Structural basis for the substrate specificity of a Bacillus 1,3-1,4-beta-glucanase

Depolymerization of polysaccharides is catalyzed by highly specific enzymes that promote hydrolysis of the scissile glycosidic bond by an activated water molecule. 1,3-1,4-beta-Glucanases selectively cleave beta-1,4 glycosidic bonds in 3-O-substituted glucopyranosyl units within polysaccharides with mixed linkage. The reaction follows a double-displacement mechanism by which the configuration of the anomeric C(1)-atom of the glucosyl unit in subsite -I is retained. Here we report the high-resolution crystal structure of the hybrid 1,3-1,4-beta-glucanase H(A16-M)(E105Q/E109Q) in complex with a beta-glucan tetrasaccharide. The structure shows four beta-d-glucosyl moieties bound to the substrate-binding cleft covering subsites -IV to -I, thus corresponding to the reaction product. The ten active-site residues Asn26, Glu63, Arg65, Phe92, Tyr94, Glu105, Asp107, Glu109, Asn182 and Trp184 form a network of hydrogen bonds and hydrophobic stacking interactions with the substrate. These residues were previously identified by mutational analysis as significant for stabilization of the enzyme-carbohydrate complex, with Glu105 and Glu109 being the catalytic residues. Compared to the Michaelis complex model, the tetrasaccharide moiety is slightly shifted toward that part of the cleft binding the non-reducing end of the substrate, but shows previously unanticipated strong stacking interactions with Phe92 in subsite -I. A number of specific hydrogen-bond contacts between the enzyme and the equatorial O(2), O(3) and O(6) hydroxyl groups of the glucosyl residues in subsites -I, -II and -III are the structural basis for the observed substrate specificity of 1,3-1,4-beta-glucanases. Kinetic analysis of enzyme variants with the all beta-1,3 linked polysaccharide laminarin identified key residues mediating substrate specificity in good agreement with the structural data. The comparison with structures of the apo-enzyme H(A16-M) and a covalent enzyme-inhibitor (E.I) complex, together with kinetic and mutagenesis data, yields new insights into the structural requirements for substrate binding and catalysis. A detailed view of enzyme-carbohydrate interactions is presented and mechanistic implications are discussed.     

Kinetics and subsite mapping of a D-xylobiose- and D-xylose-producing Aspergillus niger endo-(1----4)-beta-D-xylanase

A previously described endo-(1----4)-beta-D-xylanase produced by Aspergillus niger was allowed to react with linear unlabeled and labeled D-xylo-oligosaccharides ranging from D-xylotriose to D-xylo-octaose. No evidence of multiple attack or of condensation and trans-D-xylosylation reactions was found. Maximum rates and Michaelis constants were measured at 40 degrees and pH 4.85. The former increased with increasing chain-length from D-xylotriose through D-xylohexaose to approximately 70% of that on soluble larchwood D-xylan, and then decreased slightly for D-xyloheptaose and D-xylo-octaose. Michaelis constants decreased monotonically with increasing chain-length. Bond-cleavage frequencies were highest near the reducing end of short substrates, with the locus of highest frequencies moving towards the middle of larger substrates. These data indicated that the endo-D-xylanase has five main subsites, with the catalytic site located between the third and fourth subsites, counting from the nonreducing end of the bound substrate. The subsite to the nonreducing side of the catalytic site strongly repels its corresponding D-xylosyl residue, while the two subsites farther towards the nonreducing end of the substrate strongly attract their corresponding residues. The subsite to the reducing side of the catalytic site moderately attracts D-xylosyl residues, while the next one towards the reducing end has a high affinity for them. The residual error of the numerical estimation was allocated largely to the Michaelis constants of the different D-xylo-oligosaccharides, whose calculated values were appreciably smaller than measured values, especially for shorter substrates. This suggests that the subsite model cannot fully account for the experimental data. Estimated and measured values of maximum rates, bond-cleavage frequencies, and dissociation constant when the active site is fully occupied by substrate agreed more closely with each other.     

Determination of steady-state kinetic parameters for a xylanase-catalyzed hydrolysis of neutral underivatized xylooligosaccharides by mass spectrometry

A direct mass spectrometric approach was used for the determination of steady-state kinetic parameters, the turnover number (k(cat)), the Michaelis constant (K(M)), and the specificity constant (k(cat)/K(M)) for an enzyme-catalyzed hydrolysis of xylooligosaccharides. Electrospray ionization mass spectrometry was performed to observe product distributions and to determine k(cat), K(M), and k(cat)/K(M) values for Trichoderma reesei endo-1,4-beta-xylanase II (TRX II) with xylohexaose (Xyl(6)), xylopentaose (Xyl(5)), xylotetraose (Xyl(4)), and xylotriose (Xyl(3)) as substrates. The determined k(cat)/K(M) values (0.93, 0.37, 0.027, and 0.00015 microM(-1) s(-1), respectively) indicated that Xyl(6) was the most preferred substrate of TRX II. In addition, the obtained K(M) value for Xyl(5) (136 microM) was roughly twice as high as that for Xyl(6) (73 microM), suggesting that at least six putative subsites contribute to the substrate binding in the active site of TRX II. Previous mass spectrometric assays for enzyme kinetics have been used mostly in the case of reactions that result in a transfer of acidic groups (e.g., phosphate) into neutral oligosaccharides giving rise to negatively charged products. Here we demonstrate that such analysis is also feasible in the case of neutral underivatized oligosaccharides. Implications of the results for the catalytic mechanism of TRX II in particular are discussed.     

Crystal structure of truncated Fibrobacter succinogenes 1,3-1,4-beta-D-glucanase in complex with beta-1,3-1,4-cellotriose

Fibrobacter succinogenes 1,3-1,4-beta-D-glucanase (Fsbeta-glucanase) catalyzes the specific hydrolysis of beta-1,4 glycosidic bonds adjacent to beta-1,3 linkages in beta-D-glucans or lichenan. This is the first report to elucidate the crystal structure of a truncated Fsbeta-glucanase (TFsbeta-glucanase) in complex with beta-1,3-1,4-cellotriose, a major product of the enzyme reaction. The crystal structures, at a resolution of 2.3 angstroms, reveal that the overall fold of TFsbeta-glucanase remains virtually unchanged upon sugar binding. The enzyme accommodates five glucose residues, forming a concave active cleft. The beta-1,3-1,4-cellotriose with subsites -3 to -1 bound to the active cleft of TFsbeta-glucanase with its reducing end subsite -1 close to the key catalytic residues Glu56 and Glu60. All three subsites of the beta-1,3-1,4-cellotriose adopted a relaxed C(1)4 conformation, with a beta-1,3 glycosidic linkage between subsites -2 and -1, and a beta-1,4 glycosidic linkage between subsites -3 and -2. On the basis of the enzyme-product complex structure observed in this study, a catalytic mechanism and substrate binding conformation of the active site of TFsbeta-glucanase is proposed.     

Barley malt-alpha-amylase. Purification, action pattern, and subsite mapping of isozyme 1 and two members of the isozyme 2 subfamily using p-nitrophenylated maltooligosaccharide substrates.

Isoforms AMY1, AMY2-1 and AMY2-2 of barley alpha-amylase were purified from malt. AMY2-1 and AMY2-2 are both susceptible to barley alpha-amylase/subtilisin inhibitor. The action of these isoforms is compared using substrates ranging from p-nitrophenylmaltoside through p-nitrophenylmaltoheptaoside. The kcat/Km values are calculated from the substrate consumption. The relative cleavage frequency of different substrate bonds is given by the product distribution. AMY2-1 is 3-8-fold more active than AMY1 toward p-nitrophenylmaltotrioside through p-nitrophenylmaltopentaoside. AMY2-2 is 10-50% more active than AMY2-1. The individual subsite affinities are obtained from these data. The resulting subsite maps of the isoforms are quite similar. They comprise four and six glucosyl-binding subsites towards the reducing and the non-reducing end, respectively. Towards the non-reducing end, the sixth and second subsites have a high affinity, the third has very low or even lack of affinity and the first (catalytic subsite) has a large negative affinity. The affinity declines from moderate to low for subsites 1 through 4 toward the reducing end. AMY1 has clearly a more negative affinity at the catalytic subsite, but larger affinities at both the fourth subsites, compared to AMY2. AMY2-1 has lower affinity than AMY2-2 at subsites adjacent to the catalytic site, and otherwise mostly higher affinities than AMY2-2. Theoretical kcat/Km values show excellent agreement with experimental values.     

Characterization of exo-(1,4)-alpha glucan lyase from red alga Gracilaria chorda. Activation, inactivation and the kinetic properties of the enzyme

Exo-(1,4)-alpha glucan lyase (GLase) was purified from a red alga Gracilaria chorda. The enzyme was activated 1.3-fold in the presence of Ca(2+) and Cl(-) ions. The ions also stabilized the enzyme increasing the temperature of its maximum activity from 45 degrees C to 50 degrees C. GLase was inactivated by chemical modification with carbodiimide and a carboxyl group of the enzyme was shown essential to the lyase activity. A tryptophanyl residue(s) was also shown to be important for the activity and was probably involved in substrate binding. K(m) values of the enzyme were 2.3 mM for maltose, 0.4 mM for maltotriose and 0.1 mM for maltooligosaccharides of degree of polymerization (dp) 4-7, and the k(0) values for the oligosaccharides were similar (42-53 s(-1)). The analysis of these kinetic parameters showed that the enzyme has four subsites to accommodate oligosaccharides. The subsite map of GLase was unique, since subsite 1 and subsite 2 have large positive and small negative affinities, respectively. The subsite map of this type has not been found in other enzymes with exo-action on alpha-1,4-glucan. The K(m) and k(0) values for the polysaccharides were lower (0.03 mM) and higher (60-100 s(-1)), respectively, suggesting the presence of another affinity site specific to the polysaccharides.     

X-ray structure of acarbose bound to amylomaltase from Thermus aquaticus. Implications for the synthesis of large cyclic glucans.

As a member of the alpha-amylase superfamily of enzymes, amylomaltase catalyzes either the transglycosylation from one alpha-1,4 glucan to another or an intramolecular cyclization. The latter reaction is typical for cyclodextrin glucanotransferases. In contrast to these enzymes, amylomaltase catalyzes the formation of cyclic glucans with a degree of polymerization larger than 22. To characterize the factors that determine the size of the synthesized cycloamyloses, we have analyzed the X-ray structure of amylomaltase from Thermus aquaticus in complex with the inhibitor acarbose, a maltotetraose derivative, at 1.9 A resolution. Two acarbose molecules are bound to the enzyme, one in the active site groove at subsite -3 to +1 and a second one approximately 14 A away from the nonreducing end of the acarbose bound to the catalytic site. The inhibitor bound to the catalytic site occupies subsites -3 to +1. Unlike the situation in other enzymes of the alpha-amylase family, the inhibitor is not processed and the inhibitory cyclitol ring of acarbose, which mimicks the half chair conformation of the transition state, does not bind to catalytic subsite -1. The minimum ring size of cycloamyloses produced by this enzyme is proposed to be determined by the distance of the specific substrate binding sites at the active site and near Tyr54 and by the size of the 460s loop. The 250s loop might be involved in binding of the substrate at the reducing end of the scissile bond.     

Porcine pancreatic alpha-amylase hydrolysis of hydroxyethylated amylose and specificity of subsite binding

Hydrolysis of partially hydroxyethylated amylose by porcine pancreatic alpha-amylase gives rise to a number of hydroxyethylated di-, tri-, and tetrasaccharides, as well as larger products. No modified monosaccharides were detected. The structures of the products containing two to four D-glucose residues have been analyzed by chromatographic and enzymatic techniques. In no instance were these oligosaccharides modified in the reducing-end residue. The location of hydroxyethylated glucose residues within the oligosaccharides has been interpreted in terms of the ability of that (hydroxyethyl)glucose to bind productively at each of the five subsites of the enzyme active site. Results indicate that subsite 3, the subsite at which catalytic attack occurs, is especially sensitive to changes in the substrate and that unmodified glucose is required for productive binding at this subsite. Other subsites specifically allow binding of some (hydroxyethyl)glucose isomers, but not others. Hydroxyethylation is permitted at C-2, C-3, and C-6 for residues bound at subsite 1 and is permitted at C-6 and possibly at C-2 and C-3 for residues bound at subsite 5. However, substitution is permitted only at C-3 and C-6 for binding at subsite 2 and at C-2 and C-3 for binding at subsite 4.     

Endo-1,4-beta-xylanase B from Aspergillus cf. niger BCC14405 isolated in Thailand: purification, characterization and gene isolation

During the screening of xylanolytic enzymes from locally isolated fungi, one strain BCC14405, exhibited high enzyme activity with thermostability. This fugal strain was identified as Aspergillus cf. niger based on its morphological characteristics and internal transcribed spacer (ITS) sequences. An enzyme with xylanolytic activity from BCC14405 was later purified and characterized. It was found to have a molecular mass of ca. 21 kDa, an optimal pH of 5.0, and an optimal temperature of 55 degrees C. When tested using xylan from birchwood, it showed K(m) and V(max) values of 8.9 mg/ml and 11,100 U/mg, respectively. The enzyme was inhibited by CuSO(4) EDTA, and by FeSO(4) The homology of the 20-residue N-terminal protein sequence showed that the enzyme was an endo-1,4-beta-xylanase. The full-length gene encoding endo-1,4-beta-xylanase from BCC14405 was obtained by PCR amplification of its cDNA. The gene contained an open reading frame of 678 bp, encoding a 225 amino acid protein, which was identical to the endo-1,4-a-xylanase B previously identified in A. niger.     

Protein-carbohydrate interactions defining substrate specificity in Bacillus 1,3-1,4-beta-D-glucan 4-glucanohydrolases as dissected by mutational analysis

The carbohydrate-binding site of Bacillus macerans 1,3-1, 4-beta-D-glucan 4-glucanohydrolase has been analyzed through a mutational analysis to probe the role of protein-carbohydrate interactions defining substrate specificity. Amino acid residues involved in substrate binding were proposed on the basis of a modeled enzyme-substrate complex [Hahn, M., Keitel, T., and Heinemann, U. (1995) Eur. J. Biochem. 232, 849-859]. The effects of the mutations at 15 selected residues on catalysis and binding were determined by steady-state kinetics using a series of chromogenic substrates of different degree of polymerization to assign the individual H-bond and hydrophobic contributions to individual subsites in the binding site cleft. The glucopyranose rings at subsites -III and -II are tightly bound by a number of H-bond interactions to Glu61, Asn24, Tyr92, and Asn180. From k(cat)/K(M) values, single H-bonds account for 1.8-2.2 kcal mol(-)(1) transition-state (TS) stabilization, and a charged H-bond contributes up to 3.5 kcal mol(-)(1). Glu61 forms a bidentated H-bond in subsites -III and -II, and provides up to 6.5 kcal mol(-)(1) TS stabilization. With a disaccharide substrate that fills subsites -I and -II, activation kinetics were observed for the wild-type and mutant enzymes except for mutations on Glu61, pointing to an important role of the bidentate interaction of Glu61 in two subsites. Whereas removal of the hydroxyl group of Tyr121, initially proposed to hydrogen-bond with the 2OH of Glcp-I, has essentially no effect (Y121F mutant), side-chain removal (Y121A mutant) gave a 100-fold reduction in k(cat)/K(M) and a 10-fold lower K(I) value with a competitive inhibitor. In subsite -IV, only a stacking interaction with Tyr22 (0.7 kcal mol(-)(1) TS stabilization) is observed.     

Studies of the mechanism of the cyclisation reaction catalysed by the wildtype and a truncated alpha-cyclodextrin glycosyltransferase from Klebsiella pneumoniae strain M 5 al, and the beta-cyclodextrin glycosyltransferase from Bacillus circulans strain 8

The actions of the wildtype and a truncated alpha-cyclodextrin glycosyltransferase from Klebsiella pneumoniae strain M 5 al on malto-oligosaccharides showed no significant differences, and there was marked dependence of the kinetic parameters on the chain lengths of the substrate. The action of the beta-cyclodextrin glycosyltransferase from Bacillus circulans was less dependent on the chain length of the substrate, but Vmax of the initial cyclisation with the longer malto-oligosaccharides was only 28% of that determined for the enzyme of K. pneumoniae. The rate parameters suggested that the active site of each enzyme spans nine glucosyl residues, and that the catalytic sites are situated between subsites three and four for the K. pneumoniae enzymes and between subsites two and three for the B. circulans enzyme. The molecular binding affinities and the affinities of the 9th subsite were calculated from the rate parameters. The primary and tertiary structures of alpha-amylases and cyclodextrin glycosyltransferases are compared in the context of the reaction mechanism of the latter enzymes.     

Binding of glycogen, oligosaccharides, and glucose to glycogen debranching enzyme

The binding of glucose and a series of oligosaccharides to glycogen debranching enzyme was determined by the ability of the saccharides to decrease the rate of reaction of sulfhydryl groups with 5,5'-dithiobis(2-nitrobenzoate) (DTNB). At pH 7.2, the strength of binding increases with chain length from glucose to maltotriose to maltopentaose but not to maltohexaose, and the free energies for binding of the oligosaccharides suggest subsites of equivalent affinities for the four glucose units following the initial reducing moiety. The rate of reaction of DTNB with enzyme saturated with saccharide is the same for all compounds, suggesting that all the saccharides, including glucose, induce the same conformational state. The site of binding may be that which binds the alpha-1,6-linked side chain of the natural limit dextrin substrate. At pH 8.0, this site exhibits similar characteristics, but an additional site, which may bind the four terminal glucose units of the main chain of the natural substrate, is manifested and exhibits different characteristics, including a very low affinity for glucose itself. The binding of glycogen to the debranching enzyme was monitored by centrifugal separation from the protein and exhibits a much lower dissociation constant than that for the oligomers, suggesting that branched polymers have more than one set of subsites.     

Subsite Structure and Action Mode of the α-Amylase from Thermoactinomvces vulgaris

The subsite structure of Thermoactinomyces vulgaris α-amylase was estimated from its action mode and rate parameters of hydrolysis on maltooligosaccharides. These results led to the conclusion that this α-amylase has six subsites with the catalytic site located between the third and fourth subsites from the non-reducing end side. Subsite affinities were calculated to be 0.38, 5.46, 2.72 and 0.23 kcal/mol for subsites 1, 2, 5 and 6, respectively, and the sum of the affinities of subsite 3 and 4 to be ?3.41 kcal/mol. The unique action mode of this α-amylase on various substrates was interpreted in terms of the subsite structure.     

The effect of substrate modification on binding of porcine pancreatic alpha amylase: hydrolysis of modified amylose containing D-allose residues

A modified amylose containing 10% of tritiated D-allose residues has been hydrolyzed by porcine pancreatic alpha amylase (PPA). This reaction produced a number of radioactive oligosaccharides of low molecular weight, including modified mono-, di-, and tri-saccharides, as well as larger products. Analysis of these products by chemical and enzymic methods identified D-allose, two isomers of modified maltose, and isomers of modified maltotriose. These results may be interpreted in terms of current PPA models to indicate that D-allose residues may be productively bound at all five subsites of the active site of the enzyme. The distribution of modified residues in these products, however, further suggests that productive binding of D-allose at the subsite where catalytic attack occurs (subsite 3) is less favorable than binding of D-glucose. These results are compared with results of a series of PPA substrates having modifications at C-3 and at other positions. Trends observed in enzyme hydrolysis of these modified substrates reflect factors that contribute to PPA catalysis, with respect to steric, electronic, and hydrogen-bonding interactions between enzyme and substrate.     

Amylose chain behavior in an interacting context. III. Complete occupancy of the AMY2 barley alpha-amylase cleft and comparison with biochemical data.

In the first two papers of this series, the tools necessary to evaluate substrate ring deformations were developed, and then the modeling of short amylose fragments (maltotriose and maltopentaose) inside the catalytic site of barley alpha-amylase was performed. In this third paper, this docking has been extended to the whole catalytic cleft. A systematic approach to extend the substrate was used on the reducing side from the previous enzyme/pentasaccharide complex. However, due to the lack of an obvious subsite at the nonreducing side, an alternate protocol has been chosen that incorporates biochemical information on the enzyme and features on the substrate shape as well. As a net result, ten subsites have been located consistent with the distribution of Ajandouz et al. (E. H. Ajandouz, J. Abe, B. Svensson, and G. Marchis-Mouren, Biochimica Biophysica Acta, 1992, Vol. 1159, pp. 193-202) and corresponding binding energies were estimated. Among them, two extreme subsites (-6) and (+4), with stacking residues Y104 and Y211, respectively, have strong affinities with glucose rings added to the substrate. No other deformation has been found for the new glucose rings added to the substrate; therefore, only ring A of the DP 10 fragment has a flexible form when interacting with the inner stacking residues Y51. Global conservation of the helical shape of the substrate can be postulated in spite of its significant distortion at subsite (-1).     

On the catalytic and binding sites of porcine enteropeptidase.

The active site of porcine enteropeptidase (EC 3.4.21.9) was investigated in order to characterize better both catalytic and binding sites. The participation of a serine and a histidine residue in the catalytic process was fully confirmed and the two residues were located on the light chain of the enzyme. The binding site was found to be composed of at least 2 subsites S1 and S2. The subsite S1 (similar to the trypsin-binding site) is responsible for the interactions with the small substrates of trypsin and the lysine side chain of trypsinogen, while subsite S2 (probably a cluster of lysines) is responsible for the interactions with the polyanionic sequence found in all trypsinogens. Binding of substrate by subsite S2 led to an increased efficiency of the catalytic site which can be correlated to the known high specificity of enteropeptidase.     

Subsite mapping of Aspergillus niger endopolygalacturonase II by site-directed mutagenesis

To assess the subsites involved in substrate binding in Aspergillus niger endopolygalacturonase II, residues located in the potential substrate binding cleft stretching along the enzyme from the N to the C terminus were subjected to site-directed mutagenesis. Mutant enzymes were characterized with respect to their kinetic parameters using polygalacturonate as a substrate and with respect to their mode of action using oligogalacturonates of defined length (n = 3-6). In addition, the effect of the mutations on the hydrolysis of pectins with various degrees of esterification was studied. Based on the results obtained with enzymes N186E and D282K it was established that the substrate binds with the nonreducing end toward the N terminus of the enzyme. Asn(186) is located at subsite -4, and Asp(282) is located at subsite +2. The mutations D183N and M150Q, both located at subsite -2, affected catalysis, probably mediated via the sugar residue bound at subsite -1. Tyr(291), located at subsite +1 and strictly conserved among endopolygalacturonases appeared indispensable for effective catalysis. The mutations E252A and Q288E, both located at subsite +2, showed only slight effects on catalysis and mode of action. Tyr(326) is probably located at the imaginary subsite +3. The mutation Y326L affected the stability of the enzyme. For mutant E252A, an increased affinity for partially methylesterified substrates was recorded. Enzyme N186E displayed the opposite behavior; the specificity for completely demethylesterified regions of substrate, already high for the native enzyme, was increased. The origin of the effects of the mutations is discussed.     

Crystallographic analysis shows substrate binding at the -3 to +1 active-site subsites and at the surface of glycoside hydrolase family 11 endo-1,4-beta-xylanases

GH 11 (glycoside hydrolase family 11) xylanases are predominant enzymes in the hydrolysis of heteroxylan, an abundant structural polysaccharide in the plant cell wall. To gain more insight into the protein-ligand interactions of the glycone as well as the aglycone subsites of these enzymes, catalytically incompetent mutants of the Bacillus subtilis and Aspergillus niger xylanases were crystallized, soaked with xylo-oligosaccharides and subjected to X-ray analysis. For both xylanases, there was clear density for xylose residues in the -1 and -2 subsites. In addition, for the B. subtilis xylanase, there was also density for xylose residues in the -3 and +1 subsite showing the spanning of the -1/+1 subsites. These results, together with the observation that some residues in the aglycone subsites clearly adopt a different conformation upon substrate binding, allowed us to identify the residues important for substrate binding in the aglycone subsites. In addition to substrate binding in the active site of the enzymes, the existence of an unproductive second ligand-binding site located on the surface of both the B. subtilis and A. niger xylanases was observed. This extra binding site may have a function similar to the separate carbohydrate-binding modules of other glycoside hydrolase families.     

Structure of a pancreatic alpha-amylase bound to a substrate analogue at 2.03 A resolution.

The structure of pig pancreatic alpha-amylase in complex with carbohydrate inhibitor and proteinaceous inhibitors is known but the successive events occurring at the catalytic center still remain to be elucidated. The X-ray structure analysis of a crystal of pig pancreatic alpha-amylase (PPA, EC 3.2.1.1.) soaked with an enzyme-resistant substrate analogue, methyl 4,4'-dithio-alpha-maltotrioside, showed electron density corresponding to the binding of substrate analogue molecules at the active site and at the "second binding site." The electron density observed at the active site was interpreted in terms of overlapping networks of oligosaccharides, which show binding of substrate analogue molecules at subsites prior to and subsequent to the cleavage site. A weaker patch of density observed at subsite -1 (using a nomenclature where the site of hydrolysis is taken to be between subsites -1 and +1) was modeled with water molecules. Conformational changes take place upon substrate analogue binding and the "flexible loop" that constitutes the surface edge of the active site is observed in a specific conformation. This confirms that this loop plays an important role in the recognition and binding of the ligand. The crystal structure was refined at 2.03 A resolution, to an R-factor of 16.0 (Rfree, 18.5).