The hygiene hypothesis, first put forth by the British epidemiologist David Strachan in 1989, suggests that the cleaner our environment becomes, the more common allergic illnesses become. Under normal circumstances, cells of the immune system circulate throughout the body and recognize antigens, small fragments of microbes or parts of your body (so-called self). A correctly developed immune system is trained to distinguish self-antigen—molecules made by the body that are supposed to be there—from pathogens—foreign invaders that present a threat to the body. According to the hygiene hypothesis, a crucial part of this training process is exposure early in life to germs and bacteria that help the body recognize what outsider molecules really look like. Without this early exposure to pathogens, the immune cells skew towards an auto-reactive or over-reactive response, getting worked up when they see self-antigen or other non-threatening antigen. This overreaction can lead to allergies, asthma, and perhaps also autoimmune diseases like type 1 diabetes.
The hygiene hypothesis sounds plausible, especially in light of studies that have found increasing rates of diseases like type 1 diabetes in developed and developing countries. However, proving the hygiene hypothesis or finding the mechanisms by which it might operate has been very difficult.
But last month, Richard S. Blumberg, Dennis L. Kasper and a team of researchers at Brigham and Women’s Hospital in Boston published a study in Science that provides evidence supporting the hygiene hypothesis, as well as a potential mechanism by which it might occur.
The researchers studied the immune system of germ-free mice—mice lacking bacteria or any other microbes—and specific-pathogen-free mice—mice living in an environment without disease-causing bacteria but with normal germs and microbes. They bred both types of mice to develop forms of asthma and inflammatory bowel disease (IBD) and compared their immune systems. They found that the germ-free mice had more invariant natural killer T (iNKT) cells in the diseased tissues. iNKT cells help the body fight infection, but they are also responsible for responding to cells within the body that look damaged or diseased. In IBD, iNKT cells responded excessively to cells within the colon, leading to inflammation and dysfunction, and in the germ-free mice, many more iNKT cells accumulated and became reactive than in the normal mice. This accumulation of iNKT cells was not limited to the colon, as the researchers found similar build-up of both iNKT cells and the cytokines, the inflammatory signaling molecules they secrete, in the lungs of germ-free mice induced to develop asthma.
The researchers also found that when they exposed germ-free mice to microbes in their first few weeks of life, the mice did not develop high levels of iNKT cells, and they did not develop the inflammation in the colon and lungs seen in those kept germ-free. However, if the germ-free mice were exposed to the microbes when they were adults, there was no beneficial effect. Thus, they concluded that for benefits to occur, the exposure to germs had to happen before the mice reached adulthood. Additionally, they found that the protective effects of being exposed to microbes early in life were long-lasting, as exposure early in life to microbiota was sufficient to normalize the germ-free mice for much of their life.
This germ-iNKT cell balance, though, is a delicate one. Exposure to microbiota is important to ensure there aren’t too many iNKT cells, but having too few iNKT cells is also a problem. In fact, one of the peculiar characteristics of mouse models of autoimmune diabetes is that the mice have fewer, less potent iNKT cells, and it’s not clear why this might be, or what role it plays in the development of the disease. Blumberg has not looked at iNKT cells in diabetes, but speculates that in diabetes “NKT cell activation may protect from disease development. It is therefore possible that early life events such as those we describe might be linked to either later life risk or protection from diabetes.”
The interplay of the microbial environment, iNKT cells, and diabetes is clearly still an open question, then, but Blumberg’s recent results imply that this is an avenue worth pursuing. It’s too early to say what this research means outside of mouse models. But a real life environment that might be reflective of Blumberg’s work is the use of antibiotics in early life. This, Blumberg says, could “create a germ-reduced environment at critical phases of microbial colonization as we demonstrated [in mice]. But such observations require direct study in order to extrapolate to humans.”
Karmel Allison is science editor of ASweetLife. She writes the blog Where is My Robot Pancreas?
Jessica Apple contributed to this article.