Hyperthermophilic and salt-dependent formyltransferase from Methanopyrus kandleri.

Methanopyrus kandleri is a hyperthermophilic methanogenic archaeon, which grows on H(2) and CO(2) as its sole energy source. Its growth temperature optimum is 98 degrees C. One of the interesting characteristics of this archaeon is its high intracellular salt content. The organism has been reported to contain the trianionic cDPG (cyclic 2,3-diphosphoglycerate) and K+ at concentrations of 1.1 and 3 M, respectively. Reflecting the high cellular salt concentration, the enzymes in this organism are adapted not only to high temperature but also to high salt concentrations. The formyltransferase from M. kandleri was characterized extensively with respect to thermo- and halophilicity. The crystal structure of the formyltransferase at 1.73 A shows the enzyme to be composed of four identical subunits of molecular mass 32 kDa. The formyltransferase is thermostable and active only at relatively high concentrations of potassium phosphate (1 M) or other salts with strongly hydrated anions (strong salting-out salts). Potassium phosphate and potassium cDPG were found to be equivalent in activating and stabilizing the enzyme. At low concentrations of these salts, the enzyme is inactive and thermolabile. It was shown by equilibrium sedimentation analysis that the enzyme is in a monomer/dimer/tetramer equilibrium, the equilibrium constant being dependent on the concentration of salts: the higher oligomeric species increase with increasing salt concentrations. Evidence was provided that the monomer is both inactive and thermolabile. Experiments using a mutation which is directed to break surface ion pairs between two dimers indicated that dimerization is required for activity and tetramerization leads to thermostability.

Structures of F420H2:NADP+ oxidoreductase with and without its substrates bound.

Cofactor F420 is a 5'-deazaflavin derivative first discovered in methanogenic archaea but later found also to be present in some bacteria. As a coenzyme, it is involved in hydride transfer reactions and as a prosthetic group in the DNA photolyase reaction. We report here for the first time on the crystal structure of an F420-dependent oxidoreductase bound with F420. The structure of F420H2:NADP+ oxidoreductase resolved to 1.65 A contains two domains: an N-terminal domain characteristic of a dinucleotide-binding Rossmann fold and a smaller C-terminal domain. The nicotinamide and the deazaflavin part of the two coenzymes are bound in the cleft between the domains such that the Si-faces of both face each other at a distance of 3.1 A, which is optimal for hydride transfer. Comparison of the structures bound with and without substrates reveals that of the two substrates NADP has to bind first, the binding being associated with an induced fit.

On the mechanism of biological methane formation: structural evidence for conformational changes in methyl-coenzyme M reductase upon substrate binding.

Methyl-coenzyme M reductase (MCR) catalyzes the final reaction of the energy conserving pathway of methanogenic archaea in which methylcoenzyme M and coenzyme B are converted to methane and the heterodisulfide CoM-S-S-CoB. It operates under strictly anaerobic conditions and contains the nickel porphinoid F430 which is present in the nickel (I) oxidation state in the active enzyme. The known crystal structures of the inactive nickel (II) enzyme in complex with coenzyme M and coenzyme B (MCR-ox1-silent) and in complex with the heterodisulfide CoM-S-S-CoB (MCR-silent) were now refined at 1.16 A and 1.8 A resolution, respectively. The atomic resolution structure of MCR-ox1-silent describes the exact geometry of the cofactor F430, of the active site residues and of the modified amino acid residues. Moreover, the observation of 18 Mg2+ and 9 Na+ ions at the protein surface of the 300 kDa enzyme specifies typical constituents of binding sites for either ion. The MCR-silent and MCR-ox1-silent structures differed in the occupancy of bound water molecules near the active site indicating that a water chain is involved in the replenishment of the active site with water molecules. The structure of the novel enzyme state MCR-red1-silent at 1.8 A resolution revealed an active site only partially occupied by coenzyme M and coenzyme B. Increased flexibility and distinct alternate conformations were observed near the active site and the substrate channel. The electron density of the MCR-red1-silent state aerobically co-crystallized with coenzyme M displayed a fully occupied coenzyme M-binding site with no alternate conformations. Therefore, the structure was very similar to the MCR-ox1-silent state. As a consequence, the binding of coenzyme M induced specific conformational changes that postulate a molecular mechanism by which the enzyme ensures that methylcoenzyme M enters the substrate channel prior to coenzyme B as required by the active-site geometry. The three different enzymatically inactive enzyme states are discussed with respect to their enzymatically active precursors and with respect to the catalytic mechanism.

X-ray structure analyses of photosynthetic reaction center variants from Rhodobacter sphaeroides: structural changes induced by point mutations at position L209 modulate electron and proton transfer.

The structures of the reaction center variants Pro L209 --> Tyr, Pro L209 --> Phe, and Pro L209 --> Glu from the photosynthetic purple bacterium Rhodobacter sphaeroides have been determined by X-ray crystallography to 2.6-2.8 A resolution. These variants were constructed to interrupt a chain of tightly bound water molecules that was assumed to facilitate proton transfer from the cytoplasm to the secondary quinone Q(B) [Baciou, L., and Michel, H. (1995) Biochemistry 34, 7967-7972]. However, the amino acid exchanges Pro L209 --> Tyr and Pro L209 --> Phe do not interrupt the water chain. Both aromatic side chains are oriented away from this water chain and interact with three surrounding polar side chains (Asp L213, Thr L226, and Glu H173) which are displaced by up to 2.6 A. The conformational changes induced by the bulky aromatic rings of Tyr L209 and Phe L209 lead to unexpected displacements of Q(B) compared to the wild-type protein. In the structure of the Pro L209 --> Tyr variant, Q(B) is shifted by approximately 4 A and is now located at a position similar to that reported for the wild-type reaction center after illumination [Stowell, M. H. B., et al. (1997) Science 276, 812-816]. In the Pro L209 --> Phe variant, the electron density map reveals an intermediate Q(B) position between the binding sites of the wild-type protein in the dark and the Pro L209 --> Tyr protein. In the Pro L209 --> Glu reaction center, the carboxylic side chain of Glu L209 is located within the water chain, and the binding site of Q(B) remains unchanged compared to the wild-type structure.

Crystallization and preliminary X-ray analysis of adenylylsulfate reductase from Archaeoglobus fulgidus.

A group of anaerobic microorganisms use sulfate as the terminal electron acceptor for energy conservation. The process of sulfate reduction involves several enzymatic steps. One of them is the conversion of adenylyl sulfate (adenosine-5'-phosphosulfate) to sulfite, catalyzed by adenylylsulfate reductase. This enzyme is composed of a FAD-containing alpha-subunit and a beta-subunit harbouring two [4Fe-4S] clusters. Adenylylsulfate reductase was isolated from Archaeoglobus fulgidus under anaerobic conditions and crystallized using the hanging-drop vapour-diffusion method using PEG 4000 as precipitant. The crystals grew in space group P2(1)2(1)2(1), with unit-cell parameters a = 72.4, b = 113.2, c = 194.0 A. The asymmetric unit probably contains two alphabeta units. The crystals diffract beyond 2 A resolution and are suitable for X-ray structure analysis.

Comparison of three methyl-coenzyme M reductases from phylogenetically distant organisms: unusual amino acid modification, conservation and adaptation.

The nickel enzyme methyl-coenzyme M reductase (MCR) catalyzes the terminal step of methane formation in the energy metabolism of all methanogenic archaea. In this reaction methyl-coenzyme M and coenzyme B are converted to methane and the heterodisulfide of coenzyme M and coenzyme B. The crystal structures of methyl-coenzyme M reductase from Methanosarcina barkeri (growth temperature optimum, 37 degrees C) and Methanopyrus kandleri (growth temperature optimum, 98 degrees C) were determined and compared with the known structure of MCR from Methanobacterium thermoautotrophicum (growth temperature optimum, 65 degrees C). The active sites of MCR from M. barkeri and M. kandleri were almost identical to that of M. thermoautotrophicum and predominantly occupied by coenzyme M and coenzyme B. The electron density at 1.6 A resolution of the M. barkeri enzyme revealed that four of the five modified amino acid residues of MCR from M. thermoautotrophicum, namely a thiopeptide, an S-methylcysteine, a 1-N-methylhistidine and a 5-methylarginine were also present. Analysis of the environment of the unusual amino acid residues near the active site indicates that some of the modifications may be required for the enzyme to be catalytically effective. In M. thermoautotrophicum and M. kandleri high temperature adaptation is coupled with increasing intracellular concentrations of lyotropic salts. This was reflected in a higher fraction of glutamate residues at the protein surface of the thermophilic enzymes adapted to high intracellular salt concentrations.

A mutation affecting the association equilibrium of formyltransferase from the hyperthermophilic Methanopyrus kandleri and its influence on the enzyme's activity and thermostability.

Formyltransferase from Methanopyrus kandleri is composed of only one type of subunit of molecular mass 32 kDa. The enzyme is in a monomer/dimer/tetramer association equilibrium, the association constant being affected by lyotropic salts. Oligomerization is required for enzyme activity and thermostability. We report here on a subunit interface mutation (R261E) which affects the dimer/tetramer part of the association equilibrium of formyltransferase. With the mutant protein it was shown that tetramerization is not required for activity but is necessary for high thermostability.

Structure of coenzyme F(420) dependent methylenetetrahydromethanopterin reductase from two methanogenic archaea.

Coenzyme F(420)-dependent methylenetetrahydromethanopterin reductase (Mer) is an enzyme of the Cl metabolism in methanogenic and sulfate reducing archaea. It is composed of identical 35-40 kDa subunits and lacks a prosthetic group. The crystal structure of Mer from Methanopyrus kandleri (kMer) revealed in one crystal form a dimeric and in another a tetrameric oligomerisation state and that from Methanobacterium thermoautotrophicum (tMer) a dimeric state. Each monomer is primarily composed of a TIM-barrel fold enlarged by three insertion regions. Insertion regions 1 and 2 contribute to intersubunit interactions. Insertion regions 2 and 3 together with the C-terminal end of the TIM-barrel core form a cleft where the binding sites of coenzyme F(420) and methylene-tetrahydromethanopterin are postulated. Close to the coenzyme F(420)-binding site lies a rarely observed non-prolyl cis-peptide bond. It is surprising that Mer is structurally most similar to a bacterial FMN-dependent luciferase which contains a non-prolyl cis-peptide bond at the equivalent position. The structure of Mer is also related to that of NADP-dependent FAD-harbouring methylenetetrahydrofolate reductase (MetF). However, Mer and MetF do not show sequence similarities although they bind related substrates and catalyze an analogous reaction.

Crystal structure of cancer chemopreventive Bowman-Birk inhibitor in ternary complex with bovine trypsin at 2.3 A resolution. Structural basis of Janus-faced serine protease inhibitor specificity.

Understanding molecular recognition on a structural basis is an objective with broad academic and applied significance. In the complexes of serine proteases and their proteinaceous inhibitors, recognition is governed mainly by residue P1 in accord with primary serine protease specificity. The bifunctional soybean Bowman-Birk inhibitor (sBBI) should, therefore, interact at LysI16 (subdomain 1) with trypsin and at LeuI43 (subdomain 2) with chymotrypsin. In contrast with this prediction, a 2:1 assembly with trypsin was observed in solution and in the crystal structure of sBBI in complex with trypsin, determined at 2.3 A resolution by molecular replacement. Strikingly, P1LeuI43 of sBBI was fully embedded into the S(1) pocket of trypsin in contrast to primary specificity. The triple-stranded beta-hairpin unique to the BBI-family and the surface loops surrounding the active site of the enzyme formed a protein-protein-interface far extended beyond the primary contact region. Polar residues, hydrophilic bridges and weak hydrophobic contacts were predominant in subdomain 1, interacting specifically with trypsin. However, close hydrophobic contacts across the interface were characteristic of subdomain 2 reacting with both trypsin and chymotrypsin. A Met27Ile replacement shifted the ratio with trypsin to the predicted 1:1 ratio. Thus, the buried salt-bridge responsible for trypsin specificity was stabilised in a polar, and destabilized in a hydrophobic, environment. This may be used for adjusting the specificity of protease inhibitors for applications such as insecticides and cancer chemopreventive agents.

The biosynthesis of methylated amino acids in the active site region of methyl-coenzyme M reductase.

The global production of the greenhouse gas methane by methanogenic archaea reaches 1 billion tons per annum. The final reaction releasing methane is catalyzed by the enzyme methyl-coenzyme M reductase. The crystal structure of methyl-coenzyme M reductase from Methanobacterium thermoautotrophicum revealed the presence of five modified amino acids within the alpha-subunit and near the active site region. Four of these modifications were C-, N-, and S-methylations, two of which, 2-(S)-methylglutamine and 5-(S)-methylarginine, have never been encountered before. We have now confirmed these modifications by mass spectrometry of chymotryptic peptides. With methyl-coenzyme M reductase purified from cells grown in the presence of L-[methyl-D(3)]methionine, it was shown that the methyl groups of the modified amino acids are derived from the methyl group of methionine rather than from methyl-coenzyme M, an intermediate in methane formation. The D(3) labeling pattern was found to be qualitatively and quantitatively the same as in the two methyl groups of the methanogenic coenzyme F(430), which are known to be introduced via S-adenosylmethionine. From the results, it is concluded that the methyl groups of the modified amino acids in methyl-coenzyme M reductase are biosynthetically introduced by an S-adenosylmethionine-dependent post-translational modification. A mechanism for the methylation of glutamine at C-2 and of arginine at C-5 is discussed.

The crystal structure of methenyltetrahydromethanopterin cyclohydrolase from the hyperthermophilic archaeon Methanopyrus kandleri.

BACKGROUND: The reduction of carbon dioxide to methane in methanogenic archaea involves the tetrahydrofolate analogue tetrahydromethanopterin (H(4)MPT) as a C(1) unit carrier. In the third step of this reaction sequence, N(5)-formyl-H(4)MPT is converted to methenyl-H(4)MPT(+) by the enzyme methenyltetrahydromethanopterin cyclohydrolase. The cyclohydrolase from the hyperthermophilic archaeon Methanopyrus kandleri (Mch) is extremely thermostable and adapted to a high intracellular concentration of lyotropic salts. RESULTS: Mch was crystallized and its structure solved at 2.0 A resolution using a combination of the single isomorphous replacement (SIR) and multiple anomalous dispersion (MAD) techniques. The structure of the homotrimeric enzyme reveals a new alpha/beta fold that is composed of two domains forming a large sequence-conserved pocket between them. Two phosphate ions were found in and adjacent to this pocket, respectively; the latter is displaced by the phosphate moiety of the substrate formyl-H(4)MPT according to a hypothetical model of the substrate binding. CONCLUSIONS: Although the exact position of the substrate is not yet known, the residues lining the active site of Mch could be tentatively assigned. Comparison of Mch with the tetrahydrofolate-specific cyclohydrolase/dehydrogenase reveals similarities in domain arrangement and in some active-site residues, whereas the fold appears to be different. The adaptation of Mch to high salt concentrations and high temperatures is reflected by the excess of acidic residues at the trimer surface and by the higher oligomerization state of Mch compared with its mesophtic counterparts.

Phospholipid bound to the flavohemoprotein from Alcaligenes eutrophus.

The structurally characterized flavohemoprotein from Alcaligenes eutrophus (FHP) contains a phospholipid-binding site with 1-16 : 0-2-cyclo-17 : 0-diacyl-glycerophospho-ethanolamine and 1-16 : 0-2-cyclo-17 : 0-diacyl-glycerophospho-glycerol as the major occupying compounds. The structure of the phospholipid is characterized by its compact form, due to the -sc/beta/-sc conformation of the glycerol and the nonlinear arrangement of the sn-1- and sn-2-fatty acid chains. The phospholipid-binding site is located adjacent to the heme molecule at the bottom of a large cavity. The fatty acid chains form a large number of van der Waal's contacts with nonpolar side chains, whereas the glycerophosphate moiety, which points towards the entrance of the channel, is linked to the protein matrix by polar interactions. The thermodynamically stable globin module of FHP, obtained after cleaving off the oxidoreductase module, also contains the phospholipid and can therefore be considered as a phospholipid-binding protein. Single amino acid exchanges designed to decrease the lipid-binding site revealed both the possibility of blocking incorporation of the phospholipid and its capability to evade steric barriers. Conformational changes in the phospholipid can also be induced by binding heme-ligating compounds. Phospholipid binding is not a general feature of flavohemoproteins, because the Escherichia coli and the yeast protein exhibit less and no lipid affinity, respectively.

Crystallization and preliminary X-ray analysis of the bacterial ATP-binding-cassette (ABC) protein MalK.

The ATP-binding protein, MalK, of the bacterial ABC (ATP-binding-cassette) transport complex MalFGK2 provides the energy for the translocation of maltose and maltodextrins across the cytoplasmic membrane. The MalK protein from Salmonella typhimurium was overexpressed in Escherichia coli and crystallized by the hanging-drop method using (NH4)2SO4as a precipitant. The crystals belong to space group P6x22 (most probably x = 1 or 5) with cell dimensions a = 181.8 and c = 182.5 A, corresponding to three or four molecules per asymmetric unit. They diffract to a resolution of about 3 A on a synchrotron X-ray source and are suitable for structure determination.

Lyotropic-salt-induced changes in monomer/dimer/tetramer association equilibrium of formyltransferase from the hyperthermophilic Methanopyrus kandleri in relation to the activity and thermostability of the enzyme.

Formyltransferase from Methanopyrus kandleri is composed of only one type of subunits of molecular mass 32 kDa. The enzyme requires the presence of lyotropic salts for activity and thermostability. We report here that the enzyme is in a monomer/dimer/tetramer association equilibrium, the association constant being affected by lyotropic salts. At 0.01 M K2HPO4/KH2PO4, pH 7.2, the enzyme (0.4 mg/ml) was mainly present in a monomeric form. Upon increase of the phosphate concentration, the concentration of the dimer increased up to a phosphate concentration of 0.6 M, then decrease at the expense of tetramer formation up to a phosphate concentration of 1.0 M. The specific activity at 4 C increased from <0.1 U/mg at 0.01 M, over 1.5 U/mg at 0.6 M to 3.6 U/mg at 1.0 M. Similar results were obtained with ammonium sulfate as lyotropic salt. The findings indicate that both oligomerization and activity increase with increasing salt concentrations, suggesting that there is a causal connection. To determine this, we exploited the observation that oligomer formation was not induced by the weak lyotropic salt NaCl up to a concentration of 1.5 M and that the dissociation of the dimer into the monomer at 4 degrees C proceeded very slowly (50% in approximately 6 h). This allowed us to study the effect of NaCl on the activity of the oligomers at NaCl concentrations not sufficient to induce oligomerization. At 4 degrees C, the activity of the oligomers increased from 0.3 U/mg at 0.25 M NaCl to 3.4 U/mg at 1.0 M NaCl. At these NaCl concentrations, the monomers were inactive. The findings indicate that oligomerization is a prerequisite for enzyme activity in the presence of NaCl. The salt-dependent induction of oligomerization was parallelled by an increase in thermostability; strong lyotropic salts conferred thermostability at much lower concentrations than the weak lyotropic NaCl.

Active sites of transition-metal enzymes with a focus on nickel.

Since 1995, crystal structures have been determined for many transition-metal enzymes, in particular those containing the rarely used transition metals vanadium, molybdenum, tungsten, manganese, cobalt and nickel. Accordingly, our understanding of how an enzyme uses the unique properties of a specific transition metal has been substantially increased in the past few years. The different functions of nickel in catalysis are highlighted by describing the active sites of six nickel enzymes - methyl-coenyzme M reductase, urease, hydrogenase, superoxide dismutase, carbon monoxide dehydrogenase and acetyl-coenzyme A synthase.

Crystal structure of methyl-coenzyme M reductase: the key enzyme of biological methane formation.

Methyl-coenzyme M reductase (MCR), the enzyme responsible for the microbial formation of methane, is a 300-kilodalton protein organized as a hexamer in an alpha2beta2gamma2 arrangement. The crystal structure of the enzyme from Methanobacterium thermoautotrophicum, determined at 1.45 angstrom resolution for the inactive enzyme state MCRox1-silent, reveals that two molecules of the nickel porphinoid coenzyme F430 are embedded between the subunits alpha, alpha', beta, and gamma and alpha', alpha, beta', and gamma', forming two identical active sites. Each site is accessible for the substrate methyl-coenzyme M through a narrow channel locked after binding of the second substrate coenzyme B. Together with a second structurally characterized enzyme state (MCRsilent) containing the heterodisulfide of coenzymes M and B, a reaction mechanism is proposed that uses a radical intermediate and a nickel organic compound.

Structure at 2.7 A resolution of the Paracoccus denitrificans two-subunit cytochrome c oxidase complexed with an antibody FV fragment.

The aa3 type cytochrome c oxidase consisting of the core subunits I and II only was isolated from the soil bacterium Paracoccus denitrificans and crystallized as complex with a monoclonal antibody Fv fragment. Crystals could be grown in the presence of a number of different nonionic detergents. However, only undecyl-beta-D-maltoside and cyclohexyl-hexyl-beta-D-maltoside yielded well-ordered crystals suitable for high resolution x-ray crystallographic studies. The crystals belong to space group P212121 and diffract x-rays to at least 2.5 A (1 A = 0.1 nm) resolution using synchrotron radiation. The structure was determined to a resolution of 2.7 A using molecular replacement and refined to a crystallographic R-factor of 20.5% (Rfree = 25.9%). The refined model includes subunits I and II and the 2 chains of the Fv fragment, 2 heme A molecules, 3 copper atoms, and 1 Mg/Mn atom, a new metal (Ca) binding site, 52 tentatively identified water molecules, and 9 detergent molecules. Only four of the water molecules are located in the cytoplasmic half of cytochrome c oxidase. Most of them are near the interface of subunits I and II. Several waters form a hydrogen-bonded cluster, including the heme propionates and the Mg/Mn binding site. The Fv fragment binds to the periplasmic polar domain of subunit II and is critically involved in the formation of the crystal lattice. The crystallization procedure is well reproducible and will allow for the analysis of the structures of mechanistically interesting mutant cytochrome c oxidases.

Formylmethanofuran: tetrahydromethanopterin formyltransferase from Methanopyrus kandleri - new insights into salt-dependence and thermostability.

BACKGROUND: Formylmethanofuran: tetrahydromethanopterin formyltransferase (Ftr) from the methanogenic Archaeon Methanopyrus kandleri (optimum growth temperature 98 degrees C) is a hyperthermophilic enzyme that is absolutely dependent on the presence of lyotropic salts for activity and thermostability. The enzyme is involved in the pathway of carbon dioxide reduction to methane and catalyzes the transfer of formyl from formylmethanofuran to tetrahydromethanopterin. RESULTS: The crystal structure of Ftr, determined to a resolution of 1:73 AE reveals a homotetramer composed essentially of two dimers. Each subunit is subdivided into two tightly associated lobes both consisting of a predominantly antiparallel beta sheet flanked by alpha helices forming an alpha/beta sandwich structure. The approximate location of the active site was detected in a region close to the dimer interface. CONCLUSIONS: The adaptation of Ftr against high lyotropic salt concentrations is structurally reflected by a large number of negatively charged residues and their high local concentration on the surface of the protein. The salt-dependent thermostability of Ftr might be explained on a molecular basis by ionic interactions at the protein surface, involving both protein and inorganic salt ions, and the mainly hydrophobic interactions between the subunits and within the core.

Crystallization and preliminary X-ray diffraction studies of methyl-coenzyme M reductase from methanobacterium thermoautotrophicum.

Methyl-coenzyme M reductase isoenzyme I from the methanogenic Archaeon, Methanobacterium thermoautotrophicum (strain Marburg), was crystallized by vapor diffusion methods. Crystal form M obtained with 2-methyl-2,4-pentanediol as the precipitant displayed space group P2(1), with unit cell parameters of a=83.2 A, b=117.4 A, c=125.1 A, and beta= 92.6 degrees, and diffracted at better than 2.8 A resolution. Crystal form P grown from polyethylene glycol 400 belonged to space group P2(1), and had unit cell parameters of a=83.1 A, b=120.2 A, c=123.1 A, and beta=91.7 degrees, diffracting at least to 1.7 A resolution. Both crystal forms have one molecule per asymmetric unit and are suitable for X-ray structure analysis.

Crystal structure of the bifunctional soybean Bowman-Birk inhibitor at 0.28-nm resolution. Structural peculiarities in a folded protein conformation.

The Bowman-Birk inhibitor from soybean is a small protein that contains a binary arrangement of trypsin-reactive and chymotrypsin-reactive subdomains. In this report, the crystal structure of this anticarcinogenic protein has been determined to 0.28-nm resolution by molecular replacement from crystals grown at neutral pH. The crystal structure differs from a previously determined NMR structure [Werner, M. H. & Wemmer, D. E. (1992) Biochemistry 31, 999-1010] in the relative orientation of the two enzyme-insertion loops, in some details of the main chain trace, in the presence of favourable contacts in the trypsin-insertion loop, and in the orientation of several amino acid side chains. The proximity of Met27 and Gln48 in the X-ray structure contradicts the solution structure, in which these two side chains point away from each other. The significant effect of a Met27-->Ile replacement on the inhibitory activity of the chymotrypsin-reactive subdomain agrees with the X-ray structure. Exposed hydrophobic patches, the presence of charged amino acid residues, and the presence of water molecules in the protein interior are in contrast to standard proteins that comprise a hydrophobic core and exposed polar amino acids.