Opinions differ on Iraq, but the immune system is a definite example of a successful surge strategy. The body harbors an enormous range of T cells that are responsive to any possible foreign invader.

But because there also is an enormous range of possible foreign invaders, just a few copies of each T cell are normally present. Only when an infection actually is detected do those cells start dividing and going after the bacterial or viral invaders; at that point, the formerly "naïve" T cells become "activated" T cells.

But the immune response is not equal against all parts of a foreign pathogen - many more T cells go after some parts, or epitopes, of a pathogen than others. Scientists have long suspected that the vigor of the immune response depends on how many naïve T cells the body has available to fight a given peptide in the first place. It is, in fact, a hunch that was confirmed for B cells more than 30 years ago.

Killer T cells, however, have been harder to pin down. The problem is that they are hard to isolate from the mass of other immune cells. The lymph organs of mice typically contain roughly 100,000 naïve killer T cells - but hidden in a sea of 200 million cells total, or enough to overwhelm even the fastest cell-sorting equipment, which is limited to sample sizes of roughly a million cells.

In the August 2007 issue of Immunity, though, researchers from the University of Minnesota and the University of Colorado Health Sciences Center reported a way to chip away at the haystack in a way that brings the total cell numbers down to a manageable number. They do it by enriching the sample with only those cells that bind to proteins of the major histocompatibility complex.

Senior author Mark Jenkins, professor of microbiology at the University of Minnesota, and his team first sorted cells according to their ability to bind to the major histocompatibility complex.

Only T cells "recognize peptide-bound MHC on the surface of another cell," Jenkins told BioWorld Today, and so pulling those cells that bind to major histocompatibility complex proteins allowed the researchers to enrich the raw lymph samples and bring the total cells in the sample down to a number that cell sorters can handle.

The authors first tested the approach by spiking a sample of other cells with known amounts of specific killer T cells, and testing how many target cells had to be present for the equipment to detect them. Enrichment improved the limit of detection by two orders of magnitude; while at least 1,000 cells had to be present in an unenriched sample, the flow cytometer was able to detect as few as 10 cells in an enriched sample.

Jenkins is modest about his achievement, calling the technological advance "basically simple." But several reviewers on the website Faculty of 1,000, which in part provides an additional layer of peer review after a paper has been published, were more effusive. One said that the method "should aid significantly in our understanding of T-cell receptor repertoire selection, self-tolerance, immunodominance, autoimmunity and pathogen-specific responses."

The Immunity paper provides a few samples of such enhanced understanding. The researchers used their strategy to isolate naïve killer T cells from the lymphoid organs, and counted how many cells were specific to each of three specific peptides.

They found that the number of naïve T cells varied tenfold, from 20 to 200. And Jenkins said that the starting number, "not surprisingly, influences how strong the immune response is."

When the team immunized mice with the same peptides and tested how many activated T cells the mice had four days later, the number of killer T cells directed against each epitope were roughly proportional to the number of naïve T cells in unimmunized mice.

Jenkins gave several reasons for why the number of naïve T cells directed against a given peptide varies. Major histocompatibility complex genes vary between individuals, which is one contributing factor. But also, "T cells, as they develop in the thymus, go through both positive and negative selection," he said.

The purpose of such selection is to preserve a wide-ranging immune response without attacking the self. "Certain foreign molecules are more like self," he said, and the immune response to those foreign molecules is more likely to be deleted to avoid autoimmunity.

Asked where he plans to take his research next, Jenkins was succinct: "Humans."

He elaborated that his team plans to study in more detail how varied the immune response is between individuals. The method theoretically could be used to determine how well an individual is likely to respond to a vaccine, though Jenkins said it is still "too laborious" to be used in a mass setting. But one place where it could find a practical application in the near term is in organ transplantation: By testing a recipient's reservoir of naïve T cells to the most likely culprits in organ rejection, doctors might be able to predict the likelihood of organ rejection for an individual donor-recipient team.