Structure of a serpin-enzyme complex probed by cysteine substitutions and fluorescence spectroscopy

ABSTRACT The x-ray crystal structure of the serpin-proteinase complex is yet to be determined. In this study we have investigated the conformational changes that take place within antitrypsin during complex formation with catalytically inactive (thrombin^sub 195A^ and active thrombin. Three variants of antitrypsin Pittsburgh (an effective thrombin inhibitor), each containing a unique cysteine residue (Cys232, Cys^sub 232^, Cys^sub P3^ and Cys^sub 313^) were covalently modified with the fluorescence probe N,N'-dimethylN-(iodoacetyl)-N'-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)ethylenediamine. The presence of the fluorescent label did not affect the structure or inhibitory activity of the serpin. We monitored the changes in the fluorescence emission spectra of each labeled serpin in the native and cleaved state, and in complex with active and inactive thrombin. These data show that the serpin undergoes conformational change upon forming a complex with either active or inactive proteinase. Steady-state fluorescence quenching measurements using potassium iodide were used to further probe the nature and extent of this conformational change. A pronounced conformational change is observed upon locking with an active proteinase; however, our data reveal that docking with the inactive proteinase thrombin^sub S195A^ is also able to induce a conformational change in the serpin.

INTRODUCTION

Serpins are a unique family of proteinase inhibitors (Potempa et al., 1994) that in contrast to the standard-mechanism inhibitors, such as the small Kunitz and Kazal inhibitors (Bode and Huber, 1992), achieve inhibition of their target proteinase via a mobile reactive center loop (RCL). Although the conformation of the RCL during the inhibitory process is not known, its ability to assume different conformations is important in inhibition (Hopkins and Stone, 1995). In native serpins (Fig. 1 a) the RCL is poised at the top of the molecule ready to interact with the target proteinase. Upon proteolytic cleavage within the RCL region, the molecule opens up and the RCL is incorporated as the fourth strand of the large central A beta-sheet (Fig. 1 b). This conformational change, termed the stressed to relaxed transition (S to R) is accompanied by a dramatic increase in heat stability and resistance to denaturants such as urea or guanidine hydrochloride (Mast et al., 1992). The x-ray crystal structures of two structural intermediates involved in this transition have been determined in which the RCL is partially inserted. In native antithrombin two residues are inserted into the top of the A beta-sheet (Fig. 2 a) (Schreuder et al., 1994; Carrell et al., 1994; Skinner et al., 1997). The recently determined structure of antichymotrypsin Leu55Pro adopted an unusual conformation (termed (), in which four residues of the RCL occupy the top half of the A beta-sheet and the F-helix unwinds to occupy the bottom half of the sheet (Fig. 2 b) (Gooptu et al., 2000). The identification of these intermediates is critical, because it provides direct crystallographic evidence that it is possible for the serpin to adopt transitional states along the pathway of conformational change.

Lawrence and colleagues (1995) elegantly demonstrated that RCL cleavage at the scissile bond is critical for the formation of the final locked complex between serpin and proteinase. Furthermore, they proposed that the conformational change within the serpin is essential for trapping the proteinase at the acyl-enzyme step in the proteinase cleavage pathway. Several studies have shown that serpin conformational mobility and RCL insertion are crucial for efficient proteinase inhibition (Hopkins and Stone, 1995; Picard et al., 1999; Shore et al., 1995; Stratikos and Gettins, 1997, 1999; Wilczynska et al., 1997). For example, numerous mutations within the RCL have been shown to result in substrate-like behavior (for review see Stein and Carrell, 1995). It is proposed that such mutations disrupt efficient loop insertion and thus allow the proteinase to escape inhibition (Hopkins et al., 1993; Hopkins and Stone, 1995). Recent biophysical studies have played a major role in elucidating the structure of the serpin-proteinase complex (Wilczynska et al., 1997; Stratikos and Gettins, 1997, 1998, 1999), with a consensus emerging that the proteinase is translocated to the bottom of the serpin (Wright, 1996) (Fig. 2 c). Although it is clear that RCL insertion into the A beta-sheet is a requirement for successful inhibition, the mechanism by which loop insertion is triggered remains unclear (Stone and Le Bonniec, 1997).

In this study we use a combination of site-directed mutagenesis, fluorescence labeling and fluorescence quenching techniques to probe the structure of the complex further. Specifically we investigated the interaction between antitrypsin Pittsburgh (denoted a,PI in this study) which possesses a P1 = Arg and thrombin. This serpin was used because it has high affinity for both thrombin and inactive thrombin (thrombin^sub S195A^), therefore providing us with the opportunity to examine both the Michaelis-complex (E*I) and final covalent complex (ElI). Using both these thrombin forms allows us to gain insight into the structure of the initial docking complex (Cooperman et al., 1993; O'Malley et al., 1997; Stone and Le Bonniec, 1997) formed between the serpin and proteinase and then the conformational changes involved in final inhibition.

MATERIALS AND METHODS

N,N'-dimethyl-N-(iodoacetyl)-N'-(7-nitrobenz-2-oxa- 1,3-diazol-4-yl)ethylenediamine (IANBD) were purchased from Molecular Probes (Eugene, OR). Thrombin was purified from human plasma and characterized as previously described (Stone and Hofsteenge, 1986), and thrombin(sl,5A) was purified and characterized as previously described (Le Bonniec et al., 1993; Stone and Le Bonniec, 1997). Spectroscopic methods

Fluorescence emission spectra were recorded on a Perkin-Elmer LS50B spectrofluorimeter, using a thermostatted cuvette holder at 37'C in a 1-cm-path-length quartz cell. Excitation and emission slits were set at 2.5 nm for all spectra and a scan speed of 10 nm/min was used. The absorbance at the excitation wavelengths was monitored in all experiments and remained below 0.05 units.

Steady-state fluorescence quenching

Fluorescence quenching measurements were performed in 50 rum Tris, pH 8.0, at 37C. Aliquots of KI (2 M stock) containing 1 mM Na,S203 were added to protein solutions (200 nM) and the change in fluorescence emission intensity of the covalently bound IANBD (A., = 480 rim) was measured. All fluorescence data were corrected for sample dilution. The quenching data were analyzed by the Stern-Volmer equation as previously described by Lehrer (1971). All data were corrected for inner filter effects where necessary.

Coordinates and model building

The coordinates of wild-type native (Elliott et al., 1996) (pdb identifier 1QLP) and cleaved (Loebermann et al., 1984) (protein data bank (PDB) identifier 7API) antitrypsin, native antithrombin (Schreuder et al., 1994; Carrell et al., 1994; Skinner et al., 1997) (PDB; identifier 2ANT), and thrombin (Qiu et al., 1992) (PDB identifier 1ABJ) were obtained from the protein data bank (www.rcsb.org). Previous studies by Elliott et al. (1996) have demonstrated that the RCL of antitrypsin adopts a canonical conformation (Hubbard et al., 1991) and can be docked into the active site of chymotrypsin with relatively few steric clashes. We used similar superposition and modeling techniques to that described previously (Elliott et al., 1996; Whisstock et al., 1996) to dock antitrypsin into the active site of thrombin. Briefly, the Pl methionine residue was changed to an arginine using the mutate facility in Quanta (MSI, San Diego, CA). The PI of the proteinase inhibitor D-Phe-Pro-Arg chloromethylketone (PPACK) in the active site of thrombin was used as a template to position the Pl arginine residue of antitrypsin into the St subsite of thrombin. The PPACK molecule was then removed to leave a model of antitrypsin P1 = Arg docked to thrombin. Several side-chain clashes were observed between the proteinase and the body of the serpin, and these were resolved by subjecting the model to rounds of CHARMm minimization until convergence was reached. The stereochemistry of the model was checked and all residues found to be in allowed conformations.

To build E-I we used the x-ray crystal structure of native antithrombin as a template in the program MODELLER (Sali and Blundell, 1993) to construct a model of antitrypsin in which the RCL is partially inserted to P14. The RCL of antithrombin is three residues longer that that of antitrypsin and in a noncanonical conformation. To maintain a canonical loop in our model of partially inserted antitrypsin, the RCL was rebuilt, using the structure of native antitrypsin as a template. A similar superposition procedure to that previously described was used to generate a model between thrombin and antitrypsin Pittsburgh in which the RCL is partially inserted to P14. We observed few steric clashes between the proteinase and inhibitor, and these were resolved by rounds of CHARMm minimization. The stereochemistry of the model of E-1 was checked and all residues were in allowed conformations

RESULTS

Previous kinetic studies have identified a number of intermediates involved in the serpin-proteinase inhibitory pathway (Fig. 3) (O'Malley et al., 1997; Stone et al., 1997;

DISCUSSION

The minimal kinetic scheme presented in Fig. 3 illustrates the complexity of the serpin inhibitory pathway. Using the combination of a,PI with both active and inactive proteinase, we have been able to form specific conformations along the pathway for study. The conformations of intact a,PI (I) and RCL cleaved a,PI (I*) have been crystallographically characterized (Fig. 1) and are easily studied. The I and I* states represent the extremes of RCL insertion, i.e., no insertion (I) and full insertion (I*). The final covalent complex (El) is formed between a,PI and active thrombin, and the initial Michaelis complex (E*I) is formed with thrombin^sub S195A^. In this study we have used a combination of protein engineering and fluorescence spectroscopic techniques to examine the structure of alpha^sub 1^PI in these four states.

Analysis of the native and cleaved states of a,Pl

The fluorescence approach used here shows that upon the S to R transformation a considerable conformational change takes place. There is no structural change around CYS2321 and this is supported by structural analysis of both I and P. CYS232 is located on the B P-sheet, a region not associated with any structural reorganization during formation of the stable complex. Indeed, a structural comparison between native and cleaved antitrypsin revealed that this region forms part of the rigid scaffold upon which most serpin conformational changes occur (Stein and Chothia, 1991; James et al., 1999; Whisstock et al., 2000). The Cys residue Of alPl-CYS13', situated at the C-terminal end of the RCL, shows a small blue shift (2 mn) in km.,a, upon RCL cleavage and an increase in K, indicating increased accessibility to the quenching agent iodide. The most significant changes observed, however, were around CYS313, which is situated at the base of the serpin molecule on a loop connecting strands 5A and 6A of the A P-sheet. Upon RCL insertion there was a large red shift in the Amk.. (6 nm) of a, PI-CYS313 and an increase in K, These data suggest significant movement in this region that increases the exposure of the IANBD label to solvent and subsequently enhances its accessibility to the iodide quenching agent. A structural comparison between native and cleaved alpha^sub 1^PI revealed that the loop containing residue 313 is part of a rigid fragment that shifts significantly during the S to R transition (Whisstock et al., 2000). Thus, the fluorescent changes we observe between the native and cleaved form of the alPI-Cys313 are entirely consistent with RCL insertion and the S to R transition.

Analysis of El^

Our data describe a covalent complex in which the proteinase has moved from the top of the serpin to some position on the A beta-sheet of the molecule. Our data are not sufficient to place the proteinase in a specific position, although we are able to narrow down its location with respect to the serpin. The lambda^sub max^ of alPI-CysP3, in complex with thrombin is similar to the native state, suggesting a similar solvent environment for the probe in both states (i.e., not covered by the proteinase in the complex). The Kv value of the ElI form is the same as the cleaved state, suggesting that the C-terminal residues of the scissile bond are in a similar environment. Therefore, the proteinase has moved significantly from its initial docking position. Our data with the CPI-Cys313 indicate that thrombin is not situated directly over this residue at the base of the serpin. This is due to the increased solvent exposure of the label on alPI-Cys313 in the presence of thrombin. This is in contrast to recent results from experiments that placed a probe at Cys314 and found it covered by trypsin. Trypsin is a much smaller proteinase than thrombin and this may go some way toward explaining the difference (Stratikos and Gettins, 1999).

In conclusion, the data presented here using alpha^sub 1^]PI and thrombin clearly demonstrate that proteinase docking is enough to trigger RCL insertion into the A beta-sheet, which has allowed us to present a model of the initial docking complex. However, although we are not able to precisely map the position of the proteinase in the final covalent complex it is clear that it has significantly moved from its initial position at the top of the serpin.

We gratefully acknowledge the Clive and Vera Ramaciotti Foundation for the purchase of the spectrofluorimeter. J.C.W. thanks the NHMRC, and S.P.B. thanks the NHMRC and the ARC for their generous support. J.P.L. and J.C.W. contributed equally to the work.

[Reference]

REFERENCES

[Reference]

Bode, W., and R. Huber. 1992. Natural protein proteinase inhibitors and their interaction with proteinases. Eur. J. Biochem 204:433-451.

Bottomley, S. P., P. C. Hopkins, and J. C. Whisstock. 1998. Antitrypsin polymerisation can occur by both loop A and C sheet mechanisms. Biochem. Biophys. Res. Commun. 251:1-5.

Carrell, R. W., P. E. Stein, G. Fermi, and M. R. Wardell. 1994. Biological implications of a 3 A structure of dimeric antithrombin. Structure. 2:257-270.

Cooperman, B. S., E. Stavridi, E. Nickbarg, E. Rescorla, N. M. Schechter, and H. Rubin. 1993. Antichymotrypsin interaction with chymotrypsin: partitioning of the complex. J. Biol. Chem. 268:23616-23625.

Eftink, M. R., and C. A. Ghiron. 1981. Fluorescence quenching studies with proteins. Anal. Biochem. 114:199-227.

Elliott, P. R., D. A. Lomas, R. W. Carrell, and J. P. Abrahams. 1996. Inhibitory conformation of the reactive loop of alpha 1-antitrypsin. Nat. Struct. Biol. 3:676-681.

Gooptu, B., B. Hazes, W. S. Chang, T. R. Dafforn, R. W. Carrell, R. J. Read, and D. A. Lomas. 2000. Inactive conformation of the serpin alpha(1)-antichymotrypsin indicates two-stage insertion of the reactive loop: implications for inhibitory function and conformational disease. Proc. Natl. Acad. Sci. U.S.A. 97:67-72.

Hopkins, P. C., R. W. Carrell, and S. R. Stone. 1993. Effects of mutations in the hinge region of serpins. Biochemistry. 32:7650-7657.

Hopkins, P. C., and S. R. Stone. 1995. The contribution of the conserved hinge region residues of alphal-antitrypsin to its reaction with elastase. Biochemistry. 34:15872-15879.

[Reference]

Hubbard, S. J., S. F. Campbell, and J. M. Thornton. 1991. Molecular recognition: conformational analysis of limited proteolytic sites and serine proteinase protein inhibitors. J. Mol. Biol. 220:507-530.

James, E. L., J. C. Whisstock, M. G. Gore, and S. P. Bottomley. 1999. Probing the unfolding pathway of antitrypsin. J. BioL Chem. 274: 9482-9488.

Lawrence, D. A., D. Ginsburg, D. E. Day, M. B. Berkenpas, I. M. Verhamme, J. 0. Kvassman, and J. D. Shore. 1995. Serpin-protease complexes are trapped as stable acyl-enzyme intermediates. J. Biol. Chem. 270:25309-25312.

Le Bonniec, B. F., E. R. Guinto, R. T. MacGillivray, S. R. Stone, and C. T. Esmon. 1993. The role of thrombin's Tyr-Pro-Pro-Trp motif in the interaction with fibrinogen, thrombomodulin, protein C, antithrombin III, and the Kunitz inhibitors. J. BioL Chem. 268:19055-19061.

Le Bonnier, B. F., E. R. Guinto, and S. R. Stone. 1995. Identification of thrombin residues that modulate its interactions with antithrombin III and alpha 1-antitrypsin. Biochemistry. 34:12241-12248.

Lehrer, S. S. 1971. Solute perturbation of protein fluorescence: the quenching of the trytophyl fluorescence of model compounds and of lysozyme by iodide ion. Biochemistry. 10:3254-3263.

Loebermann, H., R. Tokuoka, J. Deisenhofer, and R. Huber. 1984. Human alpha 1-proteinase inhibitor: crystal structure analysis of two crystal modifications, molecular model and preliminary analysis of the implications for function. J. MoL BioL 177:531-557.

Mast, A. E., J. J. Enghild, and G. Salvesen. 1992. Conformation of the reactive site loop of alpha 1-proteinase inhibitor probed by limited proteolysis. Biochemistry. 31:2720-2728.

O'Malley, K. M., S. A. Nair, H. Rubin, and B. S. Cooperman. 1997. The kinetic mechanism of serpin-proteinase complex formation: an intermediate between the Michaelis complex and the inhibited complex. J. Biol. Chem. 272:5354-5359.

[Reference]

Picard, V., P. E. Marque, F. Paolucci, M. Aiach, and B. F. Le Bonniec. 1999. Topology of the stable serpin-protease complexes revealed by an autoantibody that fails to react with the monomeric conformers of antithrombin. J. BioL Chem. 274:4586-4593.

Potempa, J., E. Korzus, and J. Travis. 1994. The serpin superfamily of proteinase inhibitors: structure, function, and regulation. J. BioL Chem. 269:15957-15960.

Qiu, X., K. P. Padmanabhan, V. E. Carperos, A. Tulinsky, T. Kline, J. M. Maraganore, and J. W. D. Fenton. 1992. Structure of the hirulog 3-thrombin complex and nature of the S' subsites of substrates and inhibitors. Biochemistry. 31:11689-11697.

Sali, A., and T. L. Blundell. 1993. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234:779-815.

Schechter, L, and A. Berger. 1967. On the size of the active site in proteases. I. Papain. Biochem. Biophys. Res. Commun. 27:157-162. Schreuder, H. A., B. de Boer, R. Dijkema, J. Mulders, H. J. Theunissen,

P. D. Grootenhuis, and W. G. Hol. 1994. The intact and cleaved human antithrombin III complex as a model for serpin-proteinase interactions. Nat. Struct. Biol. 1:48-54.

Shore, J. D., D. E. Day, A. M. Francis-Chmura, I. Verhamme, J. Kvassman, D. A. Lawrence, and D. Ginsburg. 1995. A fluorescent probe study of plasminogen activator inhibitor-1: evidence for reactive center loop insertion and its role in the inhibitory mechanism. J. BioL Chem. 270:5395-5398.

Skinner, R., J. P. Abrahams, J. C. Whisstock, A. M. Lesk, R. W. Carrell, and M. R. Wardell. 1997. The 2.6 angstrom structure of antithrombin indicates a conformational change at the heparin binding site. J. Mol. Biol. 266:601-609.

[Reference]

Stein, P. E., and R. W. Carrell. 1995. What do dysfunctional serpins tell us about molecular mobility and disease? Nat. Struct. BioL 2:96-113.

Stein, P., and C. Chothia. 1991. Serpin tertiary structure transformation. J. MoL BioL 221:615-621.

Stone, S. R., and J. Hofsteenge. 1986. Kinetics of the inhibition of thrombin by hirudin. Biochemistry. 25:4622-4628.

Stone, S. R., and B. F. Le Bonnier. 1997. Inhibitory mechanism of serpins: identification of steps involving the active-site serine residue of the protease. J. Mol. Biol. 265:344-362.

Stone, S. R., J. C. Whisstock, S. P. Bottomley, and P. C. Hopkins. 1997. Serpins: a mechanistic class of their own. Adv. Exp. Med. Biol. 425: 5-15.

[Reference]

Stratikos, E., and P. G. Gettins. 1997. Major proteinase movement upon stable serpin-proteinase complex formation. Proc. Natl Acad. Sci. U.S.A. 94:453-458.

Stratikos, E., and P. G. Gettins. 1998. Mapping the serpin-proteinase complex using single cysteine variants of alphal-proteinase inhibitor Pittsburgh. J. Biol. Chem. 273:15582-15589.

Stratikos, E., and P. G. Gettins. 1999. Formation of the covalent serpinproteinase complex involves translocation of the proteinase by more than 70 A and full insertion of the reactive center loop into beta-sheet A. Proc. Natl. Acad. Sci. U.S.A. 96:4808-4813.

Whisstock, J., A. M. Lesk, and R. Carrell. 1996. Modeling of serpinprotease complexes: antithrombin-thrombin, alpha(1)-antitrypsin (358Met-Arg)-thrombin, alpha 1-antitrypsin (358Met-Arg)-trypsin, and antitrypsin-elastase. Proteins. 26:288-303.

Whisstock, J. C., R. Skinner, R. W. Carrell, and A. M. Lesk. 2000. Conformational changes in serpins. I. The native and cleaved conformations of alpha(l)-antitrypsin. J. Mol. Biol. 295:651-665.

Wilczynska, M., M. Fa, J. Karolin, P. 1. Ohlsson, L. B. A. Johansson, and T. Ny. 1997. Structural insights into serpin-protease complexes reveal the inhibitory mechanism of serpins. Nat. Struct. BioL 4:354-357.

Wright, H. T. 1996. The structural puzzle of how serpin serine proteinase inhibitors work. Bioessays. 18:453-464.

[Author Affiliation]

Justin P. Ludeman,* James C. Whisstock,* Paul C.R. Hopkins,^ Bernard F. Le Bonniec,^^ and Stephen P. Bottomley*

*Department of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria 3800, Australia; ^Gladstone Institute of Cardiovascular Disease, San Francisco, California 94141 USA; ^^INSERM, Unite 428, Univeriste Paris V, 75270 Paris, France

[Author Affiliation]

Received for publication 6 June 2000 and in final form 3 October 2000.

Address reprint requests to Dr. Stephen Bottomley, Department of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria 3800, Australia. Tel.: 61-3-9905-3703; Fax: 61-3-9905-4699; E-mail: steve.bottomley@med.monash.edu.au.