Search

A novel inhibitor against maltose-producing alpha-amylase was prepared via stepwise degradation of a high-molecular-weight acarbose (HMWA) using Thermus maltogenic amylase (ThMA). The structure of the purified inhibitor was determined to be alpha-D-glucopyranosyl-alpha-acarviosinyl-D-glucopyranose (GlcAcvGlc) by (13)C NMR and MALDI-TOF/MS. Progress curves of PNPG2 hydrolysis by various amylolytic enzymes, including MGase, ThMA, and CDase I-5, in the presence of acarbose or GlcAcvGlc indicated a slow-binding mode of inhibition. Analytical ultracentrifugation and X-ray crystallography analyses revealed that the presence of GlcAcvGlc increased the dimerization of ThMA. The formation of dimer complexed with GlcAcvGlc might induce a conformational change in ThMA, leading to a two-step inhibition process. The inhibition potency of GlcAcvGlc for MGase, ThMA, and CDase I-5 was 3 orders of magnitude higher than that of acarbose.

The thermostable phytase from Bacillus amyloliquefaciens DS11 hydrolyzes phytate (myo-inositol hexakisphosphate, IP6) to less phosphorylated myo-inositol phosphates in the presence of Ca2+. In this report, we discuss the unique Ca2+-dependent catalytic properties of the phytase and its specific substrate requirement. Initial rate kinetic studies of the phytase indicate that the enzyme activity follows a rapid equilibrium ordered mechanism in which binding of Ca2+ to the active site is necessary for the essential activation of the enzyme. Ca2+ turned out to be also required for the substrate because the phytase is only able to hydrolyze the calcium-phytate complex. In fact, both an excess amount of free Ca2+ and an excess of free phytate, which is not complexed with each other, can act as competitive inhibitors. The Ca2+-dependent catalytic activity of the enzyme was further confirmed, and the critical amino acid residues for the binding of Ca2+ and substrate were identified by site-specific mutagenesis studies. Isothermal titration calorimetry (ITC) was used to understand if the decreased enzymatic activity was related to poor Ca2+ binding. The pH dependence of the Vmax and Vmax/Km consistently supported these observations by demonstrating that the enzyme activity is dependent on the ionization of amino acid residues that are important for the binding of Ca2+ and the substrate. The Ca2+-dependent activation of enzyme and substrate was found to be different from other histidine acid phytases that hydrolyze metal-free phytate.

Δ5-3-ketosteroid isomerase (KSI) from Pseudomonas putida Biotype B catalyzes the allylic isomerization of Δ5-3-ketosteroids to their conjugated Δ4-isomers via a dienolate intermediate. Two electrophilic catalysts, Tyr-14 and Asp-99, are involved in a hydrogen bond network that comprises Asp-99 Oδ2···O of Wat504···Tyr-14 Oη···Tyr-55 Oη···Tyr-30 Oη in the active site of P. putida KSI. Even though neither Tyr-30 nor Tyr-55 plays an essential role in catalysis by the KSI, the catalytic activity of Y14F could be increased ca. 26−51-fold by the additional Y30F and/or Y55F mutation in the hydrogen bond network. To identify the structural basis for the pseudoreversion in the KSI, crystal structures of Y14F and Y14F/Y30F/Y55F have been determined at 1.8 and 2.0 A resolution, respectively. Comparisons of the two structures near the catalytic center indicate that the hydrogen bond between Asp-99 Oδ2 and C3−O of the steroid, which is perturbed by the Y14F mutation, can be partially restored to that in the wild-type e...

Delta(5)-3-Ketosteroid isomerase from Pseudomonas putida biotype B is one of the most proficient enzymes catalyzing an allylic isomerization reaction at rates comparable to the diffusion limit. The hydrogen-bond network (Asp99... Wat504...Tyr14...Tyr55...Tyr30) which links the two catalytic residues, Tyr14 and Asp99, to Tyr30, Tyr55, and a water molecule in the highly apolar active site has been characterized in an effort to identify its roles in function and stability. The DeltaG(U)(H2O) determined from equilibrium unfolding experiments reveals that the elimination of the hydroxyl group of Tyr14 or Tyr55 or the replacement of Asp99 with leucine results in a loss of conformational stability of 3.5-4.4 kcal/mol, suggesting that the hydrogen bonds of Tyr14, Tyr55, and Asp99 contribute significantly to stability. While decreasing the stability by about 6.5-7.9 kcal/mol, the Y55F/D99L or Y30F/D99L double mutation also reduced activity significantly, exhibiting a synergistic effect on k(cat) relative to the respective single mutations. These results indicate that the hydrogen-bond network is important for both stability and function. Additionally, they suggest that Tyr14 cannot function efficiently alone without additional support from the hydrogen bonds of Tyr55 and Asp99. The crystal structure of Y55F as determined at 1.9 A resolution shows that Tyr14 OH undergoes an alteration in orientation to form a new hydrogen bond with Tyr30. This observation supports the role of Tyr55 OH in positioning Tyr14 properly to optimize the hydrogen bond between Tyr14 and C3-O of the steroid substrate. No significant structural changes were observed in the crystal structures of Y30F and Y30F/Y55F, which allowed us to estimate approximately the interaction energies mediated by the hydrogen bonds Tyr30...Tyr55 and Tyr14...Tyr55. Taken together, our results demonstrate that the hydrogen-bond network provides the structural support that is needed for the enzyme to maintain the active-site geometry optimized for both function and stability.

Bacterial Delta 5-3-ketosteroid isomerase (KSI) from Pseudomonas testosteroni has been intensively studied as a prototype for understanding an enzyme-catalyzed allylic rearrangement involving intramolecular proton transfer. Asp38 serves as a general base to abstract the proton from the steroid C4-H, which is a much stronger base than the carboxyl group of this residue. This unfavorable proton transfer requires 11 kcal/mol of energy which has to be provided by favorable interactions between catalytic residues and substrate in the course of the catalytic reaction. How this energy is provided at the active site of KSI has been a controversial issue, and inevitably the enzyme mechanism is not settled. To resolve these issues, we have determined the crystal structure of this enzyme at 2.3 A resolution. The crystal structure revealed that the active site environment of P. testosteroni KSI is nearly identical to that of Pseudomonas putida KSI, whose structure in complex with a reaction intermediate analogue we have determined recently. Comparison of the two structures clearly indicates that the two KSIs should share the same enzyme mechanism involving the stabilization of the dienolate intermediate by the two direct hydrogen bonds to the dienolate oxyanion, one from Tyr14 OH and the other from Asp99 COOH. Mutational analysis of the two residues and other biochemical data strongly suggest that the hydrogen bond of Tyr14 provides the more significant contribution than that of Asp99 to the requisite 11 kcal/mol of energy for the catalytic power of KSI.

Bacterial Δ5-3-ketosteroid isomerase (KSI) catalyzes a stereospecific isomerization of steroid substrates at an extremely fast rate, overcoming a large disparity of pKa values between a catalytic residue and its target. The crystal structures of KSI from Pseudomonas putida and of the enzyme in complex with equilenin, an analogue of the reaction intermediate, have been determined at 1.9 and 2.5 A resolution, respectively. The structures reveal that the side chains of Tyr14 and Asp99 (a newly identified catalytic residue) form hydrogen bonds directly with the oxyanion of the bound inhibitor in a completely apolar milieu of the active site. No water molecule is found at the active site, and the access of bulk solvent is blocked by a layer of apolar residues. Asp99 is surrounded by six apolar residues, and consequently, its pKa appears to be elevated as high as 9.5 to be consistent with early studies. No interaction was found between the bound inhibitor and the residue 101 (phenylalanine in Pseudomonas testos...

The benefits of performing ENDOR experiments at higher microwave frequency are demonstrated in a Q-band (35 GHz) ENDOR investigation of a number of proteins with [nFe-mS] clusters, n = 2, 3, 4. Each protein displays several resonances in the frequency range of 0-20 MHz. In all instances, features are seen near v approximately 13 and 8 MHz that can be assigned, respectively, to "distant ENDOR" from 13C in natural-abundance (1.1%) and from 14N (the delta m1 = +/- 2 transitions); the nuclei involved in this phenomenon are remote from and have negligible hyperfine couplings to the cluster. In addition, a number of proteins show local 13C ENDOR signals with resolved hyperfine interactions; these are assigned to the beta carbons of cysteines bound to the cluster [A(13C) approximately 1.0 MHz]. Five proteins show resolved, local delta m1 = +/- 2 ENDOR signals from 14N with an isotropic hyperfine coupling, 0.4 less than or equal to A(14N) less than or equal to 1.0, similar to those seen in ESEEM studies; these most likely are associated with N-H...S hydrogen bonds to the cluster. Anabaena ferredoxin further shows a signal corresponding to A(14N) approximately 4 MHz. Quadrupole coupling constants are derived for both local and distant 14N signals. The interpretation of the data is supported by studies on 15N- and 13C-enriched ferredoxin (Fd) from Anabaena 7120, where the 15N signals can be clearly correlated with the corresponding 14N signals and where the 13C signals are strongly enhanced. Thus, the observation of 14N delta m1 = +/- 2 signals at Q-band provides a new technique for examining weak interactions with a cluster.(ABSTRACT TRUNCATED AT 250 WORDS)

Hyperfine 1H NMR signals of the 2Fe-2S* vegetative ferredoxin from Anabaena 7120 have been studied by two-dimensional (2D) magnetization exchange spectroscopy. The rapid longitudinal relaxation rates of these signals required the use of very short nuclear Overhauser effect (NOE) mixing times (0.5-20 ms). The resulting pattern of NOE cross-relaxation peaks when combined with previous 1D NOE results [Dugad, L. B., La Mar, G. N., Banci, L., & Bertini, I. (1990) Biochemistry 29, 2263-2271] led to elucidation of the carbon-bound proton spin systems from each of the four cysteines ligated to the 2Fe-2S* cluster in the reduced ferredoxin. Additional NOE cross peaks were observed that provide information about other amino acid residues that interact with the iron-sulfur cluster. NOE cross peaks were assigned tentatively to Leu27, Arg42, and Ala43 on the basis of the X-ray coordinates of oxidized Anabaena 7120 ferredoxin [Rypniewski, W.R., Breiter, D.R., Benning, M.M., Wesenberg, G., Oh, B.-H., Markley, J.L., Rayment, I., & Holden, H. M. (1991) Biochemistry 30, 4126-4131]. Three chemical exchange cross peaks were detected in magnetization exchange spectra of half-reduced ferredoxin and assigned to the 1H alpha protons of Cys49 and Cys79 [both of whose sulfur atoms are ligated to Fe(III)] and Arg42 (whose amide nitrogen is hydrogen-bonded to one of the inorganic sulfurs of the 2Fe-2S* cluster). The chemical exchange cross peaks provide a means of extending assignments in the spectrum of reduced ferredoxin to assignments in the spectrum of the oxidized protein.(ABSTRACT TRUNCATED AT 250 WORDS)

A sequence alignment shows that residue 332 is conserved as glutamate in maltogenic amylases (MAases) and in other related enzymes such as cyclodextrinase and neopullulanase, while the corresponding position is conserved as histidine in alpha-amylases. We analyzed the role of Glu332 in the hydrolysis and the transglycosylation activity of Thermus MAase (ThMA) by site-directed mutagenesis. Replacing Glu332 with histidine reduced transglycosylation activity significantly, but enhanced hydrolysis activity on alpha-(1,3)-, alpha-(1,4)-, and alpha-(1,6)-glycosidic bonds relative to the wild-type (WT) enzyme. The mutant Glu332Asp had catalytic properties similar to those of the WT enzyme, but the mutant Glu332Gln resulted in significantly decreased transglycosylation activity. These results suggest that an acidic side chain at position 332 of MAase plays an important role in the formation and accumulation of transfer products by modulating the relative rates of hydrolysis and transglycosylation. From the structure, we propose that an acidic side chain at position 332, which is located in a pocket, is involved in aligning the acceptor molecule to compete with water molecules in the nucleophilic attack of the glycosyl-enzyme intermediate.

The aromatic residues Phe-54, Phe-82, and Trp-116 in the hydrophobic substrate-binding pocket of Delta(5)-3-ketosteroid isomerase from Pseudomonas putida biotype B have been characterized in their roles in steroid binding and catalysis. Kinetic and equilibrium binding analyses were carried out for the mutant enzymes with the substitutions Phe-54 --Ala or Leu, Phe-82 --Ala or Leu, and Trp-116 --Ala, Phe, or Tyr. The removal of their bulky, aromatic side chains at any of these three positions results in reduced k(cat), particularly when Phe-82 or Trp-116 is replaced by Ala. The results are consistent with the binding interactions of the aromatic residues with the bound steroid contributing to catalysis. All the mutations except the F82A mutation increase K(m); the F82A mutation decreases K(m) by ca. 3-fold, suggesting a possibility that the phenyl ring at position 82 might be unfavorable for substrate binding. The K(D) values for d-equilenin, an intermediate analogue, suggest that a space-filling hydrophobic side chain at position 54, a phenyl ring at position 82, and a nonpolar aromatic or small side chain at position 116 might be favorable for binding the reaction intermediate. In contrast to the increased K(D) for equilenin, the enzymes with any substitutions at positions 54 and 116 display a decreased K(D) for 19-nortestosterone, a product analogue, indicating that Phe-54 and Trp-116 might be unfavorable for product binding. The crystal structure of F82A determined to 2.1-A resolution reveals that Phe-82 is important for maintaining the active site geometry. Taken together, our results demonstrate that Phe-54, Phe-82, and Trp-116 contribute differentially to the stabilization of steroid species including substrate, intermediate, and product.

Multinuclear two-dimensional NMR techniques were used to assign nearly all diamagnetic 13C and 15N resonances of the plant-type 2Fe.2S* ferredoxin from Anabaena sp. strain PCC 7120. Since a 13C spin system directed strategy had been used to identify the 1H spin systems [Oh, B.-H., Westler, W. M., & Markley, J. L. (1989) J. Am. Chem. Soc. 111, 3083-3085], the sequence-specific 1H assignments [Oh, B.-H., & Markley, J. L. (1990) Biochemistry (first paper of three in this issue)] also provided sequence-specific 13C assignments. Several resonances from 1H-13C groups were assigned independently of the 1H assignments by considering the distances between these nuclei and the paramagnetic 2Fe.2S* center. A 13C-15N correlation data set was used to assign additional carbonyl carbons and to analyze overlapping regions of the 13C-13C correlation spectrum. Sequence-specific assignments of backbone and side-chain nitrogens were based on 1H-15N and 13C-15N correlations obtained from various two-dimensional NMR experiments.