Monoclonal antibody P1/MCA4 and its binding to P1

Monoclonal antibody P1/MCA4, raised against the P1:1160-1518 peptide, strongly inhibited the adhesion of M. pneumoniae to red blood cells. In the presence of P1/MCA4, gliding mycoplasma cells rapidly reduce their speed and eventually halt and detach from sialylated glass surfaces (S1 Video), while non-moving cells are not affected by the presence of P1/MCA4 and remain attached to surfaces. As mentioned in the introduction, a similar behaviour had been reported for antibody P1/MCA3 [28]. It is worth mentioning that this behaviour is also observed in three additional monoclonal antibodies raised in the same batch as P1/MCA3 and P1/MCA4 (S1 Table). The P1:1160-1518 peptide corresponds to a fragment of the N-terminal domain (residues Ala1160-Thr1399) and to (almost) the whole C-domain (Ala1400-Gln1518) according to the structural information now available [19].

cDNA sequencing of P1/MCA4 mRNAs in hybridoma cells showed the presence of a single IgG heavy chain mRNA but two different light chain mRNAs (S1 Fig). N-terminus analysis of samples from antibody P1/MCA4 (obtained as indicated in the Material and Methods) confirmed the presence of one heavy chain but two different light chains (roughly 70% for the most abundant). The heterogeneity of light chains might explain the failure of numerous crystallization attempts involving P1/MCA4. Analysis by multi-angle light scattering (MALS) showed the formation of a stable and homogeneous P1-Fab(P1/MCA4) complex, in a solution with equimolecular amounts of P1/MCA4 Fabs and the whole P1ectodomain (residues Thr29-Asp1521), indicating that the two kinds of Fab found in P1/MCA4 bind efficiently to P1 (S2 Fig).

Antibody P1/MCA4 presents high affinity for a construct corresponding to the C-domain of P1, proving that this small domain contains the epitope. Moreover, western blotting analysis performed using different segments of P1 showed that residues Thr1426-Asp1438 are a key part of the epitope (S3 Fig). An additional observation that could have clinical implications is that P1/MCA4 presents affinity also for the C-domain of P140 (residues Lys1220-Asp1351) from M. genitalium, even stopping the movement of cells, although after treatments longer than with M. pneumoniae, in agreement with the high sequence and structural similarities between the C-domains of the adhesins of M. pneumoniae and M. genitalium (S4 Fig).

Cryo-EM structure of the P1-Fab (P1/MCA4) complex

The structure of the ectodomain of P1 bound to the Fab fragment of P1/MCA4 was determined by cryo-EM single particle analysis with an overall resolution in the final map of 2.4 Å (Figs 1 and S5 and S2 Table). The high quality of the map allowed accurate modelling of the structure from P1 and the Fab variable domains, defining clearly the paratope-epitope interface (Fig 2a). The Fab constant domains were less well defined, indicating some flexibility in the Fab elbow. The sequence observed for the Fab corresponded to the one with light chain L1. The structure of P1 consists of a large N-terminal domain (residues 60-1399) and a smaller C-domain (1400-1521), in full agreement with the structures of P1 reported previously [19]. Superposing the P1 structure determined here with the P1 crystal structure (PDB code 6rc9), gives a root mean square deviation (rmsd) of 1.80 Å for 1236 aligned residues (from a total of 1339 in the model) (S6 Fig). When the superposition is done separately for the N-terminal domain and C-domain, rmsds are 1.65Å (1151 aligned residues out of 1226) and 1.20Å (97 aligned out of 104), respectively. These results show the high plasticity of P1, in particular for the N-terminal domain, which experiences significant changes although it participates only indirectly in the interaction with the P1/MCA4 Fab. Comparison of the crystal (6rc9) and the new (in the complex of P1-Fab) structures of P1, indicates a 7.7° rotation of the C-domain relative to the N-terminal domain, confirming hinge movements between the two domains as had suggested the cryo-EM structure of P1 where the C-domain was disordered [19]. The structure of the P1-Fab complex shows that the epitope of P1/MCA4 is conformational, although involving residues from the C-domain only, in particular from loop Val1425-Asp1438 in agreement with the epitope mapping analysis (Figs 2a and S3). The quality of the cryo-EM map was high enough to even allow the localization in the N-terminal domain of a significant number of solvent molecules, most of them also present in the crystal structure of P1 (S5 Fig). It is worth mentioning that docking using a previous cryo-EM map determined at lower resolution, correctly predicted the overall binding interface between P1 and the Fab(P1/MCA4), with the largest deviations arising from the inaccuracy on the fitting of the P1 C-domain (S7 Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Cryo-EM structure determination of the P1-Fab(P1/MCA4) complex.

a) Representative micrograph of P1-Fab(P1/MCA4) complexes examined by CryoEM. b) Initial classes obtained after 2D classification of 2000 particles that were manually selected. c) Schematic processing overview for the P1-Fab(P1/MCA4) complex structure resolution from particles that were automatically picked from movies; CTF, Constrast transfer function. d) Fourier shell correlation of P1-Fab(P1/MCA4) complex at a final resolution of 2.39Å according to the 0.143 cut-off criteria. e) Final 3D reconstruction map color-coded according to local resolution.

https://doi.org/10.1371/journal.ppat.1012973.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Cryo-EM structure of the P1-Fab(P1/MCA4) complex.

a) Ribbon representations, with three 90° apart views, of the complex between the ectodomain of P1 (brown) and the Fab fragment of Mab P1/MCA4 (light and heavy chains in blue and green, respectively). Insets indicate the main interactions between P1 and the Fab. The epitope is located in the C-domain of P1 closed to the mycoplasma membrane, which is also displayed as a reference. b) Two views of the Fab(P1/MCA4) (shown in green) and the extracellular region of the Nap, with P1 (brown) and P40/P90 (pink transparent). In the “open” conformation of the Nap (left panel) the P1/MCA4 epitope, buried at the center of the Nap, is totally inaccessible to antibodies and a Fab placed there would catastrophically clash with the adhesins. Therefore, to expose the P1/MCA4 epitope the C-domain of P1 has to experience an important rearrangement (right panel).

https://doi.org/10.1371/journal.ppat.1012973.g002

Importantly, superposing the structure of P1-Fab(P1/MCA4) onto the Nap complex in “open” conformation available for M. genitalium [22] results in a catastrophic steric clash of the Fab against the adhesins, indicating that the epitope of P1/MCA4 is totally inaccessible to antibodies in the “open” conformation of the Nap complex (Fig 2b). Therefore, this complex must experience important structural rearrangements between the “open” conformation and when the epitope of P1/MCA4 is fully exposed.

Polyclonal antibodies against the P40/P90 and P1 ectodomains and against the N-terminal domain of P1

Polyclonal antisera against the ectodomains of P40/P90 (residues Ala25-Pro1113) and of P1(residues Thr29-Asp1521) were obtained by immunizing mice with recombinant versions of these polypeptides. Polyclonal antisera against the N-terminal domain of P1 were obtained by immunizing mice with the P1 N-terminal domain (residues Thr29-Ala1375). These antisera were tested by ELISA, Western Blotting and immunofluorescence assays on whole M. pneumoniae cells, and titers higher than 1/2000 were obtained for all the antisera. Effects of these antisera on M. pneumoniae motility and attachment were investigated by incubating M. pneumoniae cells in the presence of 3% gelatin in the SP4 medium [21]. Because M. pneumoniae cells spontaneously detach at high frequencies from the observation surface in the absence of gelatin [27], more reproducible and consistent results were obtained when gliding motility was analyzed in SP4 medium supplemented with gelatin (Tables 1 and S3). In the absence of antibodies (negative control), most of the cells remained attached and motile during the observation time (Table 1). In contrast, the P1 ectodomain antisera stopped the movement of 50% of mycoplasma cells after four minutes of incubation (Table 1 and S2 Video) with very few cells remaining motile at the end of the observation period. No significant effects on mycoplasma gliding were detected neither for the antisera against the P1 N-terminal domain nor for the P40/P90 ectodomain, even at the lowest dilution tested (Table 1 and S3 and S4 Videos). It is also worth mentioning that cell adhesion was inhibited when cells were added to media that already contained antisera against the ectodomains of either P1 or P40/P90, suggesting that these antibodies can interfere with their binding to SOs via steric hindrance, but their epitopes are not accessible when the cells are attached to surfaces.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Mycoplasma motility in the presence of antibodies.

https://doi.org/10.1371/journal.ppat.1012973.t001

Binding of antibodies to epitopes whose accessibility varies during the conformational transformation interferes with the dynamics of the adhesion complex. Accordingly, the attachment/detachment cycle is halted by antibody P1/MCA4 against the C-domain of P1 that experiences important rearrangements during the cycle. In contrast, the attachment/detachment cycle is not altered, or very little, by antibodies against P40/P90 or against the N-terminal domain of P1, indicating that accessibility of their epitopes must remain essentially unchanged during the cycle.

Mutations in the Engelman motifs

The transmembrane domains of P1 and P40/P90, predicted to be single helices, are located just after the C-terminal of the globular ectodomains of these proteins. Remarkably, these helices contain one Engelman motif in P1 and two Engelman motifs in P40/P90 (S8 Fig). These three motifs are highly conserved in the M. pneumoniae cluster of mycoplasmas. Engelman motifs are frequently involved in high-affinity interactions between membrane helices and have the sequence GXXXG, with X being any residue but in general hydrophobic [25]. We have investigated by mutational analysis possible contributions of the Engelman motifs from adhesins to the organization and functioning of the Nap complex. This study was performed with M. genitalium because simple genetic methods to introduce point mutations are available in this microorganism and transmembrane helices from orthologue adhesins of M. pneumoniae and M. genitalium present high sequence identities (S8 Fig). The Engelman motif at the C-end of the P140 transmembrane helix, close to the internal side of the cell membrane, will be referred as E1 and the two Engelman motifs in P110 close to the external and to the internal sides of the cell membrane, as E2 and E3, respectively (Fig 3a). Mutations were introduced by transposon delivery in a M. genitalium null mutant G37∆Adh (G37∆MG_191-∆MG_192), to obtain isogenic strains bearing Gly to Phe double point mutations of the glycine residues of P140 and/or P110 Engelman motifs (S4 Table). The protein profile analysis of these mutant strains showed that adhesins bearing the different mutations were expressed at similar levels that P140 and P110 in wild type (WT) cells (S9 Fig). The properties of the adhesin variants were tested by a quantitative hemadsorption assay (Fig 3b). This assay provides data about the adhesion properties of mycoplasma cells to mammalian red blood cells [29]. Cells from E1 mutant strain showed hemadsorption values very similar to WT cells and also similar to the cells from the null mutant G37∆Adh complemented by a WT copy of both adhesin coding genes. In contrast, E2 mutants showed an intermediate adherent phenotype and E3 mutants had a more severely impaired hemadsorption phenotype, characterized by low B_max and high Kd values, indicating the presence of adhesins with a lower binding affinity for cell receptors. As expected, E1-E2 and E1-E3 mutant strains exhibited phenotypes similar to those showed by E2 and E3 mutants, respectively. In addition, E2-E3 and E1-E2-E3 mutant strains exhibited a non-adherent phenotype characterized by very low B_max and very high K_d values, resembling the phenotype of the cells from the G37∆Adh null mutant strain. Interestingly, immunolabeling with a polyclonal antiserum against the Nap complex indicates that adhesin variants are located at the tip of the cells bearing the triple mutation E1-E2-E3 (S10 Fig), suggesting the presence of Nap complexes in TOs of these cells. Mutant strains were also examined by Scanning Electron Microscopy (Fig 3c). While cells from E1 mutants exhibited normal morphologies, cells with mutations in E2 or E3 showed a multiple TO (mTO) phenotype. Although the fraction of cells presenting mTOs was only slightly increased in E2 variants, most of the cells from E2-E3 and E1-E2-E3 variants exhibited mTOs (S5 Table). Nascent TOs develop adjacent to a preexistent TO, and gliding motility provides the driving force to deliver the new TOs to the opposite cell pole. This process is hindered in cells having adhesion and motility deficiencies [30], which can explain the correlation between the reduced adhesion and the increased frequency of cells showing mTOs in E2-E3 variants. Since P110 and P140 adhesins co-stabilize each other and the formation of Nap complexes is essential for TO development [31,32], the presence of TOs by SEM suggests that these adhesins are properly folded in the TOs of cells bearing Engelman motif variants. Moreover, the absence of detectable phenotypic defects in E1 adhesin variant also indicates that this mutation has no impact in the folding or in the organization of the Nap complex. Therefore, mutational analysis allows us to conclude that Engelman motifs from P110, but unexpectedly not from P140, have a critical and synergic contribution to adhesion, strongly suggesting that interactions between the transmembrane helices of the two P110 subunits are required to achieve a functional “open”, ready for binding, conformation of the Nap complex.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Engelman motifs in adhesins.

a) Transmembrane helices of adhesins from the M. pneumoniae cluster of mycoplasmas contain three highly conserved Engelman motifs (GxxxG sequences), referred as E1 (P140) and E2, E3 (P110) in M. genitalium. b) Mutational analysis, performed in the Engelman motifs of P140 and P110, indicate that motifs E2 and E3 from P110 have important and synergic contributions to adhesion. On the contrary, no effects were observed when E1 from P140 was mutated. c) Changes in adhesion correlate with increasing rates of cells presenting multiple TOs phenotypes.

https://doi.org/10.1371/journal.ppat.1012973.g003

Ethics statement

The experimental procedures to immunize mice and obtaining polyclonal sera were approved by the Ethics Committee on Animal and Human Experimentation of the Universitat Autònoma de Barcelona (UAB, procedure 1002R3R2R). Experiments growing and transforming mycoplasma cells were approved by the Biosecurity Committee of the UAB (procedure HR-640-20).

Bacterial strains and tissue culture cells

M. pneumoniae M129 strain was grown in cell culture flasks containing SP4 or PPLO media and incubated at 37 ºC with or without 5% CO₂. Surface-attached mycoplasmas were harvested using a cell scraper and resuspended in medium or suitable buffers for experiments. When growing mycoplasma cells on IBIDI 8-well chamber slides, each well was seeded with about 10⁵ CFUs and incubated for 12-24 h in 200 μL SP4 supplemented with 3% gelatin. All M. genitalium strains were grown in SP-4 broth at 37°C in a 5% CO₂ atmosphere in tissue culture flasks. SP-4 plates were prepared supplementing the medium with 0.8% agar (BD). Chloramphenicol (17 μg mL⁻¹), puromycin (3 μg mL⁻¹) or gentamycin (100 μg mL⁻¹) were added for mutant selection.

Escherichia coli strain XL-1 Blue (Agilent, Santa Clara, USA) was used for cloning and plasmid propagation purposes. The strain was grown in Luria Bertani (LB) or LB agar plates containing 100 μg mL⁻¹ ampicillin, 40 μg mL⁻¹ X-Gal, and 24 μg mL⁻¹ Isopropyl β-D-1-thiogalactopyranoside (IPTG) when needed.

NSI myeloma cells [37] were grown in RPMI 1640 medium supplemented with 10% FBS and 50 μg mL⁻¹ gentamycin (complete RPMI). Hybridomas were selected in complete RPMI supplemented with HAT media and BM-Condimed (Sigma Aldrich).

Cloning, expression and purification of P1 and P1 fragments

The coding sequences for constructs of P1 and the C-domain of P1 were amplified from the synthetic clone of MPN_141 (P1) gene from M. pneumoniae (Genscript), using primers P1F and P1R for P1 (29-1521), P1Ct1400_F and P1R for P1 C-terminal (1400-1521), P1Ct1376_F and P1R for P1 C-terminal (1376-1521) (S6 Table). The coding sequences for constructs of P40/P90 were amplified from the synthetic clone of MPN_142 (P40/P90) gene from M. pneumoniae (Genscript), using primers P40P90_F and P40P90_R (23-1114). The PCR fragments were cloned into the expression vector pOPINE (-) to add a C-terminal His-tag to the resulting constructs. Recombinant proteins were obtained after expression with B834 (DE3) cells (Merck), with IPTG at 0.8 mM and 22ºC overnight. The cell pellets were lysed in 1xPBS at pH 7.4 and 40 mM Imidazole (binding buffer) and centrifuged at 20000 RPM at 4ºC. Supernatant was charged into a HisTrap 5ml column (GE Healthcare) and eluted with a buffer containing 400 mM imidazole. Soluble aliquots were concentrated and loaded in a Superdex 200 GL 10/300 column (GE Healthcare) with a Tris·HCl 20 mM buffer (pH 7.4) and 150 mM NaCl.

Production and validation of polyclonal antibodies

Polyclonal antisera anti-P1, anti-P1 fragments and anti-P90/P40 were prepared by immunizing BALB/C mice with the respective recombinant proteins and polypeptides described above. Sera were obtained by cardiac puncture of properly euthanized mice just before splenectomy and tittered using serial dilutions of each antigen. Titers of the different polyclonal sera were determined as the IC₅₀ value from four parameter logistic plots and found to be approximately between 1/2500 and 1/4000.

Production and sequencing of the monoclonal antibody P1/MCA4

A p1 gene fragment was amplified by PCR from genomic DNA of M. pneumoniae M129 strain using primers P1F_2 and P1R_2 (S6 and S7 Tables). The amplified fragment was inserted into NcoI and XhoI site of pET-30c(+) expression vector, resulting the pP1-8 plasmid. E. coli BL21(DE3) was transformed by the pP1-8 for expression of a recombinant P1 peptide containing Ala1160 to Gln1518 of the P1 of M129 strain. The recombinant P1 was purified by His-tag affinity chromatography and used to immunize mice for production of anti-P1 monoclonal antibody. The monoclonal antibody P1/MCA4 was selected by ELISA screening and showed high specificity to P1 in Western blotting analysis and immunofluorescence microscopy of living M. pneumoniae cells. The amino acid sequence of P1/MCA4 was determined at GenScript (Tokyo, Japan) by cloning and sequencing of the cDNAs of antibody mRNA from P1/MCA4 producing hybridoma cell. The obtained sequences for a single H chain and two L chain of the P1/MCA4 (S1 Fig) were deposited into the DDBJ/ENA/GenBank databases under the accession numbers LC600310, LC600311 and LC600312.

Purification of the fab fragment

Monoclonal antibody P1/MCA4 against P1 was diluted up to 2 mg mL⁻¹ in 1xPBS buffered at pH 8.0 and mixed with 0.1mg/ml of Papain, 20 mM L-cysteine, 20 mM EDTA and 1xPBS pH: 7.0 in a relation 1:40 (Protein:Papain) and incubated during 4 hours at 37ºC. To stop digestion, 10% of the final volume of iodoacetamide was added. The digested solution was loaded onto HiTrap Protein G 5ml with Binding buffer (20 Mm Na₂HPO₄ pH:7.0) and Elution buffer (0.1 M glycine·HCl pH 2.7). Soluble aliquots were concentrated and loaded in a Superdex 75 GL 10/300 column (GE Healthcare) with a Tris·HCl 20 mM pH 7.4 and 150 mM NaCl buffer. Complexes of P1-mAb and P1-Fab were formed by mixing both proteins in a ratio 2:1 and 1:1, respectively.

Epitope mapping of monoclonal antibody P1/MCA4

For epitope mapping of P1/MCA4, a series of plasmids that express recombinant P1 peptides were constructed (S7 Table). These plasmids were generated by partial deletion of the pP1-8 plasmid using specific primer sets and PrimeSTAR Mutagenesis Basal kit (Takara Bio, Shiga, Japan). Recombinant P1 peptides were produced in E. coli BL21(DE3) harboring the plasmids, separated by 10-20% gradient-gel SDS-PAGE, and transferred to nitrocellulose membranes. Recombinant P1 peptides on membranes were reacted with 0.5 µg/ml P1/MCA4. After washing of the membranes, binding of the P1/MCA4 to P1 peptide was detected by anti-mouse IgG, HRP conjugate secondary antibody (Promega, Madison, WI, USA) and EzWestBlue W substrate (Atto, Tokyo, Japan).

Size exclusion chromatography and multi angle light scattering (SEC-MALS)

Molecular weights were measured using a Superdex 200 10/300 GL (GE HEalthcare) column in a Prominence liquid chromatography system (Shimadzu) connected to a DAWN HELEOS II multi-angle light scattering (MALS) detector and an Optilab T-REX refractive index (dRI) detector (Wyatt Technology). ASTRA 7 software (Wyatt Technology) was used for data processing and result analysis. A dn/dc value of 0.185 mL g⁻¹ (typical of proteins) was assumed for calculations.

Single-particle cryo-EM of the binary complex

To prevent preferred orientations, Octylthioglucoside (OG) was added to solution of purified P1-Fab (P1/MCA4) complex to adjust the final concentration of OG to 0.9% (w/l). 2.6 μL of sample solution was applied to a glow-discharged Au-coated Quantifoil holey carbon grid (R1.2/1.3, Cu, 200 mesh), blotted for 4 sec at 4 °C in 100% humidity and plunged into frozen liquid ethane using a Vitrobot Mark IV (Thermo Fisher Scientific). The grid was inserted into a CRYO ARM 300 (JEOL) operating at an acceleration voltage of 300 kV. CryoEM images were recorded with a K3 direct electron detector (Gatan) in CDS mode with an energy filter at a slit width of 20 eV. Data were automatically collected using the SerialEM software (https://bio3d.colorado.edu/SerialEM/) at a physical pixel size of 0.49 Å with 50 frames at a dose of 2.0 e⁻/ Å ² per frame and an exposure time of 2.86 sec per movie with a defocus ranging from -0.5 to -1.5 μm. A total of 27,122 movies were collected for P1-Fab complex.

The movie frames were subjected to beam-induced movement correction using MotionCor2.1 [38] and contrast transfer function (CTF) was evaluated using Gctf [39]. Approximately 2,000 particles were manually selected from 20 micrographs to perform two-dimensional (2D) classification. Using a good 2D class average image, a total of 4,312,408 particle images were automatically picked and 2D classifications were performed using RELION-3.1 [40]. A total of 1,252,491 particles were selected for building the initial model of the P1-Fab complex using cryoSPARC2 [41] and subjected to 3D classification into 8 classes using RELION-3.1. The selected particles were re-extracted at a pixel size of 0.49 Å and subjected to three 3D refinement, two CTF refinement and Bayesian polishing. After selection of particles using no-alignment 3D classification, a total of 352,039 particles were subjected to two 3D refinement and CTF refinement. The final 3D refinement and post-processing yielded maps with global resolutions of 2.39 Å, according to the 0.143 criterion of the Fourier shell correlation (S2 Table). Local resolution was estimated using RELION-3.1. Processing strategy is described in Fig 1.

Model building and refinement of the binary complex

The model of the P1-Fab(P1/MCA4) complex was built based on the cryoEM density map. Previous published crystal structure of P1 (PDB ID: 6RC9) was docked into the EM density map using UCSF Chimera [41]. N-domain, C-domain of P1 and Fab were manually repositioned and refined iteratively using COOT [42] and PHENIX real space refinement [43]. The statistics of the 3D reconstruction and model refinement are summarized in S2 Table. The final refined structure has been deposited in the PDB with code 8ROR.

DNA manipulation and primers

Plasmid DNA was purified using GeneJET Plasmid Miniprep Kit (Thermo Fisher Scientific). PCR products and DNA fragments were recovered from agarose gels using NucleoSpin Gel and PCR Clean-up Kit (Macherey-Nagel, Düren, Germany), and digested using the corresponding restriction enzymes (Thermo Fisher Scientific) when necessary. For transformation of M. genitalium, plasmids were purified using the GenElute HP Midiprep Kit (Sigma-Aldrich, St. Louis, USA) following the manufacturer’s instructions. All primers used in this study are listed in S8 Table.

M. genitalium mutant strains

The suicide plasmid pBE∆MG_191/MG_192 was designed to generate a G37 M. genitalium MG_191 and MG_192 null mutant strain by gene replacement. First, a 1 kb flanking upstream region (UR) to the MG_191 gene was PCR-amplified using the MgParUp-F and MgParUp-R primers. Similarly, a 1 kb flanking downstream region to the MG_192 gene was PCR-amplified with the MgParDw-F and MgParDw-R primers. In parallel, the a lox version of the CmM438 selectable marker [44] was PCR-amplified with the Lox71p438-Fwd-XhoI and CatLox66-Rev-BamHI primers, obtaining the CmM438 flanked with the lox61 and lox71 sequences. Then, the UR and DR PCR products were joined by PCR with the MgParUp-F and MgParDw-R primers. The resulting PCR product was cloned into an EcoRV-digested pBE plasmid [45]. Finally, the resulting CmM438 amplicon was digested with XhoI and BamHI restriction enzymes and ligated into the similarly digested pBE containing the UR and DR regions obtained before.

This plasmid was used to obtain the G37∆Adh chloramphenicol resistant mutant strain. Electroporation of this mutant with the pCre plasmid [46] that contains the Cre recombinase allowed us to obtain the G37∆Adh strain, free of any antibiotic selectable marker. This strain was used as the recipient strain to transform all the pMTnPac plasmids generated in this study.

MG_191 and MG_192 genes from the chromosome of M. genitalium G37 strain were amplified by PCR using COMmg191-F-ApaI and COMmg191/192-R-SalI primers. The resulting PCR product was cloned into an EcoRV-digested pBE plasmid to create the pBE-COMP140P110. At the same time, the amplicon was digested with the ApaI and SalI restriction enzymes and ligated to a similarly digested pMTnPac plasmid to create the pTnPacCOMP140P110. This plasmid was used to reintroduce the wild-type alleles of the MG_191 and MG_192 genes to the G37ΔAdh mutant. The COMmg192-F (ApaI) primer includes the upstream region (70 nucleotides) of the MG_191 gene, which contains a strong promoter identified in a previous study [47]. This promoter was used to drive the transcription of the transposon-encoded copy of the MG_191 and MG_192 genes in all mutants.

To generate P140 and P110 variants carrying specific mutations in the Engelman motifs, the target mutations in the MG_191 and or MG_192 genes were introduced by Exsite-PCR using the pBE-COMP140P110 as a template and the specific primers for each mutant (S8 Table). Then, plasmids containing the P140 and P110 alleles with the desired mutations were re-ligated and mutant P140 and P110 alleles excised by digestion with ApaI and SalI restriction enzymes. Finally, DNA fragments with mutant alleles were ligated into a pMTnPac plasmid previously digested with ApaI and SalI, to generate the corresponding pMTnPacCOMP140P110 plasmid series.

Sequencing analysis of the different TnPacCOMP140P110 constructs using primers Tnp3, RTPCR192-F, RTPCR192-R, and PacUp, ruled out the presence of additional mutations in the MG_191 and MG_192 sequences. These plasmids were transformed into the M. genitalium G37∆Adh null mutant to create the different P140 and P110 variant strains. Identification of the minitransposon insertion site in the individual clones was done by sequencing using the PacDown primer and chromosomal DNA as a template.

Transformation and screening

M. genitalium G37∆Adh null mutants were transformed by electroporation using 10 μg of plasmid DNA of the different minitransposons or 30 μg when using plasmids for gene replacement experiments, as previously described [45,48]. Puromycin-resistant colonies were picked, propagated, and stored at –80 °C. For screening purposes, strains were further propagated in antibiotic supplemented SP4 medium in 25 cm² tissue culture flasks. To obtain crude lysates from grown cultures, SP4 medium was removed and cells were lysed using 50 μL of Lysis Buffer (0.1 M Tris-HCl pH 8.5, 0.05% Tween-20, and 250 μg mL⁻¹ Proteinase K) and incubated for 1 h at 37 °C. Then, Proteinase K was inactivated at 96 °C for 10 min. M. genitalium lysates were screened by PCR or direct Sanger sequencing using the primers in S8 Table.

Sequencing reactions

DNA sequencing reactions were performed using BigDye v3.1 Cycle Sequencing kit using 2.5 μL of genomic DNA or M. genitalium lysate, following manufacturer’s instructions (Thermo Fisher Scientific). All reactions were analysed in an ABI PRISM 3130xl Genetic Analyzer at the Servei de Genòmica i Bioinformàtica (UAB).

SDS-PAGE and western blotting

Whole-cell lysates were obtained from mid-log phase cultures grown in 75 cm² flasks. Protein concentration was determined with the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific), and similar amounts of total protein were separated by SDS-PAGE following standard procedures.

For specific detection of M. genitalium proteins, SDS-PAGE gels were electrotransferred to PVDF membranes by a Semi-dry transfer System, using cold Towbin buffer (25 mM Tris, 192 mM Glycine pH 8.3-8.4, 20% Methanol (v/v)) following the standard protocols [49,50]. Membranes were probed with the appropriate primary antibody in blocking solution, which was then detected using a secondary antibody conjugated with horseradish peroxidase (Bio-Rad). Antibodies used in this study are listed in S9 Table. Bioluminescence reaction was catalyzed with Luminata Forte Western HRP substrate (Merk Millipore). Visualization and image optimization was performed in a VersaDoc Imaging System (Bio-Rad) using QuantityOne software (Bio-Rad).

Quantitative hemadsorption assay

Hemadsorption was quantified using flow cytometry as previously described [29] with few modifications. We used 10⁹ mycoplasma cells for the hemadsorption assay. Fluorescence-activated cell sorting (FACS) data were acquired using a FACSCalibur (Becton Dickinson, Franklin Lakes, USA) equipped with an air-cooled 488 nm argon laser and a 633 nm red diode laser and analysed with the CellQuest-Pro and FACSDiva software (Becton Dickinson). Binding of mycoplasma cells to red blood cells was modelled in an inverse Langmuir isothermal kinetic function:

Plots represent the best-fitting curves to a series of hemadsorption measurements obtained from at least two biological repeats for each strain. We performed a double-gating strategy, using a preliminary FL3-H/FL2-H gate following an SSC-H/FL1-H gate.

Time lapse micro-cinematography

Effects of antibody sera on mycoplasma cells were investigated by time lapse cinematography of M. pneumoniae cells growing in IBIDI 8-well chamber slides using SP4 medium supplemented with 3. Prior to observation, the medium was replaced with fresh SP4 supplemented with 3% gelatin and pre-warmed at 37°C. After incubation for 10 minutes at 37 ºC and 5% CO₂, the slide was placed in a Nikon Eclipse TE 2000-E inverted microscope equipped with a Microscope Cage Incubation System (Okolab) at 37 ºC. Images were captured at 0.5 s intervals for a total observation time of 10 min, and the different antibodies were dispensed directly into the wells after the first 50 seconds of observation. Frequencies of motile cells and detached cells before adding the different antibodies were calculated from the images collected between 0 and 50 seconds of observation. Frequencies of motile cells and detached cells after adding the different antibodies were calculated from the pictures collected between 50 seconds and 10 minutes of observation.

Immunofluorescence microscopy

The immunofluorescence staining of mycoplasma cells on chamber slides was similar to previously described [14,27], with several modifications. All mycoplasma cells were washed with PBS containing 0.02% Tween 20 (PBS-T) except cells from M. genitalium G37∆Adh strain and Engelman motif mutant strains, which were washed in PBS containing 136 mM sucrose (PBS-Suc), prewarmed at 37 °C. Then, each well was fixed with 200 μL of 3% paraformaldehyde (wt/vol) and 0.1% glutaraldehyde. Cells were washed three times with PBS-T (or PBS-Suc), and slides were immediately treated with 3% BSA in PBS-T (blocking solution) for 30 minutes. The blocking solution was removed and each well was incubated for 1 hour with 200 μL of the primary antibodies diluted in blocking solution. We have used a 1/2000 dilution for all polyclonal sera. Wells were washed three times with PBS-T or PBS-Suc and incubated for 1 hour with a 1/2000 dilution of a goat anti-mouse Alexa 555 secondary antibody (Invitrogen) in blocking solution. Wells were then washed three times with PBS-T or PBS-Suc and incubated for 20 minutes with 100 μL of a solution of Hoechst 33342 10 μg μL⁻¹ in PBS-T or PBS-Suc. Wells were finally washed once with PBS-T and replenished with 200 μL of PBS-T. Cells were examined by phase contrast and epifluorescence in an Eclipse TE 2000-E inverted microscope (Nikon). Phase contrast images, 4’,6-diamidino-2-phenylindole (DAPI, excitation 387/11 nm, emission 447/60 nm) and Texas Red (excitation 560/20 nm, emission 593/40 nm) epifluorescence images were captured with an Orca Fusion camera (Hamamatsu) controlled by NIS-Elements BR software (Nikon).

Scanning electron microscopy

For the Scanning Electron Microscopy (SEM) analysis, the M. genitalium cultures were grown to mid-long phase over glass and round coverslips in a 24-well plate with fresh SP4. For non-adherent mutant strains, the coverslips were treated in advance with 0.2 mg/mL Poly-L-Lysine to allow the cells to stick to the surface of the glass. The inoculum depended on the stock concentration and usually ranged between 5-10 µL of M. genitalium strains. The Mge-WT and the mutant strains were grown overnight and washed twice with 1 mL 0.1M PB. Then, the cells were fixed with 2% paraformaldehyde and 2.5% glutaraldehyde (v/v) dissolved in 0.1M PB for two hours in the dark and room temperature. Samples were then washed again with 0.1M PB and dehydrated sequentially with increasing ethanol solutions (25, 50, 75 and 100% ethanol), 10 minutes for each solution. Once in 100% ethanol, samples were sent to Servei de Microscòpia (UAB) and immediately critical-point dried in a K850 Critical Point Dryer (Emitech) and then they were sputter coated with gold. Samples were observed using a MERLIN FE-SEM microscope (ZEISS).