Learning a new body language

A former Bunny girl turned immunologist may have found the answer to why tumours are not attacked by the body's immune system

There is nothing conventional about Polly Matzinger, America's latest immunology star. Addressing a seminar at the recent meeting of the American Association for the Advancement of Science in Seattle, Washington, Dr Matzinger was accompanied by her pet dog Annie. Here was a scientist, whose ideas challenge the fundamentals upon which our understanding of the immune system is based, with an enthusiastic determination to spread her message, and a perfect pet to boot.

Matzinger's background - and her willingness to tell the whole of America about it - is remarkable. Four months after co-authoring a paper in Science, which, according to this leading American science journal, could mean "drastic revisions" to immunology text books, Matzinger gave an interview to the Washington Times. In it she revealed a former life - as a bass player, dog trainer, cocktail waitress and Playboy bunny.

If there was ever a Cinderella story in the science world, this must be it. While waiting tables near the University of California, Matzinger told the Washington Times, she overheard two scientists talking about a recent experiment. Playboy bunny Matzinger suggested errors in their reasoning. A friendship was forged which, says Matzinger, involved nine months of coercion by one of the two scientists to force her back to school. "He said, 'You're a scientist! You should be a scientist'." And finally she gave in.

What Matzinger and her colleagues, John Paul Ridge and Ephraim Fuchs, published 20 years later in Science should therefore perhaps not have been unexpected. Matzinger has a history of challenging convention.

Traditional research says that the body's immune system works by distinguishing between "self" and "non-self" at a very early stage in life.

A cell from the liver is self because it is part of the body at birth, but a flu virus, invading the body after birth, is non-self. Anything that the immune system fails to recognise as self, that is to say as an integral part of the body at birth, is attacked by the immune system's killer cells.

Since the first experiments in the 1950s, immunologists have uncovered unexpected complexity in the immune system. Research suggests that the immune system's killer cells - those cells that attack viral and bacterial infections - require two signals to be activated before they launch an attack. It is widely thought that two cells provide this second signal, T helper cells and antigen presenting cells.

But, says Matzinger, all these theories emerge from the self/non-self mind-set. Results which do not fit this established paradigm have been ignored. Matzinger told the AAAS: "As agraduate student I had a problem with this paradigm. I would ask my professors: if the immune system learns what self is at birth, what happens when you go through puberty (when there is a great deal of cell change)? Most of us don't kill ourselves off at puberty -if we do it's not for immunologicalreasons."

Matzinger goes further. Why is the foetus, half of whose cells are foreign to the mother, not rejected if maternal immune systems distinguish self from non-self? The same applies to food, which is rife with foreign antigens.

With these unanswered questions in mind, Matzinger and her colleagues concluded that a more effective way for the immune system to work would be for it to distinguish not between self and non-self, but between what is dangerous and what is not. But how?

"That took another year," Matzinger told the AAAS. "And one day in the bath came the idea." What Matzinger realised was the importance of the way in which a cell dies.

A natural way for the immune system to detect danger would be for it to differentiate between natural, programmed cell death when the cell shrivels up and its contents are never released, and necrosis - when a cell dies a non-natural death, such as when it is blown open by a virus, and its contents leak or spill out.

Matzinger suggested that it might be dendritic cells, the least understood type of antigen presenting cells, which are important in activating the immune system. Dendritic cells, which do not have receptors to distinguish self from non-self, may be the means by which danger in the form of violent, necrotic cell death is detected.

"It's not very different from what everybody has been saying," Matzinger admitted to the AAAS. But she added: "What's new is the idea that the way a cell dies matters."

Matzinger says her hypothesis can be used to explain events from the acceptance of pregnancy through to transplant rejection and tumours. And it has practical implications. "Babies don't look dangerous," she said, "because the cell death that goes on during foetal development is programmed cell death. It doesn't matter that this foreign body is in the womb. It does no damage. It doesn't look dangerous so the mother does not respond."

Matzinger says this may be why tumours are not suppressed by the immune system. They too, she says, don't look dangerous. All they are, to start with at least, are cells that have lost growth control. There is no unprogrammed cell death - and therefore there is no danger signal to the antigen presenting cells."

The very act of preparing a heart for transplant causes distress and triggers the dendritic cells into action, says Matzinger. She suggests that rather than giving transplant patients immuno-suppressant drugs, it would be better to find a way of blocking the signal from the dendritic cells, thus allowing the transplant to be tolerated.

Matzinger's ideas offer a radical challenge to conventional medical practice and to the science of immunology. If she is correct in her analysis, cures for some cancers could be around the corner, with more effective transplants on the horizon. And if she is wrong, at least she is at least asking some fundamental questions.