By David N. Leff

Far up in the North Atlantic, 250 miles northwest of Scotland's tip, lie the 22 Faeroe Islands - a distant outpost of Denmark - at 62 degrees north latitude. Yet the Faeroes are not as bleak as all that. Washed by warm oceanic currents, they bless their 43,000 inhabitants with mild winters and cool summers.

"Yet some time maybe 50 years ago," recalled research immunologist Andrew Caton, at the Wistar Institute in Philadelphia, "there was an outbreak of multiple sclerosis on the Faeroe Islands, presumably due to a viral infection."

For decades, viruses have been indicted - on circumstantial evidence - as perpetrators of multiple sclerosis (MS). One circumstance: MS is six times as prevalent in Winnipeg (125o N) as in New Orleans (30o N). By the same token, MS is rare between the sun-drenched tropics of Cancer and Capricorn.

"For example," Caton observed, "Streptococcus infection, as well as certain types of viruses, can induce rheumatic heart disease. However," he added, "the mechanisms and processes are not well understood."

A cover story in Monday's twice-monthly Journal of Experimental Medicine, dated Dec. 18, 2000 - of which Caton is senior author - contributes to that understanding. Its title: "Virus-induced maturation and activation of autoreactive memory B cells."

"What we've shown," Caton told BioWorld Today, "is a way in which viral infections, and other kinds of infections like them, can trigger autoimmunity. The way we've done it is to use an experimental mouse model we've developed.

"One of the difficulties with addressing the autoimmunity question in humans," he explained, "is that it's hard to know what infections they've had, and very hard to look and figure out ways in which the antibodies and T cells that have been induced by the infection can cross-react with their own 'self' proteins. Healthy people," Caton continued, "get rid of all those T cells and B cells that can recognize their own proteins. That forms a tolerance in educating the immune system not to react against self. In autoimmune diseases, it does. Lupus, for example, features the appearance of autoantibodies that react against the body's DNA."

Caton's Basic Game Plan

"When an animal - or you and I - gets an infection, we make antibodies in two ways: During that first week, we make an antibody that plays a role in getting rid of the disease. Then our immune defenses turn to forming long-term memories. But that memory formation has nothing to do with getting rid of the virus. You acquire these memory B cells and memory T cells, and they last your lifetime. That's why if you catch measles, or get a good vaccination against it, you never get it again. The rest of your life you're making antibodies."

Caton and his co-authors created a mouse strain with its own built-in virally induced autoimmune influenza. "What we did," he recounted, "was to take the hemagglutinin (HA) glycoprotein gene from the flu virus and insert it in the mouse genome. Then we bred the mice in a way that segregated those that had this influenza gene in them, from those who did not. We showed a few years ago, that they had done a good job of getting rid of the HA-specific antibodies that are produced in the first week after the infection - the first wave of antibodies.

"But there were precursors to B cell-raised antibodies that would form part of the second wave. They were still there. That was a very unexpected finding for us.

"So we had a distinction now, where we thought that maybe there were different kinds of B-cell subpopulations, some of which the body was getting rid of, but others were still available. In theory, they could be activated by viruses, and make these memory B-cell responses, as if we were vaccinating - harmfully - against the cross-reactive self. This showed that even if the mouse got rid of the first wave of B cells, it did nothing to stop the second wave.

"People over the years have speculated that there must be some way for the immune defenses to get rid of the cells that react with one's self. They had observed in lupus, during that second phase, that patients had antibodies that clearly had gone through the memory process. What we convincingly show in the case of lupus was that healthy people normally regulate that process of purging self-reactive cells during the memory phase, but for some reason it fails in lupus."

Caton then described a typical in vivo experiment employing his autoimmune influenza mice: "We immunized them with flu virus in a way not very different from giving them a flu shot. A month later, we vaccinated them again, and examined their ability to form memory B cells. That second challenge compared their responses to those in control mice that didn't have the flu antigen. It showed that the HA mice were generating antibody responses as easily as control mice that were generating memory responses.

"They got rid of the first wave of antibody," he said, "and purged the self-reactive antibodies from their systems. But the second wave was not affected. This implied that there really was nothing to prevent B cells from becoming activated after infection. If their precursor B cells are present, there's nothing to stop them from going through this mutational process, and illuminating one's own cells that might become self-reactive."

En Route To Clinical Intervention

"We are now trying to understand," Caton went on, "what is it about an autoimmune antigen in the body that turns these antibodies to something that is dangerous to the animal? We have another mouse where we can see that a similar process may be happening in those animals that develop arthritis and lung disease and anti-DNA antibodies. To make those mice, we've retained the same HA gene, but changed where it is located in the body. In the mice that get sick we expressed the HA on their dendritic and B cells instead of ubiquitously. They can now spontaneously develop an autoimmune disease.

"We think that's important because very few small animal models can develop arthritis spontaneously. We think that's where our real clinical applications will come in. Then," Caton concluded, "we can begin to dissect the cellular processes that raise targets for intervention." n