Irritable bowel syndrome (IBS) is a mysterious condition that affects an estimated 11% of the global population and causes chronic abdominal pain, diarrhea, and constipation. With IBS, the nerve endings that innervate the gut become hypersensitive, overreacting to everyday stimuli and often causing debilitating pain. Some people have a “flare” – a painful episode that can last days or weeks – in response to eating certain foods, but researchers have been at a loss as to how foods can trigger IBS pain.
Now, a study in mice and humans ties together years of work in the field to define an elegant mechanism in which an infection, such as food poisoning, can lay the foundation for IBS by inciting a localized immune reaction, with lasting consequences.
The research team, headed by Guy Boeckxstaens, KU Leuven, Belgium, show in a mouse model of IBS that a gastrointestinal infection paired with a food antigen – in this case, the egg protein ovalbumin – leads to prolonged painful sensitivity to subsequent exposure to ovalbumin, even long after the infection has been cleared. Although the food-related pain mechanism depends on multiple immune players, including mast cells, the phenomenon was strikingly localized to the colon (large intestine).
“I like this paper quite a lot,” said Karin Westlund High, a pain researcher at the University of New Mexico, Albuquerque, US, who was not involved in the work. “A lot of people have worked on this problem, and this study puts the pieces all in one place and nails down a mechanism. It’s a nice contribution to the field.”
In an accompanying News & Views, Stuart Brierly, a pain researcher at Flinders University, Adelaide, South Australia, wrote that the “study provides information on the mechanisms underlying abdominal pain, and gives added meaning to the saying, ‘you are what you eat.’”
The research and News & Views were published January 13, 2021, in Nature.
Putting the clues together
Piecing together findings from previous research, Boeckxstaens hypothesized that under certain conditions, such as a bacterial infection, immune cells would mount a response to a common and normally harmless food antigen, which could, upon re-exposure to the antigen, trigger immune cell activation, leading to pain hypersensitivity.
“One of the key findings that an immune response in the intestine could be involved dates to a milestone paper that showed there were more mast cells in the mucosa of patients with IBS than in controls,” Boeckxstaens told PRF in an interview, referring to the presence of these immune cells in the innermost layer of the intestine (Barbara et al., 2004). “That set the stage that something is wrong with the mucosal immune system in these patients.”
Later work from his own group showed that visceral nerves in rectal biopsies from IBS patients displayed heightened sensitivity to agonists of pain receptors (Wouters et al., 2016). “That was the first functional evidence that there was something different with the neuronal physiology in these patients.”
Breaking oral tolerance
Javier Aguilera-Lizarraga, co-first author along with Morgane Florens, both at KU Leuven, developed a mouse model of IBS based on the team’s hypothesis, which rests on the idea of breaking oral tolerance. The phenomenon of oral tolerance recognizes potentially harmful antigens, in this case foods, and gives them a pass so that they can be safely digested. But harmful bacteria and other pathogens that make it down the gastrointestinal tract don’t get this exception. When the immune system mounts a response to such invaders, it can also react to foods that happen to be present at the site of the crime.
To model IBS in mice, the researchers infected mice with the bacterium Citrobacter rodentium and also fed the animals drinking water containing ovalbumin (OVA). Five weeks later, after the infection had passed, exposure to OVA caused diarrhea in the mice, and the authors detected IgE antibodies specifically in the colon.
“One of the most surprising findings was that we found IgE antibodies against OVA in the large intestine, the site of the infection,” Boeckxstaens said. But they did not find IgE in the small intestine or circulating in the blood. “That’s an important finding; it’s a very localized phenomenon. This is what discriminates it from systemic food allergy.”
Following the infection, mice displayed increased pain responses to colorectal distention for up to two weeks. After the infection had cleared and pain responses had normalized, later exposure to OVA elicited long-lasting pain responses only in the mice that received the coincident OVA and infection, and not in uninfected mice or mice that did not receive OVA. Exposure to bovine serum albumin (BSA) did not trigger a pain response in the infected mice, indicating that the prior exposure to OVA was critical. And, similar to patients with IBS, the mucosal lining was more permeable in the mice with pain hypersensitivity.
The pain hypersensitivity and gut permeability were prevented by treatment with a monoclonal antibody against IgE and did not develop in transgenic mice lacking IgE. Conversely, naïve mice sensitized to an OVA-specific IgE antibody displayed pain hypersensitivity upon exposure to OVA, implicating IgE as a key mediator.
The mechanism depended on the coincidence of the infection with OVA. “Oral tolerance was broken because of this ‘bystander’ effect,” Boeckxstaens later told PRF in an email. “Because ovalbumin was present at the time of infection, by accident the immune system not only raised a response against the infectious agent – the bacterium – but also to the food antigen.”
Unsurprisingly to Aguilera-Lizarraga and Boeckxstaens, mice infected and exposed to OVA showed a shift in the makeup of their microbiome, the community of microbes that populate the gut. But visceral hypersensitivity developed independently of the shift, and it developed even in mice treated with antibiotics to deplete their microbiome, indicating the shift was purely coincidental.
Ongoing inflammation is thought to be a culprit particularly in post-infection IBS. In the hypersensitive mice, noticeable colon inflammation was resolved by seven weeks after infection, but experiments revealed that mast cells were involved in the ongoing pain hypersensitivity. The gene Tpsab1, which encodes a tryptase enzyme found only in mast cells, was overexpressed in infected mice, suggesting the cells were sensitized. When mast cells isolated from infected, OVA-exposed mice were incubated with a peptide piece of OVA, they degranulated, a process where they become activated and in this case release histamine, a major mast cell mediator.
Richard Traub, who studies visceral pain at the University of Maryland, Baltimore, US, and was not involved in the study said, “The fact that infection resulted in IBS later on is not surprising, but the mechanism was unknown. That’s what’s interesting, this new insight into the mechanism that contributes to post-infection IBS.”
Stop mast cells, prevent the pain
Certain drugs, including an agent called doxantrazole, work to stabilize mast cells from degranulating. “When we treated animals with the mast cell stabilizer, similar to in humans, we saw it had an effect in normalizing pain in these mice,” Aguilera-Lizarraga told PRF. “From then on, we tried to look for different factors that could lead to mast cell activation.”
Stopping mast cell degranulation could be a successful strategy for IBS therapeutics, Traub said, because “once the barn door is open, it doesn’t matter if you close it, the horses are out,” referring to degranulation. “Once they degranulate, further blocking degranulation doesn’t seem to make a difference. Once nerve hypersensitivity develops, it’s there. If you can block sensitization of afferents from starting, that is going to be beneficial.”
Ablation of mast cells or of B cells and plasma cells, which produce IgE antibodies, also prevented the development of pain hypersensitivity. “If they’re gone, we have no phenotype,” Boeckxstaens said. “So we have so much evidence that an adaptive immune response is responsible.”
As expected, visceral sensory nerves from mice with the pain hypersensitivity phenotype discharged more action potentials in response to mechanical stimulation of the mucosa than those from mice without pain. Similarly, nerves showed increased excitability in response to application of colonic fluid from hypersensitive mice compared to fluid from normo-sensitive mice. The nerves were also more sensitive to capsaicin, an agonist of the transient receptor potential channel TRPV1, which plays a key role in IBS pain. The increased firing was blocked when the fluid bathing the nerves also included the histamine-1 receptor (H1R) antagonist pyrilamine. Treating the hypersensitive mice with pyrilamine also reduced colorectal pain, indicating an important role for H1R downstream of immune cell activation.
Aguilera-Lizarraga next wondered whether superantigens (SAgs) could be contributing to the IBS model phenotype. SAgs are microbial proteins that cause widespread immune cell activation and have been linked to so-called atopic conditions like asthma. Mice exposed to the SAg staphylococcal enterotoxin B and subsequently exposed to OVA also developed an immune response to OVA, including production of OVA-specific IgE antibodies, and they displayed pain hypersensitivity and gut permeability similar to mice with Citrobacter rodentium infection. The mice also had upregulated Tpsab1 and other pro-inflammatory factors. The findings suggest that the SAg exposure could “break” oral tolerance to OVA, much like the infection did.
Translating to people
Switching to humans, the researchers next examined fecal samples from IBS patients and healthy volunteers for levels of SAg-producing bacteria. Nearly a quarter of the samples from the IBS patients contained Staphylococcus aureus compared to only one in 10 from the healthy controls, and nearly half of these IBS patient samples contained SAgs compared to only one in five from controls.
In an effort to recapitulate the mouse findings in people with IBS, the researchers then injected the intestinal mucosa of 12 IBS patients and eight healthy volunteers with a solution containing soy, wheat, gluten, and milk – foods that commonly cause irritation in IBS patients in the absence of food allergy. All of the patients, but only two of the healthy controls, developed local edema in reaction to the food antigens and had increased indicators of mast cell activity. Upon visual examination of biopsy samples from guts, the researchers saw similar numbers of mast cells and IgE-positive mast cells in patients and healthy controls, but those with IBS had more IgE+ cells in closer proximity to visceral nerves, which correlated with patients’ pain severity.
“It’s showing up in a number of autoimmune and inflammatory conditions that immune cells are setting off the nerves,” Westlund High said. “Having these activated mast cells right at the [nerve] terminals all the time, aggravating them, poises them to explode, really. They’ve shown that again here.”
The work describes an elegant mechanism for how food-related abdominal pain could arise in IBS, and it suggests multiple possibilities for new therapeutic approaches, from blocking IgE antibodies, to stopping mast cell degranulation, to blocking histamine signaling.
“It seems like pharma could just go on this paper and come up with a whole lot of new treatments for people,” Westlund High said. “That struck me as why this paper is so important.”
The need is great, she added. “This is such an important area for patients; it’s of broad relevance.” While some people have minor IBS symptoms, she said, “there are others who are getting by day to day, hoping not to set off a flare. It’s very debilitating and frankly disabling for people who have true IBS.”
Stephani Sutherland, PhD, is a neuroscientist and freelance journalist in Southern California. Follow her on Twitter @SutherlandPhD.
Image credit: Aguilera-Lizarraga et al., 2021