Delineation of immunodominant and cytadherence segment(s) of Mycoplasma pneumoniae P1 gene

Background Adhesion of Mycoplasma pneumoniae (M. pneumoniae) to host epithelial cells requires several adhesin proteins like P1, P30 and P116. Among these proteins, P1 protein has been inedited as one of the major adhesin and immunogenic protein present on the attachment organelle of M. pneumoniae. In the present study, we scanned the entire sequence of M. pneumoniae P1 protein to identify the immunodominant and cytadherence region(s). M. pneumoniae P1 gene was synthesized in four segments replacing all the UGA codons to UGG codons. Each of the four purified P1 protein fragment was analyzed for its immunogenicity with anti-M. pneumoniae M129 antibodies (Pab M129) and sera of M. pneumoniae infected patients by western blotting and ELISA. Antibodies were produced against all the P1 protein fragments and these antibodies were used for M. pneumoniae adhesion, M. pneumoniae adhesion inhibition and M. pneumoniae surface exposure assays using HEp-2 cells lines. Results Our results show that the immunodominant regions are distributed throughout the entire length of P1 protein, while only the N- and C- terminal region(s) of P1 protein are surface exposed and block cytadhesion to HEp-2 cells, while antibodies to two middle fragments failed to block cytadhesion. Conclusions These results have important implications in designing strategies to block the attachment of M. pneumoniae to epithelial cells, thus preventing the development of atypical pneumonia.


Background
Mycoplasmas are the smallest known self-replicating prokaryotes originally isolated from bovine pleuropneumonia and are also referred as pleuropneumonia like organisms (PPLO). A key characteristic of mycoplasma is the lack of a cell wall, which allows exchange of different components between the host membrane and the M. pneumoniae membrane after adhesion [1,2]. M. pneumoniae is a human pathogen that colonizes the ciliated upper and lower respiratory tract, causing atypical pneumonia. M. pneumoniae is also found to be associated with other respiratory tract infections such as tracheobronchitis, bronchiolitis, croup, Acute Respiratory Distress Syndrome (ARDS), Guillain-Barre Syndrome (GBS), stroke and less severe upper respiratory tract infections in older children as well as in young adults [3][4][5][6][7]. Adherence of M. pneumoniae to the human host respiratory epithelium is a prerequisite for the colonization and subsequent induction of disease [4,8]. It attaches to ciliated epithelial cells in the respiratory tract, where it induces ciliostasis that protects the M. pneumoniae from removal by the mucociliary clearance mechanism of the host [9]. M. pneumoniae is elongated and consists of a longer tail-like rear end, a thicker body part and a frontal attachment organelle.
Cytadherence requires a complex interaction of several M. pneumoniae proteins present on the attachment organelle, including the adhesins P1 (170 kDa), P30 (30 kDa), and P116 (116 kDa) and proteins HMW1 to HMW3, as well as proteins A, B and C [4,[10][11][12][13][14][15]. Protein P1 and P30 appear to be directly involved in receptor binding [8,16]. The HMW proteins and proteins A, B, and C are accessory proteins as they are not adhesins, but are required for proper attachment. The P1 protein, which is mainly concentrated at the tip of apical organelle, is one of the major adhesins in M. pneumoniae as mutants lacking the P1 protein lose cytadherence and virulence capabilities [17,18]. In addition, treatment of M. pneumoniae infection with anti-P1 antibodies has been shown to effect the gliding speed of M. pneumoniae, thus hampering the mobility of the bacterium and possibly its ability to find suitable host adhesion receptors [19]. Besides its role in M. pneumoniae cytadherence, P1 antigen is an important immunogen and is also being developed as defined and specific antigen for the serodiagnosis of M. pneumoniae infection [20]. Previous reports and we have shown that a C-terminal region of P1 antigen can comparably diagnose M. pneumoniae infection taking the Serion-Virion ELISA as the standard [14,21]. Serum samples from patients suffering from M. pneumoniae infection have also been shown to bind the peptide fragments located in the middle of the~170 kDa P1 antigens [22].
Since P1 is one of the major surface molecules on the apical organelles of M. pneumoniae, a number of studies have been performed to determine its immunogenicity as well as to characterize its role in adhesion/cytadherence. Using λgt11 recombinant DNA expression library of M. pneumoniae, Dallo et al. for the first time identified cytadherence (epitopes) at the C-terminal region of P1 gene [23]. Subsequently, in two independent studies based on topological mapping of the P1 binding sites, Gerstenecker et al. and Opitz et al. identified adherence associated region(s) across the length of P1 gene [11,24]. Jacobs et al. further defined immunodominant epitopes of 338 amino acids between leucine 801 and leucine 1139 residues [25]. In 2002, Svenstrup et al. expressed P1 fragments lacking the tryptophan codon which codes for a stop codon in M. pneumoniae and identified adhesion epitopes in the C-terminal part of M. pneumoniae P1 gene using monospecific antibodies [14].
Although these above mentioned studies identified few adhesion/cytadherence segment(s) in M. pneumoniae P1 protein, a systematic study defining the region(s) involved in these processes across the entire length of P1 protein is lacking, therefore leading to contradicting results. One of the main reasons for the lack of a systematic study is the presence of 21 UGA codons, which makes it difficult to express this protein or its fragments in heterologus systems such as Escherichia coli or mammalian systems. To circumvent this problem, PCR-based site-directed mutagenesis may have been one of method to replace TGA codons in P1 gene as mentioned by Hames et al. [26], but we decided to synthesize the entire P1 gene into four different fragments by codon optimization. This included the N-terminal (P1-I) fragment, two middle fragments P1-II and P1-III and a C-terminal (P1-IV) fragment, which have been suggested to be immunodominant and to act as adhesins [14,21,25,27]. All these fragments were cloned and expressed in an E. coli system [28][29][30]. The immunological and cytadherence characterization of all the four P1 protein fragments identified specific cytadherence regions. These results will enable to define strategies for the development of drug/vaccine against M. pneumoniae infection.

Results
Cloning, expression and purification of P1 gene fragments Four fragments of the M. pneumoniae P1 gene, i.e., P1-I, P1-II, P1-III, & P1-IV ( Figure 1), were amplified by PCR, cloned in expression vector pET28b and expressed in E. coli BL21(DE3) cells. The expressed proteins were analyzed on SDS-PAGE. As shown in Figure 2A, four proteins of molecular weights:~39 kDa,~38 kDa,~73 kDa, and~43 kDa were induced and they were mainly expressed in inclusion bodies. The expressions of recombinant proteins were further confirmed by western blot analysis using anti-6XHis antibody ( Figure 2B i & ii). The expressed proteins were purified up to near homogeneity on a Ni 2+ -NTA column ( Figure 2C). Fractions that contained single band for each of the recombinant protein were pooled, dialyzed and further characterized. The expressed and purified proteins reacted nicely with anti-6XHis antibody ( Figure 2D).
Recombinant rP1-I, rP1-II, rP1-III and rP1-IV proteins are immunogenic High antibody responses were seen against each of the four recombinant proteins. The time course response for each of the recombinant proteins showed that the antibody titers gradually increased after first and second booster and peaked after the second boost. An additional figure file [see Additional file 1] shows the time dependent response for recombinant P1-I protein.
Almost similar antigenic responses were observed for other three P1 protein fragments (data not shown). The end point titers for each protein were > 1 × 10 5 . Western blotting for all the four recombinant proteins with their respective antibodies confirmed the specificity of each antibody. All these antibodies showed major reactivity with~170 kDa band of P1 protein in M. pneumoniae lysate by ELISA ( Figure 3B) and western blotting. Anti-P1 antibodies also reacted with few additional bands in M. pneumoniae lysate. These additional bands probably represent the degraded P1 protein bands ( Figure 3A). No cross reactivity was observed between each of the four antibodies ( Figure 3C & 3D). Almost similar reactivity was observed with two other P1 protein fragments rP1-II & rP1-III (data not shown). These results indicated that all the four P1 protein fragments are immunogenic and antibodies are specific as they only recognized the corresponding protein fragment. Prebleed and control rabbit sera showed no reactivity with any of the recombinant protein fragments. An additional figure file [see Additional file 2] shows the reactivity of each protein fragment with pre-bleed sera. M. pneumoniae adhesion and surface exposure assays reveal that P1-I and P1-IV regions are surface exposed.
For the adhesion assay, HEp-2 cells were infected with M. pneumoniae and methanol fixed before exposing them with each of the four anti-P1 antibodies; Pab (rP1-I), Pab (rP1-II), Pab (rP1-III), and Pab (rP1-IV) antibody. The bound antibodies were detected with an FITC-conjugated goat anti-rabbit immunoglobulin. As shown in Figure 6 (A-E), Indirect immunofluorescence microscopy analysis showed that the antibodies, Pab (rP1-I and Pab (rP1-IV were able to identify M. pneumoniae bound to the HEp-2 cells, while other two antibodies, Pab (rP1-II) and Pab (rP1-III) failed to identify the bound organism to HEp-2 cells.
To detect the accessibility of the antibodies on the surface of the cytadhering M. pneumoniae, the primary antibodies were added before fixation with methanol. The bound primary antibodies were detected with FITCconjugated goat anti-rabbit IgG antibody followed by immunofluorescence microscopy. As seen in the case of adhesion detection assay, only the antibodies Pabs, rP1-I and rP1-IV were able to detect cytadhering M. pneumoniae, while no fluorescence was observed when antibodies Pabs, (rP1-II) and (rP1-III) were used ( Figure 6 (F-J). The inhibition was maximum at highest concentration of antibody (1:50) and inhibition decreased as concentration of antibodies decreased and almost no inhibition were seen with the minimum concentration of antibody (1:500 dilution). In an independent experiment, we also performed DAPI staining to confirm adhesion inhibition by Pab (rP1-I) and Pab (rP1-IV) antibodies [see Additional file 3]. Importantly, antibodies; Pab (rP1-II) and Pab (rP1-III) failed to block the M. pneumoniae adhesion to HEp-2 cells even at the maximum antibody concentration (1:50 dilution) ( Figures 7M & N). Taken together, these results suggested that P1-I and P1-IV regions of M. pneumoniae P1 protein are surface exposed and are involved in cytadherence.

Discussion
The human respiratory pathogen M. pneumoniae adheres to erythrocytes/respiratory epithelial cells. P1 has been shown to be a major adhesion protein [31][32][33][34]. A number of studies using synthetic peptides and monoclonal antibodies against the native P1 protein have illustrated that the P1 epitopes are involved in the adhesion and immunerecognition; however a complete topological mapping of P1-adhesin is still lacking [12,25,27,35]. In the present study, we segmented the entire P1 gene in four regions; P1-I (1069 bp), P1-II (1043 bp), P1-III (1983 bp) & P1-IV (1167 bp) beginning from start residue, ATG and ending with the stop codon. These segments were synthesized by codon optimization replacing all the UGA codons to UGG codons. The P1 fragments were expressed in E. coli system and all these fragments were expressed in inclusion bodies. A protocol was developed to purify these protein fragments to near homogeneity and to obtain these proteins in large amount.
The testing of P1 protein fragments with anti-M. pneumoniae sera and sera of M. pneumoniae infected patients revealed that all these protein fragments were recognized by these sera, thereby suggesting that the immunodominant regions are distributed across the entire length of the protein. These results are in agreement with a number of previous reports that showed the presence of immunodominant regions either in the N-, middle and C-terminal segments of P1 protein [21,23,25,27]. A number of previous reports have shown the presence of immunodominant epitopes usually in the C-terminal of M. pneumoniae P1 protein [21,23,36], but few reports also showed immunodominant regions in the middle and extreme N-terminal [25,27]. A comparative summary of these results is presented in additional figure file 4 [see Additional file 4]. However, our's is the first study that systematically scanned the full P1 protein for their immunodominant and cytadherence.
Since P1 protein is considered to be the major ligand mediating attachment, we next tested the ability of the antibodies raised against the four P1 fragments for adhesion detection, surface exposure and adhesion inhibition assays to identify the cytadherence regions. Previously, a number of studies have identified a few M. pneumoniae P1 regions involved in cytadherence. Trypsinization of M. pneumoniae P1 protein and ability of various fragments or peptides so generated to block cytadherence provided first evidence for the role of P1 protein in cytadherence [4]. Baseman et al., later showed that the treatment of M. pneumoniae with protease blocked its adherence to tracheal explants which was restored when P1 was re-generated [32]. Role of M. pneumoniae P1 protein in cytadherence was further substantiated by a study where pre-treatment of M. pneumoniae with antiserum directed against the P1 protein blocked its cytadherence to hamster tracheal ring up to 80% [37]. Gerstenecker and Jacobs [11] and Opitz and Jacobs [24], showed the involvement of N-terminal, middle and C-terminal segment of M. pneumoniae (P1) as well as M. genitalium (MgPa) in cytadherence. Although a number of above mentioned studies have highlighted the role of M. pneumoniae P1 protein in cytadherence, however a systematic study spanning the entire length of P1 protein is missing. We performed a systematic analysis of surface exposure and cytadherence region(s) for M. pneumoniae P1 protein fragments spanning the entire length. Our results showed that the antibodies against P1-I and P1-IV corresponding to N-and C-terminal regions of P1 protein recognized the surface of adhesive mycoplasmas and blocked their attachment when mycoplasmas were pre-incubated with these antibodies. Based on the result presented here, it can be concluded that the adherence regions are located in the N-terminal and C-terminal regions. Interestingly, Pab (rP1-II) and Pab (rP1-III) antibodies failed to block the cytadherence. The finding of an attachment regions located in the C-terminal part of M. pneumoniae P1 protein was consistent with a number of previous studies [11,14,23,24,38,39]. Summary of the various P1 cytadherence mapping regions is presented in additional figure file 5 [see Additional file 5].

Conclusions
Present study describes a systematic approach to delineate the immunodominant and cytadherent regions across the entire length of M. pneumoniae P1 protein.
Our results showed that the immunodominant regions are present in several positions across the entire length of the M. pneumoniae P1 protein, while the N-terminal and C-terminal regions of the protein are surface exposed and antibodies to these two regions significantly block the adhesion. This data plus data from earlier observations thus confirms the functional significance for M. pneumoniae P1 protein in adhesion and immunodiagnosis. These results may have important implications in the development of tools for anti-Mycoplasma drug/ vaccine development.

Ethics statement
The protocol of this study was approved by Institutional Animal Ethics Committee (IAEC), AIIMS, New Delhi.
Human blood samples used in this study were received from an already-existing collection approved by the Institution Ethics Committee (IEC), AIIMS, New Delhi.

Mycoplasma pneumoniae, HEp-2 cells and culture conditions
The lyophilized ampoule of M. pneumoniae standard strain (M129 strain; National Collection of Type Cultures, London, United Kingdom) was reconstituted in Edward Hayflick medium containing PPLO basal broth that was supplemented with 1% glucose (Difco) as the carbon source and 0.0002% phenol red as the indicator. Tissue culture flasks (Nunc, Roskilde, Denmark) were incubated at 37°C aerobically and inspected daily. An exponential growth phase was indicated by a change in color of the medium from red to orange. Cells were harvested at this stage, washed in phosphate-buffered saline (PBS), centrifuged, and the pellet was stored at −70°C. The organism was confirmed by sub-culturing 0.2 ml of the broth culture on PPLO agar plates (Borosil). Plates were incubated at 37°C in 5% CO 2 incubator and were  2), sodium bicarbonate, fetal calf serum 10%, 200 μg ml −1 gentamicin and 2 mM glutamine, pH 7.2. HEp-2 cell was maintained by loosening the cells with PBS containing trypsin 0.25% and EDTA 0.02% and new flask was seeded [14].

Synthesis and PCR amplification of P1 gene fragments
Entire M. pneumoniae M129 P1 gene was synthesized in four fragments; N-terminal P1-I (1069 bp), two middle fragments P1-II (1043 bp) and P1-III (1983 bp), and C-terminal P1-IV (1167 bp) fragments by codon optimization replacing 21 UGA to UGG codons (Entelechon GmbH, Germany). To express these P1 gene fragments, four sets of primers were designed, each having two restriction sites either at 5'end or 3' end; NcoI and Bam HI were inserted at 5' end or Hind III and Sal I were inserted at 3' end. Table 1 shows the sequence of each primer. PCR was performed in a 50 μl of reaction mixture containing 1U of Taq polymerase, 1X PCR buffer, 200 μM deoxynucleotide diphosphates, 1.5 mM MgCl 2 , 10 pmol of each primer and template DNA. The reaction conditions were standardized at an initial denaturation of 94°C for 5 min, followed by denaturation at 94°C for 30 sec, annealing at 60°C for 30 sec and extention at 72°C for 1 min for 30 cycles. A final extention was done at 72°C for 5 min. All the four amplified fragments were cloned in pGEM-T easy cloning vector. Cloned fragments were confirmed by restriction digestion and sequencing.

Cloning, expression and purification of P1 gene fragments
For the expression, sub-cloning of the P1 gene fragments was done in NcoI and Hind III linearised pET28b vector. Ligation mixtures were used to transform BL21(DE3) and transformants were selected on kanamycin (25 μg ml −1 ) plates. Plasmid DNA was extracted from overnight cultures and subjected to restriction digestion to check the inserts. BL21(DE3) cells containing the recombinant plasmids were cultivated in 5 ml of LB broth containing kanamycin at 37°C with shaking (250 rpm) until the optical density (OD) reached 0.4 to 0.6. Protein expression was induced by 1 mM IPTG (isopropyl-β-D-thiogalactopyranoside; Sigma). After 5 h of induction at 37°C, bacterial cells were pelleted by centrifugation and the expression of each protein was analyzed on sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) gel.
Sub-cellular localization studies were carried out to analyze the expression of protein fragments in E. coli cells. Proteins were found to be expressed in the inclusion bodies. For the preparation of inclusion bodies E. coli cells were disrupted by sonication in Tris-buffer (0.05 M Tris, pH 8.0, and 0.3 M NaCl) with 1 min pulses at 1 min intervals 10 times using mini probe (LABSONIC R M, Sartorius Stedim Biotech GmbH, Germany) . The soluble and insoluble fractions were separated by centrifugation at 14,000 × g at 4°C for 30 min and were analyzed by SDS-PAGE.
To purify the all four P1 fragments, a protocol developed by Jani et al. was followed [40]. Briefly, one liter of E. coli culture cells expressing each of the protein fragments was grown and induced with 1 mM IPTG. After the induction, the bacterial pellets were obtained by centrifugation and then suspended in 1/20 volume of sonication buffer; 0.05 M Tris (pH 8.0), 0.3 M NaCl and 1% Triton X-100. The cell suspension was sonicated and the suspension was centrifuged at 14,000 × g for 30 min at 4°C. Pellets were washed 4 times with Tris-buffer without Triton X-100 and resuspended in CAPS (N-cyclohexyl-3amino propanesulfonic acid, pH 11) buffer containing 1.5% Sarkosin and 0.3 M NaCl. Suspensions were incubated for 30 min at room temperature and were centrifuged at 14,000 × g for 10 min at 4°C. Supernatant of each protein was kept with Ni-NTA + agarose resin with constant shaking for 1 h at 4°C. After binding, each supernatant was packed in four different purification columns and the resin was washed 4 times with CAPS buffer (10% imidazole). Bound proteins were eluted with Tris-buffer (pH 8.0) containing 0.25 M imidazole (Sigma-Aldrich, USA). Each protein fragments were eluted in 5 ml of buffer collecting in ten different fraction of 0.5 ml each. Eluted protein fractions were analyzed on 10% SDS-PAGE gels and fractions containing the recombinant proteins with a high degree of purity were pooled separately. The pooled protein fractions were extensively dialyzed against PBS, pH 8.0 and the protein concentration was determined by Bradford method. The eluted recombinant proteins were denoted as rP1-I, rP1-II, rP1-III and rP1-IV for protein fragments P1-I, P1-II, P1-III and P1-IV respectively.

SDS-PAGE and western blotting
To analyze the expression of all four recombinant proteins, induced and un-induced E. coli pellets from 1 ml of grown cultures were resuspended in 100 μl of 1× SDS sample buffer (62.5 mM Tris-HCl, pH 6.8, 10% glycerol, 2.3% w/v SDS, 5% v/v β-mercaptoethanol and 0.05% w/v bromophenol blue) and boiled for 5 min. The proteins were resolved on 10% SDS-PAGE gel and subsequently stained with Coomassie brilliant blue R-250. To ascertain   Immunization of Rabbits for raising antibodies against P1 protein fragments rP1-I, rP1-II, rP1-III and rP1-IV To characterize the immunogenic potential of recombinant P1 protein fragments, New Zealand white rabbits were used for the immunization with the approval of the Animal Ethics Committee, in accordance with the rules and regulations set forth by the AIIMS Animal Ethics Committee. Immunization was carried out with 6 week old New Zealand white rabbits which were maintained in the animal facility of AIIMS. Before immunization, pre-bleed sera were collected from each of these rabbits. Rabbits were immunized with 200 μg of purified recombinant P1 protein fragments (rP1-I, rP1-II, rP1-III and rP1-IV) emulsified in equal volume (300 μl) of complete Freund's adjuvant (CFA, Sigma -Aldrich, USA) intramuscularly. Rabbits were subsequently boosted with 200 μg of same protein fragments emulsified in equal volume (300 μl) of incomplete Freund's adjuvant (CFA, Sigma -Aldrich, USA) through the same route on the 28 th and 56 th day. Each one of the control rabbit was immunized with complete or incomplete Freund's adjuvant in PBS according to the immunization schedule. Blood samples were collected from each of the rabbit by ear vein puncturing on 14, 21, 35, 49 and 63 days. The serum was separated by centrifugation and stored at −20°C for further analysis. The rabbit sera were denoted as Pab (rP1-I), Pab (rP1-II), Pab (rP1-III) and Pab (rP1-IV) respectively. IgG antibody responses against the recombinant protein fragments were analyzed by ELISA and end point titers were determined. To confirm the specificity of antisera, western blot analysis was carried-out for each of the recombinant protein fragments with each antiserum.

Comparative analysis of recombinant P1 protein fragments by western blotting
In this experiment, equal amount (1 μg) of purified recombinant P1 protein fragments (rP1-I-IV) were run in two separate SDS-PAGE. SDS-PAGE of all the four purified P1 protein fragments was transferred to two separate nitrocellulose membrane to perform western blotting. After blocking with 5% skimmed milk in PBS-T one membrane was then incubated with primary antibody (pooled sera of M. pneumoniae infected patients, 1:50) and second membrane was incubated with primary anti-M. pneumoniae antibody (1:3,000 dilutions) for 1 h. After washing with PBS-T first membrane was incubated with secondary antibody goat anti-human IgG and second membrane with secondary antibody goat anti-rabbit IgG conjugated with horseradish peroxidase (1:5000 dilutions) for 1 h. The membrane was developed with DAB and H 2 O 2 .

Reactivity of recombinant P1 protein fragments to patient sera
All the four recombinant P1 protein fragments; rP1-I, rP1-II, rP1-III and rP1-IV were analyzed for their reactivity to twenty five sera of M. pneumoniae infected patients and sixteen healthy patient sera using ELISA assay as well as fifteen sera of M. pneumoniae infected patients by western blot analysis. Western blot analysis was performed as described above using equal amount of recombinant proteins. For the ELISA analysis, 96-well microplates (Nunc, Roskilde, Denmark) were coated with 50 ng of either of the four P1 protein fragments in 0.06 M carbonate/bicarbonate buffer (pH 9.6) per well. The plates were kept overnight at 4°C and next day the well were washed with PBS-T and blocked with 5% skimmed milk in PBS-T for 2 h at room temperature. The antigen coated wells were next incubated with sera of M. pneumoniae infected patients (1:50 dilutions) for 1 h at 37°C. After incubation, plates were washed with PBS-T and incubated with secondary goat anti-human antibody conjugated with horseradish-peroxidase (1:3,000 dilutions) for another 1 h at 37°C. The enzyme reaction was developed by addition of TMB/H 2 O 2 substrate (Bangalore Genei) and was incubated in dark for 30 min at 37°C. The reaction was stopped with 2 N H 2 SO 4 and the absorbance was read at 450 nm wavelength using micro-plate ELISA reader (Bio-Tek Microplate Reader, USA).

M. pneumoniae adhesion assay
HEp-2 cells (5×10 4 HEp-2 cells ml −1 ), in RPMI-1640 medium with penicillin (100 U ml −1 ) 0.05% were added to 24-well Multi-dish plates (Nunc, Roskilde, Denmark) using sterile glass cover slips underneath. The plates were incubated overnight in 5% CO 2 at 37°C. Next day, HEp-2 cells in each well were infected with the M. pneumoniae RPMI-suspension (50 μl well −1 ) and incubated for 6 h in 5% CO 2 at 37°C. The infected HEp-2 cells were fixed in methanol 100% (1 ml well −1 ) at −20°C for 1 h and washed with PBS. To detect the adhering mycoplasmas, 100 μl per well of each of the primary antibodies anti-M. pneumoniae antibodies, Pab(rP1-I), Pab(rP1-II), Pab(rP1-III), or Pab (rP1-IV (1:500 dilutions) were added and were incubated for 1 h at 37°C. Wells were washed subsequently and later 100 μl of secondary fluorescein isothiocyanate (FITC)conjugated goat anti-rabbit IgG (whole molecule, 1:100 dilutions) (Santa Cruz Biotech, USA) was added. The cells were washed twice in PBS before and after the addition of antibodies. Cells were subsequently incubated with Evans Blue diluted 1:10 for 30 min at 37°C. Finally the cells were washed with double distilled water.

M. pneumoniae adhesion inhibition assay
For the adhesion inhibition assay, protocol developed by Svenstrup et al. was followed [14]. Briefly, the M. pneumoniae suspension (50 μl) was pre-incubated for 2 h at 37°C with 50 μl of anti-M. pneumoniae antibodies, Pab (rP1-I), Pab (rP1-II), Pab (rP1-III) or Pab (rP1-IV) in different dilutions (1:50, 1:100, 1:200 and 1:500) before incubation of the HEp-2 cells. The M. pneumoniae-antibodies suspension (100 μl) was then added to the HEp-2 cells together with 1 ml of RPMI with penicillin and incubated overnight in 5% CO 2 at 37°C. Fixation and addition of secondary antibodies were carried-out as described in the adhesion of M. pneumoniae. To further confirm the adhesion inhibition, the assay was performed as mentioned above except that DAPI was added at the end of the assay for further 30 min at room temperature.

M. pneumoniae surface exposure assay
To detect M. pneumoniae surface protein, the primary antibodies were added before methanol fixation. Otherwise, the procedure was the same as described for the M. pneumoniae adhesion assay.

Indirect immunofluorescence microscopy (IFM)
Samples prepared for M. pneumoniae adhesion assay, M. pneumoniae adhesion inhibition assay and M. pneumoniae surface exposure assay were analyzed by IFM using Olympus BX51upright fluorescence microscope. Before microscopy analysis, a drop of anti-fade solution (p-phenyldiamine dihydrochloride 1 μg ml −1 in PBS 10% and glycerol 90%, pH 9.0) was placed between the glass cover slips and the slides.