Lisa Kattenhorn, a Harvard Medical School graduate student in Ploegh's lab, discovered another cloaking mechanism. In order to be displayed at the cell surface by MHC, viral proteins must first go through a proteasome's grinders. But to get into a proteasome, the proteins need a special pass called ubiquitin; without this control, any and every protein could go into the grinder, and the cell would eventually die. Kattenhorn found that one of the first proteins to enter a cell during infection by a herpesvirus is an enzyme that can remove ubiquitin from viral proteins. No ubiquitin means no viral-protein fragments for MHC to display, so the infection is likely to be invisible to killer T cells. All herpesviruses have this enzyme, so Kattenhorn hopes her work might lead to a broadly applicable therapy.

Though Kattenhorn is a virologist, her work relies on probes made by chemists in Ploegh's lab. Howard Hang, a chemistry postdoc working with Ploegh, describes these probes as "bait for pulling out proteins" so they can be examined in detail. Each probe has the equivalent of a fishing fly that entices proteins to bite, as well as a "line" that can be used to retrieve them. Kattenhorn's probes, for example, use ubiquitin as bait to attract the enzymes that remove it.

These chemical probes work well in studies of the parts of the cell undermined by herpesviruses, but they cannot be used in live cells, Hang says. The probes are too big to enter and exit intact cells, so the cells must be pulverized before they're examined. Hang is designing a smaller, more flexible probe to do live-­imaging studies of Salmonella bacteria in action. In the cells it infects, Salmonella somehow fends off proteases, enzymes that, like proteasomes, break down proteins. It thus prevents its telltale proteins from reaching the cell surface and being seen by T cells. But Salmonella has to be intact and alive to pull off this feat.

Ploegh is using other kinds of live-cell imaging techniques to study the interconnections between the various branches of the immune system. He is also investigating immune-system cells that operate on a more general level than killer T cells do. Rather than responding to a particular strain of E. coli or to herpes simplex virus type 1, these cells recognize threats in very general categories--bacterium, virus, or fungus--and act fast. Textbooks make sharp distinctions between these two branches of the immune system, but "in real life they are intimately connected," says Ploegh. "They function on a continuous spectrum."

During an infection, a microbe tries to multiply, and its host tries to destroy it. "There you have the beginnings of a protracted battle," says Ploegh. In studying the war plans of herpesviruses and other microbes, Ploegh says, he's looking, not for a way to cure a specific disease, but for a better understanding of how the immune system works. And that understanding will better prepare us to combat any disease.