FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1997483799> ?p ?o ?g. }

• W1997483799 endingPage "9269" @default.
• W1997483799 startingPage "9264" @default.
• W1997483799 abstract "Despite several concordant structural studies on the initial non-covalent complex that serpins form with target proteinases, a recent study on the non-covalent complex between the serpin α1-proteinase inhibitor (α1PI) and anhydroelastase (Mellet, P., and Bieth, J. G. (2000) J. Biol. Chem. 275, 10788–10795) concluded that translocation of the proteinase precedes cleavage of the reactive center loop and formation of the acyl ester. Because this conclusion is diametrically opposite to those of the other structural studies on serpin-proteinase pairs, we proceeded to examine this specific serpin-proteinase complex by the same successful NMR approach used previously on the α1PI-Pittsburgh-S195A trypsin pair. Both non-covalent complex with anhydroelastase and covalent complex with active elastase were made with 15N-alanine-labeled wild-type α1PI. The heteronuclear single quantum correlation spectroscopy (HSQC) NMR spectrum of the non-covalent complex showed that the entire reactive center loop remained exposed, and the serpin body maintained a conformation indistinguishable from that of native α1PI, indicating no movement of the proteinase and no insertion of the reactive center loop into β-sheet A. In contrast, the HSQC NMR spectrum of the covalent complex showed that the reactive center loop had fully inserted into β-sheet A, indicating that translocation of the proteinase had occurred. These results agree with previous NMR, fluorescence resonance energy transfer, and x-ray crystallographic studies and suggest that a common mechanism is employed in formation of serpin-proteinase complexes. We found that preparations of anhydroelastase that are not appropriately purified contain material that can regenerate active elastase over time. It is likely that the material used by Mellet and Bieth contained such active elastase, resulting in mistaken attribution of the behavior of covalent complex to that of the non-covalent complex. Despite several concordant structural studies on the initial non-covalent complex that serpins form with target proteinases, a recent study on the non-covalent complex between the serpin α1-proteinase inhibitor (α1PI) and anhydroelastase (Mellet, P., and Bieth, J. G. (2000) J. Biol. Chem. 275, 10788–10795) concluded that translocation of the proteinase precedes cleavage of the reactive center loop and formation of the acyl ester. Because this conclusion is diametrically opposite to those of the other structural studies on serpin-proteinase pairs, we proceeded to examine this specific serpin-proteinase complex by the same successful NMR approach used previously on the α1PI-Pittsburgh-S195A trypsin pair. Both non-covalent complex with anhydroelastase and covalent complex with active elastase were made with 15N-alanine-labeled wild-type α1PI. The heteronuclear single quantum correlation spectroscopy (HSQC) NMR spectrum of the non-covalent complex showed that the entire reactive center loop remained exposed, and the serpin body maintained a conformation indistinguishable from that of native α1PI, indicating no movement of the proteinase and no insertion of the reactive center loop into β-sheet A. In contrast, the HSQC NMR spectrum of the covalent complex showed that the reactive center loop had fully inserted into β-sheet A, indicating that translocation of the proteinase had occurred. These results agree with previous NMR, fluorescence resonance energy transfer, and x-ray crystallographic studies and suggest that a common mechanism is employed in formation of serpin-proteinase complexes. We found that preparations of anhydroelastase that are not appropriately purified contain material that can regenerate active elastase over time. It is likely that the material used by Mellet and Bieth contained such active elastase, resulting in mistaken attribution of the behavior of covalent complex to that of the non-covalent complex. Serpins are a large and growing superfamily of proteins, many of which are inhibitors of serine proteinase and some of which can also inhibit cysteine proteinases (1Gettins P.G.W. Chem. Rev. 2002; 102: 4751-4804Crossref PubMed Scopus (1003) Google Scholar). X-ray structures have now been determined for 10 different serpins in their active conformations, all of which show the same overall fold and a prominent, exposed reactive center loop through which the serpin makes initial contact with the target proteinase (1Gettins P.G.W. Chem. Rev. 2002; 102: 4751-4804Crossref PubMed Scopus (1003) Google Scholar). Much early biochemical data had indicated that serpins function by a mechanism very different from that of lock-and-key type proteinase inhibitors of the Kunitz, Kazal, and Bowman-Birk families (2Patston P.A. Gettins P. Beechem J. Schapira M. Biochemistry. 1991; 30: 8876-8882Crossref PubMed Scopus (173) Google Scholar, 3Cooperman B.S. Stavridi E. Nickbarg E. Rescorla E. Schechter N.M. Rubin H. J. Biol. Chem. 1993; 268: 23616-23625Abstract Full Text PDF PubMed Google Scholar, 4Lawrence D.A. Olson S.T. Palaniappan S. Ginsburg D. J. Biol. Chem. 1994; 269: 27657-27662Abstract Full Text PDF PubMed Google Scholar, 5Lawrence D.A. Ginsburg D. Day D.E. Berkenpas M.B. Verhamme I.M. Kvassman J.-O. Shore J.D. J. Biol. Chem. 1995; 270: 25309-25312Abstract Full Text Full Text PDF PubMed Scopus (229) Google Scholar, 6Olson S.T. Bock P.E. Kvassman J. Shore J.D. Lawrence D.A. Ginsburg D. Björk I. J. Biol. Chem. 1995; 270: 30007-30017Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar, 7Hood D.B. Huntington J.A. Gettins P.G.W. Biochemistry. 1994; 33: 8538-8547Crossref PubMed Scopus (102) Google Scholar). However, prior to definitive structural studies, the details of the mechanism had remained highly controversial (8Wright H.T. Scarsdale J.N. Proteins. 1995; 22: 210-225Crossref PubMed Scopus (157) Google Scholar). In the past 6 years structural studies on the final serpin-proteinase complex and on intermediates in the pathway have led to a coherent picture of how serpins inhibit serine proteinases by a suicide substrate mechanism (Fig. 1) (9Stratikos E. Gettins P.G.W. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 4808-4813Crossref PubMed Scopus (218) Google Scholar, 10Stratikos E. Gettins P.G.W. J. Biol. Chem. 1998; 273: 15582-15589Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar, 11Stratikos E. Gettins P.G.W. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 453-458Crossref PubMed Scopus (139) Google Scholar, 12Huntington J.A. Read R.J. Carrell R.W. Nature. 2000; 407: 923-926Crossref PubMed Scopus (955) Google Scholar, 13Baglin T.P. Carrell R.W. Church F.C. Esmon C.T. Huntington J.A. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 11079-11084Crossref PubMed Scopus (186) Google Scholar, 14Ye S. Cech A.L. Belmares R. Bergstrom R.C. Tong Y. Corey D.R. Kanost M.R. Goldsmith E.J. Nat. Struct. Biol. 2001; 8: 979-983Crossref PubMed Scopus (145) Google Scholar, 15Fa M. Bergstrom F. Hagglof P. Wilczynska M. Johansson L. Ny T. Structure. 2000; 8: 397-405Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar). This mechanism involves formation of an initial noncovalent Michaelis-like complex between the proteinase and specificity-determining residues in the exposed RCL. 1The abbreviations used are: RCL, reactive center loop; α1PI, α1-proteinase inhibitor; PMSF, phenylmethylsulfonyl fluoride; PAI-1, plasminogen activator inhibitor 1; HSQC, heteronuclear single quantum correlation spectroscopy. This complex requires little or no change in conformation of the RCL, and no insertion of the RCL into β-sheet A of the serpin. The catalytic apparatus of the proteinase then attacks the scissile peptide bond in the serpin RCL leading to formation of a covalent acyl-enzyme intermediate between the peptide carbonyl of the P1 residue and Ser-195 of the proteinase and cleavage of the RCL. This RCL cleavage then permits unrestrained insertion of the RCL into β-sheet A, with concomitant movement of the attached proteinase. Following full insertion of the RCL, the proteinase is then held sufficiently tightly against the bottom of the serpin that the proteinase active site residues are disoriented, rendering the proteinase catalytically incompetent and hence trapping the proteinase in a kinetically stabilized covalent acyl-enzyme complex. The source of the energy required to distort the proteinase is believed to come from the enhanced stability of the loop-inserted conformation of the serpin (16Boudier C. Bieth J.G. Biochemistry. 2001; 40: 9962-9967Crossref PubMed Scopus (17) Google Scholar, 17Kaslik G. Kardos J. Szabó L. Závodszky P. Westler W.M. Markley J.L. Gráf L. Biochemistry. 1997; 36: 5455-5474Crossref PubMed Scopus (104) Google Scholar) but may require a coupling mechanism to store the energy until needed (18Gettins P.G.W. FEBS Lett. 2002; 523: 2-6Crossref PubMed Scopus (59) Google Scholar). This highly unusual mechanism is now supported by structural studies on a significant number of different serpin-proteinase complexes. For the final kinetically trapped covalent complex, there are fluorescence studies on the α1PI-Pittsburgh-trypsin (9Stratikos E. Gettins P.G.W. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 4808-4813Crossref PubMed Scopus (218) Google Scholar, 10Stratikos E. Gettins P.G.W. J. Biol. Chem. 1998; 273: 15582-15589Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar, 11Stratikos E. Gettins P.G.W. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 453-458Crossref PubMed Scopus (139) Google Scholar), PAI-1-trypsin, PAI-1-urokinase-type plasminogen activator, and PAI-1-tissue plasminogen activator complexes (15Fa M. Bergstrom F. Hagglof P. Wilczynska M. Johansson L. Ny T. Structure. 2000; 8: 397-405Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar, 19Backovic M. Stratikos E. Lawrence D.A. Gettins P.G.W. Protein Sci. 2002; 11: 1182-1191Crossref PubMed Scopus (14) Google Scholar); NMR studies on the α1PI-Pittsburgh-trypsin complex (20Peterson F.C. Gettins P.G.W. Biochemistry. 2001; 40: 6284-6292Crossref PubMed Scopus (44) Google Scholar); and an x-ray crystallographic study on the α1PI-trypsin complex (12Huntington J.A. Read R.J. Carrell R.W. Nature. 2000; 407: 923-926Crossref PubMed Scopus (955) Google Scholar) that all find a common pattern of full proteinase translocation. For the initial non-covalent Michaelis-like complex, there are fluorescence (9Stratikos E. Gettins P.G.W. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 4808-4813Crossref PubMed Scopus (218) Google Scholar, 10Stratikos E. Gettins P.G.W. J. Biol. Chem. 1998; 273: 15582-15589Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar, 19Backovic M. Stratikos E. Lawrence D.A. Gettins P.G.W. Protein Sci. 2002; 11: 1182-1191Crossref PubMed Scopus (14) Google Scholar), NMR (21Peterson F.C. Gordon N.C. Gettins P.G.W. Biochemistry. 2000; 39: 11884-11892Crossref PubMed Scopus (35) Google Scholar), and x-ray crystallographic studies (22Dementiev A. Simonovic M. Volz K. Gettins P.G.W. J. Biol. Chem. 2003; 278: 37881-37887Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar) on the α1PI-Pittsburgh-S195A/anhydrotrypsin complex and x-ray crystallographic studies on the serpin1K-S195A trypsin (14Ye S. Cech A.L. Belmares R. Bergstrom R.C. Tong Y. Corey D.R. Kanost M.R. Goldsmith E.J. Nat. Struct. Biol. 2001; 8: 979-983Crossref PubMed Scopus (145) Google Scholar) and heparin cofactor II-S195A thrombin (13Baglin T.P. Carrell R.W. Church F.C. Esmon C.T. Huntington J.A. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 11079-11084Crossref PubMed Scopus (186) Google Scholar) complexes that all find similar structures involving no insertion of the RCL into β-sheet A of the serpin and no movement of the proteinase. In this context of general agreement on the structures of the initial and final complexes, it is particularly surprising to encounter studies that claim very different structures for such complexes. Recently this has been the case for the covalent complex formed between α1-antichymotrypsin and chymotrypsin, in which it was proposed that little movement of the proteinase occurs (23O'Malley K.M. Cooperman B.S. J. Biol. Chem. 2001; 276: 6631-6637Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar), and conversely for the α1PI-anhydroelastase non-covalent complex for which it was proposed that full proteinase translocation occurs (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar). Because the two-dimensional NMR method that we used for examining α1PI-Pittsburgh-trypsin covalent and non-covalent complexes (20Peterson F.C. Gettins P.G.W. Biochemistry. 2001; 40: 6284-6292Crossref PubMed Scopus (44) Google Scholar, 21Peterson F.C. Gordon N.C. Gettins P.G.W. Biochemistry. 2000; 39: 11884-11892Crossref PubMed Scopus (35) Google Scholar) proved to be both very sensitive to conformational differences in the RCL and body of the serpin and to give results that were in full agreement with both earlier fluorescence resonance energy transfer studies and later x-ray studies on the same complexes, we decided to use this method to examine the specific serpin-proteinase pair examined by Mellet and Bieth (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar) to determine whether or not the nature of the pair radically affected the conformation of the complex. Expression and Purification of α1PI and 15N-Ala-labeled α1PI—The plasmid pQE30-multi8α1PI was made from pQE30-multi9α1PI (21Peterson F.C. Gordon N.C. Gettins P.G.W. Biochemistry. 2000; 39: 11884-11892Crossref PubMed Scopus (35) Google Scholar) by changing the P1 arginine residue back to the wild-type methionine. The multi7 variant of human α1PI has 7 stabilizing mutations that improve the thermal stability and prevent loop-sheet polymerization (25Im H. Seo E.J. Yu M.H. J. Biol. Chem. 1999; 274: 11072-11077Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar) but otherwise has similar inhibitory properties to that of the natural wild-type protein. The multi8 variant contains an additional change of Cys-232 to serine to prevent unwanted disulfide-mediated dimerization during refolding. α1PI was expressed in Escherichia coli, isolated as inclusion bodies, refolded, and purified as described previously (21Peterson F.C. Gordon N.C. Gettins P.G.W. Biochemistry. 2000; 39: 11884-11892Crossref PubMed Scopus (35) Google Scholar). For specific labeling with 15N-Ala, cells were grown in minimal medium supplemented with 100 mg liter-115N-Ala and 100 mg liter-1 of all other amino acids except alanine, as described previously. We use the name α1PI subsequently to represent the multi8 variant that has a natural P1 methionine. To maximize the yield of serpin, α1PI was also isolated from the supernatant and purified by the same steps as for refolded material. The concentration of α1PI was determined spectro-photometrically using an extinction coefficient at 280 nm of 0.43 for a 1 mg ml-1 solution. Preparation of α1PI-Sepharose—Anhydroelastase was purified by affinity chromatography on α1PI-Sepharose to ensure that homogeneous, inactive anhydroelastase was isolated for all subsequent experiments. α1PI was coupled to CNBr-activated Sepharose 4 Fast Flow (Amersham Biosciences) according to the manufacturer’s protocol. 20 mg of α1PI was used for coupling to each milliliter of resin. After 90 min at room temperature, 16 mg of protein was bound per ml of Sepharose. After blocking unreacted groups and washing, a 7-ml column was prepared for purification of anhydroelastase. Preparation and Purification of Anhydroelastase—Porcine pancreatic elastase from Calbiochem (catalog number 324682) was used as the starting material. A stock solution of ∼10 mg ml-1 was prepared in 5 mm sodium acetate buffer, pH 4.5, containing 1 mm EDTA and kept at 4 °C until use. Elastase concentration was determined using ∈280 = 2.1 ml·mg-1·cm-1. The stock solution was diluted to 20 μm (0.52 mg/ml) with 5 mm sodium citrate, and the pH was adjusted to 7.0 with 1 m sodium citrate. Aliquots of PMSF (stock solution of 50 mm in methanol) were added, and the mixture was incubated on ice. A total of 4–5-fold molar excess of PMSF and 16 h were necessary to obtain ≤0.5% residual activity. 1 m prechilled NaOH was added to a final concentration of 0.1 m to effect the β-elimination and generate anhydroelastase (26Ako H. Foster R.J. Ryan C.A. Biochem. Biophys. Res. Commun. 1972; 47: 1402-1407Crossref PubMed Scopus (71) Google Scholar, 27Ako H. Foster R.J. Ryan C.A. Biochemistry. 1974; 13: 132-139Crossref PubMed Scopus (78) Google Scholar). The reaction was stopped after 2 h by adjusting the pH to 6.0 with 1 m acetic acid. N-MeO-succinyl-Ala-Ala-Pro-Val-chloromethyl ketone (Sigma) was added to 2 μm (∼10% of total elastase concentration) to inhibit any residual activity. The mixture was further incubated on ice for 16 h, concentrated, and then clarified by centrifugation. About half of the crude material was lost to precipitation. The clarified crude anhydroelastase had <0.05% activity. Crude anhydroelastase (5–10 mg in 0.5–1 column volume) was applied to the α1PI-Sepharose column (7 ml) in high pH buffer (0.2 m Tris, 0.2 m NaCl, pH 8.0) and eluted by a step pH gradient. The gradients were by mixing different ratios of the high pH buffer and a low pH buffer (0.2 m sodium acetate, 0.2 m NaCl, pH 4.0). Activity of selected fractions was measured immediately after purification and also after prolonged storage at 4 °C. Anhydroelastase was found in peak II, being characterized by weak binding to α1PI-Sepharose, and lack of proteolytic activity (i.e. <0.01%) even after prolonged storage. The middle of peak II was used for our studies. The overall yield was about 30%. Kinetic Assays—Elastase activity was measured by using the chromogenic substrate succinyl-Ala-Ala-Ala-p-nitroaniline. The assay buffer was either 100 mm HEPES, 100 mm NaCl, 1 mm EDTA, 0.1% PEG-8000, pH 8.0, or 50 mm HEPES, 1 mm EDTA, 0.1% PEG-8000, pH 8.0, as indicated. The substrate concentration was 220 μm (except for KM determination where it varied between 0.22 and 4.4 mm and for the inhibition assays where it was 660 μm). Absorbance was monitored at 410 nm for 1–20 min at 25 °C. Active elastase concentrations were 1–20 nm. PMS or anhydroelastase concentrations were 0.02–2 μm. Anhydroelastase Binding to α1PI—Independent kinetic demonstration of the binding of anhydroelastase to α1PI was performed analogously to a previous demonstration of S195A tissue plasminogen activator binding to PAI-1 (28Olson S.T. Swanson R. Day D. Verhamme I. Kvassman J. Shore J.D. Biochemistry. 2001; 40: 11742-11756Crossref PubMed Scopus (89) Google Scholar, 29Delaglio F. Grzesiak A. Vuister G.W. Zhu G. Pfeifer J. Bax A. J. Biomol. NMR. 1995; 6: 277-293Crossref PubMed Scopus (11638) Google Scholar). 100 nm final α1PI was premixed with 0, 10, and 20 μm (final concentration) anhydroelastase in 50 mm HEPES, 1 mm EDTA, 0.1% PEG-8000, pH 8.0, containing 660 μm succinyl-Ala-Ala-Ala-p-nitroaniline. The reaction was started by addition of active elastase to a final concentration of 10 nm. Absorbance at 410 nm was monitored until no further increase was observed. The pseudo-first-order inhibition rate constants (kobs) were determined by nonlinear regression. The KD of the α1PI-anhydroelastase complex was calculated using the equation: KD = [EAH]·(q/(1 - q))/(1 + [S]/KM), where q = kobs/kobs,0; [S] is the substrate concentration; [EAH] is the anhydroelastase concentration, and KM is the Michaelis constant. kobs,0 is the pseudo-first-order inhibition rate constant measured in the absence of anhydroelastase. KM was determined to be 1.45 mm in 50 mm HEPES, 1 mm EDTA, 0.1% PEG-8000, pH 8.0. Preparation of Cleaved α1PI and α1PI-Elastase Covalent Complex— Active elastase and 15N-Ala-labeled α1PI were reacted in 20 mm Tris, 50 mm NaCl, 1 mm EDTA, pH 8.0 buffer. Initial concentrations were 2 μm (52 μg/ml) for elastase and 3.6 μm (162 μg/ml) for α1PI (1:1.8 molar ratio). After 20 min at room temperature, 20 μmN-MeO-succinyl-Ala-Ala-Pro-Val-chloromethyl ketone was added to inhibit any residual activity, and the mixture was stored at 4 °C until further use. The final composition of the mixture was ≤2 μm covalent complex, and ∼0.8–0.8 μm cleaved and native α1PI, respectively, together with small amounts of degradation products. The mixture was stable for several weeks due to the presence of excess native α1PI and inhibition of residual elastolytic activity. After concentration and buffer exchange, the covalent complex was purified by chromatofocusing. A mixture containing ∼ 25 mg of covalent complex in starting buffer (25 mm imidazole, pH 7.4) was applied in ≤1 column volume to a 23-ml PBE 94 (Amersham Biosciences) column and eluted with 1:8 diluted polybuffer 74 (Amersham Biosciences), pH 5.0. The first peak was pure covalent complex, and subsequent overlapping peaks were degradation products, native α1PI, and cleaved α1PI. Immediately after purification, 20 μm chloromethyl ketone and 2 μm unlabeled native α1PI (an estimated 15% molar ratio assuming 100% yield) were added to the covalent complex-containing fractions to prevent any cleavage of the complex by any free elastase that might form from complex hydrolysis. Polybuffer was removed during concentration of the sample on 30-kDa cut-off membranes. The extinction coefficient of the covalent complex was calculated from that of the components (∈280 = 1.04 ml·mg-1·cm-1). The final yield for the covalent complex was 50–60%. Fractions containing cleaved α1PI were combined (pH ≈6.0), concentrated, and heated at 90 °C for 30 min. An equal volume of 500 mm NaCl, 20 mm, 1 mm EDTA, pH 8.0, was then immediately added. Most of the contaminating proteins precipitated due to the heat treatment, except cleaved α1PI, which has a melting point of >120 °C. The mixture was clarified by centrifugation and filtration, and monomeric cleaved α1PI was separated from soluble aggregates and polybuffer by gel filtration on a Superdex-75 column (Amersham Biosciences). The overall yield for cleaved α1PI was 20–30%. 1H-15N HSQC NMR—All NMR experiments were performed at 318 K on a Bruker DRX600 spectrometer equipped with a 5-mm 1H/15N/13C triple resonance cryoprobe and pulse field gradient capability. 15N-Ala-labeled native α1PI and cleaved α1PI were both 200 μm in 10% D2O, 90% H2O, 50 mm NaCl, 10 mm sodium phosphate, pH 7.0, 0.02% NaN3. The non-covalent complex was made in the same buffer using 200 μm15N-Ala-labeled native α1PI and an excess (300 μm) of unlabeled anhydroelastase. The covalent complex was ∼350 μm in the same buffer, except that the NaCl concentration was 300 mm to avoid precipitation/aggregation, and contained ∼100 μm unlabeled native α1PI. Data collection times were 3.8, 5.7, 14.3, and 15.3 h for native α1PI, cleaved α1PI, the noncovalent complex, and the covalent complex, respectively. NMR data were processed using NMRPipe software (29Delaglio F. Grzesiak A. Vuister G.W. Zhu G. Pfeifer J. Bax A. J. Biomol. NMR. 1995; 6: 277-293Crossref PubMed Scopus (11638) Google Scholar). The Non-covalent Complex of α1PI with Anhydroelastase—We have previously used two-dimensional 1H-15N HSQC NMR of 15N-Ala-labeled α1PI-Pittsburgh as an extremely sensitive means of separately evaluating the conformations of the body of the serpin and of the reactive center loop when it forms covalent complex with trypsin or non-covalent complex with S195A trypsin (20Peterson F.C. Gettins P.G.W. Biochemistry. 2001; 40: 6284-6292Crossref PubMed Scopus (44) Google Scholar, 21Peterson F.C. Gordon N.C. Gettins P.G.W. Biochemistry. 2000; 39: 11884-11892Crossref PubMed Scopus (35) Google Scholar). The spectrum of the native serpin shows the expected number of resolved resonances, with those from the cluster of RCL alanines in the center of the spectrum, reflecting their exposed, non-structured environment. Upon RCL cleavage and insertion into β-sheet A, the resonance of every alanine changes chemical shift, although the changes are particularly pronounced for the alanines from the RCL, because they undergo the most drastic change in environment, from exposed and unstructured to being a central strand of β-sheet A. Although wild-type α1PI was used for the present study rather than the Pittsburgh variant, which contains a P1 arginine, the two-dimensional HSQC spectra of both native and cleaved forms were essentially identical to those of the Pittsburgh variant (Fig. 2, A and D). This is in keeping with x-ray structures of the native states of wild-type and Pittsburgh α1PIs, which are superimposable in both the body and RCL with minimal root mean square deviation. Accordingly, assignments obtained previously for selected alanines of the RCL and the serpin body are readily transferable to the spectra of the wild-type protein for both the native (Fig. 2A) and cleaved (Fig. 2D) states. The HSQC NMR spectrum of the non-covalent complex of 15N-Ala-labeled α1PI with anhydroelastase gave a single set of resolved resonances (Fig. 2C) that differed primarily from the spectrum of native α1PI in being a little broader, but at otherwise similar chemical shifts. Importantly, the presence of only a single set of visible resonances indicates that any uncomplexed 15N-labeled α1PI must be undetectable under the conditions of the NMR experiment. The lack of chemical shift perturbation is more clearly seen by superimposing the spectra of native α1PI and the non-covalent complex (Fig. 2B). The only resonances that show significant perturbation are those of P4 in the reactive center loop and Ala-284, which lies directly underneath P1 in the x-ray structures of both native α1PI and the non-covalent complex of α1PI-Pittsburgh with S195A trypsin. These are also the only two alanines that showed perturbation upon forming the non-covalent complex between α1PI-Pittsburgh and S195A trypsin and reflect direct contact perturbations resulting from interaction of the proteinase active site with residues in the immediate vicinity of the P1 residue. Critically, because the chemical shifts of alanines at P12, P11, and P9 shift by 3.4 to 4.5 ppm in the 15N dimension upon inserting into β-sheet A, yet are perturbed by at most 0.1 ppm here upon formation of the non-covalent complex, there can be no insertion of any residues of the reactive center loop into β-sheet A upon forming this complex. This is further demonstrated by the superpositioning of alanine resonances from alanines in the body of the serpin. Because these are also sensitive to the conformational change in the serpin body that accompanies expansion of β-sheet A to accommodate the RCL as a central strand, their lack of perturbation indicates un-equivocally that the serpin body has not changed conformation. Although the spectrum of the non-covalent complex shows both resonance broadening and chemical shift perturbation of two alanines, one at P4 and one underneath P1, each of which indicates that a complex was indeed formed under the conditions of the experiment, we wanted to independently demonstrate that the anhydroelastase prepared here bound to the RCL of α1PI. For this a competitive inhibition assay run under continuous assay conditions was carried out. α1PI was incubated with different concentrations of anhydroelastase in the presence of the chromogenic substrate succinyl-Ala-Ala-Ala-p-nitroaniline. Active elastase was then added and the hydrolysis of substrate monitored. As expected, the presence of anhydroelastase delayed the irreversible inhibition of elastase by α1PI in a concentration-dependent manner (Fig. 3), indicating a competition between anhydroelastase and elastase for the RCL of α1PI. Fitting of the data gave a KD of 33 ± 7 μm for the interaction between anhydroelastase and α1PI. Although this is significantly weaker than some other interactions between S195A serine proteinases and serpins (11Stratikos E. Gettins P.G.W. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 453-458Crossref PubMed Scopus (139) Google Scholar, 30Futamura A. Stratikos E. Olson S.T. Gettins P.G.W. Biochemistry. 1998; 37: 13110-13119Crossref PubMed Scopus (25) Google Scholar), it should be noted that human neutrophil elastase, rather than the porcine pancreatic elastase used both here and by Mellet and Bieth (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar), is the in vivo target of α1PI. There are also other examples of cognate serpin-proteinase pairs with comparable affinities. The Covalent Complex of α1PI with Elastase—Because it is still the case that covalent complexes have been examined structurally for relatively few serpin-proteinase pairs, we also wanted to use the same NMR approach to characterize the covalent complex formed between α1PI and active elastase. This would be both the first example of a covalent serpin complex with elastase examined by a primary structural technique (NMR or x-ray crystallography) and serve as a comparison to the non-covalent complex described above. Covalent complex between 15N-Ala-labeled α1PI and active pancreatic elastase was generated and purified as described under “Experimental Procedures.” Because it is well documented that the proteinase in such serpin-proteinase complexes is much more susceptible to proteolytic cleavage by free proteinase (whether self or foreign) than is uncomplexed proteinase, care must be taken to ensure that the covalent complex remains intact by inhibiting any elastase that might form by deacylation of the covalent complex to yield active proteinase and cleaved serpin. This was accomplished by addition of a small excess of unlabeled α1PI to react with any regenerated elastase. SDS-PAGE of the purified covalent complex after initial purification and at the beginning and end of NMR data collection showed that negligible deacylation had occurred and, also importantly, that the covalent complex showed no evidence for proteolytic cleavage, which would have resulted in bands with mobility intermediate between that of the 71-kDa intact covalent complex and the 45-kDa cleaved α1PI (Fig. 4). The covalent complex of 15N-Ala α1PI with elastase gave a 1H-15N HSQC NMR spectrum very different from that of the equivalently labeled non-covalent complex (Fig. 2F). Resonance positions were very different from those of native α1PI and instead were mostly identical to those of cleaved, loop-inserted α1PI. This is shown clearly in Fig. 2E in which the spectra of the covalent complex and cleaved α1PI are overlaid. Importantly, the resonances of the RCL alanines P12, P11, and P9 show identical positions to those in cleaved α1PI, showing that the RCL has inserted into β-sheet A (note that P12 is only visible at a lower contour level than shown). This is corroborated by the similar positions of most of the remaining reporter alanines from the body of the serpin. Only alanines 183, 316, 325, and 332 and P4 in the RCL showed a somewhat altered chemical shift from cleaved α1PI. However, these are the same alanines that showed equivalent differences in chemical shift in the complex of α1PI-Pittsburgh with trypsin. In that case it was found that each of the alanines has a special position that makes them sensitive to additional perturbations in the complex compared with simple cleaved α1PI. Thus, alanines 316 and 325 are at the contact interface between trypsin and α1PI in the x-ray structure of this covalent complex, whereas alanines at positions 183 and 332 and P4 are in strands s3A, s5A, and s4A, respectively, of β-sheet A, close to the proteinase and may be sensing the effect of the mutual “crushing” of proteinase against the bottom of the serpin. The spectrum thus indicates that elastase forms a covalent complex with α1PI that involves full loop insertion and consequent alteration of the serpin conformation to resemble that of cleaved α1PI. Such a structure, involving translocation of the proteinase to the distal end of the serpin, is the same as was found by NMR spectroscopy and x-ray crystallography for the complex of α1PI with trypsin and suggests a common structural organization and mechanism for inhibition of serine proteinases by serpins. Rationalization of Earlier Conflicting Structure—Given that the results presented above on the non-covalent complex of anhydroelastase with α1PI clearly indicate a structure in which the anhydroproteinase perturbs only the residues of the RCL in the immediate vicinity of the scissile bond and in which there is no loop insertion into β-sheet A and hence no translocation of the proteinase, the question arises as to why Mellet and Bieth (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar) came to the opposite conclusion using time-dependent fluorescence resonance energy transfer measurements. Their approach was to use anhydroelastase to which a fluorescein donor fluorophore had been covalently attached and α1PI to which a rhodamine acceptor fluorophore had been covalently attached. These species were then mixed in a stopped-flow apparatus, and the change in fluorescence was monitored with time. These authors found a two-phase change, with the first phase interpreted as initial association and the second as a concentration-independent step, representing translocation of the proteinase. Because anhydroelastase was used for this study, the second step was interpreted as a translocation prior to acylation. Two aspects of this study struck us as problematic. The first was that the kinetics of the two phases seen with anhydroelastase were identical to those reported earlier by the same authors, and by the same method, for formation of the covalent complex. The second was that the magnitude of the fluorescence change for the second phase, as a percentage of total fluorescence, was about 10%, whereas for formation of the covalent complex it was very much larger. A possible explanation for the results was that the preparation of anhydroelastase used by Mellet and Bieth (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar) contained a significant proportion of active proteinase. Under such circumstances, the second phase of the fluorescence change would correspond to formation of covalent serpin-proteinase complex and consequently have identical kinetics to that found previously for formation of this complex. At the outset of the present study we therefore tested this possibility, both to see if this could be the explanation for the discrepancy and, if it were, to ensure that anhydroelastase used here for NMR studies was not contaminated with active proteinase. Anhydroelastase was prepared chemically by the same β-elimination method used by Mellet and Bieth (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar). In the present study, however, the crude product was fractionated over an α1PI-Sepharose column, and fractions were assayed immediately and after several weeks storage at 4 °C for enzymatic activity. Although both the initial crude material and all eluted fractions assayed immediately after elution had negligible enzymatic activity, it was found that the later eluting fractions (Fig. 5A, peak III) assayed after storage at 4 °C recovered activity in a solution assay and were capable of forming SDS-stable complexes with α1PI (Fig. 5B, lanes 8 and 9). In contrast, fractions eluting in the central part of peak II (Fig. 5A) did not regain activity over time and were incapable of forming an SDS-stable complex with α1PI. These fractions were judged to be pure anhydroelastase and were the ones used in the present study. Because Mellet and Bieth (24Mellet P. Bieth J.G. J. Biol. Chem. 2000; 275: 10788-10795Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar) used only the absence of enzymatic activity and the ability to bind to eglin c as criteria of purity, any elastase that was still the PMSF derivative and had not undergone β-elimination might have been included in their preparation. If used immediately, there would have been no ability to form covalent complex. However, if used some time later, hydrolysis of the PMS group would have regenerated functional elastase, as we found here for fractions eluting under peak III. Conclusions—In the present study we have used a sensitive primary structural approach (NMR) to examine the conformations and organization of both non-covalent and covalent complexes of the serpin α1PI with anhydroelastase and elastase, respectively. The structure of the non-covalent complex was found to be indistinguishable from those of other serpin-proteinase pairs and to involve simple docking of serpin and proteinase. Similarly, the covalent complex was found to be like other covalent serpin-proteinase complexes and to involve dramatic conformational change of the serpin and translocation of the proteinase from one pole to the other. These new structures significantly extend the currently very limited range of serpin-proteinase pairs for which primary structural information is available and support the idea that a common mechanism of serine proteinase inhibition is used by serpins. We have also identified a very plausible explanation for the earlier contradictory conclusion on the structure of the same non-covalent complex. We thank Dr. Steven Olson for helpful discussion and for comments on the manuscript and Dr. Klavs Dolmer for invaluable help with protein expression and purification." @default.
• W1997483799 created "2016-06-24" @default.
• W1997483799 creator A5022122582 @default.
• W1997483799 creator A5051618397 @default.
• W1997483799 date "2004-03-01" @default.
• W1997483799 modified "2023-09-28" @default.
• W1997483799 title "α1-Proteinase Inhibitor Forms Initial Non-covalent and Final Covalent Complexes with Elastase Analogously to Other Serpin-Proteinase Pairs, Suggesting a Common Mechanism of Inhibition" @default.
• W1997483799 cites W1586249833 @default.
• W1997483799 cites W1588954462 @default.
• W1997483799 cites W1645568321 @default.
• W1997483799 cites W1912762104 @default.
• W1997483799 cites W1963500684 @default.
• W1997483799 cites W1971611954 @default.
• W1997483799 cites W1978220538 @default.
• W1997483799 cites W1995582010 @default.
• W1997483799 cites W1996362995 @default.
• W1997483799 cites W1996455719 @default.
• W1997483799 cites W2006293624 @default.
• W1997483799 cites W2009304073 @default.
• W1997483799 cites W2011955429 @default.
• W1997483799 cites W2015720097 @default.
• W1997483799 cites W2023454712 @default.
• W1997483799 cites W2024750173 @default.
• W1997483799 cites W2031007165 @default.
• W1997483799 cites W2032615313 @default.
• W1997483799 cites W2045776240 @default.
• W1997483799 cites W2050353991 @default.
• W1997483799 cites W2065420594 @default.
• W1997483799 cites W2070970604 @default.
• W1997483799 cites W2074729875 @default.
• W1997483799 cites W2074759290 @default.
• W1997483799 cites W2079859332 @default.
• W1997483799 cites W2100534168 @default.
• W1997483799 cites W2132657670 @default.
• W1997483799 cites W2159046147 @default.
• W1997483799 cites W2169821755 @default.
• W1997483799 cites W2170327826 @default.
• W1997483799 doi "https://doi.org/10.1074/jbc.m311731200" @default.
• W1997483799 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/14593107" @default.
• W1997483799 hasPublicationYear "2004" @default.
• W1997483799 type Work @default.
• W1997483799 sameAs 1997483799 @default.
• W1997483799 citedByCount "28" @default.
• W1997483799 countsByYear W19974837992013 @default.
• W1997483799 countsByYear W19974837992014 @default.
• W1997483799 countsByYear W19974837992015 @default.
• W1997483799 countsByYear W19974837992016 @default.
• W1997483799 countsByYear W19974837992017 @default.
• W1997483799 countsByYear W19974837992018 @default.
• W1997483799 countsByYear W19974837992020 @default.
• W1997483799 countsByYear W19974837992022 @default.
• W1997483799 countsByYear W19974837992023 @default.
• W1997483799 crossrefType "journal-article" @default.
• W1997483799 hasAuthorship W1997483799A5022122582 @default.
• W1997483799 hasAuthorship W1997483799A5051618397 @default.
• W1997483799 hasBestOaLocation W19974837991 @default.
• W1997483799 hasConcept C104317684 @default.
• W1997483799 hasConcept C111472728 @default.
• W1997483799 hasConcept C138885662 @default.
• W1997483799 hasConcept C159654299 @default.
• W1997483799 hasConcept C163764329 @default.
• W1997483799 hasConcept C178790620 @default.
• W1997483799 hasConcept C180577832 @default.
• W1997483799 hasConcept C181199279 @default.
• W1997483799 hasConcept C185592680 @default.
• W1997483799 hasConcept C21194631 @default.
• W1997483799 hasConcept C2776326425 @default.
• W1997483799 hasConcept C2776714187 @default.
• W1997483799 hasConcept C2777807008 @default.
• W1997483799 hasConcept C2778965386 @default.
• W1997483799 hasConcept C2781044401 @default.
• W1997483799 hasConcept C2909000736 @default.
• W1997483799 hasConcept C3018687534 @default.
• W1997483799 hasConcept C54355233 @default.
• W1997483799 hasConcept C55493867 @default.
• W1997483799 hasConcept C71240020 @default.
• W1997483799 hasConcept C86803240 @default.
• W1997483799 hasConcept C89611455 @default.
• W1997483799 hasConceptScore W1997483799C104317684 @default.
• W1997483799 hasConceptScore W1997483799C111472728 @default.
• W1997483799 hasConceptScore W1997483799C138885662 @default.
• W1997483799 hasConceptScore W1997483799C159654299 @default.
• W1997483799 hasConceptScore W1997483799C163764329 @default.
• W1997483799 hasConceptScore W1997483799C178790620 @default.
• W1997483799 hasConceptScore W1997483799C180577832 @default.
• W1997483799 hasConceptScore W1997483799C181199279 @default.
• W1997483799 hasConceptScore W1997483799C185592680 @default.
• W1997483799 hasConceptScore W1997483799C21194631 @default.
• W1997483799 hasConceptScore W1997483799C2776326425 @default.
• W1997483799 hasConceptScore W1997483799C2776714187 @default.
• W1997483799 hasConceptScore W1997483799C2777807008 @default.
• W1997483799 hasConceptScore W1997483799C2778965386 @default.
• W1997483799 hasConceptScore W1997483799C2781044401 @default.
• W1997483799 hasConceptScore W1997483799C2909000736 @default.
• W1997483799 hasConceptScore W1997483799C3018687534 @default.
• W1997483799 hasConceptScore W1997483799C54355233 @default.
• W1997483799 hasConceptScore W1997483799C55493867 @default.
• W1997483799 hasConceptScore W1997483799C71240020 @default.

• next