The structural basis of a conserved P2 threonine in canonical serine proteinase inhibitors

Bowman-Birk inhibitors (BBIs) are a well-studied family of canonical inhibitor proteins of serine proteinases. In nature, the active region of BBIs possesses a highly conserved Thr at the P2 position. The importance of this residue has been reemphasized by synthetic BBI reactive site loop proteinomimetics. In particular, this residue was exclusively identified for active chymotrypsin inhibitors selected from a BBI template-assisted combinatorial peptide library. A further kinetic analysis of 26 P2 variant peptides revealed that Thr provides both optimal binding affinity and optimal resistance against enzymatic turnover by chymotrypsin. Herein, we report the (1)H-NMR spectroscopic study of a 5-membered sub-set of these reactive site loop peptides representing a stepwise elimination of the Thr side-chain functionalities and inversion of its side-chain chirality. The P2 Thr variant adopts a three-dimensional structure that closely mimics the one of the corresponding region of the complete protein. This validates the use of this template for the investigation of structure-function relationships. While the overall backbone geometry is similar in all studied variants, conformational changes induced by the modification of the P2 side chain have now been identified and provide a rational explanation of the kinetically observed functional differences. Eliminating the gamma-methyl group has little structural effect, whereas the elimination of the gamma-oxygen atom or the inversion of the side-chain chirality results in characteristic changes to the intramolecular hydrogen bond network. We conclude that the transannular hydrogen bond between the P2 Thr side-chain hydroxyl and the P5' backbone amide is an important conformational constraint and directs the hydrophobic contact of the P2 Thr side chain with the enzyme surface in a functionally optimal geometry, both in the proteinomimetic and the native protein. In at least four canonical inhibitor protein families similar structural arrangements for a conserved P2 Thr have been observed, which suggests an analogous functional role. Substitutions at P2 of the proteinomimetic also affect the conformational balance between cis and trans isomers at a distant Pro-Pro motif (P3'-P4'). Presented with a mixture of cis/trans isomers chymotrypsin appears to interact preferably with the conformer that retains the cis-P3' Pro-trans-P4' Pro geometry found in the parent BBI protein.     

Selection of human elastase inhibitors from a conformationally constrained combinatorial peptide library.

A resin-bound cyclic peptide library was constructed based on the sequence of the reactive-site loop of Bowman-Birk inhibitor, a proteinase inhibitor protein. The constrained loop sequence, which incorporates the minimal proteinase-binding motif, was retained throughout the library, but selected residues known to be important for inhibitor specificity were randomised. The approach was used to create a 'one bead, one peptide' library with 8000 variants resulting from randomization at three target locations in the sequence (P4, P1 and P2'). This library allows us to examine the degree to which variations in this proteinase-binding motif can redirect activity, as well as providing information about the binding specificity of a proteinase target. Screening this library for binding to human leucocyte elastase identified sequences with a strong consensus, and on resynthesis all were found to act as inhibitors, with Ki values as low as 65 nM. Human leucocyte elastase is known to have a substrate preference for small alkyl chains at the P1 locus, with valine being preferred. However, alanine and not the expected valine was found in 21 out of 23 identified sequences. The remaining two sequences had threonine at P1, a finding that would be hard to predict based on substrate specificity alone. Further analysis of resynthesized peptides demonstrated that valine substitution results in an analogue that is hydrolysed far more rapidly than ones having library-selected P1 residues. Testing of the human leucocyte elastase-selected sequences as inhibitors of porcine pancreatic elastase demonstrates a significant difference in the specificity of the P4 locus between these two proteinases.     

The Structural Basis of a Conserved P2 Threonine in Canonical Serine Proteinase Inhibitors

Abstract

Bowman-Birk inhibitors (BBIs) are a well-studied family of canonical inhibitor proteins of serine proteinases. In nature, the active region of BBIs possesses a highly conserved Thr at the P2 position. The importance of this residue has been reemphasized by synthetic BBI reactive site loop proteinomimetics. In particular, this residue was exclusively identified for active chymotrypsin inhibitors selected from a BBI template-assisted combinatorial peptide library. A further kinetic analysis of 26 P2 variant peptides revealed that Thr provides both optimal binding affinity and optimal resistance against enzymatic turnover by chymotrypsin.

Herein, we report the H-NMR spectroscopic study of a 5-membered sub-set of these reactive site loop peptides representing a stepwise elimination of the Thr side-chain functionalities and inversion of its side-chain chirality. The P2 Thr variant adopts a three-dimensional structure that closely mimics the one of the corresponding region of the complete protein. This validates the use of this template for the investigation of structure-function relationships. While the overall backbone geometry is similar in all studied variants, conformational changes induced by the modification of the P2 side chain have now been identified and provide a rational explanation of the kinetically observed functional differences. Eliminating the γ-methyl group has little structural effect, whereas the elimination of the γ-oxygen atom or the inversion of the side-chain chirality results in characteristic changes to the intramolecular hydrogen bond network. We conclude that the transannular hydrogen bond between the P2 Thr side-chain hydroxyl and the P5′ backbone amide is an important conformational constraint and directs the hydrophobic contact of the P2 Thr side chain with the enzyme surface in a functionally optimal geometry, both in the proteinomimetic and the native protein.

In at least four canonical inhibitor protein families similar structural arrangements for a conserved P2 Thr have been observed, which suggests an analogous functional role. Substitutions at P2 of the proteinomimetic also affect the conformational balance between cis and trans isomers at a distant Pro-Pro motif (P3′-P4′). Presented with a mixture of cis/trans isomers chymotrypsin appears to interact preferably with the conformer that retains the cis-P3′ Pro-trans-P4′ Pro geometry found in the parent BBI protein.     

Design of serine proteinase inhibitors by combinatorial chemistry using trypsin inhibitor SFTI-1 as a starting structure.

A small peptide library of monocyclic SFTI-1 trypsin inhibitor from sunflower seeds modified in positions P(1) and P(4)' was synthesized using a portioning-mixing method. The peptide library was deconvoluted by the iterative approach in solution. Two trypsin ([Met(9)]-SFTI-1 and [Arg(5), Abu(9)]-SFTI-1), one chymotrypsin ([Phe(5)]-SFTI-1) and one human elastase ([Leu(5), Trp(9)]-SFTI-1) inhibitors were selected and resynthesized. The values of their association equilibrium constants (K(a)) with target enzymes indicate that they are potent inhibitors. In addition, the last two analoges belong to the most active inhibitors of this size. The results obtained show that the conserved Pro(9) residue in the Bowman-Birk inhibitor (BBI)s is not essential for inhibitory activity.     

Characterization of potato proteinase inhibitor II reactive site mutants.

Potato proteinase inhibitor II (PI-2) is composed of two sequence repeats. It contains two reactive site domains. We developed an improved protocol for the production of PI-2 using the yeast Pichia pastoris as the expression host. We then assessed the role of its two reactive sites in the inhibition of trypsin and chymotrypsin by mutating each of the two reactive sites in various ways. From these studies it appears that the second reactive site strongly inhibits both trypsin (Ki = 0.4 nM) and chymotrypsin (Ki = 0.9 nM), and is quite robust towards mutations at positions P2 or P1'. In contrast, the first reactive site inhibits only chymotrypsin (Ki = 2 nM), and this activity is very sensitive to mutations. Remarkably, replacing the reactive site amino acids of domain I with those of domain II did not result in inhibitory activities similar to domain II. The fitness for protein engineering of each domain is discussed.     

Proteinase inhibition using small Bowman-Birk-type structures

Bowman-Birk inhibitors (BBIs) are cysteine-rich and highly cross-linked small proteins that function as specific pseudosubstrates for digestive proteinases. They typically display a "double-headed" structure containing an independent proteinase-binding loop that can bind and inhibit trypsin, chymotrypsin and elastase. In the present study, we used computational biology to study the structural characteristics and dynamics of the inhibition mechanism of the small BBI loop expressing a 35-amino acid polypeptide (ChyTB2 inhibitor) which has coding region for the mutated chymotrypsin-inhibitory site of the soybean BBI. We found that in the BBI-trypsin inhibition complex, the most important interactions are salt bridges and hydrogen bonds, whereas in the BBI-chymotrypsin inhibition complex, the most important interactions are hydrophobic. At the same time, ChyTB2 mutant structure maintained the individual functional domain structure and excellent binding/inhibiting capacities for trypsin and chymotrypsin at the same time. These results were confirmed by enzyme-linked immunosorbend assay experiments. The results showed that modeling combined with molecular dynamics is an efficient method to describe, predict and then obtain new proteinase inhibitors. For such study, however, it is necessary to start from the sequence and structure of the mutant interacting relatively strongly with both trypsin and chymotrypsin for designing the small BBI-type inhibitor against proteinases.     

Albumin is a substrate of human chymase. Prediction by combinatorial peptide screening and development of a selective inhibitor based on the albumin cleavage site

Human chymase is a chymotryptic serine peptidase stored and secreted by mast cells. Compared with other chymotryptic enzymes, such as cathepsin G and chymotrypsin, it is much more slowly inhibited by serum serpins. Although chymase hydrolyzes several peptides and proteins in vitro, its target repertoire is limited compared with chymotrypsin because of selective interactions in an extended substrate-binding site. The best-known natural substrate, angiotensin I, is cleaved to generate vasoactive angiotensin II. Selectivity of angiotensin cleavage depends in major part on interactions involving substrate residues on the carboxyl-terminal (P1'-P2') side of the cleaved bond. To identify new targets based on interactions with residues on the aminoterminal (P4-P1) side of the site of hydrolysis, we profiled substrate preferences of recombinant human chymase using a combinatorial, fluorogenic peptide substrate library. Data base queries using the peptide (Arg-Glu-Thr-Tyr-X) generated from the most preferred amino acid at each subsite identify albumin as the sole, soluble, human extracellular protein containing this sequence. We validate the prediction that this site is chymase-susceptible by showing that chymase hydrolyzes albumin uniquely at the predicted location, with the resulting fragments remaining disulfide-linked. The site of hydrolysis is highly conserved in vertebrate albumins and is near predicted sites of metal cation binding, but nicking by chymase does not alter binding of Cu2+ or Zn2+. A synthetic peptidic inhibitor, diphenyl N alpha-benzoxycarbonyl-l-Arg-Glu-Thr-PheP-phosphonate, was designed from the preferred P4-P1 substrate sequence. This inhibitor is highly potent (IC50 3.8 nM) and 2,700- and 1,300-fold selective for chymase over cathepsin G and chymotrypsin, respectively. In summary, these findings reveal albumin to be a substrate for chymase and identify a potentially useful new chymase inhibitor.     

Synthetic peptides and fluorogenic substrates related to the reactive site sequence of Kunitz-type inhibitors isolated from Bauhinia: interaction with human plasma kallikrein

We have previously described Kunitz-type serine proteinase inhibitors purified from Bauhinia seeds. Human plasma kallikrein shows different susceptibility to those inhibitors. In this communication, we describe the interaction of human plasma kallikrein with fluorogenic and non-fluorogenic peptides based on the Bauhinia inhibitors' reactive site. The hydrolysis of the substrate based on the B. variegata inhibitor reactive site sequence, Abz-VVISALPRSVFIQ-EDDnp (Km 1.42 microM, kcat 0.06 s(-1), and kcat/Km 4.23 x 10(4) M(-1) s(-1)), is more favorable than that of Abz-VMIAALPRTMFIQ-EDDnp, related to the B. ungulata sequence (Km 0.43 microM, kcat 0.00017 s(-1), and kcat/Km 3.9 x 10(2) M(-1) s(-1)). Human plasma kallikrein does not hydrolyze the substrates Abz-RPGLPVRFESPL-EDDnp and Abz-FESPLRINIIKE-EDDnp based on the B. bauhinioides inhibitor reactive site sequence, the most effective inhibitor of the enzyme. These peptides are competitive inhibitors with Ki values in the nM range. The synthetic peptide containing 19 amino acids based on the B. bauhinioides inhibitor reactive site (RPGLPVRFESPL) is poorly cleaved by kallikrein. The given substrates are highly specific for trypsin and chymotrypsin hydrolysis. Other serine proteinases such as factor Xa, factor XII, thrombin and plasmin do not hydrolyze B. bauhinioides inhibitor related substrates.     

A conserved cis peptide bond is necessary for the activity of Bowman-Birk inhibitor protein

The Bowman-Birk inhibitor (BBI) family of protease inhibitors has an inhibitory region comprising a disulfide-linked nine-residue loop that adopts the characteristic canonical motif found in many serine protease inhibitors. A unique feature of the BBI loop is the presence of a cis peptide bond at the edge of the inhibitory loop. BBI-related protein fragments that encapsulate this loop retain the structure and inhibitory activity of the parent protein. The most common BBI loop sequence has a proline-proline element with a cis-trans geometry at P3'-P4'. We have examined this element by analysis of the inhibitory activity and structure for a series of synthetic fragments where each of these proline residues has been systematically replaced with alanine. The results show that only when a proline is present at P3' are potent inhibition and a cis peptide bond at that position in the solution structure observed, suggesting that this conformation is required for biological activity. Though a P4' proline is not essential for activity, it effectively stabilizes the cis conformation at P3' by suppressing alternative conformations. This is most evident from the Pro-Ala variant, which comprises a 1:1 mixture of slowly exchanging and structurally different cis and trans isomers. Monitoring the action of trypsin on this mixture by NMR shows that this protease interacts selectively with the cis P3' structure, providing direct evidence for the link between activity and the nativelike structure of the cis isomer. This is, to the best of our knowledge, the first example where cis isomer selectivity can be demonstrated for a proteinase.     

Purification, characterization and primary structure of a chymotrypsin inhibitor from Naja atra venom

A chymotrypsin inhibitor, designated NA-CI, was isolated from the venom of the Chinese cobra Naja atra by three-step chromatography. It inhibited bovine alpha-chymotrypsin with a Ki of 25 nM. The molecular mass of NA-CI was determined to be 6403.8 Da by matrix-assisted laser-desorption ionization time-of-flight (MALDI-TOF) analysis. The complete amino acid sequence was determined after digestion of S-carboxymethylated inhibitor with Staphylococcus aureus V8 protease and porcine trypsin. NA-CI was a single polypeptide chain composed of 57 amino acid residues. The main contact site with the protease (P1) has a Phe, showing the specificity of the inhibitor. NA-CI shared great similarity with the chymotrypsin inhibitor from Naja naja venom (identities=89.5%) and other snake venom protease inhibitors.     

Purification and characterization of a chymotrypsin inhibitor from the venom of Ophiophagus hannah (King Cobra)

A chymotrypsin inhibitor from the venom of Ophiophagus hannah was isolated by a combination of ion-exchange chromatography and reverse phase HPLC. Amino acid sequence analysis revealed that this protein consists of 58 amino acids, six of these being cysteine residues and is highly homologous to Kunitz-type protease inhibitors. ESI-mass spectrum showed that the protein had a mass of 6493, which is in agreement with that predicted from its primary structure. In contrast to P1 Leu, Met, Phe, Trp, and Tyr appearing in other chymotrypsin inhibitors, a P1 Asn in the novel inhibitor may cause a weak binding (Ki = 3.52 microM) with chymotrypsin. Phylogenetic analysis suggests that the functional variations of the chymotrypsin inhibitor and other Kunitz-type inhibitors probably distinguish from dendrotoxins by accelerated evolution.     

Inhibition of human beta-tryptase by Bowman-Birk inhibitor derived peptides: creation of a new tri-functional inhibitor

Bowman-Birk inhibitor proteins (BBIs), which are potent inhibitors of chymotrypsin-like proteases, do not inhibit human beta-tryptase despite this protein having a chymotrypsin-like fold. We have reported previously that, in contrast, BBI-derived peptides (whose sequences incorporate the solvent exposed reactive site loop motif) are able to inhibit human beta-tryptase. This is due to their small size, which allows them to access the restricted active site(s) of tryptase, which has an unusual tetrameric arrangement with four active sites flanking a central pore. In this paper, we have examined the possibility of creating additional interactions within this pore by adding extensions to the BBI-peptide motif. We have taken the core disulfide-bridged sequence SCTKSIPPQCY and examined a series of extensions, at both the C- and N-termini, that bear a second positively charged Lys residue at their end. The aim was to construct inhibitors that could make additional interactions in tryptase by spanning the gap between adjacent active sites in the enzyme, producing a double-headed inhibitor; a positively charged group was used as the dominant specificity of this enzyme is for a positively charged P1 residue. Both N- and C-terminal extensions are found to produce inhibitors of much increased potency, with a strong dependence of potency on chain length. Moreover, it was found that the C- and N-terminal extensions were able to synergise, with their combination on the same peptide producing an even better inhibitor with a potency 10(4)-fold greater than the original sequence. We suggest that the C- and N-terminal extensions are picking up interactions with separate additional sites on the tryptase, making the doubly extended BBI peptide a tri-functional tryptase inhibitor.     

Kinetic properties of the binding of alpha-lytic protease to peptide boronic acids

The kinetic parameters for peptide boronic acids in their interaction with alpha-lytic protease were determined and found to be similar to those of other serine proteases [Kettner, C., & Shenvi, A. B. (1984) J. Biol. Chem. 259, 15106-15114]. alpha-Lytic protease hydrolyzes substrates with either alanine or valine in the P1 site and has a preference for substrate with a P1 alanine. The most effective inhibitors are tri- and tetrapeptide analogues that have a -boroVal-OH residue in the P1 site. At pH 7.5, MeOSuc-Ala-Ala-Pro-boroVal-OH has a Ki of 6.4 nM and Boc-Ala-Pro-boroVal-OH has a Ki of 0.35 nM. Ac-boroVal-OH and Ac-Pro-boroVal-OH are 220,000- and 500-fold less effective, respectively, than the tetrapeptide analogue. The kinetic properties of the tri- and tetrapeptide analogues are consistent with the mechanism for slow-binding inhibition, E + I in equilibrium EI in equilibrium EI*, while the less effective inhibitors are simple competitive inhibitors. MeO-Suc-Ala-Ala-Pro-boroAla-OH is a simple competitive inhibitor with a Ki of 67 nM at pH 7.5. Other peptide boronic acids, which are analogues of nonsubstrates, are less effective than substrate analogues but still are effective competitive inhibitors. For example, MeOSuc-Ala-Ala-Pro-boroPhe-OH has a Ki of 0.54 microM although substrates with a phenylalanine in the P1 position are not hydrolyzed. Binding for boronic acid analogues of both substrate and nonsubstrate analogues is pH dependent with higher affinity near pH 7.5. Similar binding properties have been observed for pancreatic elastase. Both enzymes have almost identical requirements for an extended peptide inhibitor sequence in order to exhibit highly effective binding and slow-binding characteristics.(ABSTRACT TRUNCATED AT 250 WORDS)     

A Kazal-type trypsin inhibitor from the protochordate Ciona intestinalis.

A trypsin inhibitor from Ciona intestinalis, present throughout the animal, was purified by ion-exchange chromatography followed by four HPLC steps. By MS the molecular mass of the native form was determined to be 6675 Da. The N-terminal amino acid sequence was determined by protein sequencing, but appeared to be partial because the theoretical molecular mass of the protein was 1101 Da too low. Thermolysin treatment gave rise to several fragments each containing a single disulphide bridge. By sequence analysis and MS intramolecular disulphide bridges could unequivocally be assigned to connect the pairs Cys4-Cys37, Cys8-Cys30 and Cys16-Cys51. The structure of the inhibitor is homologous to Kazal-type trypsin inhibitors. The inhibitor constant, KI, for trypsin inhibition was 0.05 nM whereas chymotrypsin and elastase were not inhibited. To reveal the complete sequence the cDNA encoding the trypsin inhibitor was isolated. This cDNA of 454 bp predicts a protein of 82 amino acid residues including a 20 amino acid signal peptide. Moreover, the cDNA predicts a C-terminal extension of 11 amino acids compared to the part identified by protein sequencing. The molecular mass calculated for this predicted protein is in accordance with the measured value. This C-terminal sequence is unusual for Kazal-type trypsin inhibitors and has apparently been lost early in evolution. The high degree of conservation around the active site strongly supports the importance of the Kazal-type inhibitors.     

Preparation of photoreactive derivatives of trypsin-chymotrypsin inhibitors from soybeans and chick peas by selective modification of lysine residues

Photoreactive derivatives of the Bowman-Birk trypsin-chymotrypsin inhibitor (BBI) from soybeans and of CI, the trypsin-chymotrypsin inhibitor from chick peas, were prepared by selective modification of the epsilon-amino groups of lysine residues with 2-nitro-4(5)-azidophenylsulfenyl chlorides (2,4(5)-NAPS-C1). The ultraviolet absorption spectra of the photolabeled inhibitors indicated that three out of the five lysines of BBI and one of the seven lysines of CI were modified. The inhibitory activity of the modified inhibitors towards trypsin and chymotrypsin was not reduced even after photolysis. The specific lysine residues that constitute the trypsin-inhibitory sites of BBI and CI did not react with the photoreactive reagents. Further modification of the photoreactive derivatives of BBI and CI with maleic anhydride, directed towards the trypsin-reactive sites, resulted in almost complete loss of the trypsin-inhibiting activity without reducing the ability to inhibit chymotrypsin. A pronounced potentiation effect (approximately 2x) of the chymotrypsin inhibiting activity was noted for 2,5-NAPS-CI and it was retained even after maleylation followed by photolysis, raising the possibility of exposure of an additional chymotrypsin inhibitory site in CI.     

Subtilisin Protein Inhibitor from Potato Tubers

A protein with molecular weight of 21 kD denoted as PKSI has been isolated from potato tubers (Solanum tuberosum L., cv. Istrinskii). The isolation procedure includes precipitation with (NH4)2SO4, gel chromatography on Sephadex G-75, and ion-exchange chromatography on CM-Sepharose CL-6B. The protein effectively inhibits the activity of subtilisin Carlsberg (Ki = 1.67 +/- 0.2 nM) by stoichiometric complexing with the enzyme at the molar ratio of 1 : 1. The inhibitor has no effect on trypsin, chymotrypsin, and the cysteine proteinase papain. The N-terminal sequence of the protein consists of 19 amino acid residues and is highly homologous to sequences of the known inhibitors from group C of the subfamily of potato Kunitz-type proteinase inhibitors (PKPIs-C). By cloning PCR products from the genomic DNA of potato, a gene denoted as PKPI-C2 was isolated and sequenced. The N-terminal sequence (residues from 15 to 33) of the protein encoded by the PKPI-C2 gene is identical to the N-terminal sequence (residues from 1 to 19) of the isolated protein PKSI. Thus, the inhibitor PKSI is very likely encoded by this gene.     

Low molecular weight serine protease inhibitors from insects are proteins with highly conserved sequences

A low molecular weight protease inhibitor peptide found in ovaries of the desert locust Schistocerca gregaria (SGPI-2), was purified from plasma of the same locust and sequenced. It was named SGCI. It was found active towards chymotrypsin and human leukocyte elastase. SGCI was synthesized using a solid-phase procedure and the sequence of its reactive site for chymotrypsin was determined. Compared with an inhibitor purified earlier from another locust species, the total sequence of SGCI showed 88% identity. In particular, the sequence of the reactive site of these inhibitors was identical. Our search for a closely related peptide in an insect species far removed from locusts, the lepidopteran Spodoptera littoralis, was unfruitful but a different chymotrypsin inhibitor, belonging to the Kazal family, was found whose mass is greater than that of SGCI (20 vs 3.6 kDa). Its N-terminal sequence shares 80% identity with that of a chymotrypsin inhibitor purified earlier from the haemolymph of another lepidopteran. Conservation of the amino acid sequence in the reactive site seems to be an exception among protease inhibitors.     

Substrate-dependent inhibition kinetics of an active site-directed inhibitor of ADAMTS-4 (Aggrecanase 1)

ADAMTS-4 (aggrecanase-1) is implicated in the breakdown of articular cartilage and is an attractive target for therapeutic intervention in arthritis. Cleavage of the native substrate, aggrecan, occurs through exosite interactions and peptide sequence recognition. Although expected to be competitive with aggrecan, the hydroxamic acid, SC81956, demonstrated noncompetitive inhibition kinetics with a Ki of 23 nM. The IC50 of SC81956 did not change when aggrecan was varied from 12.8 to 200 nM (0.2-3.3 times the apparent aggrecan Km of 61 nM) but was shifted as expected for a competitive inhibitor when increasing levels of a low molecular weight peptide substrate were added to a fluorogenic peptide assay system. These observations are consistent with a model for aggrecan cleavage where substrate initially binds at an exosite, followed by binding of the appropriate peptide sequence at the active site. A peptide-competitive inhibitor could bind both free enzyme and initial substrate-enzyme exosite complex but would be excluded by the final Michaelis complex. Noncompetitive appearing kinetics for such inhibitors is predicted as long as the equilibrium between the two forms of enzyme-substrate complex significantly favors the initial exosite complex. In support, hydrolysis of a low molecular weight peptide substrate and its inhibition by SC81956 were unaffected by aggrecan concentrations substantially above the Km. These observations suggest that the apparent Km for aggrecan cleavage predominately reflects the exosite interaction. Consequently, the efficacy of active-site inhibitors of ADAMTS-4 will not be limited by competition with native substrate as predicted from the Km determined by traditional kinetic models.     

Toxicity to the pea aphid Acyrthosiphon pisum of anti-chymotrypsin isoforms and fragments of Bowman-Birk protease inhibitors from pea seeds

Aphids feed on a protein-poor diet and are insensitive to several serine protease inhibitors. However, among the Bowman-Birk family of plant trypsin inhibitors (BBI), some members display significant toxicity to the pea aphid Acyrthosiphon pisum. A BBI isoform purified from pea seeds (PsTI-2) displays an IC50 of 41 microM and a LC50 of 48 microM at 7 days. Our data show that the chymotrypsin-directed active site from these bifunctional inhibitors is responsible for this activity, and that artificial cyclic peptides bearing the Bowman-Birk anti-chymotrypsin head induce much greater toxicity and growth inhibition than their anti-trypsin counterparts. The toxic syndrome included a rapid behavioural response of aphids on diets containing the toxic peptides, with induced restlessness after only 1 h of exposure to the chymotrypsin inhibitor. Nevertheless, chymotrypsin activity was not detected in aphid guts, using two chromogenic chymotrypsin substrates, and the physiological target of the chymotrypsin inhibitor remains unknown. These data show for the first time that plant chymotrypsin inhibitors, still widely unexplored, may act as paradoxical toxicants to aphids and serve as defensive metabolites for phloem-feeding insects.     

Serpins of oat (Avena sativa) grain with distinct reactive centres and inhibitory specificity

Most proteinase inhibitors from plant seeds are assumed to contribute to broad-spectrum protection against pests and pathogens. In oat (Avena sativa L.) grain the main serine proteinase inhibitors were found to be serpins, which utilize a unique mechanism of irreversible inhibition. Four distinct inhibitors of the serpin superfamily were detected by native PAGE as major seed albumins and purified by thiophilic adsorption and anion exchange chromatography. The four serpins OSZa-d are the first proteinase inhibitors characterized from this cereal. An amino acid sequence close to the blocked N-terminus, a reactive centre loop sequence, and the second order association rate constant (ka') for irreversible complex formation with pancreas serine proteinases at 24 degrees C were determined for each inhibitor. OSZa and OSZb, both with the reactive centre scissile bond P1-P1' Thr downward arrow Ser, were efficient inhibitors of pancreas elastase (ka' > 105M-1 s-1). Only OSZb was also an inhibitor of chymotrypsin at the same site (ka' = 0.9 x 105M-1 s-1). OSZc was a fast inhibitor of trypsin at P1-P1' Arg downward arrow Ser (ka' = 4 x 106M-1 s-1); however, the OSZc-trypsin complex was short-lived with a first order dissociation rate constant kd = 1.4 x 10-4 s-1. OSZc was also an inhibitor of chymotrypsin (ka' > 106M-1 s-1), presumably at the overlapping site P2-P1 Ala downward arrow Arg, but > 90% of the serpin was cleaved as substrate. OSZd was cleaved by chymotrypsin at the putative reactive centre bond P1-P1' Tyr downward arrow Ser, and no inhibition was detected. Together the oat grain serpins have a broader inhibitory specificity against digestive serine proteinases than represented by the major serpins of wheat, rye or barley grain. Presumably the serpins compensate for the low content of reversible inhibitors of serine proteinases in oats in protection of the grain against pests or pathogens.