Escherichia coli aspartate carbamoyltransferase: the probing of crystal structure analysis via site-specific mutagenesis

Crystal structures are known for aspartate carbamoyltransferase(ATCase) in the T and R states, with and without the allostericactivator adenosine triphosphate (ATP) or inhibitor cytidinetriphosphate (CTP). Visual inspection of X-ray crystal structuresdoes not provide all of the information necessary for the determinationof structure-function relationships in protein molecules. Thisproblem is compounded because the crystalline states of themolecule may introduce effects due to crystal packing, restrictedflexibility and less than optimum enzymatic conditions. Therefore,alternative techniques are required to test mechanisms conjecturedfrom three-dimensional crystal structures of proteins. The techniqueof site-specific mutagenesis allows the researcher to test structure- function models based on threedimensional structures and toobtain further insight into characteristics of the enzyme. Site-specificmutagenesis has been used to probe residues believed to be criticalin the structure and function of ATCase. Selection of residuesto be mutated has depended extensively on three-dimensionalcrystal structures of the enzyme. To date, 48 site-specificmutations at 37 different amino acid sites have been published.Although a total of 118 mutants at 58 different sites has beencommunicated to our laboratory, only published mutants willbe considered in this review. In this paper, we compile forthe first time, review, and analyze the site-specific mutantsof ATCase. Site-specific mutagenesis of proteins has becomea powerful technique in modern-day molecular biology, especiallyin studying a molecule as large as aspartate carbamoyltransferase.In this review, the role of site-specific mutagenesis of ATCaseis discussed and improvements in the analysis are suggested.     

Changes in the specificity of antibodies by site-specific mutagenesis followed by random mutagenesis

The specificity for 11-deoxycortisol (11-DOC) of a monoclonalantibody (mAb), designated SCET, was changed to specificityfor cortisol (CS) by site-specific mutagenesis followed by randommutagenesis. The Fab form of SCET was expressed on the surfaceof a phage. During the first step, mutations were introducedat 14 amino acid positions in three complementarity-determiningregions (CDRs) of the V_H domain that seemed likely to form thesteroid-binding pocket. A clone, DcC16, was isolated from theresultant library with multiple mutations and this clone wasshown to have CS-binding activity but also to retain high 11-DOC-bindingactivity. During the second step, mutations were introducedrandomly into the entire V_H-coding region of the DcC16 cloneby an error-prone polymerase chain reaction, and CS-specificmutant antibodies were selected in the presence of 11-DOC asa competitor. Three representative clones were analyzed withthe BIAcore instrument, and each revealed a large increase inthe binding constant for CS and a decrease in that for 11-DOC.Structural models, constructed by computer simulation, indicatedthe probable molecular basis for these changes in specificity.     

Analysis of proteinprotein interactions by mutagenesis: direct versus indirect effects

Site-directed mutagenesis, including double-mutant cycles, isused routinely for studying protein–protein interactions.We now present a case analysis of chymotrypsin inhibitor 2 (CI2)and subtilisin BPN' using (i) a residue in CI2 that is knownto interact directly with subtilisin (Tyr42) and (ii) two CI2residues that do not have direct contacts with subtilisin (Arg46and Arg48). We find that there are similar changes in bindingenergy on mutation of these two sets of residues. It can thusbe difficult to interpret mutagenesis data in the absence ofstructural information.     

Isolated transmembrane helices arranged across a membrane: computational studies

A computational procedure for predicting the arrangement ofan isolated helical fragment across a membrane was developed.The procedure places the transmembrane helical segment intoa model triple-phase system `water–octanol–water';pulls the segment through the membrane, varying its `global'position as a rigid body; optimizes the intrahelical and solvationenergies in each global position by `local' coordinates (dihedralangles of side chains); and selects the lowest energy globalposition for the segment. The procedure was applied to 45 transmembranehelices from the photosynthetic reaction center from Rhodopseudomonasviridis, cytochrome c oxidase from Paracoccus denitrificansand bacteriorhodopsin. In two thirds of the helical fragmentsconsidered, the procedure has predicted the vertical shiftsof the fragments across the membrane with an accuracy of –0.15± 3.12 residues compared with the experimental data.The accuracy for the remaining 15 fragments was 2.17 ±3.07 residues, which is about half of a helix turn. The procedurepredicts the actual membrane boundaries of transmembrane helicalfragments with greater accuracy than existing statistical methods.At the same time, the procedure overestimates the tilt valuesfor the helical fragments.     

Mutational analysis of DNase I-DNA interactions: design, expression and characterization of a DNase I loop insertion mutant with altered sequence selectivity

A mutant of bovine pancreatic DNase I containing two additionalresidues in a loop next to C173 has been expressed in Escherichiacoli, purified and characterized biochemically. Modelling studiessuggest that the inserted arginine and glutamate side chainsof the modified loop sequence C173-R-E-G-T-V176 could contactthe bases 3' to the cleaved bond in the major groove of a boundDNA, and that up to 10 bp could interact with the enzyme andpotentially influence its cutting rate. The loop insertion mutanthas an 800-fold lower specific activity than wild-type and showsoverall cleavage characteristics similar to bovine pancreaticDNase I. Compared with the wild-type enzyme, the mutant showsa strongly enhanced preference for cutting the inverted repeat:5'-GACTT A AAGTC-3' CTGAA T TTCAG or close variants thereof.Unexpectedly for a minor groove binding protein, the preferredcutting sites in opposite strands are staggered by 1 bp in the5' direction, causing the cleavage of a TA and a TT step, respectively.This finding demonstrates that the sequence context is relativelymore important for the cutting frequency than the nature ofthe dinucleotide step of the cleaved bond, and clearly showsthat base recognition is involved in determining the sequenceselectivity of the mutant. The importance of the sequence 5'to the cleaved bond for the cutting rate suggests that the additionalmajor groove contacts may require a distortion of the DNA associatedwith a higher energy barrier, resulting in an increased selectivityfor flexible DNA sequences and a lower overall activity of themutant enzyme.     

Improved growth response of antibody/receptor chimera attained by the engineering of transmembrane domain

Structure-based design of antibody/cytokine receptor chimeras has permitted a growth signal transduction in response to non-natural ligands such as fluorescein-conjugated BSA as mimicry of cytokine-cytokine receptor systems. However, while tight on/off regulation is observed in the natural cytokine receptor systems, many chimeras constructed to date showed residual growth-promoting activity in the absence of ligands. Here we tried to reduce the basal growth signal intensity from a chimera by engineering the transmembrane domain (TM) that is thought to be involved in the interchain interaction of natural cytokine receptors. When the retroviral vectors encoding the chimeras with either the wild-type erythropoietin receptor (EpoR) TM or the one bearing two mutations in the leucine zipper motif were transduced to non-strictly interleukin-6-dependent 7TD1 cells, a tight antigen-dependent on/off regulation was attained, also demonstrating the first antigen-mediated genetically modified cell amplification of non-strictly factor-dependent cells. The results clearly indicate that the TM mutation is an effective means to improve the growth response of the antibody/receptor chimera.     

Protein secretion controlled by a synthetic gene in Escherichia coli

The inability of Escherichia coli to secrete proteins in growthmedium is one of the major drawbacks in its use in genetic engineering.A synthetic gene, homologous to the one coding for the kil peptideof pColE1, was made and cloned under the control of the lacpromoter, in order to obtain the inducible secretion of homologousor heterologous proteins by E.coli. The efficiency of this syntheticgene to promote secretion was assayed by analysing the productionand secretion of two proteins, the R-TEM1 ß-lactamase,and the -amylase from Bacillus licheniformis. This latter proteinwas expressed in E.coli from its gene either on the same plasmidas the kil gene or on a different plasmid. The primary effectof the induction of the kil gene is the overproduction of thesecreted proteins. When expressed at a high level, the kil genepromotes the overproduction of all periplasmic proteins andthe total secretion in the culture medium of both the ß-lactamaseor the -amylase. This secretion is semi-selective for most periplasmkproteins are not secreted. The kil peptide induces the secretionof homologous or heterologous proteins in two steps, first actingon the cytoplasmic membrane, then permeabilizing the outer membrane.This system, which is now being assayed at the fermentor scale,is the first example of using a synthetic gene to engineer anew property into a bacterial strain.     

A 3D model of the {delta} opioid receptor and ligand-receptor complexes

A model for the 3D structure of the transmembrane domain ofthe opioid receptor was predicted from the sequence divergenceanalysis of 42 sequences of G-protein coupled peptide hormonereceptors belonging to the opioid, somatostatin and angiotensinreceptor families. No template was used in the prediction steps,which include multiple sequence alignment, calculation of avariability profile of the aligned sequences, use of the variabilityprofile to identify the boundaries of transmembrane regions,prediction of their secondary structure, optimization of thepacking shape in a helix bundle, prediction of side chain conformationsand structural refinement The general shape of the model issimilar to that of the low resolution rhodopsin structure inthat the TM3 and TM7 helices are most buried in the bundle andthe TM1 and TM4 helices are most exposed to the lipid phase.An initial assessment of this model was made by determiningto what extent a binding site identified using four structurallydisparate high affinity opioid ligands was consistent withknown mutational studies. With the assumption that the pro-tonatedamine nitrogen, a feature common to all opioid ligands, interactswith the highly conserved Aspl27 in TM3, a pocket was foundthat satisfied the criteria of complementarity to the requirementsfor receptor recognition for these four diverse ligands, two selective antagonists (the fused ring naltrindole and the peptideTyr-Tic-Phe-Phe-NH₂) and the two agonists lofentanil and BW373U86deduced from previous studies of the ligands alone. These ligandscould be accommodated in a similar region of the receptor. Thereceptor binding site identified in the optimized complexescontained many residues in positions known to affect ligandbinding in G-protein coupled receptors. These results also allowedidentification of key residues as candidates for point mutationsfor further assessment and refinement of this model as wellas preliminary indications of the requirements for recognitionof this receptor.     

The use of alanine scanning mutagenesis to determine the role of the N-terminus of the regulatory chain in the heterotropic mechanism of Escherichia coli aspartate transcarbamoylase

The location of the first seven residues of the regulatory chainof Escherichia coli aspartate transcarbamoylase has been identifiedby X-ray crystallography to be near the binding site of theregulatory nucleotides. In order to determine the function ofthe N-terminus of the regulatory chain of aspartate transcarbamoylasein heterotropic regulation, alanine scanning mutagenesis wasused. Specifically, Thr2r, His3r, Asp4r, Asn5r, Lys6r and Leu7rwere each replaced with alanine. Analyses of these mutant enzymesindicate that none of these substitutions significantly alterthe catalytic properties of the enzyme. However, three of themutant enzymes, Asp4r Ala, Lys6r Ala and Leu7r Ala, exhibitednotable changes in their response to the regulatory nucleotides,while mutations at Thr2r, His3r and Asn5r exhibited only minorchanges in their heterotropic responses. For the Asp4r Alaenzyme, the responses to ATP and CTP were reduced 30 and 40%respectively, compared with the wild-type enzyme. For the Lys6r Ala enzyme, the response to ATP was reduced 70%, while theCTP response was reduced 50%. In the case of the Leu7r Alaenzyme, a 30 and 20% reduction in response to ATP and CTP respectively,was observed. The synergistk inhibition by UTP in the presenceof CTP for the Lys6r Ala enzyme was reduced 40% compared withthat of the wild type enzyme. For the Leu7r Ala enzyme, thesynergistic inhibition was abolished. In addition, UTP decreasedthe CTP binding affinity of the Leu7r Ala enzyme. Analysisof the kinetic data from these mutant enzymes suggests thatresidues Thr2r, His3r and Asn5r have little effect on the heterotropicmechanism, while residues Asp4r, Lys6r and Leu7r play a moresignificant role in the heterotropic response of the enzymetoward the nucleotides. Furthermore, residue Leu7r appears tobe directly involved in the mechanism for synergistic inhibitionof aspartate transcarbamoylase. In this study alanine scanningmutagenesis has provided a rapid method of identifying thoseresidues in the N-terminal region of the regulatory chain ofaspartate transcarbamoylase important for heterotropic regulation.     

Multiple interactions of lysine-128 of Escherichia coli glutamate dehydrogenase revealed by site-directed mutagenesis studies

A highly conserved lysine at position 128 of Escherichia coliglutamate dehydrogenase (GDH) has been altered by sitedirectedmutagenesis of the gdhA gene. Chemical modification studieshave previously shown the importance of this residue for catalyticactivity. We report the properties of mutants in which lysine-128has been changed to histidine (K128H) or arginine (K128R). Bothmutants have substantially reduced catalytic centre activitiesand raised pH optima for activity. K128H also has increasedrelative activity with amino acid substrates other than glutamate,especially L-norvaline. These differences, together with alterationsin K_m values, K_d values for NADPH and K₁ values for D-glutamate,imply that lysine-128 is intimately involved in either director indirect interactions with all the substrates and also incatalysis. These multiple interactions of lysine-128 explainthe diverse effects of chemical modifications of the correspondinglysine in homologous GDHs. In contrast, lysine-27, another highlyreactive residue in bovine GDH, is not conserved in all of thesequenced NADP-specific GDHs and is therefore not likely tobe involved in catalysis.     

The N-terminal domain of VirG of Agrobacterium tumefadens: modelling and analysis of mutant phenotypes

Fourteen mutants in the N-terminal domain of virulence factorG (VirG) were obtained by random mutagenesis. Two mutants showedan altered phenotype, all others were nonfunctional.All mutantscan still be phosphorylated and bind to DNA. A 3-D model wasbuilt based on the coordinates of chemotaxis protein Y (CheY).Many of the observed phenotypic changes of VirG are explainedqualitatively. Combination of model building and biochemicalinformation leads to the conclusion that the active sites ofVirG and CheY must partly use different residues to performthe same phosphorylation and dephosphorylation reactions.     

Sequence divergence analysis for the prediction of seven-helix membrane protein structures: I. Comparison with bacteriorhodopsin

A method using protein sequence divergence to predict the three-dimensionalstructure of the transmembrane domain of seven-helix membraneproteins is described. The key component in the multistep procedureis the calculation of a hydrophilic and lipophilic variabilityindex for each amino acid in an alignment of a family of homologousproteins. The variability profile, a plot of the calculatedvariability index versus alignment position, can be used topredict a tertiary model of the backbone conformation of thetransmembrane domain. This method was applied to bacteriorhodopsin(BR) and the model obtained was compared with the known structureof this protein. Using an alignment of the amino acid sequencesof BR and closely related (20% identity) proteins, the boundariesof the transmembrane regions, their secondary structures andorientations inside the membrane bilayer were predicted basedon the variability profile. Additional information about theshape of the helix bundle was also obtained from the averagevariability of each transmembrane helix with the assumptionthat the helices are packed sequentially and form a closed helixbundle. Correct features of the known structure of BR were foundin the model structure, suggesting that a similar strategy canbe used to predict transmembrane helices and the packing shapeof other membrane proteins with seven transmembrane helices,such as the opsins and other G-protein coupled receptors.     

Biased random mutagenesis of peptides: determination of mutation frequency by computer simulation

Cassette mutagenesis is a method of protein engineering whichgenerates a wide diversity of genetic variants that can be subjectedto either selection or screening. As long as the target sequenceto be modified is kept short (corresponding to four to six aminoacids), complete combinatorial libraries can be produced. Amajor problem arises when longer peptides are to be engineeredfor desired functions. In such situations the production ofa limited collection of variants can be helpful; thus, biasedrandom mutagenesis and ‘doping schemes’ have beenreported previously. Here we describe a computer algorithm thatenables the determination of the degree of phosphoramidite contaminationof nucleotide precursor reservoirs. Through simulation of biologicaltranslation, the algorithm allows the prediction of the effectof contamination levels on the number of mutations to occurfor any given peptide sequence. In this study the cholinergicbinding site was used as a model sequence (22 amino acids).Considerations, based on the computer program, are discussedregarding the efficient design of phage-display combinatoriallibraries.     

Thermosensitive mutants of Aspergillus awamori glucoamylase by random mutagenesis: inactivation kinetics and structural interpretation

Abstract Seven thermosensitive glucoamylase mutants generated by randommutagenesis and expressed inSaccharomyces cerevisiae were sequencedand their inactivation kinetics were determined. Wild-type glucoamylaseexpressed in S.cerevisiae was more glycosylated and more stablethan the native Aspergillus niger enzyme. All mutants had lowerfree energies of inactivation than wild-type glucoamylase. Inthe Ala39 Val, Ala302 Val and Leu410 Phe mutants, small hydrophobicresidues were replaced by larger ones, showing that increasesin size and hydrophobicity of residues included in hydrophobicclusters were destabilizing. The Gly396 Ser and Gly407 Aspmutants had very flexible residues replaced by more rigid ones,and this probably induced changes in the backbone conformationthat destabilized the protein. The Prol28 Ser mutation changeda rigid residue in an a-helix to a more flexible one, and destabilizedthe protein by increasing the entropy of the unfolded state.The Ala residue in the Ala442 Thr mutation is in the highlyO-glycosylated region surrounded by hydrophilk residues, whereitmay be a hydrophobic anchor Unking the O-glycosylated arm tothe catalytic core. It was replaced by a residue that potentiallyis O-glycosylated. In five of the seven mutations, residuesthat were part of hydrophobic microdomains were changed, confirmingthe importance of the latter in protein stability and structure     

Site-directed mutagenesis of glutathione synthetase from Escherichia coli B: mapping of the {gamma}-L-glutamyl-L-cysteine-binding site

Lysl8, Arg86, Asn283, Ser286, Thr288 and Glu292 of glutathionesynthetase from Escherichia coli B are presumed to be highlyconcerned with the substrate, -L-glutamyl-L-cysteine (-Glu-Cys),binding by X-ray crystallography and affinity labeling studies.Using site-directed mutagenesis, we investigated functionalroles of those residues for -Glu-Cys binding. The mutant enzymesof Arg86 and Asn283 altered their kinetic parameters, especiallythe Michaelis constants of -Glu-Cys. In the case of Asn283,the residue is not likely to have an essential role in -Glu-Cysbinding but its side chain would extend to make a van der Waalscontact with bound -Glu-Cys. Chemical modification of a cysteineresidue with 5,5'-dithiobis(2-nitrobenzoate) (DTNB) showed Arg86would not only be much responsible for -Glu-Cys binding butwould also have a role in maintaining the structural integrityof the enzyme. The other mutant enzymes showed little defectin their kinetic parameters of -Glu-Cys.     

The structural basis for the altered substrate specificity of the R292D active site mutant of aspartate aminotransferase from E.coli

Two refined crystal structures of aspartate aminotransferasefrom E.coli are reported. The wild type enzyme is in the pyridoxalphosphate (PLP) form and its structure has been determined to2.4 Å resolution, refined to an R-factor of 23.2%. Thestructure of the Arg292Asp mutant has been determined at 2.8Å resolution, refined to an R–factor of 20.3%. Thewild type and mutant crystals are isomorphous and the two structuresare very similar, with only minor changes in positions of importantactive site residues. As residue Arg292 is primarily responsiblefor the substrate charge specificity in the wild type enzyme,the mutant containing a charge reversal at this position mightbe expected to catalyze transamination of arginine as efficientlyas the wild type enzyme effects transamination of aspartate[Cronin,C.N. and Kirsch,J.F. (1988) Biochemistry, 27, 4572–4579].This mutant does in fact prefer arginine over aspartate as asubstrate, however, the rate of catalysis is much slower thanthat of the wild type enzyme with its physiological substrate,aspartate. A comparison of these two structures indicates thatthe poorer catalytic efficiency of R292D, when presented witharginine, is not due to a gross conformational difference, butis rather a consequence of both small side chain and main chainreorientations and the pre–existing active site polarenvironment, which greatly favors the wild type ion pair interaction.     

A sequence motif in the transmembrane region of growth factor receptors with tyrosine kinase activity mediates dimerization

A mechanism by which ligand binding to the extracellular domainof a growth factor receptor causes activation of its cytoplasmictyrosine kinase domain is that binding promotes receptor dimerization.Recently we proposed a model in which dimerization of the transmembrane-helices in one member of this family, rat neu, is mediatedby the presence of three specific residues. This paper showsthat a similar sequence motif is observed in 18 of the 20 transmembrane-helices of the tyrosine kinase family of growth factor receptors.The motif encompasses a five residue segment in which position0 (P0) requires a small side chain (Gly, Ala, Ser, Thr or Pro),P3 an aliphatic side chain (Ala, Val, Leu or Ile) and P4 onlythe smallest side chains (Gly or Ala). In addition other featuresof the transmembrane sequences are reported. It is concludedthat the dimerization of transmembrane -helices may be a generalmechanism of tyrosine kinase activation in this family of growthfactor receptors.     

A novel method for predicting transmembrane segments in proteins based on a statistical analysis of the SwissProt database: the PRED-TMR algorithm

We present a novel method that predicts transmembrane domainsin proteins using solely information contained in the sequenceitself. The PRED-TMR algorithm described, refines a standardhydrophobicity analysis with a detection of potential termini(`edges', starts and ends) of transmembrane regions. This allowsone both to discard highly hydrophobic regions not delimitedby clear start and end configurations and to confirm putativetransmembrane segments not distinguishable by their hydrophobiccomposition. The accuracy obtained on a test set of 101 non-homologoustransmembrane proteins with reliable topologies compares wellwith that of other popular existing methods. Only a slight decreasein prediction accuracy was observed when the algorithm was appliedto all transmembrane proteins of the SwissProt database (release35). A WWW server running the PRED-TMR algorithm is availableat http://o2.db.uoa.gr/PRED-TMR/     

Site-directed mutagenesis of Arg58 and Asp86 of elongation factor Tu from Escherichia coli: effects on the GTPase reaction and aminoacyl-tRNA binding

Elongation factor Tu from Escherichia coli was mutated separatelyat positions Asp86 and Arg58, in order to shed light both onthe GTPase mechanism of elongation factor Tu and on the bindingof aminoacyl-tRNA. In addition, the binding of guanine nucleotideswas investigated by determination of the dissociation and associationrate constants. The results imply that Arg58 is unimportantfor the intrinsic GTPase mechanism and the binding of guaninenucleotides, whereas it is strongly involved in the bindingof aminoacyl-tRNA and of the ribosome. Asp86 appears to be essentialfor the regulation of guanine-nucleotide affinities, and itmay also play a role in the intrinsic GTPase mechanism.     

A new method for the extracellular production of recombinant thermolysin by co-expressing the mature sequence and pro-sequence in Escherichia coli

Thermolysin, a representative zinc metalloproteinase from Bacillus thermoproteolyticus, is synthesized as inactive pre-proenzyme and receives autocatalytic cleavage of the peptide bond linking the pro- and mature sequences. The conventional expression method for recombinant thermolysin requires the autocatalytic cleavage, so that production of a mutant thermolysin is affected by its autocatalytic digestion activity. In this study, we have established a new expression method that does not require the autocatalytic cleavage. The mature sequence of thermolysin containing an NH(2)-terminal pelB leader sequence and the pre-prosequence of thermolysin were co-expressed constitutively in Escherichia coli as independent polypeptides under the original promoter sequences in the npr gene which encodes thermolysin. Unlike the conventional expression method, not only the wild-type thermolysin but also mutant thermolysins [E143A (Glu143 is replaced with Ala), N112A, N112D, N112E, N112H, N112K and N112R] were produced into the culture medium. The wild-type enzyme expressed in the present method was indistinguishable from that expressed in the conventional method based on autocatalytic cleavage, as assessed by hydrolysis of N-[3-(2-furyl)acryloyl]-glycyl-L-leucine amide and N-carbobenzoxy-L-aspartyl-L-phenylalanine methyl ester. The present method should be useful especially for preparation of active-site mutants of thermolysin, which might have suppressed autocatalytic digestion activity. The results also demonstrate clearly that the covalent linking between the pro- and mature sequences is not necessary for the proper folding of the mature sequence by the propeptide in thermolysin.