Immune Defect Makes Gut Bacteria Turn Bad

In a striking new mouse model, an immune system glitch converts mild-mannered intestinal bacteria into a pathological army that inflames the host’s colon and spreads the problem to normal mice.

The findings emphasize the importance of the interplay among the gut, its microbes, and its immune squad. The model opens new avenues to study inflammatory bowel disease and to test potential new therapies for people, said senior author Laurie Glimcher, the Irene Heinz Given professor of immunology at HSPH, and her colleagues. The study appears in the Oct. 5 Cell.

“To my knowledge, this is the first paper to show that a host immune defect can shape the composition of microflora in a way that results in pathology,” said Lora Hooper, assistant professor of immunology and microbiology at the University of Texas Southwestern Medical Center in Dallas.

“That’s interesting in and of itself, but they went one step further and induced this inflammation in healthy mice by transferring the microflora,” said Hooper, who was not involved in the study. “The implications of that are profound.”

The researchers did not identify a specific disease-causing bacterial species, said lead author Wendy Garrett, a postdoctoral fellow in the Glimcher lab. They ruled out known troublemakers, such as Helicobacter pylori, Salmonella, and E. coli. Antibiotic experiments narrowed the probable menace down to an anaerobic species. Hooper speculates that a complex change in the community of bacteria likely unleashes renegade behavior among several species. The Glimcher lab is trying to unmask the culprits in collaboration with Jeff Gordon at Washington University in St. Louis, whose group may be best known for finding evidence that intestinal bacterial composition may contribute to obesity.

Except for isolated reports, there is little evidence that inflammatory bowel disease is contagious among people, said Richard Blumberg, chief of gastroenterology at Brigham and Women’s Hospital. In the study, the maleficent microbial mix may have passed from mother to pup during birth and by feces-eating habits of unrelated mice.

No one knows exactly what causes inflammatory bowel disease. There is no cure, only treatment for symptoms, such as relieving the inflammation, and supportive therapies, such as nutrition, said Wayne Lencer, chief of gastroenterology at Children’s Hospital Boston. Antibiotics may work in some patients for one major type—Crohn’s disease—but not for the other—ulcerative colitis.

“We have pretty good genetic evidence that it’s an abnormality of the immune system at one level and that the disease is due to a dysregulated immune response within mucosal tissues to bacteria normally resident in the intestines,” said Blumberg, HMS professor of medicine. But the primary cause—immune system or bacteria—is debated, he said. This study knits together the two concepts.

“What’s amazing here is that the microbe appears to be coming from the normal commensal bacteria population,” said Lencer, HMS associate professor of pediatrics. “We live symbiotically with microbes. They’re essential. We typically don’t think of commensals as invasive. The study really makes poignant how important this dialogue is between commensals and our cells.”

From many perspectives ranging from microscopic to physical examination, the mouse disease looks remarkably like human ulcerative colitis. In adults and children, the disease presents as continuous sores in the thin lining of the large intestine beginning at the often prolapsed rectum but limited to the colon. (Crohn’s disease, in contrast, extends more deeply into the intestinal wall at sporadic locations and occurs throughout the gastrointestinal tract all the way up to the mouth, but mostly in the small intestine.)

“We did not set out to establish a model of communicable ulcerative colitis,” Glimcher said. Instead, researchers in her lab were systematically exploring the role of a transcription factor, T-bet, in adaptive and innate immunity. Discovered in her lab seven years ago, T-bet seems to single-handedly control the destiny of naive CD4+ T helper cells into type 1 and not type 2. The studies have found that this power is necessary for effective defense against pathogens and cancer cells; it is protective in asthma but pathogenic in the setting of autoimmunity.

About three years ago, Glimcher noticed her mouse facility bills going up and asked why. Co-author Geanncarlo Lugo-Villarino, a postdoctoral fellow now at the University of California, San Diego, told her that his mice had all become sick. For his thesis work elucidating the function of T-bet in the innate immune system, he was studying T-bet–deficient mice with no T cells or B cells, known in scientific shorthand as TRUC.

Postdoctoral fellow Graham Lord, now a professor of medicine at King’s College London and a co-author, took on the task of determining what made the mice sick. A formal analysis showed a distinctive disease phenotype. “I’ve looked after plenty of patients with this,” said Lord, a renal transplant physician who continues to collaborate with Glimcher. “It looks just like human ulcerative colitis.”

Garrett, an HMS instructor in medicine at Dana–Farber Cancer Institute and Brigham and Women’s Hospital, joined the lab two years ago and took over the project when Lord moved to London. She was able to detect the first signs of colonic permeability in mice as young as three and a half weeks. Electron microscopy showed holes in the colonic epithelium in mice at the same young age.

Together with colleagues, the team conducted a multitude of experiments that delineate the etiology in the mouse model with intriguing implications for human disease. They first narrowed down the innate immune cell that expressed T-bet, settling on dendritic cells after ruling out macrophages, mast cells, and natural killer cells. The transcription factor T-bet usually activates immune cells to produce factors—and contributes to T cell–mediated colitis. But in dendritic cells, T-bet binds to the promoter region of the TNF-alpha gene and suppresses production of the inflammatory cytokine. Human colonic dendritic cells also employ T-bet to suppress TNF-alpha.

In mice, both TNF-alpha antibodies and TNF-alpha antagonists can prevent and cure the disease. A recent paper from another group reported that TNF-alpha antagonists, approved to treat rheumatoid arthritis, seemed to alleviate ulcerative colitis in people. Unlike in people, antibiotics also cured the mouse colitis. That gave the researchers a major clue to what activated the dendritic cells.

“Dendritic cells can project their processes into the colonic lumen and sample the bacteria there,” Garrett said. T-bet–deficient mice with an adaptive immune system did not get sick. The team identified a protective population of T regulatory cells, a lymphocyte that suppresses effector T cells and is in clinical development as a potential therapy for other conditions, Glimcher said.

The researchers turned again to the intestinal bacteria. All of the TRUC mice spontaneously suffered the disease. Normal mice with intact immune systems acquired a milder form of the disease by living in the same cage. “We could cure it with antibiotics, but it was more interesting to us that they got sick,” Garrett said.

In people, resident bacteria outnumber human cells by as much as 100 to 1. Most of these microbes hunker down in the colon. Many of the estimated 500 unidentified commensal species perform essential symbiotic tasks of processing nutrients. The colonic epithelium layer—only one cell thick—separates the hordes of microorganisms from the underlying mucosal epithelial cells, where dendritic cells lurk.

Now, Glimcher and her colleagues are seeking the identity of the rogue bacteria. “If you think about it, one thin lining is all that is protecting the body from the millions of bacteria teeming in the gut, and the peacekeeper at the gate is the immune system,” Glimcher said. “Both the adaptive and innate immune systems want to keep the bacteria in the gut and out of mucosal organs.”

Based on the mouse model, Glimcher and her colleagues think inflammatory bowel disease starts with genetic variations that predispose people to disease. Then, a hyperactive and misbehaving immune system leads to secretion of TNF-alpha, which injures the colonic epithelial lining. The new colonic environment somehow transforms the intestinal bacteria into a harmful mix that can cause disease in a healthy gut.

“Time will tell just how faithful a model of human disease this is,” Glimcher said. “Just because it looks like human disease histologically and can be cured by agents that ameliorate the disease doesn’t mean it’s the same.”