Optimizing the search algorithm for protein engineering by directed evolution

An in silico protein model based on the Kauffman NK-landscape,where N is the number of variable positions in a protein andK is the degree of coupling between variable positions, wasused to compare alternative search strategies for directed evolution.A simple genetic algorithm (GA) was used to model the performanceof a standard DNA shuffling protocol. The search effectivenessof the GA was compared to that of a statistical approach calledthe protein sequence activity relationship (ProSAR) algorithm,which consists of two steps: model building and library design.A number of parameters were investigated and found to be importantfor the comparison, including the value of K, the screeningsize, the system noise and the number of replicates. The statisticalmodel was found to accurately predict the measured activitiesfor small values of the coupling between amino acids, K 1.The ProSAR strategy was found to perform well for small to moderatevalues of coupling, 0 K 3, and was found to be robust to noise.Some implications for protein engineering are discussed. Received January 2, 2003; revised May 13, 2003; accepted June 19, 2003.     

Structure-oriented rational design of chymotrypsin inhibitor models

Three peptides modelling a highly potent, 35-residue chymotrypsininhibitor (Schistocerca gregaria chymotrypsin inhibitor) weredesigned and synthesized by convergent peptide synthesis. Foreach model peptide, the inhibitory constant (K_i) on chymotrypsinand the solution structure were determined. In addition, moleculardynamics calculations were performed for all of them. Two modelscontaining approximately half of the parent inhibitor (17 of35 residues) were designed and subsequently found to have nosubstantial inhibitory activity (K_i values in the mM range).The third model composed of 24 amino acid residues proved tobe an effective (K_i 10^–7) inhibitor of bovine chymotrypsin.Both the solution structure properties determined by NMR spectroscopyand the dynamic behaviour of the latter model system are comparableto the native inhibitor. In contrast, the structure and dynamicsof the first two related model peptides show characteristicdifferences. We suggest that the conformation and flexibilityof the modelled protease inhibitor are crucial for its biologicalefficiency. Moreover, the structural and dynamic features ofthe binding loop (28–33) and those of the rest of themolecule appear to be interdependent. Most importantly, thesestructural characteristics can be rationally modified, at leastpartially, by peptide design. Received March 7, 2003; revised August 25, 2003; accepted August 26, 2003.     

Stabilization of a chitinase from Serratia marcescens by Gly->Ala and Xxx->Pro mutations

This paper describes attempts to increase the kinetic stabilityof chitinase B from Serratia marcescens (ChiB) by the introductionof semi-automatically designed rigidifying mutations of theGlyAla and XxxPro type. Of 15 single mutants, several displayedsignificant increases in thermal stability, whereas most mutantsshowed minor effects. All mutations with non-marginal effectson stability clustered in a limited, surface-exposed regionof the enzyme, indicating that this region is involved in apartial unfolding process that triggers irreversible thermalinactivation (aggregation). A double mutant containing two stabilizingmutations in this region (G188A, A234P) displayed a 10-foldincrease in half-life at 57°C and a 4.2°C increase inapparent T_m. These results show that entropic stabilizationworks well for ChiB and they pinpoint a region whose unfoldingmay be crucial for the kinetic stability of this enzyme. Received June 18, 2003; revised September 3, 2003; accepted September 12, 2003.     

Synthesis of a novel histidine analogue and its efficient incorporation into a protein in vivo

Proteins containing unnatural amino acids have immense potentialin biotechnology and medicine. We prepared several histidineanalogues including a novel histidine analogue, ß-(1,2,3-triazol-4-yl)-DL-alanine.These histidine analogues were assayed for translational activityin histidine-auxotrophic Escherichia coli strain UTH780. Weobserved that several histidine analogues, including our novelhistidine analogue, were efficiently incorporated into the proteinin vivo; however, other analogues were rejected. These resultssuggest that the hydrogen atom at a specific position seriouslyaffects incorporation. Received April 10, 2003; revised June 20, 2003; accepted July 22, 2003.     

Change in the crystal packing of soybean {beta}-amylase mutants substituted at a few surface amino acid residues

In spite of the high similarity of amino acid sequence and three-dimensionalstructure between soybean ß-amylase (SBA) and sweetpotato ß-amylase (SPB), their quaternary structureis quite different, being a monomer in SBA and a tetramer inSPB. Because most of the differences in amino acid sequencesare found in the surface region, we tested the tetramerizationof SBA by examining mutations of residues located at the surface.We designed the SBA tetramer using the SPB tetramer structureas a model and calculating the change of accessible surfacearea (ASA) for each residue in order to select sites for themutation. Two different mutant genes encoding SB3 (D374Y/L481R/P487D)and SB4 (K462S added to SB3), were constructed for expressionin Escherichia coli and the recombinant proteins were purified.They existed as a monomer in solution, but gave completely differentcrystals from the native SBA. The asymmetric unit of the mutantscontains four molecules, while that of native SBA contains one.The interactions of the created interfaces revealed that therewere more intermolecular interactions in the SB3 than in theSB4 tetramer. The substituted charged residues on the surfaceare involved in interactions with adjacent molecules in a differentway, forming a new crystal packing pattern. Received June 9, 2003; revised September 11, 2003; accepted September 16, 2003.     

Vicinal disulfide turns

The formation of a disulfide bond between adjacent cysteineresidues is accompanied by the formation of a tight turn ofthe protein backbone. In nearly 90% of the structures analyzeda type VIII turn was found. The peptide bond between the twocysteines is in a distorted trans conformation, the omega torsionangle ranges from 159 to –133°, with an average valueof 171°. The constrained nature of the vicinal disulfideturn and the pronounced difference observed between the oxidizedand reduced states, suggests that vicinal disulfides may beemployed as a ‘redox-activated’ conformational switch. Received December 16, 2002; revised June 30, 2003; accepted July 30, 2003.     

Immobilized oxidoreductase as an additive for refolding inclusion bodies: application to antibody fragments

Three foldases—the apical domain of GroEL (mini-chaperone)and two oxidoreductases (DsbA and DsbC) from Escherichia coli—werestudied in refolding a protein with immunoglobulin fold (immunoglobulin-foldedprotein) that had been produced as inclusion bodies in E.coli.The foldases promoted the refolding of single-chain antibodyfragments from denaturant-solubilized and reduced inclusionbodies in vitro, and also effectively functioned as alternativesfor labilizing agent and oxidizing reagent in the stepwise dialysissystem. Immobilization of the oxidoreductases enhanced refoldingand recovery of functional single-chain antibody in the dialysissystem, suggesting that immobilized oxidoreductases can be usedas an effective additive for refolding immunoglobulin-foldedproteins in vitro. Received April 7, 2003; revised June 5, 2003; accepted June 6, 2003.     

VL position 34 is a key determinant for the engineering of stable antibodies with fast dissociation rates

Predictive engineering of antibodies exhibiting fast kineticproperties could provide reagents for biotechnological applicationssuch as continuous monitoring of compounds or affinity chromatography.Based on covariance analysis of murine germline antibody variabledomains, we selected position L34 (Kabat numbering) for mutationalstudies. This position is located at the VL/VH interface, atthe base of the paratope but with limited antigen contacts,thus making it an attractive position for mild alterations ofantigen binding properties. We introduced a serine at positionL34 in two different antibodies: Fab (fragment antigen binding)57P (Asn34Ser) and scFv (single chain fragment variable) 1F4(Gln34Ser), that recognize peptides derived from the coat proteinof tobacco mosaic virus and the oncoprotein E6, respectively.Both mutated antibodies exhibited similar properties: (i) expressionlevels of active fragments in Escherichia coli were markedlyimproved; (ii) thermostability was enhanced; and (iii) dissociationrate parameters (k_off) were increased by 2- and at least 57-foldfor scFv1F4 and Fab57P, respectively, while their associationrate parameters (k_on) remained unchanged. The L34 Ala and Thrmutants of both antibody fragments did not possess these properties.This first demontration of similar effects observed in two antibodieswith different specificities may open the way to the predictivedesign of molecules with enhanced stability and fast dissociationrates. Received October 3, 2002; revised March 7, 2003; accepted April 4, 2003.     

A new computational approach for real protein folding prediction

An effective and fast minimization approach is proposed forthe prediction of protein folding, in which the ‘relativeentropy’ is used as a minimization function and the off-lattice model is used. In this approach, we only use the informationof distances between the consecutive C atoms along the peptidechain and a generalized form of the contact potential for 20types of amino acids. Tests of the algorithm are performed onthe real proteins. The root mean square deviations of the structuresof eight folded target proteins versus the native structuresare in a reasonable range. In principle, this method is an improvementon the energy minimization approach. Received June 25, 2002; revised June 6, 2003; accepted July 29, 2003.     

Involvement of surface cysteines in activity and multimer formation of thimet oligopeptidase

Thimet oligopeptidase is a metalloenzyme involved in regulatingneuropeptide processing. Three cysteine residues (246, 248,253) are known to be involved in thiol activation of the enzyme.In contrast to the wild-type enzyme, the triple mutant (C246S/C248S/C253S)displays increased activity in the absence of dithiothreitol.Dimers, purportedly formed through cysteines 246, 248 and 253,have been thought to be inactive. However, analysis of the triplemutant by native gel electrophoresis reveals the existence ofdimers and multimers, implying that oligomer formation is mediatedby other cysteines, probably on the surface, and that some ofthese forms are enzymatically active. Isolation and characterizationof iodoacetate-modified monomers and dimers of the triple mutantrevealed that, indeed, certain dimeric forms of the enzyme arestill fully active, whereas others show reduced activity. Cysteineresidues potentially involved in dimerization were identifiedby modeling of thimet oliogopeptidase to its homolog, neurolysin.Five mutants were constructed; all contained the triple mutationC246S/C248S/C253S and additional substitutions. Substitutionsat C46 or C682 and C687 prevented multimer formation and inhibiteddimer formation. The C46S mutant had enzymatic activity comparableto the parent triple mutant, whereas that of C682S/C687S wasreduced. Thus, the location of intermolecular disulfide bonds,rather than their existence per se, is relevant to activity.Dimerization close to the N-terminus is detrimental to activity,whereas dimerization near the C-terminus has little effect.Altering disulfide bond formation is a potential regulatoryfactor in the cell owing to the varying oxidation states insubcellular compartments and the different compartmental locationsand functions of the enzyme. Received March 1, 2003; revised June 17, 2003; accepted June 23, 2003.     

Improving tolerance of Candida antarctica lipase B towards irreversible thermal inactivation through directed evolution

To expand the functionality of lipase B from Candida antarctica(CALB) we have used directed evolution to create CALB mutantswith improved resistance towards irreversible thermal inactivation.Two mutants, 23G5 and 195F1, were generated with over a 20-foldincrease in half-life at 70°C compared with the wild-typeCALB (WT-CALB). The increase in half-life was attributed toa lower propensity of the mutants to aggregate in the unfoldedstate and to an improved refolding. The first generation mutant,23G5, obtained by error-prone PCR, had two amino acid mutations,V210I and A281E. The second generation mutant, 195F1, derivedfrom 23G5 by error-prone PCR, had one additional mutation, V221D.Amino acid substitutions at positions 221 and 281 were determinedto be critical for lipase stability, while the residue at position210 had only a marginal effect. The catalytic efficiency ofthe mutants with p-nitrophenyl butyrate and 6,8-difluoro-4-methylumbelliferyloctanoate was also found to be superior to that of WT-CALB. Received May 8, 2003; revised June 9, 2003; accepted June 23, 2003.     

Improvement of binding of Puumala virus neutralization site resembling peptide with a second-generation phage library

We have previously selected a peptide insert FPCDRLSGYWERGIPSPCVRrecognizing the Puumala virus (PUUV) G2-glycoprotein-specificneutralizing monoclonal antibody (MAb) 1C9 with K_d of 2.85x10^–8from a random peptide library X₂CX₁₄CX₂ expressed on the pIIIprotein of the filamentous phage fd-tet. We have now createda second-generation phage-displayed peptide library in whicheach amino acid of the peptide was mutated randomly to anotherwith a certain probability. Peptides were selected for higheraffinity for MAb 1C9 and for a common binding motif for MAb4G2 having an overlapping epitope with MAb 1C9 in G2 glycoprotein.The resulting peptides were synthesized as spots on cellulosemembrane. Amino acid changes which improved the reactivity ofthe peptides to MAb 1C9 were combined in the peptide ATCDKLFGYYERGIPLPCALwith K_d of 1.49x10^–9 in biosensor measurements. Our resultsshow that the binding properties of peptides, the affinity andthe specificity can be improved and the binding specificitydetermining amino acids and structural factors can be analyzedby combining binding assays with synthetic peptides on membranewith the use of second-generation phage display libraries. Received October 18, 2002; revised April 16, 2003; accepted May 26, 2003.     

Molecular docking of substrates and inhibitors in the catalytic site of CYP6B1, an insect cytochrome P450 monooxygenase

Furanocoumarins represent plant toxins that are used in thetreatment of a variety of skin diseases and are metabolizedby cytochrome P450 monooxygenases (P450s) existing in insectssuch as Papilio polyxenes (the black swallowtail). To elucidatethe active site in the CYP6B1 protein that is the principalP450 existing in this species, we have constructed a homologymodel of it based on sequence and structure alignments withthe bacterial CYP102 protein whose crystal structure has beendefined and with the insect CYP6B4 protein that also metabolizesfuranocoumarins. In the derived CYP6B1 model, Phe116 and His117in SRS1, Phe371 in SRS5 and Phe484 in SRS6 contribute to theformation of a resonant network that stabilizes the P450’scatalytic site and allows for interactions with its furanocoumarinsubstrates. The first two of these residues are absolutely conservedin all members of the insect CYP6B subfamily and the last twoare variable in different members of the CYP6B subfamily. Acombination of theoretical and experimental docking analysesof two substrates (xanthotoxin and bergapten) and two inhibitors(coumarin and pilocarpine) of this P450 provide significantinformation on the positioning of furanocoumarins within thiscatalytic pocket. Molecular replacement models based on theresults of variations at two of these critical amino acids providesupport for our furanocoumarin-docked model and begin to rationalizethe altered substrate reactivities observed in experimentalanalyses. Received December 4, 2002; revised June 8, 2003; accepted June 23, 2003.     

The structural roles of conserved Pro196, Pro197 and His199 in the mechanism of thymidylate synthase

We generated replacement sets for three highly conserved residues,Pro196, Pro197 and His199, that flank the catalytic nucleophile,Cys198. Pro196 and Pro197 have restricted mobility that couldbe important for the structural transitions known to be essentialfor activity. To test this hypothesis we obtained and characterized13 amino acid substitutions for Pro196, 14 for Pro197 and 14for His199. All of the Pro196 and Pro197 variants, except P197R,and four of the His199 variants complemented TS-deficient Escherichiacoli cells, indicating they had at least 1% of wild-type activity.For all His199 mutations, k_cat/K_m for substrate and cofactordecreased more than 40-fold, suggesting that the conserved hydrogenbond network co-ordinated by His199 is important for catalysis.Pro196 can be substituted with small hydrophilic residues withlittle loss in k_cat, but 15- to 23-fold increases in K_m^dUMP.Small hydrophobic substitutions for Pro197 were most active,and the most conservative mutant, P197A, had only a 5-fold lowerk_cat/K_m^dUMP than wild-type TS. Several Pro196 and Pro197 variantswere temperature sensitive. The small effects of Pro196 or Pro197mutations on enzyme kinetics suggest that the conformationalrestrictions encoded by the Pro–Pro sequence are largelymaintained when either member of the pair is mutated. Received February 24, 2003; revised June 19, 2003; accepted June 20, 2003.     

Effect of mutations involving charged residues on the stability of staphylococcal nuclease: a continuum electrostatics study

A continuum electrostatics model is used to calculate the relativestabilities of 117 mutants of staphylococcal nuclease (SNase)involving the mutation of a charged residue to an unchargedresidue. The calculations are based on the crystallographicstructure of the wild-type protein and attempt to take implicitlyinto account the effect of the mutations in the denatured stateby assuming a linear relationship between the free energy changescaused by the mutation in the native and denatured states. Agood correlation (linear correlation coefficient of 0.8) isfound with published experimental relative stabilities of thesemutants. The results suggest that in the case of SNase (i) chargedresidues contribute to the stability of the native state mainlythrough electrostatic interactions, and (ii) native-like electrostaticinteractions may persist in the denatured state. The continuumelectrostatics method is only moderately sensitive to modelparameters and leads to quasi-predictive results for the relativemutant stabilities (error of 2–3 kJ mol^–1 or ofthe order of k_BT), except for mutants in which a charged residueis mutated to glycine. Received March 24, 2003; revised August 11, 2003; accepted September 12, 2003.     

Simple consensus procedures are effective and sufficient in secondary structure prediction

We have analyzed the performance of majority voting on minimalcombination sets of three state-of-the-art secondary structureprediction methods in order to obtain a consensus prediction.Using three large benchmark sets from the EVA server, our resultsshow a significant improvement in the average Q₃ predictionaccuracy of up to 1.5 percentage points by consensus formation.The application of an additional trivial filtering procedurefor predicted secondary structure elements that are too short,does not significantly affect the prediction accuracy. Our analysisalso provides valuable insight into the similarity of the resultsof the prediction methods that we combine as well as the higherconfidence in consistently predicted secondary structure. Received March 7, 2003; revised May 24, 2003; accepted June 6, 2003.     

Starch-binding domain shuffling in Aspergillus niger glucoamylase

Aspergillus niger glucoamylase (GA) consists mainly of two forms,GAI [from the N-terminus, catalytic domain + linker + starch-bindingdomain (SBD)] and GAII (catalytic domain + linker). These domainswere shuffled to make RGAI (SBD + linker + catalytic domain),RGAIL (SBD + catalytic domain) and RGAII (linker + catalyticdomain), with domains defined by function rather than by tertiarystructure. In addition, Paenibacillus macerans cyclomaltodextringlucanotransferase SBD replaced the closely related A.nigerGA SBD to give GAE. Soluble starch hydrolysis rates decreasedas RGAII GAII GAI > RGAIL RGAI GAE. Insoluble starchhydrolysis rates were GAI > RGAIL > RGAI >> GAE RGAII > GAII, while insoluble starch-binding capacitieswere GAI > RGAI > RGAIL > RGAII > GAII > GAE.These results indicate that: (i) moving the SBD to the N-terminusor replacing the native SBD somewhat affects soluble starchhydrolysis; (ii) SBD location significantly affects insolublestarch binding and hydrolysis; (iii) insoluble starch hydrolysisis imperfectly correlated with its binding by the SBD; and (iv)placing the P.macerans cyclomaltodextrin glucanotransferaseSBD at the end of a linker, instead of closely associated withthe rest of the enzyme, severely reduces its ability to bindand hydrolyze insoluble starch. Received June 20, 2002; revised May 23, 2003; accepted June 6, 2003.     

Increased thermal stability against irreversible inactivation of 3-isopropylmalate dehydrogenase induced by decreased van der Waals volume at the subunit interface

We have investigated factors affecting stability at the subunit–subunitinterface of the dimeric enzyme 3-isopropylmalate dehydrogenase(IPMDH) from Bacillus subtilis. Site-directed mutagenesis wasused to replace methionine 256, a key residue in the subunitinteraction, with other amino acids. Thermal stability againstirreversible inactivation of the mutated enzymes was examinedby analyzing the residual activity after heat treatment. Themutations M256V and M256A increased thermostability by 2.0 and6.0°C, respectively, whereas the mutations M256L and M256Ihad no effect. Thermostability of the M256F mutated enzyme was4.0°C lower than that of the wild-type enzyme. To our surprise,increasing the hydrophobicity of residue 256 within the hydrophobiccore of the enzyme resulted in a lower thermal stability. Themutated enzymes showed an inverse correlation between thermostabilityand the volume of the side chain at position 256. Based on theX-ray crystallographic structure of Escherichia coli IPMDH,the environment around M256 in the B.subtilis homolog is predictedto be sterically crowded. These results suggest that Met256prevents favorable packing. Introduction of a smaller aminoacid at position 256 improves the packing and stabilizes thedimeric structure of IPMDH. The van der Waals volume of theamino acid residue at the hydrophobic subunit interface is animportant factor for maintaining the stability of the subunit–subunitinterface and is not always optimized in the mesophilic IPMDHenzyme. Received September 3, 2002; revised June 13, 2003; accepted June 20, 2003.     

A long insertion reverts the functional effect of a substitution in acetylcholinesterase

Proteins are thought to undertake single substitutions, deletionsand insertions to explore the fitness landscape. Nevertheless,the ways in which these different kind of mutations act togetherto alter a protein phenotype remain poorly described. We introducedincrementally the single substitution W290A and a 26 amino acidlong insertion at the 297 location in the Nippostrongylus brasiliensisacetylcholinesterase B sequence and analysed in vitro the inducedchanges in the hydrolysis rate of three hemi-substrates: pirimicarb,paraoxon methyl and omethoate. The substitution decreased thehydrolysis rate of the three hemi-substrates. The insertiondid not influence this kinetic alteration induced by the substitutionfor the former hemi-substrate, but reverted it for the two others.These results show that two different kinds of mutations caninteract together to influence the direction of a protein’sadaptative walk on the fitness landscape. Received January 28, 2003; revised April 25, 2003; accepted June 6, 2003.