Memory T cells, because of the sheer speed with which they can respond to an infection, are "the best thing you can have" to protect against a disease, Mark Davis told BioWorld Today. But memory appears to be a misnomer: Davis and his team have shown that every one of more than 20 individuals had memory T cells to viruses they had never been exposed to.

Davis said that the findings came as a surprise to his team, which identified the cells in the course of studying regulatory T cells. "We believed, like everybody else, that you only had these things if you got a disease, or a vaccination," he said.

But when Davis, who is at Stanford University, and his team looked at T cells from 26 blood donors, they found memory T cells that reacted to HIV, herpes simplex virus and cytomegalovirus in every one of them, despite the fact that those individuals had no antibodies to any of the viruses.

Davis said that the absence of HIV antibodies in particular was strong evidence that the memory T cells his team found are not, in fact, due to an infection that was somehow cleared without eliciting an antibody response. The U.S. blood supply has been monitored for HIV via antibody testing since 1985, and since that time, HIV transmission via donated blood has plummeted – the Centers for Disease Control and Prevention now estimates that the risk of contracting HIV via a blood transfusion is less than 1 in 1 million. The notion that each of two dozen individuals that the team tested would have been infected with HIV without having an antibody response is all but mathematically impossible.

Davis and his team also showed that when they vaccinated two individuals with flu vaccine, that vaccination, too, led to cross-reactive memory T cells that recognized not only the flu, but also bacterial and protozoan epitopes.

In the cord blood of newborns, on the other hand, they could identify no such memory T cells – a fact that might be part of the explanation for why babies are so vulnerable to infections.

Davis cautioned that "we have not shown that the T cells will save you from a particular disease, or even lead to a better outcome," although he "strongly suspects" that this is the case. Testing whether artificial memory T cells protect their owners is on his lab's to-do list.

Additionally, the team wants to delve into the details of when memory T cells start developing. The work could explain why measles vaccinations have been shown to lower a vaccinated child's risk of dying not just from measles, but from all infectious diseases.

Finally, Davis said, his team's work is another sign that immune system research done in mice needs to be taken with a grain – or perhaps a whole spoonful – of salt.

Between the age of most mice at the time they are used in research, and the fact that they live in facilities where germs are absent as far as possible, "they have an immune system that is more like that of a newborn" than a battle-scarred and much better armed adult.

Earlier this week, another paper showed that mice have a very different inflammatory response to everything from burns to infection. (See BioWorld Today, Feb. 12, 2013.)

"Their paper points to an evolutionary difference," between the mouse and human immune systems, Davis said. "This paper points to a difference in the environment."

But both papers suggested that mouse data may not be terribly predictive of human success – in either direction.

Failures in clinical trials after success in mice is, of course, a well-known risk of drug discovery. But Davis pointed out that the opposite may also be true, particularly if the immune system of the average laboratory mouse is weaker than that of the average adult human: Treatments that don't work in mice might still be successful in humans.