During acute inflammation, RONS from activated immune cells can promote tissue repair and regeneration in addition to their pathogen-killing effects. However, recurrent histological damage and RONS repair processes also induce permanent DNA damage (including single- and double-strand breaks, nucleotide modifications and abrogation sites) and genomic instability in proliferating intestinal epithelial cells (4).

Apart from oxygen free radicals produced by inflammatory cells, extracellular free radicals from intestinal bacteria significantly increase DNA damage in co-clonal epithelial cells, possibly associated with CRC-associated chromosomal instability (CIN) (73).

Furthermore, it has been shown that there is a positive feedback relationship between inflammation and genomic instability (74). Inflammation is attributed to mutations that damage DNA through the production of RONS, exacerbating the inflammation in turn. Despite being influenced by complex signalling networks and DNA repair mechanisms, it usually leads to carcinogenesis under long-term chronic inflammatory conditions (75,76).

The accumulation of ROS and RNS has been observed in inflammatory tissues of patients with active UC and CD (75). If RONS disrupt proto-oncogenes in the intestinal epithelium, it may cause alterations in the genetics and epigenetics of heritable intestinal epithelial cells.

The primary genetic alterations in IBD-CRC differ in frequency and timing from those in sCRC, which are thought to be closely related to genomic damage induced by oxidative stress. Metagenomic studies have found that proto-oncogene p53 mutations and loss of heterozygosity occur early in IBD-CRC (77). P53 loss of function appears to be an early event in initiating IBD-CRC pathogenesis (78). P53 critically impacts cell proliferation and prevents clonal expansion of mutant cells. Besides being found in >80% of patients with IBD-CRC (79), p53 mutations have been found in inflamed mucosa of 50% of patients with UC without cancer (80,81). In vitro studies have shown that the acquisition of mutant p53 (mutp53) function is manifested in DSS-induced colitis mice by the rapid emergence of flattened developmentally abnormal lesions that can progress to invasive carcinomas, accompanied by the accumulation of mutant p53 and enhanced activation of NF-κB. This change is similar to the developmental pattern of human IBD-CRC and may explain the early appearance of p53 mutations in human IBD-CRC (30).

Loss of function of the oncogene APC and the proto-oncogene KRAS also occur as early events in IBD-CRC but with a much lower prevalence than in sCRC (82). APC encodes an essential member of the β-catenin disruption complex. This complex critically attenuates the nuclear localisation of β-catenin and its binding to the TCF family of transcription factors, promoting Wnt signalling. Amplification of MYC proto-oncogene occurred in 26% of IBD-CRC samples compared to 26% of sCRC samples. In addition, isocitrate dehydrogenase 1 mutations were significantly more common in IBD-CRC and rare in sCRC.

Inflammation is instrumental in promoting gene mutations and genomic instability (74). Inflammation-driven genetic alterations, along with changes at the epigenetic level, may have an essential role in the tumourigenesis of CRC, particularly in IBD-CRC (83). Similar to sCRC, the major oncogenic molecular pathways in IBD-CRC include CIN, microsatellite instability (MSI) and CpG island methylator phenotype.

CIN is usually triggered by mutations in oncogenes followed by a series of duplications that ultimately lead to an altered chromosome number (aneuploidy) and chromosomal structural abnormalities (somatic copy number alterations, deletions, insertions and amplifications) (84) and possible loss of oncogene heterozygosity and chromosomal rearrangements. CIN is considered a significant feature of the pathogenesis of CAC and is found in 80% of cases (74,76). CIN is associated with a progressive accumulation of mutations in pro-oncogenes and tumour suppressor gene. Aneuploidy is a consequence of CIN and occurs before dysplasia in colonic mucosal biopsy samples from patients with IBD (85).

Furthermore, there was a statistically significant relationship between increased aneuploidy and histological progression of dysplasia/carcinoma in UC (78). This observation highlights the progressive genomic instability during IBD-CRC tumour progression.

MSI is thought to be a genomic microsatellite hypermutation phenotype due to faulty operation of the DNA mismatch repair (MMR) mechanism. Following the inactivation of MMR genes, mutations continually accumulate in the cancer genome as there is a lack of ability to repair post-DNA replicative base substitutions or slippage mistakes at microsatellite sequences. With defective MMR, there has been a significant increase in mutated genes that have not been able to be correctly repaired and this ability to increase mutation frequency is thought to shorten the typical time frame for adenoma-to-carcinoma formation from 1–2 decades to 1–2 years (86).

Damage to the DNA MMR system induced by oxidative stress would lead to the accumulation of MSI phenotypes in the colonic mucosa of patients with IBD-CRC. The MSI phenotype can be observed in ~15% of patients with IBD-CRC, a rate similar to that of sCRC (87). In patients with IBD-CRC, high MSI is closely associated with hMLH1 hypermethylation and hMSH2 expression deficiency, two mismatch repair genes (88). In IBD-CRC, proximal tumours with high mutation rates correlate with MSI and have a higher predicted neo-epitope load (89). Such tumours tend to have increased immunogenicity and may respond better to immunotherapy.

The CpG island methylation phenotype itself overlaps with MSI pathogenesis. Distinguished from MSI, the hypermethylation pathway is characterised by the CpG island methylator phenotype, resulting in the epigenetic silencing of several genes, particularly tumour suppressor genes. Several studies have explored the hypermethylation pathways in patients with IBD-CRC. For instance, promoter hypermethylation of p16^INK4a and p14^ARF, which are expected early events in IBD-CRC carcinogenesis, was found in 100 and 50% of UC-CRC, respectively, and these two genes have a crucial role in cell cycle inhibition (90). In addition, CDH1 promoter methylation was detected in 93% of patients with IBD with colonic dysplasia, compared with 6% of patients with IBD without dysplastic lesions (91). This gene is involved in biological processes that are dysregulated in cancer, such as the sequence of cell sorting, migration and differentiation. Death-associated protein kinase (DAPK), a tumour suppressor gene, also appears to undergo epigenetic modifications during IBD-CRC carcinogenesis. A study using surgical resection specimens from patients with UC for DAPK epigenetic analysis concluded that DAPK silencing induced by promoter hypermethylation may lead to the accumulation of mucosal DNA damage in UC and, therefore, may contribute to the development of UC-associated CRC (92).

A transcriptome-based analysis of tumour subtypes showed that the typical epithelial tumour subtype associated with Wnt signalling was absent in IBD-CRCs; instead, a mesenchymal stroma-rich subtype predominated (55). IBD-CRCs exhibit a switch from the epithelial consensus molecular subtype (CMS)2 to the more mesenchymal CMS4 phenotype [epithelial-mesenchymal transition (EMT)]. Dysregulation of Wnt signalling activates transforming growth factor (TGF)-β and results in an ‘immuno-inflammatory’ inhibitory microenvironment enriched with CD4+ cells. EMT involves loss of tumour cell polarity and cell-cell adhesion to enhance cell migratory and invasive properties (93).

In IBD-associated CRC, polymeric immunoglobulin receptor (PIGR) and cytokine oncostatin M (OSMR), which are involved in mucosal immunity, are dysregulated by epigenetic modifications. Low expression of PIGR may promote epithelial barrier dysfunction and persistent inflammation. Increased OSMR signalling may then favour the establishment of mesenchymal CRC subtype in patients with IBD (77). A previous single-cell analysis showed the presence of an activated mesenchymal cell population in the mesenchyme surrounding the colonic crypts of patients with IBD. This subpopulation expresses TNF superfamily member 14, fibroblastic reticulocyte-associated genes, IL-33 and lysyl oxidase. In addition, it induces factors that impair epithelial proliferation and maturation and contribute to oxidative stress and disease severity in vivo (94). Whether this process may contribute to the cancerous process requires further exploration.

Inflammation-promoting macrophages are a central and potent component of innate immunity. During colon inflammation, intestinal epithelial cells stimulated by lipopolysaccharide (LPS) release chemokines (e.g., C-X-C motif chemokine ligand-2 and C-C motif chemokine ligand 2) and recruit large numbers of monocytes and neutrophils (38). Massive macrophages infiltrate the intestinal lamina propria, triggering polarisation towards M1 or M2 in response to continuous exposure to inflammatory microenvironmental signals (97,98). Bacterial products and Th1 cytokines induce macrophage polarisation towards pro-inflammatory, cytotoxic and antigen-presenting M1, whereas Th2 cytokines polarise macrophages towards immunomodulatory, pro-angiogenic M2 (99).

TAMs are the most abundant myeloid cell population infiltrating solid tumours, predominantly displaying an M2 phenotype as crucial for promoting tumour inflammation and mediating immunosuppression (100). TAMs have been found to have poor antigen-presenting capacity and limited anti-tumour responses. Reduced killing capacity of TAMs has been associated with reduced iNOS expression (101).

The tumour-promoting effects of TAMs are multifaceted. TAMs infiltrated in the TME, on the one hand, promote the formation of the inflammatory microenvironment through a large amount of released inflammatory mediators (e.g., IL-6, TNF-α and RONS) (102); on the other hand, they further provide proliferation signals through the supply of mitogenic growth mediators [e.g., epidermal growth factor (EGF) and TGF-β] to maintain the uncontrolled proliferation of epithelial and mesenchymal cells (103). In addition, TAMs have recruitment effects on immunosuppressive cells such as myeloid-derived suppressor cells (MDSCs) and Tregs and inhibitory effects on T cells and natural killer (NK) cells, resulting in the suppression of immune activity in the local microenvironment (104,105). In the early stages of IBD-CRC, monocyte-like macrophages are recruited into the local microenvironment and promote the expansion of Th17 cells by producing IL-1β, which further exacerbates dysbiosis and inflammation, leading to a vicious cycle and contributing to tumourigenesis (102).

Similar to TAMs, MDSCs [including monocytic MDSCs and polymorphonuclear MDSCs (PMN-MDSCs)] are also premature myeloid cells. Although they overlap with monocytes and granulocytic myeloid cells, they represent two cell populations with distinct immune phenotypes. MDSCs can suppress both adaptive and innate immunity. Infiltration of MDSCs may promote tumourigenesis on the one hand through activation of the STAT3 and NF-κB pathways (106), and on the other hand, may influence the function of other cells through chemokines. MDSCs, which express C-X-C motif chemokine receptor 2, can directly inhibit cytocidal cytotoxicity of NK cells and CD8+ T cells (107,108), accelerating tumour growth (109). Furthermore, C-C motif chemokine receptor 5-expressing MDSCs can also induce local aggregation of Tregs in tumours (110), promoting immunosuppression. Mouse models with breast cancer have additionally shown that MDSCs can affect tumour immunity by increasing IL-10 production and inhibiting macrophage secretion of IL-12 (111).

T cells that promote immunosuppression. Tumour-infiltrating T cells have been associated with improved clinical prognosis and survival of patients with CRC (112,113). Tumour-killing type 1 T-helper cells, CD4+ T cells and CD8+ T cells have anti-tumour effects. However, these types of cells are regulated by other immunosuppressive cells in the tumour microenvironment, which, to a certain extent, maintains the local inflammatory state and leads to an immunosuppressive state in the local microenvironment.

The IL-23/Th17 signalling pathway has an important role in inflammation-associated tumours. IL-23 produced by inflammatory dendritic cells infiltrated in the tumour microenvironment induces the polarisation of Th17 cells. Tumour-infiltrating γδT17 cells further activate downstream inflammatory signals by secreting inflammatory factors such as IL-17A. In addition, IL-8, TNF-α and granulocyte-macrophage colony-stimulating factor secreted by Th17 cells chemotactically attract immunosuppressive PMN-MDSC and maintain immunosuppressive activity (114–116).

Treg cells modulate adaptive immunity, suppressing inflammation in chronic inflammatory diseases. Treg cells promote the clearance of pathogenic bacteria by establishing a protective immune response during intestinal infections (117,118). However, the cytotoxic effects of Treg cells on antitumourigenic CD8+ T cells have suggested that these cells are immunosuppressive in the tumour environment (119), and different subpopulations of Treg cells exhibit other immune properties in an environmentally dependent manner.

Massive infiltration of RAR-related orphan receptor (ROR)γt+Foxp3+ cells has been observed in the intestinal lamina propria of UC and colon cancer (120), a subpopulation of Foxp3+ Treg cells induced in a chronic inflammatory environment. This subpopulation of Tregs can exhibit certain features of Th17 cells, such as secretion of IL-17A, which has been shown to have a pathogenic role in the pathogenesis of infiltrative CRC (121). In addition, this group of cells promotes the production of inflammatory cytokines, such as IL-17A, IL-1, IFN-γ and TNF-α, by colitis cells (120), which supports local inflammatory responses and oxidative stress insults.

Treg cells can perform an immunosuppressive function by inhibiting the anti-tumour response of CD4+ T cells and CD8+ T cells, contributing to the formation of an immunosuppressive tumour microenvironment (122). The co-expressed transcription factors Foxp3 and RORγt on Treg cells can drive IBD-CRC growth by activating STAT3 in tumour cells and promoting uncontrolled expression of IL-6 in tumour-infiltrating dendritic cells (38).

Mesenchymal cells that promote immunosuppression. Mesenchymal stromal cells in tumours, also known as CAFs, are a significant component of the tumour stroma, which secrete growth factors and inflammatory ligands. In the tumour microenvironment, CAFs with distinct phenotypes are activated by bacterial products (e.g., LPS) and cytokines (e.g., TNF-α) (123) to promote tumour progression through the expression of a wide range of growth and inflammatory factors [e.g., IL-8, vascular endothelial growth factor (VEGF) and fibroblast growth factor 2] (124–127), resulting in a sustained inflammatory environment. Factors from diverse cellular and tumour sources shape the phenotype of heterogeneous CAFs (128). Similar to the polarisation of TAMs, CAFs show a high degree of plasticity. The function of CAFs has been shown to impact the development of the tumour microenvironment substantially and has been found to have partial anti-tumour activity. Studies on the role of fibroblasts in the IBD-CRC model are summarised in Table III (14,39,42,123–130).

The gut microbiota and its metabolites have a multifaceted role in inhibiting inflammation, maintaining the mucosal barrier and modulating immunity during interactions with the host. The gut microbiota can produce a wide range of metabolites using exogenous undigested dietary substrates (e.g., dietary fibre) and endogenous compounds [e.g., bile acid (BA) metabolites] to provide energy and nutrients to the host. Certain bacterial species, such as Bifidobacteria, are involved in the biosynthesis of various components such as vitamin K and B (135).

Several bacteria, such as Firmicutes, Bacteroidetes and certain anaerobic gut microorganisms, can provide short-chain fatty acids (SCFAs) with biologically active components through the fermentation of dietary fibre (136). SCFAs produced in the gut are rapidly absorbed by host epithelial cells. They are a significant source of energy for colonic epithelial cells and have systemic immunomodulatory and anti-inflammatory effects (136). The most abundant SCFAs in the colon are acetate, propionate and butyrate (13,15,16), which promote the proliferation of beneficial bacteria, stimulate Treg cells and macrophages to reduce the production of inflammatory mediators and increase the oxygen consumption of colonic epithelial cells, resulting in the enhancement of immunomodulation and intestinal barrier function (137). Butyrate has been found to enhance intestinal barrier function by activating AMPK, Akt and other signalling pathways to promote tight junction assembly in a dose-dependent manner (138). Furthermore, during colitis, butyrate induces T cell-independent immunoglobulin A (IgA) secretion in the colon through activation of free fatty acid receptor 3 [also known as G protein-coupled receptor (GPR)41] and carboxylic acid receptor 2 (also known as GPR109A), as well as inhibition of histone deacetylase, which restores the epithelial barrier function under inflammatory conditions (139).

In addition to producing substances that regulate energy metabolism in colonic epithelial cells, the gut microbiota can synthesise several neurochemicals that can affect both the central nervous and peripheral gut systems. For instance, γ-aminobutyric acid (GABA) is an important inhibitory neurotransmitter in the brain and numerous neuropsychiatric disorders have been linked to GABA dysfunction, which is also thought to be connected to the brain-gut axis connection.

The gut microbiota has also reportedly been involved in synthesising substances such as BA, cholesterol and conjugated fatty acids and branched-chain amino acids (140). Bile salts escaping during enterohepatic circulation become substrates for gut microbial metabolism. Bacteria, including Clostridium, Bifidobacterium, Bacteroides, Listeria and Lactobacillus, are involved in deconjugating bile acids, which is vital for the gut microbiota to mediate immune regulation. On the one hand, bile acids can produce direct antimicrobial effects by disrupting bacterial biomembrane. On the other hand, the bile acid receptors farnesoid X receptor (FXR) and Takeda G protein-coupled receptor 5 can be activated by bile acid derivatives, which are essential for the regulation of intestinal inflammation and tumourigenesis. For example, activated FXR signalling suppresses CRC progression by antagonising the Wnt/β-catenin pathway and inhibiting cytokine signalling 3 gene transcription.

Another significance of the gut microbiota for the host is its fundamental role in the differentiation of the host's immune system. Symbiotic bacteria in the gut are able to not only competitively inhibit pathogenic pathogens in the gut, but also indirectly defend against pathogens by activating the immune response. For instance, LPS and flagellin derived from the gut microbiota enhance the expression of antimicrobial peptide and RegIIIγ from epithelial cells by stimulating Toll-like receptor (TLR)4+ stromal cells and TLR5+ CD103+ dendritic cells (141). The gut microbial metabolite butyrate modulates T-cell differentiation and proliferation. It enhances Treg cell function and induces their differentiation (142), inhibits IL-17 levels as well as Th17 cells in peripheral blood and colon tissues of rats with 2,4,6-trinitrobenzenesulfonic acid-induced colitis (143), and suppresses the proliferation of CD4+ and CD8+ T cells in a dose-dependent manner (144).

Host recognition of microbial metabolites relies on pattern recognition receptors (PRRs), essential for initiating the innate immune response (145). One of the most well-studied PRRs is the TLR, a transmembrane protein expressed in intestinal epithelial cells and innate immune cells. TLRs have an essential role in activating the innate immune mechanism and maintaining the integrity of the intestinal epithelial barrier by recognising highly conserved structural molecules expressed by microbial pathogens (146). TLR4 pre-expression is upregulated in intestinal epithelial cells during UC development and is further increased in CAC, suggesting an essential role in the transition from inflammation to cancer (147,148).

The gut microbiota controls the levels of antimicrobial peptides and immunoglobulin IgA by regulating the differentiation of intestinal lymphoid tissues and the number of lymphocytes, which then handles the development and maturation of the immune system.

Dysregulation of the interaction between the host and its microbiota leads to a loss of host immune tolerance (149), which induces immune responses. This aberrant immune response is thought to link inflammation to cancer (136). For instance, an increased abundance of pathogenic bacteria that can cling to the intestinal epithelium correlates with increased intestinal permeability. Alterations in the diversity and composition of the gut microbiota would lead to the induction of intestinal inflammation by regulating the expression of inflammatory genes (150).

In patients with IBD with an impaired mucus layer of the gastrointestinal tract, the luminal microbiota can penetrate the epithelial cells, causing proliferative and inflammatory processes (151). The microbiota activates the host's innate and adaptive immune response by producing metabolites that act as antigens, which cause inflammation and promote tumour growth. Mucous membranes under the influence of inflammatory damage are more susceptible to bacterial, particularly pathogenic, stimulation, which causes the submucosa to be exposed to more antigens, which leads to a vicious, positive feedback loop of mucosal damage (152).

The onset and progression of CRC do not depend on the species prevalent but on the entire metabolic functional pathway of the microbiota. Metagenomic studies have observed altered diversity and abundance of the gut microbiota in patients with IBD and CRC. In a recent survey of patients with CAC, Richard et al (153) reported that the CAC group had a significantly altered gut microbiota and reduced α-diversity compared to a healthy population and patients with sporadic tumours. Fusobacterium and Ruminococcus genera appeared to be significantly reduced in patients with IBD-CRC compared to those with sCRC (154). Indeed, evidence suggests that patients with CAC may experience changes in the composition of their gut microbiota at different stages. Specifically, CAC can show a significant increase in Akkermansia, Fusobacterium, Peptostreptococcus, Streptococcus and Ruminococcus at advanced stages, while Granulicatella and Lactobacillus were reduced (155).

Some of these genera have been shown to contribute to colorectal carcinogenesis through direct oncogenic effects or by participating in or promoting chronic inflammation-mediated epithelial cell damage (136). Gut microbes differentially affect DNA damage, DNA methylation, chromatin structure and non-coding RNA expression in CRCs. Several genes and pathways altered by gut microbes have been implicated in the development of CRC, particularly those involved in cell proliferation and Wnt signalling.

Fusarium nucleatum is associated with developing cancers of the oral cavity and gastrointestinal tract. FadA inhibits the tumour suppressor activity of E-cadherin by binding to E-cadherin in epithelial and malignant cells, leading to increased levels of β-catenin-regulated transcription. This process leads to the expression of inflammatory molecules, such as NF-κB, IL-6, IL-8 and IL-18 and activation of TLR2/TLR4 (156). It has also been shown that F. nucleatum accelerates DNA methylation of cancer-specific genes in UC patients and inhibits cytotoxicity of NK cells through the expression of Fap2 protein, a lectin expressed by F. nucleatum (157).

Enterotoxigenic Bacteroides fragilis (ETBF) is positively associated with active IBD and CRC. Activation of Wnt/β-linker protein and NF-κB signalling by secretion of B. fragilis toxin cleaves E-calmodulin. This cleavage increases the permeability of the intestinal barrier and the signal transduction of E-cadherin and β-catenin in intestinal epithelial cells, giving the organism the potential for co-clonal oncogenic transformation (61). Furthermore, ETBF induces the STAT3 pathway and the Th17 cancer pathway by driving the local infiltration of Tregs, which promotes the pre-expression of inflammatory factors (e.g., IL-17, IL-8) and establishes an immunosuppressive tumour microenvironment, contributing to the development of IBD-CRC (158–161).

The polyketide genotoxin colistin, produced by E. coli via polyketide synthase (Pks), can cause DNA damage upon contact with epithelial cells and acts synergistically with host inflammation to create a tumour-promoting microenvironment (162). Colonic inflammation has been further shown to promote the genotoxic effects of Pks+ E. coli and facilitate E. coli colonisation of the mucosa, leading to an increase in colistin-induced DNA damage to colonic epithelial cells, which allows this bacterial strain to exert its oncogenic activity (162).

In the stool flora of patients with CRC, Enterococcus faecalis is significantly more abundant than in healthy individuals. The possible pathogenic role of E. faecalis in CRC is multifaceted. On the one hand, E. faecalis activates macrophage MMP-9, which disrupts the integrity of the intestinal monolayer and induces an inflammatory response in the intestine. On the other hand, ROS and extracellular superoxide produced by E. faecalis lead to genomic instability, causing damage to colonic DNA and inducing mutations that can lead to cancer (163).

It has been suggested that genotoxic indoleamine derived from Morganella morganii may affect multiple aspects of host biology, including direct DNA damage activity, triggering or exacerbating inflammatory processes, and cancer in abiotic mice (164). Bacteria produce genotoxicity that can increase species' competitive ability to grow in an inflammatory microenvironment, affecting microbiota composition. Dysbiotic microbiota will further contribute to the vicious cycle of inflammation and DNA damage in patients with IBD and ultimately lead to tumourigenesis.

The current consensus is that dysbiosis is present in both IBD and CRC and that dysbiosis may lead to disruption of the mucosal barrier and persistence of inflammation and cancer. It is thought that dysbiosis leads to an increase in bacteria such as E. coli and ETBF, which leads to a breakdown of the intestinal mucosal barrier, allowing more bacteria to move from the lumen of the bowel to the inside of the tissue. This condition leads to chronic tissue inflammation and the release of inflammatory and pro-cancer mediators increases the risk of developing CRC. This positive feedback loop of dysbiosis may underlie the inflammation-abnormal proliferation-cancer sequence. Further studies are required to elucidate the impact of host factors and pathogenic microbial interactions in patients with IBD-CRC.