Replacement of porB

We genetically engineered N. gonorrhoeae to remove the immunosuppressive properties of PorB and RmpM from our gonococcal vaccine. To generate N. gonorrhoeae FA1090 expressing a meningococcal porB, N. meningitidis MC58 porB (“Nm PorB”) was amplified by PCR, fused to a kanamycin resistance cassette and introduced into FA1090 by transformation (Fig 1A). N. meningitidis MC58 porB is a class 3 porB allele while N. gonorrhoeae FA1090 porB is a PorB.IB allele [44]; the proteins share 68% amino acid identity (S1 Fig). Insertion of Nm porB was confirmed by PCR and sequencing, demonstrating that the gonococcal porB gene had been entirely replaced with the meningococcal gene. The resultant strain, FA1090_{MC58 PorB}, had a reduced growth rate compared to wild type (WT) FA1090 (S2 Fig). Western blot analysis using PorB-specific typing monoclonal antibodies confirmed that FA1090_{MC58 PorB} expressed Nm PorB alone (Fig 1B). In addition, the rmpM gene was replaced with ermC’, and the deletion confirmed by PCR and Western blot analysis (Fig 1D). The deletion of rmpM exacerbated the growth defect of FA1090_{MC58 PorB} (S2 Fig).

Download:

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

Fig 1.

A) Generation of N. gonorrhoeae (Ng) FA1090 expressing MC58 Nm porB. Schematic representation of replacing the Ng porB gene with Nm porB, under the control of the native Ng promoter. Promoters shown in grey. Kanamycin resistance cassette (kanR) under its own promoter is present downstream of the open reading frame for selection. B) Western blots confirming the replacement of Ng PorB with Nm PorB, using mAbs against specific PorBs. C) Schematic representation of the deletion of rmpM in Ng FA1090 by replacement with an erythromycin resistance cassette (eryR). D) Western blot confirming deletion of rmpM. Arrows denote open reading frames. Dotted lines represent regions of homologous recombination. Figure not to scale.

https://doi.org/10.1371/journal.ppat.1012688.g001

Ng PorB OMVs and Nm PorB OMVs have additional proteome differences

OMVs were generated from FA1090ΔrmpM (Ng PorB OMVs) and FA1090_{Nm porB} ΔrmpM (Nm PorB OMVs). Upon analysis of the OMVs by denaturing SDS-PAGE, we observed different protein profiles of the Ng PorB OMVs and Nm PorB OMV preparations, in addition to the change in PorB. Differences in the OMV proteomes were consistent between independent batches (S3 Fig). To compare the OMV proteomes in detail, we analysed them by mass spectrometry. A total of 174 proteins were identified across three biological replicates of Ng PorB OMVs and Nm PorB OMVs (S2 Table). Proteins identified in only one sample, with low peptide number (< 2), or low overall coverage (< 7%), were eliminated from the analysis, leaving 86 proteins identified with high confidence (FDR of < 1%, S3 Table). Proteomic analysis confirmed that Ng PorB OMVs only contained FA1090 PorB, and Nm PorB OMVs only contained MC58 PorB. Other than PorB, one protein was found solely in Nm PorB OMVs, NGO_09965, an Opacity (Opa) family protein, and one protein was found solely in Ng PorB OMVs (NEIS0210, unknown function). A total of 31 out of the 86 proteins identified exhibited similar abundance in Nm PorB OMVs and Ng PorB OMVs, while 46 had a higher abundance (ratio ≥ 2) in Nm PorB OMVs compared to Ng PorB OMVs, with only four proteins found to have lower abundance (ratio ≤ 0.5) (Fig 2 and S3 Table).

Download:

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

Fig 2. The proteomes of OMVs derived from N. gonorrhoeae FA1090ΔrmpM and FA1090_{MC58 PorB}ΔrmpM differ.

Proteins with an abundance ratio (Nm PorB OMVs/Ng PorB OMVs) ≥ 2 are indicated in green, and those with an abundance ratio ≤ 0.5 are indicated in red. Proteins with an abundance ratio between 0.5 and 2 are indicated in blue.

https://doi.org/10.1371/journal.ppat.1012688.g002

Several groups of proteins with related functions exhibited a greater than two-fold increase in Nm PorB OMVs compared with Ng PorB OMVs. Iron acquisition and storage proteins transferrin binding protein B (TbpB), transferrin binding protein A (TbpA) and bacterioferritin (BrfB) were in higher abundance in Nm PorB OMVs. Furthermore, Nm PorB OMVs had a higher abundance of other nutrient acquisition proteins, including zinc-acquisition proteins TonB-dependent function protein-H and -J (TdfH, TdfJ), zinc-binding protein A (ZnuA, also known as MntC), and the methionine transporter MetQ. Potentially related to the increase in nutrient acquisition proteins, surface lipoprotein assembly modulator proteins 1 and 2 (Slam1 and Slam2) also had higher abundances; Slam1 is involved in the translocation of TbpB to the outer membrane [45]. The abundance of metabolism-related proteins ethanol–active dehydrogenase (AdhP), dihydrolipoamide acetyltransferase (AceF), and carbonic anhydrase (Cah) were also increased in Nm PorB OMVs. A final group of interest were proteins related to host-pathogen interactions, including Neisseria surface protein A (NspA), IgA1 protease (NEIS1959), macrophage infectivity potentiator (Ng-MIP), and Neisseria heparin binding antigen (NHBA), which were all increased in Nm PorB OMVs compared to Ng PorB OMVs.

Two Opa-related proteins, NGO_05420 and OpaD (NEIS0903) had the largest differences when comparing Nm PorB OMVs and Ng PorB OMVs, with abundance ratios (AR, Nm/Ng) of 74 and 44, respectively. Such a large difference in AR is consistent with ON:OFF phase variation, which is known to affect Opa expression [46]. An additional Opa54-related protein, NGO_06725, also had a higher abundance in Nm PorB OMVs, with an AR of 6.2. Additionally, NGO_07725, an Opa54-related protein, was identified with slightly lower abundance in Nm PorB OMVs with an AR of 0.84. Together these data suggest that phase variation and/or mutations occurred within the Opa coding regions during the construction of the OMV-producing strains, resulting in the expression of multiple Opa54 proteins. Overall, replacing the porB gene in N. gonorrhoeae FA1090 with meningococcal porB resulted in additional differences in the OMV proteomes with changes in factors involved in nutrient acquisition, metabolism, and host-pathogen interactions.

As a significant number of proteins were more abundant in Nm PorB OMVs, we used quantitative mass spectrometry to determine whether the abundance of PorB differed between Nm and Ng PorB OMVs. PorB represented 70.2% (± 9.3%) of the Ng PorB OMV proteome, compared to 36.5% (± 3.9%) of the Nm PorB OMV proteome. Quantitatively, PorB measured 506 μg/mL (± 191 μg/mL) in Ng PorB OMVs and 235 μg/mL (± 67 μg/mL) in Nm PorB OMVs (Table 1). In a 12.5 μg vaccine dose, PorB would average 8.8 μg (± 1.2 μg) in Ng PorB OMVs and 4.6 μg (± 0.5 μg) in Nm PorB OMVs. In summary, the amount of PorB is reduced in Nm PorB OMVs compared to Ng PorB OMVs, potentially reflecting the increased abundance of other proteins.

Download:

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

Table 1. Absolute quantification of PorB in Ng- and Nm-PorB OMVs using quantitative mass spectrometry.

Data are the mean ± standard deviation.

https://doi.org/10.1371/journal.ppat.1012688.t001

OMVs from FA1090_{Nm porB}ΔrmpM generate a higher antibody titre in mice

To determine the effect of replacing gonococcal PorB with meningococcal PorB on immunogenicity, mice were immunised with Ng PorB OMVs or Nm PorB OMVs together with a heterologous antigen, fHbp. N. meningitidis fHbp does not share homology to any protein in N. gonorrhoeae OMVs and was administered at a fixed amount. Anti-fHbp IgG antibody titres were significantly higher when mice were immunised with fHbp with Nm PorB OMVs when compared with Ng PorB OMVs, with endpoint titres (log₁₀) of 6.2 and 4.7, respectively (p < 0.001, Fig 3A), demonstrating that immune responses against an exogenous protein were higher in the presence of Nm rather than Ng PorB OMVs. Antibody responses against MtrE and MetQ, candidate antigens in gonococcal vaccines [24, 47], were also examined. MtrE is a surface-exposed component of a drug efflux pump which is upregulated in AMR strains of N. gonorrhoeae, while MetQ is involved in methionine uptake [24, 47]. Anti-MtrE and anti-MetQ IgG antibody titres were also significantly higher in mice immunised with Nm PorB OMVs compared to Ng PorB OMVs, with endpoint titres (log₁₀) of 5.6 and 4.7 for MtrE, and 5.8 and 4.2 for MetQ, after immunisation with Nm PorB OMVs or Ng PorB OMVs, respectively (p < 0.001, Fig 3B and 3C).

Download:

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

Fig 3. Murine antibody responses after immunisation with Ng- or Nm-PorB expressing OMVs.

A-C) Endpoint IgG antibody titres of pooled sera recognising the model antigen fHbp, and the OMV antigens MtrE and MetQ, determined by ELISA. D-F) Endpoint IgG1 and IgG2a titres recognising the model antigen fHbp, and the OMV antigens MtrE and MetQ, determined by ELISA. Data are the mean ± standard deviation. *** p < 0.001, **** p < 0.0001. G-I) Antibody deposition on whole cell N. gonorrhoeae strains FA1090, G97687 and 60755 determined by flow cytometry showing the mean fluorescence intensity.

https://doi.org/10.1371/journal.ppat.1012688.g003

To further understand the difference in IgG responses, the ratios of IgG1 to IgG2a antibody titres were examined, as an indirect measure of Th1 vs. Th2 responses [27]. IgG subclasses differed in their titres when comparing antibody responses to recombinant fHbp, as well as the two OMV-based antigens, MtrE and MetQ. For fHbp, IgG1 antibodies comprised a higher proportion of the total IgG titre after immunisation with Nm PorB OMVs, with an endpoint titre (log₁₀) of 4.5 (Fig 3D). In contrast, for MtrE and MetQ, IgG2a antibodies represented a higher proportion of total IgG responses after immunisation with Nm PorB OMVs, with endpoint titres (log₁₀) of 3.7 and 4.9, respectively (Fig 3E and 3F). When examining the IgG1/IgG2a ratios to indicate a Th1- or Th2-skew, immunisation with Ng PorB OMVs resulted in a Th1-bias for all three antigens examined, with IgG1/IgG2a ratios of 0.4, 0.3 and 0.6 for fHbp, MtrE and MetQ, respectively (Table 2). Immunisation with Nm PorB OMVs resulted in a Th2-bias against fHbp, with an IgG1/IgG2a ratio of 4.1, but a stronger Th1-bias against the OMV antigens MtrE and MetQ, with ratios of 0.1 and 0.05, respectively. Overall, Ng PorB OMVs elicited responses against fHbp and OMV antigens with a consistent, although small, Th1-bias. However, Nm PorB OMVs elicited a marked Th1-bias against OMV antigens, but a Th2-dominated response against the recombinant antigen, fHbp.

Download:

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

Table 2. The ratio of IgG1 to IgG2a geometric mean antibody titres to fHbp, MtrE and MetQ, elicited after immunisation with Nm- or Ng-PorB OMVs.

https://doi.org/10.1371/journal.ppat.1012688.t002

As higher antibody titres were observed against individual protein antigens MtrE and MetQ after immunisation with Nm PorB OMVs compared to Ng PorB OMVs, flow cytometry was used to investigate serum IgG antibody binding to whole cell N. gonorrhoeae. Serum from mice immunised with Nm PorB OMVs exhibited higher IgG antibody deposition against wild-type N. gonorrhoeae FA1090 when compared to Ng PorB OMV immunised mice (Fig 3G), even though FA1090 expresses the same PorB as in the Ng PorB OMVs that were used for immunisation. Antibody binding to two clinical isolates was also investigated, a multi-drug resistant strain G97687 (which expresses PorB.IB) and an isolate from Kenya, 60755 (which expresses PorB.IA) [48, 49]. For strain G97687, serum from Nm PorB OMV immunised mice had higher IgG antibody deposition compared to serum from Ng PorB OMV immunised mice (Fig 3H). For strain 60755, IgG antibody binding was similar between the two OMV-immunised groups (Fig 3I).

As naturally released OMVs containing unmodified lipooligosaccharide (LOS) were used in immunisations, murine immune responses to LOS were investigated. Western blot analysis of LOS extracted from N. gonorrhoeae WT FA1090 or the LOS mutant FA1090ΔlgtE with sera from mice immunised with Ng or Nm PorB OMVs showed anti-LOS antibody responses in both immunisation groups (Fig 4A). A smaller LOS species was observed in the LOS mutant FA1090ΔlgtE compared with the wild-type strain, consistent with LOS truncation in the lgtE mutant and suggesting that OMVs can elicit antibodies specific to regions deeper within the LOS structure. Further analyses focused on responses against the LOS 2C7 epitope, which is the target epitope of candidate gonococcal vaccines [50]. Western blot using an anti-2C7 monoclonal antibody (mAb) [51] confirmed the 2C7 epitope was present in both Ng and Nm PorB OMVs, as well as WT FA1090 as demonstrated previously [52] (Fig 4B). As expected, 2C7 binding was not detected in the FA1090ΔlgtE LOS mutant. Next, an ELISA was used to determine whether Ng and Nm PorB OMV immunised mice had generated antibodies against the 2C7 epitope. ELISA plates coated with LOS from WT FA1090 were first blocked with the 2C7 mAb before probing with sera from mice immunised with Ng or Nm PorB OMVs. As both the 2C7 mAb and the OMV-immunised sera are murine, we utilised the fact that the 2C7 mAb is an IgG3 isotype and used anti-IgG1 and -IgG2a secondary antibodies to detect binding of antibodies in OMV-immunised sera. A small yet significant decrease in LOS binding was observed for both Ng and Nm PorB OMV immunised groups when the 2C7 epitope was blocked by adding the mAb (Fig 4C), indicating that immunisation with naturally released OMVs elicits 2C7-specific IgG1 and/or IgG2a antibodies.

Download:

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

Fig 4. Murine antibody responses to gonococcal lipooligosaccharide (LOS) after immunisation with Ng- or Nm-PorB expressing OMVs.

A-B) Western blots probing gonococcal LOS with murine sera after immunisation with Ng PorB OMVs (left blot) or Nm PorB OMVs (right blot) (A) or with 2C7 mAb (B). C) LOS ELISA probing with murine sera from Ng- or Nm-PorB OMV immunised mice or control groups, with or without pre-blocking with the 2C7 mAb. * p < 0.05.

https://doi.org/10.1371/journal.ppat.1012688.g004

To assess the functionality of the antibody responses, serum bactericidal assays (SBA) were performed using pooled sera from five OMV-immunised mice. N. gonorrhoeae FA1090 was incubated with serial dilutions of heat-inactivated murine sera before the addition of IgG- and IgM-depleted normal human sera (NHS) as the complement source. Human complement-dependent SBA titres (inverse of the serum dilution giving ≥50% killing of N. gonorrhoeae) were similar for Ng PorB OMV sera and Nm PorB OMV sera against FA1090, only differing by a single dilution (Table 3 and S4A Fig). SBA titres were also determined using N. gonorrhoeae clinical isolates G97687 and 60755. Both Ng PorB OMV sera and Nm PorB OMV sera exhibited SBA against G97687 and 60755, with similar titres (Table 3 and S4B–S4C Fig). Overall, immunisation with Ng or Nm PorB OMVs elicited bactericidal antibodies in BALB/c mice that were cross-reactive against clinical isolates of N. gonorrhoeae that have different antigenic profiles.

Download:

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

Table 3. Murine serum bactericidal activity^a against N. gonorrhoeae strains FA1090, G97687 and 60755 after immunisation with Ng- or Nm-PorB OMVs.

https://doi.org/10.1371/journal.ppat.1012688.t003

Immunoprofiling of murine antibody responses against gonococcal antigens

To further characterise antibody responses after immunisation with OMVs, sera from individual immunised mice were used to probe microarrays containing 91 gonococcal surface proteins. The serum IgG reactivity to each individual gonococcal antigen was determined for mice immunised with i) PBS alone, ii) fHbp alone, iii) fHbp with Ng PorB OMVs, or iv) fHbp with Nm PorB OMVs and shown as a heat map or volcano plots (Figs 5 and S5, respectively). Low background responses for the PBS and fHbp-immunised groups are apparent. For sera derived from mice immunised with OMVs, reactivities against multiple gonococcal antigens were detected. When comparing the Ng PorB OMV and Nm PorB OMV immunised groups, several antigens exhibit reactivities in both groups. Antigens with stronger reactivities indicated by higher mean fluorescence intensity (MFI) included MtrE, SliC, Lipoprotein 2 (NEIS0906), GNA2091 (NEIS2071), NEIS1462, Lipoprotein 1 (NEIS1063), PilQ and NEIS1487. Of these eight proteins, the antibody reactivity was stronger in the Nm PorB OMV immunised group for seven of the antigens. PilQ was the exception, where reactivity in Ng PorB OMV immunised mice was stronger. A relatively lower MFI was observed in both OMV immunised groups for BamE, PilE, and Ton2, where again the reactivity was higher in the Nm PorB OMV immunised group for three of the four antigens; for Slam1, reactivity levels were similar between Nm PorB OMV and Ng PorB OMV immunised mice. A subset of antigens showed reactivity in only one group; NEIS2647 (NGO554) displayed reactivity in only Ng PorB OMV immunised mice, whereas a larger number of antigens, including Ape1, IgA protease, Maf1, MetQ, NspA, Potf3, NEIS1125 and NEIS1405, displayed reactivity only in Nm PorB OMV immunised mice. The stronger antibody responses to MtrE and MetQ in Nm PorB OMV immunised mice, compared to Ng PorB OMV immunised mice, shown by the protein microarray are consistent with the ELISA data (Fig 3). Overall, microarray data analysis demonstrated that total murine IgG antibody responses were higher and more diverse after immunisation with Nm PorB OMVs compared to immunisation with Ng PorB OMVs.

Download:

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

Fig 5. Antigen-specific total IgG responses in sera from individual mice immunised with Ng PorB OMVs (third section) or Nm PorB OMVs (fourth section), compared to PBS alone (first section) and fHbp alone (second section); each column is the result for an individual mouse.

Individual antigens are in indicated in each row. Colour represents the mean fluorescence intensity (MFI) value for antibody binding to each antigen.

https://doi.org/10.1371/journal.ppat.1012688.g005

The gonococcal protein microarray revealed strong reactivities to PorB and Opa variants. The microarray included PorB from FA1090 and, as expected, mice immunised with Ng PorB OMVs generated a strong antibody response to FA1090 PorB, the variant present in these OMVs. In contrast, Nm PorB OMVs elicited very weak antibody responses to FA1090 PorB. For Opa proteins, both Nm PorB OMV and Ng PorB OMV immunised mice exhibited widespread reactivity against Opa variants. The reactivity profiles to Opa variants are remarkably similar for both OMV immunised groups, particularly given the differences in Opa expression profiles revealed by proteomic analysis of the OMVs, consistent with responses elicited against epitopes shared between Opas.

To further assess murine polyclonal IgG responses after immunisation with Ng or Nm PorB OMVs, the gonococcal protein microarrays were also used to analyse the reactivity of IgG1 and IgG2a subclasses (Fig 6). Similar prominent profiles for IgG1 and IgG2a reactivity to the Opa variants were observed for both Ng- and Nm-PorB OMVs. Interestingly, some antigens tended to elicit IgG2a over IgG1 antibody responses; for example, BamE, SliC, Lipoprotein 1 (NEIS1063) and Lipoprotein 2 (NEIS0906) have relatively stronger MFI signals for IgG2a. However, this observation was independent of the Ng/Nm PorB OMV immunisation group. Overall, when examining individual antigens, reactivity in Nm PorB OMV immunised mice was stronger for both IgG1 and IgG2a, compared to Ng PorB OMV immunised mice, suggesting that increases in both subclasses contributed to the overall higher IgG responses shown in Fig 5.

Download:

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

Fig 6. Antigen-specific IgG1 and IgG2a responses in sera from mice immunised with Ng PorB OMVs or Nm PorB OMVs.

Individual antigens are in indicated in each row; each column is the result for an individual mouse. Each colour block represents the mean fluorescence intensity (MFI) value for antibody binding to each antigen.

https://doi.org/10.1371/journal.ppat.1012688.g006

Principal Component Analysis (PCA) was applied to all four immunisation groups in order to capture the variance between each serum sample and group samples with similar reactivity profiles. As both OMV groups responded strongly to nearly all Opa proteins and only Ng PorB OMVs elicited PorB reactivity, we removed PorB and Opa variants to ascertain which other antigens were contributing to the different responses after immunisation with Ng or Nm PorB OMVs. PCA of individual serum samples showed that Ng PorB OMV and Nm PorB OMV immunised mice were well separated from the fHbp alone and PBS control groups in the PC1 dimension (Fig 7A; each point is a serum sample from a single mouse). Even with the PorB and Opa variants removed, sera from Nm PorB OMV immunised mice were separated from Ng PorB OMV serum samples, further from the PBS controls. This observation is attributable to a greater amplitude of antigen responses overall in the Nm PorB OMV serum samples. The antigens contributing most strongly to this separation were Lipoprotein 2 (NEIS0906), outer membrane protein H.8, Potf3, Lipoprotein 1 (NEIS1063), SliC, MtrE, GNA2091 (NEIS2071) and NEIS1487 (Fig 7B). In summary, PCA showed that differences in murine IgG antibody responses to Ng PorB or Nm PorB OMVs are attributable to several different gonococcal antigens.

Download:

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

Fig 7. PCA applied to antigen-specific total IgG responses.

A) Separation by individual serum sample, including all antigens. The biplot is generated using the squared coordinates (cos²) for PC1 and PC2, calculated as the squared coordinates of the eigenvalues. B) Contributions of individual antigens, excluding PorB and Opa variants, to PC1 and PC2 separation.

https://doi.org/10.1371/journal.ppat.1012688.g007

Nm PorB OMVs elicit pro-inflammatory responses in ex vivo stimulated splenocytes

As the switch from Ng PorB to Nm PorB in gonococcal OMVs enhanced humoral immune responses, we investigated cellular immune responses elicited by Ng/Nm PorB OMVs by re-stimulating splenocytes from immunised mice with the vaccine antigens. Production of the pro-inflammatory cytokine IFNɣ, the effector cytokine indicative of a Th1 response [53], was significantly higher in splenocytes from Nm PorB immunised mice when compared to the Ng PorB OMV immunised group, when re-stimulated with either Ng PorB OMVs (p < 0.05) or Nm PorB OMVs (p < 0.001) (Fig 8A). Overall, the increased IFNɣ after immunisation with Nm PorB OMVs suggests that Nm PorB OMVs prime a Th1-biased response.

Download:

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

Fig 8. Cytokines produced by murine splenocytes after immunisation with Ng- or Nm-PorB OMVs, when re-stimulated with concanavalin A (ConA, positive control), Ng PorB OMVs, Nm PorB OMVs or no stimulation.

A) interferon-ɣ (IFNɣ), B-F) interleukin (IL)-4, IL-10, IL-17A, IL-6 and IL-2. Data are the mean ± standard error of the mean. * p < 0.05, *** p < 0.001, **** p < 0.0001, ns: not significant.

https://doi.org/10.1371/journal.ppat.1012688.g008

IL-4 is an effector cytokine indicative of a Th2 response [54]. There was a trend to higher IL-4 production by splenocytes from Nm PorB OMV immunised mice compared to Ng PorB OMV immunised mice when stimulated with either Ng PorB OMVs (mean 13.3 pg/mL compared to 7.7 pg/mL) or Nm PorB OMVs (mean 14.2 pg/mL compared to 9.5 pg/mL) although the differences were not statistically significant (Fig 8B). Splenocytes from both the Ng PorB OMV and Nm PorB OMV immunised groups produced significantly more IL-4 when re-stimulated with OMVs when compared to the PBS immunised control group (p < 0.0001). Together this suggests that immunisation with Ng PorB OMVs or Nm PorB OMVs primes Th2 responses to a similar extent. The production of cytokine IL-10, involved in modulating the inflammatory response, was significantly increased only in splenocytes from mice immunised with Nm PorB OMV when re-stimulated with Nm PorB OMVs (p < 0.05, Fig 8C).

As gonococcal infection drives a Th17 response [23], we examined the production of IL-17A as an indicator of Th17 responses. Splenocytes from both the Ng PorB OMV and Nm PorB OMV immunised groups produced significantly more IL-17A when re-stimulated with OMVs when compared to the PBS immunised control group (p < 0.001, Fig 8D). However, there was no significant difference in IL-17A production between the groups of mice immunised with Ng PorB OMVs or Nm PorB OMVs (Fig 8D), suggesting that changing PorB did not impact the murine Th17 response.

IL-6 is a pleiotropic cytokine that participates in both pro- and anti-inflammatory responses. After stimulation with Ng PorB OMVs, IL-6 production was similar in murine splenocytes from naïve mice and mice immunised with Ng PorB OMVs. However, a difference was observed between these groups when stimulated with Nm PorB OMVs, with the latter group producing more IL-6 (p < 0.05). IL-6 production was also significantly higher by splenocytes from Nm PorB OMV immunised mice when compared to the Ng PorB OMV immunised group, when re-stimulated with either Ng PorB OMVs (p < 0.0001) or Nm PorB OMVs (p < 0.0001) (Fig 8E). IL-2 is produced by T cells and is also a pleiotropic cytokine that does not promote any particular Th-response, shown to positively influence the differentiation, expansion, and maintenance of both Th1- and Th2-type cells, as well as T regulatory (Treg) cells and effector T cells [55,56]. Production of IL-2 was significantly higher in murine splenocytes stimulated with OMVs after immunisation with either Ng PorB OMVs or Nm PorB OMVs when compared to naïve splenocytes (p < 0.001), suggesting an antigen-specific expansion of T cells in response to immunisation with OMVs. However, there was no significant difference in IL-2 production between the groups of mice immunised with Ng PorB OMVs or Nm PorB OMVs (Fig 8F).

To summarise, immunisation with Ng or Nm PorB OMVs elicits antigen-specific T cell expansion in splenocytes after re-stimulation with OMVs, evidenced by increased IL-2 production from splenocytes compared to control groups. Splenocytes from both Ng and Nm PorB OMV immunised mice elicited similar production of IL-4, suggesting a similar level of Th2 response. However, Nm PorB OMVs elicited a more pro-inflammatory response, producing significantly more IFNɣ, indicative of a Th1-biased response, which is beneficial for protection against gonococcal infection [24,57]. Potentially related to high IFNɣ production, both regulatory cytokines IL-6 and IL-10 were significantly higher when splenocytes from Nm PorB OMV immunised mice were re-stimulated with Nm PorB OMVs.

Ethics statement

All animal experiments were carried out under protocols reviewed by the Animal Welfare and Ethical Review Board at the Sir William Dunn School of Pathology, and approved by the Home Office, UK, under licence number PPL P20CC6E82.

Bacterial growth and transformation

N. gonorrhoeae strain FA1090 was grown on gonococcal base medium (GCB) agar plates or in liquid media containing 1% Vitox at 37°C in 5% CO₂. Mutants were generated by spotting DNA fragments onto GCB plates, streaking N. gonorrhoeae FA1090 over the spot, and incubating at 37°C and 5% CO₂ for 8 hours. Growth over the spot was then plated onto selective media (kanamycin at 80 μg/mL or erythromycin at 2 μg/mL, as appropriate) and incubated overnight at 37°C in 5% CO₂. Mutants were confirmed by PCR and Sanger sequencing.

Genetic manipulation of N. gonorrhoeae

Genomic DNA from N. meningitidis strain MC58 was used as the template to amplify N. meningitidis porB (primer pair: MC58 porB F and R, all primers are listed in S1 Table), and to add downstream complementary sequences to a kanamycin resistance cassette aph(3) [74] (kanR). DNA upstream (799 bp, primers FA1090 porB upstream F and R) and downstream (690 bp, primers FA1090 porB downstream F and R) of porB was amplified from FA1090 genomic DNA by PCR adding DNA sequences complementary to N. meningitidis MC58 porB and kanR; Q5 polymerase was used throughout (New England Biolabs),. PCR products were fused by Gibson assembly (New England Biolabs), and the final construct was re-amplified and purified by gel extraction (Promega Wizard) before being used to transform N. gonorrhoeae FA1090. Insertion of the N. meningitidis porB gene was confirmed by PCR, Sanger sequencing, and Western blot analysis, with the resultant strain named FA1090_{Nm porB}.

The ermC’ erythromycin resistance cassette [74] (referred to as eryR) was amplified by PCR, with the addition of sequences complementary to regions upstream and downstream of N. gonorrhoeae FA1090 rmpM (primer pair: eryR F and R). DNA sequences downstream (620 bp, primer pair: rmpM downstream F and R) and upstream (578 bp, primer pair: rmpM upstream F and R) of rmpM were amplified from FA1090 genomic DNA. The three PCR products were fused by Gibson assembly (New England Biolabs) and used to transform N. gonorrhoeae FA1090 and FA1090_{Nm porB}. Deletion of rmpM was confirmed by PCR, Sanger sequencing and Western blot, and the resultant strains named FA1090ΔrmpM and FA1090_{Nm porB}ΔrmpM. The genomic sequences upstream and downstream of N. gonorrhoeae FA1090 lgtE were amplified by PCR (primer pairs: lgtE upstream F and lgtE upstream R; lgtE downstream F and lgtE downstream R) with the addition of complementary sequences to the kanR resistance cassette. PCR products were fused by Gibson assembly and used to transform N. gonorrhoeae FA1090 as before, and the mutant strain confirmed by Sanger sequencing.

Production of OMVs

Bacteria were grown overnight on GCB agar plates then resuspended in 50 mL liquid GCB at an optical density (OD₆₀₀) of 0.1 for FA1090ΔrmpM and 0.2 for FA1090_{Nm porB} ΔrmpM, and incubated at 37°C in 5% CO₂ with shaking at 180 rpm. Upon reaching an OD₆₀₀ of 1, bacteria were pelleted by centrifugation at 2,739 x g for 20 minutes at 4°C. Supernatants were passed through 0.22 μm pore size filters, and OMVs obtained via ultracentrifugation of the supernatant at 235,000 x g for 2 hours at 4°C. OMVs were washed with PBS before ultracentrifugation as before, then stored at 4°C. The protein content of OMV preparations was determined by BCA assay (Pierce).

Protein purification

mtrE and metQ were amplified from N. gonorrhoeae FA1090 and ligated into pET14b, generating pET14b:mtrE and pET14b:metQ, respectively. For protein expression, E. coli B834 containing pET14b:mtrE or pET14b:metQ was grown in TB medium for 24 h at 22°C and harvested by centrifugation at 5,000 × g for 30 minutes at 4°C, prior to cell lysis using an EmulsiFlex-C5 homogeniser (Avestin, 15,000 lb/in²). Lysates were centrifuged at 20,000 × g for 30 minutes at 4°C, and recombinant protein was bound to HisTrap columns (GE Healthcare), eluted with 300 mM imidazole, and further purified by size exclusion (AKTA, HiLoad 16/600 Superdex 200 pg column; GE Healthcare). Protein concentrations were estimated using a Nanodrop 2000c spectrophotometer (Thermo Scientific). Factor H binding protein (fHbp) v1.1 was expressed in E. coli B834 during growth at 22°C for 24 hours with 1 mM IPTG (final concentration). Bacteria were harvested and resuspended in Buffer A (50 mM Na-phosphate pH 8.0, 300 mM NaCl, 30 mM imidazole) and fHbp purified by Nickel affinity chromatography (Chelating Sepharose Fast Flow; GE Healthcare). Columns were washed with Buffer A, then with 80:20 Buffer A:Buffer B (50 mM Na-phosphate pH 8.0, 300 mM NaCl, 300 mM imidazole) and proteins eluted in 40:60 Buffer A:Buffer B. Proteins were dialysed overnight at 4°C into PBS, 1 mM DTT pH 8.0 with TEV protease prior to Nickel affinity chromatography to remove the HIS-GST-TEV. fHbp was eluted from Sepharose columns with Buffer B after washing with buffer C (50 mM Na-phosphate pH 6.0, 500 mM NaCl, 30 mM imidazole), and dialysed overnight at 4°C into Tris pH 8.0. Proteins were filtered through a 0.22 μm pore size filter before use in immunisations.

Immunisation of mice with OMVs

Six-week-old female BALB/c mice (five per group) were immunised by the intraperitoneal route with: i) PBS alone, ii) 10 μg fHbp alone, iii) 10 μg fHbp with 12.5 μg FA1090ΔrmpM OMVs, or iv) 10 μg fHbp with 12.5 μg FA1090_{Nm porB} ΔrmpM OMVs. All components were diluted in PBS (200 μL total volume per dose). Mice received three immunisations (on days 1, 21 and 35) before sera was obtained for analysis on day 49.

Antibody titre determination by ELISA

Microplates (Nunc Immuno Maxisorp, Thermo Scientific) were coated with the target recombinant protein (fHbp v1.1, MtrE or MetQ) by adding 50 μL of 2.5 μg/mL protein diluted in PBS and incubating overnight at 4°C. PBS was added to control wells. Plates were washed three times with PBS + 0.5% Tween20 (PBS-T) before blocking with 4% bovine serum albumin (BSA) in PBS-T for 1 hour at 37°C. Plates were washed as before, before addition of pooled sera from immunisation groups in threefold serial dilutions, and incubated for 1 hour at 37°C. Plates were washed as before, and anti-mouse IgG-HRP (Bio-Rad, 172–1011) conjugated secondary antibody was added at a 1:10,000 dilution and incubated for 1 hour at 37°C. Alternatively, anti-mouse IgG1-HRP (Abcam, ab97240) or IgG2a-HRP (Abcam, ab97245) were added at a 1:20,000 dilution, or anti-mouse IgG1-AF488 (Invitrogen, A21121) or IgG2a-AF488 (Invitrogen, A21131) were added at a 1:1000 dilution. After further washes, plates were developed by adding TMB ELISA substrate followed by stop solution (BioTechne), and the absorbance read at 450 nm in a FLUOStar Omega plate reader (BMG LabTech). For AF488 conjugated secondary antibodies, fluorescence was measured using 485/520 nm for excitation/emission, respectively. To calculate endpoint titres, the baseline value was set at three times the average of the PBS controls. The endpoint titre is reported as the log₁₀ of the reciprocal of the serum dilution that was equal to the baseline value. Statistical significance was tested by two-way ANOVA with multiple comparisons in GraphPad Prism (GraphPad Software Inc. v.10.0).

Flow cytometry

N. gonorrhoeae strains FA1090, G97687 and 60755 were grown overnight on GCB agar plates then resuspended in 5 mL liquid GCB at an OD₆₀₀ of 0.1 for FA1090 and 0.2 for G97687 and 60755, and incubated at 37°C in 5% CO₂ with shaking at 180 rpm. Upon reaching an OD₆₀₀ of 0.8, 1 mL of the bacteria suspension was pelleted by centrifugation at 16,000 x g for 5 minutes, washed once with supplemented PBS, and 100 μL bacteria suspension was incubated in a twofold serial dilution of murine sera in supplemented PBS with 1% BSA (w/v) (1:500 to 1:16,000) for 30 minutes at RT with shaking at 1250 rpm. Bacteria were washed twice as before, followed by incubation with goat anti-mouse IgG-Alexa fluor 647 secondary antibody (Invitrogen) at a 1:1000 dilution for 30 minutes at RT with shaking at 1250 rpm. Bacteria were washed twice as before and fixed using 3% paraformaldehyde before analysing on flow cytometer BD LSRFortessa X-20. FlowJo v10.10 software was used to calculate the mean fluorescence intensity for each spectra.

Lipooligosaccharide analyses

N. gonorrhoeae WT FA1090 and FA1090ΔlgtE were grown overnight on GCB agar plates then resuspended in PBS at an OD₆₀₀ of 1 and 750 μL of bacterial suspension was pelleted by centrifugation at 16,000 x g for five minutes before resuspending in 200 μL PBS. Bacterial suspensions or 15 μg of OMV preparations were digested with proteinase K (final concentration 1 mg/mL) at 59°C for a minimum of six hours. LOS resulting from proteinase K digested OMVs or bacterial suspension was separated on a 16% tricine gel, as previously described [75], and transferred onto PVDF membrane (0.22 μm, Immobilon-PQS) at 15 V for 90 minutes using a semi-dry blot (Bio-Rad). Membranes were blocked overnight at 4°C in 5% skim milk in PBS-T before probing with either an anti-2C7 monoclonal antibody (antibody registry ID: AB_2889813, Developmental Studies Hybridoma Bank, University of Iowa) [51] at a final concentration of 0.25 μg/mL or pooled sera from OMV-immunised mice at a 1:2000 dilution. After three washes in PBS-T, goat anti-mouse IgG-HRP (Bio-Rad) was added at a 1:10,000 dilution. After final washes, membranes were developed using Amersham ECL Western Blotting Analysis System (GE Healthcare) and exposed to Amersham Hyperfilm ECL (GE Healthcare). For ELISA analysis, microplates (Nunc Immuno Maxisorp, Thermo Scientific) were coated with LOS at a protein equivalent of 2 μg of proteinase K digested WT FA1090 overnight at 4°C. After blocking with 4% BSA in PBS-T, 100 μL anti-2C7 mAb at 0.5 μg/mL was added to appropriate wells prior to incubation with serial dilutions of OMV-immunised murine sera. Anti-mouse IgG1-HRP (Abcam, ab97240) and IgG2a-HRP (Abcam, ab97245) were added together at a 1:10,000 dilution, and plates were developed by adding TMB ELISA substrate followed by stop solution (BioTechne), and the absorbance read at 450 nm in a FLUOStar Omega plate reader (BMG LabTech). Wells with LOS and 2C7 mAb alone and wells with secondary antibody only were used as background controls.

SDS-PAGE gel electrophoresis and Western blotting

N. gonorrhoeae cell lysates (10 μL) or OMVs (5 μg) were separated on 12% SDS-polyacrylamide gels. SDS-PAGE gels were either Coomassie stained, or proteins were transferred to 0.45 μm nitrocellulose membranes, using a semi-dry blot at 25 V for 30 minutes (Bio-Rad). Membranes were blocked overnight in 5% milk in PBS-T at 4°C, before the addition of primary antibody and incubation at room temperature for 2 hours. Primary antibodies were diluted in 5% milk in PBS-T at 1:1000 for anti-PorB antibodies (P3.15 for meningococcal PorB and H5.2 for gonococcal PorB, NIBSC, UK) and 1:10,000 for the anti-RmpM antibody [76]. After three washes in PBS-T, membranes were incubated with polyclonal goat anti-mouse immunoglobulins secondary antibody conjugated to HRP (Dako, P0447), diluted 1:10,000. After final washes, membranes were developed using Amersham ECL Western Blotting Analysis System (GE Healthcare) and exposed to Amersham Hyperfilm ECL (GE Healthcare).

Splenocyte re-stimulation and cytokine quantification

Two weeks after the final immunisation, mouse spleens (five per group) were harvested and splenocytes were pressed through a 70 μm cell strainer, and suspended in RPMI-1640 supplemented with 10% Foetal Bovine Serum (FBS) and 50 μM 2-mercaptoethanol (complete RPMI). The cell suspension was pelleted by centrifugation at 300 x g for 10 minutes, and the pellet was resuspended in 3 mL ACK lysis buffer for 5 minutes to remove red blood cells. Cells were washed with 10 mL complete RPMI, centrifuged as before, resuspended in PBS and counted using a haemocytometer. Splenocytes were diluted to 2 x 10⁶ cells/mL in complete RPMI and 1 mL was plated per well in a 24-well plate. Splenocytes were then stimulated with 2 μL per mL concanavalin A (500X stock, eBioscience), 5 μg Ng PorB OMVs, 5 μg Nm PorB OMVs, or left unstimulated, and incubated for 72 hours at 37°C with 5% CO₂. Cell culture supernatants were stored at -80°C until analysis. The release of cytokines interferon-gamma, IL-2, IL-4, IL-6, IL-10 and IL-17A were quantified by ELISA (Abcam) according to the manufacturer’s instructions and diluting the supernatants as appropriate. Statistical significance was tested by two-way ANOVA with multiple comparisons in GraphPad Prism (GraphPad Software Inc. v.10.0).

Serum bactericidal assay

N. gonorrhoeae was grown overnight on GCB agar and colonies were resuspended in PBS to an OD₆₀₀ of 0.07 (~ 6–7 x 10³ colony forming units (CFU)/mL) and incubated with serial dilutions of pooled from five mice heat-inactivated (56°C for 1 hour) murine sera for 10 minutes at 37°C with 5% CO₂. IgG-depleted human sera (Pel-freeze) was then added at a final concentration of 3% (v/v) for strain FA1090 and 10% (v/v) for clinical isolates G97687 and 60755 [48, 49]. and incubated at 37°C with 5% CO₂ for 45 minutes. The dilutions were plated on GCB agar in triplicate as 10 μL spots and incubated for 20–28 hours at 37°C with 5% CO₂. Survival was calculated as a percentage of the colonies present in the IgG-depleted human sera only control.

Mass spectrometry analysis of OMVs

OMVs were subjected to trypsin digestion before separation and mass spectrometric (MS) analysis of the tryptic peptides was performed on an Vanquish Neo UHPLC coupled to Orbitrap Eclipse Tribrid Mass Spectrometer (Thermo Fisher Scientific). For LC-MS/MS analysis, 4 μL of tryptic sample was injected. Samples were further purified and concentrated on a trap column (C18 PepMap, 300 μm ID x 5 mm, 5 μm, 100 Å; Thermo Fisher Scientific), then separated on a 15 cm analytical nano-LC column (PepMap C18, 75 μm ID x 150 mm, 2 μm, 100 Å; Thermo Fisher Scientific) using a binary 60-minute linear gradient from 5% to 45% buffer B (80% acetonitrile (ACN), 0.1% formic acid (FA)) for 60 minutes, then to 99% buffer B over 1 minute and maintained at 99% for a further 10 minutes, at a flow rate of 300 nL/min. Solvent A is a buffer containing 2% ACN and 0.1% FA in water.

Orbitrap Eclipse Tribrid Mass Spectrometer equipped with a Thermo Easy-Spray capillary Emitter was used to analyse the separated peptides. Protein identification and label-free quantification (LFQ) were achieved using Proteome Discoverer (PD; version 2.5, Thermo Fisher Scientific). For protein identification, raw files were searched against a database containing canonical protein sequences of the N. meningitidis MC58 PorB and N. gonorrhoeae FA1090 (https://www.uniprot.org) with the Sequest HT search algorithm and using modified standard processing and consensus workflows. The false discovery rate (FDR) tolerance in the Percolator node was set to 0.01 for high confidence and 0.05 for medium confidence.

Absolute quantification of PorB in OMVs was carried out using the same method developed previously for quantifying porins in Bexsero [77]. Briefly, PorB peptide standards, a native and a heavy (isotopically labelled) analogue, were synthesised by Thermo Scientific (HeavyPeptide AQUA Ultimate service, 97% purity; 99% isotope enrichment). The heavy peptides were ¹³C¹⁵N labelled; for SDYLGVNK, the valine residue was labelled, for GQEDLGNGLK, the penultimate leucine was labelled. Labelled peptide was added to a final concentration of 70 fmol/μL to the tryptic digests of OMVs before separation by LC/MS. Peak area ratios of tryptic-released native peptide/heavy peptide were compared with a PorB peptide standard curve [78] for each peptide to provide a surrogate estimate of the molar content of PorB in the digest.

Gonococcal protein microarray

The gonococcal microarray slides were custom-made by Arrayjet Limited, UK, as described previously [79]. Each microarray slide consisted of 16 identical blocks, with each mini-array containing 91 individually purified gonococcal proteins (derived from N. gonorrhoeae strain FA1090) and 17 control samples, printed in five repeats. Three sets of control samples were used for the slides, consisting of monoclonal mouse IgG, human IgG and EBNA-1 viral protein. Mouse and human IgGs were prepared at eight concentrations, with the highest concentration of 0.5 mg/mL and serially diluted with JetStar printing buffer (Arrayjet, UK) down to 0.0035 mg/mL. EBNA-1 viral protein at 0.05 mg/mL concentration was also included as a negative control. The slides were stored at 4°C until use.

Microarray immunogenicity probing and data acquisition

The slides were blocked with 3 mL of SuperG Blocking Buffer (Grace Biolabs, US) per slide and incubated for one hour at room temperature (RT). Sera were diluted 1:300 with TBS, and 50 μL was added to each mini-array and incubated at 20°C for one hour. After washing three times with 300 μL TBS-T (0.05% Tween-20) and once in TBS for 10 minutes, slides were incubated in the dark for 1 hour at 20°C with 50 μL of goat anti-mouse IgG Fc (DyLight 650) (ab97018, Abcam, UK) diluted 1:5000 in blocking agent. The same protocol was followed for determining IgG1 and IgG2a, using a goat anti-mouse IgG1 cross-adsorbed secondary antibody (Alexa Fluor 647) (A-21240, Invitrogen) and a goat anti-mouse IgG2a cross-adsorbed secondary antibody (Alexa Fluor 488) (A-21131, Invitrogen). Slides were rinsed in de-ionised water and dried by centrifugation at 200 x g for 2 minutes. Slides were scanned in an InnoScan 710 (Innopsys, France) with the photomultiplier tube set to 40% for 635 nm. Image analysis and data quantification were carried out using Mapix—Microarray image acquisition and analysis software (v9.1.0, Innopsys, France). Microarray spot intensities were quantified using automatic background subtraction for each spot. The spot intensities for each protein were recorded in quintuplicate; arithmetic means were determined, and spot intensities for buffer-only controls were subtracted.