Multiple-Mutation Reaction: a Method for Simultaneous Introduction of Multiple Mutations into the glpK Gene of Mycoplasma pneumoniae

In Mycoplasma pneumoniae, the UGA opal codon specifies tryptophanrather than a translation stop site. This often makes it difficultto express Mycoplasma proteins in E. coli isolates. In thiswork, we developed a strategy for the one-step introductionof several mutations. This method, the multiple-mutation reaction,is used to simultaneously replace nine opal codons in the M.pneumoniae glpK gene.

INTRODUCTION

Top

Introduction

Mycoplasma pneumoniae is a pathogen that lives on mucosal surfacesand causes diseases such as mild pneumonia, tracheobronchitis,and complications affecting the central nervous system, theskin, and mucosal surfaces (9, 14, 26). This bacterium possessesone of the smallest genomes of any free-living organism knownso far. This reduced genome makes Mycoplasma spp. interestingfrom two points of view: (i) the analysis of these bacteriamay help to identify the minimal set of genes that is requiredfor independent life (7), and (ii) M. pneumoniae and its closerelative M. genitalium are well suited for the development ofthe methods of the postgenomic era (10, 27). Another interestingaspect of the small genome is the observation that several enzymesof Mycoplasma spp. are "moonlighting"; i.e., they have multipleunrelated functions (11). This was discovered for glycolytickinases, which are also active as nucleoside diphosphate kinasesin M. pneumoniae and other Mycoplasma spp. (18).

However, the analysis of proteins from Mycoplasma spp. is hamperedby a peculiarity of the genetic code of these bacteria: theyuse the UGA opal codon to incorporate tryptophan rather thanas a stop codon as in the universal genetic code (8, 21). Thus,if cloned into Escherichia coli or other hosts, the genes fromM. pneumoniae may contain many stop codons that prevent heterologousexpression. Several strategies have been developed to solvethis problem. For example, some M. pneumoniae genes, such asptsH or hprK, do not possess UGA codons and thus require nospecial care (24). Expression of mollicute genes in Spiroplasmaspp. that read the UGA as a tryptophan codon was reported, butthese bacteria are difficult to handle (23). E. coli suppressorstrains expressing an opal suppressor tRNA were developed, butthey fail if multiple opal codons are present (22). M. pneumoniaegenes containing few UGA codons have been expressed in Bacillussubtilis with low efficiency (12). In cases with only a fewopal codons, these were changed by site-directed mutagenesisto allow expression in E. coli (13, 17). The M. pneumoniae P1adhesin gene contains 21 opal codons, and a large-scale purificationof the protein, though highly desired, has so far not been possible.In this case, protein fragments were expressed and purified(3). Finally, Mycoplasma genes could be synthesized in vitrofrom oligonucleotides; this strategy is, however, quite expensive.In this work, we present a PCR-based method that allows thesimultaneous introduction of several mutations in a single step.Using this strategy, 9 of the 10 opal codons of the glpK genefrom M. pneumoniae were modified, leading to expression of glycerolkinase in E. coli.

Outline of the multiple-mutation reaction (MMR) strategy.

Top

Outline of the multiple-mutation...

Several methods for PCR-based site-directed mutagenesis havebeen developed. Among these, the combined chain reaction method(1, 2) proved to be very rapid and reliable. The principle ofthis method is the use of mutagenic primers that hybridize morestrongly to the template than the external primers. The mutagenicprimers are phosphorylated at their 5' ends, and these are ligatedto the 3' OH groups of the extended upstream primers by theaction of a thermostable DNA ligase. Moreover, the DNA polymeraseemployed must not exhibit 5' $->$ 3' exonuclease activity, to preventthe degradation of the extended primers. In our view, Pfu andPwo polymerases are both well suited (15, 19). The originalprotocol describes the introduction of two mutations simultaneously.In a previous study, we used a combined chain reaction to mutagenizefour distant bases in a DNA fragment in a one-step reaction(our unpublished results).

For the introduction of up to nine mutations in a single experiment,we developed the MMR. This method requires the efficient bindingof all the mutagenic primers to the target DNA. To ensure thatextension of a PCR product is not possible beyond the next (i.e.,more downstream) mutation site without ligation to the correspondingmutagenic primer, special care needs to be taken in primer design.This reaction is based on an accurate calculation of meltingtemperatures. For this purpose, the formula T_m (melting temperaturein °C) = 81.5 + 16.6(log[Me⁺]) + 0.41 x %G+C – (500/oligonucleotidelength) – 0.61 x % formamide was used (16). Only basesthat match between primer and template were used for the calculation.One consideration was made when designing the mutagenic primers:ligation was facilitated by placing a G or C at the 5' end ofthe oligonucleotide to favor close duplex formation betweenthe primer and the target DNA. The external primers were selectedto have melting temperatures considerably lower (about 4°C)than those of the mutagenic primers. The MMR was performed with2.5 units of Pfu DNA polymerase (MBI Fermentas, Lithuania) and15 units of Ampligase (Epicentre, Madison, WI) in MMR buffer(20 mM Tris-HCl [pH 8.5], 3 mM MgCl₂, 50 mM KCl, 0.4 mg/ml bovineserum albumin, and 0.5 mM NAD⁺) in a total volume of 50 µl.Conditions for MMR included denaturation at 95°C for 30s, primer annealing at 57°C for 30 s, and elongation at65°C for 6 min, for 35 cycles. Initially, the DNA fragment(100 ng) was denatured for 5 min at 95°C. Ten picomolesof each primer was used. The sequences and the arrangementsof the oligonucleotides used in this study are shown in Table1 and Fig. 1, respectively.

View this table:
[in this window]
[in a new window]
TABLE 1. Oligonucleotides used in this study

View larger version (8K):
[in this window]
[in a new window]
FIG. 1. Strategy for amplification and mutagenesis of the M. pneumoniae glpK gene (MPN050 [6]). The positions of the opal codons in the wild-type glpK gene (indicated by a W followed by the number corresponding to the amino acid) and the position and orientation of the external and mutagenic oligonucleotides are shown. The annealing site of each oligonucleotide is indicated by an arrow. Oligonucleotides bearing an A $->$ G transition are depicted by crossed arrows.

Cloning of M. pneumoniae glpK and expression of the protein in E. coli.

Top

Cloning of m. pneumoniae...

An analysis of growth behavior and the in vivo protein phosphorylationpattern identified glycerol as a key carbon source associatedwith regulatory phenomena. This substrate triggered in vivophosphorylation of the HPr protein of the phosphotransferasesystem by the metabolite-sensitive HPr kinase/phosphorylase(5, 24). We were therefore interested in studying glycerol metabolismand its regulation in more detail. As a first step, we intendedto purify the glycerol kinase. This enzyme is known to be akey target of catabolite regulation in gram-positive bacteria(4, 25). However, the corresponding glpK gene contains 10 opalcodons and was therefore a good subject for MMR in order tochange these codons to tryptophan codons for E. coli. The glpKgene was amplified using the oligonucleotides CH7 and CH8 andchromosomal DNA of M. pneumoniae M129 (ATCC 29342) as a template.With CH8, the most C-terminal opal codon was replaced by a TGGcodon. The amplicon was cloned between the SalI and HindIIIsites of the expression vector pWH844 (20). The resulting plasmid,pGP253, was used as a template for MMR with CH9 and CH10 asexternal primers and CH11 through CH19 as mutagenesis primers.Five independent MMRs were carried out, and the MMR productswere individually cloned as a SalI/HindIII fragment into pWH844.The inserts of one clone resulting from each MMR were sequenced.Out of the five candidates, three contained the nine desiredmutations without any additional mutations. One plasmid containedseven out of nine mutations, and the fifth plasmid bore allnine mutations and one additional undesired 1-bp deletion inone of the primer regions. Plasmids bearing all nine desiredmutations but no additional mutations were designated pGP254.pGP254 allows the expression of M. pneumoniae glycerol kinasefused to an N-terminal hexahistidine sequence under the controlof an isopropyl-ß-D-thiogalactopyranoside (IPTG)-induciblepromoter. To test the success of the mutagenesis, we comparedthe protein contents of E. coli cultures carrying either pWH844,pGP253, or pGP254. A prominent band corresponding to an approximatemolecular mass of 56 kDa is detectable in the strain bearingpGP254, while no such protein is expressed from pGP253 encodingthe unmutated glpK gene (Fig. 2A). The glycerol kinase was purifiedto apparent homogeneity by Ni²⁺-nitrilotriacetic acid chromatographyas described previously (Fig. 2B) (15). Thus, MMR was successfulin achieving efficient overproduction of M. pneumoniae glycerolkinase for biochemical studies.

View larger version (25K):
[in this window]
[in a new window]
FIG. 2. Overproduction and purification of M. pneumoniae GlpK. (A) Sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis for the detection of His₆-tagged GlpK in crude extracts of E. coli DH5 ${alpha}$ bearing either the empty expression vector pWH844 (lane 1); the expression vector including the wild-type glpK allele, pGP253 (lane 2); or the vector including the mutated glpK allele, pGP254 (lane 3). Cells were grown to an optical density at 600 nm of 0.8, and expression from the IPTG-inducible promoter was induced by addition of 1 mM IPTG (final concentration). After 2 h, cells were harvested and disrupted by sonication. The insoluble fraction was pelleted in a centrifugation step and solubilized using 6 M urea, and sample aliquots were separated on an SDS-12% polyacrylamide gel. (B) SDS-polyacrylamide gel electrophoresis to monitor the purification of His₆-tagged GlpK. Crude extract of the GlpK expression strain (E. coli DH5 ${alpha}$ bearing the plasmid pGP254) that had been grown in the presence of 1 mM IPTG was passed over a Ni²⁺-nitrilotriacetic acid superflow column (5-ml bed volume; QIAGEN) and washed extensively with a buffer containing 10 mM Tris-HCl (pH 7.4) and 200 mM NaCl, followed by elution with an imidazole gradient (from 10 to 500 mM imidazole). Aliquots of the individual fractions were separated on SDS-12% polyacrylamide gels. A prestained protein molecular mass marker (Fermentas) served as a standard (lane M). Lane 1, flowthrough; lane 2, 10 mM imidazole; lane 3, 20 mM imidazole; lane 4, 50 mM imidazole; lane 5, 100 mM imidazole; lane 6, 200 mM imidazole; lane 7, 500 mM imidazole.

This study demonstrates that MMR can be used for the rapid andhighly efficient introduction of multiple mutations into a gene.Out of five individual clones, four had the desired mutations.Of these four, only one candidate contained an extra mutation,which was most probably due to an impure oligonucleotide mix.Indeed, other experiments indicated that the quality of theoligonucleotides is the limiting factor for MMR. Obviously,this method is useful not only for the expression of Mycoplasmaspecies genes, but also to change codon usage patterns or forany other purpose that requires the introduction of many mutationsor combinations of mutations at the same time. What is the maximumnumber of mutations that can be introduced by MMR in a singlestep? Our results suggest that the target of nine mutationsis still far from a theoretical limit, and we are confidentthat this method can be made even more effective by taking careof the quality of the oligonucleotides (see above) and by usingmutagenic primers that alternate between the two strands ofthe DNA. With this method at hand, even the expression of afunctional P1 adhesin gene in E. coli, which has so far beenbeyond imagination (3), now seems feasible.

ACKNOWLEDGMENTS

We are grateful to Richard Herrmann for the gift of M. pneumoniaechromosomal DNA.

This work was supported by the Fonds der Chemischen Industrie.S.H. was supported by a personal grant from the Fonds der ChemischenIndustrie.

FOOTNOTES

* Corresponding author. Mailing address: Department of General Microbiology, Institute of Microbiology and Genetics, Georg-August University Göttingen, Grisebachstr. 8, D-37077 Göttingen, Germany. Phone: 49-551-393781. Fax: 49-551-393808. E-mail: jstuelk{at}gwdg.de.

REFERENCES

Top