The B. anthracis Picture Is Now Complete

After a three-year effort, scientists have determined the crystal structure of edema factor, a toxic protein secreted by Bacillus anthracis, the bacteria that causes anthrax.1 Edema factor (EF) works in concert with two other anthrax proteins—protective antigen (PA) and lethal factor (LF)—to kill its host cell. PA's crystal structure was reported in 19972 and LF's in November 2001.3 With this three-dimensional map now in hand, researchers are making headway into understanding how the proteins function and, more importantly, how they can be stopped. Knowledge of EF's crystal structure "is basically all you need" for developing a drug that targets it, says cell biologist Robert Liddington, from the Burnham Institute in La Jolla, Calif., who led the teams that solved the crystal structures of PA and LF. Drugs that target the toxic proteins would be a big boost for treatment against anthrax since antibiotics are not always an effective antidote. Even though most B. anthracis strains are susceptible to antibiotics, there is a certain point during the course of infection, after which so much toxic protein has accumulated, that no amount of antibiotic does any good. But if an arsenal of antitoxins existed that could be administered at this later stage of infection, such drugs could be used to "cure those unfortunate patients," according to Wei-Jen Tang, a professor at the University of Chicago's Ben-May Institute for Cancer Research and coauthor on the recent Nature paper.1 Tang says he suspects that antitoxins could have saved the lives recently lost from inhalation anthrax. In Search of Treatment Tang and his collaborators didn't necessarily set out to find such a treatment. "When we started the project, we thought that edema factor was likely to be similar to mammalian cyclases and that this similarity would ultimately teach us something about the mammalian proteins," says Tang's coauthor, Andrew Bohm of the Boston Biomedical Research Institute in Watertown, Mass. EF is one of several adenylyl cyclases, molecules that convert ATP into cyclic AMP (cAMP). Cyclic AMP exists in almost every organism outside of plants, says Tang, and it plays an important role in responding to outside stimuli. Cellular cAMP levels, for example, moderate the fight-or-flight response. And, like mammalian cyclases, EF is only active when it binds with calmodulin, a calcium-dependent catalyst that naturally occurs in the host cell. Even though researchers have been studying some 50 different calmodulin target proteins over the past decade, still "nobody had ever seen an active enzyme complexed with calmodulin," says Bohm. Using X-ray crystallography, researchers compared the crystal structure of EF in two different states (with and without the bound calmodulin) hoping to better understand the calmodulin-induced activation of EF. The results were surprising. "The EF-calmodulin interaction is very different from that of other calmodulin-binding proteins," says Bohm. Typically, calmodulin wraps around its protein target, but in this case the target wraps around calmodulin. In fact, says Tang, "the conformation of calmodulin and the way it interacts with EF is so different than how it interacts with other target proteins that we need to rethink calmodulin-protein interactions." The binding pocket on EF—a very small region of the protein where the ATP binds and is converted to cAMP—does not look anything like any known human protein. So, on one hand, the team learned nothing new about mammalian cyclases; on the other, an anti-anthrax drug design is now possible, says Bohm. The difference in structure makes it less likely that a drug targeting EF won't cross-react by binding to a human protein instead, he explains. Now, says Tang, researchers have to "find a chemical that will fit into the pocket that belongs to the bad guy." The Next Step To this end, both Bohm and Tang are pursuing structure-based EF drug-design projects. Tang is collaborating with Brian Shoichet at Northwestern University Medical School, in Evanston, Ill., in an effort similar to the high throughput docking project that Graham Richards' team at Oxford University are doing with PA.4 The goal of that project is to enlist the power of more than a million computers idling worldwide to screen a database of small molecules for potential anti-PA drugs. "Indeed, the edema factor is a tempting target for high throughput docking since its 3-D structure, the binding site, and the interactions with the ligand, are all known," says Richards' collaborator, Meir Glick. However, drug developers may not consider the search for an anti-EF drug a profitable venture, says Tang, primarily because anthrax is rare in the United States. But, argues Bohm, B. anthracis causes tens of thousands of naturally occurring cases of anthrax worldwide per year. Furthermore, EF is only one of three pathogenic bacteria that inflict havoc in the human body by secreting adenylyl cyclase ecotoxins. The other two are B. pertussis, the bacteria that causes whooping cough, and Pseudomonas aerupinosa, the bacteria that Tang says is responsible for 10% to 20% of all hospital-acquired infections. "It is likely that other pathogenic bacteria also carry this protein," says Tang, and it would be an easy next step to design drugs for use against them as well. Leslie Pray (lpray@nasw.org) is a freelance science writer in Leverett, Mass. References 1. C.L. Drum et al., "Structural basis for the activation of anthrax adenylyl cyclase exotoxin by calmodulin," Nature, 415:396-402, Jan. 24, 2002. 2. C. Petosa et al., "Crystal structure of the anthrax toxin protective antigen," Nature, 385:833-8, 1997. 3. A.D. Pannifer et al., "Crystal structure of the anthrax lethal factor," Nature, 414:229-33, Nov. 8, 2001. 4. L. Hand, "Aiming a world of computers at anthrax," The Scientist, 16:17, Feb. 4, 2002.

The B. anthracis Picture Is Now Complete

Interested in reading more?

Become a Member of