The Production of Antibodies

An evening in early August 1954 Niels Kaj Jerne had an idea while walking home from work and crossing Copenhagen's beautiful Knippelsbro—a bridge connecting the island of Amager where his laboratory was located at the vaccine-producing national serum institute SSI, and central Copenhagen. The flash of insight was completed even before he had crossed the bridge and so was the idea that would for ever change our understanding of the immune system—and perhaps one day also perception and cognition.

Niels Kaj Jerne had for a long time been worried about a single piece of evidence in the study of antibodies. Our immune system somehow produces these elegant structures that can bind to and escort out invading foreign substances like viruses or bacteria. The shape of the antibody fits the three-dimensional structure of the invader like a hand in a glove or a key in a lock.

But how was that possible: There is not nearly enough information in the genetic system to specify an antibody for each of the many million foreign entities that could enter the body. A favorite explanation at the time was that the antibody was informed by the alien substance: It took its shape from what was entering the body, so that it could bind to it and usher it out.    However, there was a problem that had been known since the 1930'ies: Small amounts of an antibody is present in the blood stream of animals—even though they have never been exposed to the foreign substance, the antigen, that the antibody is perfectly shaped to bind to. So the antibody could not have learned from the antigen.

The summer evening walk home from work gave Jerne an idea that he immediately saw as "fabulous": The body holds a collection of building blocks for antibodies. Whenever something new enters the body, the building blocks are shuffled into a vast array of antibodies. Each will be produced in one copy. One of them will bind particularly well to the shape of the invader. When the "re-cognition" has taken place, that particular antibody will be mass-produced by the immune system.

This immediately explains why it takes time to fight an infection and why vaccination helps (the shuffling and selection has already been done before the invaders arrive in large numbers).

Niels Kaj Jerne told biochemist Günther Stent about the idea. It took Stent "fifteen, maybe twenty minutes" to become convinced that the idea was right. It took thirty years before Jerne was awarded the Nobel Prize.

In a sense it is Darwinism applied to the immune system. A random production of possible shapes is subject to a selection pressure from an outside universe of shapes. When they fit, the shape from inside is multiplied in large numbers. In Darwinism we are talking organisms and the environment, in Jerne's model we are talking antibodies and the environment.

The elegance of the idea is that there is no transfer of information from the environment into the "design" of the antibody/organism. The environment is only selecting between options produced at random.

Is this not like perception: We actually do not see the world around us, but only our own "simulated visual scenery". When our fantasies are "relevant" to our behavior they are selected for and reproduced.

Is this not like cognition and thinking: We make up lots of "lego-brick" models of the empirical data, reject most of them, but eventually when something fits, we keep thinking that way?

Is Niels Kaj Jerne's idea not a prime example of itself: After playing around with numerous small toy models in his mind, discarding all of them as hopeless, suddenly the right one popped up in his mind that August evening crossing Knippelsbro?

He then kept thinking that way. Reproducing the idea again and again.

As we do, right now.