Report:

Danger: immunologists battle over self recognition

Investigators: Polly Matzinger and Melvin Cohn

Tuesday Jul 24th, 2001

by Melissa Mertl

Two competing theories for how to view the immune system came head to head tonight as two outspoken theoreticians exchanged friendly fire. Is the immune system more concerned with damage than with foreignness? In the process of trying to answer the question, theoretical biology itself stood trial.

Polly Matzinger, head of the T-cell Tolerance and Memory Section at the National Institute of Allergy and Infectious Disease, Bethesda, Maryland, promoted her updated version of the "Danger Model," which she first published seven years ago as an alternative model to the classical paradigm of immunity that asks how the body distinguishes between self and non-self.

Melvin Cohn, who had been billed as defending the hypothesis that "activation of T-cell immunity requires specific recognition of non-self," said he was trying to get away from using the terms "self" and "non-self" because of philosophical debates over what they actually mean. Instead, he chose to promote the mouthful of a phrase: "not-to-be ridden" (i.e. not to be got rid of) versus "to-be-ridden."

The terms are meant to be synonyms for self and non-self, and form the two categories into which the immune system initially sorts events. If the decision is made that an antigen is "to-be-ridden" (e.g. if it is non-self), then the decision must be made as to which effector mechanism is optimal to get rid of it: an intracellular or an extracellular response, which in turn results in several other possible pathways.

Cohn, principal investigator in the Conceptual Immunology Group at the Salk Institute in San Diego, California, discussed his group's model of the sorting processes and invited others to play with their web-accessible simulation tool to test the parameters of the model.

"Mel and I have been debating since 1975," said Matzinger. "Mel has taught me several things." One of those things was that one should always "dig up hidden assumptions and see what they predict." The assumption made in Cohn's model is that the first question to ask is how does the body sort self and non-self, she said. But the question she wanted to ask was "how does the immune system decide to respond?"

The important factor is not whether the body recognizes something as foreign or not, she said, but whether the tissue receives cell-death signals that are benign or that indicate danger.

And an answer isn't sufficient unless it covers a large portion of immunological phenomenon, such as autoimmune disease, transplant rejection, tumor non-rejection, and fetal tolerance, she added.

"Mel says that what is 'self' is prior ... when women start to lactate they make proteins they did not have before." The self/non-self model doesn't account for why the body doesn't reject a woman's own lactating breasts, she said, adding that the very concept of self changes over a lifespan. "We are not the same today as we were in utero." She did admit that the Danger Model cannot adequately address the mechanism of allergies. That problem is governed by the second level question: "When you've decided what you've got to respond to, how does the body control the class of response?" she said. "There is currently no viable theory anywhere that covers that."

"But none of this will be proved by computer modeling." She went on to give examples of studies that she said supported the idea that being foreign is not enough to trigger an immune response.

"Right now, we may not have enough data to decide who is right. Self/non-self is an old theory. Perhaps if we switch that original question to how the cell dies rather than foreign or not, we might begin to explain the next 50 years worth of questions."

With her hair dyed a surprising shade of apple green, Matzinger (equally renowned as a sheep-dog trainer and former bunny girl) struck a combative stance in preparation for Cohn's response.

"I have found that a theory that explains everything is usually wrong," said Cohn.

One can't just take a case such as mother's not rejecting their fetuses, and simply interpret this as evidence for a theory, he added. There are many mechanisms that explain why fetuses have protected immunity, and one could analyze the situation under both theories and not find much difference.

Nicholas Restifo of the National Cancer Institute stood up to object to the entire basis of the discussion. "These two quite intelligent, possibly brilliant, biologists are pretending to be physicists," he said. "They make the mistake that the immune system is governed by overarching principles."

Unlike physics, where "one can invent theories that explain everything while sitting in the bathtub," said Restifo "[biology] is messy."

Cohn and Matzinger finally found common ground by disagreeing with Restifo's objection.

In conclusion, debate chair Moller said that Matzinger and Cohn represent a dying breed of biologists. Hypothesis-driven research is being replaced by so-called discovery driven research. "I for one am glad we still have Polly and Melvin," he concluded.