By David N. Leff

Escherichia coli is the prime pathogen of urinary tract infection. To infect its victim, that microbial bacterium must stick to the cells of its target tract wall ¿ from urethra to ureter, to bladder to kidney. Where does it acquire that stickiness?

Recent research suggests that a specific wrinkle in the germ¿s 3-dimensional protein structure ¿ namely, the beta-helical fold ¿ enables it to recognize and adhere to the long, extracellular polysaccharides of a particular cell type. In microorganisms that cause human diseases, a set of proteins on their surface allows the pathogen to bind to or penetrate human cells. These proteins include adhesins, toxins and virulence factors.

¿The first protein to be discovered with this beta-helix fold,¿ recalled molecular biologist Jonathan King, at Massachusetts Institute of Technology in Cambridge, Mass., ¿was the enzyme pectate lyase. It¿s a protein of a bacterium, Erwinia chrysanthemi, that infects plants, and turns fruits and vegetables to mush. But pectate lyase, as its name suggests, was historically employed in the making of jelly. The protein we work on,¿ he continued, ¿is from a bacteriophage virus. It¿s also the mechanism by which the virus recognizes and infects Salmonella bacteria. Phages infect bacteria, but they¿re not themselves pathogenic. They sit in the world of models that scientists use to study medically important organisms. One thing the helical fold turns up in is proteins on Bordetella, the whooping-cough germ, and various other bacteria that are important in human infections.¿

King is a co-author of a paper in today¿s Proceedings of the National Academy of Sciences (PNAS), dated Dec. 18, 2001. Its title: ¿BETAWRAP: Successful prediction of parallel b-helices from primary sequences reveals an association with many microbial pathogens.¿ Its senior author is computational biologist Bonnie Berger.

¿Our first of two findings,¿ King told BioWorld Today, ¿was success in predicting ¿ apparently with very high accuracy ¿ this particular fold, the parallel beta coil, from protein sequences. This Holy Grail in genomics and proteomics is to be able to predict the 3-dimensional structure of the linear sequences. It is what holds back all the information in the genomics project because you need the 3-D fold, not the 1-D sequence. It¿s been very difficult to do this, so it¿s an important first step.¿

Bacteria Make Surprise Appearance

¿The second significant ¿ and unexpected ¿ result,¿ he continued, ¿was the association of the parallel beta coil, the beta-helix fold, with surface proteins of diverse human pathogens. We think we understand why, because this fold appears to be specialized to binding along one of its sides ¿ the long, floppy polysaccharides that sit on the surfaces of cells, and don¿t occur very much within cells. So the first step pathogens have to take is to recognize the surface of the host cell and penetrate it. It looks as if a whole group of them use this particular BetaWrap fold to recognize these surface sugars and related floppy molecules.

¿I work on protein folding,¿ King observed, ¿and my colleagues are computational biologists who are trying to understand this general relationship between linear amino acid sequences of 3-D folds, the major unsolved problem in modern biology. This class of proteins, these parallel beta helices, is topologically simple. It¿s just like a coiled spring. If you call up the image of a coiled spring, here you have an accurate picture of how these protein chains are folded.

¿We thought maybe we could make progress a little simpler than the major class of beta-sheet folds,¿ King recounted. ¿Phil Bradley, the PNAS paper¿s lead author, had developed the algorithm that found all the known structures correctly. Then he ran it on the half-million sequences in GenBank ¿ where you don¿t know the structure, you just know the linear sequence ¿ and it was a surprise when out popped Bordetella in this area, but hardly any human proteins, hardly any rat proteins, hardly any fish proteins, hardly any fly proteins ¿ a very unusual clustering.¿

All told, the PNAS article charted 41 of the top 200 BetaWrap-scoring pathogenic proteins ¿ from Bordetella to Vibrio cholerae.

¿A striking feature of the sequences identified,¿ PNAS reported, ¿is their association with known human pathogens.¿ It spelled out 18: V. cholerae (cholera); Helicobacter pylori (ulcers); Plasmodium falciparum (malaria); Chlamidia trachomatis (venereal infection); C. pneumoniae (respiratory infection); Listeria monocytogenes (listeriosis); C. abortus (genital infections); T. brucei (sleeping sickness); B. burgdorferi (Lyme disease); L. donovani (Leishmaniasis); B. bronchiseptica (respiratory infection); R. rickettsii (Rocky Mountain spotted fever); T. cruzi (sleeping sickness); B. parapertussis (whooping cough); Bacillus anthracis (anthrax); R. Japonica (Oriental spotted fever); Neisseria meningitides (meningitis); and L. pneumophilia (Legionaire¿s disease).

Leg Up For Antibacterial Drug Discovery

¿This has the potential of being an early warning,¿ King predicted. ¿On the one hand, there is this continual problem of trying to recognize new and novel pathogens before they have actually emerged. The beta-coil class of protein folds is associated with pathogenesis, and if you have an organism that¿s not well characterized, and you find a sequence like this in the genome, it suggests paying a little closer attention to this gene, in terms of antibiotic development.

¿Most proteins that bind to substrate have a structural cleft,¿ he pointed out, ¿and when you¿re making an inhibitor to the enzyme that blocks its action ¿ like penicillinase breaks down penicillin ¿ you try to design some chemical that¿s going to sit there in the cleft.

¿These BetaWrap proteins are very different. They don¿t bind with a cleft. They have a long, lateral surface, and they use its long side to grab this long, floppy molecule,¿ King said. ¿That leads to a very different strategy in terms of looking for inhibitors. So it is a new class of targets for potential antibacterial strategies.

¿This is an arena that people now call proteomics,¿ King noted. ¿It¿s an example of the rapid developments in that field, opening up new opportunities in terms of antibacterial pharmaceutical development. There¿s been a lot of hype about that, and a lot of activity, but there aren¿t that many new products yet. It will be a few more years before these tools actually deliver. But,¿ he concluded, ¿you can see them opening up these windows now.¿