Antibiotics are associated with increased large intestinal proteolytic activity and gut barrier disruption, thereby raising the risk of chronic colitis in susceptible individuals, a recent study found.
Although the association between antibiotics and chronic colitis has been previously described, this is the first study to demonstrate the causative role of high proteolytic activity, reported lead author Hongsup Yoon, PhD, chair of nutrition and immunology at Technische Universität München in Freising-Weihenstephan, Germany, and colleagues. The team’s experiments support development of antiproteolytic strategies in susceptible humans.
“In the context of IBD, several clinical studies have already revealed that early and frequent antibiotic therapies, especially metronidazole or fluoroquinolone treatments, are associated with increased risk for Crohn’s disease,” the authors wrote in Cellular and Molecular Gastroenterology and Hepatology. “However, the causal role of antibiotic therapies in the disease development and the mechanisms underlying this [potentially] serious long-term adverse effect of antibiotics on the intestinal immune homeostasis remain unknown.”
Previous studies have shown that antibiotic therapy often causes high luminal proteolytic activity in the large intestine, likely because of the elimination of antiproteolytic bacteria that normally control pancreatic protease levels. Other studies have shown that exposing murine colonic mucosa to fecal supernatants with high proteolytic activity increases gut barrier permeability, which triggers chronic inflammation via translocation of luminal antigens.
“In view of these data,” the authors wrote, “we hypothesized that the antibiotic-increased proteolytic activity in the large intestine is a relevant risk factor for the development of colitis in susceptible organisms.”
The first component of the study used transwell experiments to evaluate the impact of high proteolytic activity on gut barrier integrity. High proteolytic activity was induced by several antibiotics, including fluoroquinolones with or without an imidazole (ciprofloxacin and levofloxacin plus or minus metronidazole), a beta-lactam (amoxicillin + clavulanate), cephalosporins with or without a macrolide (azithromycin and ceftriaxone plus or minus azithromycin), and a rifamycin (rifaximin).
“All tested antibiotic classes mediated a major proteolytic activity increase in some patients but not in others,” the authors wrote, “demonstrating individual-specific vulnerability of the intestinal microbiota toward antibiotic therapies, which is likely caused by the high interindividual variability of human microbial ecosystems.”
One-quarter of patients had a 400% or greater increase in large intestinal proteolytic activity following antibiotic therapy, and several had an increase greater than 900%. Analysis indicated that proteolytic activity was caused by pancreatic proteases such as chymotrypsin and trypsin.
Subsequent cell line testing showed that stool supernatants with high proteolytic activity damaged the epithelial barrier, but samples with low proteolytic activity did not. Of note, the negative impact of high proteolytic activity on epithelial cells could be mitigated by incubating stool supernatants with a serine protease inhibitor.
In analogous experiments, mice were given a combination of vancomycin and metronidazole (V/M). In contrast with the various proteolytic activity levels observed in humans, all mice had high proteolytic activity levels following treatment, suggesting that V/M eliminated almost all antiproteolytic bacteria.
The loss of antiproteolytic bacteria was clarified by cecal microbiota transplantation tests. Transplants from untreated mice were capable of normalizing proteolytic activity levels in germ-free mice (which have high proteolytic activity levels), but transplants from V/M-treated mice were ineffective, suggesting a near-total loss of antiproteolytic bacteria. The identity of these antiproteolytic bacteria remains a mystery.
“Although our data are in line with published literature suggesting specific strains of the order Bacteroidales to play a role in the physiological inactivation of pancreatic proteases,” the authors wrote, “the identity of relevant antiproteolytic species/strains remains to be elucidated.”
The next part of the study involved wild-type and interleukin (IL)-10–/– mice, the latter of which serves as a model of human colitis. Both types of mice were given V/M with or without an oral serine protease inhibitor, a potential therapy intended to limit proteolytic activity and associated intestinal barrier damage.
Although both wild-type and IL-10–/– mice had increased intestinal permeability after V/M treatment, only IL-10–/– mice showed lasting inflammation. Of note, coadministration of an oral serine protease inhibitor with V/M protected against colitis in IL-10–/– mice.
The protective benefit of an oral serine protease inhibitor in IL-10–/– mice prompts the development of antiproteolytic strategies in humans. These would target “large intestinal proteolytic activity [e.g., oral administration of encapsulated serine protease inhibitors, commensal antiproteolytic bacteria, or genetically modified bacteria expressing protease inhibitors] to protect the large intestinal mucosa from adverse effects of antibiotic-induced or diarrhea-induced high proteolytic activity,” the authors wrote.
The study was funded by the Deutscher Akademischer Austauschdienst. No conflicts of interest were reported.
SOURCE: Yoon H-S et al. Cell Mol Gastroenterol Hepatol. 2018 May 29. doi: 10.1016/j.jcmgh.2018.05.008.