of Shigella (p = 0.025), Porphyromonas (p = 0.041), Proteus (p = 0.004), Bacteroides (p = 0.016), Alistipes (p = 0.012), Fusobacterium (p = 0.030), and Escherichia (p = 0.042), whereas the healthy control group had higher expression levels of Eubacterium (p = 0.002), Clostridium (p < 0.005), Dialister (p = 0.033), Faecalibacterium (p = 9.36E−05), Blautia (p = 0.010), Coprococcus (p = 0.012), Dorea (p = 0.008), Subdoligranulum (p = 0.035), Megamonas (p < 0.005), Roseburia (p = 0.002), and Prevotella (p = 0.018) (Figure 2e). The differentiation of dominant genera may suggest not only quantitative changes in bacterial abundance, but also potentially indicate a systemic shift within the gut microbiome.

Figure 2.

Panel a: Stacked bar chart showing relative abundances of microbial taxa by sample, with each color representing a different taxon. Panel b: Stacked bar chart comparing taxonomic composition between groups H and C. Panel c: Heatmap with hierarchical clustering depicting abundance values for specific taxa across samples, where color intensity reflects abundance. Panel d: Horizontal bar plot comparing mean proportions of taxa between groups H and C, with statistical significance indicated. Panel e: Circular cladogram showing phylogenetic relationships and group-enriched taxa, with an adjacent legend linking taxa to color-coded group labels.

Open in a new tab

Fecal composition analysis of healthy control (H) and CRC (C) groups. (a) Relative abundance plot of bacterial phyla in fecal samples from groups H and C. (b) Abundance plot of the top 30 bacterial genera with the highest relative expression levels in each group of samples. (c) A heatmap of differential species abundance; the left dendrogram shows the hierarchical clustering of species, the upper dendrogram displays the hierarchical clustering of samples, and the central heatmap illustrates the abundance of species. (d) The histogram of differential species abundance shows the mean abundance of species on the vertical axis. The top 50 species were ranked by p-values in descending order and sorted by abundance from high to low. Significant differences were observed between the two groups for the bacterial genera, including Faecalibacterium, Prevotella, Roseburia, Eubacterium, Megamonas, and Proteus. (e) LEfSe evolutionary branching diagram with the circles radiating outward, representing the taxonomic levels from kingdom to species. The diameter of the small circle is proportional to the relative abundance; different colors represent different groups. Nodes of different colors indicate microbes that play an important role within the respective groups.

3.3. Species differences and correlation between healthy individuals and patients with CRC

The random forest algorithm was employed to construct multiple decision trees to classify samples and identify microbial species that significantly contributed to the differences between samples. The resulting species importance ranking plot showed that Faecalibacterium emerged as a promising candidate feature species in this preliminary analysis. (Figure 3a). Given the small sample size, the model parameters were directly optimized using 10-fold cross-validation to enhance the accuracy of the model. Receiver operating characteristic curve analysis showed that Bacteroides (area under the curve [AUC] = 0.795), Shigella (AUC = 0.812), Fusobacterium (AUC = 0.778), Proteus (AUC = 0.872), Escherichia (AUC = 0.786), Eubacterium (AUC = 0.906), Clostridium (AUC = 0.821), Faecalibacterium (AUC = 1.000), Blautia (AUC = 0.829), Coprococcus (AUC = 0.803), Megamonas (AUC = 0.957), Roseburia (AUC = 0.957), and Prevotella (AUC = 0.863). These microorganisms demonstrate the potential to predict CRC diagnosis. (Figures 3b–m).

Figure 3.

Panel (a) shows a dot plot ranking bacterial genera by mean decrease Gini importance for a classification model. Panels (b–m) contain receiver operating characteristic (ROC) curves with area under the curve (AUC) values labeled at the center of each plot. Panel (n) is a network diagram depicting correlations among microbial taxa, where nodes vary in size based on abundance and edges are colored by correlation strength.

Open in a new tab

Species differences and correlations between the groups. (a) Random forest analysis of species importance ranking plot; the x-axis represents the species importance measure. Higher values indicate a greater decrease in classification accuracy when the species is removed, highlighting its importance in sample classification. The y-axis represents the species names ranked by importance. (b–m) Receiver operating characteristic curve, with the performance evaluated based on the area under the curve. (n) Species correlation network diagram, where circles represent species, and their size is proportional to species abundance. Lines connecting the circles represent correlations between species, and the thickness of the lines indicates the strength of the correlation. Line colors denote the type of correlation: red represents a positive correlation; green represents a negative correlation.

To explore microbial associations, we selected 80 species with the highest abundance and performed correlation analysis based on their abundance and variation in the samples. Using the Spearman algorithm (including positive and negative correlations) and statistical tests, data sets with correlations greater than 0.5 and p < 0.05 were selected. A correlation network diagram was generated using R. Significant positive correlations were observed between the following species: Alistipes, Bucyricimonas, and Odoribacter; Shigella and Escherichia; Eubacterium, Faecalibacterium, Blautia, Roseburia, and Lachnospira; Clostridium and Coprococcus; Blautia and Dorea; Faecalibacterium, Roseburia, and Megamonas; Megamonas and Roseburia; and Roseburia and Prevotella (p < 0.005, Figure 3n). These findings unveil the intricate ‘social network’ and ‘functional alliances’ within the gut microecosystem. They suggest that therapeutic interventions should extend beyond targeting single species to focusing on the holistic status of these ‘microbial modules.’ Restoring symbiotic relationships may hold greater promise than single-bacteria supplementation, which could offer valuable insights for clinical therapeutic strategies.

3.4. Overall comparison of functional changes in fecal microbiota associated with CRC

To characterize microbial gene functions and the differences between patients with CRC and non-tumor participants, we quantified the relative abundances of prokaryotic genes using KEGG, eggNOG, the pathogen-host interaction database (PHI-base), and metagenomic gene modules in the study samples. For beta diversity analysis, differences between the two groups were mapped using PCA and PCoA analyzes. Significant clustering was observed between groups (R2 = 0.18, p = 0.001; R2 = 0.12, p = 0.001; Figures 4a, b). NMDS analysis of beta distances between the two groups revealed significant differences in NMDS1 and NMDS2 dimensions (stress = 0.1181, p = 0.001, Figure 4c). Moreover, ANOSIM and PERMANOVA analyzes indicated significant differences between the two groups in fecal samples (R = 0.307, p = 0.003; R2 = 0.178, p = 0.012, Figures 4d, e). To investigate the carbohydrate utilization preferences of the microbiota and functional genes associated with antibiotic resistance, the metagenome was annotated using prokaryotic carbohydrate-active enzyme families from the CAZy database and antibiotic resistance families from the CARD database. The results showed significant differences between the two groups, with 42 carbohydrate-active enzymes, 38 antibiotic resistance ontologies, and nine antibiotic resistance genes that were significantly abundant in patients with CRC (p < 0.005, Figures 4f–h). In summary, these preliminary findings suggest that the intestinal environment of patients with colorectal cancer (CRC) may undergo systemic functional alterations. These observations could provide preliminary clues for subsequent exploration of its pathogenic mechanisms and the development of precision therapies, highlighting the potential exploratory value of this research in translational medicine.

Figure 4.

Panel a shows a PCA scatter plot comparing two groups H and C, with distinct clustering and variance explained. Panel b shows a second PCA plot, also distinguishing groups. Panel c displays an NMDS plot with group separation. Panel d and panel e show boxplots comparing distances or scores for categories Between, H, and C, highlighting statistical differences. Panels f and g display hierarchical clustered heatmaps of standardized values for features by sample, with treatment groups annotated. Panel h contains a circular chord diagram illustrating interactions or associations between categories labeled around the circumference.

Open in a new tab

Functional gut microbiome changes in healthy control (H) and CRC (C) groups. (a) PCA analysis of functional genes (R2 = 0.18, p = 0.001). (b) PCoA analysis of functional gene abundance (R2 = 0.12, p = 0.001). (c) NMDS analysis of functional genes (stress = 0.1181, p = 0.001). (d) ANOSIM plot of inter-sample distances for functional genes shows that the analysis indicates greater differences between groups than within groups (R = 0.307, p = 0.003). (e) PERMANOVA analysis of inter-sample distances for functional genes (R2 = 0.178, p = 0.012). (f) Heatmap of differential carbohydrate-active enzyme abundance showing the clustering tree of differential enzymes on the left, the sample clustering tree at the top, and the heatmap in the center. (g) Heatmap of the differential abundance of antibiotic resistance ontology. (h) Circle plot of CARD for detecting the antibiotic resistance gene composition, with the outer right half-circle representing the samples and the left half-circle representing the annotated resistance gene types. Scale represents the proportion of abundance. The inner ribbons connect the resistance genes and samples, revealing the functional composition and distribution of resistance genes in the samples. The width of the ribbons represents the proportional distribution.

3.5. Differences in the microbial-related metabolic composition between fecal samples of patients with CRC and healthy individuals

Gene ontology (GO) is a database that defines and describes gene and protein functions. Based on the GO functional annotations, the top enriched GO terms were primarily related to biological processes (e.g., metabolic process, cellular process, localization, and biological regulation), cellular components (e.g., membrane, cell, and membrane part), and molecular functions (e.g., catalytic activity, binding, and transporter activity) (Figure 5a). KEGG is a comprehensive database that collects genomic, pathway, and compound data from various organisms. Orthologous protein groups were formed by clustering sequences using KEGG. The most abundant enriched pathways in the KEGG orthologous protein groups were associated with global and overview maps, carbohydrate metabolism, nucleotide metabolism, amino acid metabolism, replication and repair, membrane transport, and translation (Figure 5b).

Figure 5.

Panel a shows a clustered bar graph comparing protein groups by functional category with colored bars. Panel b presents a horizontal bar chart of KEGG pathway levels associated with metabolism and cellular processes. Panel c features a horizontal bar graph comparing the mean proportion of various metabolic pathways between groups C and H, with statistical significance indicated. Panel d displays a horizontal bar chart comparing functional categories related to cellular processes and their proportions in groups C and H, with p-values listed. Panel e shows a horizontal bar chart of PHI categories and their proportions for groups C and H, alongside significance values.

Open in a new tab

Differential metabolic composition between groups. (a) Statistical plot of GO secondary node annotation classification, with the horizontal axis representing the GO category content and the vertical axis (on the left) showing the number of genes. (b) Statistical plot of KEGG metabolic pathway-related functional genes at level 2, with the horizontal axis showing the relative abundance of the corresponding functional genes and the vertical axis representing the classification content of KEGG levels 1 and 2. (c) Histogram of the KEGG pathway (level 3) showing the differential functional genes, with the leftmost axis indicating the mean functional abundance, the horizontal axis representing the mean proportion, and the vertical axis representing the functional names. (d) eggNOG (class-level) histogram shows the differential functional genes between the groups. (e) Histogram of PHI-base demonstrating the differential functional genes between the groups, where * indicates p < 0.05, ** indicates p ≤ 0.001, and *** indicates p ≤ 0.0001. The p-values are displayed on the rightmost side, with the data ordered by descending p-value (top 15 selected) and then by abundance, from high to low.

Intergroup differences between the control and CRC groups were analyzed based on the KEGG pathway (Level 3). Significant differences were observed in pathways such as amino acid biosynthesis, pyruvate metabolism, cysteine and methionine metabolism, 2-oxoacid metabolism, hydroxymethylglutaryl-CoA metabolism, and diphenylmethane ester metabolism (p < 0.05, Figure 5c). Additionally, the eggNOG database contains orthologous gene clusters from organisms, with proteins in each cluster assumed to originate from an ancestral protein with the same function. According to eggNOG annotations (class level), significant differences were found between the healthy control and CRC groups in processes such as cell wall/membrane/envelope biogenesis, inorganic ion transport and metabolism, energy production and conversion, signal transduction mechanisms, defense mechanisms, cell cycle control, cell division, chromosome partitioning, and the biosynthesis, transport, and degradation of secondary metabolites (p < 0.05, Figure 5d). The PHI-base includes experimentally validated gene data, primarily from fungi, oomycetes, and bacterial pathogens. The infected hosts include animals, plants, fungi, and insects. Based on PHI-base annotations, significant differences were observed between the two groups in genes such as PHI:6520 (LpdA), PHI:4613 (esaN), PHI:7990 (LysX), PHI:7577 (cstA), PHI:3913 (SpMsrAB2), PHI:3976 (MoIlv6), PHI:4910 (LptA), PHI:7740 (graS), PHI:4074 (wecA), PHI:7966 (tnaA), PHI:2975 (ipx10), PHI:5573 (envC), PHI:2322 (SidH), PHI:2633 (argB), and PHI:2570 (CYB2) (p < 0.05, Figure 5e), with LptA, tnaA, envC, and argB being closely related to Escherichia coli. In summary, this approach demonstrates potential for clinical diagnosis. It offers new insights into exploring the pathogenic mechanisms of colorectal cancer (CRC) and developing precision therapies, highlighting its exploratory significance in translational medicine research.

4. Discussion

CRC is one of the malignancies with the highest incidence and mortality rate worldwide. Commonly used early screening methods for CRC include FOBT, colonoscopy, and colon computed tomography imaging. However, each screening method has limitations. With advances in microbial community research, human fecal metagenomic sequencing has emerged as a sophisticated, non-invasive method for early CRC screening. This approach directly reflects the ecological status of the gut and aids in exploring the pathogenesis of various chronic diseases, including cancer. Research on CRC has yet to identify definitive carcinogenic pathogens similar to those found in gastric, cervical, and liver cancers. Abnormal indicators of CRC from species and functional perspectives require elucidation to identify precise diagnostic and therapeutic biomarkers. Fecal biomarker detection is expected to enhance the convenience, sensitivity, and clinical practicality of screening and improve patient compliance.

Increasing evidence (Wang et al., 2023) suggests that the gut microbiome is closely related to the onset and progression of CRC. Our metagenomic sequencing data suggest potential differences in the dominant and subordinate genera between the two groups. These findings are consistent with those of previous studies by Cao et al. and Kushkevych et al (Cao et al., 2021; Kushkevych et al., 2024), indicating that the gut microbiome associated with CRC has a higher species richness and a greater abundance of pathogenic microbes. However, these studies primarily focused on ulcerative colitis or rat models, whereas our study involved individuals with colonic lesions, providing a more representative sample. In addition, correlation analysis was performed on highly abundant species to identify significant associations within the abnormal gut microbiome. We observed that, from a biological perspective, these bacteria likely share similar ecological niches, exhibit metabolic synergism, or are subject to common regulation by the host’s diet and immune status. A high-fat diet can influence Alistipes and Odoribacter, leading to the production of novel metabolic products, including sulfonate compounds, which are closely linked to inflammatory bowel disease through the Toll-like receptor 4 (TLR4) signaling pathway (Walker et al., 2017; Older et al., 2023). Shigella and Escherichia are key microbes for food safety and clinical diagnostics. Genomic deletions can enhance the virulence of Shigella spp. and enteroinvasive Escherichia coli, increasing their pathogenicity (Maurelli et al., 1998; Liu et al., 2024). Eubacterium, Faecalibacterium, Blautia, Roseburia, and others can produce short-chain fatty acids (such as butyrate), which play important roles in immune regulation and preventing early CRC (Vacca et al., 2020; Gao et al., 2022). Changes in the gut microbiome of Clostridium sp. BR31 and Coprococcus 1 strengthen the relationship between dietary imbalance and gastrointestinal symptoms in children with autism spectrum disorder (Li et al., 2023). The correlation between Faecalibacterium and Megamonas in the gut microbiome is complex. A study in Henan showed a positive correlation between Faecalibacterium and Megamonas species. Other studies (Lai et al., 2024) have also suggested that various factors, including dietary habits, living environment, and genetic factors, influence the relative abundance of Faecalibacterium and Megamonas in the gut microbiome. These studies are consistent with our findings, although correlations between some microbiota have not been documented in the literature. It is undeniable that Faecalibacterium plays a significant role within the gut microbiota. However, while we employed rigorous cross-validation strategies to assess model performance, the AUC of 1.000 obtained in the context of a limited sample size suggests a potential risk of overfitting. The absence of an independent test set implies that the model’s generalizability has not yet been substantiated. Therefore, caution should be exercised when interpreting the potential of Faecalibacterium as a biomarker, and future studies should focus on validating this finding in independent cohorts.

Changes at the genetic level may influence microbial alterations and immune status in CRC. We identified genes associated with the gut microbiome, these genes exhibit varying degrees of association with the gut microbiome. For example, LptA, tnaA, envC, and argB are closely linked in E. coli. LptA is a periplasmic lipid A-binding protein involved in the lipopolysaccharide export pathway (Tran et al., 2008). Disruption of the LptA/LptC interaction inhibits the growth of Gram-negative bacteria. Moreover, tnaA encodes tryptophanase, an enzyme widely found in gram-negative bacteria but is present in only a few gram-positive bacteria (Boya et al., 2021). Transcription and translation occur simultaneously in E. coli, a prokaryote without a membrane-bound nucleus, resulting in a high demand for tryptophan. Enhancing key genes, such as trpR, tnaA, aroG, and trpED, increases tryptophan production. In addition, envC encodes a cell wall hydrolase activator in E. coli.

In Gram-negative bacteria, EnvC promotes cell separation and division by activating cell wall hydrolases via interactions with the FtsEX complex (Xu et al., 2023). argB encodes acetylglutamate kinase in E. coli, which catalyzes the second step in the biosynthesis of ornithine and arginine. Ornithine and arginine may help maintain the gut microbial balance and reduce the growth and infection of harmful bacteria such as E. coli. Recent studies have revealed that colorectal cancer (CRC) cells universally exhibit a phenotype of “arginine auxotrophy.” This metabolic vulnerability stems from the downregulation of argininosuccinate synthase (the mammalian homolog of the bacterial argB gene), which renders the cells incapable of de novo arginine biosynthesis (Xu et al., 2025). Consequently, CRC cells appear to become dependent on exogenous arginine uptake, an observation that further supports the potential biological characteristics associated with argB deficiency. However, no study has directly discussed the relationship between argB and E. coli. Our findings suggest that LptA, tnaA, and envC levels are elevated in CRC, whereas argB levels are reduced. This observation has not been previously reported. The current treatment options for CRC include endoscopic resection, surgery, chemotherapy, radiotherapy, targeted therapy, and immunotherapy. Given the close relationship between microbes and genes, genetic intervention may emerge as a promising new therapeutic approach.

In summary, human fecal metagenomic analysis and high-throughput sequencing through an integrative multi-omic analysis were employed to identify the microbiota and specific genes closely associated with CRC, enabling a comprehensive study. Significant differences were observed between the healthy control and CRC groups, and microbe-to-microbe and microbe-to-gene correlations were explored. We have initial constructed a multidimensional microbial diagnostic model incorporating Shigella, Porphyromonas, Proteus, Bacteroides, Alistipes, Fusobacterium, Escherichia, Eubacterium, Clostridium, Dialister, Faecalibacterium, Blautia, Coprococcus, Dorea, Subdoligranulum, Megamonas, Roseburia, and Prevotella. This model significantly enhances the sensitivity and specificity of early CRC screening. Furthermore, our preliminary exploration identifies LptA, tnaA, envC, and argB as promising novel therapeutic targets, which also hold potential as predictive biomarkers for CRC. This study highlights the advantages of combining microbiome research with CRC screening and treatment, offering new perspectives for identifying key carcinogenic pathogens in CRC. From the perspective of tumor staging, all enrolled CRC patients were classified as Stage I or II. This characteristic suggests that the constructed composite diagnostic model, along with the preliminary gene targets identified, holds potential clinical value for the early screening of colorectal cancer. However, this study also has limitations that limit the depth of the research. A small number of cases may not allow for more detailed studies such as stratified analysis. For example, it is difficult to stratify patients according to the severity of the disease and study treatment effects, making it difficult to comprehensively understand the relationship between the disease and treatment methods. Future studies will be dedicated to expanding the sample cohort and refining subject stratification. We aim to extend our scope beyond CRC patients to include individuals with enteritis, intestinal polyps, and tubular adenomas, thereby systematically characterizing the gut microbiota landscape across different stages of disease progression. Concurrently, we will investigate the expression profiles of Escherichia coli-associated genes to gain deeper insights into the mechanisms underlying disease pathogenesis. With advances in healthcare, new screening technologies and treatment methods will provide patients with more options and better treatment outcomes. However, challenges remain regarding the prevention and treatment of CRC. Therefore, there is a need to improve early diagnosis rates and strengthen public health education to increase awareness and prevent CRC.

Acknowledgments

Thank you to my colleagues in the laboratory, the polishing teacher, and all the personnel involved in this project for their tremendous technical assistance.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. The present study was supported by Jinzhou Medical University Internal Horizontal Project (Project No. 205220027).

Edited by: Cuncong Zhong, University of Kansas, United States

Reviewed by: Soumyadeep Chakraborty, University of South Florida, United States

Tamara Alejandra Piñero, CONICET Instituto de Medicina Traslacional e Ingeniería Biomédica (IMTIB), Argentina

Abbreviations: AUC, Area Under the Curve; ANOSIM, Analysis of Similarities; CAZy, Carbohydrate-Active Enzymes Database; CARD, Comprehensive Antibiotic Resistance Database; CRC, Colorectal Cancer; DNA, Deoxyribonucleic Acid; envC, Gene encoding an Escherichia coli cell wall hydrolase activator; FOBT, Fecal Occult Blood Test; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; LptA, Gene encoding a periplasmic lipid A-binding protein; MAPK/ERK, Mitogen-Activated Protein Kinase/Extracellular Signal-Regulated Kinase; MEGAHIT, A software tool for metagenomic assembly; MMseqs2, Many-against-Many sequence searching software; NMDS, Non-Metric Multidimensional Scaling; OUT, Operational Taxonomic Unit; PCA, Principal Component Analysis; PCoA, Principal Coordinate Analysis; PERMANOVA, Permutational Multivariate Analysis of Variance; PHI-base, Pathogen-Host Interaction Database; PI3K/AKT/mTOR, Phosphoinositide 3-Kinase/Protein Kinase B/Mammalian Target of Rapamycin; QUAST, Quality Assessment Tool for Genome Assemblies; ROC, Receiver Operating Characteristic; SPSS, Statistical Package for the Social Sciences; tnaA, Gene encoding tryptophanase; TLR4, Toll-Like Receptor 4.

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

Ethics statement

It approved by the Ethics Committee of the Third Affiliated Hospital of Jinzhou Medical University (approval No. KX2022036). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Author contributions

ZG: Data curation, Writing – original draft, Writing – review & editing. QT: Software, Writing – original draft, Writing – review & editing. DM: Investigation, Writing – review & editing. YZ: Investigation, Writing – review & editing. YW: Methodology, Writing – review & editing. DH: Supervision, Writing – review & editing. SC: Conceptualization, Funding acquisition, Writing – review & editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

  1. Adebisi A. A., Onobun D. E., Adediran A., Ononye R. N., Ojo E. O., Oluyi A., et al. (2024). Evaluation of morbidity and mortality in iatrogenic colonic perforation during colonoscopy: a comprehensive systematic review and meta-analysis. Cureus 16, e73302. doi:  10.7759/cureus.73302 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Antonetti L., Berrilli F., Di Cristanziano V., Farowski F., Daeumer M., Eberhardt K. A., et al. (2024). Investigation of gut microbiota composition in humans carrying blastocystis subtypes 1 and 2 and Entamoeba hartmanni. Gut Pathog. 16, 72. doi:  10.1186/s13099-024-00661-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Boya B. R., Kumar P., Lee J. H., Lee J. (2021). Diversity of the tryptophanase gene and its evolutionary implications in living organisms. Microorganisms 9, 2156. doi:  10.3390/microorganisms9102156 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cao H., Zong C., Dai W., Gao Q., Li D., Wu X., et al. (2021). The effects of Chinese medicine QRD, antibiotics, and probiotics on therapy and gut microbiota in septic rats. Front. Cell. Infect. Microbiol. 11, 712028. doi:  10.3389/fcimb.2021.712028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Capparelli R., Cuomo P., Gentile A., Iannelli D. (2023). Microbiota-liver diseases interactions. Int. J. Mol. Sci. 24, 3883. doi:  10.3390/ijms24043883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen P., Yang C., Ren K., Xu M., Pan C., Ye X., et al. (2024). Modulation of gut microbiota by probiotics to improve the efficacy of immunotherapy in hepatocellular carcinoma. Front. Immunol. 15, 1504948. doi:  10.3389/fimmu.2024.1504948 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dikeocha I. J., Al-Kabsi A. M., Chiu H. T., Alshawsh M. A. (2022). Faecalibacterium prausnitzii ameliorates colorectal tumorigenesis and suppresses proliferation of HCT116 colorectal cancer cells. Biomedicines 10, 1128. doi:  10.3390/biomedicines10051128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Esposito M. V., Fosso B., Nunziato M., Casaburi G., D'Argenio V., Calabrese A., et al. (2022). Microbiome composition indicate dysbiosis and lower richness in tumor breast tissues compared to healthy adjacent paired tissue, within the same women. BMC Cancer 22, 30. doi:  10.1186/s12885-021-09074-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fu J., Zhou L., Li S., Zheng J., Hou Z., He P. (2024). Let-7c-5p down regulates the proliferation of colorectal cancer through the MAPK-ERK-signaling pathway. Biochem. Genet. 62, 3231–3243. doi:  10.1007/s10528-023-10581-9 [DOI] [PubMed] [Google Scholar]
  10. Gao R., Wu C., Zhu Y., Kong C., Zhu Y., Gao Y., et al. (2022). Integrated analysis of colorectal cancer reveals cross-cohort gut microbial signatures and associated serum metabolites. Gastroenterology 163, 1024–1037.e9. doi:  10.1053/j.gastro.2022.06.069 [DOI] [PubMed] [Google Scholar]
  11. Hong R., Lin S., Zhang S., Yi Y., Li L., Yang H., et al. (2024). Pathogen spectrum and microbiome in lower respiratory tract of patients with different pulmonary diseases based on metagenomic next-generation sequencing. Front. Cell. Infect. Microbiol. 14, 1320831. doi:  10.3389/fcimb.2024.1320831 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Kaźmierczak-Siedlecka K., Stachowska E., Folwarski M., Przewłócka K., Makarewicz W., Bryl E. (2021). The potential of gut microbiome as a non-invasive predictive biomarker for early detection of pancreatic cancer and hepatocellular carcinoma. Eur. Rev. Med. Pharmacol. Sci. 25, 7275–7284. doi:  10.26355/eurrev_202112_27421 [DOI] [PubMed] [Google Scholar]
  13. Kushkevych I., Martínková K., Mráková L., Giudici F., Baldi S., Novak D., et al. (2024). Comparison of microbial communities and the profile of sulfate-reducing bacteria in patients with ulcerative colitis and their association with bowel diseases: a pilot study. Microb. Cell. 11, 79–89. doi:  10.15698/mic2024.03.817 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lai J., Gong L., Liu Y., Zhang X., Liu W., Han M., et al. (2024). Associations between gut microbiota and osteoporosis or osteopenia in a cohort of Chinese Han youth. Sci. Rep. 14, 20948. doi:  10.1038/s41598-024-71731-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Li H., Liu C., Huang S., Wang X., Cao M., Gu T., et al. (2023). Multi-omics analyses demonstrate the modulating role of gut microbiota on the associations of unbalanced dietary intake with gastrointestinal symptoms in children with autism spectrum disorder. Gut Microbes 15, 2281350. doi:  10.1080/19490976.2023.2281350 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Li Z., Ke H., Cai J., Ye S., Huang J., Zhang C., et al. (2024). MTHFD1 regulates autophagy to promote growth and metastasis in colorectal cancer via the PI3K-AKT-mTOR signaling pathway. Cancer Med. 13, e70267. doi:  10.1002/cam4.70267 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Liu J., Huang X., Chen C., Wang Z., Huang Z., Qin M., et al. (2023). Identification of colorectal cancer progression-associated intestinal microbiome and predictive signature construction. J. Transl. Med. 21, 373. doi:  10.1186/s12967-023-04119-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Liu S., Ma J., He F. (2024). A new SPQC biosensor for the detection of a new target of Escherichia/Shigella genera based on a novel method of synthesizing long-range DNA. Anal. Chem. 96, 9826–9833. doi:  10.1021/acs.analchem.4c00071 [DOI] [PubMed] [Google Scholar]
  19. Maurelli A. T., Fernández R. E., Bloch C. A., Rode C. K., Fasano A. (1998). Black holes" and bacterial pathogenicity: a large genomic deletion that enhances the virulence of Shigella spp. and enteroinvasive Escherichia coli. Proc. Natl. Acad. Sci. U.S.A. 95, 3943–3948. doi:  10.1073/pnas.95.7.3943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Monge C., Waldrup B., Carranza F. G., Velazquez-Villarreal E. (2024). WNT and TGF-beta pathway alterations in early-onset colorectal cancer among Hispanic/Latino populations. Cancers (Basel) 16, 3903. doi:  10.3390/cancers16233903 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Nakayama M., Saito H., Murakami K., Oshima H., Oshima M. (2025). Missense mutant p53 transactivates Wnt/β-catenin signaling in neighboring p53-destabilized cells through the COX-2/PGE2 pathway. Cancer Res. Commun. 5, 13–23. doi:  10.1158/2767-9764.CRC-24-0471 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Older E. A., Zhang J., Ferris Z. E., Xue D., Zhong Z., Mitchell M. K., et al. (2023). Biosynthetic enzyme-guided disease correlation connects gut microbial metabolites sulfonolipids to inflammatory bowel disease involving TLR4 signaling. Nat Commun. 15 (1), 9371. doi:  10.1038/s41467-024-53670-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Phiri R. M., Ertuğrul M. B., Bozdoğan B., Hoşbul T. (2024). Evaluation of virulence genes in Proteus strains isolated from diabetic foot infections and urinary tract infections. J. Infect. Dev. Ctries 18, 1559–1565. doi:  10.3855/jidc.18928 [DOI] [PubMed] [Google Scholar]
  24. Sadri S., Aghajani A., Soleimani H., Ghorbani Kalkhajeh S., Nazari H., Brouki Milan P., et al. (2024). Exploring the role of the TGF-β signaling pathway in colorectal precancerous polyps biochemical genetics. Biochem. Genet. 63, 1116–1148. doi:  10.1007/s10528-024-10988-y [DOI] [PubMed] [Google Scholar]
  25. Saraswat I., Goel A. (2024). Therapeutic modulation of the microbiome in oncology: current trends and future directions. Curr. Pharm. Bio/Technol. 26, 680–699. doi:  10.2174/0113892010353600241109132441 [DOI] [PubMed] [Google Scholar]
  26. Sevcikova A., Martiniakova M., Omelka R., Stevurkova V., Ciernikova S. (2024). The link between the gut microbiome and bone metastasis. Int. J. Mol. Sci. 25, 12086. doi:  10.3390/ijms252212086 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Shah A. B., Baiseitova A., Zahoor M., Ahmad I., Ikram M., Bakhsh A., et al. (2024). Probiotic significance of Lactobacillus strains: a comprehensive review on health impacts, research gaps, and future prospects. Gut Microbes 16, 2431643. doi:  10.1080/19490976.2024.2431643 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Song Q., Gao Y., Liu K., Tang Y., Man Y., Wu H. (2024). Gut microbial and metabolomics profiles reveal the potential mechanism of fecal microbiota transplantation in modulating the progression of colitis-associated colorectal cancer in mice. J. Transl. Med. 22, 1028. doi:  10.1186/s12967-024-05786-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Su X., Wang X., Lai J., Mao S., Li H. (2024). Unraveling a novel hippo-associated immunological prognostic signature: the contribution of SERPINE1 in facilitating colorectal cancer progression via the notch signaling pathway. Genomics 116, 110794. doi:  10.1016/j.ygeno.2024.110794 [DOI] [PubMed] [Google Scholar]
  30. Tran A. X., Trent M. S., Whitfield C. (2008). The LptA protein of Escherichia coli is a periplasmic lipid A-binding protein involved in the lipopolysaccharide export pathway. J. Biol. Chem. 283, 20342–20349. doi:  10.1074/jbc.M802503200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Vacca M., Celano G., Calabrese F. M., Portincasa P., Gobbetti M., De Angelis M. (2020). The controversial role of human gut Lachnospiraceae. Microorganisms 8, 573. doi:  10.3390/microorganisms8040573 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Walker A., Pfitzner B., Harir M., Schaubeck M., Calasan J., Heinzmann S. S., et al. (2017). Sulfonolipids as novel metabolite markers of Alistipes and Odoribacter affected by high-fat diets. Sci. Rep. 7, 11047. doi:  10.1038/s41598-017-10369-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wang P., Huang Q., Zhu Y., Chen L., Ye K. (2024). Fusobacterium nucleatum promotes microsatellite instability in colorectal carcinoma through up-regulation of miRNA-155-5p-targeted inhibition of MSH6 via the TLR4/NF-κB signaling pathway. Adv. Biol. (Weinh) 8, e2400293. doi:  10.1002/adbi.202400293 [DOI] [PubMed] [Google Scholar]
  34. Wang Z., Dan W., Zhang N., Fang J., Yang Y. (2023). Colorectal cancer and gut microbiota studies in China. Gut Microbes 15, 2236364. doi:  10.1080/19490976.2023.2236364 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Xu X., Li J., Chua W. Z., Pages M. A., Shi J., Hermoso J. A., et al. (2023). Mechanistic insights into the regulation of cell wall hydrolysis by FtsEX and EnvC at the bacterial division site. Proc. Natl. Acad. Sci. U.S.A. 120, e2301897120. doi:  10.1073/pnas.2301897120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Xu Y., Wu X., Li Y., Liu X., Fang L., Jiang Z. (2024). Probiotics and the role of dietary substrates in maintaining the gut health: use of live microbes and their products for anticancer effects against colorectal cancer. J. Microbiol. Biotechnol. 34, 1933–1946. doi:  10.4014/jmb.2403.03056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Xu S., Zhang Y., Ding X., Yang Y., Gao J., Zou N., et al. (2025). Intestinal microbiota affects the progression of colorectal cancer by participating in the host intestinal arginine catabolism. Cell Rep. 44, 115370. doi:  10.1016/j.celrep.2025.115370 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Data Availability Statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

Articles from Frontiers in Cellular and Infection Microbiology are provided here courtesy of Frontiers Media SA

RESOURCES