The high resolution crystal structure for class A beta-lactamase PER-1 reveals the bases for its increase in breadth of activity

The treatment of infectious diseases by beta-lactam antibiotics is continuously challenged by the emergence and dissemination of new beta-lactamases. In most cases, the cephalosporinase activity of class A enzymes results from a few mutations in the TEM and SHV penicillinases. The PER-1 beta-lactamase was characterized as a class A enzyme displaying a cephalosporinase activity. This activity was, however, insensitive to the mutations of residues known to be critical for providing extended substrate profiles to TEM and SHV. The x-ray structure of the protein, solved at 1.9-A resolution, reveals that two of the most conserved features in class A beta-lactamases are not present in this enzyme: the fold of the Omega-loop and the cis conformation of the peptide bond between residues 166 and 167. The new fold of the Omega-loop and the insertion of four residues at the edge of strand S3 generate a broad cavity that may easily accommodate the bulky substituents of cephalosporin substrates. The trans conformation of the 166-167 bond is related to the presence of an aspartic acid at position 136. Selection of class A enzymes based on the occurrence of both Asp(136) and Asn(179) identifies a subgroup of enzymes with high sequence homology.     

Active-site residues of the transpeptidase domain of penicillin-binding protein 2 from Escherichia coli: similarity in catalytic mechanism to class A beta-lactamases.

By means of amino acid sequence alignment with class A beta-lactamases, the residues essential for the catalytic activity of the peptidoglycan transpeptidase of penicillin-binding protein 2 (PBP2) have been predicted to be Lys333, Asp447, and Lys544, in addition to the acylation site residue for the acyl-enzyme mechanism, Ser330. Accordingly, these residues were replaced by site-directed mutagenesis, and the resultant mutants were examined as to penicillin-binding activity and genetic complementation, which represent only the acylation step and the total reaction during transpeptidation, respectively. All the mutants at position 333 showed the complete loss of both the binding and complementation activities. Most of the mutants at position 447 retained the binding activity but lost the complementation activity, the exception being the D447E mutant, which retained both. The binding rates for various penicillins of the D447N mutant, which had lost the complementation activity, were almost identical to those of the wild type. The binding of the mutants at position 544 tended to require a higher penicillin concentration, and that of the K544H mutant required a lower pH. When the roles of the counterpart residues, Lys73, Glu166, and Lys234, in class A beta-lactamases were considered, the results suggested that Lys333 and Asp447 are essential for the acylation and acyl-transfer steps, respectively, and that Lys544 stabilizes the Michaelis complex through its side-chain positive charge.     

Selection of catalytically active biotin ligase and trypsin mutants by phage display.

Phage display has been shown to facilitate greatly the selection of polypeptides with desired properties by establishing a direct link between the polypeptide and the gene that encodes it. However, selection for catalytic activities displayed on phage remains a challenge, since reaction products diffuse away from the enzyme and make it difficult to recover catalytically active phage-enzymes. We have recently described a selection methodology in which the reaction substrate (and eventually the reaction product) is anchored on calmodulin-tagged phage-enzymes by means of a calmodulin binding peptide. Phage displaying a catalytic activity are physically isolated by means of affinity reagents specific for the product of reaction. In this study, we investigated the efficiency of selection for catalysis by phage display, using a ligase (the Escherichia coli biotin ligase BirA) and an endopeptidase (the rat trypsin His57--> Ala mutant) as model enzymes. These enzymes could be displayed on phage as fusion proteins with calmodulin and the minor coat protein pIII. Both the display of functional enzyme and the efficiency of selection for catalysis were significantly improved by using phage vectors, rather than phagemid vectors. In model selection experiments, phage displaying BirA were consistently enriched (between 4-fold and 800-fold) per round of panning, relative to negative controls. Phage displaying the trypsin His57-->Ala mutant, a relatively inefficient endopeptidase which cleaves a specific dipeptide sequence, were enriched (between 15-fold and 2000-fold), relative to negative controls. In order to improve the catalytic properties of the trypsin His57-->Ala mutant, we constructed a combinatorial phage display library of trypsin mutants. Selection of catalytically active phage-enzymes was evidentiated by increasing phage titres at the different rounds of panning relative to negative control selections, but mutants with catalytic properties superior to those of trypsin His57-->Ala mutant could not be isolated. The results obtained provide evidence that catalytic activities can be recovered using phage display technology, but stress the importance of both library design and stringent biopanning conditions for the recovery of novel enzymes.     

Interaction between the peripheral site residues of human butyrylcholinesterase, D70 and Y332, in binding and hydrolysis of substrates

Human butyrylcholinesterase displays substrate activation with positively charged butyrylthiocholine (BTC) as the substrate. Peripheral anionic site (PAS) residues D70 and Y332 appear to be involved in the initial binding of charged substrates and in activation control. To determine the contribution of PAS residues to binding and hydrolysis of quaternary substrates and activation control, the single mutants D70G/Y and Y332F/A/D and the double mutants Y332A/D70G and Y332D/D70Y were studied. Steady-state hydrolysis of the charged substrates, BTC and succinyldithiocholine, and the neutral ester o-nitrophenyl butyrate was measured. In addition, inhibition of wild-type and mutant enzymes by tetramethylammonium was investigated, at low concentrations of BTC. Single and double mutants of D70 and Y332 showed little or no substrate activation, suggesting that both residues were important for activation control. The effects of double mutations on D70 and Y332 were complex. Double-mutant cycle analysis provided evidence for interaction between these residues. The category of interaction (either synergistic, additive, partially additive or antagonistic) was found to depend on the nature of the substrate and on measured binding or kinetic parameters. This complexity reflects both the cross-talk between residues involved in the sequential formation of productive Michaelian complexes and the effect of peripheral site residues on catalysis. It is concluded that double mutations on the PAS induce a conformational change in the active site gorge of butyrylcholinesterase that can alter both substrate binding and enzyme acylation.     

Saturation mutagenesis of Asn152 reveals a substrate selectivity switch in P99 cephalosporinase

In class C beta-lactamases, the strictly conserved Asn152 forms part of an extended active-site hydrogen-bonding network. To probe its role in catalysis, all 19 mutants of Enterobacter cloacae P99 cephalosporinase Asn152 were simultaneously constructed and screened in Escherichia coli for their in vivo activity. The screen identified the previously uncharacterized mutants Asn152Ser, Asn152Thr, and Asn152Gly, which possess significant activity and altered substrate selectivity. In vitro measurement of Michaelis-Menten kinetic constants revealed that the Asn152Ser mutation causes a selectivity switch for penicillin G versus cefoxitin. Asn152Thr showed a 63-fold increase in k (cat) for oxacillin, a slow substrate for wild-type cephalosporinase. The results contribute to a growing body of data showing that mutation of highly conserved residues in the active site can result in substrate selectivity changes. The library screening method presented here would be applicable to substrate selectivity determination in other readily screenable enzymes.     

Characterization of the beta-lactam binding site of penicillin acylase of Escherichia coli by structural and site-directed mutagenesis studies

The binding of penicillin to penicillin acylase was studied by X-ray crystallography. The structure of the enzyme-substrate complex was determined after soaking crystals of an inactive betaN241A penicillin acylase mutant with penicillin G. Binding of the substrate induces a conformational change, in which the side chains of alphaF146 and alphaR145 move away from the active site, which allows the enzyme to accommodate penicillin G. In the resulting structure, the beta-lactam binding site is formed by the side chains of alphaF146 and betaF71, which have van der Waals interactions with the thiazolidine ring of penicillin G and the side chain of alphaR145 that is connected to the carboxylate group of the ligand by means of hydrogen bonding via two water molecules. The backbone oxygen of betaQ23 forms a hydrogen bond with the carbonyl oxygen of the phenylacetic acid moiety through a bridging water molecule. Kinetic studies revealed that the site-directed mutants alphaF146Y, alphaF146A and alphaF146L all show significant changes in their interaction with the beta-lactam substrates as compared with the wild type. The alphaF146Y mutant had the same affinity for 6-aminopenicillanic acid as the wild-type enzyme, but was not able to synthesize penicillin G from phenylacetamide and 6-aminopenicillanic acid. The alphaF146L and alphaF146A enzymes had a 3-5-fold decreased affinity for 6-aminopenicillanic acid, but synthesized penicillin G more efficiently than the wild type. The combined results of the structural and kinetic studies show the importance of alphaF146 in the beta-lactam binding site and provide leads for engineering mutants with improved synthetic properties.     

Retinoic acid receptor: a simulation analysis of retinoic acid binding and the resulting conformational changes.

The binding/escape mechanism of all- trans retinoic acid with respect to the ligand-binding domain of the nuclear receptor RARgamma has been studied by molecular dynamic simulations. The entry/exit channel was shown to be on the side of the activation helix by the use of multiple copy dynamics. Three independent minimum energy paths from the liganded structure to a model for the unliganded structure were calculated with the conjugate peak refinement method. Ligand escape takes place in the early steps of the transition during rearrangement of the binding pocket; the latter involves inward motion of the beta-sheet and outward motions of the Omega-loop and helix H6. The correlated rearrangements involved in the escape phase are similar and occur in the same order for the different paths. After the escape phase, the conformational changes affect primarily the C-terminal helices H11-H12 and the Omega-loop. The three paths are significantly different for this reorganization phase and reveal a multiplicity of possibilities, in agreement with the idea that the apo state is structurally less constrained. The present calculations extend the crystallographic results, confirming the "mouse trap" mechanism and stressing the importance of the helix H3 conformation and of the contacts between the Omega-loop and helices H11 and H6. They are in good agreement with known mutants and point to other functionally important residues, especially in helices H3 and H11, suggesting mutations that may affect the ligand-binding function and the associated conformational changes.     

Function of arginine-166 in the active site of Escherichia coli alkaline phosphatase

The function of arginine residue 166 in the active site of Escherichia coli alkaline phosphatase was investigated by site-directed mutagenesis. Two mutant versions of alkaline phosphatase, with either serine or alanine in the place of arginine at position 166, were generated by using a specially constructed M13 phage carrying the wild-type phoA gene. The mutant enzymes with serine and alanine at position 166 have very similar kinetic properties. Under conditions of no external phosphate acceptor, the kcat for the mutant enzymes decreases by approximately 30-fold while the Km increases by less than 2-fold. When kinetic measurements are carried out in the presence of a phosphate acceptor, 1.0 M Tris, the kcat for the mutant enzymes is reduced by less than 3-fold, while the Km increases by more than 50-fold. For both mutant enzymes, in either the absence or the presence of a phosphate acceptor, the catalytic efficiency as measured by the kcat/Km ratio decreases by approximately 50-fold as compared to the wild type. Measurements of the Ki for inorganic phosphate show an increase of approximately 50-fold for both mutants. Phenylglyoxal, which inactivates the wild-type enzyme, does not inactivate the Arg-166----Ala enzyme. This result indicates that Arg-166 is the same arginine residue that when chemically modified causes loss of activity [Daemen, F.J.M., & Riordan, J.F. (1974) Biochemistry 13, 2865-2871]. The data reported here suggest that although Arg-166 is important for activity is not essential. The analysis of the kinetic data also suggests that the loss of arginine-166 at the active site of alkaline phosphatase has two different effects on the enzyme. First, the binding of the substrate, and phosphate as a competitive inhibitor, is reduced; second, the rate of hydrolysis of the covalent phosphoenzyme may be diminished.     

Functional expression and affinity selection of single-chain cro by phage display: isolation of novel DNA-binding proteins

A robust selection system affording phage display of the DNA-binding helix-turn-helix protein Cro is presented. The aim of the work was to construct an experimental system allowing for the construction and isolation of Cro-derived protein with new DNA-binding properties. A derivative of the phage lambda Cro repressor, scCro8, in which the protein subunits had been covalently connected via a peptide linker was expressed in fusion with the gene 3 protein of Escherichia coli filamentous phage. The phage-displayed single-chain Cro was shown to retain the DNA binding properties of its wild-type Cro counterpart regarding DNA sequence specificity and binding affinity. A kinetic analysis revealed the rate constant of dissociation of the single-chain Cro-phage/DNA complex to be indistinguishable from that of the free single-chain Cro. Affinity selection using a biotinylated DNA with a target consensus operator sequence allowed for a 3000-fold enrichment of phages displaying single-chain Cro over control phages. The selection was based on entrapment of phage/DNA complexes formed in solution on streptavidin-coated paramagnetic beads. The expression system was subsequently used to isolate variant scCro8 proteins, mutated in their DNA-binding residues, that specifically recognized new, unnatural target DNA ligands.     

High-density mutagenesis by combined DNA shuffling and phage display to assign essential amino acid residues in protein-protein interactions: application to study structure-function of plasminogen activation inhibitor 1 (PAI-I)

The identification of specific amino acid residues involved in protein-protein interaction is fundamental to understanding structure-function relationships. Supported by mathematical calculations, we designed a high-density mutagenesis procedure for the generation of a mutant library of which a limited number of random clones would suffice to exactly localize amino acid residues essential for a particular protein-protein interaction. This goal was achieved experimentally by consecutive cycles of DNA shuffling, under error prone conditions, each followed by exposure of the target protein on the surface of phages to screen and select for correctly folded, functional mutants. To validate the procedure, human plasminogen activator inhibitor 1 (PAI-1) was chosen, because its 3D structure is known, many experimental tools are available and it may serve as a model protein for structure-function studies of serine proteinases and their inhibitors (serpins). After five cycles of DNA shuffling and selection for t-PA binding, analysis of 27 randomly picked clones revealed that PAI-1 mutants contained an average of 9.1 amino acid substitutions distributed over 114 different positions, which were preferentially located at the surface of the protein. This limited collection of mutant PAI-1 preparations contained multiple mutants defective in binding to three out of four tested anti-PAI-1 monoclonal antibodies. Alignment of the nucleotide sequence of defective clones permitted assignment of single dominant amino acid residues for binding to each monoclonal antibody. The importance of these residues was confirmed by testing the properties of single point mutants. From the position of these amino acid residues in the 3D structure of PAI-1 and the effects of the corresponding monoclonal antibodies on t-PA-PAI-1 interaction, conclusions can be drawn with respect to this serpin-serine proteinase interaction.     

The importance of a critical protonation state and the fate of the catalytic steps in class A beta-lactamases and penicillin-binding proteins

Beta-lactamases and penicillin-binding proteins are bacterial enzymes involved in antibiotic resistance to beta-lactam antibiotics and biosynthetic assembly of cell wall, respectively. Members of these large families of enzymes all experience acylation by their respective substrates at an active site serine as the first step in their catalytic activities. A Ser-X-X-Lys sequence motif is seen in all these proteins, and crystal structures demonstrate that the side-chain functions of the serine and lysine are in contact with one another. Three independent methods were used in this report to address the question of the protonation state of this important lysine (Lys-73) in the TEM-1 beta-lactamase from Escherichia coli. These techniques included perturbation of the pK(a) of Lys-73 by the study of the gamma-thialysine-73 variant and the attendant kinetic analyses, investigation of the protonation state by titration of specifically labeled proteins by nuclear magnetic resonance, and by computational treatment using the thermodynamic integration method. All three methods indicated that the pK(a) of Lys-73 of this enzyme is attenuated to 8.0-8.5. It is argued herein that the unique ground-state ion pair of Glu-166 and Lys-73 of class A beta-lactamases has actually raised the pK(a) of the active site lysine to 8.0-8.5 from that of the parental penicillin-binding protein. Whereas we cannot rule out that Glu-166 might activate the active site water, which in turn promotes Ser-70 for the acylation event, such as proposed earlier, we would like to propose as a plausible alternative for the acylation step the possibility that the ion pair would reconfigure to the protonated Glu-166 and unprotonated Lys-73. As such, unprotonated Lys-73 could promote serine for acylation, a process that should be shared among all active-site serine beta-lactamases and penicillin-binding proteins.     

Neutron Diffraction Studies of a Class A β-Lactamase Toho-1 E166A/R274N/R276N Triple Mutant

β-Lactam antibiotics have been used effectively over several decades against many types of bacterial infectious diseases. However, the most common cause of resistance to the β-lactam antibiotics is the production of β-lactamase enzymes that inactivate β-lactams by rapidly hydrolyzing the amide group of the β-lactam ring. Specifically, the class A extended-spectrum β-lactamases (ESBLs) and inhibitor-resistant enzymes arose that were capable of hydrolyzing penicillins and the expanded-spectrum cephalosporins and monobactams in resistant bacteria, which lead to treatment problems in many clinical settings. A more complete understanding of the mechanism of catalysis of these ESBL enzymes will impact current antibiotic drug discovery efforts. Here, we describe the neutron structure of the class A, CTX-M-type ESBL Toho-1 E166A/R274N/R276N triple mutant in its apo form, which is the first reported neutron structure of a β-lactamase enzyme. This neutron structure clearly reveals the active-site protonation states and hydrogen-bonding network of the apo Toho-1 ESBL prior to substrate binding and subsequent acylation. The protonation states of the active-site residues Ser70, Lys73, Ser130, and Lys234 in this neutron structure are consistent with the prediction of a proton transfer pathway from Lys73 to Ser130 that is likely dependent on the conformation of Lys73, which has been hypothesized to be coupled to the protonation state of Glu166 during the acylation reaction. Thus, this neutron structure is in agreement with a proposed mechanism for acylation that identifies Glu166 as the general base for catalysis.     

Kinetic and mechanistic studies of penicillin-binding protein 2x from Streptococcus pneumoniae.

High molecular weight penicillin-binding proteins (PBPs) are bifunctional enzymes that build bacterial cell walls from the glycopeptide lipid II [GlcNAc-MurNAc(L-Ala-gamma-D-Glu-L-Lys-D-Ala-D-Ala)-pyrophosphate-undecaprenol] by a process involving disaccharide polymerization and peptide cross-linking. The latter reaction involves acyl-transfer chemistry in which the penultimate (D)Ala first acylates the active-site serine, with release of the terminal (D)Ala, and is then transferred to the epsilon-amine of a Lys on a neighboring pentapeptide chain. These enzymes also catalyze hydrolysis of specific thioester substrates and acylation by beta-lactam antibiotics. In this paper, we explore these latter two reactions and report mechanistic experiments on the reaction of Streptococcus pneumoniae PBP 2x with N-benzoyl-(D)Ala-thioacetic acid [Bz-(D)Ala-(S)Gly] and penicillin G. For these experiments, we used PBP 2x, a soluble form of PBP 2x in which the transmembrane domain was deleted. The following results are significant: (1) pH dependencies for acylation of PBP 2x by penicillin G and Bz-(D)Ala-(S)Gly are identical, suggesting that the same ionizable residues are involved in both reactions and that these residues play the same catalytic role in the two processes. On the basis of these results, we propose a mechanistic model that is also consistent with recently published structural data [Gordon, E., et al. (2000) J. Mol. Biol. 299, 477-485]. (2) Pre-steady-state experiments for the PBP 2x-catalyzed hydrolysis of Bz-(D)Ala-(S)Gly at pH 6.5 indicate that k(c) is principally rate-limited by acylation with some contribution from deacylation. The contribution of these steps to rate limitation is pH-dependent, with acylation entirely rate-limiting at pH values less than 5.5 and deacylation principally rate-limiting above pH 8.5. (3) Results of solvent isotope effect and proton inventory experiments for acylation suggest a complex process that is at least partially rate-limited by chemistry with some involvement of changes in solvation and/or enzyme conformation. (4) Analysis of activation parameters suggests that during the acylation of PBP 2x by penicillin G the inherent chemical stability of penicillin's amide bond, as manifested in the enthalpy of activation, is offset by a favorable entropy term that reflects penicillin's rotationally constrained bicyclic system, which presumably allows a less energetically demanding entry into the transition state for acylation.     

Engineering E. coli Alkaline Phosphatase Yields Changes of Catalytic Activity,Thermal Stability and Phosphate Inhibition

To investigate the function of aspartic acid residue 101 and arginine residue 166 in the active site of Escherichia coli alkaline phosphatase (EAP), two single mutants D101S (Asp 101 &#77 Ser) and R166K (Arg 166 &#77 Lys) and a double mutant D101S/R166K of EAP were generated through site-directed mutagenesis based on over-lap PCR method. Their enzymatic kinetic properties, thermal stabilities and possible reaction mechanism were explored. In the presence of inorganic phosphate acceptor, 1 M diethanolamine buffer, the k cat for D101S mutant enzyme increased 10-fold compared to that of wild-type EAP. The mutant R166K has a 2-fold decrease of k cat relative to the wild-type EAP, but the double mutant D101S/R166K was in the middle of them, indicative of an additive effect of these two mutations. On the other hand, the catalytic efficiencies of mutant enzymes are all reduced because of a substantial increase of K m values. All three mutants were more resistant to phosphate inhibitor than the wild-type enzyme. The analysis of the kinetic data suggests that (1) the D101S mutant enzyme obtains a higher catalytic activity by allowing a faster release of the product; (2) the R166K mutant enzyme can reduce the binding of the substrate and phosphate competitive inhibitor; (3) the double mutant enzyme has characteristics of both quicker catalytic turnover number and decreased affinity for competitive inhibitor. Additionally, pre-steady-state kinetics of D101S and D101S/R166K mutants revealed a transient burst followed by a linear steady state phase, obviously different from that of wild-type EAP, suggesting that the rate-limiting step has partially change from the release of phosphate from non-covalent E-Pi complex to the hydrolysis of covalent E-Pi complex for these two mutants.     

Significance of Asn-77 and Trp-78 in the catalytic function of undecaprenyl diphosphate synthase of Micrococcus luteus B-P 26

The primary structures of cis-prenyltransferases are completely different from those of trans-prenyltransferases. To obtain information about amino acid residues relating to catalytic function, random mutation of the undecaprenyl diphosphate synthase gene of Micrococcus luteus B-P 26 was carried out to construct a mutated gene library using an error-prone polymerase chain reaction. From the library, the mutants showing poor enzymatic activity were selected by the colony autoradiography method. Among 31 negative clones selected from 3,000 mutants, two clones were found to contain only one amino acid substitution at either Asn-77 or Trp-78. To determine the functional roles of these interesting residues, we prepared six mutated enzymes with substitutions at residues Asn-77 or Trp-78 by site-directed mutagenesis. Substitution of Asn-77 with Ala, Asp, or Gln resulted in a dramatic decrease in catalytic activity, but the K(m) values for both allylic and homoallylic substrates of these mutant enzymes were comparable to those of the wild-type. On the other hand, three Trp-78 mutants, W78I, W78R, and W78D, showed 5-20-fold increased K(m) values for farnesyl diphosphate but not for Z-geranylgeranyl diphosphate. However, these mutants showed moderate levels of enzymatic activity and comparable K(m) values for isopentenyl diphosphate to that of the wild-type. These results suggest that the Asn-Trp motif is involved in the binding of farnesyl diphosphate and enzymatic catalysis.     

Polymyxin B neutralizes bacteria-released endotoxin and improves the quality of boar sperm during liquid storage and cryopreservation

Gram negative bacteria are the predominant type detected in boar semen. Since these bacteria release lipopolysaccharide (LPS), and because polymyxin B (PMB) neutralizes LPS activity, the objective was to improve techniques for long-term storage of boar sperm by testing the beneficial effects of PMB. In the present study, LPS bound directly to the head region of sperm, decreased sperm motility, and induced sperm apoptosis. The addition of 100 μg/mL PMB suppressed LPS binding activity and blocked the negative effects of LPS on sperm quality. Additionally, when PMB treatment was combined with penicillin G (PenG), sperm motility was increased after 10 d of liquid storage or in frozen-thawed sperm (P<0.05). When the PMB- and PenG-treated sperm was used for artificial insemination, the conception rate was increased relative to that of artificial insemination with sperm treated by PenG alone in both liquid (62 vs. 81%) and cryopreserved forms (50 vs. 80%, P < 0.05). We concluded that PMB suppressed LPS-induced low sperm motility and apoptosis via the reduction of LPS binding to Toll-like receptors (TLRs). Thus, in order to enhance sperm quality for artificial insemination, a combined treatment with PMB and PenG immediately after ejaculation seemed appropriate to maintain sperm motility and function during both liquid storage and cryopreservation.     

Direct selection of EGF mutants displayed on filamentous phage using cells overexpressing EGF receptor.

Understanding receptor-ligand interactions, and the signal transduction pathways they activate, is of great interest for the discovery of novel antagonists and agonists. In this report we describe a rapid and efficient procedure to evaluate the importance of several different epidermal growth factor (EGF) residues for the binding and activation of its receptor (EGFR). We constructed an EGF mutant library randomized at positions 13, 15 and 16 and expressed them on filamentous phages. Phage display is a powerful system, allowing rapid isolation of binding mutants. Since many of the most pharmacologically interesting receptors cannot be produced in a soluble form, we developed a technique to rapidly select receptor-binding molecules directly on cells. A luciferase assay, simple to perform, was then used to test their biological transduction activity and to rapidly detect mutants of interest. Analysis of the resulting sequences revealed that the wild-type amino acids at positions 13, 15 and 16 are optimized for binding and activity. EGF mutants with agonist properties were also isolated and tolerated substitutions were identified.     

Site-directed mutagenesis of beta-lactamase I. Single and double mutants of Glu-166 and Lys-73.

Two single mutants and the corresponding double mutant of beta-lactamase I from Bacillus cereus 569/H were constructed and their kinetics investigated. The mutants have Lys-73 replaced by arginine (K73R), or Glu-166 replaced by aspartic acid (E166D), or both (K73R + E166D). All four rate constants in the acyl-enzyme mechanism were determined for the E166D mutant by the methods described by Christensen, Martin & Waley [(1990) Biochem. J. 266, 853-861]. Both the rate constants for acylation and deacylation for the hydrolysis of benzylpenicillin were decreased about 2000-fold in this mutant. In the K73R mutant, and in the double mutant, the rate constants for acylation were decreased about 100-fold and 10,000-fold respectively. All three mutants also had lowered values for the rate constants for the formation and dissociation of the non-covalent enzyme-substrate complex. The specificities of the mutants did not differ greatly from those of wild-type beta-lactamase, but the hydrolysis of cephalosporin C by the K73R mutant gave 'burst' kinetics.     

Binding of phage displayed Bacillus subtilis lipase A to a phosphonate suicide inhibitor

Phage display can be used as a protein engineering tool to select proteins with desirable binding properties from a library of randomly constructed mutants. Here, we describe the development of this method for the directed evolution of Bacillus subtilis lipase A, an enzyme that has marked properties for the preparation of pharmaceutically relevant chiral compounds. The lipase gene was cloned upstream of the phage g3p encoding sequence and downstream of a modified g3p signal sequence. Consequently, the enzyme was displayed at the surface of bacteriophage fd as a fusion to its minor coat protein g3p. The phage-bound lipase was correctly folded and fully enzymatically active as determined from the hydrolysis of p-nitrophenylcaprylate with K(m)-values of 0.38 and 0.33 mM for the phage displayed and soluble lipase, respectively. Both soluble lipase and lipase expressed on bacteriophages reacted covalently with a phosphonate suicide inhibitor. The phage does not hamper lipase binding, since both soluble and phage-bound lipase have a similar half-life of inactivation of approximately 5 min. Therefore, we conclude that the Bacillus lipase can be functionally expressed on bacteriophages as a fusion to the phage coat protein g3p. The specific interaction with the suicide inhibitor offers a fast and reproducible method for the future selection of mutant enzymes with an enantioselectivity towards new substrates.     

Improved autoprocessing efficiency of mutant subtilisins E with altered specificity by engineering of the pro-region.

Modification of substrate specificity of an autoprocessing enzyme is accompanied by a risk of significant failure of self-cleavage of the pro-region essential for activation. Therefore, to enhance processing, we engineered the pro-region of mutant subtilisins E of Bacillus subtilis with altered substrate specificity. A high-activity mutant subtilisin E with Ile31Leu replacement (I31L) as well as the wild-type enzyme show poor recognition of acid residues as the P1 substrate. To increase the P1 substrate preference for acid residues, Glu156Gln and Gly166Lys/Arg substitutions were introduced into the I31L gene based upon a report on subtilisin BPN' [Wells et al. (1987) Proc. Natl. Acad. Sci. USA 84, 1219-1223]. The apparent P1 specificity of four mutants (E156Q/G166K, E156Q/G166R, G166K, and G166R) was extended to acid residues, but the halo-forming activity of Escherichia coli expressing the mutant genes on skim milk-containing plates was significantly decreased due to the lower autoprocessing efficiency. A marked increase in active enzyme production occurred when Tyr(-1) in the pro-region of these mutants was then replaced by Asp or Glu. Five mutants with Glu(-2)Ala/Val/Gly or Tyr(-1)Cys/Ser substitution showing enhanced halo-forming activity were further isolated by PCR random mutagenesis in the pro-region of the E156Q/G166K mutant. These results indicated that introduction of an optimum arrangement at the cleavage site in the pro-region is an effective method for obtaining a higher yield of active enzymes.