Sera collection

A total of 162 serum samples, collected from Tunisian patients of reproductive age between 2019 and 2024, were included in this study. All samples had previously been tested using an in-house IIF assay based on crude total antigens derived from urogenital mycoplasmas (MH, Uspp., Mycoplasmoides genitalium, and Mycoplasmopsis fermentans), as well as Mycoplasmoides pneumoniae, included to assess cross-reactivity (Supplementary information Fig. S1).

The IIF technique showed positivity rates of 10.5%, 27%, and 38% for MH, UP, and UU, respectively. All sera included in this study were negative for M. genitalium, M. fermentans and M. pneumoniae. Western blot analyses were performed to verify the IIF results and to characterize the antigenic recognition profiles of patient sera using total antigens from all tested urogenital mycoplasma species and M. pneumoniae.

Bacterial strains and culture conditions

Reference Mycoplasma strains; MH PG21 (ATCC 23114), UU (ATCC 27618) and UP (ATCC 27813) were cultured in SP4 liquid and solid media as previously described [8]. Culture and antigen preparation were performed as previously described [9] and were stored at −20 °C until use. Escherichia coli (E. coli) strains DH5α and BL21 (Promega, USA) were transformed with recombinant plasmids and grown at 37 °C in 2YT media supplemented with 100 µg/ml of ampicillin [10].

Cloning of the genes encoding the major surface proteins of M. hominis, U. parvum, and U. urealyticum

Polymerase chain reaction (PCR) was used to amplify the genes encoding major surface proteins of mycoplasma species of interest. Specific primers targeting conserved regions of each surface protein were used (Table 1). The MBA-derived recombinant proteins (MBA3 and MBA8) were designed based on conserved regions of the N-terminal domain, which are shared across multiple serovars. This approach was intentionally chosen to enhance cross-serovar reactivity and minimize bias toward specific serovars. To validate this strategy, sequence alignments were performed against representative serovars of UP (SV1, SV3, SV6, SV14) and UU (SV2, SV4, SV5, SV7–13) using GenBank reference sequences, confirming a high degree of conservation within the selected regions (Supplementary information Fig. S2 and S3).

Table 1 Oligonucleotide primers used for the amplification of the genes encoding the major surface proteins of M. hominis, U. parvum, and U. urealyticum

The same approach was used to design the Vaa and P120’derived recombinant proteins; conserved regions of the N-terminal domain were targeted. Sequence alignments were performed for each protein using available ATCC reference strains retrieved from GenBank (Supplementary information Fig. S4 and S5). The PCR was conducted in a thermocycler 2720 (Applied Biosystem, USA) under the following conditions: 35 cycles of denaturation at 95 °C for 45 s (s), annealing at 56 °C for 45 s and elongation at 72 °C for 45 s in a reactional mix containing 2.5 mM of MgCl2 (Invitrogen, USA), 10 mM of dNTP (Thermo scientific, Lithuania), 20 µM each primer (forward and reverse) (Carthagenomics, Tunisia) and 5 U/µl Taq polymerase (Invitrogen, USA).

PCR products were then purified using the QIAquick PCR Purification Kit (QIAGEN, Germany) before digestion with XhoI and BamHI (Thermo Scientific, Lithuania), whose corresponding sites were added at the 5’ end of the forward and reverse primers, respectively. The three purified genes were ligated to the pGEX-4T-1 plasmid (GE Healthcare) and then used to transform E. coli DH5α. The integrity of the recombinant plasmids was verified by PCR amplification and restriction enzyme analysis, and the correct reading frame was preserved by cloning into the pGEX-4T-1 vector; additionally, all PCR products were confirmed by Sanger sequencing to verify sequence accuracy and exclude frameshift mutations prior to protein expression. The recombinant plasmids were introduced into E. coli BL21. All E. coli strains were cultured overnight at 37 °C in 2YT medium supplemented with 100 µg/ml ampicillin to select for recombinant clones.

Recombinant protein expression and purification

Overnight culture of recombinant E. coli BL21 was diluted 1:10 and incubated at 37 °C with agitation at 200 rpm. Once the optical density at 600 nm (A600) reached 0.6–0.8, protein expression was induced with 100 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG) (GE Healthcare, Lithuania), with incubation at 30 °C with agitation for 3 hours (GST-Vaa), 2.5 hours (GST-P120’), or 3.5 h for (GST-MBA3 and GST-MBA8). Cells were pelleted, resuspended in ice-cold PBS, and disrupted by sonication (three cycles of 10 s) using a Branson Sonifier 450 (Fisher Bioblock Scientific, Illkirch, France). Subsequently, lysates were centrifuged at 5000 rpm, and recombinant protein expression was confirmed by 10% SDS-PAGE.

The purification of the GST-tagged recombinant proteins was performed by affinity chromatography using Glutathione-Sepharose 4B beads from the GST Bulk Kit (GE Healthcare, USA) according to the manufacturer’s instructions. Protein purity and quality were confirmed by SDS-PAGE, while specificity and immunogenicity were assessed by Western blot, using rabbit-derived specific antisera against target proteins (anti-GST-MBA3, anti-GST-MBA8, anti-GST-Vaa and anti-GST-P120’).

Production of polyclonal antisera to GST-Vaa, GST-P120’, GST-MBA3 and GST-MBA8 proteins and M. hominis, U. parvum, and U. urealyticum crude antigens

The production of rabbit polyclonal antibodies against the GST-fusion recombinant proteins and the total antigens of MH, UU, and UP was performed as detailed elsewhere [10]. Briefly, New Zealand White rabbits were subcutaneously injected with 250 µg of each purified GST-tagged recombinant protein, notably, GST-Vaa, GST-P120’, GST-Mba3 and GST-Mba8. For the first administration, recombinant proteins were mixed with Freund’s complete adjuvant; for the additional 4 booster injections, the incomplete formulation was used. Booster injections were administered at 15-day intervals. The production of the hyperimmune sera against total antigens of each species, MH, UP and UU was conducted according to the protocol outlined by our group [9]. All antibody titers were determined by immunoblotting. The antisera samples were stored at −20 °C until further use. All rabbit immunization procedures and collection of sera were conducted in accordance with the protocol of the Center for Biologics Evaluation and Research/Food and Drug Administration Institutional Animal Care and Use Committee. The ethics statement number for this study is “2015/16/1/LR11IPT01/V0”.

Development of a crude antigen-based ELISA

Before the establishment of MUREAPLEXE, we initially developed an indirect ELISA based on the interaction between the crude antigens of MH, UP, and UU and patient sera. This test was mainly optimized for comparative purposes and to assess the reactivity of Tunisian patients’ antisera against the total proteins of the studied mycoplasma species.

ELISA optimization was performed using a checkerboard titration approach to systematically define the optimal assay conditions [10]. Serial antigen concentrations, ranging from 1000 to 7.8 ng/100 µL, were coated onto 96-well microplates (Sarstedt, USA) and incubated overnight at 4°C. Optimal antigen concentrations for MH (7.8 ng/100µL), UP (20 ng/100µL), and UU (15 ng/100µL) were determined using specific rabbit polyclonal antisera at 1:2000 dilution. Nonspecific binding sites were blocked following washing with a casein-based buffer, and different human sera dilutions (1:200, 1:400, 1:600, and 1:800) were evaluated for the optimal working dilution, followed by incubation with a horseradish peroxidase (HRP)-conjugated rat anti-human IgG (Abcam, UK), tested at 1:10,000, 1:15,000, and 1:20,000 dilutions. Each incubation step was optimized for duration and temperature. Signal detection was performed using 100 µl of the 3,3’,5,5’ tetramethyl-benzidine (TMB) (Abcam, UK) substrate, and the reaction was terminated with 100 µl of 7% sulfuric acid prior to optical density (OD450) measurement at 450 nm using MULTISKAN GO microplate reader (ThermoScientific).

The assay performance was assessed using the index of positivity (IP), as previously described by Yacoub et al. [11]. IP was defined as the sample OD to cut-off OD ratio, established using a reference positive serum with an OD at least threefold higher than the negative control.

$$IP=Absorbance\;of\;the\;tested\;serum\div Absorbance\;of\;the\;cut-off$$

Based on IP values, samples were classified as negative (IP < 0.7), positive (IP > 0.7), or borderline (IP = 0.7).

Development of the multiplex recombinant proteins-based ELISA: MUREAPLEXE

The same optimization strategy was applied for the development of MUREAPLEXE multiplex recombinant protein-based ELISA. Checkerboard titration allowed the determination of the optimal concentrations of GST-Vaa, GST-P120’, GST-MBA3, and GST-MBA8 recombinant antigens, as well as the appropriate serum and conjugate dilutions and incubation conditions, ensuring optimal assay sensitivity and specificity.

Statistical analysis

Statistical analyses were conducted to assess the correlation between MUREAPLEXE and the crude antigen-based ELISA. Pearson correlation analysis was used to calculate correlation coefficients and to generate regression models. Bland–Altman analysis was performed to assess agreement between the two assays. Diagnostic performance was evaluated using paired t-tests and receiver operating characteristic (ROC) curve analysis, from which sensitivity and specificity were estimated. Graphical outputs and statistical analyses were generated in RStudio (version 30.4.0).