Does the Microbiome Cause Ankylosing Spondylitis?

James T. Rosenbaum, MD

James T. Rosenbaum, MD is Chief of the Division of Arthritis & Rheumatic Diseases, and Edward E Rosenbaum Professor of Inflammation Research at Oregon Health & Science University; Richard Chenoweth Chair of Ophthalmology at Legacy Devers Eye Institute in Portland, Oregon; and former Chair &a

Physicians and scientists relentlessly search for clues as to what causes or affects a disease. Is it more common in women or in men? Is it more common in certain geographic areas? Is there any obvious environmental trigger such as an infection? Such clues can speak volumes about pathogenesis, a fancy medical term for causation. And understanding causation holds promise for finding improved therapy.

Few clues in medicine are as dramatic as the relationship between HLA-B27 and the development of ankylosing spondylitis. HLA-B27 is a genetic trait that one is born with, much as your blood type, A, B, AB, or O, is determined at birth. While there are only four blood types, there are hundreds of HLA types. About one person in every 14 in the U.S. has the B27 type. Nearly 40 years ago, a group in Los Angeles and a group in London discovered that the B27 type increased the odds of developing ankylosing spondylitis by about 100 fold. This is a phenomenal clue because most genetic factors increase the risk of developing a complex disease less than two fold. We now know a great deal about the HLA-B27 molecule. We think that we know its function, which is to help some cells in the immune system communicate or present antigen to a type of white blood cell called a T lymphocyte. And we think that white blood cells are the basis for a variety of immune mediated diseases including ankylosing spondylitis. We know that there are at least 65 closely related variants of HLA-B27 and all but two of these increase the risk to develop ankylosing spondylitis. We know that rats that are genetically manipulated to express HLA-B27 along with an important cofactor called beta 2 microglobulin spontaneously develop spinal arthritis that mimics many aspects of ankylosing spondylitis. But with all these tools to study HLA-B27, we still have not figured out why it increases the likelihood to develop ankylosing spondylitis so dramatically.

Since conventional approaches to understand the relationship between ankylosing spondylitis and B27 have failed, the time seems ripe for a more novel theory. Enter the concept of the microbiome.

The term, microbiome, has been attributed to the Nobel laureate, Joshua Lederberg. It refers to the microbial world that lives in us and on us usually to mutual benefit. For example, most of the vitamin K which we need to make clotting factors is provided by bacteria in our gut. It is estimated that our bodies contain ten trillion cells. But we are host to 100 trillion bacterial cells. And for every RNA message that one of our cells is producing, the bacteria within us are producing 100 more. Surely these bacteria are important to our health.

Recent studies in mice have revealed that bacteria profoundly shape our immune system. It is possible to raise a mouse or a rat in a sterile environment such that bacteria never have a chance to grow inside the intestine. Without these bacteria, the immune system never develops. If the HLA-B27 positive rat is raised in this type of “germ free” environment, it develops very little arthritis. Mouse models of various diseases like asthma and colitis can be effectively treated just by feeding mice specific bacteria, and sometimes even just feeding products from those bacteria. The B27 positive rat develops reduced arthritis if it receives antibiotics. And this remission is sustained if it swallows a specific strain of lactobacillus but not if it swallows a different strain.  Lactobacilli are ingested by millions daily in the form of yogurt. And the strain of lactobacillus used to treat the B27 positive rat can be purchased in a health food store or in specific yogurts.

Ankylosing spondylitis clearly is a disease caused by the immune system. For a century we have known that certain bacteria can trigger a reactive arthritis, especially in HLA-B27 positive individuals. Now that we know that intestinal bacteria shape our immune system. Because all HLA molecules affect the immune response, it is highly likely that HLA-B27 helps to determine which bacteria are in our microbiome and that in turn affects the likelihood that we will develop ankylosing spondylitis.

Before you rush out to buy yogurt or lactobacillus, there are a number of holes to fill in this hypothesis. First, if HLA-B27 does change the bacteria that live within us, we need to define this change. But this is no simple task. A major reason why we are just now learning about the microbiome is that most of the bacteria in the gut do not grow on a culture plate. They are very fastidious, meaning that they only grow when they have specific conditions within the bowel. New molecular techniques have allowed scientists to study the bacteria without culture. This is done similarly to the way a forensic pathologist would attempt to identify a person from a single hair or a drop of blood. Second, the easiest bacteria to study are those that are present in feces, but the bacteria in the appendix, or the cecum, or the ascending colon might be the critical ones to study. Third, people are not mice and while we can extrapolate a great deal from the immune system of mice to the immune system of man, there are clearly differences. And finally, very closely related bacteria might have different effects. In the rat studies, one strain of lactobacillus worked in the B27 rats, but lactobacillus can worsen another rodent model of arthritis.

Psoriatic arthritis and inflammatory bowel disease each have features that overlap with ankylosing spondylitis and many experts believe that each is triggered by a reaction to bacteria.

So if it is true that HLA-B27 shapes the microbiome and the microbiome shapes the immune response and thus leads to ankylosing spondylitis, how can we test this hypothesis? Animal studies will be invaluable although diseases in mice and rats resemble human disease but never exactly duplicate it. We could design a study in which a probiotic like a specific lactobacillus or a diet was used to change the microbiome and thus prevent ankylosing spondylitis. The challenge with such a study is to determine when ankylosing spondylitis begins. The onset of AS is often so insidious that it might be impossible to know which subjects were truly prevented from developing inflammatory low back pain. The uveitis or eye inflammation associated with ankylosing spondylitis is much easier to date in terms of onset because the eye typically becomes affected over a few days and new redness, pain, and light sensitivity develop. Trying to prevent uveitis attacks in patients with AS by using a diet or probiotic might be an excellent surrogate for trying to prevent the spinal arthritis.

Even in the year 2011 when we have achieved so much scientifically, we still have many unknowns and many frontiers.  The microbiome is a vast frontier about which very little is known. Alteration of the microbiome potentially could be used to treat or prevent many diseases caused by the immune system including ankylosing spondylitis.