A biological weapon of an entirely different sort is what Harvard Medical School researchers are trying to build from the potent toxin of Bacillus anthracis. Exploiting the toxin's ability to transport molecules into cells, the researchers have used it to develop an experimental vaccine directed against a model pathogen. What's more, the vaccine was found to protect mice against infection by that pathogen, the researchers report in the October 29 Proceedings of the National Academy of Sciences.

Though still in early stages, the vaccine may lead to an entirely new class of human vaccines against most viruses, certain bacteria, and parasites. Moreover, the approach may be useful in developing cancer vaccines and therapies, says senior author Michael Starnbach, assistant professor of microbiology and molecular genetics. "This study provides the proof of principle that this strategy works," he says.

It also represents the first successful attempt to engineer a protein-based vaccine that works by priming the immune system's killer T cells to respond against infection and to generate a specific immunological memory for future protection. Most current protein-based vaccines, such as the one commonly used against tetanus, stimulate B cells, which then churn out antibodies. The trouble is, B cells can detect invading pathogens only as long as they are outside of cells. Once the pathogen has snuck past this line of defense and slipped inside cells of the body Harnessing the killer T cells' power for vaccination has been difficult, says Starnbach, because they require that the antigens against which they act be displayed to them from inside infected cells. And delivering a vaccine into cells is much more complex than simply injecting it into a person's bloodstream.

A current approach to solving this problem, using so-called naked DNA, harbors the danger of introducing foreign genetic material that could possibly insert itself into the human's own DNA. "The safety of protein-based vaccines is one of their main attractions," says Starnbach.

To engineer an intracellular vaccine, Starnbach collaborated with John Collier, professor of microbiology and molecular genetics at Harvard Medical School, who had, for years, studied the way in which the anthrax toxin managed to do exactly what Starnbach needed: ferry proteins across the cell membrane and into the cytoplasm. Collier had already developed a technique to manipulate some of the toxin's components so they became innocuous but could in theory transport any protein Armed with this technology, the researchers genetically fused a harmless snippet of the anthrax toxin to a snippet of their model pathogen required to stimulate T cells but unable to cause full-blown disease. Then they mixed this construct with the transporter component of the anthrax toxin and injected it into mice. When they infected vaccinated and unvaccinated animals with the model pathogen Much work remains to be done, however, before the researchers can claim victory. For starters, they need to test if their vaccine can protect against death, not just reduce bacterial load, and whether it can do so in diverse strains of mice, not only the inbred strain that was studied. These are necessary steps toward testing a vaccine in humans.

Next they need to apply their method to more medically important pathogens than the bacteria used in this study. First candidates could be the cytomegalovirus that causes retinitis in people with AIDS, and the bacterium that causes dysentery, says Starnbach.

It may seem ironic to us that we should allow any part of Bacillus anthracis into our bodies to protect us from disease. But it is also oddly fitting that this organism opens the door to solving a knotty problem in vaccine development, because it was one of the first pathogens against which a vaccine was successfully made The author of the PNAS article, in addition to Collier and Starnbach, is Jimmy D. Ballard, a postdoctoral fellow in the Department of Microbiology and Molecular Genetics at Harvard Medical School.