Stephen Leppla, Ph.D.

This is an archive page. The links are no longer being updated.

Testimony
Before the House Committee on Government Reform
Quickening the Pace of Research In Protecting Against Anthrax and Other Biological Terrorist Agents-A Look at Toxin Interference
Statement of
Stephen Leppla, Ph.D.
Senior Investigator, Oral Infection and Immunity Branch, Division of Intramural Research,
National Institute of Dental and Craniofacial Research, National Institutes of Health
For Release on Delivery
February 28, 2002

Mr. Chairman and Members of the Committee:

I appreciate the opportunity to appear before you today to describe my research regarding anthrax toxin receptors and the role of the protease furin in anthrax toxin action. Included in my remarks will be some discussion about the possible use of furin inhibitors to block anthrax toxin action, and the potential this holds for the treatment of anthrax infections.

The National Institute of Allergy and Infectious Diseases (NIAID) spearheads the bioterrorism research effort at the National Institutes of Heath (NIH) and supported the recent studies by Dr. John Collier, Harvard University, and Dr. John Young, University of Wisconsin Medical School, who are present today, which elucidated the mechanisms by which anthrax toxin destroys cells. The information gained through these studies will likely hasten the development of new drugs to treat anthrax.

NIAID also supports a number of other drug development efforts for anthrax and other Category A agents of bioterrorism. Earlier in the month, NIAID sponsored a Blue Ribbon Panel on Bioterrorism and its Implications for Biomedical Research, which brought together a distinguished panel of leaders from the biomedical research community and experts in Category A agents of bioterrorism to obtain expert advice and input on NIAID's Counter-Bioterrorism Research Agenda. This group helped NIAID assess its current research efforts to counter bioterrorism and identified goals for NIAID to implement on an immediate and intermediate/long-term basis. Indeed, one of the immediate goals identified for anthrax research is to encourage exploration of new targets for antimicrobial therapies, including strategies to prevent germination of spores, the synthesis or neutralization of toxins, and interference with attachment and entry of toxins into host target cells, which will build upon the findings of Drs. Young and Collier.

First I wish to briefly discuss our work on the cellular receptor for anthrax toxin. Bacterial toxins that attack animal cells must first bind to the surface of those cells. Several toxins do this by interacting with a single specific protein present on the cell surface. By definition, this protein is the toxin receptor. The receptor is typically a normal cellular protein that has a recognized role in cell function. It is only by accident that this protein is used by a bacterial toxin to enter the cell and damage it. I began studies intended to identify the anthrax toxin receptor some years ago while a researcher at the United States Army Medical Research Institute of Infectious Diseases (USAMRIID) in Frederick, MD, and continued this work after transferring to the National Institute of Dental and Craniofacial Research (NIDCR) in 1989. Our early work showed that all types of cells have anthrax toxin receptors, that cells typically have about 10,000 receptor molecules on their surface, and that the receptors were probably proteins. We generated cultured cell mutants lacking functional anthrax toxin receptors. Over a period of years, we worked intermittently to identify the receptor, using various biochemical and genetic methods, but our efforts were unsuccessful. Several years ago, a group at Harvard University began similar efforts, and they were successful in identifying the receptor last summer. This work, led by Drs. John Young and John Collier and supported by the NIAID, was published in the journal Nature several months ago. This important work showed that anthrax toxin uses as its receptor a protein named tumor endothelial marker 8, or TEM8. This cell surface protein had been described just one year ago in work from the laboratory of Dr. Ken Kinzler of Johns Hopkins University, work supported by the National Cancer Institute. The protein was identified as one that is highly expressed in tumor endothelial cells.

As Drs. Young and Collier pointed out in their publication, and as I mentioned earlier, their important discovery opens potential several avenues toward development of new therapies for anthrax infection. Specifically, they showed that a portion of the receptor, produced as a recombinant protein in the common bacterium Esherichia coli, was able to act as a receptor decoy and block the action of toxin in cultured cells. A precedent for receptor decoys being effective therapeutic agents is provided by the tumor necrosis factor soluble receptor, marketed as the product Enbrel, which is used in treating rheumatoid arthritis.

Now let me turn to describing work performed in my own laboratory on furin. Furin is a cellular protease which is required for the processing of many proteins that a cell secretes or delivers to its cell surface. Furin is a member of a family of similar enzymes that include those required for generating the final, active forms of peptide hormones such as insulin. It is an essential enzyme, as indicated by the fact that inactivation of the gene in mice causes death during an early stage of embryonic development.

When I began work on anthrax toxin, there were already several examples of bacterial toxins that require proteolytic activation. That is, the toxins had to be cut at a specific site by a protease enzyme to be made fully active. During our first efforts to purify the protective antigen protein of anthrax toxin, we noted that it was very easily cleaved by proteases at a single site. By sequence analysis of the fragments, we determined that the protein was cleaved following a sequence of four amino acids, arginine-lysine-lysine-arginine. We then showed that removal of this cleavage site inactivated the toxin. This was convincing proof that cleavage at this site is essential for anthrax toxin action. Because uncleaved toxin was fully active when added to cells, we suspected that the cells were causing proteolytic activation of the toxin. In effect, the toxin appeared to be using a cellular protease to achieve its own activation.

We then set out to identify the cellular protease required for anthrax toxin activation. We changed the amino acids within the arginine-lysine-lysine-arginine sequence, replacing each amino acid with many different ones. By comparing the toxicity of more than 30 mutated proteins, we found that any toxin protein having arginine at both the first and fourth positions was toxic to cells. The identity of the amino acids in the two middle positions occupied by lysine in the original sequence could be changed to other amino acids with little effect. At the time this work was being done, other researchers had finally identified the long-sought human proteases which process biosynthetic precursors of hormones such as insulin. It had been thought that these proteases recognize only paired basic amino acid residues such as the sequence arginine-arginine. However, the new evidence suggested that one of these proteases, named furin, cleaves proteins having arginines at the first and fourth positions. Because this sequence exactly matched the one we had identified, we speculated that furin was the cellular protease that was needed to activate anthrax toxin protective antigen. We then contacted Dr. Gary Thomas, of the Vollum Institute, University of Oregon. Dr. Thomas was already a recognized expert in study of these cellular proteases, and he agreed to collaborate in further studies. He quickly determined that purified furin rapidly cleaves the anthrax toxin protective antigen protein.

Subsequently, we generated mutated cultured cells lacking functional furin, and showed that these were highly resistant to anthrax toxin. Similar mutant cells had been produced some years earlier by Thomas Moehring, University of Vermont, but the genetic defect in the cells was not known. We went on to show that the furin-deficient cells are also resistant to several other bacterial proteins that require protease activation, and Dr. Moehring had already shown that such cells are resistant to certain viruses. In was later shown through work in other laboratories that furin is involved in the activation of many viral envelope proteins, including those of influenza virus and HIV. My lab has not been actively working on furin in the last several years, although we continue to aid others investigators in this field by supplying the furin-deficient cultured cells when requested. My original collaborator in the furin studies, Dr. Gary Thomas, has continued to work actively and productively in this field, and he can provide details about the current state of research on furin.

I now would like to offer some comments comparing possible therapeutic opportunities for anthrax infections. Researchers working on anthrax have identified at least eight distinct stages at which one theoretically could interfere with anthrax toxin action. Studies in cell culture models have demonstrated in principle that each of these stages can be blocked. Drs. Collier, Young, and Dr. Arthur Friedlander, USAMRIID, have provided much of the data proving that these separate stages each represent a target for therapeutic interventions.

In trying to find targets for therapeutic intervention in infectious diseases, most researchers focus on identifying target molecules that are unique to the pathogen. For example, one attractive target is the anthrax toxin lethal factor. Bacillus anthracis bacteria lacking lethal factor are greatly weakened in their ability to cause anthrax. Because of the success with which AIDS is treated by inhibitors of the HIV viral protease, many researchers believe that there is a great opportunity for treatment of anthrax with inhibitors of the lethal factor protease. Pharmaceutical companies and academic researchers have extensive experience in developing inhibitors of proteases, and some of that expertise is being redirected toward developing lethal factor inhibitors. NIAID has for several years been supporting at least two research groups studying lethal factor structure and inhibitor development. An important advance in this area occurred several months ago with the publication of the crystal structure of the lethal factor protease, work done in the laboratory of Robert Liddington, Burnham Institute, La Jolla, CA. Dr. Collier and I were collaborators in that work. The availability of the complete crystal structure of lethal factor has encouraged many researchers to begin new efforts or intensify existing efforts to develop lethal factor inhibitors. My lab is providing purified lethal factor protein to several of these groups so as to facilitate their work. I have considerable hope that a carefully selected lethal factor inhibitor will prove to be an effective therapeutic for anthrax.

The other protease involved in anthrax toxin action is the one discussed above, the cellular protease furin. Because of the important role furin has in normal physiological processes, NIH has supported many studies involving furin and the family of proteases which are closely related to it. For example, NIH has supported the work of Dr. Thomas over a number of years, during which time he developed the potent inhibitor which he calls the "Portland" inhibitor. Potent furin inhibitors have also been developed by two other NIH-funded researchers, Drs. Iris Lindberg, of Louisana State University, and Robert Fuller, of the University of Michigan. The inhibitors developed by these three NIH-funded researchers employ three different approaches to inhibitor design, and together identify a number of opportunities for development of even more potent furin inhibitors. It should be mentioned that NIH intramural researchers have also made important contributions in furin research. Drs. David FitzGerald and Ira Pastan of the National Cancer Institute proved that furin has an essential role in the action of Pseudomonas exotoxin. Dr. Juan Bonifacino of the National Institute of Child Health and Human Development has provided important knowledge about the movement of furin between various compartments within a cell. Several other NIH-funded studies include analysis of the properties and functions of furin as a part of larger studies of various disease processes. This portfolio of investigator-initiated extramural and intramural research is producing a strong knowledge base on which to base therapies for those diseases in which furin plays a role.

I mentioned earlier that drug developers prefer to target molecules that are unique to a pathogen. For this reason, furin has received less attention as a target for drug development. The expectation has been that inhibition of this enzyme, which plays an essential role in many normal processes, might cause significant physiological damage to normal tissue. Consistent with that prediction is the fact mentioned above, that genetic inactivation of furin causes death of mouse embryos. Nevertheless, I do believe that inhibition of furin should be examined as one possible therapy for anthrax. Given the renewed interest in anthrax, I anticipate that the furin inhibitors mentioned above, as well as others, will be evaluated for anthrax toxin inhibition in appropriate cell culture models in the near future.

That concludes my testimony. I would be happy to respond to any questions that you or Members of the Committee may have.

Last revised: March 7, 2002