Properdin binds to sulfatide [Gal(3-SO4)beta 1-1 Cer] and has a sequence homology with other proteins that bind sulfated glycoconjugates

Properdin, which stabilizes the C3 convertase during the activation of the alternate complement pathway, contains amino acid sequence homologies with several proteins that bind sulfated glycoconjugates, including the adhesive protein thrombospondin and the leech salivary protein antistasin. This homology is based around the sequence Cys-Ser-Val-Thr-Cys-Gly-X-Gly-X-X-X-Arg-X-Arg. To determine if these homologous amino acid sequences are sulfated glycoconjugate-binding domains, purified native properdin, as well as activated properdin (a high molecular weight form of properdin), were examined for binding to various lipids in solid phase radioimmunoassays. Of the lipids tested, both native and activated properdin bind with high affinity only to sulfatide [Gal(3-SO4)beta 1-1 Cer], but not to comparable levels of cholesterol-3-SO4, or several neutral glycolipids, gangliosides, and phospholipids. Sulfatide binding by both forms of properdin is inhibited by dextran sulfate (Mr = 500,000) or fucoidan, whereas only the activated form is inhibited by dextran sulfate (Mr = 5,000) or heparin. Comparable levels of chondroitin sulfates A, B, and C, keratan sulfate, dextran (Mr = 90,000), or hyaluronic acid do not inhibit binding. Taken together, these data suggest that properdin, like antistasin and thrombospondin, binds sulfated glycoconjugates and supports the conclusion that the homologous sequences are sulfated glycoconjugate-binding domains.     

Subunit interactions in hybrids of native, carboxypeptidase-treated and citraconylated rabbit muscle aldolase

1. The kinetic properties of hybrids of native (or carboxypeptidase-treated) and citraconylated rabbit muscle aldolase are compared with those of equivalent mixtures of the parental enzymes. 2. In the hybrids, the native subunits function slightly less well than in the homotetramer, but the citraconylated subunits have enhanced activity. 3. Subunits of carboxypeptidase-treated aldolase behave essentially as expected in a hybrid environment, but the citraconylated subunits do not show the same enhancement of activity found in the hybrids of native and citraconylated enzyme. The apparent affinity for fructose 1,6-diphosphate of the citraconylated subunits in hybrids of carboxypeptidase-treated and citraconylated aldolase is increased. 4. These results are interpreted in terms of a substrate-induced conformational difference between native and carboxypeptidase-treated aldolase. 5. This conformational change can take place within a single native subunit in the hybrids and does not require a similar conformational change to occur simultaneously in the other three subunits.     

The involvement of proteolysis in conformational stability of the carbamoyl-phosphate synthetase/aspartate carbamoyltransferase enzyme of Neurospora crassa

The pyrimidine-3 gene of Neurospora crassa codes for a bifunctional enzyme catalysing the first two steps of the pyrimidine biosynthetic pathway. Difficulties have been experienced in purification due to the lability of the enzyme. The enzyme loses carbamoyl-phosphate synthetase (carbon-dioxide: ammonia ligase (ADP-forming, carbamate-phosphorylating), EC 6.3.4.16) activity and undergoes a change in apparent molecular weight from the native 650,000 to 100,000 of the only detectable fragment. Attempts have been made therefore to stabilize the enzyme so as to minimise these effects. Elastinal, a protease inhibitor, reduces the effects, as do certain ultraviolet-sensitive mutant strains which lack a minor protease. The nature of the loss of carbamoyl-phosphate synthetase suggests an instability in the tertiary structure of the enzyme which can be reduced by the use of glycerol. Glycerol also exhibits a protease-inhibiting effect in this system. Although a range of protease inhibtors and use of uvs mutants can reduce the rate of decay of carbamoyl-phosphate synthetase activity, only glycerol can stabilize the native molecular weight. Our results support the hypothesis that the loss of carbamoyl-phosphate synthetase activity and change in molecular weight of the enzyme is a three-step sequence of proteolysis, conformational shift and cleavage of a further non-covalent bond.     

Kinetic evidence that allosteric activation of antithrombin by heparin is mediated by two sequential conformational changes

The serpin, antithrombin, requires allosteric activation by a sequence-specific pentasaccharide unit of heparin or heparan sulfate glycosaminoglycans to function as an anticoagulant regulator of blood clotting proteases. Surprisingly, X-ray structures have shown that the pentasaccharide produces similar induced-fit changes in the heparin binding site of native and latent antithrombin despite large differences in the heparin affinity and global conformation of these two forms. Here we present kinetic evidence for similar induced-fit mechanisms of pentasaccharide binding to native and latent antithrombins and kinetic simulations which together support a three-step mechanism of allosteric activation of native antithrombin involving two successive conformational changes. Equilibrium binding studies of pentasaccharide interactions with native and latent antithrombins and the salt dependence of these interactions suggest that each conformational change is associated with distinct spectroscopic changes and is driven by a progressively better fit of the pentasaccharide in the binding site. The observation that variant antithrombins that cannot undergo the second conformational change bind the pentasaccharide like latent antithrombin and are partially activated suggests that both conformational changes contribute to allosteric activation, in agreement with a recently proposed model of allosteric activation.     

Heat-labile enterotoxin crystal forms with variable A/B5 orientation. Analysis of conformational flexibility.

A new native crystal form of heat-labile enterotoxin (LT) has two AB5 complexes in the asymmetric unit with different orientations of the A subunit with respect to the B pentamer. Comparison with other crystal forms of LT shows that there is considerable conformational freedom for orientating the A subunit with respect to the B pentamer. The rotations of A in different crystal forms do not follow one specific axis, but most of them share a hinge point, close to the main interaction area between A and B5. Analysis of the two high-resolution structures available shows that these rotations cause very little change in the actual interactions between A and B5.     

Analysis of the natural polymeric forms of human properdin and their functions in complement activation

Many of the anomalies observed in studies or properdin may be explained on the basis of its ability to form a series of multi-subunit polymers and by differences in the functions of these forms of properdin. Dimers (P2), trimers (P3), tetramers (P4), and higher Mr polymers (Pn) of the 46,000-Da subunit were separated by gel filtration or by cation exchange chromatography of purified properdin. The specific activity of each form was measured in two assays. The native properdin activity of P4 was 10 times that of P2 (on a molar basis) with the order: P4 greater than P3 greater than P2 greater than Pn. During C activation P4 was found to be consumed first, P3 second, and P2 last, consistent with their measured specific activities. Assays for activated properdin showed that only Pn caused fluid phase C consumption when incubated in serum at normal concentrations. Pn accumulated during long term storage of purified P and freezing rapidly converted the smaller oligomers to Pn. The isolated oligomers were extremely stable, but did redistribute after denaturation-renaturation cycles by using low pH or guanidine. Renaturation after exposure of any species to denaturing conditions yielded mixtures of 20:54:26 (P4:P3:P2). This distribution was almost identical to that found in fresh normal human serum or plasma, suggesting that a distinct distribution of oligomers exists in blood that provides the C system with an apparently advantageous range of specific activities.     

Biochemical characterization of murine leukaemia inhibitory factor produced by Krebs ascites and by yeast cells

A cDNA library was constructed using mRNA from Krebs ascites tumor cells that was shown by Northern blot hybridization to contain mRNA for murine leukemia inhibitory factor (LIF). This library was screened with an oligonucleotide corresponding to the 3' end of a partial LIF cDNA clone, and an overlapping cDNA clone isolated. Nucleotide sequence analysis of this latter clone allowed the complete sequence of LIF to be derived. A cDNA molecule encoding the entire mature LIF protein was installed in a yeast expression vector, and LIF produced up to about 100 ng/ml in the growth medium. The LIF produced by yeast cells has the same biologic properties as native LIF and competes with native 125I-LIF for binding to specific cellular receptors. Two forms of native LIF, distinguishable by their chromatographic behavior on DEAE-Sepharose, were converted by neuraminidase treatment to a form with similar chromatographic behavior, suggesting that the major difference between these two species is the content of sialic acid on the carbohydrate portion. Moreover, yeast-derived recombinant LIF appears to display a different pattern of glycosylation to both forms of native LIF. From in vitro experiments, we conclude that the nature of the glycosylation is not crucial to biologic activity.     

The role of retinal in the thermal stability of the purple membrane.

Differential scanning calorimetry demonstrates that the bleached form of the purple membrane does not possess any measurable thermal transition in water, up to 105 degrees C, whereas in 0.1 M phosphate pH 7.5 it shows a transition at about 82 degrees C, with an enthalpy of 110 kJ/mol. In the latter medium, the native membrane shows the main transition at 97 degrees C, with an enthalpy of 390 kJ/mol. The reduced form of the purple membrane shows two small transitions in water, as well as in 0.1 M phosphate, which do not seem to be related to the main thermal transition of the native membrane. Fourier-transform infrared spectra in D2O show that the two modified samples, as well as the native one, undergo similar secondary structural changes upon thermal denaturation. These changes appear to extend through a wide temperature range for both modified forms, particularly for the bleached one. The results suggest that the main thermal transition in the purple membrane is due to a cooperative conformational change involving the disruption of the network of electrostatic and hydrogen-bonding interactions which originate from the protonated Schiff base. In the two modified membranes, these conformational changes appear to proceed smoothly through a rather low or non-cooperative process. The thermal behaviour of the bleached membrane in water resembles that of the molten globule state described for several globular proteins.     

Purification of prostate-specific membrane antigen using conformational epitope-specific antibody-affinity chromatography

Prostate-specific membrane antigen (PSMA) is a type II membrane protein that has attracted significant attention as a target for immunioscintigraphic and radioimmunotherapeutic applications for prostate cancer. However, definitive studies on its substrate and inhibitor specificity as well as protein-protein interactions have been somewhat limited by difficulties in the purification of native PSMA. In this study, we optimized the purification of native PSMA from LNCaP cells using conformational epitope-specific antibody-affinity chromatography. Western blot analysis and an HPLC-based enzymatic activity assay were used to compare the yield and activity of PSMA purified by different methods. The ratio of purified PSMA in a native and active conformation was determined by quantifying the amount of non-native PSMA not retained in a second antibody-affinity isolation. The addition of both a neutralization step and the inclusion of Zn(2+) to the equilibration buffer in desalting step provides considerable enhancement in the yield of active PSMA from LNCaP cells.     

Purification of a 31,000-dalton insulin-like growth factor binding protein from human amniotic fluid. Isolation of two forms with different biologic actions

Human amniotic fluid has been shown to contain a protein that binds insulin-like growth factor I and II (IGF-I and IGF-II). Partially purified preparations of this protein have been reported to inhibit the biologic actions of the IGFs. In these studies our laboratory has used a modified purification procedure to obtain a homogeneous preparation of this protein as determined by polyacrylamide gel electrophoresis and amino acid sequence analysis. During purification the ion exchange chromatography step resulted in two peaks of material with IGF binding activity termed peaks B and C. Each peak was purified separately to homogeneity. Both peaks were estimated to be 31,000 daltons by polyacrylamide gel electrophoresis and their amino acid compositions were nearly identical. Amino acid sequence analysis showed that both peaks had identical N-terminal sequences through the first 28 residues. Neither protein had detectable carbohydrate side chains and each had a similar affinity for radiolabeled IGF-I (1.7-2.2 x 10(10) liters/mol). In contrast, these two forms had marked differences in bioactivity. Concentrations of peak C material between 2 and 20 ng/ml inhibited IGF-I stimulation of [3H]thymidine incorporation into smooth muscle cell DNA. In contrast, when peak B (100 ng/ml) was incubated with IGF-I there was a 4.4-fold enhancement of stimulation of DNA synthesis. Additionally, pure peak B was shown to adhere to cell surfaces, whereas peak C was not adherent. The non-adherent peak C inhibited IGF-I binding to its receptor and to adherent peak B. We conclude that human amniotic fluid contains two forms of IGF binding protein that have very similar physiochemical characteristics but markedly different biologic actions. Since both have similar if not identical amino acid compositions, N-terminal sequences, and do not contain carbohydrate, we conclude that they differ in some other as yet undefined post-translational modification.     

Native states of adenylate kinase are two active sub-ensembles

There are two kinds of conformational forms of adenylate kinase (AK) in equilibrium in solution with different ANS-binding properties. Furthermore, the nature of AP(5)A inhibition suggests also that the native forms of AK for binding with different substrates pre-exist in the absence of substrates. In the present study, a kinetics approach was used to explore the native forms distinguished by ANS-binding properties and by the nature of AP(5)A inhibition. The results revealed that the native forms distinguished by ANS probe are two conformational sub-ensembles. Both sub-ensembles are active and consist of a series of forms, which pre-exist in solution and can bind with different substrates. The K(m) values of N(1) for AMP, ADP and MgATP are larger than that of N(2), indicating that the N(2) sub-ensemble is more specific for binding substrates. This is consistent with the previous observation that the activity of N(2) is about 1.8-fold of that of N(1).     

Circular dichroism studies of native and chemically modified Ca2+-dependent protein modulator.

The structural features of the native Ca2+-dependent protein modulator and two chemically modified derivatives, namely, nitrotyrosyl modulator and alkylated modulator, were examined by circular dichroism. The binding of Ca2+ to the native molecule was accompanied by an increase in helical content from 40 to 49%, with little effect on the local environments of aromatic residues in the modulator. The Mg2+ and Mn2+ do not elicit the conformational change induced by the binding of Ca2+, which also stabilizes the modulator against urea denaturation. The overall secondary structure of nitrotyrosyl modulator is indistinguishable from that of the native protein and undergoes a similar conformational change upon binding Ca2+. These observations are in agreement with the fact that nitration has no effect on modulator functions. Furthermore, nitrotyrosyl modulator interacts with troponin I only in the presence of Ca2+, as detected by circular dichroism (cd). On the other hand, alkylation of five methionine residues on the modulator with benzyl bromide affects protein conformation, as evidenced by a reduced helical content of only 35%. Alkylated modulator retains the ability of the native protein to bind Ca2+ although the affinity of this derivative for Ca2+ is reduced some three orders of magnitude relative to the native protein, with Kd = 3.2 X 10(-4) M. The results with the alkylated modulator, in conjunction with previous cd studies on N-chlorosuccinimide oxidized modulator are utilized to advance a model for the Ca2+ activation of modulator protein, based on three conformational states of the molecule.     

S-ovalbumin, an ovalbumin conformer with properties analogous to those of loop-inserted serpins.

Most serpins are inhibitors of serine proteinases and are thought to undergo a conformational change upon complex formation with proteinase that involves partial insertion of the reactive center loop into a beta-sheet of the inhibitor. Ovalbumin, although a serpin, is not an inhibitor of serine proteinases. It has been proposed that this deficiency arises from the presence of a charged residue, arginine, at a critical point (P14) in the reactive center region, which prevents loop insertion into the beta-sheet and thereby precludes inhibitory properties. To test whether loop insertion is prevented in ovalbumin we have examined the properties of two forms of ovalbumin: the native protein and S-ovalbumin, a form that forms spontaneously from native ovalbumin and has increased stability. Calorimetric measurements showed that S-ovalbumin was more stable than ovalbumin by about 3 kcal mol-1. CD spectra, which indicated that S-ovalbumin had less alpha-helix than native ovalbumin, and 1H NMR spectra, which indicated very similar overall structures, suggest limited conformational differences between the two forms. From comparison of the susceptibility of the reactive center region of each protein to proteolysis by porcine pancreatic elastase and by subtilisin Carlsberg, we concluded that the limited native-to-S conformational change specifically affected the reactive center region. These data are consistent with a structure for S-ovalbumin in which part of the reactive center loop has inserted into beta-sheet A to give a more stable structure, analogously to other serpins. However, the rate of loop insertion appears to be very much lower than for inhibitory serpins.(ABSTRACT TRUNCATED AT 250 WORDS)     

The dimeric and trimeric solution structures of the multidomain complement protein properdin by X-ray scattering, analytical ultracentrifugation and constrained modelling

Properdin regulates the alternative pathway of the complement system of immune defence by stabilising the C3 convertase complex. It contains six thrombospondin repeat type I (TSR-1 to TSR-6) domains and an N-terminal domain. Properdin exists as either a dimer, trimer or tetramer. In order to determine the solution structure of multiple TSR domains, the molecular structures of dimeric and trimeric properdin were studied by X-ray scattering and analytical ultracentrifugation. Guinier analyses showed that the dimer and trimer have radii of gyration R(G) values of 7.5 nm and 10.3 nm, respectively, and cross-sectional radii of gyration R(XS) values of 1.3 nm and 1.5 nm, respectively. Distance distribution functions showed that the maximum lengths of the dimer and trimer were 25 nm and 30 nm, respectively. Analytical ultracentrifugation gave sedimentation coefficients of 5.1S and 5.2S for the dimer and trimer forms, respectively. Homology models for the TSR domains were constructed using the crystal structure of the TSP-2 and TSP-3 domains in human thrombospondin as templates. Properdin could be represented by seven TSR domains, not six as believed, since the crystal structure determined for TSP-2 and TSP-3 showed that the N-terminal domain (TSR-0) could be represented by a truncated TSR domain with the same six conserved Cys residues found in TSR-1 to TSR-6. Automated constrained molecular modelling revealed the solution conformations of multiple TSR domains in properdin at medium resolution. The comparison of 3125 systematically generated conformational models for the trimer with the X-ray data showed that good curve fits could be obtained by assuming that the linker between adjacent TSR domains possessed limited flexibility. Good trimer models correspond to partially collapsed triangular structures, and extended triangular shapes do not fit the data. The corresponding 3125 models for the dimer revealed a similar outcome in which a partially collapsed TSR structure gave good fits. The models account for the effect of mutations that cause properdin deficiencies, and suggest that the biologically active TSR-4, TSR-5 and TSR-6 domains are exposed for protein-protein interactions. The role of the other TSR domains in properdin may be to act as spacers to make TSR-4, TSR-5 and TSR-6 accessible for function.     

Characterization of recombinant CEL-I, a GalNAc-specific C-type lectin, expressed in Escherichia coli using an artificial synthetic gene

CEL-I is a C-type lectin isolated from the Holothuroidea Cucumaria echinata. This lectin shows very high N-acetylgalactosamine-binding specificity. We constructed an artificial gene encoding recombinant CEL-I (rCEL-I) using a combination of synthetic oligonucleotides, and expressed it in Escherichia coli cells. Since the recombinant protein was obtained as inclusion bodies, the latter were solubilized using urea and 2-mercaptoethanol, and the protein was refolded during the purification and dialysis steps. The purified rCEL-I showed comparable hemagglutinating activity to that of native CEL-I at relatively high Ca(2+)-concentrations, whereas it was weaker at lower Ca(2+)-concentrations due to decreased Ca(2+)-binding affinity. rCEL-I exhibited similar carbohydrate-binding specificity to native CEL-I, including strong GalNAc-binding specificity, as examined by hemagglutination inhibition assay. Comparison of the far UV-CD spectra of recombinant and native CEL-I revealed that the two proteins undergo a similar conformational change upon binding of Ca(2+). Single crystals of rCEL-I were also obtained under the same conditions as those used for the native protein, suggesting that they have similar tertiary structures. Although native CEL-I exhibited strong cytotoxicity toward cultured cells, rCEL-I showed low cytotoxicity. These results indicate that rCEL-I has a tertiary structure and carbohydrate-binding specificity similar to those of native CEL-I. Howeger, there is a subtle difference in the properties between the two proteins probably due to the additional methionine residue at the N-terminus of rCEL-I.     

Substitutions in a flexible loop of horse liver alcohol dehydrogenase hinder the conformational change and unmask hydrogen transfer.

When horse liver alcohol dehydrogenase binds coenzyme, a rotation of about 10 degrees brings the catalytic domain closer to the coenzyme binding domain and closes the active site cleft. The conformational change requires that a flexible loop containing residues 293-298 in the coenzyme binding domain rearranges so that the coenzyme and some amino acid residues from the catalytic domain can be accommodated. The change appears to control the rate of dissociation of the coenzyme and to be necessary for installation of the proton relay system. In this study, directed mutagenesis produced the activated Gly293Ala/Pro295Thr enzyme. X-ray crystallography shows that the conformations of both free and complexed forms of the mutated enzyme and wild-type apoenzyme are very similar. Binding of NAD(+) and 2,2, 2-trifluoroethanol do not cause the conformational change, but the nicotinamide ribose moiety and alcohol are not in a fixed position. Although the Gly293Ala and Pro295Thr substitutions do not disturb the apoenzyme structure, molecular modeling shows that the new side chains cannot be accommodated in the closed native holoenzyme complex without steric alterations. The mutated enzyme may be active in the "open" conformation. The turnover numbers with ethanol and acetaldehyde increase 1.5- and 5.5-fold, respectively, and dissociation constants for coenzymes and other kinetic constants increase 40-2,000-fold compared to those of the native enzyme. Substrate deuterium isotope effects on the steady state V or V/K(m) parameters of 4-6 with ethanol or benzyl alcohol indicate that hydrogen transfer is a major rate-limiting step in catalysis. Steady state oxidation of benzyl alcohol is most rapid above a pK of about 9 for V and V/K(m) and is 2-fold faster in D(2)O than in H(2)O. The results are consistent with hydride transfer from a ground state zinc alkoxide that forms a low-barrier hydrogen bond with the hydroxyl group of Ser48.     

Purine metabolism in Trypanosoma brucei gambiense

The procyclic forms of Trypanosoma brucei gambiense do not incorporate glycine or serine into ribonucleotides. Although de novo purine synthesis does not occur, all purine bases and ribonucleotides are interconverted, indicating the presence of active salvage pathways. Guanine is actively deaminated to xanthine by guanase activity. Purine ribonucleosides are cleaved to their respective free bases. The order of salvage efficiency for purine bases and their respective ribonucleotides is: adenine > hypoxanthine > guanine > xanthine.     

Mimicry of native peptide antigens by the corresponding retro-inverso analogs is dependent on their intrinsic structure and interaction propensities

Retro-inverso (ri) analogs of model T cell and B cell epitopes were predictively designed as mimics and then assayed for activity to understand the basis of functional ri-antigenic peptide mimicry. ri versions of two MHC class I binding peptide epitopes, one from a vesicular stomatitis virus glycoprotein (VSV(p)) and another from OVA (OVAp), exhibit structural as well as functional mimicry of their native counterparts. The two ri peptides exhibit conformational plasticity and they bind to MHC class I (H-2K(b)) similar to their native counterparts both in silico and in vivo. In fact, ri-OVAp is also presented to an OVAp-specific T cell line in a mode similar to native OVAp. In contrast, the ri version of an immunodominant B cell peptide epitope from a hepatitis B virus protein, PS1, exhibits no structural or functional correlation with its native counterpart. PS1 and its ri analog do not exhibit similar conformational propensities. PS1 is less flexible relative to its ri version. These observed structure-function relationships of the ri-peptide epitopes are consistent with the differences in recognition properties between peptide-MHC vs peptide-Ab binding where, while the recognition of the epitope by MHC is pattern based, the exquisitely specific recognition of Ag by Ab arises from the high complementarity between the Ag and the binding site of the Ab. It is evident that the correlation of conformational and interaction propensities of native L-peptides and their ri counterparts depends both on their inherent structural properties and on their mode of recognition.     

Complete purification and studies on the structural and kinetic properties of two forms of yeast valyl-tRNA synthetase.

Two forms of baker's yease valyl-tRNA synthetase have been purified to apparent homogeneity by classical methods. It was demonstrated that one of the two forms of the enzyme originates from the other by proteolysis, the respective amounts of each form depending on the physiological state of the yeast. The species mainly isolated from exponential growing yeast cells is a monomer of 130,000 daltons molecular weight. In stationary phase cells or in commercial yeast the major species is a degraded monomer of 120,000 daltons molecular weight ; however when the purification is carried out in the presence of phenylmethyl-sulphonyl fluoride, or diisopropylfluorophosphate large amounts of the not - degreded monomer can be obtained. Of great practical usefulness is the fact that large amounts of the native enzyme can be obtained pure after only two chromatographic steps on DEAE-cellulose and hydroxylapatite. The kinetic constants for valine, ATP and tRNAVal were determined, as well as the optimum aminoacylation conditions. It was found that the specific activity of the nondegraded valyl-tRNA synthetase is higher than that of the proteolysed enzyme for the aminoacylation reaction. On the contrary, both forms have the same ATP-pyroposphate exchange activity. The amino acids composition of the native enzyme was established. The tryptic fingerprints of the two valyl-tRNA synthetases were studied. Essentially similar maps were obtained. The number of the spots in the fingerprints indicates that the enzymes contain a high proportion of repeated sequences.     

Conformational changes in serpins: I. The native and cleaved conformations of alpha(1)-antitrypsin

The serpins (SERine Proteinase INhibitors) are a family of proteins with important physiological roles, including but not limited to the inhibition of chymotrypsin-like serine proteinases. The inhibitory mechan- ism involves a large conformational change known as the S-->R (stressed-->relaxed) transition. The largest structural differences occur in a region around the scissile bond called the reactive centre loop: In the native (S) state, the reactive centre is exposed, and is free to interact with proteinases. In inhibitory serpins, in the cleaved (R) state the reactive centre loop forms an additional strand within the beta-sheet. The latent state is an uncleaved state in which the intact reactive centre loop is integrated into the A sheet as in the cleaved form, to give an alternative R state.The serpin structures illustrate detailed control of conformation within a single protein. Serpins are also an unusual family of proteins in which homologues have native states with different folding topologies. Determination of the structures of inhibitory serpins in multiple conformational states permits a detailed analysis of the mechanism of the S-->R transition, and of the way in which a single sequence can form two stabilised states of different topology.Here we compare the conformations of alpha(1)-antitrypsin in native and cleaved states. Many protein conformational changes involve relative motions of large rigid subunits. We determine the rigid subunits of alpha(1)-antitrypsin and analyse the changes in their relative position and orientation. Knowing that the conformational change is initiated by cleavage at the reactive centre, we describe a mechanism of the S-->R transition as a logical sequence of mechanical effects, even though the transition likely proceeds in a concerted manner.