...of protein Z-dependent protease inhibitor (ZPI) by protein Z

来自 : 发布时间：2025-01-10

(PDF) Structural basis for catalytic activation of protein Z-dependent protease inhibitor (ZPI) by protein Z THROMBOSIS AND HEMOSTASISStructural basis for catalytic activation of protein Z–dependent protease inhibitor(ZPI) by protein Z*Xin Huang,1*Yahui Yan,2Yizheng Tu,3Jeffrey Gatti,1George J. Broze Jr,3Aiwu Zhou,4and Steven T. Olson11Center for Molecular Biology of Oral Diseases, University of Illinois at Chicago, Chicago, IL; 2Department of Haematology, University of Cambridge, Cambridge,United Kingdom; 3Division of Hematology, Washington University, St Louis, MO; and 4Key Laboratory of Cell Differentiation and Apoptosis of Ministry ofEducation of China, Shanghai JiaoTong University School of Medicine, Shanghai, ChinaThe anticoagulant serpin, proteinZ-dependent protease inhibitor (ZPI), iscatalytically activated by its cofactor, pro-tein Z (PZ), to regulate the function ofblood coagulation factor Xa on mem-brane surfaces. The X-ray structure of theZPI-PZ complex has shown that PZ bindsto a unique site on ZPI centered on helixG. In the present study, we show byAla-scanning mutagenesis of the ZPI-binding interface, together with nativePAGE and kinetic analyses of PZ bindingto ZPI, that Tyr240 and Asp293 of ZPI arecrucial hot spots for PZ binding. Comple-mentary studies with protein Z–proteinC chimeras show the importance of bothpseudocatalytic and EGF2 domains ofPZ for the critical ZPI interactions. Tounderstand how PZ acts catalytically, weanalyzed the interaction of reactive loop–cleaved ZPI (cZPI) with PZ and deter-mined the cZPI X-ray structure. The cZPIstructure revealed changes in helicesA and G of the PZ-binding site relative tonative ZPI that rationalized an observed6-fold loss in PZ affinity and PZ catalyticaction. These findings identify the keydeterminants of catalytic activation of ZPIby PZ and suggest novel strategies forameliorating hemophilic states throughdrugs that disrupt the ZPI-PZ interaction.(Blood. 2012;120(8):1726-1733)IntroductionProtein Z (PZ)–dependent protease inhibitor (ZPI), an anticoagu-lant protein of the serpin superfamily, is activated by its cofactor,PZ, to rapidly and specifically inhibit membrane-associated coagu-lation factor Xa. ZPI also rapidly inhibits factor XIa, but PZactivation is not required for this inhibition and cleaved ZPI is thepredominant product.1-3 The importance of ZPI for regulation offactor Xa activity is evident from the observation that ZPI or PZdeficiency results in an increased risk of thrombosis, especiallywhen combined with the factor V Leiden mutation or other riskfactors.4-6 ZPI circulates in blood plasma as a tight complex withPZ,7a vitamin K–dependent protein with a domain structuresimilar to that of factors VII, IX, and X and protein C, but with anonfunctional protease domain.8Biochemical and structural stud-ies have suggested that PZ promotes ZPI inhibition of membrane-associated factor Xa primarily though a bridging mechanism inwhich the binding of PZ to ZPI enables the ZPI-PZ complex to bindto a membrane surface and encounter membrane-bound factorXa.9,10 PZ acts catalytically in promoting the ZPI–factor Xareaction, as evidenced by the observation that once ZPI forms aninhibited complex with factor Xa, it dissociates from PZ.11 Thiscatalytic action may be important for sparing PZ, because normalplasma concentrations of PZ are limiting relative to ZPI.2,12The X-ray structure of the ZPI-PZ complex was determinedrecently by 2 different groups of investigators.9,10 The structureshows that ZPI interacts with PZ through 3 clusters of salt bridgesinvolving residues D74, D238, K239, and D293 and throughhydrophobic interactions involving residues M71 and Y240 at aunique site centered on helix G. In the present study, we performedAla-scanning mutagenesis of these 6 ZPI contact residues in thebinding interface to determine their relative contributions to PZbinding. Kinetic analysis of the effect of these mutations on thecofactor-dependent ZPI–factor Xa reaction, together with nativePAGE analysis of PZ binding to the ZPI mutants, revealed largedifferential contributions of these residues to PZ binding, with only2 residues accounting for the bulk of the binding energy. Comple-mentary studies with PZ chimeras confirmed that the key ZPIresidues interacted with both pseudocatalytic and EGF2 domains ofPZ. Kinetic competition studies showed that reactive loop–cleavedZPI (cZPI) bound PZ with a 6-fold lower affinity than the nativeserpin. Solution of the crystal structure of cZPI and its alignmentwith native ZPI in the ZPI-PZ complex revealed that changes inhelices Aand G of the PZ-binding site accounted for the loss in ZPIaffinity for PZ. These findings provide new insights into themechanism of catalytic activation of ZPI by PZ and have implica-tions for the design of small molecules that could disrupt theZPI-PZ interface and potentially ameliorate hemophilic states.MethodsProteinsPlasma-derived human factor Xa, ZPI, PZ, and factor XIa were obtainedfrom Enzyme Research Laboratories or were purified as described previ-ously.3Recombinant human ZPI and ZPI mutants were expressed inbaculovirus-infected insect cells and purified as described previously.9,11A human ZPI Y387R variant and wild-type mouse ZPI were expressed inEscherichia coli and purified as described previously.10 The reactive centerloop cleaved form (cZPI) was prepared from a Y387R ZPI variant withSubmitted March 22, 2012; accepted July 3, 2012. Prepublished online asBlood First Edition paper, July 11, 2012; DOI 10.1182/blood-2012-03-419598.*X.H. and Y.Y. contributed equally to this work.The online version of this article contains a data supplement.The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 USC section 1734.© 2012 by The American Society of Hematology1726 BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom enhanced substrate reactivity1,9 by reacting with catalytic factor Xafollowed by purification.10 DNA mutagenesis was carried out by PCR usingthe Quick Exchange kit (Stratagene). Mutations were confirmed by DNAsequencing. Molar concentrations of ZPIs were determined from the absorbanceat 280 nm using a calculated extinction coefficient of 31 525M⫺1s⫺1.13Protease concentrations were determined by standard activity assays thatwere calibrated based on active-site titrations.11Recombinant PZ and protein Z/protein C (PZ/PC) chimeric proteinswere produced by manipulating PZ cDNA to encode the sequence forPZ/PC chimeric proteins using standard techniques. The pre-pro-leadersequence of PZ was replaced with that of prothrombin to improve proteinexpression levels, and the ␥-carboxyglutamic acid domain (Gla), epidermalgrowth factor 1 domain (EGF1), epidermal growth factor 2 domain (EGF2),or Gla through EGF2 domains (Gla-EGF2) of PZ were replaced with thecorresponding domains of PC. cDNAs encoding the PZ/PC chimericproteins and wild-type PZ were engineered into the pcDNA4/TO (Invitro-gen) vector with placement of a 6-His tag at their C-termini and transfected(lipofectamine; Invitrogen) into T-Rex-293 cells (Invitrogen). Cells werecultured in DMEM with tetracycline-reduced FBS (10%, Invitrogen),L-glutamine (2mM), vitamin K (10 ␮g/mL), basticidin (5 ␮g/mL), andZeocin (300 ␮g/mL). At confluence, cells were washed and the mediareplaced with FreeStyle 293 expression medium (Invitrogen) with tetracy-cline (1 ␮g/mL) supplemented with the same levels of glutamine, vitaminK, basticidin, and Zeocin. After 24 hours, conditioned medium wascollected, benzamidine added (5mM), and the expressed proteins wereisolated by metal chelate (Talon; Clontech), Resource Q, and Superose12 (GE Health Care) chromatography. The chimeric proteins were ex-pressed at levels similar to PZ wild-type and reacted with the followingavailable mAbs in a fashion equivalent to that of native PZ or PC. ThemAbs tested included: anti-PZ EGF1 domain (mAb 2048.ED9), anti-PZpseudocatalytic domain (mAb 2306.AC5, mAb 2306.BF12, mAb 4260.2B9,and mAb 4260.5F5), anti-PC Gla domain (mAb CaC-11), and anti-PCEGF1 domain (mAb 1518.EG9).PhospholipidsSmall, unilamellar phospholipid vesicles were prepared from a 7:3 mixture(by weight) of dioleyl phosphatidylcholine and dioleoyl phosphatidylserine(Avanti Polar Lipids) as described previously.9,11 The phospholipid concen-tration was determined by colorimetric assay.14Kinetics of ZPI-FXa reactionsThe rate of factor Xa inactivation by wild-type and mutant ZPIs in theabsence and presence of cofactors was measured under pseudo-first-orderconditions in 50mM HEPES, 0.1M NaCl, 0.1% PEG 8000 buffer, pH 7.4, at25°C by discontinuous or continuous assays as described previously.9,11 Inthe discontinuous assay, reactions were quenched at various times by 10- to20-fold dilution into substrate (100␮M Spectrozyme FXa [AmericanDiagnostica] or 50␮M Pefafluor FXa [Centerchem]) and the residual factorXa activity was measured from the initial rate of absorbance change(405 nm) or fluorescence change (␭ex 380 nm, ␭em 440 nm). Quenchsolutions contained 10mM EDTA for reactions with calcium. Because ofthe slow rate of dissociation of the ZPI–factor Xa complex during the assay,initial velocities of substrate hydrolysis by residual factor Xa weredetermined from computer fits by a second-order polynomial function.11Observed pseudo-first-order rate constants (kobs) were obtained by fittingthe loss of protease activity by an exponential decay function with anonzero end point that reflected ⬍10% degraded protease more resistant toinhibition. For continuous assays, reactions included 50␮M Pefafluor FXasubstrate and progress curves of protease inhibition were obtained bycontinuously monitoring the decrease in rate of substrate hydrolysis to asteady-state level of inhibited enzyme activity under conditions in whichsubstrate consumption was ⬍10% and the rate of hydrolysis was linear inthe absence of inhibitor. Reaction progress curves were fit over5-10 half-lives by an exponential plus steady-state equation to obtain kobs.11Binding of wild-type and mutant ZPIs to PZ was quantitated fromkinetic titrations of the accelerating effect of PZ on kobs for the ZPI–factorXa reaction in the presence of 25␮M lipid and 1mM calcium. To ensure thatkobs was linearly dependent on the ZPI-PZ complex concentration, reactions weredone at ZPI-PZ complex concentrations below the KMof approximately 50nMfor PZ-ZPI-FXa ternary complex formation.11 To avoid lipid oxidation andachieve reproducible reaction rates, frozen aliquots of phospholipid vesiclesprepared by sonication were thawed and used without additional sonication.Second-order association rate constants for reactions in the absence (kass,uncat)orpresence of PZ, lipid, and calcium cofactors (kass,cat) and the KDfor the ZPI-PZcomplex interaction were obtained by fitting the dependence of kobs on the ZPI orPZ concentration by the equation:kobs ⫽kdiss ⫹(kass,cat [ZPI-PZ] ⫹kass,uncat [ZPI]f)/(1 ⫹[S]o/KM,S)where kdiss is the first-order rate constant for ZPI-FXa complexdissociation, [ZPI-PZ] and [ZPI]f(free ZPI) are functions of total ZPI andPZ concentrations and the KDfor the ZPI-PZ interaction as given by thequadratic equilibrium binding equation,11 [S]ois the fluorogenic substrateconcentration, and KM,S is the Michaelis constant for substrate hydrolysisby factor Xa. A KM,S of 119 ⫾3␮M was determined under the conditions ofthe kinetic experiments. For ZPI variants with high PZ affinity, theuncatalyzed reaction term and kdiss could be neglected. For ZPI variantswith low PZ affinity, kass,uncat and kdiss were fixed at values determinedindependently for the ZPI–factor Xa reaction in the absence of PZ but in thepresence of 25␮M lipid and 1mM CaCl2, and kass,cat was assumed to beequal to the wild-type value to allow fitting of KD.Experiments examining the competitive effect of cZPI on thePZ-dependent reactions of native ZPI or M71A ZPI with factor Xa weredone at 5.5nM ZPI, 5.5nM PZ, variable cZPI (0-500nM), 0.1nM factor Xa,25␮M lipid, and 5mM CaCl2. Reactions were initiated by adding a mixtureof ZPI and cZPI to a preincubated solution of PZ, factor Xa, lipid, andcalcium, and progress curves were measured by the discontinuous assay toobtain kobs. The decrease in kobs as a function of cZPI concentration was fitby the kinetic equation above except that the quadratic equation for the[ZPI-PZ] term was replaced by the cubic equation that defines the ZPI-PZcomplex concentration in the presence of a competitor.15 M71A ZPI–factorXa reaction data were fit by fixing the KDfor the M71A ZPI-PZ interactionat the measured value and fitting the KDfor the cZPI-PZ interaction as aparameter. Wild-type ZPI–factor Xa reaction data were fit by fixing thecZPI-PZ interaction KDand fitting the wild-type ZPI-PZ interaction KDas aparameter. Stoichiometric factors for ZPI and cZPI interactions with PZwere assumed to be 1. Simulations of ZPI-PZ complex reactions with factorXa were done using Explorer Version 2.5 software (KinTech).Stoichiometry of ZPI-protease reactionsThe stoichiometry of inhibition of factor Xa or factor XIa by ZPI wasdetermined by end point titrations of approximately 100nM protease withincreasing molar ratios of ZPI to protease from 3:1 to 8:1 in the absence orpresence of PZ (equimolar with ZPI), phospholipid (25 ␮m), and CaCl2(5mM) under the conditions of kinetic studies.9,11PAGENondenaturing PAGE was performed at 4°C using the Laemmli buffersystem16 with 5.5% gels and running times of 2-5 hours at 100-150 V.SDS-PAGE used the Laemmli buffer system and 7.5%-10% polyacryl-amide gels. Protein bands were detected by Coomassie blue staining.Characterization of PZ/PC chimeric proteinsPlasma PZ, recombinant wild-type PZ (rPZ), and the rPZ/PC chimericproteins were tested for their ability to accelerate the inhibition of factorXa (0.5nM) by plasma ZPI (40nM) in the presence of phospholipid (15␮Mrabbit brain cephalin; Pentapharm) and Ca2⫹(4mM) in 0.1M NaCl, 0.02MHEPES, pH 7.4, with 1 mg/mL of BSA, similar to previous studies.2PZ, factor Xa, phospholipid, and Ca2⫹were incubated for 3 minutes at 37°Cand then ZPI was added to start the reaction. Alternatively, ZPI, PZ,phospholipid, and Ca2⫹were incubated for 3 minutes at 37° and then factorXa was added. After 60 seconds, a sample (50 ␮L) was added to prewarmedbuffer (100 ␮L) and CaCl2(25mM, 50 ␮L) and the clotting time wasmeasured after the addition of factor X–deficient plasma (George KingCATALYTIC ACTIVATION OF ZPI BY PROTEIN Z 1727BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom Biomedical) in a fibrometer (BBL). Remaining factor Xa activity wasdetermined based on a factor Xa standard curve.The rPZ/PC chimeric proteins were tested for their ability to bind ZPIwith a microtiter-plate assay. rPZ/PC proteins (2 ␮g/mL in 0.1M NaCl,0.02M HEPES, pH 7.4) were allowed to bind to the microtiter wells(100 ␮L, 2 hours) followed by washing and blocking with PBS containingTween 20 (0.05%; PBST). ZPI in PBST was then added (0-2 ␮g/mL,100 ␮L) and incubated for 30 minutes. After washing with PBST, boundZPI was detected using biotin-conjugated mAb 4249.3 anti-ZPI (2 ␮g/mL),avidin-HRP (Pierce), and 3,3⬘,5,5⬘-tetramethylbenzidine (Sigma-Aldrich)with PBST washes between steps. The same method and a mAb against thepseudocatalytic domain of PZ (mAb 2306.AC5) showed comparablebinding of the PZ/PC chimeric proteins in replicate microtiter wells.Crystallization and X-ray structure determination of ZPIsCrystals of cZPI were grown in 1.8M sodium/potassium phosphate, pH 5,or 10% 2-propanol, 0.2M Li2SO4, 0.1M Na acetate, pH 4.5, using the sittingdrop method. Crystals of mouse ZPI were grown in 4M NaCl, 50mM MESbuffer, pH 6. cZPI crystals were quickly soaked in cryo-solution (theprecipitant plus 15% ethylene glycol) and snap-frozen in liquid nitrogen.Mouse ZPI crystals were snap-frozen without cryo-solution. Diffractiondata for cZPI and mouse ZPI were collected from single crystals at theDiamond Synchrotron stations I02 and I04-1 and processed with MosflmVersion 7.0.5 software.17 The cZPI and mouse structures were solved bymolecular replacement with Phaser Version 2.3.0 software18 using thereactive loop cleaved antitrypsin structure (pdb 7API) and native humanZPI (pdb 3F1S) as search models. cZPI crystals with space groups I2 andP41 were observed, each with one copy of cZPI in the correspondingasymmetric unit. Structures were built in COOT Version 0.6.2,19 refinedusing Refmac Version 5.6.0117,20 and validated by MolProbity Version3.1921 (supplemental Table 1, available on the Blood Web site; see theSupplemental Materials link at the top of the online article). All residues ofthe I2 cZPI structure apart from 1-40 and 387 were built. In the P41 cZPIstructure, residues 1-34, 115-116, 288, and 387-389 were not built due topoor density. Mouse ZPI was built with residues 47-425 (except for the7 reactive loop residues 380-386 due to poor density).ResultsEngineering and functional activity of ZPI variantsZPI variants were engineered in which the 3 clusters of salt bridgesand hydrophobic interactions in the ZPI-PZ interface involvingMet71 and Asp74 in helix A and its C-terminal extension; Asp238,Lys239, and Tyr240 in the loop connecting strands 3 and 4 of sheetC; and Asp293 in helix G were each mutated to Ala (Figure 1).A double Asp74/Asp293 mutant was also engineered. The mutantswere expressed and purified as in previous studies and yieldedamounts of protein comparable to wild-type.9All mutants inhibitedfactor Xa or factor XIa in the absence of cofactors at a rate similarto wild-type (Figure 2A and supplemental Figure 1). Moreover, theinhibition stoichiometries for reaction of the variant ZPIs withthese proteases were similar to or only modestly elevated relative towild-type reactions (Table 1 and supplemental Table 2). Theseresults indicated that the mutations did not perturb the cofactor-independent inhibitory function of ZPI.Native PAGE analysis of ZPI-PZ interactionsTo determine whether the ZPI mutations impaired the ability tobind PZ, we analyzed binding by nondenaturing PAGE.9,11 Bindingof PZ to ZPI induces a large shift in ZPI electrophoretic mobility,resulting in a complex band migrating intermediate between that ofZPI and PZ (Figure 3). Aminor complex band with slower mobilitythan the major complex band likely reflects the binding of a minorthrombin-cleaved Gla-domainless PZ species to ZPI.9When mixedwith a slight molar excess of PZ, bands for D238A and K239AZPIvariants were shifted completely to positions corresponding tocomplex, similar to the behavior of wild-type ZPI. Slight variationsin complex mobility reflected the altered charge of ZPI variants(supplemental Figure 2). In contrast, PZ caused partial band shiftsof M71A, D74A, and Y240A variants and no band shift for theD293A and D74A/D293AZPI variants.These results indicated thatmutation of ZPI binding-interface residues produces variableimpairments in PZ binding.Cofactor-dependent activity of ZPI variantsTo confirm the PZ-binding defects of the ZPI variants, the kineticsof the ZPI–factor Xa reaction were analyzed in the presence of PZequimolar with ZPI and with lipid and calcium cofactors. Contrast-ing the similar rates at which the mutant ZPIs inhibited factor Xa inthe absence of cofactors, the mutants showed strikingly differentcofactor-dependent rates of factor Xa inhibition (Figure 2B).Whereas D238A and K239A mutants showed rapid rates of factorXa inhibition similar to wild-type ZPI, the M71A, D74A, Y240A,D293A, and D74A/D293A mutants showed progressively greaterreductions in the cofactor-dependent rate of factor Xa inhibition.Relative to the wild-type ZPI reaction, the pseudo-first-orderinhibition rate constant (kobs) was decreased for mutant ZPIreactions by 2-fold for M71A, 5-fold for D74A, 50-fold for Y240A,400-fold for D293A, and 1300-fold for D74A/D293A. Stoichiom-etries of inhibition for the mutant ZPI reactions in the presence ofcofactors showed small or no changes from wild-type (Table 1).These results suggested markedly different contributions of ZPIcontact residues in the PZ-binding interface to bind PZ andaccelerate the reaction of ZPI with membrane-associated factor Xa.These contributions increased in the order D293 ⬎Y240 ⬎D74 ⬎M71, with D238 and K239 residues making no detectablecontributions. Such findings were in agreement with the nativePAGE analysis.Figure 1. ZPI contact residues in the ZPI-PZ complex interface. ZPI-PZ complexstructure (pdb 3F1S) with ZPI in green and PZ pseudocatalytic and EGF2 domains incyan. The reactive loop (RCL, red), sheet A (blue), helix A (orange), and helixG (purple) of ZPI are highlighted. The 6 ZPI residues of the contact interface in helixA, helix G, and sheet C are represented in stick.1728 HUANG et al BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom Quantification of ZPI-PZ interactionsTo quantify the PZ-binding defects of the mutant ZPIs, weperformed kinetic titrations of the accelerating effect of PZ on theZPI–factor Xa reaction in the presence of lipid and calcium. Forwild-type, D238A, and K239A ZPIs, kobs increased with increasingZPI concentration to an end point reached at a ZPI concentrationequimolar with PZ, indicating tight stoichiometric binding (Figure4A). The end point for the K239A variant was somewhat reducedfrom wild-type and D238A variants because of an increasedstoichiometry of inhibition (Table 1). Fitting of the data by theequilibrium-binding equation indicated a poorly determined sub-nanomolar affinity in all cases. For M71A and D74A variants, kobsshowed a more gradual increase to an end point similar towild-type, consistent with a significant weakening of PZ-bindingaffinity. Fitting of these titrations provided KDsof17⫾5nM and34 ⫾5nM, respectively, indicating minimal 20- to 30-fold de-creases in affinity for these mutants. Titrations with Y240A,D293A, and D74A/D293A ZPI variants showed barely detectableincreases in kobs over the same ZPI concentration range, indicatingweak binding. Extending the ZPI concentration range revealed aclear saturable increase in kobs for the Y240 variant, from which aKDof 1000 ⫾100nM was determined (Figure 4B). However,D293A and D74A/D293A variants still showed no detectableincreases in kobs over this range of ZPI concentrations relative towild-type ZPI in the absence of PZ. By titrating the PZ-dependentrate enhancement at a fixed level of ZPI and increasing PZconcentrations, significant linear increases in kobs were detectablefor these mutants, verifying a weak, PZ-dependent enhancement ofthe inhibition rate constant (Figure 4B). Assuming that thesemutations only affected PZ binding to ZPI and that the maximalrate constant at saturating PZ was equivalent to wild-type, esti-mates of KDfor the binding of Y240A, D293A, and D74A/D293AZPI variants to PZ of 1500 ⫾100nM, 40 000 ⫾14 000nM, and240 000 ⫾160 000nM, respectively, were made. These resultsconfirmed that massive binding defects result from mutations ofY240 and D293.ZPI binding to PZ/PC chimerasOf the 4 ZPI contact residues in which mutations resulted inimpaired binding to PZ, 3 contacted the pseudocatalytic domain ofPZ, whereas one, Y240, bound in a hydrophobic pocket formed bythe pseudocatalytic domain and the EGF2 domain of PZ (Figure 1).To assess the importance of the EGF2 domain in this interaction,PZ chimeras in which the Gla, EGF1, and EGF2 domains of PZwere replaced with the homologous domains of protein C individu-ally or together were tested for their ability to promote ZPIinhibition of factor Xa in the presence of lipid and calcium. Thepurity and appropriate size of the chimeras was confirmed bySDS-PAGE (supplemental Figure 3). The chimeric proteins wereexpressed at the same level as PZ wild-type and reacted in a fashionFigure 2. Kinetic analysis of mutant ZPI reactionswith factor Xa in the absence and presence of cofac-tors. (A) Progress curves for reactions of 480nM wild-type and mutant ZPIs as indicated with 3nM factor Xa inthe absence of cofactors. (B) Progress curves for reac-tions of 16nM wild-type and mutant ZPIs as indicated with0.25nM factor Xa in the presence of 13nM PZ, 25␮Mphospholipid, and 5mM Ca2⫹. Solid lines are fits by asingle exponential function.Table 1. Kinetic constants and stoichiometries of inhibition for ZPI-factor Xa reactions in the absence and presence of cofactors and KDsfor ZPI-PZ interactionsZPINo cofactors Cofactorskass,uncat,Mⴚ1sⴚ1SI, molI/mol Ekass,uncat ⴛSI, Mⴚ1sⴚ1kass,cat,Mⴚ1sⴚ1SI, molI/mol Ekass,cat ⴛSI, Mⴚ1sⴚ1*KDZPI-PZ, MWild-type 1.0 ⫾0.1 ⫻1043.6 ⫾0.4 3.6 ⫾0.8 ⫻1043.1 ⫾0.1 ⫻1062.8 ⫾0.2 8.7 ⫾0.9 ⫻1061.2 ⫾0.1 ⫻10⫺9M71A 8.6 ⫾0.7 ⫻1033.7 ⫾0.1 3.2 ⫾0.3 ⫻1042.8 ⫾0.3 ⫻1062.9 ⫾0.1 8.1 ⫾1.1 ⫻1061.7 ⫾0.5 ⫻10⫺8D74A 9.5 ⫾0.1 ⫻1034.1 ⫾0.1 3.9 ⫾0.1 ⫻1043.3 ⫾0.2 ⫻1062.6 ⫾0.1 8.6 ⫾0.9 ⫻1063.4 ⫾0.5 ⫻10⫺8D238A 6.0 ⫾0.4 ⫻1035.5 ⫾0.4 3.3 ⫾0.5 ⫻1043.1 ⫾0.1 ⫻1062.9 ⫾0.1 9.0 ⫾0.9 ⫻106ⱕ1⫻10⫺9K239A 9.3 ⫾0.2 ⫻1034.0 ⫾0.1 3.7 ⫾0.2 ⫻1042.4 ⫾0.1 ⫻1064.2 ⫾0.1 10 ⫾1⫻106ⱕ1⫻10⫺9Y240A 5.8 ⫾0.1 ⫻1034.7 ⫾0.1 2.7 ⫾0.1 ⫻104ND ND 1-1.5 ⫻10⫺6D293A 7.6 ⫾0.6 ⫻1033.8 ⫾0.1 2.9 ⫾0.3 ⫻104ND ND 4.0 ⫾1.4 ⫻10⫺5D74A/D293A 1.1 ⫾0.1 ⫻1044.7 ⫾0.1 5.2 ⫾0.6 ⫻104ND ND 2.4 ⫾1.6 ⫻10⫺4Rate constants in the absence of cofactors (kass,uncat) were obtained from the slope of linear plots of pseudo-first-order rate constants (kobs) on ZPI concentration in therange of 100-800nM. Rate constants for reactions in the presence of cofactors (kass,cat) were obtained from the data shown in Figure 4. kass,cat could not be determined for ZPIvariants with low PZ affinity (Y240A, D293A, and D74A/D293A). Stoichiometry of inhibition (SI) values were determined from stoichiometric titrations of factor Xa with ZPI asdescribed in \"Stoichiometry of ZPI-protease reactions.” The product of kass,uncat or kass,cat and SI represents the corrected association rate constant for reaction through theinhibitory pathway. KDs for ZPI-PZ interactions were determined from fits of the kinetic titrations shown in Figures 4 and 6 using the equation provided in \"Kinetics of ZPI-FXareactions.”ND indicates not determined.*Wild-type rate constants are somewhat lower than those measured in previous studies9,11 because of the reduced calcium concentration.CATALYTIC ACTIVATION OF ZPI BY PROTEIN Z 1729BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom equivalent to native PZ with a bank of mAbs, suggesting that theywere folded appropriately. The EGF1 chimera enhanced the rate ofZPI inhibition of factor Xa to the same extent as recombinant orplasma forms of wild-type PZ, indicating that this domain was notimportant for PZ rate enhancement (Figure 5A). The Gla domainchimera showed a reduced rate-enhancing effect, which is consis-tent with previous studies showing that a PZ–Gla domain interac-tion with the factor Xa Gla domain made an important contributionto the PZ-dependent acceleration of the ZPI–factor Xa reaction on amembrane surface.3,11,22 The EGF2 chimera or combined Gla,EGF1, and EGF2 chimeras showed no ability to enhance the rate ofthe ZPI reaction with membrane-associated factor Xa, which isconsistent with an important role for the EGF2 domain in thePZ-dependent rate enhancement. Solid-phase binding assays of thePZ chimeras to ZPI showed similar strong interactions of ZPI withwild-type and EGF1 or Gla-domain chimeras and no detectablebinding of the EGF2 or combined Gla, EGF1, and EGF2 chimeras(Figure 5B). These results are consistent with an important role forthe PZ EGF2 domain, but not the EGF1 or Gla domains, forbinding ZPI in addition to the well-established role of thepseudocatalytic domain for binding.22Binding of cZPI to PZPZ dissociates from ZPI after ZPI forms an inhibited complex withfactor Xa, allowing PZ to recycle as a catalyst.11 To determine theextent of the PZ affinity loss for ZPI in the ZPI–factor Xa complex,we analyzed PZ binding to cZPI. This was based on findings thatthe serpin moiety of the serpin-protease complex is structurallyindistinguishable from cleaved serpin23,24 and that both cZPI andZPI–factor Xa complex products of the ZPI–factor Xa reactionsimilarly fail to bind PZ at nM levels at which native ZPI binds.11cZPI reduced kobs for the reaction of 5nM native ZPI-PZ complexwith factor Xa in the presence of lipid and calcium in a dose-dependent manner (Figure 6A), with 50% reduction occurring atapproximately 50nM cZPI (Figure 6B). This was consistent withcZPI competing with native ZPI for binding PZ, thereby reducingthe level of ZPI-PZ complex competent to inhibit membrane-associated factor Xa. Because the KDfor the wild-type ZPI-PZinteraction was not well determined, we also analyzed the effect ofcZPI on the membrane-dependent reaction of M71A ZPI-PZcomplex with factor Xa. As expected, cZPI competed moreeffectively with M71A ZPI for PZ, as evidenced by the 50%reduction in kobs at approximately 10nM cZPI (Figure 6B). Fittingof both datasets by the cubic equation for competitive binding15 byfixing the KDfor the M71A ZPI-PZ interaction at 17nM (Figure 4)indicated KDs for wild-type ZPI-PZ and cZPI-PZ interactions of1.2 ⫾0.1nM and 6.8 ⫾0.5nM, respectively. Reactive loop cleav-age of ZPI thus reduces PZ affinity by approximately 6-fold. Theability of cZPI to bind PZ was confirmed by native PAGE analysis(supplemental Figure 4). Simulations of the reaction of the ZPI-PZcomplex with factor Xa at physiologic concentrations of ZPI andPZ, assuming the measured KDs for native ZPI and the cZPI–factorXa complex, showed that the decrease in PZ affinity is sufficient toaccount for the catalytic effect of PZ (supplemental Figure 5).Structure of cZPITwo crystal forms of cZPI were obtained from different crystalliza-tion conditions, one from the I2 space group diffracting up to2.09 Å and a second from the P41 space group diffracting up to2.65 Å. The statistics of data collection and refinements are shownin supplemental Table 1. Overall, the 2 structures share similarfeatures except for the position of helix D and confirm that ZPIundergoes the typical stressed to relaxed conformational transitionof serpins on reactive loop cleavage (Figure 7A and supplementalFigure 6A). Compared with the structure of the ZPI-PZ complex(Figure 1), the cZPI structures show that there are marked changesin both helix Aand helix G apart from the expansion of sheetAthataccompanies reactive loop insertion into sheet A. HelixA, which isconsiderably longer in ZPI than in other serpins, is bent in theZPI-PZ complex, presumably to allow hydrophobic interactionsbetween W46 and neighboring leucines. These interactions arebroken in cZPI by the repositioning of helix A when sheet Aexpands, allowing helix A to straighten (supplemental Figure 6B).Helix G is partially unwound in both cZPI structures relative to theFigure 4. Kinetic titrations of mutant ZPI binding toPZ. (A) Titrations of the PZ-dependent increase in kobs forwild-type and mutant ZPI reactions with factor Xa (0.04nM)in the presence of 8.8nM PZ, variable ZPI, 25␮M phospho-lipid, and 1mM Ca2⫹. (B) Titrations of the low-PZ-affinityvariants of ZPI from panel A over an extended range ofZPI concentrations, along with control wild-type ZPI in theabsence of PZ. Inset shows titrations of the PZ-dependent increase in kobs for reactions of low-PZ-affinityZPI variants with factor Xa at 55nM ZPI, variable PZ,25␮M phospholipid, and 1mM Ca2⫹.kobs values repre-sent averages of 2-3 independent measurements. Solidlines are fits of data by the equation given in \"Kinetics ofZPI-FXa reactions” from which values of KDfor theZPI-PZ interaction and kass,cat were determined (Table 1).For ZPI variants with low PZ affinity, kass,cat was fixed atthe wild-type value.Figure 3. Native PAGE analysis of mutant ZPI binding to PZ. Lane 1 is wild-typeZPI; lanes 2-9, wild-type or mutant ZPIs (approximately 3␮M), as indicated, mixedwith a molar excess of plasma PZ; and lane 10, PZ.1730 HUANG et al BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom ZPI-PZ complex, indicating that this structural change is not anartifact of crystal packing. To determine whether uncomplexed ZPIadopts a similar conformation, we solved the structure of mouseZPI. Mouse ZPI has a typical native serpin structure that resembleshuman ZPI in the ZPI-PZ complex, including helix G, and differsonly in the reactive loop and position of helix A. Helix Gunwinding thus appears to be induced in cZPI by the stressed torelaxed transition (supplemental Figure 7). When the cZPI struc-tures are overlaid with ZPI in the ZPI-PZ complex, movements ofthe relative positions of the 4 key residues involved in binding PZare observed (Figure 7B). Specifically, Y240 in sheet C is slightlyshifted away from D74 in the loop extending from helix A andD293 of helix G is shifted closer to M71 of helix A, with the latterchange being the most significant (supplemental Figure 8).DiscussionIn the present study, we have elucidated the relative contributionsof 6 ZPI residues that form prominent salt-bridge and hydrophobiccontacts with PZ in the X-ray structure of the ZPI-PZ complex tobinding PZ and to reducing PZ affinity when ZPI undergoescleavage in the ZPI–factor Xa complex. These residues werechosen for mutagenesis based on their complementary interactionswith PZ residues in the binding interface and their conservation inmouse, rat, and chicken ZPIs. Surprisingly, only 2 of the 6 contactresidues on ZPI, D293 on helix G and Y240 in the gate region loopof sheet C (Figure 1), were found to be critical for binding PZ.Unitary binding energies25 of ⫺8.4 kcal/mol for Y240 and⫺10.4 kcal/mol for D293 can be calculated for these residuesbased on estimated KDs for Y240A and D293A interactions, whichtogether account for essentially all of the binding energy of theinteraction. Such \"hot spot” binding residues are common inprotein-protein interactions and indicate that binding energy is notdistributed additively over the binding interface, but instead isrealized through cooperative interactions of key residues.26 Theimportance of these residues was not predictable from the X-raystructures of the ZPI-PZ complex based on the energetics of burialof contact residues (supplemental Table 3). Salt bridges of D293 inZPI with R298 and H210 of PZ and the hydrophobic interaction ofY240 of ZPI in the hydrophobic cavity formed by the EGF2 andpseudocatalytic domains of PZ thus appear to be crucial determi-nants of the ZPI-PZ interaction.We found that replacement of the EGF2 domain of PZ with thehomologous domain of protein C completely abrogates the abilityof PZ to bind ZPI or to enhance the rate of the ZPI reaction withmembrane-associated factor Xa. This finding reinforces the criticalimportance of the ZPI Y240 interaction with the PZ pseudocata-lytic and EGF2 domains for the ZPI-PZ interaction. Surprisingly, aprevious study of a PZ-factor Xa chimera in which the Gla, EGF1,and EGF2 domains of PZ were replaced with those of factor Xashowed normal binding to ZPI.22 In contrast to the work presentedherein, the solid phase-binding assay used in that study detectedsubstantially weaker interactions of adsorbed ZPI with solution-phase PZ and reported that the binding could be detected in thepresence of calcium ions, but not EDTA, implying a divalent cationdependence of the PZ-ZPI interaction. The reason for thesediscrepancies is not clear. Whereas D293 andY240 residues of ZPIFigure 5. Characterization of cofactor activities ofPZ-PC chimeric proteins. (A) Comparison of the abilityof PZ and PZ-PC chimeras to accelerate the inhibition of0.5nM factor Xa by 40nM plasma ZPI in the presence ofphospholipid (15␮M) and Ca2⫹(4mM). Reactions wereinitiated by adding ZPI last and were allowed to proceedfor 60 seconds before measuring residual factor Xaactivity by coagulation assay. Reactions initiated withfactor Xa showed similar relative ZPI reactivities (notshown). (B) Analysis of the binding of ZPI to the recombi-nant forms of PZ as assessed in a microtiter-plate assaydescribed in \"Characterization of PZ/PC chimeric pro-teins.”Figure 6. Competitive effect of cZPI on the PZ-dependent ZPI–factor Xa reaction. (A) Progress curvesfor the reaction of 5.5nM ZPI with 0.1nM factor Xa in thepresence of 5.5nM PZ, 25␮M lipid, 5mM Ca2⫹, andincreasing concentrations of cZPI. Solid lines are fits to asingle exponential function from which kobs was deter-mined. (B) Percentage decrease in kobs for the PZ-accelerated reactions of wild-type ZPI (F) and M71A ZPI(Œ) with factor Xa plotted as a function of the competitorcZPI concentration. Solid lines are fits by the cubicequation for competitive binding of cZPI and wild-type orM71A ZPI to PZ after fixing the KDfor the M71A ZPI-PZinteraction at the value determined from the data shownin Figure 4.CATALYTIC ACTIVATION OF ZPI BY PROTEIN Z 1731BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom are of overriding importance for binding PZ, interactions of M71and D74 in helix A of ZPI with PZ also were found to makesignificant, albeit much smaller, contributions to binding. Notably,our results demonstrate that D238 and K239 contact residues ofZPI do not contribute to PZ binding.Our determination of the structure of cleaved ZPI and studies ofits interactions with PZ have provided new insight into how PZ actsas a catalyst in promoting ZPI inhibition of membrane-associatedfactor Xa. This catalysis results from a loss in PZ affinity for ZPIafter it is cleaved in the ZPI–factor Xa complex, causing PZ todissociate and bind another ZPI molecule.11 Kinetic competitionstudies revealed that cZPI binds PZ with a 6-fold lower affinitythan native ZPI. Simulations showed that such an affinity lossaccounts for PZ catalytic action and the sparing of PZ underphysiologic conditions in which PZ levels are limiting relative toZPI.2Whereas the reduction in ZPI affinity is not sufficient to cause100% of bound PZ to dissociate from the ZPI–factor Xa complex,this may not be necessary given that ZPI levels only modestlyexceed those of PZ, so the fraction of PZ that does dissociate issufficient to ensure that all ZPI is activatable by PZ. Indeed,catalytic levels of PZ are sufficient to accelerate the reaction ofmuch higher levels of ZPI and factor Xa in the presence of lipid andcalcium.11Alignment of the structures of cleaved and native ZPI suggestedhow structural changes in ZPI induced by factor Xa cleavage causethe observed 6-fold loss in PZ affinity. The structure of theZPI-binding site was largely maintained in cZPI, which is consis-tent with the secondary structural elements of the binding sitemoving as a rigid unit in the conformational transition accompany-ing cleavage, as occurs in other cleaved serpins.27 Nevertheless,structural changes unique to cZPI, including the straightening ofhelix A and the unwinding of helix G, caused the relative positionsof the key binding residue side chains of these helices to besimilarly altered in both cleaved structures, suggesting that thesechanges were induced by the native to cleaved conformationaltransition rather than by an inherent flexibility of the side chains(supplemental Figure 8). The modestly reduced affinity of PZ forcleaved ZPI may thus arise from a reduced complementarity of thesecondary Met71 and Asp74 residues in the binding site once thecritical Tyr240 and Asp293 residues have bound PZ.The results of the present study have additional implications fordesigning small molecules that could disrupt the ZPI-PZ interac-tion and thereby reduce the anticoagulant regulation of factor Xa atsites of factor X activation. Such molecules might be targeted todisrupt either of the 2 \"hot spot” interactions that mediate theZPI-PZ interaction. This could possibly restore a balance betweenanticoagulant and procoagulant factors in certain hemophiliadisorders to support normal hemostasis. The feasibility of such anapproach has been demonstrated with several other protein-proteininteractions.28AcknowledgmentsThe authors thank S. Paul Bajaj (University of California-LosAngeles) for providing the mAb against the PC Gla domain(CaC-11), and Peter Gettins (University of Illinois-Chicago) forproviding critical comments on the manuscript.This study was supported by the National Institutes of Health(grants R37 HL39888 to S.T.O. and HL60782 to G.J.B.), theAmerican Heart Association (Scientist Development GrantSDG4880022 to X.H.), and the British Heart Foundation (grantPG/09/072 to A.Z.). A.Z. was supported by The Program forProfessor of Special Appointment (Eastern Scholar) at ShanghaiInstitutions of Higher Learning and by Innovation Program ofShanghai Municipal Education Commission (no. 12ZZ113).AuthorshipContribution: X.H., G.J.B., A.Z., and S.T.O. designed the experi-ments; X.H., Y.Y., Y.T., and J.G. performed the experiments; andX.H., Y.Y., G.J.B., A.Z., and S.T.O. analyzed the data and wrote themanuscript.Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.Figure 7. Comparison of the structures of cZPI andZPI. (A) Ribbon structure of cZPI (I2 space group) in graywith sheet Ain blue, the cleaved reactive loop in red, helixA in orange, and helix G in purple. (B) Superposition ofthe PZ-binding sites of native (cyan) and cleaved (green)ZPI by aligning strand 6 of sheet A. The 4 key PZ-interacting residues are represented in stick.1732 HUANG et al BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom The current affiliation for Y.T. is Astellas Research Institute ofAmerica, Skokie, IL.Correspondence: Steven Olson, Center for Molecular Biologyof Oral Diseases, University of Illinois at Chicago, 801 S PaulinaSt, Chicago, IL 60612; e-mail: stolson@uic.edu; or Aiwu Zhou,Key Laboratory of Cell Differentiation and Apoptosis of ChineseMinistry of Education, and Shanghai Key Laboratory of TumorMicroenvironment and Inflammation, Shanghai Jiao-Tong UniversitySchool of Medicine, Chongqing South Road 280, Shanghai 200025,China; e-mail: aiwuzhou@googlemail.com; or George J. Broze Jr,Division of Hematology, Washington University, 660 S Euclid Ave, StLouis, MO 63110; e-mail: gbroze@dom.wustl.edu.References1. Han X, Huang ZF, Fiehler R, Broze GJ Jr. Theprotein Z-dependent protease inhibitor is a ser-pin. Biochemistry. 1999;38(34):11073-11078.2. Han X, Fiehler R, Broze GJ Jr. Characterizationof the protein Z-dependent protease inhibitor.Blood. 2000;96(9):3049-3055.3. Han X, Fiehler R, Broze GJ Jr. Isolation of a pro-tein Z-dependent plasma protease inhibitor. ProcNatl Acad Sci U S A. 1998;95(16):9250-9255.4. Zhang J, Tu Y, Lu L, Lasky N, Broze GJ Jr. Pro-tein Z-dependent protease inhibitor deficiencyproduces a more severe murine phenotype thanprotein Z deficiency. Blood. 2008;111(10):4973-4978.5. Sofi F, Cesari F, Tu Y, et al. Protein Z-dependentprotease inhibitor and protein Z in peripheral arte-rial disease patients. J Thromb Haemost. 2009;7(8):731-735.6. Kemkes-Matthes B, Nees M, Kuhnel G, Matzdorff A,Matthes KJ. Protein Z influences the prothrom-botic phenotype in Factor V Leiden patients.Thromb Res. 2002;106(4-5):183-185.7. Tabatabai A, Fiehler R, Broze GJ Jr. Protein Zcirculates in plasma in a complex with proteinZ-dependent protease inhibitor. Thromb Hae-most. 2001;85(4):655-660.8. Sejima H, Hayashi T, Deyashili Y, Nishioka J,Suzuki K. Primary structure of vitamin K-dependenthuman protein Z. Biochem Biophys Res Com.1990;171(2):661-668.9. Huang X, Dementiev A, Olson ST, Gettins PG.Basis for the specificity and activation of the ser-pin protein Z-dependent proteinase inhibitor (ZPI)as an inhibitor of membrane-associated factorXa. J Biol Chem. 2010;285(26):20399-20409.10. Wei Z, Yan Y, Carrell RW, Zhou A. Crystal struc-ture of protein Z-dependent inhibitor complexshows how protein Z functions as a cofactor inthe membrane inhibition of factor X. Blood. 2009;114(17):3662-3667.11. Huang X, Swanson R, Broze GJ Jr, Olson ST.Kinetic characterization of the protein Z-dependentprotease inhibitor reaction with blood coagulationfactor Xa. J Biol Chem. 2008;283(44):29770-29783.12. Miletich JP, Broze GJ Jr. Human plasma proteinZ antigen: range in normal subjects and the effectof warfarin therapy. Blood. 1987;69(6):1580-1586.13. Gill SC, von Hippel PH. Calculation of protein ex-tinction coefficients from amino acid sequencedata. Anal Biochem. 1989;182(2):319-326.14. Stewart JC. Colorimetric determination of phos-pholipids with ammonium ferrothiocyanate. AnalBiochem. 1980;104(1):10-14.15. Lindahl P, Raub-Segall E, Olson ST, Bjo¨rkI.Pa-pain labelled with fluorescent thiol-specific re-agents as a probe for characterization of interac-tions between cysteine proteinases and theirprotein inhibitors by competitive titrations.Biochem J. 1991;276(Pt 2):387-394.16. Laemmli UK. Cleavage of structural proteins dur-ing the assembly of the head of bacteriophageT4. Nature. 1970;227(5259):680-685.17. Evans PR. In: Sawyer L, Isaacs N, Baily S, eds.Proceedings of the CCP4 Study Weekend DataCollection and Processing. Darsbury, UnitedKingdom: Darsbury Laboratory; 1993.18. McCoy AJ, Grosse-Kunstleve RW, Storoni LC,Read RJ. Likelihood-enhanced fast translationfunctions. Acta Crystallogr D Biol Crystallogr.2005;61(4):458-464.19. Emsley P, Cowtan K. Coot: model-buildfing toolsfor molecular graphics. Acta Crystallogr D BiolCrystallogr. 2004;60(Pt 12 Pt 1):2126-2132.20. Winn MD, Isupov MN, Murshudov GN. Use ofTLS parameters to model anisotropic displace-ments in macromolecular refinement. Acta Crys-tallogr D Biol Crystallogr. 2001;57(Pt 1):122-133.21. Davis IW, Leaver-FayA, Chen VB, et al. Molpro-bity: all-atom contacts and structure validation forproteins and nucleic acids. Nucleic Acids Res.2007;35(Web Server issue):W375-W383.22. Rezaie AR, Bae J-S, Manithody C, Qureshi SH,Yang L. Protein Z-dependent protease inhibitorbinds to the C-terminal domain of protein Z. J BiolChem. 2008;283(29):19922-19926.23. Huntington JA, Read RJ, Carrell RW. Structure ofa serpin-protease complex shows inhibition bydeformation. Nature. 2000;407(6806):923-926.24. Peterson FC, Gettins PGW. Insight into themechanism of serpin-proteinase inhibition from2D (1H-15N) NMR studies of the 69 kDa alpha1-proteinase inhibitor Pittsburgh-trypsin covalentcomplex. Biochemistry. 2001;40(21):6284-6892.25. Lewis SD, Shields PP, Shafer JA. Characteriza-tion of the kinetic pathway for liberation of fibrino-peptides during assembly of fibrin. J Biol Chem.1985;260(18):10192-10199.26. Clackson T, Wells JA. A hot spot of binding en-ergy in a hormone-receptor interface. Science.1995;267(5196):383-386.27. Whisstock JC, Skinner R, Carrell RW, LeskAM.Conformational changes in serpins: I. The nativeand cleaved conformations of alpha1-antitrypsin.J Mol Biol. 2000;296(2):685-699.28. Shahian T, Lee GM, Lazic A, et al. Inhibition of aviral enzyme by a small-molecule dimer disruptor.Nat Chem Biol. 2009;5(9):640-646.CATALYTIC ACTIVATION OF ZPI BY PROTEIN Z 1733BLOOD, 23 AUGUST 2012 䡠VOLUME 120, NUMBER 8For personal use only.on May 8, 2015. by guest www.bloodjournal.orgFrom