Characterization of three gut microbes and their toxins in cancerous and healthy tissues of Iranian patients with colorectal cancer Authors: Abbas Yadegar, Masoumeh Azimirad, Somayeh Jahani-Sherafat, Hadi Ahmadi Amoli, Masoumeh Farahani, Mohammad Reza Zali, Saeed Abdi, Afsaneh Sharifian Published: 2025 ABSTRACT AbstractAim:This study aimed to investigate the presence of three gut bacteria and their toxin-encoding genes in cancerous tissues and their adjacent healthy tissues in Iranian patients with colorectal cancer (CRC).Background:CRC is a leading gastrointestinal malignancy influenced by genetic and environmental factors. Gut microbiota dysbiosis, including overgrowth of pathogenic bacteria and loss of beneficial species, contributes to carcinogenesis through toxins, inflammation, and DNA damage.Methods:Totally, 64 biopsy samples including 22 cancerous and 22 adjacent healthy tissues from CRC patients and 20 biopsy samples from healthy controls were obtained. The presence of difficile (C. difficile), enterotoxigenic Bacteroides fragilis (ETBF), Helicobacter pylori (H. pylori) and their toxin-encoding genes were detected by using culture and polymerase chain reaction, respectively.Results:Totally, 19 (43.2%) toxigenic C. difficile isolates were detected in cancerous specimens and healthy adjusted tissues. Most of the patients (36.4%) with high-grade tumors, 9.1% with moderate and 13.6% with low-grade tumor were positive for C. difficile culture. The frequency of B. fragilis in cancerous tissues, healthy adjacent and healthy control tissues were 72.7%, 59.1% and 90%, respectively. The prevalence of bft gene within B. fragilis isolates from CRC patients was significantly higher than the other groups (P=0.009). H. pylori was detected only in one cancerous tissue with moderate grade of tumor, and this isolate was negative for cagA, but carried vacA s1m2 alleles.Conclusion:Our study demonstrated a significant increase in C. difficile and ETBF colonization in CRC tissues compared to control samples. Further studies with greater cohorts of CRC patients are required to better evaluate the role of gut bacteria and their toxins in CRC pathogenesis and development.Key Words: Colorectal cancer, Cancerous tissue, C. difficile, B. fragilis, H. pylori, Toxin genes, ETBF Introduction Colorectal cancer (CRC) is the third most common cancer of the gastrointestinal (GI) tract with the fourth highest mortality rate worldwide (1, 2). Several genetic and environmental factors have been described for the etiology of CRC. Based on recent studies, alteration in the composition of gut microbiota is considered as a triggering factor in induction of CRC (3). Although much is known about the genetic pathways in development of CRC, the role of intestinal bacteria is not clearly understood. However, the concept of "driver" or "keystone" bacteria in the pathogenesis of CRC has been suggested in recent years (4, 5). Previous studies have shown that gut microbiota dysbiosis is associated with the development of precancerous status (3). In addition, the overgrowth of some bacteria including Bacteroides fragilis, enterotoxigenic B. fragilis (ETBF), Fusobacterium nucleatum, Enterococcaceae or Campylobacter, Peptostreptococus, Enterococus faecalis, Escherichia coli, Shigella, Streptococcus gallolyticus, Streptococcus bovis, Helicobacter pylori, and decrease in the abundance of Faecalibacterium, Blautia, Clostridium, Bifidobacterium and Roseburia were reported to be engaged in the progression of CRC (6-8). Among these, H. pylori has been considered as a strong risk factor for development of gastric adenocarcinoma and MALT lymphoma by the International Agency for Research on Cancer (IARC), a subordinate organization of the World Health Organization (WHO) (9). H. pylori can induce various oncogenesis-associated changes, from inflammation to intestinal metaplasia via stimulating intracellular signaling pathways by its cytotoxin-associated gene A (CagA) (10). Like CagA, previous studied have established that cytotoxin VacA can deregulates multiple cellular pathways that may lead to inflammation (11-13). Even though the role of H. pylori in the induction of gastric cancer is well known, few studies have reported the possible role of this bacterium in CRC (14-16). Moreover, C. difficile is a non-invasive gut microbe responsible for difficile infection (CDI) (17). Indeed, two factors affect the clinical manifestations of CDI, including bacterial pathogenicity and host immunity responses. The pathogenicity of C. difficile is directly related to the toxigenic nature of the infecting strains. However, the role of C. difficile in colon carcinogenesis has not yet been well investigated, and few studies have been performed to explore the role of this bacterium in the development of CRC (16, 18). Some studies have shown that up to 17% of the CRC patients carried C. difficile (19). Moreover, these studies have concluded that antibiotics therapy was a major risk factor for the development of CDI among CRC patients (20). Additionally, some studies have demonstrated that Escherichia coli and Bacteroides spp. were the main bacteria found in CRC patients (21, 22). It has been shown that the toxins produced by intestinal bacterial pathogens can increase the risk of cancer via induction of DNA damages (21). Also, the production of reactive oxygen species (ROS) in response to intestinal inflammation can mediate DNA damage (6, 23, 24). Furthermore, bacterial toxins can affect the carcinogenesis pathways within the colon, as well as in other tissues via systemic effects (25-27). Herein, we aimed to investigate the presence of three gut bacteria and their toxin-encoding genes in cancerous tissues and their adjacent healthy tissues in Iranian patients with colorectal cancer (CRC). Methods Patients and sample collection In this cross-sectional study, totally 64 tissue samples were collected from 22 CRC patients and 20 healthy subjects from February 2017 to October 2018. Twenty-two cancerous and 22 healthy adjacent tissues were collected from patients diagnosed with CRC who underwent colorectal surgery. Additionally, 20 healthy tissue samples were collected as healthy controls. Informed consent was obtained from all patients under a standard protocol and the study was approved by the Ethical Review Committee of the Research Institute for Gastroenterology and Liver Diseases at Shahid Beheshti University of Medical Sciences. A standardized questionnaire including demographic and clinical data was fulfilled for all participants. Healthy controls were outpatients who had no history of CRC, polyps, irritable bowel syndrome (IBS) and inflammatory bowel disease (IBD). Patients ≥18 years with stage I–IV CRC, a life expectancy ≥2 years and ability to sign consent were included in this study. Patients with a history of radiation, and/or chemotherapy and treated with antibiotics in 3 months prior to surgery, life expectancy less than 2 years due to concomitant disease, refusal to perform informed consent and other malignancy were excluded in this study. Histopathological evaluations were used to confirm CRC, tumor stage and tumor grade. C. difficile culture and identification Colonic biopsies were transported to the laboratory in thioglycolate broth Merck, Darmstadt, Germany), and immediately homogenized with a mechanical homogenizer and cultured on proper media. For identification of C. difficile, 100 microliters of homogenized biopsy specimens were cultured on C. difficile medium (CCFA: cycloserine-cefoxitin fructose agar) (Mast, London, United Kingdom) supplemented with D-cycloserine (250 mg/ liter), cefoxitin (8 mg/liter), lysozyme (5 mg/liter) (Mast, London, United Kingdom), and 7% horse blood under anaerobic conditions (CO2 5%, H2 10%, N2 85%) at 37oC for 48 hrs (28). B. fragilis culture and identification For isolation of B. fragilis 100 microliters of homogenized biopsy specimens were cultured on B. fragilis-bile-esculin agar (Difco) with 5% sheep blood and incubated at 37°C for 48-72 hrs under aforementioned anaerobic conditions (29). The anaerobic conditions were generated by an anaerobic system using Anoxomat® Gas Exchange System (Mart MicrobiologyBV, Holland). Suspected isolates were identified by using characteristic colony morphology and conventional methods including Gram staining, biochemical tests (indole, catalase, esculin hydrolysis). The isolates were further identified by PCR using species-specific primers. All isolates were stored at -70oC in thioglycolate broth with 15% glycerol for further analysis (29). H. pylori culture and identification For investigation of H. pylori, 100 microliters of homogenized biopsy specimens were cultured on Brucella agar plates supplemented with 7% horse blood, Campylobacter selective supplement (vancomycin 2.0 mg, polymyxin 0.05 mg, trimethoprim 1.0 mg) and amphotericin B (2.5 mg/L). The cultured plates were incubated under microaerophilic conditions at 37°C for 3-5 days. Colony morphology, Gram staining, oxidase, catalase, and urease activity tests were performed to identify the suspected colonies, which were further confirmed by PCR (30). Genomic DNA extraction Total DNA was extracted from pure colonies grown on specific media using DNeasy Blood and Tissue Kit (Qiagen, Germany) based on the manufacturer's instructions. The quality and purity of the extracted DNA samples were measured using NanoDrop (Thermo Scientific, Asheville, NC, USA). PCR assays For molecular identification of C. difficile isolates, PCR was performed using species-specific genes, including cdd and 16S rRNA. The presence of C. difficile toxins was examined by PCR using tcdA and tcdB specific primers as described previously (31). For identification of B. fragilis and detection of the enterotoxins, PCR amplification was performed as described previously (29). H. pylori colonies were confirmed by amplification of 16S rRNA, vacA, and cagA genes (32). The oligonucleotide sequences, amplicon size and PCR conditions for each amplification assay are presented in Table 1. Table 1Oligonucleotide sequences used in this studyTarget genePrimerOligonucleotide sequence (5′-3′)Product size (bp)PCR conditionsReference16S rRNAPS13PS14GGAGGCAGCAGTGGGGAATATGACGGGCGGTGTGTACAAG106294°C 10 min,35 cycles(94°C 50 sec;54°C 40 sec;72°C 50 sec),72°C 5 min(33) tcdA tcdA-F3345tcdA-R3969GCATGATAAGGCAACTTCAGTGGTAAGTTCCTCCTGCTCCATCAAATG629 tcdB tcdB-F5670tcdB-R6079AtcdB-R6079BCCAAARTGGAGTGTTACAAACAGGTGGCATTTCTCCATTCTCAGCAAAGTAGCATTTCTCCGTTTTCAGCAAAGTA410 cdd3 Tim 6Struppi 6TCCAATATAATAAATTAGCATTCCGGCTATTACACGTAATCCAGATA62495°C 5 min,35 cycles(95°C 1 min;57.1°C 1 min;72°C 1 min),72°C 5 min B. fragilis BaFRABaFRATGATTCCGCATGGTTTCATTCGACCCATAGAGCCTTCATC24394°C 5 min,40 cycles(94°C 20 sec;55°C 1 min, 72°C 5 min), 72°C 5 min(31) bft BFTFBFTRCGCGGCATTATTAGCTGCATGTTCTAATGGATACATCAGCTGGGTTGTAGACATCCCA1 KB94°C 5 min,40 cycles(94°C 1min;62 °C 2 min,72°C 1min), 72°C 5 min vacA s1/s2vacA m1/m2VA1FVA1RVAG-FVAG-FATGGAAATACAACAAACACACCTGCTTGAATGCGCCAAACCAATCTGTCCAATCAAGCGAGGCGTCAAAATAATTCCAAGG259/286570/64594°C 5 min,33 cycles(94°C 1 min;53°C 33 sec; 72°C 1min), 72°C 5 min(34) cagA CagAF CagARAATACACCAACGCCTCCAAGTTGTTGCCGCTTTTGCTCTC40094°C 5 min,33 cycles(94°C 1min;59°C1min;72°C 1min),72°C 5 min Oligonucleotide sequences used in this study Statistical analysis Statistical analysis was performed using SPSS software version 21 (SPSS Inc., USA). Statistical differences between the groups were analyzed by Fisher’s exact test and the results were considered to be significant at a P value of ≤0.05. Results Patient characteristics In this study, CRC patients consisted of 10 females (45.4%) and 12 males (54.6%), within the age range of 19 to 79 years (mean 58 ± 12.2 years). Healthy control group included 11 males (55%) and 9 females (45%) within the age range of 21 to 65 years (mean 38.2 ± 22.2 years). Among the CRC patients, high-grade tumor was diagnosed in 8 (36.4%) cases, moderate-grade in 4 (18.2%), and low-grade in 10 (45.4%). Among these patients, 10 (45.5%) were diagnosed with CRC stage I, 7 (31.8%) with stage II, and 5 (22. 7%) with stage III. Baseline demographics and clinical characteristics of the cohorts are summarized in Table 2. Table 2Demographic and clinical characteristics of CRC patients and healthy controls.VariablesCRC patients(n=22), n (%)Healthy controls(n=20), n (%)GenderMaleFemale12 (54.6%)10 (45.4%)11 (55%)9 (45%)AgeMean ± SDRange58 ± 12.219-7938.2 ± 22.221-69TNM stageIIIIII10 (45.5%)7 (31.8%)5 (22.7%)---Tumor gradeHighModerateLow8 (36.36%)4 (18.18%)10 (45.45%)---CRC, colorectal cancer; TNM, tumor node metastasis. Demographic and clinical characteristics of CRC patients and healthy controls. CRC, colorectal cancer; TNM, tumor node metastasis. Prevalence of bacterial isolates and toxin genes In total, C. difficile was isolated from 19/44 (43.2%) biopsy samples of CRC patients, including 13/22 (59.1%) from cancerous specimens and 6/22 (27.3%) from healthy adjusted tissues. In contrary, C. difficile was isolated only from 2/20 (10%) healthy controls, which showed a significant association when compared to cancerous specimens (P=0.04). Most of the patients (8/8) with high-grade tumor and 2/8 with moderate and 3/10 with low-grade tumor were positive for C. difficile culture. All C. difficile isolates within the CRC patients were toxigenic and harbored one or both tcdA and tcdB toxin-encoding genes. B. fragilis was isolated from 16/22 (72.7%) cancerous tissue specimens and 13/22 (59.1%) healthy adjusted tissues of CRC patients. In healthy controls, 18 samples (90%) were positive for B. fragilis culture. The isolation rate was not statistically different between cancerous and healthy adjusted tissue specimens from CRC patients (P>0.05), however, our findings showed that B. fragilis was isolated significantly and more frequently in healthy controls than CRC patients (P=0.04). The bft gene was detected in 6/22 (27.3%) cancerous tissues of CRC patients, 2/22 (15.4%) of healthy adjusted tissue of CRC patients, and 2/20 (10%) samples of the healthy controls. Accordingly, the prevalence of bft gene among B. fragilis isolates from CRC tissues was significantly higher than healthy adjusted tissues in CRC patients and biopsies from healthy controls (P=0.009). Moreover, around 50% of ETBF isolates from cancerous samples belonged to high-grade CRC samples. There was a significant relationship (P=0.044) between the presence of bft toxin-encoding gene and the grade of CRC. H. pylori was detected only in one cancerous tissue (4.5%), and this isolate was negative for cagA, but carried vacA s1m2 alleles. The frequency of selected bacteria and their toxin-encoding genes in CRC and healthy control biopsy samples are presented in Table 3. Table 3Presence of the selected bacteria and their toxin-encoding genes in CRC and healthy control biopsy samples.IsolatesCRC (n=22)Normal tissue (n=22)Total (n=44)Healthy control(n=20)HighModerateLowTotal (n/%) C. difficile tcdA + B + tcdA + B - tcdA - B - 8 (61.5)7 (58.3)1 (100)02 (15.4)2 (16.6)003 (23.1)3 (25)00 (100)12 (100)1 (100)06 (27.3)4 (18.2)2 (9.1)019 (43.2)16 (36.4)3 (6.8)02/20 (10)-------2 (10) B. fragilis bft 5 (31.25)3 (50)3 (18.75)1 (16.6)8 (50)2 (33.4)16 (100)6 (100)13 (59.1)2 (9.1)29 (65.9)8 (18.2)18 (90)2 (10) H. pylori cagA vacA 0001 (100)01 (100)0001 (100)000001 (2.27)01000 Presence of the selected bacteria and their toxin-encoding genes in CRC and healthy control biopsy samples. Discussion There have been previous reports of increased intestinal bacterial overgrowth in CRC patients compared to healthy controls (33, 34). In the present study, we investigated the existence of three enteric pathogens including C. difficile, B. fragilis and H. pylori, and their toxin-encoding genes in biopsy samples obtained from Iranian patients suffering from CRC. Importantly, in our study, the cancerous and healthy adjusted tissues were obtained from the same patients, which minimized the bias in the adjustment of samples, and therefore, some factors such as nutrition, age, sex, and lifestyle could not interfere with the results of the current study. In this study, a significant increase in the colonization of C. difficile in the cancerous tissues and high-grade tumors were observed compared to adjacent healthy tissues. There are insufficient studies investigating the role of C. difficile among CRC patients. In a study conducted by Zheng et al. in 2017, C. difficile was isolated from 16% of stool samples of patients with CRC and 97% of the isolates were toxigenic (19). The differences observed in these findings might be related to the study population characteristics. Moreover, the isolation rate of C. difficile from intestinal mucosal tissue has been reported to be higher than feces due to the favorable conditions for bacterial growth. Additionally, there was a significant difference in the colonization of C. difficile between the cancerous tissues and their adjacent tissues. The possible explanation is that bleeding from cancerous lesions may alter the microbiota population and lead to a change in C. difficile colonization. Furthermore, blood and its components may also play a role in the colonization of C. difficile (35, 36). Our findings regarding the high proportion of C. difficile in the high-grade tumor samples may also suggest the favorable conditions for this bacterium in such tumor microenvironments. Further studies are however needed to clarify the relationship between bleeding in the cancerous tissues and C. difficile colonization. It is likely that the increase in C. difficile colonization in the cancerous tissues results from the nature of opportunistic behavior of the bacterium, so-called "bacterial passenger". This process occurs after the onset of the neoplastic progression, which is due to the changes in the host physiological status such as inflammation and increased free radical’s oxygen toxicity in the colon environment that facilitate the overgrowth of opportunistic or passengers’ bacteria. It has been described that the presence and overgrowth of C. difficile in CRC patients, especially during chemotherapy, may cause CDI and severe diarrhea that can lead to cessation of chemotherapy treatment (35). Therefore, C. difficile colonization should be investigated for all patients before CRC operation as part of the risk factors for the effective cancer treatment. As already mentioned, it has been proposed that ETBF strains may simplify the creation of dysbiotic microbiota populations in CRC (6, 37). ETBF strains, which are associated with inflammatory diarrheal disease, (38) produce an enterotoxin known as BFT. BFT is also considered as a pro-oncogenic factor, which has the ability to influence the localized mucosal immune responses (34). It has been suggested that a long-term colonization of ETBF in colon elevates the chance of CRC, because the high-level expression of BFT can disrupt E-cadherin junctions and leads to an increase in the B-catenin signaling and IL-8 secretion in the epithelial cells of colon (37, 39). Recent microbiome analysis has shown that dysbiosis can occur in non-tumor colorectal tissue in addition to tumor tissue in CRC patients (40). In this study, B. fragilis was isolated more frequently in the cancerous tissues than healthy adjusted tissues. However, the isolation rate of B. fragilis was not statistically different between cancerous tissue specimens and healthy adjusted tissues of CRC patients, which was in accordance to the results reported by Toprak et al. in Turkey (41). As B. fragilis accounts for two percent of the total Bacteroides population in the gut, further investigations with greater sample size are recommended to determine the precise role of this bacterium in CRC (41). Additionally, BFT is a major virulence factor involved in the pathogenesis of B. fragilis. In our study, bft gene was detected significantly higher in colon mucosal samples of CRC patients compared to controls. Our results are similar to other studies suggesting that mucosal bft expression may be a risk factor for the development of CRC. In accordance with our findings, in which 27% of B. fragilis in CRC tissues harbored the bft gene, Viljoen et al (38) reported that 26% of mucosal samples from CRC patients were bft-positive. In contrary to these results, there is a report indicating a high percentage of ETBF in CRC patients (41). According to our findings, H. pylori was isolated from only one sample of cancerous tissue, and this pathogen was not detected in the healthy tissues of the control group which is on par with previous studies (34). Our PCR assay revealed that this positive sample of H. pylori was negative for cagA toxin but positive for vacA s1m2 alleles. In a study conducted by Bulajic et al., H. pylori was detected in 1.2% of tumor tissues by molecular assays which is in agreement with the results of this study (42). In another study, the DNA of H. pylori was found in 27% of cancerous tissue and 22% of healthy tissue samples, but no significant difference was observed (43). Some studies also reported the presence of H. pylori in CRC patients by serologic and immunohistochemical methods, but the association of this pathogen with CRC cannot be verified due to high rate of false positives in these methods (44-46). Some other serological studies, however, did not find any significant association between H. pylori and CRC (44). Accordingly, based on the results of the present study and previous researches, probably no association can be made between H. pylori infection and colon cancer. However, the effect of H. pylori infection in the development of CRC should also be explained through the alterations in the acidity of the stomach and increasing the permeability of intestines, which can result in changes in the gut microbiota composition (47). Conclusion Our study showed a significant increase in C. difficile colonization in CRC tissues compared to control samples, especially in high-grade tumor samples. This may be due to favorable conditions in CRC lesions after the formation of CRC. Our results also showed significant differences in the presence of bft gene in CRC and healthy control samples. Based on our study no association was found between H. pylori infection and increased risk of CRC. In the current study, we only examined the presence or absence of toxin-encoding genes, and we well know that their positivity by PCR is not a definite cause for the production of bacterial toxins as the genes may exist but not expressed. Further studies with greater sample size that include large cohorts of CRC patients and healthy individuals are required to evaluate the exact role of gut microbes and their toxin production in CRC pathogenesis and development. Acknowledgments The authors would like to acknowledge the assistance of the nursing staff from participating hospitals in Tehran during this study period. Conflict of interests There is no conflict of interest for authors of this article. References