Long-term immunity isn't always beneficial.

Come down with a case of chicken pox and, after you recover, your body seems to wear an invisible suit of armor that protects you from getting the disease again. Catch a cold, on the other hand, and the protective armor seems to fall away quickly.

Common sense indicates that the longer your immune memory lasts, the healthier you will be. Now, a mathematical model indicates that there may be a good reason that you quickly lose your protection against the sniffles. The endless succession of colds that results may protect you from far nastier bugs.

When a person becomes infected by most pathogens, the immune system instantly goes on the attack. After the infection is vanquished, the immune response subsides, but not all the way down to its original level. Long-lived sentries called memory cells remain ready to pounce if the bug reappears.

Dominik Wodarz of the Fred Hutchinson Cancer Research Center in Seattle has considered a scenario in which two different pathogens--call them A and B--threaten a host population. He assumed that pathogen A is far more deadly to its host than pathogen B is. Pathogen B, he also assumed, is fitter than pathogen A, meaning that if the two pathogens in the host have to compete for resources on an even playing field, pathogen B will win out.

If an individual becomes infected with pathogen B, the immune system will create memory cells against B, but that will tilt the playing field in A's favor. This makes the host vulnerable to the more virulent A. Thus, long-lasting memory of pathogen B can actually work against the health of the host.

Wodarz carried out calculations showing that, in this scenario, the host population will indeed evolve toward a short immunological memory of pathogen B infections. He reports his findings in the Sept. 16 Current Biology.

"Wodarz' paper suggests that the naive assumption--that the longer memory lasts, the better--may be wrong," says Charles Bangham, an immunologist at Imperial College in London.

Over the long term, Wodarz says, a host population would probably cycle between long and short immune memory. In the above scenario, for instance, once the host population evolved to have a short memory of pathogen B, pathogen A might become extinct. That would eliminate the competition and permit the host's memory of pathogen B to gradually lengthen. Eventually, the door would open for another pathogen like A to spring up, and the cycle would begin again.

The findings could have important public health implications, Wodarz says, since vaccines are essentially humanmade immune-memory boosters against diseases. "If you vaccinate a population, it may backfire and allow the invasion of pathogens that are more virulent," he says.

However, he notes, it would be hard to test this hypothesis in people, since epidemiological studies of immune memory are difficult and slow. "It is hard even to measure the duration of memory in a controlled way," Wodarz says. "You have to identify when a person is infected, then draw blood every year."

For now, mathematical modeling may be the best way to gain insight into the duration of immune memory, says Derek Smith, a computational biologist at the University of Cambridge in England. For instance, he says, it would be interesting to figure out time scales on which a population would cycle between short and long memory.

The current study is a good start, Smith says. "It's a very sound and well-worked-out analysis."