Fecal Carriage of Colibactin-Encoding Escherichia coli Associated With Colorectal Cancer Among a Student Populace

Abstract Fecal carriage of the colibactin (clb) gene in Escherichia coli is described as a source that could promote carcinogenesis, progressing to colorectal cancer. The present study investigated the demographic, dietary, and antibiotic consumption variables as correlates for fecal carriage of clb+/E coli among the student populace. In a randomized cross-sectional survey, E coli (N = 136) from the fecal samples of eligible students were characterized and evaluated for antibiotic resistance, β-lactamase (blm), biofilm, virulence factor production, and strain tryptophan reverse mutagenic activity. The encoded clb+/E coli were analyzed for correlates with principal component analysis. Of all the E coli strains, a low rate of 2 clb+/E coli (1.5%) and higher rates of biofilm (13.2%) and blm producers (11.8%) were recorded among the mutant strains as compared with the nonmutant types. All the clb+/E coli showed complete resistance to amoxicillin, Augmentin (amoxicillin and clavulanate), gentamicin, and trimethoprim/sulfamethoxazole. The fecal clb-encoded E coli (1.5%) were not associated with demographic status, fiber-based food (odds ratio [OR], 1.03; 95% CI, 56.74–138.7; P = .213), alcohol (OR, 1.27; 95% CI, 61.74–147.1; P = .221), antibiotic consumptions (OR, 1.11; 95% CI, 61.29–145.3; P = .222), and handwashing (OR, 1.17; 95% CI, 60.19–145.5; P = .216). The hierarchical cluster of blm+/E coli revealed high-level resistance with a multiantibiotic resistance index ≥0.2 (P < .05). Only 12% of all strains were tryptophan mutant/blm+, and 1.5% of clb+/ECblm+ were observed in fecal samples with a 452–base pair size. Trimethoprim/sulfamethoxazole and biofilm production positively regressed with clb expression (P > .05). Principal component analysis score plot indicated an association of clb+/ECblm+ with dietary pattern, alcohol, blm, and hemolysin production. The combined activity of blm and biofilm production in the gut microbiota could promote clb+/E coli colonization, facilitating genotoxin production and possible colorectal cancer induction.


INTRODUCTION
According to the World Health Organization, colorectal cancer (CRC) is identified as one of the leading causes of death [1] and is mostly diagnosed in men as compared with women [1,2].In Europe, the early onset of CRC incidence was reported among individuals aged 20 to 39 years, of which diet and lifestyle play a significant role [2,3].In the recent past, the CRC incidence rate is getting higher in developing countries with the burden of infectious diseases, and this further predisposes several people to the risk of CRC [4,5].Genetic inheritance cannot be ruled out as a significant risk for CRC, while several reports have indicated the following to be associated with the incidence of CRC among the African populace: alcohol, smoking, chemical condiments as food spice (organic or inorganic type), bacteria, oncogenic viruses, diet, hormonal imbalance, and environmental pollution [6][7][8].To date, the association of the dietary pattern and lifestyle of students with CRC is yet to be confirmed, but plant-based diets and few dairy products are reported to lower the risk of CRC while diets rich in meats, refined grains, and sugar tend to increase gut susceptibility [9].The association between CRC and alcohol remains unclear.
Antibiotic use was identified to possibly increase CRC risk by altering the profile of the gut microbiota [10].The persistent misuse of antibiotics for the treatment of gastrointestinal diseases is suggestive of the functional role of different antibiotic classes (quinolones and sulfonamides and/or trimethoprims) to increase the gut's chances for proximal colon cancer and possible specific colorectal carcinogenesis [11].Hygiene practices with regular handwashing after use of the toilet and before meals were significantly associated with the development of CRC in a survey conducted among the Pakistani population [12].In most communities in developing countries with open Fecal Carriage of Colibactin-Encoding E coli and Colorectal Cancer • OFID • 1 Open Forum Infectious Diseases M A J O R A R T I C L E defecation and poor hand hygiene after use of the toilet [13], there is possible transmission of fecal pathogenic Escherichia coli causing severe intestinal morbidity.
Globally, there is limited information on the association of β-lactamase resistance enzymes with CRC, but there is a need to investigate the contribution of β-lactamase production in E coli and the induction of CRC.Fecal carriage of β-lactamaseproducing E coli demonstrating multidrug resistance poses a serious challenge to therapeutic options for colon cancer management and could enhance the severity of CRC in many individuals [14].Among the gut microbiomes, E coli is commonly found as normal flora in patients with CRC and healthy individuals; however, few pathotypes were reported to induce specific cytopathic changes (megalocytosis) in the human gut via the production of toxins [15].Colibactin is an important genotoxin produced from a 54-kb pks island, which consists of a clbA-S gene cluster encoding nonribosomal peptides and polyketide synthases with other accessory enzymes [12].Colibactin is predominantly found in the pks genomic island, mostly reported in the E coli B2 phylogenetic group [15], and is associated with cytopathic changes that enhance the progression of CRC [12].Mutant E coli were identified to be associated with CRC, showing invasive colonization of the lumen epithelium and producing different toxins called cyclomodulins (eg, colibactin, cytolethal distending toxin, cycle inhibiting factor, and cytotoxic necrotizing factor), as previously reported in patients with CRC [15,16].
Among the students, disparities in dietary patterns, misuse of antibiotics, and hygiene status have not been studied as major factors that facilitate fecal carriage of colibactin-encoded E coli.The present study aims to investigate the demography and dietary pattern, biofilm, β-lactamase, and virulence factor production as correlates for fecal carriage of clb-encoded E coli as potential agents for CRC among the student populace.

Study Population
This is a randomized cross-sectional study of fecal carriage of clb-encoded E coli among the student population, including undergraduate and postgraduate students.The inclusion criteria are based on their consent to participate and their status as full-time undergraduate and postgraduate students.The study excluded pregnant students, any students on medication, and students who were acutely ill or had known diseases.From the estimation of >12 000 students, the sample size (N = 136) was determined by the Fischer formula (n = z2 pq/d 2 ) [17].A prevalence of the colibactin gene among the tertiary students could not be found, but the prevalence rate in a previous study was used to derive the sample size [18].

Ethical Permission
The study was approved by the Covenant Health Research Committee, Covenant University, Ota, Nigeria (protocol NHREC/CU-HREC/11/04/2023). The informed consent of recruited individuals to participate in the study was integrated into the questionnaire, which was structured to include select demography data, feeding patterns, antibiotic use, alcohol consumption, condiment use, and hand hygiene patterns.

Sampling and Fecal Collection
A random sampling for the fecal sample collection from the student populace was based on the exclusion criteria for the selection of respondents.Structured questionnaires were administered to obtain data on age, gender, feeding pattern and dietary practices, antibiotic use, alcohol consumption, and hygiene practices.A separate code number was given to each respondent's questionnaire, as well as to each respective fecal sample.

Isolation and Biotyping
Collected fecal samples (N = 136) were homogenized in sterile normal saline and plated on MacConkey lactose agar (Oxoid) for selective isolation of gram-negative bacteria and further differentiation for lactose fermentation.The plates were incubated at 37 °C for 24 hours.Based on the colonial and cellular morphology, suspected E coli colonies that showed lactose fermentation (pink coloration) and were slightly raised and glossy were subcultured onto eosin methylene blue agar.After incubation at 37 °C overnight as previously described [19], colonies showing phenotypic colonial appearance of differential metallic sheen were selected.Pure colonies were cultured on 7% sheep blood agar and incubated at 37 °C overnight to evaluate their hemolytic reaction.Each strain was biotyped by emulsifying a loopful of a pure colony in sterile phosphate-buffered saline and added to an API 20E panel of biochemical tests (BioMérieux) for biochemical characterization.

Antibiotic Resistance and β-Lactamase Phenotyping
The susceptibility pattern of each characterized strain to selected antibiotics was based on the report of commonly listed antibiotics from the questionnaires obtained from the respondents.In line with the recommendation and guideline of the Clinical and Laboratory Standards Institute [20], the Kirby-Bauer disc diffusion method [21] was used to evaluate the susceptibility pattern of E coli to different classes of antibiotics [22].Briefly, overnight 0.5 MacFarland turbid broth suspensions were spread smoothly onto Mueller-Hinton agar with a sterile swab stick.The following antibiotic discs were gently placed on the plate and incubated at 37 °C for 24 hours: cefprozil (10 µg), ofloxacin (30 µg), Augmentin (30 µg), nitrofurantoin (30 µg), ciprofloxacin (30 µg), ceftazidime (30 µg), gentamicin (10 µg), and ampicillin (30 µg).E coli strain ATCC 25922, serving as a control strain, was used as a reference.The antibiogram results were used to classify the isolates as resistant, intermediate resistant, and susceptible according to guidelines of the Clinical and Laboratory Standards Institute [20].All isolates showing resistance to ≥3 classes of antibiotics were classified as multidrugresistant strains.The multiantibiotic resistance index of each strain to antibiotics was also determined [23,24].The select multidrug-resistant strains were tested for β-lactamase production for definitive assessment of enzyme inactivation as a resistance mechanism with the starch iodide acidimetric method as previously described [25,26] and further identified by the modified method of Stokes and Ridgway [27].

Biofilm and Virulence Factor Detection
The ability of the multidrug-resistant strains to produce biofilm was detected by microtiter plate bioassay [28], and the level of biofilm produced was quantitatively measured by an enzyme-linked immunosorbent assay spectrophotometric reader as previously described [29].Briefly, from overnight broth culture, 1:40 diluted broth in tryptic soy broth containing 0.25% glucose was prepared, and 200 µL from the diluted broth was inoculated in a sterile microtiter well and then incubated at 37 °C for 24 hours for biofilm production.After incubation, the medium was removed and washed twice with 0.2 mL of phosphate-buffered saline (pH 7.4) and stained with 0.1% crystal violet solution for 30 minutes.The color intensity produced after the addition of 95% ethanol was measured at 560 nm by the enzyme-linked immunosorbent assay plate reader.The absorbance of the test was compared with the negative control to determine the level of biofilm produced.Production of hemolysin was demonstrated on 5% defibrinated sheep blood agar overlaid with the Nutrient Agar base (Oxoid) after streaking a single colony and incubating the plates at 37 °C for 72 hours [30].Detection of lipase production was performed on a tributyrin agar plate as described [30].A single colony was cultured on the tributyrin agar plate and incubated at 37 °C for 24 hours.The observed zone of hydrolysis around the colony indicates lipase production [31].The phenotypic assay for protease production was carried out according to Suganthi et al [32].Briefly, 100 µL of 0.5 MacFarland turbid broth was gently placed on skim milk agar supplemented with 1% casein and allowed to be adsorbed, followed by incubation at 37 °C for 24 hours.A clear zone around the inoculum spot indicates positive casein hydrolysis.

Strain Mutagenicity and clb Genotyping
An Ames test was used to describe the E coli tryptophan (trp) reverse mutagenic activity via the blocking of trp biosynthesis prior to the production of anthranilic acid as described by Arun Nagendran [33].Overnight broth culture of E coli was mixed with 0.05 mg of trp in a sterile test tube that served as a test (trp+ E coli) and a control tube containing E coli mixed with normal saline (trp-E coli).Both tubes were incubated at 37 °C for 20 minutes.Five milliliters of the trp+ E coli and trp-E coli suspension were spread on separate agar-agar plates and incubated at 37 °C for 48 hours.After 48 hours, the number of colonies in each plate was counted.The mutagenicity of trp is proportional to the number of colonies observed.If there is a large number of colonies on the test plate in comparison with the control, then the isolates were suspected mutant strains.The carriage of clbA was determined by genotyping as described by McCarthy et al [34].clbA is an important gene cluster in colibactin synthesis and noted as a phosphopantetheinyl transferase-encoded gene,  (3) biofilm and virulence factor production.Eigenvalues >1.00 were accounted for the number of selected principal components according to the Kaiser criterion [35].Components 1 and 2 were extracted for the correlates to provide the possible and significant association of clb carriage with other correlates shown by spatial visualization of the data distribution.

Cluster Analysis of β-Lactamase-Producing E coli Strains
In Figure 2, all the identified β-lactamase-positive strains were grouped into 3 clusters.High resistance patterns to all the antibiotics, except chloramphenicol and ciprofloxacin, and high production of protease and lipase were observed among strains in cluster 1. Cluster 2 revealed a low susceptibility to >8 antibiotics with very low virulence production.Among the strains in cluster 3, low virulence factor production was recorded with intermediate susceptibility to fluoroquinolones (pefloxacin, ofloxacin, sparfloxacin, and ciprofloxacin) and chloramphenicol.

Colibactin Detection Among β-Lactamase-Resistant E coli
A significant rate of 13.2% strains with a multiantibiotic resistance index ≥0.2 was observed among β-lactamase-producing strains (n = 23) vs non-β-lactamase producers, suggesting a high level of resistance to select commonly used antibiotics (Figure 3A).A considerable rate of strains (12%) was trp mutant β-lactamase+ while others were nonmutant strains (Figure 3B).Only 1.5% of clb+/EC bl+ (n = 2) was observed in fecal samples obtained from a participant, which was significantly different from the 12.4% of clb-/EC bl+ strains (P = .027),while no clb+/EC bl-was recorded (Figure 3C).In Figure 2D, the agarose gel showed the 452-base pair polymerase chain reaction amplicon band of the clb gene from the β-lactamase-producing E coli.

Implication of Identified Variables for clb Carriage
None of the antibiotic variables obtained from the evaluation of phenotypic resistance pattern showed a significant correlation with clb expression (P > .05),but only the SXT resistance variables provided a prediction for clb carriage (P = .045;95% CI, .001-.046;Table 2).Virulence factors such as lipase, protease, and hemolysin could not show significant correlations or predictive carriage of clb among the isolates.A positive correlation of biofilm production with clb induction in E coli strains was observed (r = 0.464, P = .026).
To explore the implication of dietary patterns, handwashing practice, antibiotic resistance patterns, and virulence factor production for possible induction of the clb gene among the E coli strains, principal component analysis was utilized.The clb-encoded strains were grouped with the variables for the consumption of fast food and alcohol (Figure 4A) and the resistance pattern of SXT, gentamicin, streptomycin, amoxicillin, Augmentin, pefloxacin, and ofloxacin (Figure 4B).To investigate the contribution of the produced virulence factors on the expression of clb, the score plot further showed the collection of variables of β-lactamase and biofilm production with clb-encoded E coli (Figure 4C, Table 2).

DISCUSSION
The present study investigated the correlates for fecal carriage of colibactin-encoded E coli among the student populace, which could trigger the induction of CRC.Considering the confounding factors, the gender, marital status, and age of the respondents were not risk factors for the carriage of the clb gene.There was no observed association between fecal clb/E coli and consumption of fiber-based food, fast food, or chemical condiment-fortified food or alcohol intake.A fiberenriched diet among the African populace was reported to reduce the risk of CRC and adenoma [8] as a result of the fiber adsorbing the potential fecal carcinogens, altering bile acid metabolism, and thereby reducing the colonic transit time and pH and increasing production of short-chain fatty acids [36].Several studies could not confirm the role of alcohol in clb in fecal E coli, but aging of the gut and persistent consumption of alcoholic beverages could predispose to CRC induction [3,37].Antibiotics alter the gut microbiota population and aid colonization of resistance types, which could enhance the risk for the clb gene.The burden of antibiotic resistance caused by a high consumption rate as prophylactics and misuse for intestinal infections supports the evidence of oral antibiotics changing the gut microbiota composition and altering essential host immune responses, possibly leading to development and exchange of resistance genes among the gut microbiota [37,38].In contrast, long-term antibiotic use in early to middle adulthood could be associated with an increased risk of colorectal adenoma [39].The burden of intestinal infection would continue to be a leading cause of gut morbidity, which is disproportionately higher among the student populace due to multiple etiologic agents.Improved hand hygiene practice is considered a cost-effective intervention needed to reduce the burden of intestinal diseases, and it would reduce the possible transfer of fecal clb/E coli [40].
The continuous dissemination of fecal β-lactamase-producing E coli is a public health challenge, driving increasing resistance to commonly used penicillin and cephalosporin antibiotic classes.Phenotypic detection of β-lactamase in clustered fecal E coli strains showing extended resistance to chloramphenicol and ciprofloxacin suggests indiscriminate use of these antibiotics, which was evident from the high occasional use of antibiotics among the student populace (Table 1).The produced β-lactamase enzymes would further enhance highlevel efflux pump activity, alter the drug target (modification of penicillin-binding proteins), and decrease membrane permeability [41,42].A separate clustering of β-lactamase E coli-producing lipase and protease with low susceptibility to fluoroquinolones and chloramphenicol would describe an imminent pool of resistance enteric strains.This suggests impending intestinal infection severity through stages of tissue  damage that involve invasion and necrotization of intestinal epithelia mucosa, leading to gut inflammation and colitis [43].
In addition, the recorded multiantibiotic resistance index ≥0.2calls for the extension of antibiotic stewardship in tertiary institutions to forestall the spread of antibiotic resistance originating from high-risk fecal sources that could contaminate the environment, particularly the water bodies.To influence or initiate carcinogenesis, detected trp mutant E coli could trigger the production of intestinal inflammatory signaling molecules (including interferon γ and interleukin 4), capable of inducing the expression of IDO1 and altering trp metabolism [33,44], leading to production of high-level indoles from trp metabolism.The produced indole from trp metabolism via the indolic pathway is a key step in inflammatory conditions in colon carcinogenesis, as reported in the fecal samples of individuals with CRC [44].
Recording a low proportion of β-lactamase-producing E coli strain encoded with a 452-base pair clb gene in the fecal sample of a participant provides indication for possible onset of early CRC.The carriage of clb in the fecal sample of apparently healthy individuals possibly initiates colibactin, acting as a cyclomodulin and causing alteration of the eukaryotic cell cycle, leading to successive enlargement of the nucleus and eventual cell death [45].Not only could the cytopathic effect of clb mediate cell death by clb+/EC bl+ strains, but direct contact of the strain with the host cell could necessitate dysfunctional metabolic activity, giving rise to microbial gene products that affect intestinal homeostasis and gut barrier function and leading to microbial dysbiosis, which is commonly found in inflammatory bowel disease and CRC [12,46].To date, the epidemiology and prevalence of colibactin in fecal E coli are yet to be explored, including their association with intestinal infections in student populations from various geographic locations [12,47,48].To better understand the transmission and prevention of fecal clb/E coli, future investigation must include the genetic role of associated virulence factors (lipase and protease), β-lactamase enzymes, and risk factor assessment.
Since SXT is one of the commonly used prophylactics for intestinal and extraintestinal infections in Nigeria as compared with other countries [49], the present study revealed the association of SXT resistance with clb genes.The SXT-resistant mutant would become dominant and survive through the antibiotic selective pressure as mutant strains.The use of SXT to treat intestinal E coli infection has occasionally enhanced the development of high-level resistance to trimethoprim in enterobacteria populations due to changes in cell permeability, loss of drug-binding capacity, and alterations in dihydrofolate reductase, which are mediated by mutation of the chromosomal DHFR gene (dfrA and dfrB) [50].Persistent DHFR mutants in gut microbiota could further enhance mediation of the clb gene in the pks island [50].The significant association of biofilm formation with fecal clb expression could promote the protective mucosal barrier for effective interaction of the host epithelium, with clb+/E coli facilitating colonization and invasion of the mucosa epithelial cells [51].In the tumor microenvironment, as reported in biopsies from patients with CRC, there is high biofilm production by mucosal-invasive E coli, showing significant binding affinity to mucus-secreting cells [45].In vitro, biofilm formation is evidence of cell-to-cell contact, which is important for the genotoxic activity of colibactin, thereby inducing DNA damage that exacerbates mucosal disruption [45,[51][52][53].
The clustering of the clb-encoded strains with the consumption of fast food and alcohol as variables further affirms the role of poor dietary patterns and alcohol in induction of the clb gene.The implication of the β-lactamase production and biofilm aggregation in clb-encoded E coli is to enhance the strain intestinal morbidity through invasion and protection from the host immunity and effect of antibiotics, thereby complicating clb expression in the gut mucosa.Further investigations are needed to understand the impact of alcohol on the combined activity of β-lactamase enzymes and biofilm in fecal clb expression in healthy individuals.
The limitations of this present study include inadequate provision of demographic data and the limited number of eligible participants providing consent to participate in the study.The inability to collect fecal samples in replicate to aid adequate representation limits the rates of the detected colibactin gene.

CONCLUSION
The respondents' demography, dietary, antibiotic, and hand hygiene patterns are not conclusive variables as risk factors for fecal clb induction.The association of SXT with clb induction further suggests the antibiotic possible mediation of genotoxic colibactin in the E coli pks island.The use of SXT for clinical management or as prophylactics for intestinal infection needs to be controlled.The combined activity of β-lactamase enzyme and biofilm production in the gut microenvironment could promote induction of clb in gut E coli, thereby facilitating genotoxin production serving as a risk for CRC.Therefore, subsequent genomic studies would provide insight into the colibactin structural mode of induction associated with β-lactamase activity and pks regulation in healthy gut and the physiologic influence in CRC induction.
[34]ajorly required for the biosynthesis and maturation of colibactin and usually associated with pathogenicity and cancer[34].Extracted chromosomal DNA template from overnight pure broth culture was obtained with a commercial kit following the manufacturer's instruction.Polymerase chain reaction was carried out in a total volume of 20 µL containing 12 µL of 10 × Master Mix buffer, 1.0 µL of clb forward primer (clbA-F, CAG ATA CAC AGA TAC CAT TCA), 1.0 µL of clb reverse primer ). AM, amoxicillin; AU, Augmentin (amoxicillin and clavulanate); bfl, biofilm; blm, β-lactamase; CH, chloramphenicol; clb, colibactin; CN, gentamicin; CPX, ciprofloxacin; hly, hemolysin; lip, lipase; OFX, ofloxacin; PEF, pefloxacin; prt, protease; S, streptomycin; SP, sparfloxacin; SXT, trimethoprim/sulfamethoxazole.whichData AnalysisData from distributed questionnaires were analyzed and calculated for 95% CI and odd ratio for significant differences (P < .05) in estimates of the demographic status, dietary pattern, antibiotic use, alcohol consumption, and hygiene practices as independent variables.The hierarchical clustering of β-lactamase+/E coli was evaluated with a heat map that showed the strain antibiotic susceptibility and virulence factor production defining different clusters based on the Euclidean distances from the dendrogram construct generated from the UPGMA evaluation.The significance of β-lactamase-producing E coli strains and fecal carriage rates of clb+/EC bl+ , clb-/EC bl+ , clb+/EC bl-, and clb-/EC bl-was determined by a t test and analysis of variance, while the rates of clb+/EC bl+ and clb-/EC bl+ were compared with a chi-square test (P < .05).

Table 2 . Analysis of Variables With clb Detection
Fecal Carriage of Colibactin-Encoding E coli and Colorectal Cancer • OFID • 7