Food allergies and our normal microorganisms
A news article that made the rounds through the popular press this week caught my eye: “Commensal bacteria protect against food allergen sensitization,” which appears in the early access section of the journal of the National Academy of Sciences. I have been a big fan of this type of research for a while now. The basic premise is this: our modern lifestyle has potentially begun to diminish the numbers and variety of microorganisms that live on our bodies in the absence of disease (the normal microbiota), and as a consequence, benefits that these benign organisms can confer to us are being lost. So far, loss of diversity of the normal microbiota have been correlated with a long list of ailments including potentially autism and cancer.
An opinion piece in this week’s Nature warns against drawing too many conclusions from these studies, and suggests that over reporting of some of them by the press reinforces the need to ensure that the public understand the distinction between “correlation” and “causation”–these concepts are frequently confused, and the distinction is sometimes not clear. Indeed, the editorial in Nature suggests that reporting of microbiome analysis and human disease should be tempered by asking 5 questions:
- Can experiments detect differences that matter? Characterization of microbiomes is generally accomplished by sequencing very highly related genes, and this analysis may hide real differences.
- Does the study show causation or just correlation? Many of the cases of a disease association with certain microorganisms may be the result of conditions in the body becoming favorable for the microbe, meaning the disease caused the microbes to alter.
- What is the mechanism? Demonstrating causation is important, however without an explanation of how a change occurs, it is not sufficient.
- How much do experiments reflect reality? Many of the putative effects of the microbiome on health involve germ free mice; that is mice that have been raised to have no normal microorganisms of their own, as this makes interpreting the effects somewhat easier. However, mice and humans are not the same, and the microorganisms that live on each are not the same.
- Could anything else explain the results? Many things can cause disease, and other factors should be considered and tested.
With this in mind, I read the article on food allergies linked up at the top. The authors carried out the study to address the hypothesis that the normal microbiota of the gastrointestinal tract are able to guide adaptive immunity at this site. The intestinal tract of animals hosts an incredible variety of organisms in the absence of disease. The immune system needs to be non-responsive to these organisms, as well as to all of the food antigens that enter the digestive tract. Immune cells in lymphoid tissue along the digestive tract modulate signals between the microbiota and the epithelial barrier of the digestive tract, which helps to prevent an ongoing inflammatory response, and thereby promote a homeostatic relationship between the microbiota and the host.
The researchers first experiment was to treat neonatal wild-type mice with an antibiotic regimen prior to weaning to eliminate intestinal microbiota, then sensitized by gastric administration of Peanut Antigen (PN). Three weeks later, the mice were challenged with the antigen and allergic responses were measured a day later by collecting blood. Control mice had essentially undetectable levels of allergic responses, while antibiotic treated mice showed highly elevated levels of IgE. Analysis of the bacteria from feces of mice at the same time intervals also showed that the antibiotic treated mice had lowered levels of fecal bacteria, and greatly diminished diversity of fecal bacteria. Specifically, members of the prokayotic phyla Bacteriodetes and Firmicutes, present under normal conditions, were essentially absent in the antibiotic treated mice. These bacteria were replaced with members of Lactobacilli, a result consistent with another recent report examining changes in the microbiota of antibiotic fed mice. The results outlined above were achieved using outbred mice strains housed in pathogen-free, but not germ free conditions; therefore this study addresses one of the critiques above with the use of outbred mice.
This paper was also significant, in that the authors also propose a mechanism for how the immune modulation occurs. Recolonization of antibiotic fed mice with a group of Firmucutes from genus Clostridium (the major genus of the Firmucutes from normal mice), prevented the allergic response produced by peanuts. Dissection of the intestines from these animals indicated that specific T cells involved in adaptive immune regulation are more prevalent in Clostridia colonized mice. Additionally, mice colonized with Clostridia in comparison to germ free mice and control mice exhibited high levels of an immune cytokine Interleukin 22 (IL-22). The authors propose that IL-22 (induced by the presence of Clostridia) causes the intestinal epithelial barrier to be reinforced, reducing the permeability to dietary proteins. To address this possibility, they then measured the levels of food allergens in the bloodstream after intragastric gavage. Colonization by Clostridia resulted in significantly lower levels of these allergens in comparison to germ free mice, supporting this hypothesis.
The major conclusions of this paper support the important role of the benign normal microbiota in promoting health. Their model argues that tolerance to food antigens is aided by the presence of those antigens along with specific components of the normal microbiota. To translate this work to human therapies, the role of Clostridia needs to be confirmed in humans. Indeed, other work has shown that Clostridia species isolated from human feces do induce the same immune regulatory cells discussed above when transferred to germ free mice, suggesting that they may be playing similar roles in both species.