Probing In Vivo Promoter Activities in Mycoplasma pneumoniae: a System for Generation of Single-Copy Reporter Constructs

The nucleotide sequences that control transcription initiationand regulation in Mycoplasma pneumoniae are poorly understood.In this work, we developed a lacZ-based reporter plasmid thatcan be used to integrate fusions of promoter fragments to apromoterless lacZ gene into the chromosome of M. pneumoniae.

INTRODUCTION

Top

Introduction

Mycoplasma pneumoniae is a human pathogen. The bacteria liveon mucosal surfaces of the respiratory tract and cause diseasessuch as mild pneumonia and tracheobronchitis. In addition, somenonrespiratory complications affecting the skin, mucosa, thecentral nervous system, the heart, and other organs were reported(14, 18, 24). During the last few years, M. pneumoniae and relatedcell wall-less bacteria of the class Mollicutes have attractedconsiderable scientific interest since these bacteria possessone of the smallest genomes of any free-living organism knownso far (9). The minimal genetic complement of M. pneumoniaeand its close relative, M. genitalium, has prompted studiesto identify the essential gene set required for independentlife (5, 10). Moreover, the molecular details of the interactionof M. pneumoniae with the host tissues that lead to pathogenesisare far from being understood. Similarly, not much is knownabout gene expression in M. pneumoniae. A few global studiesof gene expression in M. pneumoniae have been reported (15,21, 23, 28). In good agreement with the life of M. pneumoniaein a rather constant environment is the small number of regulatoryproteins encoded by the genomes of these bacteria. While thetranscription of some individual genes was studied in M. pneumoniae(1, 6, 11, 12, 27), nothing is known of regulatory mechanismsin these bacteria.

The molecular analysis of M. pneumoniae has been hampered bythree problems. First, the genes of M. pneumoniae and relatedmollicutes use the UGA opal codon to incorporate tryptophanrather than as a stop codon as in the universal genetic code.This makes it difficult to express proteins from Mycoplasmaspp. in heterologous hosts in order to make them available forbiochemical analysis (3, 13). Recently, a method for the simultaneousreplacement of multiple opal codons was developed and used forthe expression of M. pneumoniae glycerol kinase in Escherichiacoli (7). A second major problem is the lack of genetic systemsthat allow the efficient targeted generation of M. pneumoniaemutants. Therefore, genetic research with these bacteria dependson the use of mutant strains that have been isolated in conventionalscreens or even by chance (26, 29). Finally, transcription inM. pneumoniae can so far be studied by only RNA-based methods,such as transcriptome analyses, Northern blotting, reverse transcriptasePCR, or primer extension for the determination of 5' ends oftranscripts (1, 6, 11, 12, 25, 27, 28). The molecular analysisof transcription regulatory mechanisms has so far not been possibledue to the lack of appropriate reporter systems that can beused to study the activity of promoter fragments and their mutantderivatives in vivo. In this study, we report a system for thegeneration of fusions of M. pneumoniae promoters to a promoterlesslacZ gene that can be integrated into the M. pneumoniae chromosome.

Construction of the M. pneumoniae reporter system.

Top

Construction of the m....

Since no genetic system allowing the targeted integration ofDNA fragments into the M. pneumoniae chromosome is available,we made use of a derivative of Tn4001 (8), which lacks the transposasegene (mini-Tn4001). The mini-Tn4001 used here contains an originof replication that functions in E. coli but not in M. pneumoniaeand a aac-aphD gentamicin resistance gene which can be usedto select for gentamicin or kanamycin resistance in M. pneumoniaeor E. coli, respectively (22). Tn4001 is known to insert randomlyinto the chromosome of M. pneumoniae (10). The mini-Tn4001,together with the tnp gene encoding transposase, was presenton plasmid pMT85 (30). As the reporter, we selected the lacZgene of E. coli encoding ß-galactosidase, which isone of the most popular reporter enzymes, due to the possibilityof getting a quick qualitative impression of the enzymatic activityin colonies using plates containing X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside),and the quantitative assay using o-nitrophenyl-ß-D-galactopyranosideas the chromogenic substrate (20). In addition, lacZ-based reportersystems were already established in other mollicutes, such asAcholeplasma oculi, Mycoplasma pulmonis, Mycoplasma arthritidis,and Mycoplasma capricolum (4, 16, 17). To facilitate cloningof the promoter fragments and their detection in E. coli, wemade use of a lacZ gene devoid of a ribosomal binding site.With such a reporter, a Shine-Dalgarno sequence must be providedwith the cloned promoter fragment to obtain a functional lacZfusion. These requirements are met by the lacZ gene presentin plasmid pAC5 (19). The lacZ gene of pAC5 was amplified usingthe primers SH44 (5' TATTTAAGTACTATAATAAGGGTAACTATTGCCG) andSH45 (5' GAACTAGTACATAATGGATTTCCTTAC). The resulting fragmentwas digested with BcuI and ScaI (these sites were introducedupon PCR and are underlined in the primer sequences) and clonedbetween the BcuI and OliI sites of pMT85. The resulting plasmidwas pGP353 (Fig. 1).

View larger version (21K):
[in this window]
[in a new window]
FIG. 1. Map of plasmid pGP353. The plasmid was constructed as described in the text. Abbreviations: IR, inverted repeats; tnp, transposase gene; aac-aphD, gentamicin/kanamycin resistance gene; lacZ, ß-galactosidase; ori colE1, gram-negative origin of replication. Restriction sites that were available for the construction of translational promoter lacZ fusions and the BcuI site used to construct pGP353 are indicated.

Construction and analysis of an ldh-lacZ fusion.

Top

Construction and Analysis of...

We are interested in carbon metabolism in M. pneumoniae andits regulation. These bacteria catabolize a few sugars suchas glucose, fructose, and glycerol via glycolysis but lack acitric acid cycle (6, 9). The NADH⁺ formed in glycolysis canbe reoxidized by the formation of lactate from pyruvate. Theldh gene (MPN674) encoding lactate dehydrogenase is one of thefew genes in M. pneumoniae which is transcribed in the oppositedirection relative to the genes located up- and downstream.This suggests that ldh is transcribed monocistronically witha promoter upstream of the gene. Indeed, a primer extensionassay revealed the presence of a promoter similar to the consensussequence of the single M. pneumoniae ${sigma}$ factor (our unpublishedresults; 27). To fuse the ldh promoter region to the promoterlesslacZ gene present in pGP353, the region from –160 to +81relative to the ldh transcription start point was amplifiedusing the oligonucleotides SH46 (5' AGAATTCAAACTGCATCGTGGTATCTG)and SH47 (5' TAGGATCCGCGTAGAGAAAGCTGGTGC) and cloned betweenthe EcoRI and BamHI sites of pGP353 (these sites are underlinedin the primer sequences). In the resulting plasmid pGP354 (seeFig. 2), the lacZ gene was fused in frame to the 21st codonof the lactate dehydrogenase gene. The promoter fragment presentin pGP354 contains two internal HindIII sites. These sites wereused to delete the promoter resulting in plasmid pGP364 (seeFig. 2).

View larger version (6K):
[in this window]
[in a new window]
FIG. 2. Scheme of the ldh promoter fragments of M. pneumoniae that have been used for the construction of pGP354 and pGP364. Putative –10 and –35-boxes of the ldh gene (MPN674) and the MPN673 gene (conserved hypothetical gene) are indicated (27). Numbering is relative to the transcriptional start point of the ldh gene.

The plasmids pGP353, pGP354, and pGP364 were used to electroporateM. pneumoniae M129 according to the protocol described previously(2). Transformants were selected on plates containing gentamicinwith or without X-Gal. On X-Gal-containing plates, all transformantsthat were obtained with pGP354 formed blue colonies, whereastransformants that were obtained with pGP353 and pGP364 formedwhite colonies (see Fig. 3A). This was the first indicationthat the promoter fragment was present in the original fragmentand that promoter activity was lost upon deletion of the internalHindIII fragment. To be unbiased in the further analysis, fivecolonies of each transformation were reisolated from platesthat did not contain X-Gal. These colonies were cultivated inorder to obtain DNA for the verification of the presence ofthe fusion in the M. pneumoniae cells and to prepare cell extractsfor the quantitative determination of ß-galactosidaseactivity. As shown in Fig. 4, the presence of all three fusionswas demonstrated and the M. pneumoniae strains were designatedGPM53, GPM54, and GPM64 (transformations with pGP353, pGP354,and pGP364, respectively). To eliminate positional effects oftransposon integration at random sites in the chromosome, fiveindividual clones per plasmid were chosen and their ß-galactosidaseactivities were determined (20) in cell extracts that were preparedas described previously (6). These individual clones were designatedGPM53/1 through GPM53/5. As shown in Fig. 3B, no ß-galactosidasewas detectable in the five GPM53 clones. This confirms thatthe "empty" reporter cloning vector does not confer any expressionof the promoterless lacZ gene. In contrast, a high activitywas observed in all GPM54 clones containing the ldh promoterand Shine-Dalgarno sequence upstream of the lacZ gene. Thisfinding demonstrates that the E. coli lacZ gene can be efficientlytranslated in M. pneumoniae. In the M. pneumoniae GPM64 clones,the core of the ldh promoter is deleted from the fragment upstreamof lacZ. This resulted in a 7.5-fold reduction of ß-galactosidaseactivity compared to that for the GPM54 clones containing thecomplete promoter fragment. Even though the transposons insertedat different positions, the ß-galactosidase activitiesdriven by the two DNA fragments were quite consistent as indicatedby the standard deviations.

View larger version (13K):
[in this window]
[in a new window]
FIG. 3. ß-Galactosidase activity of the promoterless lacZ gene and two different ldh-lacZ fusions in M. pneumoniae. (A) Single colonies of M. pneumoniae after transformation with pGP353, pGP354, or pGP364 on M. pneumoniae agar plates containing 150 µg/ml X-Gal and 80 µg/ml gentamicin. (B) ß-Galactosidase activity in crude extracts of M. pneumoniae that had been transformed with either pGP353, pGP354, or pGP364. Cultures were grown in 10 ml of modified Hayflick medium with 80 µg/ml gentamicin for 7 days at 37°C and assayed for ß-galactosidase activity. ß-Galactosidase activities were determined at 28°C for five individual clones (GPM53/1 through GPM53/5) that were obtained for each plasmid, and average values are shown as bars. Error bars indicate standard deviations.

View larger version (45K):
[in this window]
[in a new window]
FIG. 4. Agarose gel electrophoresis to confirm the presence of the promoterless lacZ gene in strain GPM53/1 (lane 1) and of the ldh-lacZ fusions of pGP354 and pGP364 in strains GPM54/1 (lane 3) and GPM64/1 (lane 5) by colony PCR. Oligonucleotides for the detection of the promoterless lacZ gene in strain GPM53/1 were SH44 and SH45; for the detection of the ldh promoter lacZ fusions in strains GPM54/1 and GPM64/1, the oligonucleotides SH45 and SH46 were used. PCRs with the plasmid pGP353 (lane 2), pGP354 (lane 4), or pGP364 (lane 6) as template using the respective oligonucleotides served as controls. The colony PCR was performed with all five individual clones from each transformation and gave the same results as those shown here.

This study demonstrates that lacZ fusions can be a useful toolfor the analysis of promoter fragments of M. pneumoniae genesin vivo. They will help to study regulatory events at the molecularlevel using promoter mutants. Moreover, the blue-white screencan be helpful in the isolation of trans-acting transposon mutantsthat affect the expression of the gene of interest.

ACKNOWLEDGMENTS

We are grateful to Richard Herrmann for the generous gift ofplasmid pMT85 and to Julia Busse for excellent technical assistance.Stephan Seiler is acknowledged for his help with photographingof Mycoplasma colonies.

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