1. Froelich, Warren

Article Content

Strains of a pathogenic bacterium residing in the gut's microbiome have been shown to play a key role in the initiation and progression of colon tumors in preclinical studies led by researchers at Johns Hopkins School of Medicine.

Colorectal Cancer. C... - Click to enlarge in new windowColorectal Cancer. Colorectal Cancer

The results, presented during a special virtual session on colorectal cancer by the American Association for Cancer Research (AACR), add to a growing body of evidence suggesting that isolated bacteria in the gut may trigger the growth of colon tumors.


In the current study, Clostridioides difficile (C. difficile), the leading cause of hospital-based or nosocomial infective diarrhea in the world, was identified as a potential novel "driver" that contributes to genetic changes underlying the initiation of tumor formation in the colons of laboratory mice. Severe cases of C. difficile are linked to severe colitis and mortality, with about 15,000 deaths per year attributed to this infection in the United States.


Julia Drewes, PhD, Assistant Professor of Medicine at Johns Hopkins, said the results were particularly surprising, since the handful of previous studies in this field were based on the hypothesis that colon cancer might increase the risk of these dangerous C. difficile infections, not the other way around.


"Given that there was zero evidence in the literature to suggest C. difficile was pro-tumorigenic, to be honest we initially assumed that these experiments wouldn't work, and we were quite skeptical," said Drewes, who presented the study's results during the AACR conference. "In retrospect, it seems almost obvious that a pathogen such as C. difficile could be pro-tumorigenic in animal models, given what we've seen with other enteric pathogens in the literature. So, I think we've gone from being skeptical to being surprised that no one else had stumbled across this before."


According to the American Cancer Society, colorectal cancer is the third-leading cause of cancer-related mortality among men and women in the United States, excluding skin cancer, and is expected to cause about 52,980 deaths during 2021. For this reason, researchers are hoping to uncover new clues for how the disease forms and develops, with the goal of creating more effective treatments.


For her part, Drewes cautioned that several bacteria have been shown to drive cancer in laboratory mice. But demonstrating pro-tumorigenesis in patients "is a whole other level," she said, "that requires longitudinal prospective studies."


Examining the Research

In her talk, Drewes outlined the many "twists and turns" over several years that led to this latest discovery. A variety of preclinical studies have identified species of bacteria capable of penetrating the mucous shield that lines the colon, including colibactin-producing E. coli, enterotoxigenic Bacteroides fragilis, and Fusobacterium nucleatum, all acting as potential drivers of colon cancer. Untill now, microbial studies in colorectal cancer have largely focused on these three species, referred to at times as "the big three."


However, other studies show that not all colorectal cancer patients harbor these organisms, suggesting that either some colorectal cancer cases are not related to these microbes or additional pro-carcinogenic bacteria have yet to be identified. Previous work in the laboratory of Cynthia Sears, MD, Professor of Medicine at Johns Hopkins, also recently established that microbial organization into mucus-invasive biofilms in the colon was strongly associated with colorectal cancer.


In collaboration with researchers from the University of Florida, Drewes and colleagues performed an experiment, published in the Journal of Clinical Investigation (2019;, where biofilm-positive tumors from five patients with colorectal cancer were removed, ground up, and mixed together to form a slurry which were fed to germ-free mice. Within 10-15 weeks, the mixture of five biofilm-positive tumors induced "robust tumorigenesis" in the colons of these mice. In contrast, biofilm-negative slurries from healthy control patients did not induce tumors.


"At the time, we perhaps naively presumed that all biofilms from colorectal cancer samples were tumorigenic," Drewes said. "However, when we started to test individual patient slurries, we actually saw that this was not the case and there was quite a bit of variability in the tumorigenic potential of these different patient tumor slurries."


From six individual patient tumor slurries, the team identified two to be the most tumorigenic. But it wasn't clear why this was the case.


"There was no consistent pattern here between the overall microbial composition of the samples that were tumorigenic versus those that were not, so in order to further define which microbes were potentially responsible, we focused on just one patient, 3728T (one of the tumor slurries), a 69-year-old male with Stage I adenocarcinoma in the ascending colon," Drewes said.


This sample induced highly consistent colonic tumors and even reassembled into biofilms in the mice. "This led us to be really confident we had a consistent model of tumorigenesis and that the microbiome was likely playing a role," she noted. "So, the question was, do you need all the bacteria in this biofilm or are there particular alpha bugs that are driving this phenomenon?"


To tease out which bacteria were important, the researchers cultured out bacteria from the distal colons of germ-free mice inoculated with their pro-tumorigenic slurry.


From several hundred bacterial colonies screened, 30 unique bacterial species emerged, including C. difficile. To test if C. difficile could be a key tumorigenic factor in the 30-bacterial mixture, Drewes and colleagues inoculated germ-free mice with all 30 bacteria. When C. difficile was included in the mix, all mice developed robust colonic tumors. However, when C. difficile was not included, the mice did not develop tumors. Curiously, none of the so-called "big three" bacteria were present in the tumorigenic isolate mixture.


"That does not take away the potential importance of other putative pro-carcinogenic organisms," Drewes said. "It just highlights the diverse microbial paths that might lead to colorectal cancer."


She added that studies at Johns Hopkins and Vanderbilt University show that C. difficile, and a toxin harbored by the bacteria called TcdB, appear to be associated with numerous pro-tumorigenic mechanisms commonly found in microbial-induced cancer models, including induction of Wnt signaling, oxidative stress pathways, and tumor immune microenvironment dominated by IL-17, whose presence is often associated with a poor prognosis for colon cancer.


If chronic or recurrent C. difficile infections do prove to be associated with an increased risk of developing colorectal cancer, a variety of tools already available ranging from fecal microbiota transplantation or antibodies against TcdB could be utilized in at-risk populations. Drewes cautioned, however, there isn't enough data yet to suggest wider use of these therapies at this time. She added that research also has shown that the microbiome may impact chemotherapy of immunotherapy or even may predict treatment side effects. "I think this is going to be a really interesting topic in the field."


Warren Froelich is a contributing writer.