It's the concept on which our understanding of autoimmunity is based - that the immune system mistakenly sees the body's own tissue as foreign and mounts a response against it.

That image of an immune system fighting foreign invaders - defined as anything we're not born with - is hardly foreign to anyone who has studied, read about or even heard of the body's system of defense. The notion has been taught, cited and studied since the 1940s. But according to Polly Matzinger, it should be forgotten.

Who is Polly Matzinger? If you had known her 25 years ago, you might have answered "a dog trainer," "a composer," "a jazz musician," or "a cocktail waitress." Today she is all these things (except a cocktail waitress; she had to give something up).

She is also chief of an immunology lab at the National Institutes of Health in Bethesda, Md., a tireless researcher, author of about 40 scientific papers and producer of two films on the immune system (one of which won the top award in educational films at the Cologne Film Festival).

But Matzinger may best be known by her colleagues in the scientific community as a maverick. Her radical explanation of the immune system's purpose and function questions the fundamental view of what we now refer to as autoimmunity and challenges the beliefs that leading scientists have held dear for as long as half a century.

From Lounge to Lab

Curiously enough, it was during her stint as a waitress that Matzinger became interested in science. One day a group of professors from the nearby University of California at

Davis came into the bar where she worked. They were discussing a recent research project, and Matzinger, who overheard their conversation, began to question their assumptions.

While some researchers might not welcome a cocktail waitress's comments on their work, one of the group - the late Robert Schwab, PhD, who at the time was chairman of the department of wildlife and fisheries at Davis - was intrigued by her astute comments. The two struck up a friendship.

"He would walk into the bar and leave an article [from a scientific journal]," Matzinger recalls. "Then a few days later he would come back as the bar was closing and we would go to Denny's and drink coffee at 3 in the morning and talk about the article, and he would say, 'Look, science never gets boring.' "

Until Matzinger met Schwab, she says she was pretty much convinced that every job gets boring - thus her checkered resume. "I had about decided I was going to be a cocktail waitress forever, because I was really good at it and I could work at night and do whatever I wanted during the day."

But Schwab encouraged Matzinger to become a scientist and finally changed her mind. "It took him nine months to convince me, but finally I said, 'Why not? I'll try it.' That was 25 years ago and it hasn't gotten boring yet!"

Questioning Long-Held Beliefs

Ironically, the same type of questions that endeared Matzinger to her mentor were met less than enthusiastically by some of her professors at the University of California at San Diego, where she majored in biology. Like all young scientists, she was taught the concept that the immune system fights what is foreign and doesn't attack what is self.

That explanation, however, which rose from the Nobel Prize-winning research of Sir Peter Medawar and Macfarlane Burnet in the 1940s, left Matzinger with questions. "Why then," she asked her professors, "does a pregnant woman not reject her fetus? Why don't lactating women reject breast milk? Why don't our bodies reject the food we eat?

"And what about autoimmune disease?" she asked. "Why does our immune system sometimes attack the organs we are born with?"

Her questions, simple, yet profound, eventually set her on a quest for answers that have the potential not only to treat cancer and organ transplant patients, but possibly to change the way scientists view the immune system and the so-called autoimmune diseases.

At the time, however, they were dismissed by professors who clung to the safe, time-honored explanation of immunity, says Matzinger. "I didn't know how to find the answers, so I gave up asking - at least for a while."

Of Baths, Border Collies and Inspiration

It wasn't until Matzinger had earned her PhD and was working in a lab at the National Institutes of Health that the questions she had asked 10 years earlier again began to trouble her.

"That was when I met Ephraim Fuchs," she says, "a young oncologist working as a postdoc in the lab next door. He was asking the same questions, only he didn't give up." Matzinger admired his tenacity. The two formed a natural team.

She and Fuchs often argued over specific points, but on at least one major concept they agreed: "It made no sense for the immune system to fight everything that was foreign."

"Finally, after a year of arguing we came up with the hypothesis that an evolutionarily useful immune system would really be one that fought things that were dangerous," she says. "After all, there's no reason to fight anything that's harmless, no matter how foreign it is."

With that concept formed, the two began to face a succession of even harder questions, beginning with "How does the immune system detect what is dangerous?"

"We spent two more years trying to figure that one out," says Matzinger. "We finally figured out that it couldn't, there was no way the immune system could discriminate between something that's dangerous and something that isn't."

So the next logical question was, "How can the immune system distinguish between harmful and harmless things?" The answer to that one came one day in 1993 when Matzinger was soaking in her bathtub. She realized, she says, that "dangerous things do damage, and harmless things don't! It was that simple! If there's no danger, there's no damage, and if there is no damage, there is no immune response."

However, there was no obvious way for the immune system itself to discriminate between something that causes damage and something that doesn't. So the next logical question was who - or what - does?

The answer to that question was spurred by an experience with her border collie, Annie. Matzinger and Annie were sitting on a hill watching a small flock of peacefully grazing sheep.

Annie was quietly curled up with her tail tucked under her nose - until a barely audible rustling in the brush aroused the sheep. As the sheep's heads came up and they started to panic, Annie was up in a flash to investigate.

"The whole picture suddenly fell into place," says Matzinger. "It was the panic of the sheep that aroused Annie, not the sound in the bushes. In the same way, the immune system doesn't need to be able to distinguish between dangerous and harmless things; it can listen to its sheep.

"I realized there is a type of cell in the body - called a dendritic cell - that acts much like a sleeping sheepdog," she recalls. The dendritic cells, each of which has little fingerlike projections that bring it into direct physical contact with about 50 to 500 cells in the body, lie peacefully until aroused by signals coming from a nearby cell. However, the neighboring cells, like quietly grazing sheep, don't usually send any alarm signals as long as they are healthy or programmed to die a normal death.

When cells die a normal death (a process called apoptosis), which happens every day in every part of our bodies, they shrivel up unobtrusively and a carefully orchestrated cellular cleanup crew eliminates them from the body. When a cell dies a tragic death, however, the results are entirely different. Whether punctured by a thorn or infected by a virus, these "murdered" cells burst open, spewing their contents into their environment. What better signal that something's amiss?

The dendritic cell, awakened by the leaking contents (or by other alarm signals) then starts the chain of events that sets an immune response in motion.

This is the scenario Matzinger uses as the basis for what has become known as the Danger Model.

Offering Answers, Stirring Controversy

With the Danger Model, Matzinger is offering answers to the questions she asked as a student and later debated with Fuchs. "Why doesn't a mother reject her fetus?" she asks. "Because a healthy fetus doesn't do damage and therefore doesn't present a danger."

On the other hand, an infected fetus - which truly does present a danger to the mother - is usually eliminated from the body." What about breast milk? Doesn't present a danger. What about the food we eat? Unless contaminated, it usually doesn't present a danger.

What about autoimmune diseases? Of all questions posed, this remains the most difficult for Matzinger to answer. Her best educated guess: that in most of what we now know as autoimmune diseases, there is no defect in the immune system, as has been thought for half a century.

"I think there are at least five categories of diseases that people refer to collectively as autoimmune disease. And I think in four of the five there is absolutely nothing wrong with the immune response," she says. In that other category she believes such problems as infection or a defect in apoptosis are the driving force.

Infection is the more likely culprit in diseases that flare, such as rheumatoid arthritis and multiple sclerosis, she suspects. "It could be that some forms of arthritis are caused by infections that are common in the environment and every time your body clears that infection, your disease flares," she says.

In some cases, the responsible infection may be in the affected tissue. For example, an infection in the joint causes joint inflammation. In other diseases, however, the infection may have no proximity to the tissue damage it causes. In these cases a similarity between the affected tissue's proteins - which are different in every person and determined genetically - and certain viruses or bacteria might cause a harmful immune response in the tissue.

The idea that infection might trigger autoimmune disease is nothing new. But Matzinger contends that infection may not only trigger the disease but drive it as well. Eliminate the danger - by clearing up the infection - and the disease will resolve, she suspects.

Another category of disease, which Matzinger refers to as "bad death," may explain lupus, she says. Bad death diseases, she explains, may be caused by a defect in apoptosis, or normal programmed cell death.

With no fewer than 30 genes in the body controlling various aspects of apoptosis, the presence of a defect in any one of those genes could cause the body to view even normal cell death as dangerous.

The Bold and the Unreasonable

Many scientists agree that Matzinger's views - as well as her way of presenting them -are pretty bold, yet not altogether unreasonable. Matzinger's first 40-page paper describing the Danger Model aroused little comment, but when her first experimental challenge to the old self-non-self model, published in the prestigious journal Science in March 1996, received mixed reviews.

Princeton University's Albert Bendalac, MD, PhD, for one, was impressed, calling it a "well thought through conceptual challenge to the way we've looked at the immune system."

Others saw it and became angry, says Matzinger - "partly because it was challenging their fundamental beliefs, but partly because scientists must be somewhat conservative. After all, we can't follow every new harebrained idea. It's only when the old models begin to fall apart that there's mental room for a new one."

Although rheumatology researchers don't know where Matzinger's model fits into the understanding of autoimmunity, some admit that the Danger Model does raise good points. Yes, when it comes to autoimmunity, the old model is starting to fall apart.

In recent years scientists have begun to realize that even perfectly healthy people can have autoreactive antibodies (proteins that can react against the body's own tissue and are a hallmark of autoimmune disease). Yet they have not understood why these antibodies only cause disease in some people.

The Danger Model, which suggests that something damaging in tissue is calling these self-directed antibodies and white blood cells called T cells into action, may offer as good explanation. But it is not complete, says David Fox, MD, professor of internal medicine and chief of the division of rheumatology at the University of Michigan Medical Center in Ann Arbor.

"I think the most important contribution of Matzinger's hypothesis so far has been to raise some questions and point out the inadequacies of the [self-non-self hypothesis]," he says.

In some other areas of medicine Matzinger is crusading for treatment changes that have the potential to literally save lives. In the case of organ transplants, for example, she tells doctors they can block danger signals that are caused largely by the damage that occurs to the organ during its removal from the donor and placement into the recipient. Once that damage caused by the transplant process has healed, they can stop the drug - forever.

For cancer, she advises getting danger signals going until tumors are completely gone and then some. Because tumor cells don't normally die traumatically, they are not recognized by the immune system as damaging, she says.

For diseases such as rheumatoid arthritis and lupus, however, she has no immediate advice, except perhaps to suggest a change in the point of view driving the search for a cause. "Instead of trying to understand why the body fails to distinguish self from non-self, scientists should be looking for the danger signal in the so-called autoimmune disease," says Matzinger. "In other words, they should be asking 'What is driving the response?' "

After all, it is only through asking the right questions - and continuing to ask them even though it would be easier to play it safe - that one can find the most meaningful answers.

Mary Anne Dunkin, senior editor of Arthritis Today, has covered arthritis-related medical and research news for 11 years.