Biologists have devoted many years to studying how the anthrax toxin invades a cell and does its killing work. In the past two decades, researchers have discovered how the toxin forms a pore that provides an entryway for two other toxin proteins to get into a cell's inner recesses and wreak havoc.

The toxin's breaking-and-entering strategy has not gone unnoticed by scientists seeking new ways to deliver vaccines into cells. And the lethality of one of the proteins has attracted the attention of investigators who have shown an interest in marshalling its killing power to combat tumors.

The idea of using a notorious pathogen as a therapeutic tool did not immediately generate an enthusiastic audience. Yichen Lu, a researcher at the Harvard University School of Public Health, remembers telling the director of the vaccine branch of the U.S. Food and Drug Administration in 1999 of his plans to use anthrax in an HIV vaccine. He described how a version of the toxin could be genetically engineered to eliminate its pathogenicity and how it could deliver a protein from the AIDS virus into an immune cell called an antigen-presenting cell. Once processed there, pieces of the protein would be displayed on the cell surface. That would induce production of killer T cells by the immune system that would then attack and kill the virus throughout the body. Lu remembers clearly the response of the FDA official when he suggested how important it was to move ahead with a safety test of the anthrax-derived vaccine on army soldiers: "She said, 'You're crazy.'"

As early as this fall, however, the U.S. Army plans to initiate a clinical trial to test the safety of just such a vaccine on a few dozen civilian volunteers from the Washington, D.C., metropolitan area. At about the same time, a safety test of an HIV vaccine using an anthrax delivery system might be launched in Botswana as part of a research program the Harvard School of Public Health maintains there. Despite its association with bioterror, such a vaccine is expected to have a good safety profile. It does not use an altered version of a live virus, which might revert to its pathogenic form, and it should also be relatively easy to produce in large quantities.

Still, the desultory results for AIDS vaccines so far dictates caution [see "Hope in a Vial," by Carol Ezzell; Scientific American, June 2002]. John G. McNeil, deputy director of the division of retrovirology at the Walter Reed Army Institute of Research, notes that the vaccine developed by Lu and his associates has produced promising results in mice and rabbits but adds that it is still too early to make any pronouncements about its effectiveness in humans. "We don't have a clue what the immune effects will be," he says.

Interest in anthrax as an immunodelivery truck extends beyond HIV. Before taking a research position at Harvard in 1999, Lu had headed development of anthrax research at Avant Immunotherapeutics, a biotechnology company in Needham, Mass. There he studied using the toxin to transport antigens--the active part of a vaccine that generates an immune response--for not only HIV but also hepatitis B. If the initial army-sponsored trials prove a success, Lu plans to use the modified anthrax toxin in vaccines for hepatitis B and Epstein Barr virus at Nan Kai University in China, where he holds a position as head of the university's vaccine laboratory.

Penetrating Cells

Michael N. Starnbach of Harvard Medical School has several graduate students working on vaccines that marshal the anthrax toxin's cell-penetrating abilities. In particular, his laboratory has studied vaccines for the pathogenic bacteria Listeria monocytogenes and for Chlamydia trachomatis. Many of these investigations have centered at Harvard, taking advantage of basic research on the biology of the anthrax toxin carried out there by biologist R. John Collier. [See "Attacking Anthrax," by John A. Young and R. John Collier; Scientific American, March 2002.] A collaboration between Starnbach and Collier produced, in 1996, the first published study on using anthrax as a delivery system for a vaccine.

Anthrax therapeutics has not been confined to the role of messenger. The inherent lethality of the toxin has intrigued researchers, who, in the laboratory, have begun to use it to target cancer cells. A report in the March 5, 2002, Proceedings of the National Academy of Sciences by scientists at the Van Andel Research Institute in Grand Rapids, Mich., and the National Institutes of Health noted that an engineered version of the toxin caused partial or complete regression of melanoma in mice grafted with the tumors. "I'm quite hopeful," says Stephen H. Leppla of the NIH, one of the researchers on the study and a pioneer in the basic biology of the anthrax toxin. "I think it shows in studies done so far very good efficacy against tumors and low toxicity."

The National Cancer Institute discovered the chemotherapeutic potential of anthrax when a component of the toxin, a protein called lethal factor, was screened for its likeness to an already known compound that inhibits tumor growth. It was discovered that lethal factor blocks the activity of an enzyme, mitogen-activated protein kinase kinase (MAPKK), that triggers another enzyme used by cancer cells for replication and the spreading of tumors. Hindering MAPKK causes a cell to die prematurely. Leppla and his colleagues found that dose levels of lethal factor injected in mice grafted with human melanomas were high enough to cause the cancer cells to die but spared normal cells.

Targeting Cancer Cells

The researchers, however, want to offer further protection to ensure that lethal factor targets only cancer cells. They have engineered another anthrax toxin protein, called protective antigen, so that it reacts with an enzyme, urokinase plasminogen activator (uPA), which is more abundant on the surface of cancer cells than on normal ones. Once cut by the enzyme, protective antigen assists lethal factor to get into the cytoplasm of tumor cells, inducing the cells' death. The engineered toxin cannot penetrate cells that do not generate uPA. Commercial development of anthrax chemotherapeutics may be in the offing. Several pharmaceutical companies have shown interest in licensing the technology, says Nicholas S. Duesbery, an investigator at Van Andel.

Scientists have also hypothesized that lethal factor's inhibition of MAPKK may also be the mechanism by which the anthrax toxin exerts its deadly effect as a pathological agent. Further research is needed to confirm this conjecture, but at least one pharmaceutical company is interested in using a test developed by Leppla and Duesbery to determine whether lethal factor is active in a cell by measuring its ability to block MAPKK. This might enable screening for therapeutic compounds that inhibit lethal factor and thus serve as an antitoxin for anthrax. The next few years should show whether decades of basic research on the toxin will help treat those infected with anthrax and whether, in the case of vaccines and chemotherapies, a killer can be turned into a healer.

Gary Stix is special projects editor at Scientific American. He conducted some of the reporting for this article during a weeklong CASE fellowship at Harvard.