|Home | About | Journals | Submit | Contact Us | Français|
The P1, P40, and P90 proteins of Mycoplasma pneumoniae and the MgPa and P110 proteins of Mycoplasma genitalium are immunogenic adhesion proteins that display sequence variation. Consequently, these proteins are thought to play eminent roles in immune evasive strategies. For each of the five proteins, a similar underlying molecular mechanism for sequence variation was hypothesized, i.e., modification of the DNA sequences of their respective genes. This modification is thought to result from homologous recombination of parts of these genes with repeat elements (RepMp and MgPar elements in M. pneumoniae and M. genitalium, respectively) that are dispersed throughout the bacterial genome. Proteins that are potentially involved in homologous DNA recombination have been suggested to be implicated in recombination between these repeat elements and thereby in antigenic variation. To investigate this notion, we set out to study the function of the RecA homologs that are encoded by the M. pneumoniae MPN490 and M. genitalium MG339 genes. Both proteins, which are 79% identical on the amino acid level, were found to promote recombination between homologous DNA substrates in an ATP-dependent fashion. The recombinational activities of both proteins were Mg2+ and pH dependent and were strongly supported by the presence of single-stranded DNA binding protein, either from M. pneumoniae or from Escherichia coli. We conclude that the MPN490- and MG339-encoded proteins are RecA homologs that have the capacity to recombine homologous DNA substrates. Thus, they may play a central role in recombination between repetitive elements in both M. pneumoniae and M. genitalium.
Mycoplasma pneumoniae is a human pathogen that causes a range of respiratory infections, including atypical pneumonia. This bacterium causes up to 40% of all community-acquired pneumonias and approximately 18% of cases requiring hospitalization for children (for reviews, see references 1 and 35). The closest known relative of M. pneumoniae on the basis of sequence similarity is Mycoplasma genitalium, which is an etiological agent of various diseases of the human reproductive tract, such as urethritis (see reference 28 for a review).
Although M. pneumoniae and M. genitalium represent the smallest self-replicating species regarding both cellular dimensions and genome size, it is interesting to note that a significant part of their genomes consists of repeated DNA elements. In M. pneumoniae strain M129 (6, 12), approximately 8% of the 816-kb genome is composed of variants of four different types of repetitive DNA elements (RepMP1, RepMP2/3, RepMP4, and RepMP5) (29, 34, 36), while in M. genitalium strain G-37T, 4% of the 580-kb genome consists of MgPa repeats (or MgPar sequences) (10, 26, 27). Common features of the two types of repetitive elements are that (i) their representatives are similar but not identical in sequence and (ii) they are also contained in open reading frames (ORFs) encoding surface-exposed, antigenic proteins. Among these proteins is the M. pneumoniae P1 protein, which plays an essential role in bacterial adhesion to host cells (2). The ORF encoding the P1 protein, MPN141, contains both a RepMP4 element and a RepMP2/3 element. It has been hypothesized that homologous recombination between these RepMP elements with elements elsewhere in the genome could generate sequence changes within MPN141. These changes could subsequently lead to amino acid sequence variation of the antigenic P1 protein and thereby contribute to bacterial evasion of the host's immune system. Strong evidence, albeit indirect, for recombination among RepMP sequences has come from the observation that all naturally occurring sequence variations within the MPN141 gene originate from RepMP2/3 and RepMP4 elements located at distant sites within the M. pneumoniae genome (32). In addition, several RepMP2/3 and RepMP4 elements outside of the MPN141 gene, as well as RepMP1 elements, appeared to have recombined in a number of strains (24, 32). In each of these cases, RepMP sequence information seemed to be copied from the donor site to the recipient site in a unidirectional fashion, which is indicative of a gene conversion-like mechanism of homologous DNA recombination (18, 32).
In analogy, the M. genitalium MgPar elements are thought to provide a pool of sequence variation of the mgpB and mgpC genes (16). These genes encode the proteins MgPa and P110, respectively, which are antigenic proteins involved in host cell attachment (3, 7, 13). Interestingly, recombination between MgPar sequences in M. genitalium appeared to be mediated by reciprocal homologous DNA recombination events rather than by unidirectional gene conversion-like processes (15, 16).
In order to elucidate the molecular mechanisms that underlie recombination between RepMP elements in M. pneumoniae, we previously initiated a study aimed at the identification of proteins that may be involved in DNA recombination in this bacterium (31). In the current study, we have focused on the putative enzymes from both M. pneumoniae and M. genitalium that may play central roles in homologous DNA recombination and DNA repair (for a review, see the work of Carvalho et al. ). These enzymes, which are encoded by the M. pneumoniae MPN490 ORF (12) and the M. genitalium MG339 ORF, show significant sequence similarity with RecA proteins from other organisms. Here we have shown that the proteins encoded by MPN490 and MG339 promote recombination between homologous DNA substrates and may therefore play a central role in recombination between RepMP and MgPar elements in M. pneumoniae and M. genitalium, respectively.
Bacterial DNA was purified from cultures of M. pneumoniae and M. genitalium as described before (31). Neither the MPN490 ORF nor the MG339 ORF contains any UGA codons (encoding tryptophan in mycoplasmas), which would lead to truncated protein products upon expression of these ORFs in Escherichia coli. The MPN490 ORF (12) was amplified by PCR as follows. The PCR mixture (50 μl) contained 0.3 μM of the oligonucleotide primer RecA_FW (5′-GGTCGTCATATGGTACAAAAA-GAAATGATTAA-3′; the sequence in bold indicates a unique NdeI restriction endonuclease recognition site, which overlaps with the translation initiation codon [underlined] of MPN490), 0.3 μM of primer RecA_RV (5′-GCAGCCGGATCCCTAGCTCGCTGGGGGTGG-3′; the sequence in bold indicates a unique BamHI restriction site, which is adjacent to the translation termination codon [anticodon underlined] of MPN490), 10 ng of M. pneumoniae genomic DNA, 0.2 mM of each deoxynucleoside triphosphate, 1 U of Pfu DNA polymerase (Fermentas), and 1× Pfu buffer containing MgSO4 [20 mM Tris-HCl (pH 8.8), 10 mM (NH4)2SO4, 10 mM KCl, 0.1% Triton X-100, 0.1 mg/ml bovine serum albumin, and 2 mM MgSO4]. PCR was performed using the following conditions: 3 min at 95°C, followed by 30 cycles of 30 s at 95°C, 30 s at 50°C, and 2 min at 72°C. The resulting 522-bp PCR fragment was cloned into the pJET1/blunt cloning vector (Fermentas) using the GeneJET PCR cloning kit (Fermentas). From the generated plasmid, pJET1-MpnRecA, the MPN490 ORF was excised by digestion with NdeI and BamHI and cloned into the NdeI- and BamHI-digested expression vector pET-11c (Novagen), resulting in the plasmid pET-11c-MpnRecA. In this plasmid, the MPN490 ORF is cloned so as to express the M. pneumoniae MPN490-encoded protein (RecAMp) in its natural form.
The cloning of the MG339 ORF was performed in a fashion similar to that described for MPN490, albeit that an NdeI site (5′-CATATG-3′) within the MG339 ORF (10) was first converted into the sequence 5′-CCTATG-3′. This modification, which does not have an effect on the coding content of the MG339 ORF, was carried out in order to facilitate downstream cloning steps. First, two separate PCRs were performed using either the primer combination MGrecApetfw (5′-CATATGGCTCAAAAAGAAATAATTAAT-3′, which includes an NdeI site [in italics] and the translation initiation codon of MG339 [underlined]) and MGRAmutrv (5′-GATTTAGCATAGGCTAAATCAAGTGcaccttc-3′), which amplifies the 5′ part of the MG339 gene up to the NdeI site, or the primer combination MGrecApetrv (5′-GGATCCTTAGCTAGCTGTTTGTTGAAATG-3′, which contains a BamHI site and the translation termination codon [anticodon underlined] of MG339) and MGRAmutfw (5′-CACTTGATTTAGCCTATGCTAAATCaattg-3′), which amplifies the 3′ part of the gene. With the exception of the sequences depicted in lowercase letters, the sequences of the primers MGRAmutrv and MGRAmutfw completely overlap. The mutated NdeI site that is introduced by these primers is indicated by the boldface letters. The two resulting PCR products were purified and used together as overlapping templates in a PCR with the primers MGrecApetfw and MGrecApetrv. The full-length PCR product was first cloned into HincII-digested pBluescript SK(−) (Stratagene). The MG339 ORF was subsequently digested from these constructs using NdeI and BamHI (which were introduced by the primers MGrecApetfw and MGrecApetrv, respectively) and ligated into the NdeI- and BamHI-digested vector pET-11c to yield the expression construct pET-11c-MgeRecA. From this construct, the M. genitalium MG339-encoded protein (RecAMg) is expressed in its natural form. The integrity of all DNA constructs used in this study was verified by DNA sequencing.
The constructs pET-11c-MpnRecA and pET-11c-MgeRecA were introduced into E. coli BL21(DE3), and the resulting strains were grown overnight at 37°C in LB medium containing 100 μg/ml ampicillin. The cultures were diluted 1:100 in 600 ml LB medium with ampicillin and grown at 37°C to an optical density at 600 nm of 0.5. Protein expression was then induced by the addition of isopropyl-β-d-thiogalactopyranoside to a final concentration of 0.3 mM. After 2 h of incubation at 30°C, the bacteria were harvested by centrifugation and stored at −20°C.
Frozen bacterial pellets were resuspended in 20 ml of buffer A (20 mM Tris-HCl [pH 7.4], 0.1 mM EDTA, 1 mM dithiothreitol [DTT]) containing 1 mg/ml of lysozyme. The suspension was sonicated on ice and clarified by centrifugation for 20 min at 12,000 × g (4°C). All subsequent purification steps were carried out either on ice or at 4°C. To the pellet, which contained the majority of the expressed RecA proteins, was added 10 ml of buffer B (50 sodium phosphate [pH 7.4], 1 M NaCl, 0.1 mM EDTA, 1 mM DTT). After resuspension of the protein pellet, the suspension was homogenized on ice using a Heidolph Potter homogenizer. Then, the suspension was rotated for 1 h, followed by centrifugation for 20 min at 12,000 × g. To the supernatant, which contained the majority of the RecA proteins, saturated ammonium sulfate was added on ice until the solution turned cloudy. The suspension was then centrifuged for 35 min at 12,000 × g. The RecA-containing supernatant was dialyzed against 2 liters of buffer C (20 mM sodium phosphate [pH 7.4], 150 mM NaCl, 0.1 mM EDTA, 10% glycerol, 1 mM DTT), after which the solution was loaded on a 2-ml Heparin Sepharose 6 Fast Flow (GE Healthcare) column previously equilibrated in buffer D (20 mM sodium phosphate [pH 7.4], 0.1 M NaCl, 0.1 mM EDTA, 10% glycerol, 1 mM DTT). The column-bound RecA protein was eluted by a 30-ml linear gradient from buffer D to buffer E (20 mM sodium phosphate [pH 7.4], 1 M NaCl, 0.1 mM EDTA, 10% glycerol, 1 mM DTT). Fractions of 1 ml (each) were collected and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The RecA-containing fractions were pooled, dialyzed against 200 ml of buffer F (20 mM Tris-HCl [pH 7.4], 0.4 M NaCl, 0.05 mM EDTA, 50% glycerol, 1 mM DTT), and stored at −20°C. Purified RecAMp and RecAMg were estimated to have a homogeneity of 95% or greater. The concentrations of the proteins used in this study refer to monomeric protein concentrations throughout this article.
Proteins were separated on SDS-polyacrylamide gels, essentially as described previously (21). After electrophoresis, gels were stained with Coomassie brilliant blue, destained in 40% methanol-10% acetic acid, and photographed using a GelDoc XR system (Bio-Rad). Digital images were processed using Quantity One 1-D analysis software (Bio-Rad).
DNA strand transfer reactions were carried out using ϕX174 DNA in a fashion similar to that described previously (5, 31, 33). Standard reaction mixtures (45 μl) contained 70 mM Tris-HCl (pH 7.6), 10 mM MgCl2, 5 mM DTT, 5 mM ATP, 0.08 U/μl pyruvate kinase (PK) preparation (type VII, from rabbit muscle [Sigma]), 3.2 mM phosphoenolpyruvate (PEP; Sigma), 1 nM single-stranded circular ϕX174 DNA, 2 nM PstI-digested, double-stranded ϕX174 DNA, 0.2 μg/μl of either RecAMp, RecAMg, or E. coli RecA (RecAEc) New England (Biolabs), and 3 ng/μl of either E. coli SSB (SSBEc; Epicentre Biotechnologies) or 6.5 ng/μl of SSBMp (31). Samples of 10 μl were taken at different time points of incubation at 37°C, and reactions were terminated by the addition of 2.5 μl of a solution of 5% SDS-75 mM EDTA (pH 8.0) and 1 μl of 10 mg/ml proteinase K, followed by incubation for 10 min at 55°C. After the addition of loading dye, the samples were separated on 0.6% agarose gels in 0.5× Tris-borate-EDTA. The gels were stained with ethidium bromide and photographed using a GelDoc XR system.
The single-stranded DNA (ssDNA)-dependent ATPase activities of the RecA proteins were determined by using a β-NAD reduced form (NADH) coupled assay on a VersaMax tunable microplate reader (Molecular Devices), as described previously (23). In short, this assay is based on the PK-catalyzed transfer of a phosphate group from PEP to ADP, yielding pyruvate. Subsequently, pyruvate is converted to lactate by lactate dehydrogenase. The latter step requires NADH, which is oxidized to NAD+. Because NADH absorbs strongly at 340 nm but NAD+ does not, the consumption of NADH in the reaction can be monitored by measuring the absorbance at 340 nm. The conversion of 1 molecule of NADH to NAD+ corresponds to the production of 1 molecule of ADP by the ATPase activity of RecA. Reactions were carried out in duplicate in a volume of 150 μl containing 20 mM potassium phosphate (pH 6.8), 1 mM MgCl2, 50 mM NaCl, 1 mM ATP, 0.15 mM NADH, 1.5 mM PEP, 20 U/ml PK, and 20 U/ml of lactate dehydrogenase and various concentrations of either E. coli RecA, RecAMp, or RecAMg. To start the reaction, ϕX174 ssDNA was added to a final concentration of 1.5 nM. The absorption of NADH was monitored at 340 nm for 30 min at intervals of 30 s.
The M. pneumoniae MPN490 ORF and M. genitalium MG339 ORF have previously been identified as putative RecA genes on the basis of sequence comparisons (10, 12, 27). Indeed, as shown in the multiple sequence alignment in Fig. Fig.1A,1A, significant similarities exist between the amino acid sequences encoded by these ORFs and the sequences of known RecA proteins from other species (Fig. (Fig.1A).1A). The similarity between the proteins predicted to be encoded by MPN490 and MG339, which will be referred to as RecAMp and RecAMg, respectively, is remarkably high (79% identity). Both the RecAMp and RecAMg proteins share 41% amino acid sequence identity with RecAEc.
The MPN490 and MG339 ORFs were amplified by PCR and cloned into the protein expression vector pET-11c. This vector allowed expression of both RecAMp and RecAMg in their native forms in E. coli. The two proteins were purified using a similar protocol. Briefly, crude bacterial lysates were generated in a low-salt buffer, followed by a centrifugation step. The RecA proteins, which were present in the pellet fraction, were extracted with a high-salt buffer, dialyzed, and subjected to heparin Sepharose affinity chromatography. Using this procedure, both RecAMp and RecAMg were purified to near-homogeneity (Fig. (Fig.1B).1B). The estimated molecular masses of the purified proteins correspond to their theoretical molecular masses of 36.9 kDa and 37.4 kDa for the 336-amino-acid RecAMp and the 340-amino-acid RecAMg, respectively.
To study the potential of RecAMp and RecAMg to catalyze DNA recombination reactions, three-strand exchange assays were performed. In these assays, the RecA protein may promote the transfer of one of the strands of a linear, double-stranded DNA (dsDNA) molecule (donor) to a complementary, circular ssDNA molecule (acceptor), resulting in a double-stranded, circular product and a linear, single-stranded product (Fig. (Fig.2A).2A). As a control, commercially available E. coli RecA (RecAEc) was used. Figure Figure2B2B shows that, like RecAEc, RecAMp readily stimulates three-strand DNA exchange in the presence of E. coli ssDNA binding protein (SSBEc). RecAMp is active at a rather narrow concentration range, with optimal reaction efficiencies at 0.1 μg/μl and 0.2 μg/μl of protein (Fig. (Fig.2B,2B, lanes 8 to 13). Activity was virtually undetectable at concentrations of 0.4 μg/μl and 50 ng/μl. RecAMg was found to have activities that were virtually indistinguishable from those of RecAMp (Fig. (Fig.2C2C).
It has previously been shown that RecAEc does not efficiently catalyze DNA strand exchange in the absence of SSB (for a review, see reference 20). However, in the presence of SSBEc (or M. pneumoniae SSB [SSBMp] ), the strand transfer activity of RecAEc is readily detectable (Fig. (Fig.2B,2B, lanes 2 to 4). To investigate the influence of SSB on the activities of RecAMp and RecAMg, these proteins were assayed in the absence or presence of either SSBEc (from a commercial source) or SSBMp (31). As shown in Fig. Fig.3A,3A, RecAMp strand transfer activity was not detectable in the absence of SSB (lanes 11 to 13), whereas activity was readily observed at as early as 15 min of incubation in the presence of either SSBEc (lane 6) or SSBMp (lane 9). The latter protein appeared to be somewhat more efficient than SSBEc in the stimulation of the strand transfer activity of RecAMp (compare lanes 5 to 7 to lanes 8 to 10 in Fig. Fig.3A).3A). However, this difference in efficiency may be due either to the difference in SSB concentrations used (3 ng/μl and 6.5 ng/μl for SSBEc and SSBMp, respectively) or to differences in purity between the SSB protein samples. Similarly to the case for RecAMp, activity of RecAMg was detected only in the presence of either SSBEc or SSBMp (Fig. (Fig.3B3B).
To determine the pH optimum and optimal Mg2+ concentration for the DNA strand exchange activity of the RecA proteins, three-strand transfer assays were performed at a range of pH values as well as Mg2+ concentrations. RecAMp demonstrated optimal activity at a pH of 7.5 (Fig. (Fig.3C).3C). Although some activity was also seen at pH 7.0, it was significantly less efficient than that at pH 7.5. At pH 6.5 and pH 8.0, recombination end products were not generated and only minor DNA species representing reaction intermediates could be observed.
Like other previously studied RecA proteins, RecAMp and RecAMg were found to require Mg2+ for activity. The optimal Mg2+ concentration for DNA strand exchange activity of RecAMp was determined to be around 10 to 15 mM (Fig. (Fig.3D,3D, lanes 6 to 13). RecAMp activity was significantly lower at Mg2+ concentrations of 7.5 mM or lower (Fig. (Fig.3D,3D, lanes 2 to 5, and data not shown) or at concentrations larger than or equal to 20 mM (data not shown). Mn2+ was unable to efficiently replace Mg2+ as a divalent cation; although reaction intermediates could be observed in the presence of 5 mM Mn2+, the full-length end products of three-strand transfer were not detected (data not shown). Similar results were obtained with the RecAMg protein (data not shown).
Both RecAMp and RecAMg were found to require ATP for DNA strand exchange activity; neither reaction intermediates nor recombination end products could be detected after incubation in the absence of ATP (data not shown). Moreover, an ATP regeneration system, in our assays supplied by PK and phospho(enol)pyruvic acid, was necessary for strand transfer reactions to run efficiently (data not shown). To further investigate the ATPase activity of the RecA proteins from both mollicute species, the ATP consumption of these proteins was measured in real time. As shown in Fig. Fig.4A,4A, RecAMp shows a concentration-dependent and ssDNA-dependent ATPase activity, which is similar to that of the E. coli RecA protein. In comparison with the latter protein, RecAMp displays a somewhat higher background level of ATP consumption in the absence of ssDNA. Again, the RecAMg protein was found to have activities similar to those of RecAMp, displaying an ssDNA- and concentration-dependent ATPase activity (Fig. (Fig.4B4B).
RecA proteins from a wide range of species, including E. coli, were shown to play crucial roles in genetic recombination and DNA repair. It is highly likely that RecAMp and RecAMg, which indeed appear to represent the homologs of RecAEc regarding both primary sequence and enzymatic activities, serve similarly vital functions in the biology of M. pneumoniae and M. genitalium, respectively. As outlined in the introduction, both proteins may also play an additional, explicit role in the pathogenesis of infection by allowing recombination between repetitive DNA elements leading up to variation of antigenic surface proteins and subsequently evasion of the host's humoral immune response. Such a mechanism of immune evasion likely constitutes a crucial part of the life cycle of both M. pneumoniae and M. genitalium and may be a major factor in the success of these species as human pathogens.
Aided by accessory proteins, such as SSB, RecAMp and RecAMg may represent the key players in the recombination between the RepMP elements in M. pneumoniae and between the MgPar elements in M. genitalium. Other putative accessory proteins involved in homologous DNA recombination in mollicutes include homologs of RuvA, RuvB, and RecU (4, 6, 10, 12). To date, only two of these accessory proteins from mollicute species have been characterized, i.e., the M. pneumoniae RuvA homolog (14) and the SSB homolog (31). The MPN535-encoded RuvA homolog (RuvAMp) was reported to bind Holliday junctions and other branched DNA structures in a manner similar to that of E. coli RuvA (14). The MPN229-encoded SSB homolog (SSBMp) was shown to selectively and efficiently bind ssDNA and stimulate E. coli RecA-promoted homologous DNA recombination (31). In this study we found, as could be expected, that SSBMp also stimulates homologous DNA recombination catalyzed by RecAMp and RecAMg. The SSB homolog from M. genitalium, which is encoded by the ORF MG091, may show similar “cross-species promiscuity,” since it displays a relatively high level of sequence similarity to SSBMp (61% identity) (31). The characterization of other proteins putatively involved in the recombination between RepMP elements in M. pneumoniae is ongoing at our laboratory.
Despite the obvious similarities between RecAMp and RecAMg in both sequence and in vitro enzymatic activities, the recombination between MgPar elements in M. genitalium and RepMP elements in M. pneumoniae appears to differ in two major aspects. First, the recombination between MgPar elements seems to occur more frequently than recombination between RepMP elements (15, 16, 24, 32). Second, recombination among MgPar elements was hypothesized to take place in a reciprocal fashion (16), whereas recombination between RepMP elements seems to occur in a nonreciprocal, unidirectional way by a mechanism reminiscent of gene conversion (18, 32). Whether these two mechanisms of recombination in M. pneumoniae and M. genitalium are mutually exclusive and whether they represent differently organized enzymatic machineries has yet to be determined. It is clear, however, that reciprocal and nonreciprocal DNA exchange processes are highly complex and cannot be discerned in the three-strand DNA transfer assay employed in this study. In addition to the in vitro characterization of (putative) recombination enzymes, as described here, the development of novel (in vivo) assays is required in order to elucidate the mechanistic differences between M. pneumoniae and M. genitalium in recombination of repetitive elements.
Although the RecAMp and RecAMg proteins are deemed to have vital roles in processes such as DNA recombination and repair, it is clear that the in vivo relevance of both proteins will still need to be determined. A so-called proteogenomic approach has indicated, however, that the RecAMp protein is expressed by M. pneumoniae strain FH in culture (17). In addition, an important role for RecA in DNA repair was previously demonstrated for two distantly related mollicute species, i.e., Acholeplasma laidlawii and Mycoplasma pulmonis. Disruption of the putative RecA genes in these species resulted in strains having a DNA repair-deficient phenotype (8, 9, 11). Nonetheless, no data have been reported on the “DNA recombinatorial potential” of these RecA-deficient strains. Thus, studies of MG339-deficient M. genitalium strains and MPN490-deficient M. pneumoniae strains will have to shed more light on the in vivo functions of the RecA proteins from these pathogens and their roles in DNA recombination in general and the recombination of repetitive elements in particular.
A.M.C.V.R. is supported by grants from the European Society for Pediatric Infectious Diseases, ZonMW, and the Erasmus MC.
Editor: A. Camilli
Published ahead of print on 8 September 2009.