Evaluation of the surface lipoprotein assembly modifier (Slam) as a vaccine candidate against Neisseria gonorrhoeae infection

Abstract

The surface lipoprotein assembly modifier (Slam) plays an essential role in the biogenesis and assembly of outer membrane components in Gram-negative bacteria. This study evaluated the potential of recombinant NG Slam as a vaccine candidate. Bioinformatics confirmed its high conservation (including with its Neisseria meningitidis homolog) and outer membrane localization. Immunofluorescence demonstrated antibody accessibility to Slam in fixed bacterial cells. Antigenic epitope mapping revealed that dominant B- and T-cell epitopes are primarily within α-helical and random coil regions. Molecular docking suggested potential interactions with Toll-like receptors TLR2 and TLR4, which warrant experimental validation. The recombinant protein corresponding to the C-terminal surface-exposed domain of Slam was successfully expressed and used to immunize BALB/c mice. Immunization elicited high titers of specific antibodies with potent complement-dependent serum bactericidal activity and significant inhibition of bacterial adhesion to epithelial cells. Furthermore, splenocyte proliferation and enhanced secretion of IL-4, IL-17 A, and IFN-γ were observed, with IL-17 A production markedly exceeding that induced by whole-cell antigens. These findings demonstrate that Slam is immunogenic and elicits functional antibody alongside a Th17-skewed cellular response, the protective relevance of which requires further evaluation in challenge models.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-48171-5.

Introduction

Gonorrhea, caused by Neisseria gonorrhoeae (NG), remains a global public health threat with rising antimicrobial resistance, underscoring the urgent need for an effective vaccine^1–3. Surface-exposed lipoproteins are key virulence factors and promising vaccine antigens in Neisseria, as demonstrated by the success of factor H-binding protein in meningococcal vaccines^4,5.

The surface display of many such lipoproteins depends on the Surface Lipoprotein Assembly Modifier (Slam), an outer membrane β-barrel translocon that is conserved in Neisseria⁵. In N. meningitidis, Slam is essential for transporting crucial virulence factors like TbpB and fHbp to the cell surface, and its deletion attenuates virulence. Slam homologs are widespread among Gram-negative bacteria, indicating a conserved functional mechanism across species^5–7.

Despite its established biological role, the potential of Slam itself as a vaccine target remains unexplored. In NG, the Slam ortholog (NGO555, UniProt: WP_010951081.1) is functionally uncharacterized and remains an unreviewed entry with limited annotation in the UniProt database. Given its predicted surface localization, critical function in virulence factor assembly, and high sequence conservation, we hypothesized that Slam could serve as a novel vaccine antigen against gonorrhea.

This study therefore aimed to comprehensively evaluate the immunogenicity and protective efficacy of recombinant NG Slam as a vaccine candidate. Our integrated strategy combined bioinformatics analysis, assessment of Slam’s antibody accessibility on native bacteria, and detailed evaluation of the immune responses elicited in a mouse model. The results demonstrate that Slam is an accessible target and induces potent functional antibodies along with a Th17-biased cellular response, supporting its development as a novel vaccine candidate against gonorrhea.

Materials and methods

Materials

Experimental animals

Fifteen specific pathogen-free (SPF) female BALB/c mice (6–8 weeks old) were purchased from Sibeifu Biotechnology Co., Ltd. (Beijing, China). All experimental protocols were approved by the Institutional Animal Care and Welfare Ethics Committee of Dali University (Approval Number: 2024-PZ-028) and conducted in accordance with the Guide for the Care and Use of Laboratory Animals and the Chinese National Standard GB/T 39,760 − 2021. This study is reported in accordance with the ARRIVE guidelines.

Main reagents and instruments

NG strain FA1090 (ATCC). plasmid extraction kit (TianGen Bio-Chemical Technology Co., Ltd.). Restriction endonucleases BamH I and Xho I (Thermo Fisher Scientific). Iron Hammer™ Super Lysis Buffer (Boyi Biotechnology Co., Ltd.). Mouse anti-His tagged monoclonal antibody (Beijing Solabio Technology Co., Ltd.). HRP-conjugated goat anti-mouse IgG and IgA secondary antibodies (Abcam Co., Ltd.). FITC-conjugated goat anti-mouse IgG (H + L) (Proteintech Co., Ltd.). Freund’s complete and incomplete adjuvants (Sigma-Aldrich, USA). CCK8 assay kit and red blood cell lysis buffer (Biyun Tian Biotechnology Co., Ltd.). ELISA kits for IL-4, IL-17 A, and IFN-γ (biotin-streptavidin detection system; 4 A Biotech Co., Ltd.). Guinea pig serum (Beijing Solabio Technology Co., Ltd.). E. coli Rosetta (DE3) competent cells, pET32a(+) expression vector, and mouse anti-NG polyclonal antibody were prepared and stored in this laboratory. Fully automated gel imaging system (Syngene Limited, UK), ultrasonic cell disruptor (Xinzhi Biotechnology Co., Ltd.), multi-mode microplate reader (Gene Co., Ltd.), and fluorescence microscope Leica DM2000 LED (Leica Microsystems Co., Ltd.).

Methods

Functional annotation and epitope analysis of Slam

Structural analysis and functional annotation

The amino acid sequence of Slam from NG FA1090 (UniProt accession: WP_010951081.1) was analyzed using BLAST (NCBI) for homology searches and Clustal Omega for multiple sequence alignment. Subcellular localization was predicted using SignalP-6.0, DeepTMHMM, and LipoP. Functional annotation was performed using NetGO 4.0 within the Gene Ontology (GO) framework, covering biological process, molecular function, and cellular component categories⁸.

Antigenic epitope prediction

The Protean module in DNAStar was used to predict the B-cell antigenic epitopes of Slam. Hydrophilicity was analyzed using the Hopp-Woods and Kyte-Doolittle algorithms, solvent accessibility was evaluated with the Emini Surface Accessibility algorithm, conformational flexibility was assessed via the Karplus-Schulz algorithm, and antigenicity indices were calculated using the Jameson-Wolf algorithm.Furthermore, the IEDB online platform was employed to refine the prediction of dominant B-cell epitopes. The cytotoxic T lymphocyte (CTL) and helper T (Th) epitopes of Slam were comprehensively predicted by integrating MHC-I and MHC-II binding servers from the IEDB platform⁹ with the SYFPEITHI tool. Higher prediction scores indicate a greater likelihood of epitope dominance in the immune response.

Structural and epitope visualization

Physicochemical properties, secondary structure, and domain composition were analyzed using ProtParam, NPS@SOPMA, and InterPro. Tertiary structure modeling was performed with SWISS-MODEL and validated using ProSA-web and SAVES v6.0. Predicted epitopes were mapped onto the tertiary structure using PyMOL 2.6.

In silico prediction of Slam interaction with Toll-like receptors

To generate hypotheses regarding the potential mechanism of immune activation by Slam, protein-protein docking was performed using the HDOCK web server¹⁰(http://hdock.phys.hust.edu.cn). The SWISS-MODEL-predicted tertiary structure of Slam served as the ligand. The crystal structures of human TLR2 (PDB ID: 6NIG) and the TLR4/MD-2 complex (PDB ID: 3FXI) were retrieved from the RCSB Protein Data Bank and used as receptors. Docking simulations were run under default HDOCK parameters, which employ an FFT-based global search algorithm followed by iterative scoring using a knowledge-based potential function. The top-ranked binding conformations for each complex were selected for analysis of the predicted interaction interfaces. HDOCK produces a confidence score ranging from 0 to 1, with scores > 0.7 indicating reliable docking poses.

Prokaryotic expression and purification of recombinant Slam

Construct design, plasmid construction, and verification

A DNA fragment encoding the C-terminal β-barrel domain of Slam (residues 199–497, corresponding to amino acids 199–497 of the full-length protein) was synthesized and cloned into the pET32a(+) vector to generate pET32a-slam. The construct was transformed into E. coli Rosetta (DE3) competent cells. Positive clones were selected by PCR and confirmed by restriction digestion and sequencing.

Expression and purification

Recombinant protein expression was induced with 0.1 mM IPTG at 16 °C overnight. Cells were lysed, and inclusion bodies were solubilized in denaturing buffer (8 M urea). The protein was refolded by stepwise dialysis and purified by Ni-NTA affinity chromatography using imidazole elution (5–80 mM). Purification was monitored by SDS-PAGE.

Western blot analysis

Purified protein was separated by SDS-PAGE, transferred to PVDF membrane, and probed with anti-His tag and anti-NG polyclonal antibodies.

Immunogenicity assessment

Preparation of NG whole-cell protein antigen

NG strain FA1090 was cultured on TM agar at 37 °C under 5% CO₂ for 18–20 h. Bacterial cells were harvested with 2 mL of sterile saline, washed twice by centrifugation (6,000 × g for 2 min at 4 °C), and resuspended. A 500 µL aliquot of the bacterial suspension was sonicated on ice for 10 min (5 s pulse on, 5 s pulse off), followed by centrifugation at 12,000 × g for 15 min at 4 °C. The supernatant was collected as the whole-cell lysate. Protein concentration was determined using a BCA assay, adjusted to 10 mg/mL with PBS (pH 7.4), and stored at − 20 °C for use as the positive control immunogen in subsequent immunization experiments.

Mouse immunization

Female BALB/c mice (6–8 weeks, n = 5/group) were immunized with 50 µg Slam recombinant protein, 50 µg NG whole-cell protein (positive control), or PBS (negative control) via subcutaneous and intramuscular routes on days 0, 14, and 28, using complete Freund’s adjuvant (CFA) for priming and incomplete Freund’s adjuvant (IFA) for boosting. Seven days post-final immunization (day 35), mice exhibited normal body weights across all groups (Slam-immunized: 18.6 ± 0.46 g; positive control: 18.0 ± 0.33 g; negative control: 19.0 ± 0.35 g). Under isoflurane anesthesia (3–5% induction, 1.5–2.5% maintenance), blood was collected retro-orbitally, sera were separated, and stored at -80 °C. Mice were then euthanized by cervical dislocation under deep anesthesia (Chinese National Standard GB/T 39760 − 2021), and spleens were aseptically harvested for immunological assays.

IgG and IgA detection by ELISA

Microplates were coated with recombinant Slam or whole-cell NG antigen. Serum antibodies were detected using HRP-conjugated anti-mouse IgG/IgA and TMB substrate. Antibody titers were defined as the highest dilution yielding absorbance 2.1-fold above negative control.

Lymphocyte proliferative capacity assay

Splenocytes (2 × 10⁶ cells/mL) were stimulated with 10 µg/mL Slam or whole-cell antigen. After 72 h, CCK-8 reagent was added, and absorbance was measured at 450 nm. Stimulation index (SI) was calculated as:

Cytokine measurement

Supernatants from stimulated splenocytes were collected at 24 h (IL-4, IL-17 A) and 72 h (IFN-γ). Cytokine levels were quantified using commercial ELISA kits.

Immunoprotective evaluation

Serum bactericidal assay

NG was cultured to mid-log phase and resuspended to 1 × 104 CFU/mL. Heat-inactivated immune sera (56 °C, 30 min) were serially diluted and first incubated with NG for 30 min at 37 °C to allow antibody binding. Exogenous complement (10% guinea pig serum, a standard source in Neisseria SBA studies) was then added, followed by a second 30 min incubation. All incubations were performed in 5% CO₂ at 37 °C to maintain NG viability. Controls included active complement control (bacteria with active complement but no immune serum), heat-inactivated complement control (to account for any complement-independent bactericidal activity), serum control (immune serum with inactivated complement), and bacterial growth control. Aliquots (50 µL) were plated on TM agar and cultured at 37 °C in 5% CO₂ for 36–48 h. The bacterial survival percentage for each serum dilution was calculated as (CFU at dilution / CFU of heat-inactivated complement control) × 100%. The bactericidal titer was then defined as the reciprocal of the highest serum dilution achieving ≤ 50% bacterial survival (i.e., ≥ 50% killing) relative to the heat-inactivated complement control. Titers are expressed as − log₁₀ values.

Serum antibody adhesion inhibition assay

HeLa cells (1 × 10⁵ cells/well) were seeded in 24-well plates and allowed to form a monolayer. NG was grown to mid-log phase, resuspended, and pre-incubated with heat-inactivated immune serum (at a dilution of 1:40, determined by prior titration) for 30 min at 37 °C. The bacteria-serum mixture was then added to the HeLa cells at a multiplicity of infection (MOI) of 100. After 3 h of co-incubation at 37 °C with 5% CO₂, non-adherent bacteria were removed by washing three times with PBS. The adherent bacteria were released by lysing the HeLa cells with 1% saponin, serially diluted, and plated on GCB agar for colony counting. An untreated control (bacteria incubated without any serum) was included in each experiment and set as 100% adhesion. The adhesion rate for each serum-treated group was calculated as (CFU of the serum-treated group / CFU of the untreated group) × 100%. The adhesion inhibition rate was then derived as 100% minus the adhesion rate.

Assessment of Slam antibody accessibility by immunofluorescence

To assess the accessibility of Slam to specific antibodies on NG, an indirect immunofluorescence assay was performed. Briefly, NG strain FA1090 cells in the logarithmic growth phase were smeared onto flame-sterilized glass slides, air-dried, and fixed with acetone for 10 min. The fixed samples were then incubated in a humidified chamber with primary antibodies (1:20 dilution) at 37 °C for 2 h. The primary antibodies used were serum from Slam-immunized mice (experimental group), serum from whole-cell NG protein-immunized mice (positive control), and serum from PBS-immunized mice (negative control). Following three 5-min washes with PBS, the slides were incubated with FITC-conjugated rabbit anti-mouse secondary antibody (1:300 dilution) at 37 °C in the dark for 2 h. After three additional PBS washes, fluorescence was visualized using a fluorescence microscope. This assay was designed to evaluate the binding capacity of Slam-specific antibodies to their target in a whole-bacterium context. Since acetone fixation permeabilizes bacterial membranes and allows antibody access to internal compartments, a positive signal, while confirming Slam accessibility, does not alone permit precise discrimination between its surface-exposed and periplasmic localization.

Statistical analysis

Antibody titers are expressed as –log₁₀. Proliferation and cytokine data are presented as mean ± standard deviation ( ± SD). Group comparisons were made using one-way ANOVA in GraphPad Prism 8.0, with P < 0.05 considered significant.

Results

Bioinformatics analysis of Slam

Homology analysis revealed that Slam from NG FA1090 (WP_010951081.1) shares 91% amino acid identity with its homolog from N. meningitidis (accession: WP_118019856.1) (Fig. 1A). SignalP-6.0 predicted a signal peptide (probability: 99%) with a cleavage site between residues Ala18 and Ala19 (probability: 98%) (Fig. 1B). DeepTMHMM predicted 14 transmembrane helices, with seven extracellular and seven periplasmic segments, and residues Ala19-Trp198 localized to the periplasm (Fig. 1C). LipoP supported classification as a lipoprotein (score: 8.06) (Fig. 1D). Functional annotation identified five high-confidence molecular function terms and eight cellular component terms, including strong support for outer membrane localization (Fig. 1E, F). No high-confidence biological process terms were identified.

Fig. 1.

Bioinformatic characterization of the surface lipoprotein assembly modifier (Slam) from NG. (A) Multiple sequence alignment of Slam with its homolog from Neisseria meningitidis (WP_118019856.1). (B) Signal peptide prediction by SignalP-6.0. (C) Transmembrane topology prediction by DeepTMHMM. (D) Lipoprotein signal peptide prediction by LipoP. (E) Gene Ontology (GO) molecular function annotations. (F) GO cellular component annotations.

Antigenic epitope prediction

Integrated analysis identified 14 putative B-cell epitope regions (Fig. 2), characterized by high hydrophilicity, surface accessibility, flexibility, and antigenicity. To evaluate the potential cross-reactivity of these epitopes, we analyzed their conservation in Slam from 16 diverse NG strains (Table 1). Ten out of the 14 predicted B-cell epitopes (71.4%) were completely conserved across all strains. The remaining epitopes also showed high conservation (81.2–93.8%). This high degree of overall sequence conservation suggests that a vaccine based on the FA1090 Slam sequence has the potential to elicit broad protection against heterologous NG strains.

Fig. 2.

Prediction of B-cell epitopes in Slam. (A) Composite epitope prediction profile generated by DNAStar, integrating analyses of hydrophilicity, surface accessibility, flexibility, and antigenicity. (B) Linear representation of the protein sequence showing the fourteen predicted dominant B-cell epitope regions (highlighted in yellow) as identified by the IEDB analysis.

Table 1.

Conservation analysis of the predicted B-cell epitopes across 16 NG strains.

Epitope	Position (aa) (FA1090)	Conserved strains	Conservation (%)	Predominant amino acid substitution(s)
1	Pro20 - Glu34,	15/16	93.8%	Ser21Pro, Ala24Pro
2	Arg37 - Gly62	15/16	93.8%	Gly62Ser
3	Tyr71 - Glu78	16/16	100%	None
4	Leu106 - His109	16/16	100%	None
5	Pro143 -Asp144	16/16	100%	None
6	Asn160 - Glu164	16/16	100%	None
7	Ala179 - Val185	15/16	93.8%	Arg182Gly
8	Gly193 - Trp198	16/16	100%	None
9	Ala237 - Thr242	16/16	100%	None
10	Asn271 - Ser280	16/16	100%	None
11	Ala300 - Arg303	16/16	100%	None
12	Gly352 - Ser357	16/16	100%	None
13	Ala390 - Lys402	16/16	100%	None
14	Ser438 - Lys446	13/16	81.2%	Arg444Ser

Furthermore, prediction with IEDB and SYFPEITHI identified two high-score cytotoxic T lymphocyte (CTL) epitopes and four helper T (Th) epitopes (Table 2). Together, these results indicate a broad immunogenic coverage of the Slam protein.

Table 2.

Predicted dominant CTL and Th epitopes in Slam.

Dominant epitope type	Site	Sequence	IEDB score	SYFPEITHI score
CTL	65–73	SIGGEIYQV	0.90	27
CTL	177–185	IPAETRPIV	0.83	20
Th	125–139	EGDFRAAGNSFQTAL	0.94	28
Th	433–447	AASFVSDNKRRRDKQ	0.91	26
Th	445–459	DKQYIMMAAAGFPQW	0.87	35
Th	243–257	FSSYSATAEKTVPLK	0.80	32

Structural modeling and epitope localization

The full-length slam gene (1,494 bp) encodes a 497-residue protein with a predicted molecular mass of 54 kDa. Secondary structure analysis of the full-length protein indicated a composition of 52.31% random coils, 36.42% α-helices, and 11.27% extended strands, suggesting structural flexibility conducive to antigenic exposure (Fig. 3A, B). A high-quality tertiary model was obtained (ProSA-web Z-score: − 5.63) (Fig. 3C) and validated by Ramachandran analysis (99.8% residues in allowed regions) (Fig. 3D). Domain architecture comprised an N-terminal signal peptide (residues Met1 to Ala18), a central TPR domain (Ala19 to Trp198), and a C-terminal β-barrel belonging to DUF560 (His199 to Phe497) (Fig. 3E). Structural localization suggested a periplasmic orientation for the TPR domain and surface exposure of the β-barrel (Fig. 3F). Predicted B- and T-cell epitopes were predominantly located in α-helical and random coil regions (Fig. 3G), supporting the immunogenic potential of Slam. Notably, the majority of predicted B- and T-cell epitopes map to the C-terminal β-barrel domain (residues 199–497), which is predicted to be surface-exposed, whereas the N-terminal TPR domain (residues 19–198) resides in the periplasm (Fig. 3F, G). This bioinformatic analysis suggested that the β-barrel domain represents the most accessible and immunologically relevant region of Slam. Therefore, for subsequent recombinant expression and immunization studies, we focused on this domain.

Fig. 3.

Structural model and epitope mapping of Slam. (A) Graphical output of the NPS@SOPMA secondary structure prediction. (B) Linear schematic of secondary structure elements. (C) ProSA-web Z-score plot for the tertiary structure model. (D) Ramachandran plot from SAVES v6.1 quantifying residue distribution in favored, allowed, and outlier regions. (E) Domain architecture predicted by InterPro. (F) Predicted membrane integration model illustrating periplasmic orientation of the Ntd and surface exposure of the β-barrel domain. (G) Tertiary structure model colored by epitope prediction: B-cell (red) and T-cell (blue) epitopes, showing colocalization with flexible regions.

In silico prediction of potential Slam interaction with TLR2 and TLR4

We performed computational docking to explore whether Slam could potentially engage with key innate immune receptors. The docking analysis predicted stable complexes with both TLR2 and TLR4, with HDOCK scores of 0.9991 for TLR2 and 0.9986 for TLR4 (Fig. 4). The modeled binding interfaces suggested complementary electrostatic and hydrophobic interactions. These in silico results propose a hypothetical mechanism wherein Slam could act as a ligand for TLR2 and/or TLR4; however, this prediction necessitates direct experimental confirmation through biophysical and functional assays.

Fig. 4.

In silico docking prediction of Slam with Toll-like receptors. Modeled. Modeled binding conformations between Slam and (A) TLR2 or (B) TLR4 generated by computational docking. High docking scores (> 0.99) suggest the possibility of complex formation.

Recombinant Slam expression and purification

Based on the bioinformatic prediction that the C-terminal β-barrel domain is surface-exposed and contains the majority of predicted epitopes (Fig. 3G), we selected this region as the immunogen for recombinant expression. This truncated construct has a predicted molecular mass of ~ 34 kDa (excluding the fusion tag) and ~ 52 kDa when including the N-terminal TrxA-His₆ tag from the pET32a(+) vector.

The recombinant plasmid pET32a-slam was successfully constructed and verified by sequencing and restriction digestion (Fig. 5A). Expression in E. coli yielded a protein of ~ 52 kDa, consistent with the predicted size including the fusion tag, with predominant accumulation in inclusion bodies (Fig. 5B). Purification under denaturing conditions followed by refolding yielded highly pure protein (Fig. 5C). Western blot confirmed specific reactivity with anti-His and anti-NG polyclonal antibodies (Fig. 6). Notably, the mouse anti-NG polyclonal antiserum (raised against whole-cell protein) also specifically recognized the recombinant Slam protein (Fig. 6B), This confirms that immunization with whole-cell NG antigen elicits antibodies against SLAM. No reactivity was observed with serum from PBS-immunized mice (Fig. 6C).

Fig. 5.

Construction, expression, and purification of recombinant Slam C-terminal domain. (A) Identification of the pET32a-slam plasmid by restriction digestion. Lane M: DNA marker with sizes bp; Lane 1: Empty pET32a(+) vector; Lane 2: Recombinant plasmid digested with BamHI and XhoI, releasing the ~ 897 bp insert. (B) SDS-PAGE analysis of expression and solubility. Lane M: Protein marker with sizes (kDa) indicated on the left; Lane 1: Total protein from IPTG-induced cells; Lane 2: Soluble fraction; Lane 3: Insoluble fraction (inclusion bodies). Lane 4: affinity chromatography flow-through; Lane 5: affinity chromatography binding fraction (intercepted from the nickel column) used to verify whether the target protein successfully bound to the affinity resin; Lanes 6–9: flow-through fractions after refolding with Solutions I-IV, respectively, demonstrating no protein loss during refolding. (C) SDS-PAGE analysis of Ni-NTA purification under denaturing conditions. Lane M: Protein marker with sizes (kDa) indicated on the left; Lane 1–3: Successive washes with 5, 10, and 20 mM imidazole; Lane 4: Elution with 80 mM imidazole; Lane 5: Concentrated protein from the 80 mM eluate.

Fig. 6.

Immunoblot validation of recombinant Slam. Purified recombinant Slam was probed with (A) anti-His tag monoclonal antibody, (B) mouse anti-NG polyclonal antiserum, and (C) serum from PBS-immunized mice (negative control). Lane M: prestained protein marker with molecular weights (kDa) shown on the left in all panels; Lane 1: Purified recombinant Slam in all panels.

Immunogenicity of recombinant Slam

Serum antibody responses

Immunization with recombinant Slam elicited significantly high IgG and IgA titers compared to the negative control (P < 0.05). While IgG titers against whole-cell NG antigen were lower than the positive control (P < 0.05), IgA responses were comparable, indicating the induction of antibodies recognizing native NG antigens (Fig. 7).

Fig. 7.

Antigen-specific serum antibody responses. Endpoint titers of (A) IgG and (B) IgA in mice immunized with recombinant Slam, whole-cell NG protein (positive control), or PBS (negative control). Sera from the Slam group were tested against both recombinant Slam and NG whole-cell lysate; control groups were tested against NG whole-cell lysate. The negative control (PBS) bar represents the calculation baseline for endpoint titer determination (see Methods); therefore, error bars are not applicable. Data are presented as -log₁₀ of the endpoint titer. “a” indicates P < 0.05 versus the negative control group; “b” indicates P < 0.05 between the Slam and positive control groups.

Notably, despite using only the C-terminal β-barrel domain as the immunogen, the elicited antibodies recognized native NG whole-cell antigens (Fig. 7), indicating that this domain contains key epitopes accessible on the bacterial surface. This validates the rational truncation strategy and confirms that the β-barrel domain is sufficient to induce antibodies targeting native Slam.

Lymphocyte proliferation and cytokine secretion

Stimulation with Slam induced a significantly higher stimulation index (SI) in splenocytes from the Slam group compared to both control groups (P < 0.05). Cytokine analysis showed significantly enhanced secretion of IL-4, IL-17 A, and IFN-γ compared to the negative control. Notably, the Slam group exhibited significantly higher IL-17 A and lower IL-4 levels than the positive control (P < 0.05), suggesting a Th17-biased immune polarization (Fig. 8).

Fig. 8.

Slam-specific T cell immune responses in immunized mice. Splenocytes isolated from mice immunized with recombinant Slam, whole-cell NG protein (positive control), or PBS (negative control) were stimulated ex vivo with recombinant Slam or NG whole-cell lysate. (A) Lymphocyte proliferation was measured by CCK-8 assay and expressed as the Stimulation Index (SI). (B-D) Concentrations of IL-4 (B), IL-17 A (C), and IFN-γ (D) in culture supernatants were quantified by ELISA. Data are presented as mean ± SD. The symbol “a” indicates P < 0.05 versus the negative control group; “b” indicates P < 0.05 versus the positive control group.

Functional protective effects of Slam-induced immunity

Sera from Slam-immunized mice exhibited complement-dependent bactericidal activity comparable to the positive control and significantly greater than the negative control (P < 0.05) (Fig. 9A). In the adhesion inhibition assay, the adhesion level of the untreated control (bacteria without serum pre-incubation) was defined as 100%. Relative to this baseline, sera from PBS-immunized mice (negative control) exhibited a non-specific inhibition of bacterial adhesion (approximately 40%). This background effect, likely attributable to innate serum components such as complement, is commonly observed in similar in vitro adhesion assays¹¹. Importantly, sera from Slam-immunized mice demonstrated a significantly greater inhibition of NG adhesion to HeLa cells than both the negative and positive controls (P < 0.05) (Fig. 9B), indicating a specific and potent antibody-mediated protective effect beyond any non-specific interference.

Fig. 9.

Functional protective effects of Slam-induced antibodies. (A) Complement-dependent serum bactericidal activity. The 50% bactericidal titer (− log₁₀) against NG in the presence of exogenous complement is shown. (B) Serum antibody-mediated inhibition of bacterial adhesion to HeLa cells. The adhesion inhibition rate was calculated as described in Methods (section “Serum antibody adhesion inhibition assay”), with the adhesion level of the untreated control (bacteria without serum) set as 100%. Data are presented as mean percentage inhibition ± SD. The symbol “a” indicates P < 0.05 versus the negative control (PBS) group; “b” indicates P < 0.05 versus the positive control (whole-cell NG protein) group.

These functional activities demonstrate that the C-terminal β-barrel domain contains protective epitopes capable of inducing antibodies that interfere with critical steps in NG pathogenesis, further validating the domain-targeted immunization strategy.

Antibody accessibility of Slam assessed by immunofluorescence

To complement the bioinformatic predictions, we performed an indirect immunofluorescence assay (IFA) to assess the antibody accessibility of Slam in N. gonorrhoeae. NG cells incubated with antiserum from Slam-immunized mice exhibited strong and specific peripheral fluorescent labeling (Fig. 10A). The signal intensity was comparable to the positive control using anti-whole cell NG serum (Fig. 10B) and distinctly different from the negative control with PBS-immunized serum (Fig. 10C). These results demonstrate that Slam is accessible to specific antibodies in fixed bacterial cells. Coupled with the consistent bioinformatic predictions of its outer membrane lipoprotein nature (Fig. 1), the IFA data indicate the potential of Slam as an immune-accessible target.

Fig. 10.

Immunofluorescence detection of Slam on NG. (A) Specific labeling with anti-recombinant Slam serum. (B) positive control using anti-whole cell NG serum. (C) negative control with PBS-immunized serum. All samples were incubated with an FITC-conjugated rabbit anti-mouse secondary antibody. The scale bar represents 20 μm.

Discussion

The high reinfection rate following natural NG infection underscores the lack of durable, sterilizing immunity, largely due to the pathogen’s sophisticated immune evasion mechanisms^12–15.This reality necessitates the exploration of novel vaccine targets, particularly highly conserved outer membrane proteins essential for bacterial survival and virulence². Reverse vaccinology has proven successful in this endeavor, as exemplified by the meningococcal B vaccine^16,17, and guided our initial identification of Slam (NGO0555) as a candidate.

Our findings establish Slam as a natural target of the humoral immune response against NG. We observed a reciprocal antigenic relationship: antibodies induced by recombinant Slam recognize native NG antigens (Fig. 7), and conversely, immunization with whole-cell NG antigens elicits antibodies that specifically bind recombinant Slam (Fig. 6B). This result is particularly relevant in the context of outer membrane vesicle (OMV) vaccines, which contain a complex mixture of native membrane proteins^18–20. The presence of immunogenic Slam in OMVs likely contributes to the functional antibody responses elicited by such platforms. Therefore, targeted immunization with recombinant Slam may offer a strategy to focus and potentially enhance immunity against this conserved, functionally critical antigen.

Our study builds upon the foundational studies of the Slam protein family. In N. meningitidis, Slam was identified as an integral outer membrane protein essential for the surface display of key lipidated virulence factors, such as TbpB, fHbp, and HpuA, through a dedicated translocation pathway⁶. Subsequent bioinformatic analyses indicate that Slam homologs and their dependent surface lipoproteins (SLPs) constitute a widespread “plug-and-play” system across diverse Gram-negative pathogens⁷. Here, we extend this knowledge to NG, providing a comprehensive evaluation of Slam’s potential as a vaccine antigen against gonorrhea.

Consistent with its role in N. meningitidis, our bioinformatic analyses strongly predict NG Slam to be an outer membrane lipoprotein. Although immunofluorescence confirmed the accessibility of Slam to antibodies in fixed bacteria (Fig. 10), we acknowledge that acetone fixation permeabilizes all membranes, limiting the ability to definitively distinguish surface‑exposed from periplasmic localization. More precise techniques, such as proteinase K surface shaving or subcellular fractionation—as employed in seminal Slam studies⁶—would be required for nanometer-scale mapping. Nevertheless, the combined evidence from computational prediction, antibody accessibility, and the functional immune profile supports Slam as a target accessible to the immune system.

The role of Th17 immunity in NG infection is complex and context-dependent. While IL-17 is crucial for neutrophil recruitment and mucosal defense^21,22, and its induction is generally viewed as a positive indicator for mucosal vaccines, NG appears to exploit this pathway. The pathogen may augment Th17 responses while suppressing protective Th1/Th2 pathways, thereby promoting inflammatory tissue damage that facilitates its survival and transmission without eliciting lasting protective memory^13,23,24. Thus, the Th17 bias induced by Slam immunization warrants careful interpretation. Although the Freund’s adjuvant (CFA/IFA) used here contains components (e.g., trehalose dimycolate) known to promote Th17 responses^25–27, the fact that Slam immunization elicited significantly higher IL-17 A levels than whole-cell NG protein immunization under identical adjuvant conditions suggests that the polarization is primarily driven by Slam’s intrinsic immunogenic properties—potentially linked to its predicted interactions with TLR2/TLR4. Whether this represents a desirable mucosal adjuvant effect or carries a risk of immunopathology requires further evaluation in challenge models.

In the context of vaccine development, Slam presents a compelling profile. Its high conservation with the meningococcal homolog and its surface accessibility are advantageous. The functional antibodies it induces target key virulence processes, such as iron acquisition (via potential SLP interactions) and adhesion. Importantly, if anti‑Slam antibodies can block the protein’s transport function, they could inhibit the surface display of multiple Slam‑dependent SLPs, thereby crippling NG’s ability to acquire iron and evade complement—a multi‑mechanistic inhibitory effect. This approach differs from targeting individual, often variable SLPs and may offer broader protective advantage.

The use of a truncated immunogen corresponding to the surface-exposed C-terminal β-barrel domain (residues 199–497) was supported by structural prediction and validated by functional data: antibodies elicited by this construct recognized native NG antigens and mediated potent bactericidal and adhesion-inhibitory activities. These results confirm that key protective epitopes reside within this domain and align with the epitope mapping analysis (Fig. 3G). Future studies comparing full-length and truncated immunogens could further elucidate the breadth of immune responses.

In conclusion, this study validates Slam as a novel and immunogenic vaccine candidate against NG. It elicits a multifaceted immune response comprising functional antibodies and a pronounced cellular component. While the observed Th17 polarization highlights the need for careful adjuvant selection and efficacy assessment, the strategic targeting of a conserved outer membrane assembly factor such as Slam offers a promising avenue for disrupting multiple virulence pathways. Future work should focus on confirming its precise localization, elucidating its interactions with NG SLPs, and evaluating its protective efficacy in stringent mucosal challenge models to fully realize its vaccine potential.

Limitations and future perspectives

This study has several limitations that point to important future directions. First, while immunofluorescence confirmed antibody accessibility, more definitive techniques such as immunogold electron microscopy or surface proteolysis are required to conclusively demonstrate surface exposure of Slam. Second, the interaction between Slam and TLR2/4 remains a computational prediction; future studies using surface plasmon resonance (SPR) to measure binding affinity, along with functional assays in TLR‑deficient cellular or animal models, are essential to validate this interaction and clarify its role in immunogenicity. Finally, the protective efficacy of Slam immunization and the net impact of the Th17-polarized response must be evaluated in a rigorous mucosal challenge model to establish translational relevance.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Ya He: Writing—original draft, methodology, conceptualization, data curation, validation, and formal analysis. Yu Huang: Methodology, Writing review, and editing. Feng Dong: Methodology, Resources. Weiyuan Wang: Methodology. **Li** Zhang: Writing review and editing, conceptualization, supervision. Lei Zhang: Writing review and editing, conceptualization, supervision, project administration, resources, and funding acquisition.

Funding

This work was supported by the National Natural Science Foundation of China (Grant No. 81760303).

Data availability

All data supporting the findings of this study are available from the corresponding author upon reasonable request.

Competing interests

The authors declare no competing interests.

Ethics approval

This study was approved by the Animal Ethics Committee of Dali University (Approval No. 2024-PZ-028).