RCA_errorProne

One-step random mutagenesis by error-prone rolling circle amplificationRyota Fujii,Motomitsu Kitaoka*and Kiyoshi HayashiNational Food Research Institute,2-1-12Kannondai,Tsukuba,Ibaraki,305-8642,JapanReceived August 18,2004;Revised and Accepted October 10,2004ABSTRACTIn vitro random mutagenesis is a powerful tool for altering properties of enzymes.We describe here a novel random mutagenesis method using rolling circle amplification,named error-prone RCA.This method consists of only one DNA amplification step followed by transformation of the host strain,without treatment with any restriction enzymes or DNA ligases,and results in a randomly mutated plasmid library with 3–4mutations per kilobase.Specific pri-mers or special equipment,such as a thermal-cycler,are not required.This method permits rapid prepara-tion of randomly mutated plasmid libraries,enabling random mutagenesis to become a more commonly used technique.INTRODUCTIONRandom mutagenesis,coupled with genetic selection or high-throughput screening,is a technique for developing enzymes with novel properties,including altered substrate specificity,enantioselectivity,stability and reaction specificity (1,2).This technique has the advantage of enabling the development of new enzymatic properties without a structural understanding of the targeted enzyme,and often yields unique mutations that could not have been predicted.In addition,further improvements can be expected by repeating the mutagenesis and selection (screening)processes in a manner mimicking Darwinian evolution.This approach,called directed evolution,is a principle method for biomolecular engineering (1–3).The most commonly used random mutagenesis method is error-pronePCR(4),whichintroducesrandommutationsduring PCR by reducing the fidelity of DNA polymerase.The fidelity of DNA polymerase can be reduced by adding manganese ions or by biasing the dNTP e of the compromised DNA polymerase causes mis-incorporation of incorrect nucleo-tides during the PCR reaction,yielding randomly mutated pro-ducts.To convert the product to a suitable form for transforma-tion of a host strain,at least three steps are required:digestion of the product with restriction enzymes,separation of the frag-ments by agarose gel electrophoresis and ligation into a vector.Although these steps do not constitute special techniques,they require almost an entire day of handling time.Further,the liga-tion step can sometimes be troublesome because low ligationefficiency can cause loss of the library.For these reasons,it is desirable to simplify these steps.Another useful random mutagenesis method is the bacterial mutator strain method (5).The most popular mutator strain is Escherichia coli XL1-Red (Stratagene,La Jolla,CA),which lacks three of the primary DNA repair pathways,MutS ,MutD and MutT ,resulting in a random mutation rate $5000-fold higher than in wild type.The protocol for using the mutator strain is composed of two steps:transformation of the mutator strain and recovery of the mutant from the transformant.This protocol is much simpler than error-prone PCR,and a ligation step is unnecessary.However,the mutation frequency is low under the standard conditions (0.5mutations per kilobase)(5),and a cultivation period longer than 24h is often required for introducing multiple mutations.Rolling circle amplification (RCA)(6–8)is an isothermal method that amplifies circular DNA by a rolling circle mechan-ism (9),yielding linear DNA composed of tandem repeats of the circular DNA sequence.This method has several advan-tages over conventional methods for amplifying DNA,such as PCR.For example,it does not require specific primers because random hexamers can be used as a universal primer for any template (10),nor does it require a thermal-cycler because the amplification reaction proceeds at a constant temperature.In addition to the ease of amplifying circular DNA,RCA products have a unique feature in that they can be used for direct trans-formation of E.coli (Fujii,R.,Kitaoka,M.and Hayashi,K.,manuscript submitted)and yeast (11),yielding re-circularized template DNA in the transformants.Therefore,the amplified product can be used directly to transform a host strain.We here describe the ‘simplest’random mutagenesis method using RCA,named error-prone RCA.This method consists of only one RCA step followed by direct transformation of the host strain,and yields mutants with an adequate mutation frequency for in vitro evolution experiments (3–4mutations per kilobase).No restriction enzymes,ligases,specific primers or special equipment such as a thermal-cycler are required.Therefore,this method is much more convenient than any other random mutagenesis methods.This method will enable random muta-genesis to become a more commonly used technique.MATERIALS AND METHODS MaterialsE.coli strains TOP10[F ÀmcrA D (mrr -hsdRMS -mcrBC )f 80lacZ D M15D lacX 74recA 1deoR araD 139D (ala -leu )7697*To whom correspondence should be addressed.Tel:+81298388071;Fax:+81298387321;Email:mkitaoka@nfri.affrc.go.jpNucleic Acids Research,Vol.32No.19ªOxford University Press 2004;all rights reservedNucleic Acids Research,2004,Vol.32,No.19e145doi:10.1093/nar/gnh147Published online October 26, 2004 at :: on May 19, 2011 Downloaded fromgalU galK rpsL (Str R )endA 1nupG ]and DH5a [F Àf 80lacZ D M15D (lacZYA-argF )U169deoR recA 1endA 1hsdR 17(r K Àm K +)phoA supE 44l Àthi -1gyrA 96relA 1]were purchased from Invitrogen (Carlsbad,CA)and TaKaRa (Shiga,Japan),respectively.f 29DNA polymerase was purchased from New England Biolabs (Beverly,MA),and the restriction enzyme BamHI was purchased from TaKaRa.TempliPhi 100DNA amplification kit was purchased from Amersham Biosciences (Piscataway,NJ).Ampicillin sodium salt and ceftazidime pentahydrate were purchased from Nacalai Tesque (Kyoto,Japan)and Sigma (St Louis,MO),respectively.The MinElute Reaction Cleanup and QIAprep miniprep kits were purchased from QIAGEN (Hilden,Germany).Error-prone rolling circle amplificationRCA was performed using the TempliPhi 100DNA amplifica-tion kit,which has a sample buffer containing random hexa-mers that prime DNA synthesis nonspecifically;an enzyme mix containing f 29DNA polymerase and random hexamers,and a reaction buffer containing deoxyribonucleotides.Although the exact composition of the TempliPhi kit is not known,the RCA reaction can be reproduced using f 29DNA polymerase and exonuclease-resistant random hexamers with thiophosphate linkages for the two 30terminal nucleotides (10).In a total volume of 10m l,the final concentrations were 1U/m l of f 29DNA polymerase and 4pmol/m l of exonuclease-resistant random hexamers in 50mM Tris–HCl buffer (pH 7.5),containing 10mM MgCl 2,10mM (NH 4)2SO 4,200ng/m l BSA,4mM DTT,0.2mM dNTP and template DNA (data not shown).As a template for the RCA reaction,we used purified pUC19dissolved in water or E.coli TOP10harboring pUC19,a colony of which was picked from a Luria–Bertani (LB)medium plate using a pipette tip and suspended in 50m l of 10mM Tris–HCl buffer (pH 8.0)containing 1mM EDTA.An aliquot (0.5m l)of the latter was mixed with 5m l of sample buffer,and the mixture was heated at 95 C for 3min to denature the plasmid and to lyse the cells.The sample was cooled to room temperature immediately.The amplification reaction was started by adding a premix from the TempliPhi kit of 5m l of reaction buffer,0.2m l of enzyme mix and 1m l of MnCl 2(0–20mM),followed by incubation at 30 C.The mix-ture was subsequently heated at 65 C for 10min to inactivate the enzyme,and a 5m l aliquot of the product was purified by using the MinElute Reaction Cleanup kit.The amount of amplified DNA was estimated by measuring its absorbance at 260nm with a NanoDrop ND-1000spectrophotometer (NanoDrop Technologies,Rockland,DE).Transformation of E.coli with RCA productElectrocompetent E.coli (50m l)were electroporated with a 1m l aliquot of the RCA product using an E.coli Pulser (Bio-Rad,Hercules,CA)in a 0.1cm electrode cuvette at a voltage setting of 1.8kV.The cells were incubated in 1ml of SOC medium at 37 C for 1h while reciprocal shaking at 160r.p.m.and then plated onto a LB plate containing 20ng/m l ampicillin sodium salt.The plate was incubated at 37 C for 16h.Characterization of plasmid in the transformantColonies on the plate were inoculated into LB medium containing 20m g/ml ampicillin sodium salt,and were incu-bated at 37 C overnight.Plasmid DNA was isolated from the culture medium using a QIAprep miniprep kit.A 1m l aliquot of the isolated plasmid (approx.100ng/m l)was digested with BamHI,and both the digested and undigested plasmids were analyzed by agarose gel electrophoresis.All plasmids of correct size on the agarose gels were sequenced using a CEQ-2000DNA Analysis System with a DTCS quick start kit (Beckman Coulter,Fullerton,CA).To measure mutation frequency in the recovered plasmid,plasmid DNA sequence was determined using an M13(À47)primer.DNA sequences of the other regions were determined using forward primers corresponding to DNA positions 841,1358,1750,2351and 2661.Improvement of ceftazidime resistance of TEM-1b -lactamase using error-prone RCAPlasmid pUC19,which contains TEM-1b -lactamase as a mar-ker for ampicillin,was amplified by error-prone RCA in buffer containing 1.5mM MnCl 2in a volume of 10m l.The product was precipitated with 70%ethanol and used to transform E.coli DH5a in 1ml medium.To estimate the total number of trans-formants,a 5m l aliquot of medium was spread on a LB plate containing 20ng/m l ampicillin sodium salt,and the residual mediumwasspreadonaLB plate containing1ng/m lceftazidime pentahydrate,which completely inhibits the growth of E.coli containing wild-typepUC19(12).After 16hincubationat 37 C,colonies formed on the ceftazidime plate were inoculated into LB liquid medium containing 1ng/m l ceftazidime.The recovered plasmids were subjected to agarose gel electrophor-esis,and the DNA sequence of the TEM-1b -lactamase gene in each transformant was analyzed.RESULTSError-prone rolling circle amplificationRCA is a laboratory method to amplify circular DNA by the rolling circle mechanism,yielding linear DNA composed of tandem repeats of the circular DNA sequence (10).The RCA product has a unique feature in that it can be used for the direct transformation of E.coli ,yielding transformants containing a plasmid identical to the RCA template (Fujii,R.,Kitaoka,M.and Hayashi,K.,manuscript submitted).We therefore con-structed a simple and rapid method for introducing random mutations into plasmid DNA.Plasmid DNA was amplified by the rolling circle mechanism in the presence of manganese ions,which has been shown to reduce the fidelity of DNA polymerases and cause random mutagenesis during RCA.E.coli was directly transformed with the RCA product,resulting in colonies containing a randomly mutated plasmid library.Although the band mobility of the recovered plasmids on agarose gel electrophoresis was mostly identical with pUC19,there were some plasmids with lower mobility than pUC19.These plasmids with lower mobility could be multi-mers,which are circular DNAs having two or more repeated sequences of pUC19(Fujii,R.,Kitaoka,M.and Hayashi,K.,e145Nucleic Acids Research,2004,Vol.32,No.19P AGE 2OF5at :: on May 19, 2011 Downloaded frommanuscript submitted).We analyzed a total of 174clones by agarose gel electrophoresis,resulting in 25clones (14%)being identified as multimers.The proportion of the monomer to multimer was almost constantly independent of mutation frequency (data not shown).When the multimers were sequenced,the mutated residues were often overlapped wild-type residues.This is probably due to the multimeric structure,which contains at least two different plasmid DNA sequences.Therefore,we used only monomers in the following experi-ments characterizing mutations.Mutation frequencyWe found that the mutation frequency increased when the concentration of manganese increased or when the initial amount of template decreased (Table 1).A too high mangan-ese concentration,however,was found to reduce the efficiency of amplification (2mM MnCl 2for 25pg pUC19).The max-imum mutation frequency was 3.5–1.0mutations/kilobase.When we examined the relationship between mutation frequency and amplification rate,we found that,as the amplification reaction proceeded,the mutation frequency increased (Table 2).A considerable number of mutations were introduced after 8h of incubation,and the highest muta-tion frequency was obtained after 24h of incubation.Library sizeIn determining the size of the mutant library produced by the error-prone RCA method (Table 3),we found that,in the presence of 1and 1.5mM MnCl 2,2200and 8900colonies,respectively,were obtained from 1m l of the RCA product.These values were lower than that obtained under error-free conditions (38000),indicating that increasing the concentra-tion of MnCl 2decreased the numbers of colonies.This was due to a decrease in the yield of the RCA product and was inde-pendent of the transformation efficiencies (Table 3).This indi-cates that the effect of mutagenesis on the genes of the plasmid replication system may be low.Characterization of mutationsTo analyze the distribution and variation of mutations caused by error-prone RCA,we isolated seven mutated plasmids from the mutant library constructed in the presence of 1.5mM MnCl 2,and each was sequenced.When we examined the distribution of mutations (Figure 1and Table 4),we found that the mutation frequency in the region from nucleotides 800to 2500was slightly lower than that in other regions.Because this region encodes genes critical for the plasmid (nucleotides 867–1455is the origin of replication,and nucleo-tides 1626–2486is the ampicillin-resistance gene),some mutations in this region may have been lethal and wereTable 1.Effect of RCA conditions on mutation frequency pUC19(pg)MnCl 2(mM)Sequenced base pair Number of mutations Mutation frequency (mutations/kb)a 250505800250.5597610.2–0.225161168 1.3–0.525 1.5376613 3.5–1.0252—b —b —b 2.50.5727210.1–0.12.51440012 2.7–0.82501488410.2–0.22502279031.1–0.6Plasmid pUC19(2.5,25or 250pg)was amplified by TempliPhi 100DNA amplification kit with varying concentrations of MnCl 2(0–2mM)for 24h.The reaction mixture was heated at 65 C for 10min to stop the reaction,and E.coli TOP10was transformed with the amplified product.Plasmids were recovered from the isolated transformants,and their DNA sequence was analyzed using the M13(À47)forward primer.aStandard errors were calculated assuming that the values follow a Poisson distribution.bNo transformants obtained.Table 2.Correlation between reaction time and mutation frequency Reaction time (h)Sequenced base pair Number of mutations Mutation frequency (mutations/kb)a 0—b —b —b4601330.5–0.38696811 1.6–0.5243766133.5–1.0Plasmid pUC19(25pg)was amplified by TempliPhi 100DNA amplification kit with 1.5mM MnCl 2for 0–24h.The reaction mixture was heated at 65 C for 10min to stop the reaction,and E.coli TOP10was transformed with the ampli-fied product.Plasmids were recovered from the isolated transformants,and their DNA sequence was analyzed using the M13(À47)forward primer.aStandard errors were calculated assuming that the values follow a Poisson distribution.bNo transformants obtained.Table 3.Library size MnCl 2(mM)RCA product (ng/m l)Number of colonies Transformation efficiency (c.f.u./m g)0120380003·10512189004·1051.5922002·105Plasmid pUC19(25pg)was amplified by TempliPhi 100DNA amplification kit with varying concentrations of MnCl 2(0,1and 1.5mM)for 24h.The reaction mixture was heated at 65 C for 10min to stop the reaction,and E.coli TOP10was transformed with 1m l of the amplified product.Table 4.Types of mutation Mutation type Number of mutations Percentage in substitutions G to A,C to T 4366G to C,C to G 46G to T,C to A 711A to T,T to A 58A to G,T to C 58A to C,T to G 12Insertion 6Deletion 0Total71100Figure 1.Distribution of mutations.The seven clones produced by RCA in the presence of 1.5mM MnCl 2from 25pg pUC19were sequenced.Lines indicate mutations in the pUC19sequence.Types of mutations are described in Table 4.P AGE 3OF5Nucleic Acids Research,2004,Vol.32,No.19e145at :: on May 19, 2011 Downloaded fromtherefore excluded from the library.The transformation efficiency did not decrease much under error-prone conditions (Table3),however,indicating that the influence of the lethality was trivial for the mutant library.When we examined the diversity of mutagenesis(Table4), we found that the direction of the mutations was biased in favor of C to T and G to A mutations(66%),and that the transition/transversion ratio was2.7.Improvement of ceftazidime resistance of TEM-1b-lactamaseTo verify that the error-prone RCA method can be used for evolutionary experiments,we altered the substrate specificity of TEM-1b-lactamase using error-prone RCA.TEM-1 b-lactamase is an enzyme that hydrolyzes the b-lactam ring and is generally used as a marker for b-lactam antibiotics such as ampicillin.In contrast,this enzyme works poorly against third-generation cephalosporins,such as ceftazidime.We introduced random mutations into the TEM-1b-lactamase gene of pUC19using error-prone RCA in the presence of1.5mM MnCl2.The RCA product was used to transformE.coli DH5a,and mutants with high ceftazidime-hydrolyzing activity were selected.We found that10colonies grew on the LB plate containing1m g/ml ceftazidime,compared with 10000on the ampicillin plate.Of the10plasmids recovered from the ceftazidime plate,agarose gel electrophoresis showed that7had bands of the same size as pUC19,whereas the other 3showed lower mobility than pUC19,indicating that they were probably multimeric plasmid DNA(data not shown). The seven plasmids of the same size as pUC19were sequenced to identify mutations in the b-lactamase gene[Table5;(13)]. All seven plasmids had mutations at R164(to H,G or C)or at D179(to G),all of which are known to increase the ceftazi-dime resistance of TEM-1b-lactamase(14,15).Therefore, we have successfully used error-prone RCA to introduce mutations that altered the substrate specificity of b-lactamase,indicating the applicability of this method for in vitro evolution experiments.DISCUSSIONRandom mutagenesis is a powerful tool for altering the proper-ties of enzymes(1,2).In this study,we have developed a random mutagenesis method using the RCA technique.This method is composed of only one DNA amplification step, followed by direct transformation of the host strain.Target DNA was amplified by the RCA technique in the presence of manganese ions,which reduces thefidelity of DNA polymer-ase and causes the mis-incorporation of nucleic acids(4).This product is used to transform E.coli,yielding colonies that contain randomly mutated plasmids.Although we used E.coli as a host strain in this report,other hosts,including yeast(11)and probably any other host having a DNA recomb-ination system,may also be used.The prime advantage of error-prone RCA is its rapidity. This method consists of only one RCA step,indicating that it is much quicker than the conventional random mutagenesis methods,such as error-prone PCR or mutator strain method (Figure2).The error-prone RCA reaction mixture can be prepared within several minutes,followed by isothermal incu-bation,to yield an amplified DNA suitable for direct transfor-mation of a host strain.Specific primers for target DNA are not necessary because random hexamers can be used as the universal primer for any plasmid.Furthermore,setting of ther-mal-cycling conditions,as in PCR,is not necessary,because the RCA reaction proceeds under isothermal conditions. Therefore,it is no exaggeration to say that error-prone RCA is the simplest of all the random mutagenesis methods. Using error-prone RCA,we introduced random mutations into plasmid DNA at a frequency of up to3.5–1.0mutations per kilobase.This mutation frequency corresponds to almost one amino acid mutation per kilobase and is therefore appro-priate for in vitro evolution experiments(3).In addition,the mutation frequency could be controlled by varying the con-centration of manganese ions.This concentration,however, was limited to below2mM because excess MnCl2decreasedTable5.Mutationsof theTEM-1b-lactamasesequencein ceftazidime-resistant variants produced by error-prone RCAColony no.MutationDNA a Amino acid b1G485A R164H2C82T,G485A,C674T silent,R164H,A227V3T113G,C484G,G852A L40W,R164G,silent4C484T R164C5C273T,G485A,A511G,C526Tsilent,R164H,I173V,R178C6A530G D179G7A530G D179GPlasmid pUC19(25pg)was amplified by TempliPhi100DNA amplification kit with1.5mM MnCl2in10m l volume.After24h incubation at30 C, the reaction was stopped by heating at65 C for10min.The product was precipitated with70%ethanol,and each precipitate was dissolved in1m l water and used to transform E.coli DH5a by electroporation.The transformants were spread on a LB plate containing1ng/m l of ceftazidime and incubated for 16h at37 C.The plasmids were recovered from the colonies,their size was analyzed by agarose gel electrophoresis,and plasmids of the correct size were sequenced to identify mutations in the TEM-1b-lactamase gene.a DNA position based on the TEM-1b-lactamase sequence.b Amino acid position based on the standard numbering for class A b-lactamase(13). at :: on May 19, 2011 Downloaded fromthe RCA e of a modified f 29DNA polymerase without 30–50exonuclease activity may increase the mutation frequency (16).For example,introduction of an H61R muta-tion into f 29DNA polymerase was found to result in a 16-fold increase in mis-incorporation efficiency,as well as a 6-to 23-fold increase in polymerization efficiency.These findings indicate that both mutation frequency and yield may be improved by using the H61R mutant of f 29DNA polymerase.While all types of substitution mutations were found in error-prone RCA variants,the mutation direction of error-prone RCA with f 29DNA polymerase was biased in favor of C to T and G to A mutations (66%).In contrast,these mutations are less favored in error-prone PCR using Taq DNA polymerase (14%)(17).These findings indicate that error-prone RCA could give rise to a wide range of amino acid substitutions not observed in error-prone PCR.Mutations are introduced throughout the entire plasmid by error-prone RCA,as well as in mutator strain mutagenesis (5),and mutations in regions other than the targeted gene might cause unexpected effects.The transformation efficiencies of varying concentrations of MnCl 2were almost constant,however,indicating that these mutations did not have a dele-terious effect on the plasmid replication system.There are many successful reports for improving enzymatic properties by mutating the entire region of the plasmid DNA by mutator strain mutagenesis (18–20),and therefore,the influence of mutations in other regions was probably negligible.To verify that error-prone RCA can be used to alter enzy-matic properties,we used this method to increase the cefta-zidime resistance of TEM-1b -lactamase.The plasmid pUC19,which has TEM-1b -lactamase gene,was mutated by error-prone RCA,and E.coli DH5a transformed with the RCA product was cultured on a ceftazidime plate.Of the seven mutant pUC19plasmids with improved ceftazidime resist-ance,all had mutations at R164(to H,G or C)or D179(to G).Both of these amino acids are located at the root of the V -loop of the TEM-1b -lactamase structure,which forms part of the substrate-binding domain,and mutations in these residues are known to improve ceftazidime resistance (14,15).There-fore,we have demonstrated that error-prone RCA can be used for altering enzymatic properties,indicating the applicability of this method for in vitro evolution experiments.In conclusion,we have developed a simple method for constructing a randomly mutated plasmid library using RCA (error-proneRCA).ThismethodwascomposedofoneRCAstep followed by direct transformation of a host strain.A target plas-mid was amplified by fidelity-reduced RCA,and the host strain was transformed directly with the product to give a mutant library with up to 3.5–1.0mutations per e of this method will save considerable labor in the introduction of random mutants.This is the simplest protocol for the prepara-tionofarandomlymutated plasmid librarytoourknowledgeand will make random mutagenesis more common.ACKNOWLEDGEMENTSThis study was supported by a grant from the Bio-oriented Technology Research Advancement Institution.REFERENCES1.Arnold,F.H.,Wintrode,P.L.,Miyazaki,K.and Gershenson,A.(2001)How enzymes adapt:lessons from directed evolution.Trends Biochem.Sci.,26,100–106.2.Brakmann,S.(2001)Discovery of superior enzymes by directed molecular evolution.Chembiochem.,2,865–871.3.Reetz,M.T.and Jaeger,K.E.(1999)Biocatalysis—From Discovery to Application .Springer-Verlag Berlin,Berlin,Vol.200,pp.31–57.4.Leung,D.W.,Chen,E.and Goeddel,D.W.(1989)A method for random mutagenesis of a defined DNA segment using a modified polymerase chain reaction.Techniques ,1,11–15.5.Greener,A.,Callahan,M.and Jerpseth,B.(1996)In Vitro Mutagenesis Protocols .Humana press,NJ.6.Fire,A.and Xu,S.Q.(1995)Rolling replication of short DNA circles.Proc.Natl A ,92,4641–4645.7.Liu,D.Y.,Daubendiek,S.L.,Zillman,M.A.,Ryan,K.and Kool,E.T.(1996)Rolling circle DNA synthesis:small circular oligonucleotides as efficient templates for DNA polymerases.J.Am.Chem.Soc.,118,1587–1594.8.Lizardi,P.M.,Huang,X.,Zhu,Z.,Bray-Ward,P.,Thomas,D.C.and Ward,D.C.(1998)Mutation detection and single-molecule counting using isothermal rolling-circle amplification.Nature Genet.,19,225–232.9.Kornberg,A.and Baker,T.(1992)DNA Replication,2nd edn .W.H.Freeman &Company,NY.10.Dean,F.B.,Nelson,J.R.,Giesler,T.L.and Lasken,R.S.(2001)Rapidamplification of plasmid and phage DNA using Phi 29DNA polymerase and multiply-primed rolling circle amplification.Genome Res.,11,1095–1099.11.Ding,X.,Snyder,A.K.,Shaw,R.,Farmerie,W.G.and Song,W.Y.(2003)Direct retransformation of yeast with plasmid DNA isolated fromsingle yeast colonies using rolling circle amplification.BioTechniques ,35,774–779.12.Gaytan,P.,Osuna,J.and Soberon,X.(2002)Novel ceftazidime-resistanceb -lactamases generated by a codon-based mutagenesis method and selection.Nucleic Acids Res.,30,e84.13.Ambler,R.P.,Coulson,A.F.W.,Frere,J.M.,Ghuysen,J.M.,Joris,B.,Forsman,M.,Levesque,R.C.,Tiraby,G.and Waley,S.G.(1991)A standard numbering scheme for the class A b -lactamases.Biochem.J.,276,269–270.14.Vakulenko,S.B.,Taibi-Tronche,P.,Toth,M.,Massova,I.,Lerner,S.A.andMobashery,S.(1999)Effects on substrate profile by mutational substitutions at positions 164and 179of the class A TEM pUC19b -lactamase from Escherichia coli .J.Biol.Chem.,274,23052–23060.15.Vakulenko,S.B.,Toth,M.,Taibi,P.,Mobashery,S.and Lerner,S.A.(1995)Effects of Asp-179mutations in TEM pUC19b -lactamase on susceptibility to b -lactams.Antimicrob.Agents Chemother.,39,1878–1880.16.de Vega,M.,Lazaro,J.M.and Salas,M.(2000)Phage f 29DNApolymerase residues involved in the proper stabilisation of the primer-terminus at the 30–50exonuclease active site.J.Mol.Biol.,304,1–9.17.Shafikhani,S.,Siegel,R.A.,Ferrari,E.and Schellenberger,V.(1997)Generation of large libraries of random mutants in Bacillus subtilis by PCR-based plasmid multimerization.BioTechniques ,23,304–310.18.Camps,M.,Naukkarinen,J.,Johnson,B.P.and Loeb,L.A.(2003)Targeted gene evolution in Escherichia coli using a highly error-prone DNA polymerase I.Proc.Natl A ,100,9727–9732.19.Henke,E.and Bornscheuer,U.T.(1999)Directed evolution of an esterasefrom Psueudomonas fluorescens .Random mutagenesis by error-prone PCR or a mutator strain and identification of mutants showing enhanced enantioselectivity by a resorufin-based fluorescence assay.Biol.Chem.,380,1029–1033.20.Bornscheuer,U.T.,Altenbuchner,J.and Meyer,H.H.(1998)Directedevolution of an esterase for the stereoselective resolution of a key intermediate in the synthesis of epothilones.Biotechnol.Bioeng.,58,554–559.P AGE 5OF5Nucleic Acids Research,2004,Vol.32,No.19e145at :: on May 19, 2011 Downloaded from。