A good half of all the deaths from disease in the industrialized worldcan be laid at the door of a single protein. Yet without tissue factor,mammalian life couldn't exist in the first place.Tissue factor (TF), a.k.a. plasma thromboplastin or coagulation factorIII, summons the circulation's clotting guards to stanch loss or leakageof blood from a wound. But like guards stationed to protect strategicbuildings, who recklessly or treacherously shoot up innocentbystanders, TF can be guilty of such pathogenic overkill asatherosclerosis, cancer, infection and septic shock.Drug developers seeking new pathways to find or fabricate compoundsthat could curb the death-dealing excesses of tissue factor got a newroad-map today. Nature (Aug. 25) carries an article titled "Crystalstructure of the extracellular region of human tissue factor."As one of its two principal co-authors, coagulation specialist EdwardG. D. Tuddenham, told BioWorld Today yesterday, "TF is the hotbutton that triggers the blood's clotting cascade. Now that ourcrystallographers have solved its molecular structure, it provides thestructural design for pharmaceutical drug development."Tuddenham, professor of hemostasis at the University of London'sRoyal Postgraduate Medical School, has been working with South SanFrancisco-based Genentech Inc. since 1982 to develop anti-coagulantdrugs. In 1984, he cloned and expressed factor VIII, now arecombinant remedy for hemophilia."In current opinion," he explained, "the body clearly and beautifullyplaces TF around the outside of blood vessels and various organs.When a vessel leaks, it contacts TF, which grabs factor VII by a`finger' we have identified, and sticks to it with one of the tightestbonds known, forces it to activate, and together they initiate theclotting cascade."Thomas Edgington, who heads the vascular biology research unit atScripps Research Institute, San Diego, collaborates with the U.K.group. by electronic mail. "TF is not normally expressed in thevasculature," he told BioWorld Today. "When the TF gene is induced,then transcribed, it comes to the cell surface. Nature wisely separated itfrom the requisite serine protease co-factor VII. Otherwise, we'd all bea solid clot immediately. One molecule of TF can generate 250,000molecules of thrombin per minute."In the absence of a blood-letting injury, Tuddenham continued, TF canstill enter the bloodstream, rafted in by macrophages. There it lands onan atheroma, an early stage of atherosclerosis. When a break occurs onthe atheroma's cover, TF goes to work, and initiates coronarythrombosis.Macrophages also release TF in inflammation, causing extra-vascularcoagulation. Cancer cells express the protein too, and so do gram-negative bacteria, with septic shock as their end-point.Human tissue factor is a cell-surface glycoprotein receptor. Of its 263amino acids, the first 219 hang out beyond the cell membrane. Thecrystal structure of this soluble protein domain, said the Nature article,had been solved to a resolution of 2.2A. "This is the first reportedstructure," it stated, "of a representative of the class 2 cytokine receptorfamily, which also includes interferon-a, interferon-g and interleukin-10 receptors."The British group's crystal-charting endeavor began, Tuddenhamrecalled, some three years ago, when Genentech expressed TF in E.coli, and sent him the cell paste. From this recombinant startingmaterial, he said, "we purified it to 99.999 percent, and grew largecrystals of the protein."They collected diffraction data sets at the Synchroton in Daresbury,U.K. and the Photon Factory in Japan's Tsukuba Science City, thenperformed heavy-metal soaking at Oxford University's Centre forMolecular Sciences. The entire crystallographic undertaking wasdirected by principal co-author William Boys of Edinburgh UniversityMedical School."The solved TF structure," Tuddenham said, "maps very preciselywhere the protein's binding sites to factor VII are located. Themolecule's atomic structure will be deposited in the Protein Data Bankat Brookhaven National Laboratory, Upton, N.Y. "Anyone concernedwith blood coagulation and thrombosis, he added, will find that datauseful."Tuddenham's collaborator at Scripps, Thomas Edgington, said, "Weput our mutagenesis and physical chemistry data together with the U.K.people's three-dimensional crystallographic data," adding, "I certainlyhope for continued collaboration with Tuddenham et al.""For the beginning of a rational drug-design program," he said, thethree-dimensional structure of TF's docking site with Factor VII isnecessary knowledge." The Scripps scientist has just submitted a paperto the Journal of Biological Chemistry, "which maps out the firstfootprint on TF for the binding of its cognate ligand, factor VIIa." It isbased on "more than 170 mutagenesis changes on the TF protein."Tuddenham describes as "friendly rivalry" the fact that Genentech hasalso solved the TF's molecular structure, almost simultaneously withthe British group, "but to a resolution of only 2.4 A."Genentech crystallographer Abraham De Vos called it "friendlycompetition." His version will appear in the journal Biochemistry onSept. 6, 1994, as an accelerated publication.De Vos, a senior scientist in the South San Francisco company'sprotein engineering department, told BioWorld Today that the workreported in today's Nature "is clearly a very important first step." Itwill help drug designers do "more extensive and complete mutagenesisanalysis of the TF protein, to identify those amino acid residues thatinteract with factor VII and its activated form, VIIa."He explained, "Without the structure, there is always groping in thedark. With it, you can do a good job, because then you know exactlywhat's on the outside, to identify those few residues that do have aneffect on binding."Without Complex, Don't Even Think DrugsDe Vos continued, "Then you can, at least conceptually, think aboutdesigning a molecule that would compete with the TF/factor VIIcomplex."But the success of such an approach, he pointed out, "would dependvery much on having a fairly small binding epitope, because thesurface of TF that interacts with factor VII is very large, and you can'tdesign a small molecule to do the same thing."So beyond the British "very important first step," (and Genentech'ssimultaneous one), De Vos sees an even more important second step:Namely, a structural analysis of the binary molecular complex betweenTF and factor VII."Since the role of this drug design is to block TF from binding to VII,you need to know where these proteins interact, which you don't knowwith the TF structure by itself." He added, "Whenever you talk topeople who think of doing rational drug design on this system,everybody almost inevitably says, `We'd like to see a structure of thecomplex before we would seriously contemplate designing a drug.' "Genentech is already embarked on a project to solve the complex'sbinary structure, but De Vos does not foresee further collaboration inthis direction with Boys and Tuddenham. Hoffmann-La Roche, ofBasel, Switzerland, he explained, "has also started working in the samearea. Considering the relationship between our two companies [Rocheis majority stockholder in Genentech], I think our collaboration will bewith them."Edgington observed, "The world is not waiting for every binarycomplex to be solved crystallographically." But he quickly added, "Imust say that probably the greatest amount of TF information willcome from the binary complex, TF-plus-VIIa."Granting, with De Vos, that this is "the next step," he emphasized that"the actual functional unit is a ternary [triple] complex of TF-plus-VIIa-plus factor X. The actual protein substrate that's going to beactivated comes and sits in there." n082594
-- David N. Leff Science Editor
(c) 1997 American Health Consultants. All rights reserved.