Crystal structure of cleaved bovine antithrombin III at 3.2 A resolution

Serpin alpha 1proteinase inhibitor probed by intrinsic tryptophan fluorescence spectroscopy

Various conformational forms of the archetypal serpin human alpha 1proteinase inhibitor (alpha 1PI), including ordered polymers, active and inactive monomers, and heterogeneous aggregates, have been produced by refolding from mild denaturing conditions. These forms presumably originate by different folding pathways during renaturation, under the influence of the A and C sheets of the molecule. Because alpha 1PI contains only two Trp residues, at positions 194 and 238, it is amenable to fluorescence quenching resolved spectra and red-edge excitation measurements of the Trp environment. Thus, it is possible to define the conformation of the various forms based on the observed fluorescent properties of each of the Trp residues measured under a range of conditions. We show that denaturation in GuHCl, or thermal denaturation in Tris, followed by renaturation, leads to the formation of polymers that contain solvent-exposed Trp 238, which we interpret as ordered head-to-tail polymers (A-sheet polymers). However, thermal denaturation in citrate leads to shorter polymers where some of the Trp 238 residues are not solvent accessible, which we interpret as polymers capped by head-to-head interactions via the C sheet. The latter treatment also generates monomers thought to represent a latent form, but in which the environment of Trp 238 is occluded by ionized groups. These data indicate that the folding pathway of alpha 1PI, and presumably other serpins, is sensitive to solvent composition that affects the affinity of the reactive site loop for the A sheet or the C sheet.

Modeling of serpin-protease complexes: antithrombin-thrombin, alpha 1-antitrypsin (358Met-->Arg)-thrombin, alpha 1-antitrypsin (358Met-->Arg)-trypsin, and antitrypsin-elastase

Based on the most recent available crystal structures and biochemical studies of protease complexes of normal and mutant serine protease inhibitors (serpins), we have built models of the complexes: alpha 1-antitrypsin + human neutrophil elastase; alpha 1-antitrypsin Pittsburgh (358Met-->Arg) (Scott et al., J. Clin. Invest. 77:631-634, 1986) + tyrpsin; alpha 1-antitrypsin Pittsburgh (358Met-->Arg) + thrombin; and antithrombin + thrombin. All serpin sequences correspond to human molecules. The models show correct stereochemistry and no steric clashes between protease and inhibitor. The main structural differences in the serpins from the parent structures are: (1) the reactive center loop is inserted into the A-sheet as far as P12; (2) strand s1C is removed from the C-sheet; and (3) the C-terminus has changed conformation and interacts with the protease. In the absence of an X-ray structure determination of a serpin-protease complex, the demonstration that insertion of the reactive center loop into the A-sheet as far as P12 is stereochemically feasible provides structures of a protease-bound conformation of intact serpins with which to rationalize the properties of mutants, guide the design of experiments, and form a basis for further modeling studies, such as the investigation of the interaction of heparin with serpin-protease complexes.

The native strains in the hydrophobic core and flexible reactive loop of a serine protease inhibitor: crystal structure of an uncleaved alpha1-antitrypsin at 2.7 A

BACKGROUND

The protein alpha1-antitrypsin is a prototype member of the serpin (serine protease inhibitor) family and is known to inhibit the activity of neutrophil elastase in the lower respiratory tract. Members of this family undergo a large structural rearrangement upon binding to a target protease, involving cleavage of the reactive-site loop. This loop is then inserted into the main body of the enzyme following the opening of a central beta sheet, leading to stabilization of the structure. Random mutageneses of alpha1-antitrypsin identified various mutations that stabilize the native structure and retard the insertion of the reactive-site loop. Structural studies of these mutations may reveal the mechanism of the conformational change.

RESULTS

We have determined the three-dimensional structure of an uncleaved alpha1-antitrypsin with seven such stabilizing mutations (hepta alpha1-antitrypsin) at 2.7 A resolution. From the comparison of the structure with other serpin structures, we found that hepta alpha1-antitrypsin is stabilized due to the release of various strains that exist in native wild type alpha1-antitrypsin, including unfavorable hydrophobic interactions in the central hydrophobic core. The reactive-site loop of hepta alpha1-antitrypsin is an extended strand, different from that of the previously determined structure of another uncleaved alpha1-antitrypsin, and indicates the inherent flexibility of the loop.

CONCLUSIONS

The present structural study suggests that the uncleaved alpha1-antitrypsin has many folding defects which can be improved by mutations. These folding defects seem to be utilized in a coordinated fashion in the regulation of the conformational switch of alpha1-antitrypsin. Some of the defects, represented by the Phe51 region and possibly the Met374 and the Thr59 regions, are part of the sheet-opening mechanism.

Arginine substitutions in the hinge region of antichymotrypsin affect serpin beta-sheet rearrangement

A hallmark of serpin function is the massive beta-sheet rearrangement involving the insertion of the cleaved reactive loop into beta-sheet A as strand s4A. This structural transition is required for inhibitory activity. Small hydrophobic residues at P14 and P12 positions of the reactive loop facilitate this transition, since these residues must pack in the hydrophobic core of the cleaved serpin. Despite the radical substitution of arginine at the P12 position, the crystal structure of cleaved A347R antichymotrypsin reveals full strand s4A insertion with normal beta-sheet A geometry; the R347 side chain is buried in the hydrophobic protein core. In contrast, the structure of cleaved P14 T345R antichymotrypsin reveals substantial yet incomplete strand s4A insertion, without burial of the R345 side chain.

A Spectroscopic Study of the Structures of Latent, Active and Reactive-Center-Cleaved Type-1 Plasminogen-Activator Inhibitor

Type-1 plasminogen-activator inhibitor (PAI-1) was studied by Fourier-transform infrared spectroscopy, far-ultraviolet CD spectroscopy, and fluorescence-emission spectroscopy, with the aim to obtain structural information about its active form. The spectra of latent, active and reactive-center-cleaved forms of PAI-1 produced by HT-1080 cells were different. While the cleaved and the latent forms were similar with regard to their beta-structure content, comparison of the spectra of these forms with the spectra of active PAI-1 suggested a much higher degree of unordered structure for the active form compared with the latent and reactive-center-cleaved forms than previously assumed. We discuss our results with reference to the known three-dimensional X-ray structures of latent PAI-1, of reactive-center-cleaved serpins, including reactive-center-cleaved PAI-1, and of intact serpins, and with reference to previous results on the differences in the affinity of mAbs for the different PAI-1 forms. We interpret our results in favor of a global rearrangement of secondary structure during latency transition and reactive-center cleavage in PAI-1, not only involving the reactive-center loop and parts of beta-sheets A and C, but also the "rear' side of the molecule, such as helices H and G. Thus, we suggest flexibility in serpin structural elements that were previously regarded as rigid.

Identification of an epitope in antithrombin appearing on insertion of the reactive-bond loop into the A beta-sheet

Previous work has shown that insertion of the reactive-bond loop of antithrombin into the main beta-sheet of the inhibitor, the A sheet, leads to exposure of epitopes that are not present in intact antithrombin. Identical epitopes are exposed in antithrombin-proteinase complexes, inferring that the reactive-bond loop is inserted into the A beta-sheet also in these complexes. Loop insertion thus presumably is involved in the mechanism of inhibition of target proteinases. In this work, we have identified a linear epitope in bovine antithrombin that reacts with antibodies specific for loop-inserted forms of the inhibitor. This epitope is a hexapeptide sequence comprising residues 342-347, Glu-Asp-Leu-Phe-Ser-Pro, and is located on the surface of the protein just carboxy-terminal of helix I. The Phe residue of this epitope is highly conserved in members of the serpin superfamily and appears to stabilize the region of the epitope in antithrombin and other serpins by interacting with the protein core. The conformational change involving expansion of the A beta-sheet following insertion of the reactive-bond loop is presumably transmitted through this Phe residue to the epitope region, thereby rendering this region accessible to antibodies.

Is the binding of beta-amyloid protein to antichymotrypsin in Alzheimer plaques mediated by a beta-strand insertion?

A growing body of experimental evidence demonstrates that the serpin antichymotrypsin plays a regulatory role in Alzheimer plaque physiology by interacting with the 42 residue beta-amyloid protein, and we have used molecular modeling and energy minimization techniques to study this interaction. Based on the unique plasticity of beta-sheet elements in antichymotrypsin (as well as other serpins), we conclude that the interaction of the two proteins is mediated by insertion of the N-terminus of beta-amyloid into beta-sheet C of antichymotrypsin as a pseudo-strand s1C. This beta-strand insertion requires the displacement of native antichymotrypsin strand s1C, which is known to occur partially or completely at different stages of serpin function. Thus, the association of the two proteins in vivo may be facilitated by a particular functional state of the serpin, e.g., the native or protease-complexed state.

Mechanism of heparin activation of antithrombin. Evidence for reactive center loop preinsertion with expulsion upon heparin binding

A heparin-induced conformational change is required to convert antithrombin from a slow to a fast inhibitor of factor Xa. It has been proposed [van Boeckel et al. (1994) Nat. Struct. Biol. 1, 423-425] that the reactive center residue P14 is inserted into beta-sheet A in native antithrombin and is displaced from the beta-sheet by heparin binding, thereby altering the conformation of the reactive center and making it a better target for factor Xa binding. To test this hypothesis, we have characterized a P14 serine --> tryptophan antithrombin variant. From changes in tryptophan fluorescence upon heparin binding, increased affinity for heparin, and partial activation of the variant against factor Xa, we conclude that the proposed mechanism of heparin activation is correct with respect to loop expulsion and that it may consequently be possible to create more highly activated antithrombin variants through suitable hinge region substitutions.

The structural puzzle of how serpin serine proteinase inhibitors work

Serine proteinase cleavage of proteins is essential to a wide variety of biological processes and is primarily regulated by protein inhibitors. Many inhibitors are conformationally rigid simulations of optimal serine proteinase substrates, which makes them highly efficient competitive inhibitors of target proteinases. In contrast, members of the serpin family of serine proteinase inhibitors display extensive flexibility and polymorphism, particularly in their reactive site segments and in beta-sheet secondary structure, which can take up and expel strands. Reactive site and beta-sheet polymorphism appear to be coupled in the serpins and may account for the extreme stability of serpin-proteinase complexes through the insertion of the reactive site strand into a beta-sheet. These unusual properties may have opened an adaptive pathway of proteinase regulation that was unavailable to the conformationally rigid proteinase inhibitors.

Molecular genetics of antithrombin deficiency

Antithrombin is the major proteinase inhibitor of thrombin and other blood coagulation proteinases. Antithrombin has two functional domains, a heparin binding site and a reactive centre (that complexes and inactivates the proteinase). Its deficiency results in an increased risk of venous thromboembolism. Appreciable progress has been made in recent years in understanding the structure and function of this protein, the genetic cause of inherited deficiency and its clinical consequence. The structure of antithrombin is now considered in terms of the models derived from X-ray crystallography, which have provided explanations for the function of its heparin interaction site and of its reactive loop. The structural organization of the antithrombin gene has been defined and numerous mutations have been identified that are responsible for antithrombin deficiency: these may reduce the level of the protein (Type I deficiency), alter the function of the protein (Type II deficiency, altering heparin binding or reactive sites), or even have multiple or 'pleiotropic effects' (Type II deficiency, altering both functional domains and the level of protein).

13C-NMR relation study of heparin-disaccharide interactions with tripeptides GRG and GKG

Heparin is a polydisperse sulphated copolymer consisting mostly of 1-->4 linked glucosamine and uronic acid residues, i.e. 2-deoxy-2-sulphamido-D-glucopyranose 6-sulphate and L-idopyranosyluronic acid 2-sulphate. 13C NMR has been used to study the interactions of heparinase-derived and purified heparin disaccharide with N- and C-terminally-blocked tripeptides GRG and GKG. Titration of the disaccharide with peptide indicates that GRG binds the disaccharide more strongly than does GKG, with interactions in either case being stronger at uronate ring positions. In the presence of GRG, a carboxylate pKa depression suggests electrostatic interactions between the arginine guanidinium group and the uronate carboxylate group. 13C relaxation data have been acquired for all disaccharide and peptide carbons in the presence and absence of GRG and GKG. 13C relaxation rates for the disaccharide are significantly faster in the presence of peptide, especially with GRG. Analysis of these relaxation data has been done in terms of molecular diffusion constants, D [symbol: see text] and D parallel, and an angle alpha between D parallel and a molecular frame defined by the moment of inertia tensor calculated for an internally rigid disaccharide. Disaccharide conformational space in these calculations has been sampled for both uronate half-chair forms (2H1 and 1H2) and over a range of glycosidic bond angles defined by motional order parameters and inter-residue nuclear Overhauser effects (+/- 30 degree from the average). In the absence of peptide, the ratio D [symbol: see text] /D parallel falls between 0.4 and 0.7; therefore molecular diffusion occurs preferentially about D parallel, which runs through both disaccharide rings. In the presence of peptide, D [symbol: see text] /D parallel is decreased, indicating that GRG is oriented along D parallel and proximal to the uronic acid ring. A model for this is shown.

Significance of secondary structure predictions on the reactive center loop region of serpins: a model for the folding of serpins into a metastable state

To address how serpins might fold so as to adopt the mechanistically required metastable conformation we have compared the predicted secondary structures of the reactive center loops (RCLs) of a large number of serpins with those of the equivalent regions of other non-serpin protein proteinase inhibitors. Whereas the RCLs of non-serpin inhibitors are predicted to be loop or beta-strand, those of inhibitory serpins are strongly predicted to be alpha-helical. However, non-inhibitory serpins, which also adopt the metastable conformation, show no consistent preference for alpha-helix. We propose that the RCL primary structure plays little role in promoting the metastable serpin conformation. Instead we hypothesize that preference for the metastable state results from the incorporation of part of the RCL into beta-sheet C, which as a consequence precludes incorporation of the RCL into beta-sheet A to give the most stable conformation. Consequently the RCL must be exposed and by default will adopt the most stable conformation in this particular context, which is likely to be an alpha-helix irrespective of the primary structure. Thus the observed correlation between inhibitory properties in serpins and prediction of alpha-helix in the RCL may instead reflect a need for alanine residues between positions P12 and P9 for functioning as an inhibitor rather than a structural or mechanistic requirement for alpha-helix.

Probing serpin reactive-loop conformations by proteolytic cleavage

Several crystal structures of intact members of the serine proteinase inhibitor (or serpin) superfamily have recently been solved but the relationship of their reactive-loop conformations to those of circulating forms remains unclear. Here we examine reactive-loop conformational changes of anti-trypsin and anti-thrombin by using limited proteolysis and binary complex formation with synthetic homologous reactive-loop peptides. Proteolysis at the P10-P9, P8-P7 and P7-P6 of anti-trypsin was distorted by binary complex formation. The P1'-P2' bond in anti-thrombin was more accessible to proteolysis after binary complex formation, whereas cleavage at the P4-P3 bond was variably altered by synthetic peptide insertion. The proteolytic accessibility of the reactive-site P1-P1' bond of anti-trypsin and anti-thrombin binary complexes was identical with that of the native form and no cleavage was observed in the hinge region (P15-P10) of either protein, whether native or as binary complexes. these results fit with the proposal that the hydrophobic reactive loop of serpins adopts a modified helical conformation in the circulation, with the hinge region being partly incorporated into the A beta-pleated sheet. This loop can be displaced by peptides and induced to adopt a new conformation similar to the three-turn helix of ovalbumin. Both the native and binary complexed forms of anti-thrombin showed a greatly increased proteolytic sensitivity in the presence of heparin, indicating that heparin either induces a conformational change in the local structure of the helical reactive loop or facilitates the approximation of enzyme and inhibitor.

Molecular genetics of human antithrombin deficiency

Human antithrombin is the major plasma inhibitor of thrombin both in the presence and absence of heparin. Its physiological importance is emphasised by the recurrent thromboses that individuals with a deficient or functionally abnormal protein are prone to develop. Such deficiencies are estimated to affect as many as 1:630 of the general population and between 3% and 5% of patients with thrombotic disease. The gene for antithrombin (AT3) has been cloned and shown to map to the long arm of chromosome 1 at 1q23-25. The gene consists of seven exons and six introns and spans 13,477bp of DNA. Advances in molecular genetic techniques have facilitated identification of the underlying DNA mutation(s) in > 80 families with antithrombin deficiency. Such work has proved invaluable in structure-function studies and in helping to provide informed genetic counselling to "at-risk" individuals based upon the natural history of similar variants.

The biostructural pathology of the serpins: critical function of sheet opening mechanism

The serpins illustrate the way in which the study of a protein family as a whole can clarify the functions of its individual members. Although the individual serpins have become remarkably diversified by evolution they all share a common structural pathology. We have previously shown how plotting of the dysfunctional natural mutations of the serpins on a template structure defines the domains controlling the mobility of the reactive centre loop of the molecule. Here we compare these natural mutations with reciprocal mutations in recombinants that restore the inhibitory stability of a labile member of the family, plasminogen activator inhibitor-1 (PAI-1). The combined results emphasise the critical part played by residues involved in the sliding movement that opens the A-sheet to allow reactive loop insertion. It is concluded that changes in these residues provide the prime explanation for the ready conversion of PAI-1 to the inactive latent state. The consistency of the overall results gives confidence in predicting the likely consequences of mutations in individual serpins. In particular the two common polymorphic mutations present in human angiotensinogen are likely to affect molecular stability and hence may be contributory factors to the observed association with vascular disease.

Heparin binding to platelet factor-4. An NMR and site-directed mutagenesis study: arginine residues are crucial for binding

Native platelet factor-4 (PF4) is an asymmetrically associated, homo-tetrameric protein (70 residues/subunit) known for binding polysulphated glycosaminoglycans like heparin. PF4 N-terminal chimeric mutant M2 (PF4-M2), on the other hand, forms symmetric tetramers [Mayo, Roongta, Ilyina, Milius, Barker, Quinlan, La Rosa and Daly (1995) Biochemistry 34, 11399-11409] making NMR studies with this 32 kDa protein tractable. PF4-M2, moreover, binds heparin with a similar affinity to that of native PF4. NMR data presented here indicate that heparin (9000 Da cut-off) binding to PF4-M2, while not perturbing the overall structure of the protein, does perturb specific side-chain proton resonances which map to spatially related residues within a ring of positively charged side chains on the surface of tetrameric PF4-M2. Contrary to PF4-heparin binding models which centre around C-terminal alpha-helix lysines, this study indicates that a loop containing Arg-20, Arg-22, His-23 and Thr-25, as well as Lys-46 and Arg-49, are even more affected by heparin binding. Site-directed mutagenesis and heparin binding data support these NMR findings by indicating that arginines more than C-terminal lysines, are crucial to the heparin binding process.

Heparin binding by fibronectin module III-13 involves six discontinuous basic residues brought together to form a cationic cradle

The thirteenth type III domain of fibronectin binds heparin almost as well as fibronectin itself and contains a so-called heparin-binding consensus sequence, Arg6-Arg7-Ala8-Arg9 (residues 1697-1700 in plasma fibronectin). Barkalow and Schwarzbauer (Barkalow, F.J., and Schwarzbauer, J.E. (1991) J. Biol. Chem. 266, 7812-7818) showed that mutation of Arg6-Arg7 in domain III-13 of recombinant truncated fibronectins abolished their ability to bind heparin-Sepharose. However, synthetic peptides containing this sequence have negligible affinity for heparin (Ingham, K.C., Brew, S.A., Migliorini, M. M., and Busby, T.F. (1993) Biochemistry 32, 12548-12553). We generated a three-dimensional model of fibronectin type III-13 based on the structure of a homologous domain from tenascin. The model places Arg23, Lys25, and Arg54 parallel to and in close proximity to the Arg6-Arg7-Ala8-Arg9 motif, suggesting that these residues may also contribute to the heparin-binding site. Domain III-13 and six single-site mutants containing Ser in place of each of the above-mentioned basic residues were expressed in Escherichia coli. All of the purified mutant domains melted reversibly with a Tm near that of the wild type indicating that they were correctly folded. When fluorescein-labeled heparin was titrated at physiological ionic strength, the wild type domain increased the anisotropy in a hyperbolic fashion with a Kd of 5-7 microM, close to that of the natural domain obtained by proteolysis of fibronectin. The R54S mutant bound 3-fold weaker and the remaining mutants bound at least 10-fold weaker than wild type. The results point out that the Arg6-Arg7-Ala8-Arg9 consensus sequence by itself has little affinity for heparin under physiological conditions, even when presented in the context of a folded domain. Thus, the heparin-binding site in fibronectin is more complex than previously realized. It is formed by a cluster of 6 positively charged residues that are remote in the sequence but brought together on one side of domain III-13 to form a "cationic cradle" into which the anionic heparin molecule could fit.

alpha 1-Antitrypsin Mmalton (Phe52-deleted) forms loop-sheet polymers in vivo. Evidence for the C sheet mechanism of polymerization

The Z (Glu342-->Lys) and Siiyama (Ser53-->Phe) deficiency variants of alpha 1-antitrypsin result in the retention of protein in the endoplasmic reticulum of the hepatocyte by loop-sheet polymerization in which the reactive center loop of one molecule is inserted into a beta-pleated sheet of a second. We show here that antitrypsin Mmalton (Phe52-deleted), which is associated with the same liver inclusions, is also retained at an endoglycosidase H-sensitive stage of processing in the Xenopus oocyte and spontaneously forms polymers in vivo. These polymers, obtained from the plasma of an Mmalton/QO (null) bolton heterozygote, were much shorter than other antitrypsin polymers and contained a reactive center loop-cleaved species. Monomeric mutant antitrypsin was also isolated from the plasma. The monomeric component had a normal unfolding transition on transverse urea gradient gel electrophoresis and formed polymers in vitro more readily than M, but less readily than Z, antitrypsin. The A beta-sheet accommodated a reactive center loop peptide much less readily than Z antitrypsin, which in turn was less receptive than native M antitrypsin. The nonreceptive conformation of the A sheet in antitrypsin Mmalton had little effect on kinetic parameters, the formation of SDS-stable complexes, the S to R transition, and the formation of the latent conformation. Comparison of the results with similar findings of short chain polymers associated with the antithrombin variant Rouen VI (Bruce, D., Perry, D., Borg, J.-Y., Carrell, R. W., and Wardell, M. R. (1994) J. Clin. Invest. 94, 2265-2274) suggests that polymerization is more complicated than the mechanism proposed earlier. The Z, Siiyama, and Mmalton mutations favor a conformational change in the antitrypsin molecule to an intermediate between the native and latent forms. This would involve a partial overinsertion of the reactive loop into the A sheet with displacement of strand 1C and consequent loop-C sheet polymerization.

Structural basis for serpin inhibitor activity

The mechanism of formation and the structures of serpin-inhibitor complexes are not completely understood, despite detailed knowledge of the structures of a number of cleaved and uncleaved inhibitor, noninhibitor, and latent serpins. It has been proposed from comparison of inhibitor and noninhibitor serpins in the cleaved and uncleaved forms that insertion of strand s4A into preexisting beta-sheet A is a requirement for serpin inhibitor activity. We have investigated the role of this strand in formation of serpin-proteinase complexes and in serpin inhibitor activity through homology modeling of wild type inhibitor, mutant substrate, and latent serpins, and of putative serpin-proteinase complexes. These models explain the high stability of the complexes and provide an understanding of substrate behavior in serpins with point mutations in s4A and of latency in plasminogen activator inhibitor I.

Identification of angiotensinogen and complement C3dg as novel proteins binding the proform of eosinophil major basic protein in human pregnancy serum and plasma

In sera from pregnant women, pregnancy-associated plasma protein-A (PAPP-A) circulates as a disulfide-bound complex (approximately 474 kDa) with the proform of eosinophil major basic protein (proMBP) (Oxvig, C., Sand, O., Kristensen, T., Gleich, G. J., and Sottrup-Jensen, L. (1993) J. Biol. Chem. 268, 12243-12246). We have produced monoclonal antibodies (mAbs) against the PAPP-A.proMBP complex and established a radioimmunoassay utilizing a mAb recognizing the PAPP-A subunit. Surprisingly, serum levels of proMBP exceed those of PAPP-A four to 10-fold on a molar basis throughout pregnancy. This result prompted an investigation of the status of proMBP in pregnancy. Using a proMBP-specific mAb two novel proMBP complexes have been isolated by chromatographic techniques. Based on sequence analysis, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and reaction with specific antibodies, one is shown to be a 2:2 disulfide-bound complex (approximately 200 kDa) between proMBP and angiotensinogen. The other is a 2:2:2 complex (approximately 300 kDa) between proMBP, angiotensinogen, and complement C3dg. Circulating proMBP in pregnancy is thus present in three types of complexes. These results suggest that specific interactions between the complexed proteins occur in pregnancy, and the possibility is raised that their interactions are important in the pathophysiology of pregnancies associated with hypertension.

Inhibitory mechanism of serpins. Interaction of thrombin with antithrombin and protease nexin 1

The mechanism for the inhibition of thrombin by the serpins antithrombin and protease nexin 1 has been investigated using several kinetic techniques at pH 7.9 and 37 degrees C with an ionic strength of 0.3 M. Rapid kinetic studies demonstrated that a two-step mechanism for the formation of the stable thrombin-serpin complex applied to both serpins. The inhibition constant for the initial thrombin-antithrombin complex was 265 microM, and the rate constant for the conversion of this complex to the final one was 3.9 s-1; the corresponding values for PN1 were 3.4 microM and 6.0 s-1. By using slow-binding kinetics, it was possible to obtain estimates of the second-order rate constants for the formation of the stable thrombin-serpin complexes (1.2 x 10(4) and 1.5 x 10(6) M-1 s-1 for antithrombin and protease nexin 1, respectively) and the dissociation constants for these complexes (< 1 nM for both serpins). The influence of viscosity on the reactions indicated that the rate of interaction of both serpins with thrombin was diffusion-controlled. Moreover, the results indicated that the initial complex reacted more rapidly to form the stable complex than it dissociated to free enzyme and inhibitor; i.e., the behavior of the serpins was analogous to that of "sticky" substrates. By using the results from slow-binding, viscosity, and rapid kinetic studies, it was possible to set values for all of the rate constants for the interactions of antithrombin and protease nexin 1 with thrombin.(ABSTRACT TRUNCATED AT 250 WORDS)

Preparation and characterization of latent alpha 1-antitrypsin

Members of the serine proteinase inhibitor or serpin superfamily have a common molecular architecture based on a dominant five-membered A beta-pleated sheet and a mobile reactive center loop. The reactive center loop has been shown to adopt a range of conformations from the three turn alpha-helix of ovalbumin to the cleaved or latent inhibitor in which the reactive center loop is fully inserted into the A sheet of the molecule. While the cleaved state can be achieved in all inhibitory serpins only plasminogen activator inhibitor-1 and, more recently, antithrombin have been shown to adopt the latent conformation. We show here that the archetypal serpin, alpha 1-antitrypsin, can also be induced to adopt the latent conformation by heating at high temperatures in 0.7 M citrate for 12 h. The resulting species elutes at a lower sodium chloride concentration on an anion-exchange column and has a more cathodal electrophoretic mobility on non-denaturing polyacrylamide gel electrophoresis and isoelectric focusing than native M antitrypsin. Latent antitrypsin is inactive as an inhibitor of bovine alpha-chymotrypsin, is stable to unfolding with 8 M urea, and is more resistant to heat-induced loop-sheet polymerization than native but less resistant than cleaved antitrypsin. The reactive center loop of latent antitrypsin is inaccessible to proteolytic cleavage, and its occupancy of the A sheet prevents the molecule accepting an exogenous reactive center loop peptide. The activity of latent antitrypsin may be increased from < 1% to approximately 35% by refolding from 6 M guanidinium chloride.

Enzymatic degradation of glycosaminoglycans

Glycosaminoglycans (GAGs) play an intricate role in the extracellular matrix (ECM), not only as soluble components and polyelectrolytes, but also by specific interactions with growth factors and other transient components of the ECM. Modifications of GAG chains, such as isomerization, sulfation, and acetylation, generate the chemical specificity of GAGs. GAGs can be depolymerized enzymatically either by eliminative cleavage with lyases (EC 4.2.2.-) or by hydrolytic cleavage with hydrolases (EC 3.2.1.-). Often, these enzymes are specific for residues in the polysaccharide chain with certain modifications. As such, the enzymes can serve as tools for studying the physiological effect of residue modifications and as models at the molecular level of protein-GAG recognition. This review examines the structure of the substrates, the properties of enzymatic degradation, and the enzyme substrate-interactions at a molecular level. The primary structure of several GAGs is organized macroscopically by segregation into alternating blocks of specific sulfation patterns and microscopically by formation of oligosaccharide sequences with specific binding functions. Among GAGs, considerable dermatan sulfate, heparin and heparan sulfate show conformational flexibility in solution. They elicit sequence-specific interactions with enzymes that degrade them, as well as with other proteins, however, the effect of conformational flexibility on protein-GAG interactions is not clear. Recent findings have established empirical rules of substrate specificity and elucidated molecular mechanisms of enzyme-substrate interactions for enzymes that degrade GAGs. Here we propose that local formation of polysaccharide secondary structure is determined by the immediate sequence environment within the GAG polymer, and that this secondary structure, in turn, governs the binding and catalytic interactions between proteins and GAGs.

Coagulation factors and their inhibitors

A comprehensive three-dimensional picture of the coagulation process is beginning to emerge. Crystallographic structure determinations of prothrombin, factor Xa, factor IXa, tissue factor and factor XIII represent important advances in our understanding of the coagulation cascade. Similarly, structures of antithrombin, tissue factor pathway inhibitor and thrombomodulin provide details of endogenous anticoagulatory mechanisms. NMR spectroscopy of multiple domains of coagulation proteins represents an important contribution to the analysis of flexibility and rigidity of modular proteins. Thrombin, as the prime candidate for antithrombotic drug design, continues to be an object of intense efforts in applied crystallography.

Thromboembolic disease due to thermolabile conformational changes of antithrombin Rouen-VI (187 Asn-->Asp)

A new variant of antithrombin (Rouen-VI, 187 Asn-->Asp) with increased heparin affinity was shown to have normal inhibitory activity which decreased slowly at 4 degrees C and rapidly at 41 degrees C. On electrophoresis the freshly isolated variant had an anodal shift relative to native antithrombin due to the mutation. A further anodal transition occurred after either prolonged storage at 4 degrees C or incubation at 41 degrees C due to the formation of a new inactive uncleaved component with properties characteristic of L-form (latent) antithrombin. At the same time, polymerization also occurred with a predominance of di-, tri-, and tetra-mers. These findings fit with the observed mutation of the conserved asparagine (187) in the F-helix destabilizing the underlying A-sheet of the molecule. Evidence of A-sheet perturbation is provided by the increased rate of peptide insertion into the A-sheet and by the decreased vulnerability of the reactive loop to proteolysis. The spontaneous formation of both L-antithrombin and polymers is consistent with our crystal structure of intact antithrombin where L-form and active antithrombin are linked together as dimers. The nature of this linkage favors a mechanism of polymerization whereby the opening of the A-sheet, to give incorporation of the reactive center loop, is accompanied by the bonding of the loop of one molecule to the C-sheet of the next. The accelerated lability of antithrombin Rouen-VI at 41 versus 37 degrees C provides an explanation for the clinical observation that episodes of thrombosis were preceded by unrelated pyrexias.

Three novel mutations of antithrombin inducing high-molecular-mass compounds

We have identified three novel mutations of the antithrombin (AT) gene in patients with thrombotic complications: a Cys 128 --> Tyr mutations, a G --> A mutation in the intervening sequence 4 (IVS4) 14 nucleotide 5' to exon 5, and a 9 bp deletion in the 3' end of exon 6 resulting in a short aberrant sequence after Arg 425. The latter mutation was associated with an Arg 47 --> His mutation in two compound heterozygous brothers. These three mutations led to the expression in the circulation of small amounts of inactive molecules with a high molecular mass in immunoblot analysis. In reducing conditions, these variant molecules had a normal molecular mass, which led us to postulate that these mutations prevent the formation of one intramolecular disulfide bond and allow the formation of intermolecular disulfide bonds. Plasma from a heterozygous patients bearing the Cys 128 --> Tyr mutation and from a compound heterozygote bearing the Arg 47 --> His mutation and the 9 bp deletion in exon 6 were passed through a heparin-sepharose column. In both cases a population of high-molecular-weight AT molecules with no binding affinity and no AT activity was separated from a population of normal molecules in the first patient, together with a population of molecules with a reduced binding affinity for heparin due to the substitution of Arg 47, in the compound heterozygote. The common feature of these three mutations is that they lead to partial misfolding and to the formation of intermolecular disulfide bonds with other plasma components, inducing the pleiotropic phenotypes observed.

Lysine-heparin interactions in antithrombin. Properties of K125M and K290M,K294M,K297M variants

Lysine residues in two different regions of antithrombin have been proposed to be involved in heparin binding and heparin-mediated acceleration of proteinase inhibition. Lysine 125 has been implicated as an essential heparin binding residue from chemical modification studies [Peterson, C. B., Noyes, C. M., Pecon, J. M., Church, F. C., & Blackburn, M. N. (1987) J. Biol. Chem. 262, 8061-8065] whereas lysines 290, 294, and 297 have been proposed from model building studies to constitute the heparin binding site [Villanueva, G. B. (1984) J. Biol. Chem. 259, 2531-2536]. To evaluate both of these proposals, we have prepared two variant human antithrombins, K125M and K290M,K294M,K297M, in which these lysines have been changed by site-directed mutagenesis to methionines. The K290M,K294M,K297M variant had properties very similar to those of wild-type recombinant antithrombin in affinity for heparin, and in rates of inhibition of thrombin and factor Xa. In contrast, K125M antithrombin had reduced affinity for both heparin pentasaccharide and full-length heparin, corresponding to delta delta Gs of 3.1 and 2.0 kcal mol-1, respectively. However, this variant was still able to inhibit both thrombin and factor Xa. Whereas the rate of thrombin inhibition was similar to that of wild-type antithrombin, the rate of factor Xa inhibition was enhanced between 2- and 3-fold, suggesting a role for lysine 125 in the allosteric coupling between the heparin binding site and the reactive center region. At saturation with either heparin pentasaccharide or full-length high-affinity heparin, the rates of inhibition of both proteinases were similar to those of wild-type antithrombin for both the K125M and K290M,K294M,K297M variants.(ABSTRACT TRUNCATED AT 250 WORDS)

Biological implications of a 3 A structure of dimeric antithrombin

BACKGROUND

Antithrombin, a member of the serpin family of inhibitors, controls coagulation in human plasma by forming complexes with thrombin and other coagulation proteases in a process greatly accelerated by heparin. The structures of several serpins have been determined but not in their active conformations. We have determined the structure of intact antithrombin in order to study its mechanism of activation, particularly with respect to heparin, and the dysfunctions of this mechanism that predispose individuals to thrombotic disease.

RESULTS

The crystal structure of a dimer of one active and one inactive molecule of antithrombin has been determined at 3 A. The first molecule has its reactive-centre loop in a predicted active conformation compatible with initial entry of two residues into the main beta-sheet of the molecule. The inactive molecule has a totally incorporated loop as in latent plasminogen activator inhibitor-1. The two molecules are linked by the reactive loop of the active molecule which has replaced a strand from another beta-sheet in the latent molecule.

CONCLUSION

The structure, together with identified mutations affecting its heparin affinity, allows the placement of the heparin-binding site on the molecule. The conformation of the two forms of antithrombin demonstrates the extraordinary mobility of the reactive loop in the serpins and provides insights into the folding of the loop required for inhibitory activity together with the potential modification of this by heparin. The mechanism of dimerization is relevant to the polymerization that is observed in diseases associated with variant serpins.

Crystal structure of an uncleaved serpin reveals the conformation of an inhibitory reactive loop

The three-dimensional structure of an uncleaved serpin, a variant of human antichymotrypsin engineered to be an inhibitor of human neutrophil elastase, has been determined by X-ray crystallographic methods and is currently being refined at 2.5 A resolution. It contains an intact reactive loop in a distorted helical conformation. A comparison of the current model with that of its cleaved counterpart suggests that the conformational 'stress' of the serpin in its uncleaved and uncomplexed state may not be confined solely to the reactive loop or beta-sheet A. It is intriguing that strand s4A is not pre-inserted into beta-sheet A of the native serpin, and this has profound implications for the mechanism of serpin function.

Antithrombin and its inherited deficiencies

Human antithrombin is the major inhibitor of the coagulation serine proteases accounting for approximately 80% of the thrombin inhibitory activity of plasma. It is a member of the serpin family of serine protease inhibitors and in common with some other members of this family it undergoes a dramatic increase in its inhibitory activity in the presence of heparin and other sulphated glycosaminoglycans. Two functional domains in antithrombin are recognised, the reactive site domain which interacts with the active site serine residue of the protease and the heparin binding domain. The gene for antithrombin has been cloned and its entire nucleotide sequence determined. A deficiency or functional abnormality of antithrombin may result in an increased risk of thromboembolic disease. Such deficiencies are estimated to affect as many as 1:300 of the general population and 3 to 5% of patients with thrombotic disease. On the basis of functional and immunological antithrombin assays, antithrombin deficiency may be subdivided into Types I and II. Type I disease is due to a wide variety of heterogeneous DNA mutations whilst in Type II disease missense mutations leading to single amino acid substitutions have been identified in all cases. Clinically, Type I antithrombin deficiency is associated with recurrent thromboembolic disease whereas in Type II deficiency the risk of thrombosis is closely related to the position of the mutation within the protein. Thus, heterozygotes with mutations within the heparin binding domain of antithrombin have a relatively low risk of thrombosis compared to those with mutations at or close to the reactive site of the molecule.

The intact and cleaved human antithrombin III complex as a model for serpin-proteinase interactions

Antithrombin is a member of the serine proteinase inhibitor (serpin) family which contain a flexible reactive site loop that interacts with, and is cleaved by the target proteinase. In cleaved and latent serpins, the reactive site loop is inserted into a large central beta-sheet in the same molecule, whereas in ovalbumin, a nonfunctional serpin, the reactive site loop is completely exposed and in an alpha-helical conformation. However, in neither conformation can the reactive site loop bind to target proteinases. Here we report the structure of an intact and cleaved human antithrombin complex. The intact reactive site loop is in a novel conformation that seems well suited for interaction with proteinases such as thrombin and blood coagulation factor Xa.

Regulation of proteolytic activity in tissues

Degradation of tissue proteins is controlled by multiple means. These include regulation of the synthesis of proteinases, activation of the zymogen forms, the activity of the mature proteinase, and the degradation of these enzymes and the substrates. Mature proteinases can be controlled by pH, calcium ions, ATP, lipids and the formation of complexes with other proteinases, proteoglycans, and inhibitors.