Diet, inflammation and microbial genotoxins: modifiable drivers of colorectal cancer

Article Review: Thakur BK, Malaise Y, Choudhury SR, Neustaeter A, Turpin W, Streutker C, Copeland J, Wong EO, Navarre WW, Guttman DS, Jobin C. Dietary fibre counters the oncogenic potential of colibactin-producing Escherichia coli in colorectal cancer. Nature Microbiology. 2025. doi: 10.1038/s41564-025-01938-4.

Colorectal cancer (CRC) is now the leading cause of death in individuals below 50 years of age in the United States.1 While dietary fiber has been a topic of contention, the verdict is in with this paper from NatureMicrobiology. Thakur and colleagues present comprehensive results from meticulously performed experimental studies, describing the link between a diet low in fiber, bacteria producing a CRC-causing toxin and an increased risk of colonic neoplasia. Interestingly, the bacterial species that they focus on is a CRC-associated Escherichia coli, with a polyketide synthase island (polyketide synthase–positive (pks⁺) E. coli), encoding for a genotoxin called colibactin that causes DNA damage.2 Although these bacteria have been detected in 20% of healthy individuals, their enrichment in 40% of inflammatory bowel disease (IBD) patients and approximately 60% of CRC patients, suggests that host and environmental factors may be required to unmask their oncogenic potential.3

The experimental model and findings:

A murine CRC model was used to evaluate the effect of different diets and colibactin-producing E. coli on colonic polyp formation. When mice were fed either a low carbohydrate diet deficient in fiber, standard rodent chow or a high fat, high sugar Western style diet, a greater number of colon polyps developed with the low carbohydrate/low fiber diet, which further increased when animals were infected with colibactin-producing E. coli. In this cohort, gut colonization with these bacterial species was also higher, than with the other two diets. An increased abundance of Enterobacteriaceae accompanied by a decrease in Bifidobacterium, Turcibacter and Marvinbryantia was observed. The latter three bacteria are known to exert an anti-inflammatory effect, partly through their production of short chain fatty acids (SCFAs) during fermentation of soluble fiber.4,5 Consumption of a low carbohydrate/low fiber diet was also associated with thinning of the mucus layer and reduced transcription of Muc2, a key mucin-encoding gene. Regardless of infection status, mice fed the low carbohydrate/low fiber diet exhibited increased colonic inflammation, characterized predominantly by neutrophils and Th17 cells. Nitrate levels in the lumen of the colon were higher in these mice as compared with those fed normal chow. This intraluminal nitrate that favors Enterobacteriaceae, is derived from nitric oxide produced by the enzyme inducible nitric oxide synthase(iNOS) in response to inflammatory stimuli.6 Consistently, mice on the low fiber diet showed elevated expression of Nos2 -the gene encoding iNOS- in both colonic tissue and epithelial cells.

Peroxisome proliferator-activated receptor gamma (PPAR-) is a nuclear receptor that is highly expressed in the colon. It is of particular significance in the colon, owing to its anti-inflammatory effect, inhibiting the expression of pro-inflammatory genes. As mentioned earlier, SCFAs are produced by fermentation of soluble fiber by intestinal bacteria. These in turn are recognized by PPAR-, which downregulates iNOS expression and prevents expansion of Enterobacter in the mouse colon.6 Mice fed the low carbohydrate/low fiber diet had lower colonic transcript levels of both PPAR- and Angpt14, a gene controlled by PPAR-. Consistently, addition of the PPAR- agonist, rosiglitazone to the diet led to decreased intraluminal nitrate levels, decreased colonic Nos2 expression, reduced colonization with the colibactin-producing E. coli as well as reduced polyp formation. The authors further demonstrated a reversal of this cascade by adding inulin, a soluble fiber, that restored PPAR- signaling and decreased colonic inflammation.

Fecal calprotectin and E. coli levels were evaluated in stool samples from individuals participating in the “Diet Intervention Examining the Factors Interacting with Treatment Success” (DIETFITS) study.7 Among individuals assigned to a low-carbohydrate diet, fecal calprotectin levels were increased at both 3 and 12 months, while E. coli was elevated at 3 months compared with baseline. In the Genetic Environmental Microbial (GEM) cohort, a prospective cohort of first-degree relatives of IBD patients- fecal calprotectin negatively correlated with fiber intake and showed a positive correlation with E. coli.

In mice lacking the mismatch repair gene Msh2, exposure to colibactin-producing E. coli resulted in a greater burden of colonic and small intestinal polyps, compared with exposure to non-colibactin-producing strains. This was accompanied by an increased number of cells being positive for phosphorylated histone, a proxy for double-stranded DNA damage. These changes were linked to the development of a senescence-associated secretory phenotype, a finding supported by treatment of the mice with a senolytic agent, that selectively induced apoptosis in senescent cells. This led to a reduction in the number and size of polyps, reinforcing the role of host genetics in this nexus.

Although most of the data presented in this study are derived from animal models, the experiments have been carefully performed with appropriate controls. In many cases, mechanistic findings have been confirmed by utilizing inhibitors or agonists, thus corroborating suspected effects. The appeal of their results centers around two very easily modifiable environmental risk factors for CRC, namely, diet and the microbiome. Studies reporting dietary recommendations and their impact on CRC-specific mortality and survival to date have been heterogeneous.8

Dietary fiber can be broadly categorized into soluble and insoluble forms. Soluble fiber is fermented by the intestinal microbiota to generate SCFAs, whereas insoluble fiber increases stool bulk and accelerates colonic transit. To summarize, increased production of SCFAs optimizes PPAR- signaling, reducing intraluminal nitrate levels and Nos2 transcription, thereby attenuating colonic inflammation. This milieu also decreases the CRC-associated bacteria while favoring colonization of SCFA-producing bacterial genera. This self-perpetuating cycle thus ultimately culminates in a lower incidence of colonic polyps and colorectal cancer.

Recent findings have highlighted the urgency to overcome the obesity epidemic for CRC prevention.9 This is the need of the hour, given the current US trends in obesity- increasing from 19.3% in 1990 to 42.3% in 2022, with a projected rise to 46.9% by 2035.10 High dietary fiber intake by obese patients has been shown to lower all-cause mortality, along with a reduction in the incidence of obesity itself.11 Studies emphasize the need to characterize fiber-gut microbiota interactions to combat obesity by fiber-based precision nutraceuticals.12 This further highlights the clinical relevance of the mechanistic findings illustrated in this manuscript.

The therapeutic potential of SCFAs and the need to modulate microbiota has been explored in IBD as well.13 This, coupled with the above data linking a low-carbohydrate, low-fiber diet to reduced mucus layer thickness and decreased Muc2 transcription suggests, that supplementation with soluble fiber may warrant consideration as a treatment strategy for patients with IBD.

Notably, this study also ties in with a growing body of evidence showing an association between types of bacteria, their metabolites and favorable responses to checkpoint immunotherapy, thus opening avenues for dietary interventions.14 This is particularly significant, especially since in this model, a MMRd host genetic background, which is inherently more responsive to immunotherapy, potentiates the genotoxicity of the bacteria.

A recent study from China reportedly decreased the burden of early-onset CRC by increasing dietary fiber intake.15 This is an exciting prospect and the findings from this study may lay the foundation for modifying lifestyle and environmental conditions to mitigate CRC.

This study should be viewed against the background of interpersonal diversity of the gut microbiome. This model focuses on one pks⁺E. coli strain; however, it is worthwhile to note that other CRC-associated bacteria such as Streptococcus bovis, enterotoxigenic Bacteroides fragilis, Fusobacterium nucleatum, Enterococcus faecalis, Escherichia coli, and Peptostreptococcus anaerobius play a crucial pathogenic role as part of the intestinal microbiome, which is often considered a “forgotten organ.”16 While this article renders the reader amply convinced of the indispensable role of dietary fiber, one is compelled to ponder about the standalone effect of soluble fiber. Multiple dietary combinations were employed in this model. A design in which soluble fiber was the sole variable, within an otherwise identical diet across study arms, would have strengthened the conclusions. As presented, it is difficult to attribute the relative contributions of carbohydrate restriction versus soluble fiber deficiency to the observed effects, despite the inclusion of an animal experiment, in which soluble fiber supplementation restored PPAR-γ signaling. Furthermore, although some findings were corroborated using human datasets - specifically the association between low carbohydrate/low fiber diets and elevated fecal calprotectin levels -additional human data examining the prevalence of colibactin-producing E. coli based on sequencing of CRC patient samples would have been valuable. The desirability of this is enhanced by the authors’ observations, that another colibactin-producing E. coli strain has been commercially used as a probiotic for several years.

This study provides invaluable insight into colorectal cancer pathogenesis as a result of the interplay between diet, gut microbiome and genetic host factors, with inflammation playing a central role. By elucidating the mechanism by which dietary fiber restrains the oncogenic potential of colibactin producing E. coli, the authors have paved the way for further investigations into focused lifestyle and dietary modifications, applicable in a preventive and therapeutic setting in colorectal cancer.