Digestion of gluten by microbiota overrides diet-based protection against diabetes

New research by Alexander Chervonsky shows that a casein-based diet can protect against development of type 1 diabetes, but this protection is interrupted when gluten is introduced.

Understanding the root cause of type 1 diabetes (T1D) has been elusive because it involves a complex series of factors, including genetics, lifestyle, and history of related chronic conditions. Diet and the makeup of commensal bacteria living in the digestive system can also affect the development of T1D, but how those two factors work together has been unclear.

In a new study published in Cell Host & Microbe, researchers from the University of Chicago showed that a diet based on casein, a milk protein, protects mice from developing T1D, improving the function of insulin producing cells in the pancreas and limiting the autoimmune response. However, when gluten (the cereal protein commonly associated with celiac disease, another autoimmune disorder) is added to their diet, it overrides the protective effects of casein. The mice continue to produce an autoimmune reaction and develop T1D, but only when certain types of bacteria are present.

The research, said Alexander Chervonsky, MD, PhD, Professor of Pathology at UChicago and senior author of the study, highlights the intricate interplay among diet, the immune system, and its microbial partners in facilitating—or protecting against—disease.

“Gluten has been shown to affect type 1 diabetes in humans, and we know it’s proinflammatory, causing celiac disease. So, it was an obvious choice to test out here,” Chervonsky said. “And sure enough, it reversed the protective effect of casein. But we also see that there must be bacteria present for gluten to have this effect now. That’s when it became most interesting.”

Dietary interventions depend on the microbiome

Microbes that live in the digestive system can have a significant impact on health, generating metabolites, producing signaling molecules, and interacting with the host immune system. The shape of the microbial community can change rapidly in response to diet, so researchers believe that any dietary interventions for treating autoimmune diseases would also be highly dependent on the microbiome.

To study the interactions between diet and microbes leading up to autoimmunity, Chervonsky’s team worked with the non-obese diabetic (NOD) mouse model of T1D, a commonly used strain of mice that are prone to developing diabetes. Once they observed the protective effects of a casein diet, followed by the overriding effect of gluten, they wanted to see how microbes played a role.

The researchers fed the casein diet to germ-free (GF) mice, which are specially raised so they don’t harbor any microbes, and specific-pathogen free (SPF) mice, which don’t have any common disease-causing pathogens. The diet also protected those groups from T1D—the incidence was significantly reduced and in the few mice that became diabetic, the onset was delayed. This means that the protection offered by the diet was microbiota-independent, since GF mice and SPF mice responded the same way. The researchers believe that casein reduces insulin secretion from islet cells in the pancreas, which limits their exposure to autoimmune responses that would otherwise incorrectly attack them.

The researchers next isolated colonies of bacteria to identify which were likely to be involved with digesting gluten, singling out Enterococcus faecalis (E. faecalis), a common commensal bacteria found in the gut. This discovery helped them unravel the chain of events that explains how gluten overcomes the protective effects of the casein diet for these mice. The bacteria secrete protease enzymes that digest gluten. This process releases lipopolysaccharides (LPS), which are components of the outer membrane of gram-negative bacteria. The presence of LPS leads to the activation of macrophages that stimulate more immune T cell responses, which overpowers any protection offered by casein and leads to autoimmunity.

The catch is that E. facecalis are gram-positive bacteria, which have a different kind of outer membrane than gram-negative bacteria. Where is the LPS coming from?

“Finding LPS was a surprise that we definitely did not expect, but it must be coming with gluten somehow,” Chervonsky said. He and his team actually bought different samples of flour with gluten at a supermarket and screened them for LPS—sure enough, it was everywhere. “There are gram negative bacteria everywhere in the soil, so maybe it’s associated with the soil where the grains are grown, or it is contaminated in storage. But we see that it is required in this protein digestion process to stimulate the innate immune response and activation of macrophages.”

Chervonsky said there is more to learn more about what happens to LPS biochemically during this process. He also wants to investigate whether peptides digested by the bacteria somehow activate T cells that cross react with islet cell proteins or just induce inflammation to activate the immune system in general. The team is also planning more studies to understand the biochemistry of how casein provides protection against T1D, potentially through neural regulation of insulin secretion.

The chain of events is still incredibly complex, but with each new study like this, the links among diet, microbiota, and the immune system—and how they contribute to autoimmune disease—are becoming clearer.

The study, “Microbiota-dependent proteolysis of gluten subverts diet-mediated protection against type 1 diabetes,” was supported by the Carlsberg Foundation, the South-Eastern Norway Regional Health Authority, the National Institutes of Health, the Department of Defense, and the Juvenile Diabetes Research Foundation. Additional authors include Matthew C. Funsten, Leonid A. Yurkovetskiy, Andrey Kuznetsov, Camilla H.F. Hansen, Katharine I. Senter, Jean Lee, and Aly A. Khan from the University of Chicago; Derek Reiman from the Toyota Technological Institute at Chicago; Jeremy Ratiu and David Serreze from the Jackson Laboratory; Shiva Dahal-Koirala and Ludvig M. Sollid from the Univeristy of Oslo, Norway; Dionysios A. Antonopoulos from Argonne National Laboratory; and Gary M. Dunny from the University of Minnesota.

Explore the Biological Sciences Division