小島 修一

L-Amino acid ligase (Lal) catalyzes the formation of a dipeptide from two L-amino acids in an ATP-dependent manner and belongs to the ATP-grasp superfamily. Bacillus subtilis YwfE, the first identified Lal, produces the dipeptide antibiotic bacilysin, which consists of L-Ala and L-anticapsin. Its substrate specificity is restricted to smaller amino acids such as L-Ala for the N-terminal end of the dipeptide, whereas a wide range of hydrophobic amino acids including L-Phe and L-Met are recognized for the C-terminal end in vitro. We determined the crystal structures of YwfE with bound ADP-Mg2+-Pi and ADP-Mg2+-L-Ala at 1.9 and 2.0 angstrom resolutions, respectively. On the basis of these structures, we generated point mutants of residues that are considered to participate in the recognition of L-Ala and measured their ATPase activity. The conserved Arg328 is suggested to be a crucial residue for L-Ala recognition and catalysis. The mutation of Trp332 to Ala caused the enzyme to hydrolyze ATP, even in the absence of L-Ala, and the structure of this mutant protein appeared to show a cavity in the N-terminal substrate-binding pocket. These results suggest that Trp332 plays a key role in restricting the substrate specificity to smaller amino acids such as L-Ala. Moreover, Trp332 mutants can alter the substrate specificity and activity depending on the size and shape of substituted amino acids. These observations provide sufficient scope for the rational design of Lal to produce desirable dipeptides. We propose that the positioning of the conserved Arg residue in Lal is important for enantioselective recognition of L-amino acids.

Bacillus subtilis YwfE, an l-amino-acid ligase, catalyzes the formation of an a-dipeptide from l-amino acids in an ATP-dependent manner. In order to elucidate the substrate-recognition mode and the reaction mechanism of this ligase, native and selenomethionine-derivatized (SeMet) crystals of YwfE in the presence of ADP, MgCl2 and the dipeptide l-Ala-l-Gln were obtained using the similar to hanging-drop vapour-diffusion method. These crystals diffracted to 1.9 and 2.8 angstrom resolution, respectively. Preliminary SAD phase calculations using the data set from the SeMet crystal suggested that the crystal belonged to the hexagonal space group P6522, with unit-cell parameters a = b = 90.85, c similar to=similar to 250.31 angstrom, and contained one molecule in the asymmetric unit with a solvent content of 57.3%.

P14C/N39C is the disulfide variant of the ovomucoid third domain from silver pheasant (OMSVP3) introducing an engineered Cys(14)-Cys(39) bond near the reactive site on the basis of the sequence homology between OMSVP3 and ascidian trypsin inhibitor. This variant exhibits a narrower inhibitory specificity. We have examined the effects of introducing a Cys(14)-Cys(39) bond into the flexible N-terminal loop of OMSVP3 on the thermodynamics of the reactive site peptide bond hydrolysis, as well as the thermal stability of reactive site intact inhibitors. P14C/N39C can be selectively cleaved by Streptomyces griseus protease B at the reactive site of OMSVP3 to form a reactive site modified inhibitor. The conversion rate of intact to modified P14C/N39C is much faster than that for wild type under any pH condition. The pH-independent hydrolysis constant (K-hyd degrees) is estimated to be approximately 5.5 for P14C/N39C, which is higher than the value of 1.6 for natural OMSVP3. The reactive site modified form of P14C/N39C is thermodynamically more stable than the intact one. Thermal denaturation experiments using intact inhibitors show that the temperature at the midpoint of unfolding at pH 2.0 is 59 degrees C for P14C/N39C and 58 degrees C for wild type. There have been no examples, except P14C/N39C, where introducing an engineered disulfide causes a significant increase in K-hyd degrees, but has no effect on the thermal stability. The site-specific disulfide introduction into the flexible N-terminal loop of natural Kazal-type inhibitors would be useful to further characterize the thermodynamics of the reactive site peptide bond hydrolysis. Copyright (C) 2011 European Peptide Society and John Wiley & Sons, Ltd.

P14C/N39C is the disulfide variant of the ovomucoid third domain from silver pheasant (OMSVP3) introducing an engineered Cys(14)-Cys(39) bond near the reactive site on the basis of the sequence homology between OMSVP3 and ascidian trypsin inhibitor. This variant exhibits a narrower inhibitory specificity. We have examined the effects of introducing a Cys(14)-Cys(39) bond into the flexible N-terminal loop of OMSVP3 on the thermodynamics of the reactive site peptide bond hydrolysis, as well as the thermal stability of reactive site intact inhibitors. P14C/N39C can be selectively cleaved by Streptomyces griseus protease B at the reactive site of OMSVP3 to form a reactive site modified inhibitor. The conversion rate of intact to modified P14C/N39C is much faster than that for wild type under any pH condition. The pH-independent hydrolysis constant (K-hyd degrees) is estimated to be approximately 5.5 for P14C/N39C, which is higher than the value of 1.6 for natural OMSVP3. The reactive site modified form of P14C/N39C is thermodynamically more stable than the intact one. Thermal denaturation experiments using intact inhibitors show that the temperature at the midpoint of unfolding at pH 2.0 is 59 degrees C for P14C/N39C and 58 degrees C for wild type. There have been no examples, except P14C/N39C, where introducing an engineered disulfide causes a significant increase in K-hyd degrees, but has no effect on the thermal stability. The site-specific disulfide introduction into the flexible N-terminal loop of natural Kazal-type inhibitors would be useful to further characterize the thermodynamics of the reactive site peptide bond hydrolysis. Copyright (C) 2011 European Peptide Society and John Wiley & Sons, Ltd.

The crystal structure of flavoredoxin from Desulfovibrio vulgaris Miyazaki F was determined at 1.05 A resolution and its ferric reductase activity was examined. The aim was to elucidate whether flavoredoxin has structural similarity to ferric reductase and ferric reductase activity, based on the sequence similarity to ferric reductase from Archaeoglobus fulgidus. As expected, flavoredoxin shared a common overall structure with A. fulgidus ferric reductase and displayed weak ferric reductase and flavin reductase activities; however, flavoredoxin contains two FMN molecules per dimer, unlike A. fulgidus ferric reductase, which has only one FMN molecule per dimer. Compared with A. fulgidus ferric reductase, flavoredoxin forms three additional hydrogen bonds and has a significantly smaller solvent-accessible surface area. These observations explain the higher affinity of flavoredoxin for FMN. Unexpectedly, an electron-density map indicated the presence of a Mes molecule on the re-side of the isoalloxazine ring of FMN, and that two zinc ions are bound to the two cysteine residues, Cys39 and Cys40, adjacent to FMN. These two cysteine residues are close to one of the putative ferric ion binding sites of ferric reductase. Based on their structural similarities, we conclude that the corresponding site of ferric reductase is the most plausible site for ferric ion binding. Comparing the structures with related flavin proteins revealed key structural features regarding the discrimination of function (ferric ion or flavin reduction) and a unique electron transport system.

The crystal structure of flavoredoxin from Desulfovibrio vulgaris Miyazaki F was determined at 1.05 A resolution and its ferric reductase activity was examined. The aim was to elucidate whether flavoredoxin has structural similarity to ferric reductase and ferric reductase activity, based on the sequence similarity to ferric reductase from Archaeoglobus fulgidus. As expected, flavoredoxin shared a common overall structure with A. fulgidus ferric reductase and displayed weak ferric reductase and flavin reductase activities; however, flavoredoxin contains two FMN molecules per dimer, unlike A. fulgidus ferric reductase, which has only one FMN molecule per dimer. Compared with A. fulgidus ferric reductase, flavoredoxin forms three additional hydrogen bonds and has a significantly smaller solvent-accessible surface area. These observations explain the higher affinity of flavoredoxin for FMN. Unexpectedly, an electron-density map indicated the presence of a Mes molecule on the re-side of the isoalloxazine ring of FMN, and that two zinc ions are bound to the two cysteine residues, Cys39 and Cys40, adjacent to FMN. These two cysteine residues are close to one of the putative ferric ion binding sites of ferric reductase. Based on their structural similarities, we conclude that the corresponding site of ferric reductase is the most plausible site for ferric ion binding. Comparing the structures with related flavin proteins revealed key structural features regarding the discrimination of function (ferric ion or flavin reduction) and a unique electron transport system.

One feature of the alpha 3-peptide, which has the amino acid sequence of (Leu-Glu-Thr-Leu-Ala-Lys-Ala)(3), that distinguishes it from many other alpha-helix-forming peptides is its ability to form fibrous assemblies that can be observed by transmission electron microscopy. In this study, the effects of Ala -> Gln substitution at the e (5th) or g (7th) position in the above heptad sequence of the alpha 3-peptide on the formation of alpha-helix and fibrous assemblies were investigated by circular dichroism spectral measurement and atomic force microscopy. The 5Q alpha 3-peptide obtained by Ala -> Gln substitution at the e position of the alpha 3-peptide was found to form very short fibrils with long- elliptical shape, whereas the 7Q alpha 3-peptide with Gln residues at the g position lost its ability to form such assemblies, in spite of alpha-helix formation in both peptides; the stabilities of both peptides decreased. These results indicate that Ala residues at the g position in the heptad sequence of the alpha 3-peptide are key residues for the formation of fibrous assemblies, which may be due to hydrophobic interactions between alpha-helical bundle surfaces.

We previously demonstrated that Pleurotus ostreatus proteinase A inhibitor I (POIA1) could function as an intramolecular chaperone of subtilisin BPN', as in the case of the propeptide of subtilisin BPN', and that its Phe44 -> Ala mutant, which lost its tertiary structure, could not assist the refolding of subtilisin BPN'. In this study, we examined the effects of hydrophobic amino acid substitutions at other sites and substitutions of Phe44 with other hydrophobic residues on the structure and functions of POIA1. These mutations were introduced into POIA1cm that had been obtained by the substitution of the C-terminal six residues of POIA1 with those of the propeptide of subtilisin BPN'. When IIe32 or Ile64 was substituted with Ala, the tertiary structure of the resultant mutant was markedly destroyed, and the activities as a protease inhibitor and an intramolecular chaperone were significantly lowered. Among the position 44 mutants, the Phe44 -> Val mutant was a much less effective intramolecular chaperone with conversion to a digestible inhibitor, possibly owing to destruction of the tertiary structure. On the other hand, the Phe44 -> Leu or Ile mutant maintained its tertiary structure, and hence could function as a more effective intramolecular chaperone than the Phe44 -> Val mutant. Furthermore, since the Phe44 -> Leu mutant was a more susceptible inhibitor than POIA1cm, the halo formed around a colony of Bacillus cells transformed with a plasmid encoding this mutant was larger than others. These results clearly show the close relationship between the tertiary structure and functions of POIA1 as a protease inhibitor and an intramolecular chaperone, and that a combination of such inhibitory properties and intramolecular chaperone activity of POIA1 might affect the diameter of the halo formed around Bacillus colonies in vivo.

We previously demonstrated that Pleurotus ostreatus proteinase A inhibitor I (POIA1) could function as an intramolecular chaperone of subtilisin BPN', as in the case of the propeptide of subtilisin BPN', and that its Phe44 -> Ala mutant, which lost its tertiary structure, could not assist the refolding of subtilisin BPN'. In this study, we examined the effects of hydrophobic amino acid substitutions at other sites and substitutions of Phe44 with other hydrophobic residues on the structure and functions of POIA1. These mutations were introduced into POIA1cm that had been obtained by the substitution of the C-terminal six residues of POIA1 with those of the propeptide of subtilisin BPN'. When IIe32 or Ile64 was substituted with Ala, the tertiary structure of the resultant mutant was markedly destroyed, and the activities as a protease inhibitor and an intramolecular chaperone were significantly lowered. Among the position 44 mutants, the Phe44 -> Val mutant was a much less effective intramolecular chaperone with conversion to a digestible inhibitor, possibly owing to destruction of the tertiary structure. On the other hand, the Phe44 -> Leu or Ile mutant maintained its tertiary structure, and hence could function as a more effective intramolecular chaperone than the Phe44 -> Val mutant. Furthermore, since the Phe44 -> Leu mutant was a more susceptible inhibitor than POIA1cm, the halo formed around a colony of Bacillus cells transformed with a plasmid encoding this mutant was larger than others. These results clearly show the close relationship between the tertiary structure and functions of POIA1 as a protease inhibitor and an intramolecular chaperone, and that a combination of such inhibitory properties and intramolecular chaperone activity of POIA1 might affect the diameter of the halo formed around Bacillus colonies in vivo.

The polypeptide alpha 3, which was synthesized by us to produce an amphipathic helix structure, contains the regular three times repeated sequence (LETLAKA)(3), and alpha 3 forms a fibrous assembly. To clarify how the side chains of amino acid residues affect the formation of a helix, Leu residues, which are located in the hydrophobic surface of an amphipathic helix, were replaced by other hydrophobic aliphatic amino acid residues systematically, and the characters of the resulting polypeptides were studied. According to the circular dichroism (CD) spectra, the Ile-substituted polypeptides formed a helix like alpha 3. However, their helix formation ability was weaker than that of alpha 3 under some conditions. The Val-substituted polypeptides formed alpha helix only under restricted condition. The Ala-substituted polypeptides did not form alpha helix under any condition. Thus, it is clear that the order of the alpha helix formation ability is as follows: Leu >= Ile > Val > Ala. The formation of a helix was confirmed by Fourier Transform Infrared (FTIR) spectra. Through electron microscopic observation, it was clarified that the formation of the alpha helix structure correlates with the formation of a fibrous assembly. The amphipathic alpha helix structure would be stabilized by the formation of the fibrous assembly.

We have examined whether the long third intracellular loop (i3) of the muscarinic acetylcholine receptor M-2 subtype has a rigid structure. Circular dichroism (CD) and nuclear magnetic resonance spectra of M(2)i3 expressed in and purified from Escherichia coli indicated that M(2)i3 consists mostly of random coil. In addition, the differential CD spectrum between the M-2 and M(2)Delta i3 receptors, the latter of which lacks most of i3 except N- and C-terminal ends, gave no indication of secondary structure. These results suggest that the central part of i3 of the M-2 receptor has a flexible structure. (c) 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

We have examined whether the long third intracellular loop (i3) of the muscarinic acetylcholine receptor M-2 subtype has a rigid structure. Circular dichroism (CD) and nuclear magnetic resonance spectra of M(2)i3 expressed in and purified from Escherichia coli indicated that M(2)i3 consists mostly of random coil. In addition, the differential CD spectrum between the M-2 and M(2)Delta i3 receptors, the latter of which lacks most of i3 except N- and C-terminal ends, gave no indication of secondary structure. These results suggest that the central part of i3 of the M-2 receptor has a flexible structure. (c) 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

Pleurotus ostrearus proteinase A inhibitor I (POIA1), which was discovered as a protease inhibitor, is unique in that it shows sequence homology to the propeptide of subtilisin, which functions as an intramolecular of a cognate protease. In this study, we demonstrate that POIA1 can function as an intramolecular chaperone of subtilisin by in vitro and in vivo experiments. The specific cleavage between POIA1 and the mature region of subtilisin BPN' occurred in a refolding process of a chimera protein, and Bacillus cells transformed with a chimera gene formed a halo on a skim milk plate. The mutational analyses of POIA1 in the chimera protein suggested that the tertiary structure of POIA1 is required for such a function, and that an increase in its ability to bind to subtilisin BPN' makes POIA1 a more effective intramolecular chaperone. (c) 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

Pleurotus ostrearus proteinase A inhibitor I (POIA1), which was discovered as a protease inhibitor, is unique in that it shows sequence homology to the propeptide of subtilisin, which functions as an intramolecular of a cognate protease. In this study, we demonstrate that POIA1 can function as an intramolecular chaperone of subtilisin by in vitro and in vivo experiments. The specific cleavage between POIA1 and the mature region of subtilisin BPN' occurred in a refolding process of a chimera protein, and Bacillus cells transformed with a chimera gene formed a halo on a skim milk plate. The mutational analyses of POIA1 in the chimera protein suggested that the tertiary structure of POIA1 is required for such a function, and that an increase in its ability to bind to subtilisin BPN' makes POIA1 a more effective intramolecular chaperone. (c) 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

The α 3-peptide, which comprises three repeats of the sequence Leu-Glu-Thr-Leu-Ala-Lys-Ala and forms an amphipathic of-helix, is unique among various α-helix-forming peptides in that it assembles into fibrous structures that can be observed by transmission electron microscopy. As part of our investigation of the structure-stability relationships of the α 3-peptide, we synthesized the α 3-peptide, whose amino acid sequence is the reverse of that of the α 3-peptide, and we investigated the effects of sequence reversal on α-helix stability and the formation of fibrous structures. Unexpectedly, the r3-peptide formed a more-stable α-helix and longer fibers than did the α 3-peptide. The stability of the r3-peptide helix decreased when the ionic strength or the buffer was increased and when the pH of the buffer was adjusted to 2 or 12. These results suggest that the r3-peptide underwent a "magnet-like" oligomerization and that an increase in the charge-distribution inequality may be the driving force for the formation of fibrous structures. © 2005 Elsevier Inc. All rights reserved.

The kinetics of the folding of a small protein, POIA1, and its two variants lacking proline residues, has been investigated by using stopped-flow measurements. Although all these proteins fold and unfold in a two-state manner, logarithms of the observed rate constants display a curvilinear dependence against denaturant concentration. Two possible mechanisms that can explain these phenomena are proposed. One is the two-state model with moving transition, which shows that the energy surface depicted from these curved chevron plots for each protein is broad and the transition state moves toward the native state with increasing concentration of the denaturant on the reaction coordinate. The shift in the transition state either increases or decreases depending on the mutation sites, reflecting changes in the high-energy structures. The other is the three-state model with a high-energy intermediate state, in which relative energies for the two transition states vary depending on the denaturant concentrations and mutation sites. Comparison of these two models sheds light on underlying differences in the folding mechanisms for these proteins. (C) 2004 Elsevier B.V. All rights reserved.

We previously noted that bovine apolipoprotein A-II (apoA-II) had a bactericidal effect causing morphological changes in the cytoplasm. To determine whether and how apoA-II and apoA-I, which have acidic isoelectric points (pIs), enter cells, we determined the rates of uptake of FITC-labeled proteins by fibroblast cells and found that they entered cells more easily at low pH than at neutral pH under conditions where endocytosis was inhibited. The enhanced uptake of proteins at low pH was also observed for other proteins examined regardless of the molecular weight (M-r) or pI in a time-dependent manner, although the efficiency of uptake varied among the proteins. Furthermore, a pH gradient was shown to be the main driving force for the translocation. As cells were viable above pH 4 for 2 h at 4degreesC and internalized beta-galactosidase was active under these conditions, we suggest that this procedure is applicable to the injection of proteins into cells without the use of an apparatus such as a microinjector.

We previously noted that bovine apolipoprotein A-II (apoA-II) had a bactericidal effect causing morphological changes in the cytoplasm. To determine whether and how apoA-II and apoA-I, which have acidic isoelectric points (pIs), enter cells, we determined the rates of uptake of FITC-labeled proteins by fibroblast cells and found that they entered cells more easily at low pH than at neutral pH under conditions where endocytosis was inhibited. The enhanced uptake of proteins at low pH was also observed for other proteins examined regardless of the molecular weight (M-r) or pI in a time-dependent manner, although the efficiency of uptake varied among the proteins. Furthermore, a pH gradient was shown to be the main driving force for the translocation. As cells were viable above pH 4 for 2 h at 4degreesC and internalized beta-galactosidase was active under these conditions, we suggest that this procedure is applicable to the injection of proteins into cells without the use of an apparatus such as a microinjector.

The gene encoding an enolase from Desulfovibrio vulgaris (Miyazaki F) was cloned and overexpressed in Escherichia coli. A 2.1-kb DNA fragment, isolated from D. vulgaris (Miyazaki F) by double digestion with PstI and BamHI, contained an enolase gene (eno) and part of the methylenetetrahydrofolate dehydrogenase gene (folD). The nucleotide sequence of eno indicates that the protein monomer is composed of 434 amino acids. An expression system for eno under control of the T7 promoter was constructed in E. coli. The purified His-tagged enolase formed a homooctamer and was active in the formation of phosphoenolpyruvate (PEP) as well as in the reverse reaction, the formation of D-(+)-2-phosphoglyceric acid (2-PGA). The pH dependence and kinetic properties of the recombinant enolase from the sulfate-reducing bacterium were also studied. The amounts of eno mRNA when the bacterium was grown on glycerol or glucose were compared to that when D. vulgaris was grown on lactate. (C) 2003 Elsevier B.V. All rights reserved.

The gene encoding an enolase from Desulfovibrio vulgaris (Miyazaki F) was cloned and overexpressed in Escherichia coli. A 2.1-kb DNA fragment, isolated from D. vulgaris (Miyazaki F) by double digestion with PstI and BamHI, contained an enolase gene (eno) and part of the methylenetetrahydrofolate dehydrogenase gene (folD). The nucleotide sequence of eno indicates that the protein monomer is composed of 434 amino acids. An expression system for eno under control of the T7 promoter was constructed in E. coli. The purified His-tagged enolase formed a homooctamer and was active in the formation of phosphoenolpyruvate (PEP) as well as in the reverse reaction, the formation of D-(+)-2-phosphoglyceric acid (2-PGA). The pH dependence and kinetic properties of the recombinant enolase from the sulfate-reducing bacterium were also studied. The amounts of eno mRNA when the bacterium was grown on glycerol or glucose were compared to that when D. vulgaris was grown on lactate. (C) 2003 Elsevier B.V. All rights reserved.

The ovomucoid third domain from silver pheasant (OMSVP3), a typical Kazal-type inhibitor, strongly inhibits different serine proteases of various specificities, i.e., chymotrypsin, Streptomyces griseus protease, subtilisin, and elastase. Structural studies have suggested that conformational flexibility in the reactive site loop of the free inhibitor may be related to broad specificity of the ovomucoid. On the basis of the structural homology between OMSVP3 and ascidian trypsin inhibitor (ATI), which has a cystine-stabilized alpha-helical (CSH) motif in the sequence, we prepared the disulfide variant of OMSVP3, introducing an engineered disulfide bond between positions 14 and 39 near the reactive site (Met(18)-Glu(19)) by site-directed mutagenesis. The disulfide variant P14C/N39C retained potent inhibitory activities toward alpha-chymotrypsin (CHT) and S. griseus proteases A and B (SGPA and SGPB), while this variant lost most of its inhibitory activity toward porcine pancreatic elastase (PPE). We determined the solution structure of P14C/N39C, as well as that of wild-type OMSVP3, by two-dimensional nuclear magnetic resonance (2D NMR) methods and compared their structures to elucidate the structural basis of the inhibitory specificity change. For the molecular core consisting of a central alpha-helix and a three-stranded antiparallel beta-sheet, essentially no structural difference was detected between the two (pairwise rmsd value = 0.41 Angstrom). In contrast to this, a significant difference was detected in the loop from Cys(8) to Thr(17), where in P14C/N39C it has drawn approximately 4 Angstrom nearer the central helix to form the engineered Cys(14)-Cys(39) bond. Concomitantly, the Tyr(11)-Pro(12) cis-peptide linkage, which is highly conserved in ovomucoid third domains, was isomerized to the trans configuration. Such structural change in the loop near the reactive site may possibly affect the inhibitory specificity of P14C/N39C for the corresponding proteases.

The ovomucoid third domain from silver pheasant (OMSVP3), a typical Kazal-type inhibitor, strongly inhibits different serine proteases of various specificities, i.e., chymotrypsin, Streptomyces griseus protease, subtilisin, and elastase. Structural studies have suggested that conformational flexibility in the reactive site loop of the free inhibitor may be related to broad specificity of the ovomucoid. On the basis of the structural homology between OMSVP3 and ascidian trypsin inhibitor (ATI), which has a cystine-stabilized alpha-helical (CSH) motif in the sequence, we prepared the disulfide variant of OMSVP3, introducing an engineered disulfide bond between positions 14 and 39 near the reactive site (Met(18)-Glu(19)) by site-directed mutagenesis. The disulfide variant P14C/N39C retained potent inhibitory activities toward alpha-chymotrypsin (CHT) and S. griseus proteases A and B (SGPA and SGPB), while this variant lost most of its inhibitory activity toward porcine pancreatic elastase (PPE). We determined the solution structure of P14C/N39C, as well as that of wild-type OMSVP3, by two-dimensional nuclear magnetic resonance (2D NMR) methods and compared their structures to elucidate the structural basis of the inhibitory specificity change. For the molecular core consisting of a central alpha-helix and a three-stranded antiparallel beta-sheet, essentially no structural difference was detected between the two (pairwise rmsd value = 0.41 Angstrom). In contrast to this, a significant difference was detected in the loop from Cys(8) to Thr(17), where in P14C/N39C it has drawn approximately 4 Angstrom nearer the central helix to form the engineered Cys(14)-Cys(39) bond. Concomitantly, the Tyr(11)-Pro(12) cis-peptide linkage, which is highly conserved in ovomucoid third domains, was isomerized to the trans configuration. Such structural change in the loop near the reactive site may possibly affect the inhibitory specificity of P14C/N39C for the corresponding proteases.

Single amino acid mutations of Met103 in the hydrophobic core of a serine protease inhibitor, Streptomyces subtilisin inhibitor, caused little change in the inhibitory activity, as measured by the inhibitor constant, although some altered the thermodynamic stability of the protein considerably. H-1 NMR investigations showed that the conformational stress caused by the replacement of Met103 with Gly, Ala, Val, and Ile, namely, the effects of the cavities generated by replacements with smaller side-chains and of the steric distortions generated by beta-branched side-chains, caused considerable changes in the structural arrangement of the side-chains within the core. However, these structural changes were absorbed within the hydrophobic core, without distorting the structure of the reactive site essential for the protein function. These results provide an excellent example of the conformational flexibility of a protein core and the degree of its tolerance of an amino acid replacement. The results also reveal the crucially designed structural relationship between the core of the inhibitor and the enzyme-binding segment with the reactive site in a serine protease inhibitor.

Single amino acid mutations of Met103 in the hydrophobic core of a serine protease inhibitor, Streptomyces subtilisin inhibitor, caused little change in the inhibitory activity, as measured by the inhibitor constant, although some altered the thermodynamic stability of the protein considerably. H-1 NMR investigations showed that the conformational stress caused by the replacement of Met103 with Gly, Ala, Val, and Ile, namely, the effects of the cavities generated by replacements with smaller side-chains and of the steric distortions generated by beta-branched side-chains, caused considerable changes in the structural arrangement of the side-chains within the core. However, these structural changes were absorbed within the hydrophobic core, without distorting the structure of the reactive site essential for the protein function. These results provide an excellent example of the conformational flexibility of a protein core and the degree of its tolerance of an amino acid replacement. The results also reveal the crucially designed structural relationship between the core of the inhibitor and the enzyme-binding segment with the reactive site in a serine protease inhibitor.

We previously showed that bovine apolipoprotein A-II (apoA-II) has antimicrobial activity against Escherichia coli in PBS, and its C-terminal residues 49-76 are responsible for the activity using synthetic peptides. In order to understand the structural requirements of peptide 49-76 for the antimicrobial activity, the N- or C-terminus was truncated and then the charged (Lys or Asp) or Ser residues were replaced by Ala. Deletion of the first or last three amino acids and replacement of Lys-54/55 or 71/72 by Ala caused a substantial decreases in alpha-helical content in 50% TFE, showing the possible presence of helices in N- and C-terminal regions, respectively. The anti-Escherichia coli activity of the peptide correlated with its liposome-binding activity. Replacement of Lys-54/55 or 71/72 by Ala resulted in an almost complete loss of anti-E. coli activity with a substantial decrease in liposome-binding activity. Moreover, deletion of the last three amino acids caused a reduction to 1/17 of the original anti-E. coli activity with a moderate decrease in liposome-binding activity. In contrast, replacement of Ser-65/66, Asp-59, or Asp-69 by Ala hardly affected the anti-E. coli activity. These findings suggest that Lys-54/55 and Lys-71/72 on the putative helices are critical for antimicrobial activity, and the C-terminal 3 amino acids are important for the structural integrity of the C-terminal region for effective antimicrobial activity.

Pleurotus ostreatus proteinase A inhibitor 1 (POIA1) has been shown to be unique among the various serine protease inhibitors in that its C-terminal region appears to be the reactive site responsible for its inhibitory action toward proteases. To investigate in more detail the mechanism of inhibition by POIA1, we have been studying its structural requirements for stable inhibition of proteases. In this study, we focused on hydrophobic Phe residues, which are generally located in the interior of protein molecules. A Phe-->Ala replacement at position 44 or 56 was introduced into a `parent' mutant of POIA1 that had been converted into a strong and resistant inhibitor of subtilisin BPN' by replacement of its six C-terminal residues with those of the propeptide of subtilisin BPN' and the effects on inhibitory properties and structural stability were examined. Both of the mutated POIA1 molecules not only were found to exhibit decreased ability to bind to subtilisin BPN' (80-fold for the F44A mutant and 13-fold for the F56A mutant), but were also converted to temporary inhibitors that were degraded by the protease. The structural stability of the mutated POIA1 was also lowered, as shown by a 13degreesC decrease in melting temperature for the F56A mutant. In particular, the F44A mutant was found to lose its tertiary structure, as judged from the circular dichroism spectrum, demonstrating that Phe44 is a strict requirement for structural formation by the POIA1 molecule. These results clearly indicate that stabilization of POIA1 by hydrophobic residues in its molecular interior is required for stable inhibition of the protease. This requirement for a stable tertiary structure is shared with other serine protease inhibitors, but other structural requirements seem to differ, in that strong binding with the protease is required for POIA1 whereas conformational rigidity around the reactive site is essential for many other protease inhibitors.

Pleurotus ostreatus proteinase A inhibitor 1 (POIA1) has been shown to be unique among the various serine protease inhibitors in that its C-terminal region appears to be the reactive site responsible for its inhibitory action toward proteases. To investigate in more detail the mechanism of inhibition by POIA1, we have been studying its structural requirements for stable inhibition of proteases. In this study, we focused on hydrophobic Phe residues, which are generally located in the interior of protein molecules. A Phe-->Ala replacement at position 44 or 56 was introduced into a `parent' mutant of POIA1 that had been converted into a strong and resistant inhibitor of subtilisin BPN' by replacement of its six C-terminal residues with those of the propeptide of subtilisin BPN' and the effects on inhibitory properties and structural stability were examined. Both of the mutated POIA1 molecules not only were found to exhibit decreased ability to bind to subtilisin BPN' (80-fold for the F44A mutant and 13-fold for the F56A mutant), but were also converted to temporary inhibitors that were degraded by the protease. The structural stability of the mutated POIA1 was also lowered, as shown by a 13degreesC decrease in melting temperature for the F56A mutant. In particular, the F44A mutant was found to lose its tertiary structure, as judged from the circular dichroism spectrum, demonstrating that Phe44 is a strict requirement for structural formation by the POIA1 molecule. These results clearly indicate that stabilization of POIA1 by hydrophobic residues in its molecular interior is required for stable inhibition of the protease. This requirement for a stable tertiary structure is shared with other serine protease inhibitors, but other structural requirements seem to differ, in that strong binding with the protease is required for POIA1 whereas conformational rigidity around the reactive site is essential for many other protease inhibitors.

Solution structure of POIA1 (Pleurotus ostreatus proteinase A inhibitor 1), which functions as an intramolecular chaperone and as an inhibitor to subtilisin, was determined. By making use of the fact that POIA1 is the only structured protein that shows homology to the propeptide of subtilisin, which is unstructured by itself, foldability of this protein was elucidated. It became clear that the evolutionarily conserved residues play two important roles, one for the maintenance of its own structure, and the other for the interaction with subtilisin. Structural softness and mutational tolerance contained in the POIA1 structure makes it an ideal material for designing a foldable protein. (C) 2002 Elsevier Science Ltd.

Solution structure of POIA1 (Pleurotus ostreatus proteinase A inhibitor 1), which functions as an intramolecular chaperone and as an inhibitor to subtilisin, was determined. By making use of the fact that POIA1 is the only structured protein that shows homology to the propeptide of subtilisin, which is unstructured by itself, foldability of this protein was elucidated. It became clear that the evolutionarily conserved residues play two important roles, one for the maintenance of its own structure, and the other for the interaction with subtilisin. Structural softness and mutational tolerance contained in the POIA1 structure makes it an ideal material for designing a foldable protein. (C) 2002 Elsevier Science Ltd.

The propeptide of subtilisin BPN', which functions as an intramolecular chaperone and a temporary inhibitor of subtilisin, is unique in that it acquires its three-dimensional structure by formation of a complex with the cognate protease. We previously showed that the successive amino acid replacements Ala47 --> Phe, Gly13 --> Ile, and Val65 --> Ile in the propeptide to increase its hydrophobicity resulted in formation of a tertiary structure, accompanied by increased ability to bind to the protease and increased resistance to proteolysis. In this study, we examined the effects of these tertiary-structure-forming mutations on the intramolecular chaperone activity of the propeptide. The successive amino acid replacements mentioned above were introduced into pro-subtilisin*, possessing a Ser221 --> Ala mutation in the catalytic residue. Refolding experiments were started by rapid dilution of the denatured pro-subtilisin*, and formation of tertiary structure in subtilisin was monitored kinetically by increase in tryptophan fluorescence. The wild-type pro-subtilisin* was found to refold with a rate constant of 4.8 x 10(-3) s(-1) in the equation describing an intramolecular process. The Ala47 --> Phe replacement in the propeptide resulted in a 1.2-fold increase in the rate constant of subtilisin refolding. When the additional replacement Gly13 --> Ile was introduced, refolding of subtilisin was substantially accelerated, and its kinetics could be fitted to a double exponential process composed of a fast phase with a rate constant of 2.1 x 10(-2) s(-1) and a slow phase with a rate constant of 4.5 x 10(-3) s(-1). The rate constant of the fast phase was increased slightly by a further replacement, Val65 --> Ile. Since the slow phase is considered to correspond to proline isomerization, we concluded that tertiary-structure-forming mutations in the propeptide produce positive effects on its intramolecular chaperone activity through acceleration of the propeptide-induced formation of the tertiary structure of subtilisin BPN'.

The propeptide of subtilisin BPN', which functions as an intramolecular chaperone and a temporary inhibitor of subtilisin, is unique in that it acquires its three-dimensional structure by formation of a complex with the cognate protease. We previously showed that the successive amino acid replacements Ala47 --> Phe, Gly13 --> Ile, and Val65 --> Ile in the propeptide to increase its hydrophobicity resulted in formation of a tertiary structure, accompanied by increased ability to bind to the protease and increased resistance to proteolysis. In this study, we examined the effects of these tertiary-structure-forming mutations on the intramolecular chaperone activity of the propeptide. The successive amino acid replacements mentioned above were introduced into pro-subtilisin*, possessing a Ser221 --> Ala mutation in the catalytic residue. Refolding experiments were started by rapid dilution of the denatured pro-subtilisin*, and formation of tertiary structure in subtilisin was monitored kinetically by increase in tryptophan fluorescence. The wild-type pro-subtilisin* was found to refold with a rate constant of 4.8 x 10(-3) s(-1) in the equation describing an intramolecular process. The Ala47 --> Phe replacement in the propeptide resulted in a 1.2-fold increase in the rate constant of subtilisin refolding. When the additional replacement Gly13 --> Ile was introduced, refolding of subtilisin was substantially accelerated, and its kinetics could be fitted to a double exponential process composed of a fast phase with a rate constant of 2.1 x 10(-2) s(-1) and a slow phase with a rate constant of 4.5 x 10(-3) s(-1). The rate constant of the fast phase was increased slightly by a further replacement, Val65 --> Ile. Since the slow phase is considered to correspond to proline isomerization, we concluded that tertiary-structure-forming mutations in the propeptide produce positive effects on its intramolecular chaperone activity through acceleration of the propeptide-induced formation of the tertiary structure of subtilisin BPN'.

We identified a gene encoding a catalase from the anaerobic bacteria Desulfovibrio vulgaris (Miyazaki F), and the expression of its gene in Escherichia coli, The 3.3-kbp DNA fragment isolated from D. vulgaris (Miyazaki F) by double digestion with EcoRI and SalI was found to produce a protein that binds protoheme IX as a prosthetic group in E. coli, This DNA fragment contained a putative open reading frame (Kat) and one part of another open reading frame (ORF-1), The amino acid sequence of the amino terminus of the protein purified from the transformed cells was consistent with that deduced from the nucleotide sequence of Kat in the cloned fragment of D. vulgaris (Miyazaki F) DNA, which may include promoter and regulatory sequences, The nucleotide sequence of Kat indicates that the protein is composed of 479 amino acids per monomer, The recombinant catalase was found to be active in the decomposition of hydrogen peroxide, as are other catalases from aerobic organisms, but its K-m value was much greater. The hydrogen peroxide stress against D. vulgaris (Miyazaki F) induced the activity for the decomposition of hydrogen peroxide somewhat, so the catalase gene may not work effectively in vivo.

Pleurotus ostreatus proteinase A inhibitor 1 (POIA1), which is composed of 76 residues without disulfide bridges, is a unique inhibitor in that it exhibits sequence similarity to the propeptides of subtilisins. In order to elucidate the inhibitory mechanism of POIA1, we constructed an expression system for a synthetic POIA1 gene. The wild-type POIA1 was found to inhibit subtilisin BPN ' with an inhibitor constant (K(i)) of 3.2 x 10(-9) M, but exhibited a time-dependent decrease of inhibitory activity as a consequence of degradation by the protease, showing that the wild-type POIA1 was a temporary inhibitor when subtilisin BPN ' was used as a target protease. Since POIA1 shows sequence similarity to the propeptide of subtilisin, which is known to inhibit the protease via its C-terminal region, the C-terminal six residues of POIA1 were replaced with those of the propeptide of subtilisin BPN '. The mutated POIA1 inhibited subtilisin BPN ' with a K(i) value of 2.8 x 10(-11) M and did not exhibit time-dependent decrease of inhibitory activity, showing about 100-fold increases in binding affinity for, and resistance to, the protease. These results clearly indicate that the C-terminal region of POIA1 plays an important role in determining the inhibitory activity toward the protease, and that the increase in binding ability to the protease is closely related to resistance to proteolytic degradation. Therefore, the inhibitory properties of POIA1 can be altered by mutation of its C-terminal region. (C) 2001 Academic Press.

We identified a gene encoding a catalase from the anaerobic bacteria Desulfovibrio vulgaris (Miyazaki F), and the expression of its gene in Escherichia coli, The 3.3-kbp DNA fragment isolated from D. vulgaris (Miyazaki F) by double digestion with EcoRI and SalI was found to produce a protein that binds protoheme IX as a prosthetic group in E. coli, This DNA fragment contained a putative open reading frame (Kat) and one part of another open reading frame (ORF-1), The amino acid sequence of the amino terminus of the protein purified from the transformed cells was consistent with that deduced from the nucleotide sequence of Kat in the cloned fragment of D. vulgaris (Miyazaki F) DNA, which may include promoter and regulatory sequences, The nucleotide sequence of Kat indicates that the protein is composed of 479 amino acids per monomer, The recombinant catalase was found to be active in the decomposition of hydrogen peroxide, as are other catalases from aerobic organisms, but its K-m value was much greater. The hydrogen peroxide stress against D. vulgaris (Miyazaki F) induced the activity for the decomposition of hydrogen peroxide somewhat, so the catalase gene may not work effectively in vivo.

Pleurotus ostreatus proteinase A inhibitor 1 (POIA1), which is composed of 76 residues without disulfide bridges, is a unique inhibitor in that it exhibits sequence similarity to the propeptides of subtilisins. In order to elucidate the inhibitory mechanism of POIA1, we constructed an expression system for a synthetic POIA1 gene. The wild-type POIA1 was found to inhibit subtilisin BPN ' with an inhibitor constant (K(i)) of 3.2 x 10(-9) M, but exhibited a time-dependent decrease of inhibitory activity as a consequence of degradation by the protease, showing that the wild-type POIA1 was a temporary inhibitor when subtilisin BPN ' was used as a target protease. Since POIA1 shows sequence similarity to the propeptide of subtilisin, which is known to inhibit the protease via its C-terminal region, the C-terminal six residues of POIA1 were replaced with those of the propeptide of subtilisin BPN '. The mutated POIA1 inhibited subtilisin BPN ' with a K(i) value of 2.8 x 10(-11) M and did not exhibit time-dependent decrease of inhibitory activity, showing about 100-fold increases in binding affinity for, and resistance to, the protease. These results clearly indicate that the C-terminal region of POIA1 plays an important role in determining the inhibitory activity toward the protease, and that the increase in binding ability to the protease is closely related to resistance to proteolytic degradation. Therefore, the inhibitory properties of POIA1 can be altered by mutation of its C-terminal region. (C) 2001 Academic Press.

We have found two isoforms of the eukaryotic translation initiation factor 4E (eIF4E) in Xenopus laevis. These proteins differ in length by 18 amino acids. Overexpression of either of the two eIF4E proteins modestly increase translation in Xenopus oocytes. The results suggest that both of these two isoforms function in translation.

We have found two isoforms of the eukaryotic translation initiation factor 4E (eIF4E) in Xenopus laevis. These proteins differ in length by 18 amino acids. Overexpression of either of the two eIF4E proteins modestly increase translation in Xenopus oocytes. The results suggest that both of these two isoforms function in translation.

We have previously shown that replacing the Pi-site residue (Ala) of chicken ovomucoid domain 3 (OMCHI3) with a Met or Lys results in the acquisition of inhibitory activity toward chymotrypsin or trypsin, respectively. However, the inhibitory activities thus induced are not strong. In the present study, we introduced additional amino acid replacements around the reactive site to try to make the P1-site mutants more effective inhibitors of chymotrypsin or trypsin. The amino acid replacement Asp-->Tyr at the P2' site of OMCHI3(P1Met) resulted in conversion to a 35000-fold more effective inhibitor of chymotrypsin with an inhibitor constant (Ki) of 1.17x10(-11) M, The Ki value of OMCHI3(P1Met, P2'Ala) indicated that the effect on the interaction with chymotrypsin of removing a negative charge from the P2' site was greater than that of introducing an aromatic ring. Similarly, enhanced inhibition of trypsin was observed when the Asp-->Tyr replacement was introduced into the P2' site of OMCHI3(P1Lys). Two additional replacements, Asp-->Ala at the P4 site and Arg-->Ala at the P3' site, made the mutant a more effective inhibitor of trypsin with a K(i) value of 1.44x10(-9) M, By contrast, Arg-->Ala replacement at the P3' site of OMCHI3(P1Met, P2'Tyr) resulted in a greatly reduced inhibition of chymotrypsin, and Asp-->Ala replacement at the P4 site produced only a small change when compared with a natural variant of OMCHI3. These results clearly indicate that not only the PI-site residue but also the characteristics, particularly the electrostatic properties, of the amino acid residues around the reactive site of the protease inhibitor determine the strength of its interactions with proteases, Furthermore, amino acids with different characteristics are required around the reactive site for strong inhibition of chymotrypsin and trypsin.

We have previously shown that replacing the Pi-site residue (Ala) of chicken ovomucoid domain 3 (OMCHI3) with a Met or Lys results in the acquisition of inhibitory activity toward chymotrypsin or trypsin, respectively. However, the inhibitory activities thus induced are not strong. In the present study, we introduced additional amino acid replacements around the reactive site to try to make the P1-site mutants more effective inhibitors of chymotrypsin or trypsin. The amino acid replacement Asp-->Tyr at the P2' site of OMCHI3(P1Met) resulted in conversion to a 35000-fold more effective inhibitor of chymotrypsin with an inhibitor constant (Ki) of 1.17x10(-11) M, The Ki value of OMCHI3(P1Met, P2'Ala) indicated that the effect on the interaction with chymotrypsin of removing a negative charge from the P2' site was greater than that of introducing an aromatic ring. Similarly, enhanced inhibition of trypsin was observed when the Asp-->Tyr replacement was introduced into the P2' site of OMCHI3(P1Lys). Two additional replacements, Asp-->Ala at the P4 site and Arg-->Ala at the P3' site, made the mutant a more effective inhibitor of trypsin with a K(i) value of 1.44x10(-9) M, By contrast, Arg-->Ala replacement at the P3' site of OMCHI3(P1Met, P2'Tyr) resulted in a greatly reduced inhibition of chymotrypsin, and Asp-->Ala replacement at the P4 site produced only a small change when compared with a natural variant of OMCHI3. These results clearly indicate that not only the PI-site residue but also the characteristics, particularly the electrostatic properties, of the amino acid residues around the reactive site of the protease inhibitor determine the strength of its interactions with proteases, Furthermore, amino acids with different characteristics are required around the reactive site for strong inhibition of chymotrypsin and trypsin.

Transition protein 2 (TP2; 137 amino acid residues) from boar late spermatid nuclei has three potential zinc finger motifs in the N-terminal 3/4 region. Gel shift assays revealed that boar TP2 recognized a CpG island sequence in a zinc-dependent manner. However, there was some nonspecific recognition of the oligonucleotide. Then, we constructed the expression system of zinc-binding domain of TP2 (TP2Z) (residues 1-103) in Escherichia coli. Double-stranded DNA fragments encoding TP2Z were synthesized as 18 fragments with 103 residues, annealed, and cloned into the expression plasmid pET11d. TP2Z was expressed upon induction with 1 mM isopropylthiogalactoside and extracted with acid including 0.71 M 2-mercaptoethanol. TP2Z was purified by ion-exchange chromatography on Fractogel EMD SO3- and HPLC on Nucleosil 300 7C18 and on Diol-120. Atomic absorption and CD spectroscopy showed that TP2Z bound three atoms of zinc per molecule of the protein and underwent a zinc-dependent conformational change in a manner similar to that for intact TP2. Gel shift assays indicated that TP2Z recognized a CpG island sequence more specifically than intact TP2 and that the specificity is dependent on zinc. (C) 1999 Academic Press.

Transition protein 2 (TP2; 137 amino acid residues) from boar late spermatid nuclei has three potential zinc finger motifs in the N-terminal 3/4 region. Gel shift assays revealed that boar TP2 recognized a CpG island sequence in a zinc-dependent manner. However, there was some nonspecific recognition of the oligonucleotide. Then, we constructed the expression system of zinc-binding domain of TP2 (TP2Z) (residues 1-103) in Escherichia coli. Double-stranded DNA fragments encoding TP2Z were synthesized as 18 fragments with 103 residues, annealed, and cloned into the expression plasmid pET11d. TP2Z was expressed upon induction with 1 mM isopropylthiogalactoside and extracted with acid including 0.71 M 2-mercaptoethanol. TP2Z was purified by ion-exchange chromatography on Fractogel EMD SO3- and HPLC on Nucleosil 300 7C18 and on Diol-120. Atomic absorption and CD spectroscopy showed that TP2Z bound three atoms of zinc per molecule of the protein and underwent a zinc-dependent conformational change in a manner similar to that for intact TP2. Gel shift assays indicated that TP2Z recognized a CpG island sequence more specifically than intact TP2 and that the specificity is dependent on zinc. (C) 1999 Academic Press.

Yeast proteinase B inhibitor 2 (YIB2), which is composed of 74 amino acid residues, is an unusual serine protease inhibitor, since it lacks disulfide bonds. To identify its reactive site for proteases, we constructed an expression system for a synthetic YIB2 gene and then attempted to change the inhibitory properties of YIB2 by amino acid replacements. The purified wild-type YIB2 inhibited the activity of subtilisin BPN', a protein homologous to yeast proteinase B, although its binding ability was not strong, and a time-dependent decrease in its inhibitory activity was observed, demonstrating that wild-type YIB2 behaves as a temporary inhibitor when subtilisin BPN' is the target protease. Since YIB2 exhibits sequence homology to the propeptide of subtilisin, which inhibits a cognate protease using its C-terminal region, we replaced the six C-terminal residues of YIB2 with those of the propeptide of subtilisin BPN' to make the mutant YIB2m1. This mutant exhibited markedly increased inhibitory activity toward subtilisin BPN' without a time-dependent decrease in its inhibitory activity. Replacement of only the C-terminal Asn of YIB2 by Tyr, or deletion of the C-terminal Tyr. Of YIB2m1, inhibited subtilisin, but the ability of these mutants to bind subtilisin and their resistance to proteolytic attack were weaker than those of YIB2ml, indicating that the C-terminal residue contributes to the interaction with the protease to a greater extent than the preceding five residues and that the resistance of YIB2 to proteolyic attack is closely related to its ability to bind a protease. These results demonstrate that YIB2 is a unique protease inhibitor that involves its C-terminal region in the interaction with the protease. (C) 1999 Academic Press.