Studies of babies growing up in Estonia and Sweden, together with data from germ-free mice, indicate that allergy is the result of an unregulated, rather than a biased, immune system. The conclusion, drawn today by Bengt Björkstén, executive director of the Karolinska Institute in Stockholm, stems from his hypothesis that gut flora is essential for immune regulation. The findings could have implications for irritable bowel disease and type 1 diabetes.

The prevalence of allergy in Estonia and other Eastern European countries has not followed the dramatic increases observed in the West. In recent years this has been explained by the "hygiene hypothesis." The hypothesis contends that an obsession with cleanliness has reduced Westerners' exposure to infections that are fought off with Th1 cells; instead, the immune system more often uses Th2 cells, which orchestrate allergic reactions.

The hypothesis appears to be supported by the observation that individuals with type 2 allergic responses are also less likely to suffer classic Th1-mediated immune diseases, including type 1 diabetes.

But in countries where the prevalence of type 1 diabetes is highest, the prevalence of allergic disease is also highest, says Björkstén. What is true for the individual is not true for the population.

"Until recently the so-called hygiene hypothesis was that if you have the Th2 skew you got allergy and the lack of microbial pressure would explain why you don't have type 1," said Björkstén. "However, it would not explain that if you took it the other way round then you would have type 1 diabetes. In traditional societies both of the diseases are uncommon."

Experiments carried out in Björkstén's lab using germ-free animals have shed some light on the problem.

"Germ-free animals produce higher levels not only of IL-4, a Th2 cytokine, but also of interferon gamma [a Th1 cytokine] compared with conventional animals," said Björkstén. In the total absence of microbial pressure these animals appear to lack an immune "dampening mechanism," he says. Both type 2 and type 1 immune responses are overactive.

"The real interesting aspect is that IL-10 and TGF-beta (transforming growth factor beta) will dampen this." IL-10 and TGF-beta are produced by so-called regulatory T cells, which tone down immune responses.

"This is what we see in Estonia," he said. Estonian children who have very high levels of IL-4, which would be associated with allergy in the West, also have high levels of IL-10, which is not seen in the West. This appears to prevent the IL-4 from triggering type 2 allergic responses, he says.

Regulatory T cells are probably trained up early in the development of the immune system during exposure to harmless gut microflora contends Björkstén.

His hunch is backed up by studies he has carried out on babies in Sweden and Estonia.

"The Estonian gut flora was similar to that described in Western Germany and Sweden in the early 1960s," said Björkstén. Within days of birth, healthy Estonian babies are colonized 3000 times more heavily with Staphylococcus than are Swedish babies. At one month of age, colonization with Lactobacillus is also significantly raised.

"I suggest that microbial pressure can actually drive the [immune] system into a balanced response," concluded Björkstén. This could be a balanced response to allergens, he says, but it could equally be a balanced response to autoantigens, such as those that would otherwise trigger diabetes, or even to normal gut bowel contents, which could otherwise cause inflammatory bowel disease, another scourge of Western countries.