INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Adherence of bacterial pathogens to eukaryotic cells is a critical step in colonization and successful infection (11). Cytadherence of pathogenic mycoplasmas is mediated by an unusual terminal structure, designated the attachment organelle, which is comprised of a complex network of interacting proteins (3, 5, 19). In Mycoplasma genitalium, a 140-kDa (MG191) surface protein was implicated as a major adhesin based on its immunological cross-reactivity with the P1 adhesin (169-kDa protein encoded by orf1627) of Mycoplasma pneumoniae (16, 22). Isolation of spontaneously arising nonadhering populations of M. genitalium that lacked MG191 reinforced its critical function in cytadherence (21). The gene encoding MG191 was subsequently cloned and sequenced, which revealed its location in an operon situated between genes encoding 29-kDa (MG190) and 114-kDa (MG192) proteins (17). The arrangement of the mg190 operon of M. genitalium structurally resembles the p1 operon of M. pneumoniae where p1 is flanked by two similar genes, orf324 (28 kDa) and orf1218 (130 kDa) (5). Further, the existence of several partial repeats of the mg191 regions in the M. genitalium genome, as observed with p1 of M. pneumoniae (25), is consistent with a role in cytadherence (7). For example, the repeat regions may serve as reservoirs to regulate the structural and functional properties of the MG191 adhesin through recombination events, which may circumvent host immune responses. In addition to MG191, M. genitalium possesses numerous genes that are homologues of M. pneumoniae genes encoding cytadherence-related proteins P30 (ORF274), HMW1 (ORF1018), HMW2 (ORF1818), and HMW3 (ORF672) (5, 12). These homologues likely exhibit important functions in the expression, assembly, positioning, and maintenance of the adhesin(s) at the M. genitalium tip organelle (3, 4).

In an attempt to further identify and characterize cytadherence-related proteins in M. genitalium, we disrupted through homologous recombination the gene encoding MG218 (190 kDa), which is a homologue of the cytadherence accessory protein, HMW2 of M. pneumoniae (9). While mutants of M. genitalium completely lacking MG218 displayed cytadherence-negative phenotypes, mutants expressing truncated MG218 (80% of MG218) were cytadherence positive. Examination of the mg218 mutants that completely lacked MG218 revealed that MG191 and MG192 proteins were posttranslationally unstable. In contrast, mg218 mutants expressing truncated MG218 exhibited stable MG191 and MG192 proteins, suggesting that MG218 plays a critical role in M. genitalium cytadherence. To further examine the specific role of MG218 in cytadherence, we compared the stability of MG191 and MG192 proteins in mg218 disruption mutants with spontaneously arising noncytadhering isogenic M. genitalium mutants either lacking MG191 (class I) or exhibiting abnormal MG191 processing (class II) (21). Unexpectedly, both spontaneous mutant classes possessed intact MG218 (10). The absence of MG191 in the class I mutants was due to posttranslational instability of MG191 and MG192 and not deletions or mutations in mg191 (10), implying that additional proteins or factors are involved in the stability and structural integrity of the MG191 and MG192 proteins.

In this report, we describe our continued effort to identify new loci that regulate mycoplasma cytadherence. The msrA (mg408) locus appears unrelated to the previously identified cytadherence loci of M. genitalium and encodes the antioxidant repair enzyme peptide methionine sulfoxide reductase (MsrA). By targeted disruption of msrA, we demonstrate that MsrA is important for the maintenance of cytadherence and virulence potential in M. genitalium.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and culture conditions. M. genitalium G37 was grown in 100 ml of SP-4 medium at 37°C for 72 h in 150-cm² tissue culture flasks (Corning, Corning, N.Y.). Surface adherent mycoplasmas were washed four times with phosphate-buffered saline (PBS) (pH 7.2) and collected by centrifugation at 20,000 × g for 20 min at 4°C. Escherichia coli strain DH5 harboring plasmids pISM2061 and pCR2.1 was grown at 37°C in Luria-Bertani (LB) broth or LB agar plates containing 100 µg of ampicillin/ml. Plasmid pISM2061 that carries transposon Tn4001 was a kind gift of C. Minion, Iowa State University, and plasmid pCR2.1 was purchased from Invitrogen, Carlsbad, Calif.

DNA manipulations. Chromosomal DNA from M. genitalium was isolated as reported earlier (8). msrA disruption constructs were made as follows. Initially, four primers, MSRA1 (5'-TTTGAAATGTATTGAAATAATGAGC-3'), MSRA2 (5'-TTTCTTCATATGCAACTTTTACAGCTTCAACAG-3), MSRA3 (5'-AAGTTGCATATGAAGAAAAGAAATTTATCT-3'), and MSRA4 (5'-TTATGAAATGGAGGACCAATCTAATAGGTC-3'), were custom synthesized based on the M. genitalium genome sequence to amplify msrA (mg408) and its flanking regions (mg407 and mg409) (13). In primers MSRA2 and MSRA3, nucleotides in bold represent modifications of the sequence to create NdeI sites. Using genomic DNA of M. genitalium as template, we amplified DNA designated fragment A (680 bp, contains part of mg408 and its upstream mg407) with primers MSRA1 and MSRA2 and DNA designated fragment B (834 bp, which contains part of mg408 and its downstream mg409) with primers MSRA3 and MSRA4. These fragments were independently cloned into the pCR2.1 vector, and the orientation of insert DNAs was assessed by digesting the miniprep plasmids with NdeI and EcoRV. Clone pMSRAU, which contained fragment A and did not release insert DNA upon cutting with NdeI and EcoRV, and clone pMSRAD, which contained fragment B and released insert DNA upon cutting with NdeI and EcoRV, were selected. In the latter case, the released fragment B from plasmid pMSRAD was separated on agarose gels, eluted, and cloned into NdeI- and EcoRV-cut pMSRAU to result in plasmid pMSRA1. This plasmid has a unique NdeI site within the coding region of msrA. The final msrA disruption construct was created by cutting plasmid pMSRA1 with NdeI, treating with Klenow, and then ligating the resultant fragment to the 2.5-kb DNA fragment containing the gentamicin resistance gene to result in plasmid pMSRA2. The 2.5-kb gentamicin-resistance gene fragment was obtained from plasmid pISM2061 by cutting with HindIII (9). Plasmid pMPMSRA, an E. coli overexpression construct, was generated by modifying the ends of M. pneumoniae msrA in PCR with primers MPMF1 (5'-TAAATTTAGCATATGAAACAAATC-3'; NdeI site is underlined) and MPMR2 (5'-CATCATTAAGGGATCCCTGTTTGT-3'; BamHI site is underlined). The product was cut with NdeI and BamHI and inserted into similarly cut T7 expression vector pET16-b (Novagen, Madison, Wis.). Sequencing of DNA fragments was performed using an automated cycle sequencing system (Applied Biosystem model 373, Center for DNA Technology Core Facility of University of Texas Health Science Center at San Antonio [UTHSCSA]) with fluorescent terminators and by an Amplicycle sequencing kit (Perkin-Elmer, Branchburg, N.J.) using a PCR machine. Southern hybridization was performed under high-stringency conditions at 68°C. Probes for Southern hybridization were labeled with [-³²P]dCTP by using the random primer method (1).

Electroporation and transformation. Transformation of M. genitalium with the gentamicin resistance gene cloned in the suicidal plasmid pMSRA2 was performed by electroporation as described earlier (24). Briefly, M. genitalium cells grown to log phase in 100 ml of SP-4 medium were washed three times with cold electroporation buffer (8 mM HEPES and 272 mM sucrose, pH 7.4), scraped, suspended in the same buffer, and pelleted by centrifugation. Mycoplasma cells were suspended in 1 ml of electroporation buffer, and 100 µl (10⁸ cells) was aliquoted to each electroporation cuvette (0.2-cm electrode gaps; Bio-Rad, Hercules, Calif.). Plasmid DNA (30 µg) in 10-µl volumes was mixed with mycoplasma cells in identical cuvettes, and control cuvettes received only 10 µl of electroporation buffer. After a 15-min incubation on ice, individual cuvettes were transferred to a Gene Pulser Electroporator cuvette holder for electroporation. The settings for electroporation were 2.5 kV with a resistance of 100  and a capacitance of 25 µF. After the addition of 0.9 ml of SP-4 medium to each cuvette, the cuvettes were incubated for 2 h at 37°C, and mycoplasmas were plated on SP-4 agar supplemented with 100 µg of gentamicin/ml.

HA test. The ability of msrA mutants to adhere to sheep erythrocytes was tested by flooding SP-4 plates containing mycoplasma colonies with 2 ml of diluted (1:50) sheep erythrocytes (40% erythrocytes in 60% Alsever's solution; Bio-Whittaker, Walkersville, Md.). Hemadsorption (HA) was monitored after a 1-h incubation at 37°C by washing mycoplasma colonies repeatedly with PBS and observing HA microscopically.

Disk inhibition assay. The sensitivity of M. genitalium strains to oxidants was tested by a disk inhibition assay. M. genitalium strains were grown to log phase in SP-4 broth and diluted 10-fold with fresh SP-4 broth. One hundred microliters of diluted cultures was plated on SP-4 plates. Filter disks (7-mm diameter) were impregnated with 10-µl volumes of different concentrations of H₂O₂, paraquat (methyl viologen), or t-butyl hydroperoxide. Plates were incubated for 5 days, and the zone of growth inhibition was measured.

Overexpression and purification of M. pneumoniae msrA. Since M. genitalium msrA has two TGA codons within the coding region and overexpression of this gene in E. coli would result in truncation, we overexpressed M. pneumoniae msrA, which contains no TGA codons and exhibits 77% nucleotide sequence identity with M. genitalium msrA. Thus, plasmid pMPMSRA was constructed as described in the previous section to overexpress M. pneumoniae MsrA in E. coli. Expression of MsrA with this construct produced a fusion with 10 histidines (His₁₀ tag) located at the N' terminal region of MsrA, which enabled its purification by nickel affinity column chromatography. Plasmid pMPMSRA was transformed into E. coli strain BL21(DE3)LysS, and individual colonies were inoculated into LB broth (2 ml) for overnight culture. One milliliter of each culture was inoculated into 200 ml of fresh LB broth, and E. coli cells were grown at 37°C to an optical density of 0.4 at A₆₀₀. Isopropyl--D-thiogalactopyranoside (IPTG; 1 mM) was added, and incubation continued for another 2 h at 37°C. Bacteria were harvested by centrifugation at 3,000 rpm, resuspended (5 ml/g [wet weight]) in Tris-HCl buffer (20 mM, pH 7.9) containing 500 mM NaCl, and lysed by three passages through a French press cell at 4°C. All subsequent steps were carried out at 4°C. Lysates were first clarified by centrifugation at 2,000 × g for 15 min, and then supernatants were centrifuged for an additional 30 min at 20,000 × g. The resulting supernatants contained most of the His₁₀-MsrA as assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Fig. 1A). Slurries of Ni-nitrilotriacetic acid (NTA) resin (2 ml Ni-nitrilotriacetic acid; Qiagen, Valencia, Calif.) and supernatants were stirred for 3 h, then packed into a column, and washed with 30 ml of wash buffer (20 mM Tris-HCl [pH 7.9] containing 60 mM imidazole and 500 mM NaCl). Finally, His₁₀-MsrA was eluted with 3 ml of elute buffer (20 mM Tris-HCl [pH 7.9] containing 300 mM imidazole and 500 mM NaCl).

View larger version (67K): [in this window] [in a new window]   FIG. 1.   SDS-PAGE and Western blot profiles. (A) Overexpression and purification of M. pneumoniae MsrA. Lane 1, molecular size markers; lane 2, E. coli BL21 cells harboring M. pneumoniae msrA overexpression construct pMPMSRA before the addition of IPTG; lane 3, E. coli BL21 cells harboring the M. pneumoniae msrA overexpression construct pMPMSRA 2 h after the addition of 1 mM IPTG; lane 4, E. coli cells overexpressed M. pneumoniae MsrA after purification with a nickel affinity column (Ni-NTA agarose). MsrA* indicates the overexpressed and purified His10 tag MsrA of M. pneumoniae. (B) Western blotting showing reactivity of anti-M. pneumoniae MsrA antibodies. Lane 1, wild-type M. genitalium strain G37; lane 2, msrA mutant M. genitalium strain MS5; lane 3, overexpressed and purified His10 tag MsrA protein of M. pneumoniae. MsrA* indicates the overexpressed and purified MsrA of M. pneumoniae, and MsrA indicates the native MsrA protein of M. genitalium.

Production of M. pneumoniae MsrA antiserum. Polyclonal antiserum against M. pneumoniae MsrA was produced in a New Zealand White rabbit immunized with 100 µg of purified His₁₀-MsrA in complete Freund's adjuvant. The immunized animal was boosted twice with 100 µg of purified protein suspended in incomplete Freund's adjuvant at 2-week intervals. One week after the second boost, antiserum was collected and tested against M. pneumoniae and M. genitalium total proteins to confirm its reactivity.

Virulence studies. The virulence potential of M. genitalium strains was studied in Syrian golden hamsters based on mycoplasma survival in lungs, which serves as an indicator of pathogenicity (6). M. genitalium cells were grown in SP-4 medium as described earlier, harvested, centrifuged, and resuspended in PBS. Mycoplasma suspensions were passed through 26-G needles to disperse clumps of organisms and diluted in PBS to desired concentrations. Hamsters were anesthetized (2), and approximately 10⁹ CFU of M. genitalium in 50 µl of PBS were inoculated intranasally. Seven days postinfection, hamsters were sacrificed and lungs were removed and homogenized. Homogenates were serially diluted, inoculated into 2 ml of SP-4 medium, and incubated at 37°C for 7 to 10 days. Mycoplasma growth was expressed as color change units (CCU)/gram of wet tissue (14).

SDS-PAGE and immunoblottings. SDS-PAGE and immunoblottings were performed by following standard protocols (1). Proteins were visualized by staining with Coomassie brilliant blue. Proteins transferred to nitrocellulose membranes were probed with polyclonal rabbit antiserum raised against M. pneumoniae MsrA. Alkaline phosphatase-conjugated anti-rabbit antibodies were used as second antibodies, and color development was performed using standard procedures (1).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Disruption of M. genitalium msrA. The published M. genitalium genome sequence indicates that the gene encoding MsrA, designated mg408 (predicted protein size of 18.4 kDa, 157 amino acids), is located between genes encoding enolase (mg407) upstream and a putative regulatory protein (mg409) downstream (12). The amino acid sequence of M. genitalium MsrA is 79% identical to the MsrA sequence of M. pneumoniae, and these mycoplasma MsrAs exhibit significant homology with MsrAs of both prokaryotes and eukaryotes (20). We first reported targeted gene disruption in M. genitalium by using suicidal plasmids (9). In a similar approach, plasmid construct pMSRA2 described in Materials and Methods was used to disrupt msrA in order to evaluate its function in M. genitalium. Electroporation of disruption construct pMSRA2 into M. genitalium cells resulted in the identification of gentamicin-resistant transformants. Eight transformants that were consistently gentamicin resistant were subjected to single-cell cloning (27). These clones were grown to late log phase in SP-4 broth, and protein extracts from these clones were screened by immunoblotting using anti-MsrA (M. pneumoniae) antiserum to further identify msrA mutants. All eight transformants were negative for MsrA, and a representative profile is shown in Fig. 1B.

View larger version (90K): [in this window] [in a new window]   FIG. 2.   Southern hybridization profiles of DNA from M. genitalium strains. G37, wild-type M. genitalium; MS5, M. genitalium msrA mutant. Genomic DNA from G37 and MS5 were digested with restriction enzymes EcoRV (E), StyI (S), and EcoRV and StyI (E+S) together. Southern blots were probed with a 1.5-kb M. genitalium msrA gene fragment (A) and a 2.5-kb gentamicin resistance gene fragment (B).

View larger version (13K): [in this window] [in a new window]   FIG. 3.   Schematic representation of msrA locus in M. genitalium strains G37 (wild type) and MS5 (msrA mutant). Stippled boxes represent msrA region, a lightly hatched box represents the gentamicin resistance gene, and open boxes represent flanking regions of mg218 locus. The different sizes of EcoRV and StyI fragments observed in Southern hybridization are represented below the msrA locus of each strain. EI, EcoRI, EV, EcoRV; H, HindIII; N, NdeI; St, StyI; X, XbaI.

M. genitalium adherence to erythrocytes. A previous report indicated that MsrA influences the cytadherence capacity of pathogenic bacteria (30). Since cytadherence is a basic requirement for successful colonization by mycoplasmas and HA serves as an indicator system of mycoplasma-target cell surface parasitism, we plated M. genitalium wild-type G37 and msrA mutant MS5 strains on SP-4 plates for 5 to 7 days at 37°C and monitored HA using sheep erythrocytes. Figure 4 compares the erythrocyte adherence patterns displayed by wild-type and msrA mutant strains. HA with wild-type M. genitalium colonies was complete and uniform, whereas HA with msrA mutant MS5 colonies was incomplete and patchy, indicating that the lack of MsrA significantly affected HA. Furthermore, differences in HA existed among smaller and larger colonies of mutant MS5 reflected in more extensive adsorption of erythrocytes to the former. Since the larger colonies were fully established and the smaller ones were still growing, the differences in HA between these colonies may be age related. Other msrA mutants exhibited HA patterns similar to that of MS5.

View larger version (78K): [in this window] [in a new window]   FIG. 4.   Mycoplasma HA assay using sheep erythrocytes. (A) M. genitalium wild-type strain G37; (B) M. genitalium msrA mutant strain MS5.

Survival of M. genitalium in hamsters. We inoculated hamsters intranasally with wild-type M. genitalium and msrA mutant strain MS5 in order to assess the ability of M. genitalium to colonize hamster lungs. After 7 days postinfection, mycoplasma growth (0.54 × 10⁵ CCU/gram of wet tissue) was readily detected in lungs infected with wild-type G37. However, no mycoplasma growth was observed in lungs infected with mutant MS5, indicating that MsrA affects the survival of M. genitalium in vivo.

Sensitivity to oxidative stress. Although MsrA is not a component controlled by the classical oxidative stress response system regulators oxyR and soxRS in E. coli, its role in defense against oxidative stress is well established (23). Since mycoplasmas, particularly M. genitalium and M. pneumoniae, lack the classical oxidative stress response system, we presumed that MsrA played an effective role in defending mycoplasma cells against oxidative stress. Therefore, we tested the sensitivity of msrA mutant MS5 and its wild-type parent G37 to oxidative radicals in a disk inhibition assay. As seen in Table 1, MS5 was much more sensitive to H₂O₂ and t-butyl hydroperoxide than wild-type G37. However, neither G37 nor MS5 was affected by paraquat.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

M. genitalium was first isolated from the urine of two male patients with nongonococcal urethritis (28) and subsequently, along with M. pneumoniae, from throat specimens of pneumonia patients (4) and from synovial fluid of a patient with polyarthritis (29). Although Jensen et al. (18) recently used a cell culture system to isolate M. genitalium from urethral specimens, routine isolation of this fastidious pathogen from humans has been very difficult. Nevertheless, mounting PCR evidence reinforces its association with urethritis and other sexually transmitted diseases (26). With a limited genome size of 580 kb (12), M. genitalium is the smallest self-replicating microorganism reported to date. M. pneumoniae, which causes primary atypical pneumonia in humans, is closely related genetically to M. genitalium and has a genome size of 816 kb (15). Both Mycoplasma species are limited metabolically and are deficient in genes common to other pathogenic bacteria, particularly genes related to cell wall synthesis, iron acquisition, oxidative stress, and two-component regulatory systems that play critical roles in pathogenic mechanisms. How mycoplasmas circumvent host immune responses and establish infections and associated pathologies remains unclear. Nonetheless, targeted disruption of specific mycoplasma genes permits selective assessment of potential virulence determinants and pathogenic mechanisms.

In this study, we show that msrA disruption mutants of M. genitalium exhibit reduced biological and pathogenic activities, such as decreased HA and survival in hamster lungs and increased sensitivity to H₂O₂ killing. Although these effects could be due to polar effect on genes adjacent to msrA (mg408), like mg407 and mg408, reverse transcription-PCR (RT-PCR) analysis (data not shown) showed that the expression of these genes were not affected in M. genitalium mutant strain MS5. MsrA catalyzes the reversible oxidation reduction of methionine sulfoxide to methionine and is a highly conserved protein (20). Methionine in proteins can be oxidized by biologically reactive oxygen intermediates, such as superoxide, hydrogen peroxide, and hydroxyl radicals, which are metabolic by-products. Proteins with oxidized methionines lose biological activity, which can be restored by MsrA (23). One explanation for reduced HA of M. genitalium msrA mutants may be the loss of biological activity of proteins involved in cytadherence. As mentioned earlier, M. genitalium cytadherence is a complex event involving numerous surface membrane adhesins and adherence-related proteins. The major M. genitalium cytadhesin MG191 (i.e., P140) possesses 13 methionine residues, many of which may be exposed externally and vulnerable to oxidation by exogenous oxidative agents. Consistent with the possible role of MsrA in virulence, varied effects of MsrA on adherence of pathogenic bacteria have been reported (25). For example, in Streptococcus pneumoniae cells, loss of msrA reduced bacterial adherence, which was due to defects in surface ligands responsible for binding to eukaryotic cells. In enteropathogenic E. coli (EPEC), loss of msrA decreased type I fimbriae-mediated hemagglutination, and restoration of MsrA activity by the introduction of a plasmid containing msrA in EPEC returned hemagglutination activity to wild-type levels (30). In contrast, msrA mutants of Neisseria gonorrhoeae exhibited hyperpiliated and hyperadherent phenotypes. Although this observation appears to contradict the previous explanation of adhesin impairment by oxidation of methionines, the effects of MsrA may be at different levels in N. gonorrhoeae. For example, gonococcal proteins impaired in msrA mutants may control pilin expression, leading to overexpression of pilins and increased cytadherence. Consistent with this hypothesis, MsrA in N. gonorrhoeae is encoded by pilB, which is part of the pilA-pilB locus and which regulates transcription of the expression locus of pilin (30). Interestingly, the msrA gene of M. genitalium was initially named pilB because of its close homology with pilB of N. gonorrhoeae.

Both M. genitalium and M. pneumoniae lack antioxidants like catalase and superoxide dismutase, and MsrA may serve as a substitute for these enzymes. In the disk inhibition assay, msrA disruption mutants of M. genitalium exhibited hypersensitivity to H₂O₂ and t-butyl hydroperoxide. Paraquat, an uncoupler of oxidative phosphorylation, which leads to superoxide production in vivo, was without effect. Similarly, msrA mutant strains of E. coli (23) and Erwinia chrysanthemi (13) exhibited hypersensitivity to H₂O₂. These observations indicate that MsrA directly or indirectly plays a role in regulating oxidative stress responses of these bacteria. Mycoplasmas must possess mechanisms to resist the effects of exogenous as well as endogenous oxidants in vivo. Thus, it appears that MsrA represents a class of enzymes involved in protein maintenance that critically impacts on the survival and disease potential of pathogenic mycoplasmas.