We know that the proteins called gluten, found in wheat and other grains, provoke celiac disease. And we know how to treat the illness: a gluten-free diet. But the rapidly increasing prevalence of celiac disease, which has quadrupled in the United States in just 50 years, is still mystifying.
Scientists are pursuing some intriguing possibilities. One is that breast-feeding may protect against the disease. Another is that we have neglected the teeming ecosystem of microbes in the gut — bacteria that may determine whether the immune system treats gluten as food or as a deadly invader.
Celiac disease is generally considered an autoimmune disorder. The name celiac derives from the Greek word for “hollow,” as in bowels. Gluten proteins in wheat, barley and rye prompt the body to turn on itself and attack the small intestine. Complications range from diarrhea and anemia to osteoporosis and, in extreme cases, lymphoma. Some important exceptions notwithstanding, the prevalence of celiac disease is estimated to range between 0.6 and 1 percent of the world’s population.
Nearly everyone with celiac disease has one of two versions of a cellular receptor called the human leukocyte antigen, or H.L.A. These receptors, the thinking goes, naturally increase carriers’ immune response to gluten.
This detailed understanding makes celiac disease unique among autoimmune disorders. Two factors — one a protein, another genetic — are clearly defined; and in most cases, eliminating gluten from the patient’s diet turns off the disease.
Yet the more scientists study celiac disease, the more some crucial component appears in need of identification. Roughly 30 percent of people with European ancestry carry predisposing genes, for example. Yet more than 95 percent of the carriers tolerate gluten just fine. So while these genes (plus gluten) are necessary to produce the disease, they’re evidently insufficient to cause it.
Animal studies have reinforced that impression. In mice engineered to express those H.L.A.’s, tolerance to gluten must be deliberately “broken.” Without an immunological trigger of some kind, the rodents happily tolerate the protein.
A recent study, which analyzed blood serum from more than 3,500 Americans who were followed since 1974, suggested that such a trigger could strike adults at any time. By 1989, the prevalence of celiac disease in this cohort had doubled.
“You’re talking about an autoimmune disease in which we thought we had all the dots connected,” says Alessio Fasano, head of the Center for Celiac Research and Treatment at the Massachusetts General Hospital for Children in Boston, and the senior author of the study. “Then we start to accumulate evidence that there was something else.”
Identifying that “something else” has gained some urgency. In the United States, improved diagnosis doesn’t seem to explain the rising prevalence. Scientists use the presence of certain self-directed antibodies to predict celiac disease. They have analyzed serum stored since the mid-20th century and compared it to serum from Americans today. Today’s serum is more than four times as likely to carry those antibodies.
BLAME for the increase of celiac disease sometimes falls on gluten-rich, modern wheat varietals; increased consumption of wheat, and the ubiquity of gluten in processed foods.
Yet the epidemiology of celiac disease doesn’t always support this idea. One comparative study involving some 5,500 subjects yielded a prevalence of roughly one in 100 among Finnish children, but using the same diagnostic methods, just one in 500 among their Russian counterparts.
Differing wheat consumption patterns can’t explain this disparity. If anything, Russians consume more wheat than Finns, and of similar varieties.
Neither can genetics. Although now bisected by the Finno-Russian border, Karelia, as the study region is known, was historically a single province. The two study populations are culturally, linguistically and genetically related. The predisposing gene variants are similarly prevalent in both groups.
Maybe more telling, this disparity holds for other autoimmune and allergic diseases. Finland ranks first in the world for Type 1 autoimmune diabetes. But among Russian Karelians, the disease is nearly six times less frequent. Antibodies indicative of autoimmune thyroiditis are also less prevalent, and the risk of developing allergies, as gauged by skin-prick tests, is one-fourth as common.
What’s the Russians’ secret?
“It’s a remote territory of Russia,” says Heikki Hyoty, a scientist at the University of Tampere in Finland. “They live like Finns 50 years ago.”
At the time of this research, roughly a decade ago, Russia’s per-capita income was one-fifteenth of Finland’s. Analysis of house dust and potable water suggests that the Russian Karelians encountered a greater variety and quantity of microbes, including many that were absent in Finland.
Not surprisingly, they also suffered from more fecal-oral infections. For example, three of four Russian Karelian children harbored Helicobacter pylori, a corkscrew-shaped bacterium, while just one in 20 Finnish children did. The bacterium can cause ulcers and stomach cancer, but mounting evidence suggests that it may also protect against asthma.
Professor Hyoty suspects that Russian Karelians’ microbial wealth protects them from autoimmune and allergic diseases by, essentially, strengthening the arm of the immune system that guards against such illnesses.
Meanwhile, Yolanda Sanz, a researcher at the Institute of Agrochemistry and Food Technology in Valencia, Spain, makes a compelling case for the importance of intestinal microbes.
Years ago, Dr. Sanz noted that a group of bacteria native to the intestine known as bifidobacteria were relatively depleted in children with celiac disease compared with healthy controls. Other microbes, including native E. coli strains, were overly abundant and oddly virulent.
How to determine cause or consequence?
In a test tube, she found that those E. coli amplified the inflammatory response of human intestinal cells to gluten. But bifidobacteria switched the response from inflammation to tolerance.
In rats, the E. coli again intensified inflammation to gluten, prompting what’s sometimes called a “leaky gut” — the milieu suspected of contributing to celiac disease. Conversely, bifidobacteria protected the intestinal barrier. Microbes, it seemed, could influence the immune response to gluten.
Bifidobacteria occur naturally in breast milk, which, along with protective antibodies and immune-signaling proteins, conveys hundreds of prebiotic sugars. These sugars selectively feed certain microbes in the infant gut, particularly bifidobacteria. Breast-fed infants tend to harbor more bifidobacteria than formula-fed ones.
All of which may explain a curious historical phenomenon — an “epidemic” of celiac disease that struck Sweden some 30 years ago. Anneli Ivarsson, a pediatrician at Umea University, recalled a sudden wave of “terribly sick” infants.
Sleuthing revealed that, just before the spike, official guidelines on infant feeding had changed. In an effort to prevent celiac disease, paradoxically, parents were instructed to delay the introduction of gluten until their babies were six months old. That also happened to be when many Swedish mothers weaned their children. Coincidentally, companies had increased the amount of gluten in baby food.
This confluence produced an unwitting “experiment with a whole population,” says Dr. Ivarsson — a large quantity of gluten introduced suddenly after weaning. Among Swedes born between 1984 and 1996, the prevalence of celiac disease tripled to 3 percent. The epidemic ebbed only when authorities again revised infant-feeding guidelines: keep breast-feeding, they urged, while simultaneously introducing small amounts of gluten. Food manufacturers also reduced the gluten content of infant foodstuffs. Dr. Ivarsson found that, during the epidemic, the longer children breast-fed after their first exposure to gluten, the more protected they were.
Not all subsequent studies have found nursing protective, but partly as a result of Sweden’s experience, the American Academy of Pediatrics now recommends that infants start consuming gluten while still breast-feeding.
Research by Dr. Sanz of Spain again illuminates how this may work. Some years back, she began following a cohort of 164 newborns with celiac disease in the immediate family. By four months, children with celiac-associated genotypes — 117 of them — had accrued a microbial community with fewer bifidobacteria compared to those without. If bifidobacteria help us tolerate gluten, these children appeared to be edging toward intolerance.
There was one notable exception: Breast-feeding “normalized” the microbes of at-risk children somewhat, boosting bifidobacterial counts.
Dr. Fasano of Boston has made another potentially important find. He followed 47 at-risk newborns, regularly collecting microbes from 16 of them, which he analyzed after two years. Like Dr. Sanz, he found these genetically at-risk children to accumulate a relatively impoverished, unstable microbial community.
But it’s a secondary observation that has Dr. Fasano particularly excited. Two of these children developed autoimmune disease: one celiac disease, another Type 1 diabetes, which shares genetic susceptibility with celiac disease. In both cases, a decline of lactobacilli preceded disease onset.
Assuming that the pattern holds in larger studies, “imagine what would be the unbelievable consequences of this finding,” he says. “Keep the lactobacilli high enough in the guts of these kids, and you prevent autoimmunity.”
The caveats here are numerous: the tiny sample size in Dr. Fasano’s study; Dr. Sanz hasn’t yet revealed who actually developed celiac disease in her cohort; and even if these microbial shifts reliably precede disease onset — as they do in larger studies on allergic disease — they’re still bedeviled by the old “chicken or the egg” question: Which comes first, the aberrant microbial community, or the aberrant immune response?
Bana Jabri, director of research at the University of Chicago Celiac Disease Center, notes that immune disturbances change the microbial ecosystem. But here’s the catch: Even if the chicken comes first, she says, the egg can contribute. Rodent experiments show that intestinal inflammation can select for unfriendly bacteria that further inflame. “You can have a positive feedback loop,” she says.
SO your microbes change you, but your genes also shape your microbes — as do environment, breast milk, diet and antibiotics, among many other factors.
Such complexity both confounds notions of one-way causality and suggests different paths to the same disease. “You have the same endpoint,” Dr. Jabri says, “but how you get there may be variable.”
The intricacies don’t stop there.
Not all breast milk is the same. It varies according to diet and other factors. One study found that milk from overweight mothers had fewer of those bifidobacteria than milk from thinner mothers. Another observed that breast milk from farming mothers, who inhabit a microbially enriched environment, carried more anti-inflammatory proteins compared with urban mothers’ milk. “All these things are going to matter,” Dr. Jabri says. And they’re all potential nudge points in the quest to prevent disease.
The tangled web of possibilities should not, however, distract us from the facts on the ground. In a far-flung corner of Europe, people develop celiac disease and other autoimmune diseases as infrequently as Americans and Finns did a half-century ago. The same genes exposed to the same quantity of gluten do not, in that environment, produce the same frequency of disease.
“We could probably prevent celiac disease if we just give the same environment to the Finnish children as they have in Karelia,” says Dr. Hyoty. “But there’s no way to do it now, except to move the babies there.”