Colorectal cancer landscape from gut microbiota: Insights into mutations, epigenetic dysregulation, and immune microenvironment alterations

ABSTRACT

This review explores how gut microbiota reshapes colorectal cancer (CRC) molecular landscape, driving initiation/progression via key mechanisms. Microbes/metabolites cause driver mutations (DNA damage, signaling interference) and alter epigenetics (DNA methylation, histone modifications, ncRNAs) to boost tumor cell proliferation/survival. Dysbiosis disrupts tumor immune microenvironment (TIME) by impairing immune cells and increasing immunosuppressive factors, fostering immune evasion.Integrating molecular biology, microbiology, and immunology, we show microbial changes are causal in oncogenesis, with species acting distinctly across tumor stages. These insights clarify CRC pathogenesis and support microbiota-based prevention, diagnosis, treatment. Targeting microbes/metabolites or signaling pathways could cut CRC risk, improve early detection, and boost therapy. The review highlights microbiota-targeted therapy’s promise and guides future research/clinical translation.

Graphical Abstract

Gut microbiota and its role in colorectal cancer progression. This graphical abstract illustrates the interplay between gut microbiota and colorectal cancer (CRC). Gut microbiota influences CRC progression through the modulation of driver mutations, epigenetic regulation, and the tumor immune microenvironment (TIME).

Introduction

Colorectal cancer (CRC) represents a significant global health burden, contributing disproportionately to worldwide cancer incidence and mortality. Current epidemiological data indicate it accounts for approximately 10% of all annually diagnosed malignancies and cancer-associated deaths [1]. Globally, CRC ranks as the second most frequently diagnosed malignancy among women and the third most common among men [2,3]. Notably, age-standardized incidence and mortality rates in women are approximately 25% lower than those observed in males. These epidemiological patterns exhibit substantial geographic heterogeneity, with the highest age-standardized rates consistently documented in highly developed nations [4]. As developing regions undergo demographic and economic transitions, projections indicate a concerning upward trajectory in CRC incidence, with models forecasting approximately 2.5 million new cases annually by 2035 [5]. In the recent years, economic developments and changes in lifestyle and diet have increased the incidence and mortality rates of CRC in China [6].

Over the past decades, remarkable advances have been achieved in the clinical management of CRC, including the refinement of surgical techniques, the optimization of chemotherapy regimens, and, more recently, the introduction of targeted therapies and immunotherapies [7–9]. These developments have collectively contributed to significant improvements in the prognosis of certain subsets of patients. Nevertheless, despite these encouraging achievements, the overall clinical outcome for patients with advanced-stage disease remains far from satisfactory. Specifically, the 5-year survival rate in this population continues to be poor [10,11], with tumor metastasis and recurrence representing the predominant factors underlying therapeutic failure [12,13]. These limitations underscore an urgent need to gain deeper mechanistic insights into CRC pathogenesis and to explore novel therapeutic targets and innovative treatment strategies, which may ultimately contribute to improved prevention, earlier diagnosis, and more effective intervention [14,15].

In parallel with these clinical challenges, the past decade has witnessed unprecedented progress in high-throughput sequencing and multi-omics technologies, including metagenomics, metabolomics, and molecular biology approaches. These advances have dramatically expanded our capacity to interrogate complex biological systems at previously unattainable levels of resolution [16]. Within this context, the human gut microbiota has emerged as a highly complex and dynamic “invisible organ,” the significance of which in human health and disease is now increasingly appreciated [17]. The gut microbiota comprises a vast ecological community of microorganisms – including bacteria, fungi, viruses, and archaea – that inhabit the gastrointestinal tract. Remarkably, the number of microbial cells exceeds that of human somatic and germ cells by more than tenfold, and their collective genetic repertoire, often referred to as the “microbiota,” contains approximately 100 times more genes than the human genome itself [18]. Through long-term coevolution with the host, this microbial ecosystem has established a finely tuned symbiotic relationship, which is indispensable for a wide spectrum of physiological processes. These include the regulation of nutrient metabolism and energy harvesting [19], modulation of innate and adaptive immune responses [20], and the maintenance of intestinal barrier structure and function [21].

The preservation of gut microbial homeostasis is therefore essential for sustaining intestinal and systemic health. Perturbations of this delicate equilibrium – induced by factors such as diet, pharmacological interventions, environmental exposures, or host genetic variation – can lead to compositional and functional alterations in the microbial community, a condition commonly termed “dysbiosis” [22]. A rapidly growing body of evidence demonstrates that microbial dysbiosis is intimately linked with the onset and progression of a broad spectrum of human diseases, ranging from inflammatory bowel disease (IBD) [23], obesity [24], and type 2 diabetes mellitus [25], to cardiovascular pathologies [26] and neurological disorders [27]. Particularly noteworthy is the emerging recognition of a close association between gut microbiota dysregulation and CRC. This area has recently become a major focus of intensive research, with accumulating experimental and clinical evidence demonstrating that alterations in gut microbial composition and function play a pivotal role not only in CRC initiation and progression, but also in metastatic dissemination and therapeutic response [28] (Table 1).

Table 1.

Impact of the microbiome throughout the stages of CRC.

Stage	Key mechanisms	Microorganisms
Initiation	Driver gene mutations and epigenetic alterations: 1. Induced DNA damage: Virus-producing strains directly disrupt the stability of the host genome. 2. Epigenetic modification: Altering DNA methylation and histone modification, silencing tumor suppressor genes.	Pks+ E. Coli; Fusobacterium nucleatum
Progression to Invasive Disease	Reshape the immune microenvironment and metastasis: 1. Immune escape: Recruit immunosuppressive cells (such as Tregs, MDSCs) to inhibit cytotoxic T cells. 2. Promote metastasis: Induce epithelial-mesenchymal transition (EMT) and angiogenesis.	Fusobacterium nucleatum; Peptostreptococcus anaerobius; enterotoxigenic ETBF; Clostridioides difficile
Treatment Response	Affect the efficacy of chemotherapy and immunotherapy: 1. Chemotherapy resistance: Reducing chemotherapy sensitivity by regulating autophagy, apoptotic pathways or metabolizing drugs. 2. Radiotherapy response: Affects the occurrence of intestinal mucositis and anti-tumor immunity.	Fusobacterium nucleatum; Akkermansia muciniphila; certain Ruminococcaceae; Salmonella strains

The interplay between gut microbiota and CRC is highly complex, involving multiple dimensions and hierarchical levels of regulation. At the molecular level, gut microbes and their metabolites can exert profound influence on the tumorigenic process through several distinct yet interrelated mechanisms. These include the induction of driver gene mutations [29], the modulation of epigenetic modifications that regulate oncogene and tumor suppressor gene expression [30], and the disruption of immune microenvironmental homeostasis [31]. Metabolic mechanisms by microbiome are also available for the development of CRC. The combined analysis of heterogeneous CRC cohorts identified an association of microbiome choline metabolism with CRC [32]. The cross-study comparisons indicated that it existed a metabolic link between cancer-associated gut microbes and a fat- and meat-rich diet within CRC metagenomes [33].

Collectively, these mechanisms reshape the molecular landscape of CRC, thereby dictating key aspects of tumor biology such as proliferation, invasion, and resistance to therapy[34]. Recognition of these microbiota-driven processes not only expands our current understanding of CRC pathogenesis, but also opens up entirely new conceptual avenues for the development of microbiota-informed preventive and therapeutic interventions [35].

Notably, while the molecular mechanisms of microbiota-driven carcinogenesis are being elucidated, it is equally critical to understand the antecedent risk factors that drive the transition from eubiosis to dysbiosis over the human lifespan. Emerging trends highlight a novel mechanism where early-life exposures such as diet and environmental pollutants induce persistent ecological impact on the microbiota among CRC populations [36]. This long-term perspective shifts the focus from mere bacterial presence to the functional trajectory of the microbiome, where exterior risk factors act as selective pressures, ultimately sculpting a pro-carcinogenic ecosystem. Ecological collapse leads to the loss of colonization resistance, making it easier for pathogenic bacteriato colonize and occupy ecological niches, which may last for several years, continuously releasing genetic toxins like Colibactin to enhance immunosuppressive lipid overload in CRC [37]. A notable enhancement in patient prognosis can be achieved by detecting and eliminating modifiable risk factors, as well as by applying personalized screening strategies for those at high risk of CRC based on different panels of genes and mRNA [38]. The prospective cohort incident-tumor biobank method (PCIBM) on longitudinal prospective cohort studies, namely the Nurses’ Health Study and the Health Professionals Follow-up Study, is available for the research of CRC [39]. Researchers evaluated the associations of CRC microbial dietary score (CMDS) with increased CRC risk according to Fusobacterium nucleatum via PCIBM [40]. The TIPC algorithm allows for unsupervised discovery of microgeometric tissue patterns and new tumor subtypes, which can enhance our understanding of the tumor immune microenvironment and potentially influence precision cancer immunotherapy for CRC [41].

Given the critical involvement of gut microbiota in virtually every stage of CRC development – from tumor initiation and progression to metastasis and treatment response – the present review seeks to provide a comprehensive and systematic synthesis of current knowledge [42]. Specifically, we summarize the molecular mechanisms through which gut microbiota reshapes the oncogenic landscape of CRC, with particular emphasis on driver mutations, epigenetic modifications, and immune microenvironmental regulation[43]. We also highlight major advances in this rapidly evolving field, identify existing knowledge gaps and unresolved questions, and discuss the future prospects and challenges associated with microbiota-targeted strategies for CRC prevention and treatment. By integrating these insights, this review aims to establish a theoretical framework that may deepen our mechanistic understanding of CRC pathogenesis and ultimately facilitate the development of novel therapeutic paradigms.

Gut microbiota in health

Basic characteristics

The adult human gastrointestinal tract harbors an extraordinarily complex and diverse community of microorganisms, collectively referred to as the gut microbiota, which encompasses bacteria, archaea, viruses, and fungi. It has been estimated that the number of microbial cells in the gut approaches 10¹⁴, thereby outnumbering host cells by at least an order of magnitude and contributing to a metabolic and genetic repertoire far greater than that of the human genome itself [44]. Although the taxonomic composition of the gut microbiota displays a degree of stability along the intestinal tract, the absolute density of microorganisms varies markedly, with microbial load generally increasing from the oral cavity toward the distal colon and rectum [45]. Importantly, inter-individual variability is a well-recognized feature, reflecting influences from genetic, environmental, nutritional, and developmental factors. Colonization begins at birth, when neonates acquire commensal organisms from maternal sources such as the skin, vaginal tract, and fecal flora [46]. The community undergoes rapid expansion and dynamic shifts during the first two years of life, a period regarded as critical for immune system maturation and host – microbe co-adaptation [47,48]. After this developmental window, the microbial ecosystem tends to stabilize and maintain a relatively consistent structure, although perturbations can occur in response to aging, dietary modifications, antibiotic exposure, or pathological conditions [49]. In older adults, gradual remodeling of community structure is observed, yet many core metabolic and physiological functions appear to be conserved [50,51].

The establishment of a diverse and balanced microbial community in early life is considered essential for the proper development of host immunity, as illustrated by profound immune abnormalities in germ-free animal models raised under sterile conditions [52]. Within the intestinal tract, the colon represents the most densely colonized region, harboring nearly 10³ distinct microbial species, predominantly bacteria [53]. The bacterial population in this region is estimated at approximately 10¹⁴ cells, accounting for nearly 70% of the host’s total microbial biomass [54]. Consequently, this review primarily emphasizes the role of colonic bacteria in the context of colorectal carcinogenesis [55].

Because the majority of bacterial species are refractory to conventional culturing techniques, molecular approaches have become indispensable for characterizing microbial communities. High-throughput methods, particularly 16S ribosomal RNA (rRNA) gene sequencing of fecal or intestinal tissue samples, enable precise identification and classification of uncultivable taxa[56]. Depending on their anatomical localization, the microbiota can be subdivided into luminal flora, residing freely within the intestinal lumen, and mucosa-associated flora, which colonize or infiltrate the mucus layer overlaying epithelial cells [57]. The intestinal mucus layer serves as a crucial barrier that prevents excessive microbial – epithelial interactions and protects enterocytes from constant exposure to dietary antigens, thereby reducing the risk of aberrant immune activation [58]. Notably, the ratio of anaerobic to aerobic organisms differs between these compartments, with mucosal surfaces generally harboring lower proportions of strict anaerobes compared with luminal communities. Fecal sampling, which is minimally invasive, provides a practical representation of distal colonic flora; however, it may not accurately reflect the proximal mucosa-associated microbiota [59].

Experimental comparisons further suggest that murine gut microbiota shares considerable similarities with human communities, thereby supporting the translational relevance of rodent models in gastrointestinal research [60,61]. Current surveys indicate that more than 50 microbial phyla and up to 500 bacterial species may comprise the commensal human gut microbiota, though the precise number and degree of inter-individual variability remain incompletely defined [62]. Factors such as diet, host genotype, environmental exposures, and lifestyle are believed to exert profound influences on community composition [63]. Despite this variability, a core set of taxa is consistently detected across individuals, dominated by three phyla: Firmicutes (30%–50%), Bacteroidetes (20%–40%), and Actinobacteria (1%–10%). Within these groups, strict anaerobes – including Bacteroides, Eubacterium, Bifidobacterium, Fusobacterium, Peptostreptococcus, and Atopobium—constitute the majority of the gut microbiome, whereas facultative anaerobes such as Lactobacillus, Enterococcus, Streptococcus, and members of Enterobacteriaceae occur at markedly lower abundances, often 1,000-fold less than strict anaerobes [64]. It is also noteworthy that microbial composition varies substantially along the gastrointestinal tract. For instance, Bacteroidetes and Actinobacteria together account for over 90% of colonic taxa, whereas in the small intestine they represent only about 50%, with Firmicutes comprising approximately 40% of the community [65]. In CRC, it was evidenced by a novel exploration on tumor location-associated fecal microbes, including Veillonella parvula for right-sided CRC (rCRC), Streptococcus angionosus for left-sided CRC (lCRC), and Peptostreptococcus anaerobius for rectal cancer (RC), while Fusobacterium nucleatum is enriched in all tumor locations [66]. Another meta-analysis determined the commensal Ruminococcus bicirculans and Faecalibacterium prausnitzii as late-stage CRC signatures [67].

Microbiota and gut homeostasis

The intestinal microbiota serves as a crucial component of the host’s natural defense system, functioning not only as a protective barrier against pathogenic invasion but also as an active participant in a wide array of structural and metabolic processes that sustain intestinal epithelial physiology and preserve gut homeostasis[68]. Beyond its role in defense, this microbial consortium contributes to multiple essential physiological functions that are indispensable for the maintenance of gastrointestinal health [69]. The majority of insights into these roles have been derived from germ-free animal models, which provide a unique opportunity to examine host physiology in the absence of microbial colonization[70]. Such animals display a markedly altered phenotype: they are highly vulnerable to infections and exhibit impaired vascular development, diminished activity of digestive enzymes, thinner intestinal muscle walls, reduced cytokine secretion, lower serum immunoglobulin concentrations, underdeveloped Peyer’s patches, and a reduced population of intraepithelial lymphocytes [71]. Strikingly, reintroducing commensal microbes into these germ-free systems is sufficient to restore mucosal immune function [72] and also modulates host gene expression involved in nutrient absorption, metabolic activity, angiogenesis, epithelial barrier integrity, and even the formation of the enteric nervous system [73]. Collectively, these findings underscore the critical role of commensal organisms in shaping host immunity, including both innate and adaptive responses [74]. Indeed, specific microbial taxa have been shown to modulate humoral immune components [75], alter T-cell repertoires, and influence T-helper cell cytokine profiles [76,77]. These experimental data strongly suggest that variations in gut microbiota composition may underlie inter-individual differences in immune competence[78].

In addition to their immune functions, the structural impact of the microbiota on the intestinal epithelium has become increasingly apparent, again largely through observations in germ-free mice. Such animals often present elongated villi associated with crypt atrophy, delayed epithelial cell turnover, and impaired angiogenesis. Moreover, reductions in mucosal and muscular wall thickness have been consistently reported in the absence of microbial colonization. Importantly, experimental data in murine models predisposed to CRC have demonstrated that the presence of a functional microbiota enhances crypt cell turnover, suggesting a link between microbial signaling, epithelial renewal, and tumor susceptibility [79].

The metabolic contribution of the gut microbiota represents another fundamental dimension of host – microbe symbiosis [69]. Microbial communities engage in a broad range of biochemical processes, including (i) anaerobic carbohydrate fermentation that generates gases such as CO₂, H₂, and CH₄, along with short-chain fatty acids (SCFAs) such as butyrate, propionate, and acetate [80]; and (ii) proteolytic fermentation, which yields bioactive metabolites including phenolic compounds, amines, ammonia, N-nitroso compounds, and indoles. These metabolites exert profound influences on intestinal physiology by regulating gene expression, directing epithelial differentiation and proliferation, facilitating vitamin biosynthesis, enhancing ion absorption, and stimulating mucus secretion [81,82]. In addition, SCFAs serve as major energy substrates for colonocytes, while also modulating systemic lipid metabolism and contributing to overall host energy balance. Through these metabolic interactions, the microbiota not only optimizes the extraction and storage of energy from dietary sources but also supplies absorbable nutrients for both host and microbial growth, thereby sustaining the symbiotic ecosystem [83,84]. However, it should be noted that not all microbial metabolites are beneficial; by-products of proteolytic fermentation, for instance, may exert toxic or carcinogenic effects on the host epithelium [85].

Beyond immune, structural, and metabolic functions, the gut microbiota confers a protective phenomenon known as “colonization resistance” or “microbial interference,” whereby commensal microorganisms prevent pathogenic bacteria from establishing residence in the gastrointestinal tract [86]. Although the precise mechanisms underlying this protective effect remain incompletely understood, several potential pathways have been proposed. These include competition for binding sites on epithelial receptors, reinforcement of mucosal barrier integrity, competition for essential nutrients, and the production of antimicrobial peptides or bacteriocins [69]. Disruption of colonization resistance – for example, through broad-spectrum antibiotic use or pathogenic invasion – has been associated with increased susceptibility to gastrointestinal disorders and infections, further highlighting the indispensable role of the commensal microbiota in maintaining intestinal equilibrium [87].

Taken together, accumulating evidence demonstrates that the gut microbiota plays multifaceted roles in host defense, epithelial structure, metabolic regulation, and immune function (Figure 1). These findings not only expand our understanding of the physiological interplay between microbes and the host but also emphasize the potential consequences of microbial imbalance in intestinal diseases, including CRC.

Figure 1.

Gut microbiota and intestinal homeostasis. Schematic representation of the healthy gut microbiota maintaining intestinal homeostasis. Commensal bacteria interact with the mucus layer and epithelial cells, producing metabolites such as short-chain fatty acids (SCFAs) that support epithelial integrity, regulate immune balance.

Gut microbiota and driver mutations

CRC is a heterogeneous malignancy driven by the stepwise accumulation of genetic alterations. Driver mutations in key oncogenes and tumor suppressor genes – such as APC [88], KRAS [89,90], p53 [91], BPAF [92] and PIK3CA [92] —are widely recognized as fundamental events in tumor initiation and progression [93]. Recent advances in microbiota research have revealed that the gut microbiota, a dynamic ecosystem of bacteria, viruses, fungi, and archaea inhabiting the human intestine, exerts profound influence on the mutational landscape of CRC [94,95]. Multi-kingdom gut microbiome dysbiosis was identified in CRC across cohorts with including 11 bacterial, 4 fungal and 1 archaeal feature, which exhibited good diagnostic potential for CRC [94]. The combination of archaeal and bacterial markers also holds promise as noninvasive diagnostic tools for CRC [96]. Increasing evidence suggests that microbial dysbiosis, characterized by alterations in microbial composition and function, contributes to the induction and selection of driver mutations through multiple mechanisms (Figure 2).

Figure 2.

Gut microbiota and driver mutations in colorectal cancer (CRC). Gut microbiota contributes to CRC driver mutations, including mutations in APC, KRAS, BRAF, and TP53. It accelerates mutations via: 1) Genotoxins (e.g. BFT from ETBF and colibactin from Escherichia coli) that cause DNA damage and genomic instability; 2) disruption of signaling pathways (e.g. fusobacterium nucleatum activating Wnt/β-catenin and NF-κB); 3) interaction with specific mutations.

This emerging field highlights a paradigm shift in CRC research, wherein the gut microbiota is no longer considered a passive bystander but rather an active participant in shaping tumor evolution. Importantly, linking specific microbial taxa and their metabolites to mutational events provides new opportunities for precision oncology [97]. For instance, the detection of microbiota-derived mutational footprints could serve as early diagnostic biomarkers, while microbiota-targeted interventions – such as probiotics, antibiotics, dietary modulation, or fecal microbiota transplantation – may complement existing therapeutic strategies by reducing mutagenic pressure and restoring microbial homeostasis. Despite these advances, several challenges remain, including disentangling causality from correlation, understanding inter-individual variability, and integrating microbiota data with genomic and epigenomic profiles in a clinically meaningful manner.

In summary, the interplay between the gut microbiota and driver mutations in CRC represents a rapidly expanding frontier in cancer biology. By elucidating how microbial factors contribute to mutagenesis and tumor evolution, researchers are uncovering novel insights into CRC pathogenesis and opening avenues for microbiota-informed strategies in prevention, early detection, and therapy. This integrated perspective not only deepens our mechanistic understanding of CRC but also holds promise for translating microbiota science into tangible clinical benefits for patients.

Gut microbiota and their metabolites induce DNA damage

Certain bacterial species within the intestinal microbiota are capable of producing genotoxic metabolites that directly or indirectly damage host cell DNA, thereby inducing genetic mutations and laying the groundwork for colorectal carcinogenesis [98]. A well-characterized example is enterotoxigenic Bacteroides fragilis (ETBF), which secretes the toxin known as Bacteroides fragilis toxin (BFT). BFT has the ability to cleave tight junction proteins between intestinal epithelial cells, thereby compromising the structural integrity of the intestinal barrier and increasing epithelial permeability. This disruption facilitates closer contact between luminal bacteria, as well as their secreted metabolites, and the epithelial surface, thereby augmenting the likelihood of DNA damage and subsequent mutational events [99,100]. Experimental studies have demonstrated that infection of mice with ETBF results in elevated levels of DNA double-strand breaks within colonic epithelial cells and is associated with mutations in critical tumor suppressor genes such as p53 [91], highlighting the mutagenic potential of bacterial toxins in vivo.

In addition to ETBF, certain strains of Escherichia coli possess the pks genomic island, which encodes the biosynthetic machinery for the production of colibactin, a hybrid polyketide – nonribosomal peptide genotoxin with well-documented mutagenic activity. Colibactin is capable of intercalating into host DNA double helices, where it induces alkylation damage, leading to mismatched or deleted base pairs, and ultimately driving mutation accumulation [101]. Clinical and metagenomic studies have further revealed that colibactin-producing E. coli strains are enriched in both tumor tissues and fecal samples of patients with CRC, compared with healthy individuals, strongly suggesting a contributory role of this toxin-producing bacterium in CRC initiation and progression [102,103]. This observation not only underscores the pathogenic capacity of colibactin but also implicates microbial-derived genotoxins as key environmental drivers of CRC-related mutational signatures.

Beyond B. fragilis and E. coli, other bacterial pathogens have also been implicated in CRC through the production of genotoxic factors. For instance, Campylobacter jejuni has been shown to promote colorectal tumorigenesis by secreting cytolethal distending toxin (CDT), a protein complex with DNase-like activity that induces DNA double-strand breaks in host cells [98]. The accumulation of such DNA lesions not only compromises genomic stability but also provides fertile ground for the emergence of driver mutations that propel malignant transformation.

In addition, dietary patterns exert continuous pressure on the host genome through microbial metabolic products. Excessive intake of high-fat diet can enrich intestinal bacteria that can improve intestinal deoxycholic acid (DCA) levels, which further exacerbated colonic inflammation [104]. Long-term exposure to DCA can induce specific gene mutation signatures in intestinal stem cells, especially those involving the RAS and p53 pathways [105,106]. An observational study indicated the dietary interactions with gut sulfur-metabolizing bacteria in the incidence of early-onset colorectal carcinogenesis [107]. Consuming significant amounts of red meat is associated with a heightened risk of CRC. A possible cause for the association is the detrimental impact of heme iron present in red meat, which could disturb homeostasis, hinder the renewal of colonic epithelial cells, and foster the formation of mutagenic and carcinogenic agents [108]. On the other hand, consuming a lot of dietary fiber can partially mitigate the negative impacts of meat by various mechanisms, such as decreasing intestinal transit time and diluting carcinogenic substances, functioning protective effects on the risk of developing colorectal cancer [109]. Apart from diet, the interaction between environmental pollutants and the microbiome is an emerging research hotspot. Disruptions in gut microbiota composition and diversity due to environmental exposures, especially microplastics and nanoplastics, can cause dysbiosis, chronic inflammation, and oxidative stress, which are associated with an increased risk of CRC [110].

Collectively, these findings highlight a critical mechanistic link between specific gut bacterial pathogens, their genotoxin production, and the induction of host genomic instability. By directly damaging DNA and fostering mutations in key oncogenes and tumor suppressor genes, these microbes contribute to the molecular landscape of CRC pathogenesis. Insights on these microbe – host interactions offer new avenues for the development of microbiota-based biomarkers and therapeutic interventions aimed at mitigating genotoxin-induced mutagenesis and preventing colorectal tumor initiation.

Gut microbiota disrupt cellular signaling pathways to promote the accumulation of gene mutations

The gut microbiota and its metabolites exert profound effects on host cellular physiology, in part by interfering with critical intracellular signaling pathways that regulate proliferation, differentiation, and DNA damage repair. Dysregulation of these processes increases the likelihood of genetic mutations and facilitates their accumulation within epithelial cells, thereby predisposing to malignant transformation. Among the signaling cascades most relevant to colorectal tumorigenesis, the Wnt/β-catenin pathway plays a central role. Under physiological conditions, this pathway is tightly regulated, ensuring the maintenance of intestinal epithelial homeostasis and normal tissue renewal [111]. However, when microbial composition and metabolic activity become imbalanced, certain bacterial species and their metabolites have been shown to aberrantly activate this signaling axis.

A particularly well-studied example involves Fusobacterium nucleatum (Fn), a pathogen frequently enriched in the CRC microenvironment. Upon infection of intestinal epithelial cells, F. nucleatum orchestrates a molecular network of the Toll-like receptor, microRNAs, and autophagy to clinically, biologically, and mechanistically control colorectal cancer chemoresistance [112]. Moreover, the activation of NF-κB leads to upregulation of multiple genes associated with the Wnt signaling pathway, thereby amplifying β-catenin signaling activity [113]. Mechanistically, this process promotes the accumulation of β-catenin in the cytoplasm, preventing its degradation. Once stabilized, β-catenin translocates into the nucleus, where it forms complexes with transcription factors of the TCF/LEF family [114]. These complexes act as transcriptional activators for a variety of oncogenic target genes, including c-Myc [115] and Cyclin D1 [116], both of which are key drivers of uncontrolled cellular proliferation and tumorigenesis.

Sustained aberrant activation of the Wnt/β-catenin pathway under these conditions not only triggers excessive and dysregulated epithelial cell proliferation but also interferes with the fidelity of DNA damage repair mechanisms [117,118]. As a consequence, the genomic stability of intestinal epithelial cells is compromised, thereby increasing the probability of mutation occurrence and promoting their accumulation over time. Such cumulative genetic alterations can ultimately drive the initiation and progression of colorectal neoplasia. These findings highlight the intricate interplay between gut microbial dysbiosis, microbial-derived metabolites, and host intracellular signaling, emphasizing how microbial perturbations can serve as a catalyst for carcinogenic transformation through chronic pathway activation.

A recent systematic investigation has provided important insights into the interplay between microbial composition, T cell receptor (TCR) repertoires, and somatic mutations in patients with CRC [119]. In this large-scale study, researchers profiled the expressed TCRβ repertoires in the peripheral blood of CRC patients (n = 107) in comparison with healthy donors (n = 30). The analysis revealed that individuals with CRC exhibited a markedly higher proportion of expanded, large TCRβ clones, accompanied by a significant reduction in overall TCR diversity. Such alterations in the adaptive immune landscape suggest profound immune remodeling during tumorigenesis and highlight the potential diagnostic value of TCR repertoire signatures. Parallel metagenomic sequencing further revealed striking differences in the gut microbiome between CRC patients and healthy controls. Specifically, the relative abundance of Fn, Escherichia coli, and even viral elements such as Dasheen mosaic virus was consistently elevated in the CRC cohort (n = 97) compared with controls (n = 30). In contrast, health-associated commensals such as Faecalibacterium prausnitzii and Roseburia intestinalis, both of which are known producers of anti-inflammatory metabolites like butyrate, were significantly depleted in CRC patients. This microbial shift underscores a transition from a protective, anti-inflammatory environment toward one enriched in pathobionts and potential carcinogenic drivers. Beyond taxonomic differences, the study examined the relationship between somatic mutations in driver genes (16 genes analyzed in n = 79 CRC patients) and both TCR clonality and microbial biomarkers. This integrative approach uncovered correlations that suggest host immune repertoire alterations may be functionally linked to tumor-associated microbial dysbiosis and mutational landscapes. To further translate these findings into a clinical setting, the investigators constructed a random forest classifier incorporating 15 key features derived from microbial profiles and TCR repertoire data. This machine-learning model demonstrated potential utility as a noninvasive tool for CRC detection and screening, offering a proof-of-concept for multi-omics – based diagnostic strategies. Importantly, the study also conducted a dedicated analysis of F. nucleatum-specific TCR repertoires, thereby providing a unique window into host – microbe immune interactions at the tumor interface. Collectively, this large-cohort, multi-omics study highlights the potential of integrating microbial, immunological, and genomic data to discover novel biomarkers of both etiological and diagnostic significance. Such findings not only advance our mechanistic understanding of CRC pathogenesis but also hold promise for informing precision medicine approaches in cancer detection and early intervention.

Effects of gut microbiota on mutations in specific driver genes

In the CRC, several recurrent driver gene mutations – most notably in KRAS [120], BRAF [121], and TP53 [122]—play pivotal roles in tumor initiation and progression. Increasing evidence indicates that the gut microbiota is not merely a passive bystander in this mutational landscape but actively contributes to the acquisition, selection, and functional consequences of these mutations.

With respect to KRAS mutations, multiple studies have elucidated microbiota-driven mechanisms influencing their frequency and oncogenic activity. SCFAs, a major class of metabolites produced by commensal bacteria, can modulate intracellular redox balance and inflammatory signaling, thereby indirectly affecting the mutational burden of KRAS [123]. Of particular interest, Fn has been shown to exhibit a striking preferential enrichment in CRC patients harboring the KRAS p.G12D mutation. Mechanistically, F. nucleatum can invade colonic epithelial cells and interact with DHX15, an RNA helicase expressed in CRC tumor cells, thereby promoting carcinogenesis through activation of the ERK/STAT3 signaling axis [124]. Moreover, KRAS mutations themselves appear to facilitate microbial colonization within the tumor microenvironment; specifically, oncogenic KRAS can regulate the miRNA-3655/SURF6/IRF7/IFN-β signaling cascade, which in turn promotes the intratumoral persistence of the pathogenic bacterium Bacteroides fragilis [89]. This bidirectional relationship underscores the intertwined nature of microbial ecology and oncogenic signaling in CRC.

In the context of BRAF mutations, distinct microbial signatures have been reported to segregate CRC patients according to their BRAF mutational status. In particular, BRAFV600E-mutant tumors demonstrate microbial profiles correlated with pathway-specific molecular markers associated with inflammation, innate immune activation, and epithelial – mesenchymal transition (EMT) [125]. Notably, two bacterial taxa—Prevotella enoeca, enriched in BRAFV600E-positive patients, and Ruthenibacterium lactatiformans, more abundant in BRAF wild-type patients – emerged as promising microbial biomarkers capable of distinguishing BRAF status with considerable accuracy (AUROC = 0.72 and 0.74, respectively, with 95% confidence intervals). These findings highlight the potential of microbiome-based classifiers in refining molecular subtyping of CRC and tailoring personalized treatment strategies.

Regarding TP53 mutations, the interaction with the gut microbiota appears to be mediated through both direct and indirect mechanisms. On one hand, bacterial genotoxins can induce DNA damage, thereby driving TP53 mutagenesis. On the other hand, chronic inflammation arising from gut microbial dysbiosis establishes a pro-tumorigenic microenvironment that accelerates the accumulation of TP53 mutations [91]. As TP53 serves as a crucial genomic guardian, the loss of its tumor-suppressive function facilitates unchecked cellular proliferation, impaired apoptosis, and enhanced genomic instability, thereby contributing substantially to CRC progression.

Beyond KRAS, BRAF, and TP53, the gut microbiota may also influence mutations in other critical CRC driver genes, such as APC. As a negative regulator of the Wnt signaling pathway [126], APC mutation represents an early and defining event in colorectal tumorigenesis. Although direct evidence linking microbial activity to APC mutation is currently limited, several lines of research suggest potential indirect influences. For example, dysregulated microbial communities can aberrantly activate Wnt/β-catenin signaling, which may synergize with APC mutations to drive tumor progression [127]. Additionally, impaired DNA repair pathways resulting from microbiota-driven inflammation and genotoxic stress may increase the probability of APC mutagenesis [128,129]. These findings collectively support the notion that microbial alterations not only modulate the tumor microenvironment but may also intersect with the mutational trajectories of key driver genes in CRC.

Taken together, accumulating evidence suggests a complex, bidirectional crosstalk between the gut microbiome and canonical CRC driver mutations. Oncogenic alterations may shape microbial colonization and metabolic outputs within the tumor niche, while, conversely, microbial products and inflammatory mediators can promote or accelerate mutagenesis in epithelial cells. This dynamic interplay highlights the importance of incorporating microbiome analysis into the study of CRC molecular pathogenesis, with the ultimate goal of identifying novel therapeutic and diagnostic opportunities.

Gut microbiota and epigenetic regulation

Accumulating evidence suggests that the gut microbiota profoundly shapes the epigenetic landscape of CRC, thereby influencing tumor initiation and progression [130,131]. Microbial metabolites such as SCFAs, particularly butyrate, can function as histone deacetylase inhibitors, altering chromatin accessibility and transcriptional activity of genes regulating cell proliferation, apoptosis, and immune surveillance [132,133]. Dysbiosis has also been associated with aberrant DNA methylation, exemplified by promoter hypermethylation of tumor suppressors like MLH1 and APC, as well as the CpG island methylator phenotype linked to Fn enrichment [134,135]. Moreover, inflammatory signaling pathways such as NF-κB and STAT3, activated by microbial imbalance, recruit chromatin-modifying enzymes to sustain pro-tumorigenic transcriptional programs [136,137]. Recent studies further highlight microbiota-driven modulation of non-coding RNAs, including oncogenic and tumor-suppressive microRNAs, which fine-tune epithelial barrier integrity, immune responses, and metastatic potential [138–140]. Collectively, these findings reveal a bidirectional interplay between the microbiome and host epigenome that not only provides mechanistic insights into CRC pathogenesis but also offers promising avenues for biomarker discovery and the development of microbiome-epigenome-based therapeutic strategies (Figure 3).

Figure 3.

Epigenetic regulation by gut microbiota in CRC. Gut microbiota modulates host epigenetics in CRC [1]. DNA methylation: SCFAs inhibit DNMT activity, resulting in demethylation and tumor suppressor gene reactivation [2]. histone modification: butyrate inhibits HDACs, leading to histone acetylation and gene activation [5]. non-coding RNA.

DNA methylation

DNA methylation represents one of the most stable and extensively studied forms of epigenetic regulation, mediated primarily by DNA methyltransferases (DNMTs) that catalyze the transfer of a methyl group to the fifth carbon of cytosine, thereby generating 5-methylcytosine (5mC) [141]. In eukaryotes, the methylation status of CpG islands located within gene promoter regions plays a pivotal role in the regulation of transcription. Under physiological conditions, promoter hypermethylation could recruit methyl-CpG binding proteins and repressive chromatin remodeling complexes, which prevent transcription factor binding and suppress gene expression. Conversely, promoter hypomethylation facilitates the assembly of transcriptional machinery, thereby enhancing gene activation [142,143]. Three major DNMTs orchestrate these processes: DNMT1 maintains existing methylation patterns following DNA replication, while DNMT3A and DNMT3B are responsible for de novo methylation, particularly during embryogenesis. Importantly, DNA methylation is a reversible process regulated by both DNMTs and demethylating enzymes. Demethylation may occur passively due to failure of methyl group maintenance during replication, or actively via ten-eleven translocation (TET) proteins, which oxidize 5mC to generate intermediates such as 5-hydroxymethylcytosine, 5-formylcytosine, and 5-carboxylcytosine, eventually excised and repaired through base excision repair pathways [144].

The gut microbiota, through its metabolic activity and interactions with host epithelial cells, exerts profound influence on DNA methylation landscapes in colorectal epithelial cells, thereby altering the expression of cancer-related genes. Among microbial metabolites, SCFAs stand out as the most extensively studied regulators [80,145]. SCFAs, including acetate, propionate, and butyrate, are the primary fermentation products of dietary fiber by commensal bacteria. Beyond serving as key energy sources for colonocytes, SCFAs exhibit epigenetic functions by modulating enzyme activities [146,147]. Specifically, they can competitively inhibit DNMT activity, thereby reducing global and promoter-specific DNA methylation. Butyrate, in particular, is recognized as a potent DNMT inhibitor; by occupying the catalytic pocket of DNMTs, it prevents their interaction with DNA substrates, ultimately leading to hypomethylation of tumor suppressor gene promoters and restoration of their expression [148,149].

In patients with colorectal cancer, the gut microbiota undergoes significant compositional changes, notably characterized by depletion of SCFA-producing bacteria. This shift reduces SCFA availability in the colonic lumen, weakening their inhibitory effect on DNMTs, thereby enhancing DNMT activity and driving aberrant DNA methylation of key regulatory genes. For example, promoter hypermethylation of the MLH1 gene, frequently observed in CRC, leads to transcriptional silencing and impaired mismatch repair capacity, thereby accelerating mutational accumulation and tumor initiation. Similarly, hypermethylation of the APC promoter results in its functional inactivation, which triggers aberrant activation of the Wnt/β-catenin signaling pathway and uncontrolled cellular proliferation. Additional tumor suppressors, including p16 and RUNX3, are also frequently silenced by promoter hypermethylation in association with microbial dysbiosis, collectively contributing to CRC progression [148,149].

Beyond SCFAs, other microbial metabolites may also influence DNA methylation. For instance, bacterial protein fermentation produces ammonia, which alters colonic luminal pH and potentially modulates DNMT activity [150]. Secondary bile acids, another class of microbial metabolites, can induce oxidative stress, thereby indirectly regulating the expression and activity of methylation-associated enzymes [151,152]. Collectively, these findings underscore that the gut microbiota reshapes the epigenetic architecture of CRC not only through SCFAs but also via multiple metabolite-driven pathways, positioning microbial metabolism as a critical determinant of DNA methylation dynamics and colorectal tumorigenesis.

Histone modifications

Post-translational modifications of histone N-terminal tails, including acetylation, phosphorylation, methylation, and ubiquitination, collectively constitute the so-called “histone code,” which governs chromatin architecture and thereby influences transcriptional activity [153,154]. Among these, acetylation and methylation have been most intensively investigated, as they function synergistically to regulate chromatin dynamics and gene expression [155,156]. Accumulating evidence indicates that the gut microbiota can modulate these modifications either directly or through metabolic products, thereby altering the transcriptional regulation of CRC – associated genes [157,158]. The balance of histone acetylation is primarily maintained by histone acetyltransferases (HATs) and histone deacetylases (HDACs), which coordinate key cellular processes such as transcription, translation, and DNA repair [131]. SCFAs, particularly butyrate, represent a major class of microbial metabolites that affect HDAC activity [159]. Butyrate functions as a potent HDAC inhibitor; in the context of CRC, where tumor cells preferentially utilize glycolysis via the Warburg effect, unmetabolized butyrate accumulates and enhances histone acetylation. This not only influences cellular proliferation and apoptosis but has also been shown to elevate histone acetylation levels in human rectal adenocarcinoma [160].

Beyond SCFAs, the microbiota also influences histone acetylation and methylation through the uptake and secretion of trace elements and cofactors such as zinc, cobalt, and iodine, which are critical for the activity of chromatin-modifying enzymes [161]. In vivo studies further support this relationship: conventionally raised mice display markedly elevated histone acetylation and methylation compared with germ-free counterparts, including a 12-fold increase in acetylation at histone H4 and a 1.5-fold increase in methylation at histone H3, highlighting the substantial contribution of microbial colonization to epigenetic regulation in the colon [162]. These findings underscore that gut microbes, by modulating nutrient metabolism, act as important regulators of epigenetic programs, thereby influencing CRC development and progression.

Therapeutically, these insights have opened new avenues for targeting histone modifications in CRC management. Conventional treatments, such as chemotherapy, surgery, and radiotherapy, have limited efficacy, whereas epigenetic interventions exploit the reversible nature of histone modifications. HDAC and histone methyltransferase/demethylase inhibitors are currently the most extensively explored in CRC. For instance, Verticillin A, a selective histone methyltransferase inhibitor, has been shown not only to suppress metastatic CRC cell growth but also to enhance cytotoxic T lymphocyte (CTL) – mediated immunotherapy, thereby impeding tumor progression and metastasis [163]. Similarly, JIB-04, a novel histone demethylase inhibitor, has demonstrated strong activity against colorectal cancer stem cells in vitro, reducing their proliferation, invasion, and migration capacity [164]. Collectively, these findings establish histone modifications as both critical mediators of microbe – host interactions and promising therapeutic targets for CRC, particularly in combination with established clinical regimens.

Non-coding RNA-mediated epigenetic regulation

Non-coding RNAs (ncRNAs), including microRNAs (miRNAs), long non-coding RNAs (lncRNAs), and circular RNAs (circRNAs), lack protein-coding capacity but play central roles in post-transcriptional regulation through base pairing or protein interactions, thereby representing key epigenetic regulators [165]. Growing evidence indicates that the gut microbiota contributes to CRC initiation and progression by modulating the expression and function of ncRNAs [166,167].

Among them, miRNAs regulate gene expression primarily by binding to the 3′ untranslated region of target mRNAs, leading to translational repression or mRNA degradation. Dysregulated microbial communities, such as Fn and Prevotella spp., markedly alter intestinal epithelial miRNA profiles, with downstream consequences on oncogenic signaling pathways [168]. For example, F. nucleatum infection upregulates miR-21, suppressing the tumor suppressor PTEN and activating Akt phosphorylation, thereby promoting CRC cell proliferation and invasion [169], while enterotoxigenic ETBF elevates miR-155 levels, silencing SOCS1 and enhancing NF-κB signaling to exacerbate intestinal inflammation [170]. Conversely, microbial metabolites such as butyrate augment miR-200c expression, leading to Bmi-1 suppression and reduced metastatic potential, supporting the therapeutic relevance of sodium butyrate in CRC [171]. Beyond tumorigenesis, altered miRNA expression is linked to metabolic disorders; for instance, downregulation of miR-146a and miR-215 in obese CRC patients suggests potential biomarker value for disease progression [172,173]. Moreover, CpG methylation – mediated silencing of tumor suppressive miRNAs, including miR-34a/b/c, highlights their role in genomic surveillance and their potential as therapeutic targets [112,174].

LncRNAs, typically longer than 200 nucleotides, also exhibit diverse regulatory mechanisms by interacting with DNA, RNA, or proteins [175]. The gut microbiota and its metabolites critically influence lncRNA expression in CRC. For instance, colibactin produced by toxigenic Escherichia coli induces DNA damage and activates HOTAIR transcription, which in turn modulates tumor suppressor gene expression and drives malignant proliferation [101,102]. Similarly, depletion of short-chain fatty acids elevates NORAD expression, which sequesters miRNAs, thereby releasing Cyclin D1 from repression and accelerating cell cycle progression [80]. Emerging evidence also implicates circRNAs in microbiota – host interactions; for example, gut microbiota – induced hsa_circ_0126925 targets BCAT2, a key enzyme in branched-chain amino acid metabolism, to facilitate CRC progression [176,177].

Collectively, these findings highlight ncRNAs as critical mediators of microbiota-driven epigenetic remodeling in CRC, providing promising avenues for biomarker discovery and therapeutic targeting [28]. The gut microbiota profoundly influences colorectal cancer by reshaping gene expression and the immune landscape through epigenetic regulation, offering novel avenues for early diagnosis and precision therapy.

Disruption of the immune microenvironment

The tumor immune microenvironment (TIME) refers to the complex and dynamic milieu surrounding tumor cells, composed of infiltrating immune cells, cytokines, chemokines, stromal components, and extracellular matrix [178]. Its functional state directly influences tumor initiation, progression, and therapeutic responsiveness. Under physiological conditions, the immune system exerts tumor surveillance by recognizing and eliminating transformed cells, thereby preventing malignant progression. However, in the CRC, this delicate equilibrium is disrupted, leading to the establishment of an immunosuppressive microenvironment that fosters immune evasion and promotes tumor growth [179].

The gut microbiota, the largest microbial community in the human body, plays a pivotal role in shaping host immunity through long-term coevolution with the immune system. Dysbiosis alters this symbiotic balance, reshaping the CRC tumor immune microenvironment via multiple mechanisms. Pathogenic bacteria release metabolites, toxins, and proteins that disrupt the intestinal epithelial barrier, trigger chronic inflammation, and modulate immune cell activity (Figure 4).

Figure 4.

Gut microbiota and the tumor immune microenvironment (TIME). Gut microbiota shapes the TIME through immune modulation. ETBF induces Th17 polarization via STAT3; fusobacterium nucleatum recruits myeloid-derived suppressor cells (MDSCs) and Tregs; SCFAs enhance regulatory T cell differentiation but also support antitumor immunity in some contexts. Microbiota-driven immune remodeling influences immune evasion and response to immunotherapy.

Current research efforts are therefore focused on developing combinatorial strategies that integrate conventional therapies with microbiota modulation to overcome CRC immune resistance. These include fecal microbiota transplantation (FMT), next-generation probiotics, and dietary interventions aimed at enriching immunostimulatory taxa and metabolites. Such approaches seek to reprogram the TIME toward an immunologically “hot” phenotype, characterized by enhanced antitumor immunity and improved therapeutic responsiveness. In the long term, a deeper mechanistic understanding of the microbiota – TIME axis will enable the development of precision microbiome-targeted therapies, providing new opportunities to synergize with immunotherapy and achieve durable clinical benefits in CRC management [180].

The crucial immune cells in TIME

Tumor-associated macrophages (TAMs)

TAMs are crucial regulators in the CRC microenvironment, primarily exhibiting an M2-like pro-tumor phenotype [181,182]. Through secretion of cytokines (e.g. TGF-β, IL-10), growth factors such as VEGF, and exosomes, TAMs promote angiogenesis, inhibit T cell function, and facilitate tumor progression and metastasis [183,184]. Targeting PD-1/PD-L1 can enhance phagocytosis and T cell activity, delaying tumor growth in a macrophage-dependent manner [185]. Therapeutic strategies aiming to reprogram M2 TAMs toward an M1 phenotype – using CD40 agonists (e.g. selicrelumab) [186,187], TLR3 agonists [188], or CSF-1 R inhibitors [189]—are under investigation. TAMs represent a key player and promising target in CRC immunotherapies.

Tumor-infiltrating T cells

T cells play a critical role in antitumor immunity, yet in the tumor microenvironment, they often become dysfunctional, characterized by elevated expression of inhibitory receptors such as PD-1, CTLA-4, LAG-3, and TIM-3, along with impaired production of effector molecules including IFN-γ, TNF, granzymes, and perforin [190]. Tregs exert inhibitory effects in the TME through a wide range of factors. These cells suppress the functions of effector T cells and natural killer (NK) cells, thereby promoting immune escape [191]. In CRC, CD4+FOXP3+ Tregs can be divided into two categories according to the level of FOXP3 expression. The prognosis of CRC with a large infiltration of FOXP3(lo) T cells is significantly better than that of CRC mainly infiltrated by FOXP3(hi) Tregs [192]. This may facilitate more precise CRC typing and treatment. Immune checkpoint blockade can reinvigorate these exhausted T cells, enhancing cytotoxic activity and antitumor responses [193]. Th17 cells exhibit dual roles in CRC: certain cytokines like IL-17 and IL-22 promote tumor growth via STAT3/NF-κB [194,195], while protective Th17 responses can inhibit colitis-associated tumorigenesis [196]. Their plasticity and regulation by cytokines (e.g. IL-1β, IL-6) and stromal interactions make them a potential therapeutic target [196,197]. Additionally, CD8+ T cells recognize tumor antigens via TCR – MHC I interactions and mediate cytotoxicity through granzyme, perforin, IFN-γ, and TNF-α [198].

Cancer-associated fibroblasts (CAFs)

CAFs are key stromal components in the CRC microenvironment, exhibiting functional heterogeneity across subtypes such as myofibroblastic, inflammatory, and pro-angiogenic populations [199]. CAFs promote immunosuppression by driving M2 macrophage polarization and inducing T cell dysfunction, partly through antigen presentation without costimulation [200]. They also facilitate ECM remodeling, metastasis, and treatment resistance via TGF-β, metalloproteinases, and cytokine signaling [201]. CAFs are prognostic markers and therapeutic targets in CRC, with strategies focusing on FAP, LOXL2, and IL1R1 [180].

Tumor-associated neutrophils (TANs)

TANs exhibit plasticity in the TME, polarizing into anti-tumor N1 or pro-tumor N2 phenotypes. In CRC, TANs often co-localize with CD8+ T cells and enhance their cytotoxicity, correlating with improved prognosis [202]. However, TAN-derived AGR2 promotes metastasis via CD98hc/xCT signaling and is linked to poor outcomes [203]. Recruitment and polarization of TANs are regulated by IL-8 (which also induces NET formation) and TGF-β—the latter promoting the N2 phenotype, while its inhibition favors N1 polarization [204,205]. TANs can suppress NK cell function through NET release and cholesterol competition [206]. A high neutrophil-to-lymphocyte ratio (NLR > 3) predicts poor response to immune checkpoint inhibition, even in MSI-H CRC [207]. Targeting TAN recruitment and polarization via CXCL8 or TGF-β represents a promising therapeutic strategy.

The influence of gut microbiota on immune cell function

ETBF

ETBF promotes colorectal tumorigenesis primarily through bacterially secreted BFT, which disrupts E-cadherin, activating β-catenin and NF-κB/Wnt pathways, increasing epithelial permeability and inducing mucosal inflammation [208]. ETBF also drives IL-17-dependent adenoma formation through STAT3 activation and Th17 polarization [209]. It recruits monocytic myeloid-derived suppressor cells and facilitates Th17 differentiation via exosomal miR-149-3p suppression, contributing to immune suppression and CRC progression [210]. Interestingly, ETBF may improve anti-PD-1 efficacy by recruiting IFNγ-producing CD8+ T cells [211].

Pks⁺ E. Coli

Pks⁺ E. Coli produces colibactin, inducing DNA double-strand breaks and mutations [212]. It is enriched in CRC, promotes tumorigenesis, and fosters an immunosuppressive TME by reducing CD3⁺ /CD8⁺ T cell infiltration, impairing anti-PD-1 efficacy [102]. Co-infection with ETBF may synergistically enhance CRC development [213].

Fn

Fn promotes colorectal carcinogenesis through its FadA adhesin binding to E-cadherin, activating β-catenin and NF-κB signaling pathways, thereby enhancing tumor proliferation and inflammatory responses [214]. Fn also facilitates CRC progression via the TLR4/NF-κB/miR-21 axis [215] and expands both MDSCs and Th17 cells, contributing to an immunosuppressive tumor microenvironment [216]. Notably, Fn upregulates PD-L1 expression through the cGAS – STING pathway and enhances infiltration of IFN-γ⁺ CD8⁺ T cells, thereby increasing sensitivity to anti-PD-L1 therapy [217].

Clostridioides difficile

Clostridioides difficile, particularly its toxigenic strains producing TcdA and TcdB, has recently been implicated in colorectal carcinogenesis. Chronic colonization promotes tumorigenesis in mice through TcdB-dependent activation of Wnt signaling, ROS generation, and induction of pro-tumor immune responses including IL-17-producing lymphocytes [218]. Secondary bile acids produced by commensal bacteria inhibit C. difficile spore germination and support colonization resistance [219]. Fecal microbiota transplantation (FMT) restores microbial diversity and secondary bile acid metabolism, effectively treating recurrent C. difficile infection [220]. Further investigation is needed to clarify its role in human CRC.

Salmonella enterica

Salmonella enterica, notably through its effector AvrA, enhances colorectal carcinogenesis by suppressing Wnt1 and inhibiting pro-apoptotic innate immune pathways via acetylation of MAPKKs, thereby facilitating tumor development [221,222]. Conversely, attenuated Salmonella strains show significant promise in cancer biotherapy. Engineered constructs such as shRNA-targeting IDO, flagellin B-expressing, and 4-1BBL-bearing Salmonella activate innate immunity, promote M1 macrophage polarization, enhance neutrophil and cytotoxic T cell infiltration, and reduce immunosuppressive cells, yielding potent antitumor responses in CRC models [223,224].

Peptostreptococcus anaerobius

Peptostreptococcus anaerobius is enriched in CRC patients and promotes tumorigenesis through multiple mechanisms. It activates PCWBR2–integrin α2/β1–PI3K – Akt – NF-κB signaling, stimulates TLR2/4–ROS – cholesterol synthesis, and produces the tryptophan metabolite IDA, which inhibits ferroptosis via the AHR – ALDH1A3–FSP1–CoQ10 axis [225,226]. Peptostreptococcus anaerobius also recruits MDSCs via CXCL1–CXCR2 and enhances their immunosuppression through Slamf4 binding, facilitating EMT and chemoresistance [227]. Targeting integrin α2β1, Slamf4, or the IDA – AHR pathway may overcome Peptostreptococcus anaerobius-mediated immunotherapy resistance [228], while probiotics like R. gnavus and B. producta can attenuate its tumorigenic effects [229].

Future perspectives

The gut microbiota, a complex and dynamic microbial ecosystem within the host intestine, has emerged as a critical determinant of CRC pathogenesis. Far from being a passive bystander, the microbiota functions as an active “hidden driver,” reshaping the molecular, epigenetic, and immunological landscapes of the host. Its influence extends across multiple dimensions: it contributes to the genetic foundation of carcinogenesis by inducing or accelerating driver mutations; it modulates epigenetic processes such as DNA methylation, histone modifications, and non-coding RNA activity, thereby rewiring transcriptional programs; and it reshapes the tumor immune microenvironment, fostering a permissive niche for immune evasion. These intertwined mechanisms act in concert to transform normal colonic epithelium into malignant tissue, underscoring the central role of the microbiota in CRC development and progression. Future investigations must dissect these mechanisms with higher resolution, as doing so will not only illuminate CRC etiology but also provide new opportunities for prevention, diagnosis, and therapy (Figure 5).

Figure 5.

Future perspectives on gut microbiota in CRC treatment. Future research will focus on multi-omics, single-cell, and spatial transcriptomics to uncover more detailed microbiota-host interactions. Clinically, microbial interventions (e.g. fecal microbiota transplantation) combined with immunotherapy (e.g. chemotherapy, radiotherapy) hold potential for improving CRC treatment outcomes.

At the genetic level, the gut microbiota contributes to CRC initiation by inducing genotoxic stress and promoting somatic mutations in key oncogenes and tumor suppressor genes. Certain bacterial strains, such as colibactin-producing Escherichia coli and enterotoxigenic ETBF, generate genotoxins that directly induce DNA double-strand breaks, oxidative stress, or chromosomal instability. These mutagenic insults provide the genetic foundation upon which clonal expansion and malignant transformation can occur [230]. Recent studies have shown that colibactin alkylates DNA and promotes p53 mutations, a hallmark of human colorectal tumors, thereby linking microbial metabolism with canonical cancer genetics. However, the contribution of microbial genotoxins is unlikely to be uniform across disease stages. It is plausible that DNA damage induced by microbes may act as an initiating event in early tumorigenesis, while subsequent tumor evolution relies on other microbiota-mediated processes such as epigenetic reprogramming and immune modulation[231]. Future research should apply long-read sequencing and lineage-tracing technologies to map the temporal sequence of microbial-induced mutations in vivo. Moreover, a deeper understanding of host genetic susceptibility – such as polymorphisms in DNA repair pathways that may amplify the carcinogenic effects of microbiota – will be critical for defining high-risk populations. Integration of host genetics with microbial functional genomics may ultimately allow for personalized risk prediction and targeted prevention strategies.

Beyond mutations, the gut microbiota profoundly influences CRC through epigenetic regulation. Microbial metabolites such as SCFAs act as histone deacetylase inhibitors, thereby modifying chromatin accessibility and transcriptional programs. Butyrate, for instance, is known to induce histone hyperacetylation at promoters of tumor suppressor genes, leading to anti-proliferative and pro-apoptotic effects in colonocytes [232]. Conversely, depletion of SCFA-producing bacteria has been associated with aberrant histone modifications and enhanced tumorigenic potential. DNA methylation patterns are also reshaped by microbial products. Dysbiosis often correlates with hypermethylation of CpG islands within promoters of tumor suppressor genes, leading to their transcriptional silencing. For example, Fn infection has been linked to CpG island methylator phenotype (CIMP)-positive tumors, suggesting that microbial colonization may underlie distinct epigenetic subtypes of CRC [134]. ncRNAs is another layer of epigenetic regulation influenced by the microbiota. Microbial colonization alters intestinal epithelial miRNA profiles, which in turn modulate host gene expression and even interkingdom communication. Dysregulation of oncogenic miRNAs, such as miR-21 and miR-155, has been linked to microbial infection and promotes pathways [170,233]. Recent studies have also identified circRNAs as microbial targets; for instance, gut microbiota – induced circRNA hsa_circ_0126925 modulates branched-chain amino acid metabolism to accelerate CRC progression [176,177]. Future work should leverage single-cell epigenomics and spatial transcriptomics to capture the precise cell-type – specific and spatially resolved effects of microbial signals on host chromatin. Such efforts could reveal stage-specific and context-dependent epigenetic vulnerabilities that may be exploited for therapy. Furthermore, ncRNAs shaped by the microbiota hold potential as both biomarkers and therapeutic targets, as their expression can be detected in stool or plasma and manipulated through synthetic mimics or inhibitors [234].

A third critical dimension of microbiota – CRC interplay lies in immune regulation. The intestinal immune system is finely tuned to maintain tolerance toward commensal microbes while remaining vigilant against pathogens. Disruption of this balance can generate a pro-inflammatory microenvironment that fuels carcinogenesis. Pathobionts such as Fn not only trigger chronic inflammation but also directly impair anti-tumor immunity. SCFAs, particularly butyrate and propionate, promote regulatory T-cell differentiation and enhance anti-inflammatory responses. Microbiota also modulate antigen presentation and influence the efficacy of immune checkpoint blockade (ICB) therapy. Clinical studies have demonstrated that patients with favorable microbial signatures – characterized by high abundance of Akkermansia muciniphila and certain Ruminococcaceae—exhibit improved responses to PD-1 inhibitors. These findings underscore the potential of microbiota modulation as a strategy to enhance immunotherapy outcomes in CRC. Future research should map the “microbiota – immune – tumor axis” with greater mechanistic clarity. Identifying microbial taxa or metabolites that drive immune resistance could inform the development of microbiota-based adjuvants for immunotherapy. Rationally designed interventions – ranging from next-generation probiotics to fecal microbiota transplantation (FMT) – should be standardized through rigorous donor selection, safety profiling, and mechanistic validation. Combining microbiota modulation with ICB, chemotherapy, or radiotherapy may prove synergistic by simultaneously targeting genetic, epigenetic, and immunological vulnerabilities of CRC.

From a translational perspective, the integration of microbiome science into CRC management holds immense promise. In diagnostics, stool-based microbial biomarkers – such as enrichment of Fn and depletion of SCFA producers – may enable noninvasive early detection. Artificial intelligence and machine learning applied to large, multicenter cohorts could refine predictive models with high sensitivity and specificity, paving the way for microbiome-informed risk stratification [235]. In therapeutics, precision microbiome engineering represents a new frontier. Engineered bacterial consortia, designed to deliver anti-inflammatory molecules or oncolytic compounds directly within the colon, could complement existing therapies. Moreover, advances in synthetic biology may allow for the design of “smart probiotics” capable of sensing tumor-associated metabolites and releasing targeted therapeutics in situ. Nevertheless, challenges remain. The inter-individual variability of microbiota composition poses significant hurdles for reproducibility and generalization. Standardized methodologies for microbiome sampling, sequencing, and functional annotation are urgently needed. Furthermore, ethical and regulatory frameworks must evolve to address safety concerns associated with FMT and engineered microbes.

Conclusion

The gut microbiota reshapes CRC biology at the levels of genetic initiation, epigenetic reprogramming, and immune regulation. These multidimensional interactions render the microbiota not merely a modulator but a central architect of colorectal carcinogenesis. The next decade should focus on integrating microbiome insights with precision oncology, developing microbiota-informed diagnostics, and engineering therapeutic strategies that harness microbial ecology for clinical benefit. Ultimately, by bridging basic mechanistic understanding with translational innovation, the microbiota may transform CRC from a late-diagnosed and treatment-refractory malignancy into a disease with actionable windows for early detection, interception, and cure.

Funding Statement

The author(s) reported there is no funding associated with the work featured in this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All data are presented in the manuscript.