Hybrid response to SARS-CoV-2 and Neisseria meningitidis C after an OMV-adjuvanted immunization in mice and their offspring

ABSTRACT

COVID-19, caused by SARS-CoV-2, and meningococcal disease, caused by Neisseria meningitidis, are relevant infectious diseases, preventable through vaccination. Outer membrane vesicles (OMVs), released from Gram-negative bacteria, such as N. meningitidis, present adjuvant characteristics and may confer protection against meningococcal disease. Here, we evaluated in mice the humoral and cellular immune response to different doses of receptor binding domain (RBD) of SARS-CoV-2 adjuvanted by N. meningitidis C:2a:P1.5 OMVs and aluminum hydroxide, as a combined preparation for these pathogens. The immunization induced IgG antibodies of high avidity for RBD and OMVs, besides IgG that recognized the Omicron BA.2 variant of SARS-CoV-2 with intermediary avidity. Cellular immunity showed IFN-γ and IL-4 secretion in response to RBD and OMV stimuli, demonstrating immunologic memory and a mixed Th1/Th2 response. Offspring presented transferred IgG of similar levels and avidity as their mothers. Humoral immunity did not point to the superiority of any RBD dose, but the group immunized with a lower antigenic dose (0.5 μg) had the better cellular response. Overall, OMVs enhanced RBD immunogenicity and conferred an immune response directed to N. meningitidis too.

KEYWORDS:

• SARS-CoV-2
• receptor-binding domain (RBD)
• outer membrane vesicles (OMVs)
• Neisseria meningitidis
• adjuvants
• cross-reaction

Introduction

The Coronavirus Disease 2019 (COVID-19), caused by SARS-CoV-2, was a health emergency that lasted for 3 years and was controlled with vaccines of different technologies: inactivated, viral-vectors, mRNA, and protein-subunit.¹ However, SARS-CoV-2 is still circulating and COVID-19 remains a public health issue, meaning that prevention strategies are still relevant, especially focusing on a long-lasting immune response and cellular activation; cross-reaction with variants; and protection of certain groups, such as newborns.^1–4 The gap in newborn vaccines could be explained by the lower COVID-19 rates in children, but they may suffer a significant burden of this disease.^5,6

Despite the efficacy of current SARS-CoV-2 vaccines, the search for new preparations, combining SARS-CoV-2 antigens with different adjuvants, would probably address these issues.^2,3 Also, technically and economically feasible adjuvants would support vaccine production in the developing world and one good example of adjuvant that falls within this scope is outer membrane vesicles (OMVs). They are released from Gram-negative bacteria; they contain several pathogen-associated molecular patterns (PAMPs), which activate the immune response and were employed as vaccines for meningococcal disease in the past^7,8 and as SARS-CoV-2 adjuvants recently,^9–11 proving to be immunogenic and safe for use. OMVs from Neisseria meningitidis strains are particularly interesting: besides the adjuvancity potential, the presence of specific antigens can elicit protection against meningococcal disease.⁷

Here, immunized isogenic mice using antigenic preparations with different RBD doses and adjuvanted by OMVs from N. meningitidis C:2a:P1.5 strain and Aluminum hydroxide (AH); aiming to investigate the potential of such OMVs as adjuvant and if mice would present an immune response to both SARS-CoV-2 and N. meningitidis. To do so, we assessed the humoral (IgG and avidity) and cellular (IL-4 and IFN-Υ) response to the antigens RBD and OMVs. We also studied the passive-transference of antibodies to the offspring.

Material and methods

Antigens and adjuvants

The plasmids to produce recombinant receptor-binding domain (RBD) of SARS-CoV-2 were donated by Dr Florian Krammer (Icahn School of Medicine, Mount Sinai, New York, NY, USA) (RBD-Wuhan) and by Dr Silvia Boscardin (Parasitology Department, Institute of Biomedical Sciences, University of São Paulo, São Paulo, SP, Brazil) (RBD-Omicron BA.2). The proteins were produced in a mammalian expression system of HEK 293 F cells, using the ExpiFectamine 293 Transfection Kit (Thermo Fischer Scientific), and purified by immobilized metal ion affinity chromatography (IMAC) with HisTrap Excel columns (GE HealthCare Life Sciences) on a fast protein liquid chromatography system (ÄKTA Pure, Cytiva Life Sciences). The procedures were described in detail before.¹²

OMVs from a N. meningitidis C:2a:P1.5 strain were obtained according to De Gaspari & Zollinger.¹³ The strain was incubated with a 0.1 M sodium acetate-0.2 M lithium chloride solution (pH 5.8) and glass beads, taken to a shaker at 300 rpm, for 2 h at 45°C; and OMVs were isolated by centrifugation at 12,000 rpm for 15 min at 4°C (Sorvall Instruments). The vesicles were detoxified using a Polymyxin B Sepharose-4B column (Pierce) and tested for LPS contamination with the Limulus Amebocyte Lysate (LAL) Gel-Clot Endosafe Kit (Charles River Laboratories).¹⁴ The electrophoretic characterization of RBD and OMVs, using SDS-PAGE in denaturing conditions, was published before.^11,14

RBD and OMVs were adsorbed in aluminum hydroxide (AH) (Rehydragel). Final concentrations of each dose were as follows: group 1A–0.5 µg RBD +0.5 µg OMVs +0.1 mM AH; group 1B–1.0 µg RBD +0.5 µg OMVs +0.1 mM AH; group 1C–1.5 µg RBD +0.5 µg OMVs +0.1 mM AH; group 2 – naïve control; group 3–1.5 µg RBD; group 4–0.5 µg OMVs +0.1 mM AH.

Mice, immunization, and sampling

Isogenic BALB/c mice (H2^d) received three subcutaneous (SC) doses, administered at days 7, when they were considered neonates, and on days 27 and 42, when mice were young (Figure 1).¹⁵ Bleds were performed in pools of two individuals each, resulting in different samples that consider the whole group: group 1A – n = 14, 7 samples; group 1B – n = 16, 8 samples; group 1C – n = 14, 7 samples; group 2 – n = 10, 5 samples; group 3 – n = 8, 4 samples; group 4 – n = 8, 4 samples. Thirty days after the last dose, two mice of each group were anesthetized (xilazin/ketamine), euthanized, and spleens were collected for ELISpot. Two couples (one male for two females) of groups 1A, 1B, 1C, and group 2 were put to mating. The offspring was bled 21 days after birth.

Figure 1. Experimental calendar with BALB/c mice. The line below explains mice’s age. SC: subcutaneous. Figure created using BioRender.

All animal experimentation was performed in accordance with Brazilian National Council of Animal Experimentation Control (CONCEA) legislation and approved by the Ethical Committee of Animal Use from Adolfo Lutz Institute (CEUA/IAL number 01/2023).

Indirect-ELISA

IgG levels and its isotypes were assessed by in house ELISA. High-binding plates (Costar) were coated with 1 µg/ml of RBD-Wuhan, or 1 µg/ml RBD-Omicron,¹¹ or 2 µg/ml of OMVs¹⁴ diluted in carbonate-bicarbonate buffer (pH 9.6) overnight, at 4°C and blocked with skimmed-milk 5% (La Sereníssima) for 2 h at 37°C. Pooled samples of each group were serially diluted in the plates and incubated overnight at 4°C to obtain the IgG titration curve (suppl. Figures S1–S4). Afterward, the samples were diluted at 1/100 for RBD-Wuhan and OMVs immune response or at 1/25 for RBD-Omicron, and incubated overnight at 4°C. Secondary-antibodies were added at the following concentrations: HRP-anti-mouse IgG-γ at 1/20,000 (Kierkegaard & Perry Laboratories, product 074–1802), biotin-anti-mouse IgG1 (product OASB01568) at 1/5,000, HRP-anti-mouse IgG2a (product OAIA00396), IgG2b (product OAIA00399), or IgG3 (product OAIA00404) at 1/10,000 (Aviva Systems Biology), incubated for 2 h at 37°C. For anti-IgG1, it was added HRP-Streptavidin at 1/2,000 (Zymed, product 43–4323) for 1 h at 37°C. The enzymatic reaction was performed incubating 3,3’5,5’ tetramethylbenzidine (TMB) (Sigma-Aldrich) at 37°C for 20 min, stopped using H₂SO₄ 1 N and the Optical density (OD) was read at 450 nm (Molecular Devices). The results were plotted as ELISA Index (EI), calculated as the ratio between the OD of the sample/cutoff.¹² Cutoffs were considered the mean OD of naive sera + 3 standard deviations (SD).

IgG avidity

The ELISA-avidity was carried out as described above, but adding potassium thiocyanate (KSCN) (Sigma-Aldrich) 1.5 M for 20 min, at room temperature (RT) (20–25ºC) after the sera incubation.¹⁶ Avidity index (AI) is the ratio between the OD with the KSCN/OD without the KSCN and classified as low (<30%), intermediary (30–49%) or high (≥50%).¹⁷

ELISA-avidity provides reliable results only when immune sera is tested,¹⁸ so it would not be adequate to evaluate AI of RBD group for OMV; of OMV+AH group for RBD-Wuhan and Omicron; of either RBD or OMV+AH groups for RBD-Omicron; and of naive sera for any antigen.

IL-4 and IFN-γ ELISpot

The assays were performed using Mouse-IL-4 or Mouse-IFN-γ ELISpot Kit ^PLUS (MabTech, products 3311-3PW and 3321-3PW, respectively), according to the manufacturer’s instructions, using pooled splenocytes from two mice of each group. After euthanasia, the spleens were harvested and 1 × 10⁷cells were resuspended in Iscove’s medium supplemented with 10% fetal calf serum (Cripion Biotecnologia Ltda.), antibiotics and antimycotics (Sigma-Aldrich). Cells were added in triplicate, stimulated with 1 µg/well of RBD-Wuhan or 2 µg/well of OMVs for specific response and incubated for 18 h at 37°C, 5% CO₂. Control wells were either incubated without any stimuli or with 2.5 µg/well of the mitogen Concanavalin A (Sigma-Aldrich). After incubation, the cells were removed, and the plates were incubated with 1 mM EDTA (Sigma-Aldrich) solution for 15 min at RT to remove debris. The detection steps were performed according to the manufacturer. After drying, the spots were counted using the AID ELISpot Reader v. 7.0 (Autoimmun Diagnostika).

Statistical analysis

The differences between the groups were assessed using One-way ANOVA and Tukey’s post test. For ELISpot data, the non-parametric Kruskal-Wallis and Dunn’s post test was applied. p values ≤ .05 were considered significant. Analyses were performed on GraphPad Prism v. 8 (GraphPad Softwares).

Results

Humoral response to RBD suggests enhanced immunogenicity, Th2 isotypes, high avidity, and cross-reaction with Omicron variant

OMV+AH as adjuvant contributed to the immune response regardless the RBD concentration; RBD alone was not as immunogenic as the adjuvanted-protein and naïve mice did not present specific antibodies. Considering the RBD-Wuhan response, the first immunization did not induce detectable levels of IgG and, when using 1.0 or 1.5 μg/dose, some individuals presented antibodies after the second dose. After three doses, IgG response was comparable between all RBD dosages (Figure 2). RBD is the main mutation site of SARS-CoV-2,¹ so we assessed the cross-reaction with Omicron (BA.2) RBD. After three doses, groups immunized with adjuvants presented cross-reactive antibodies.

Figure 2. IgG response to RBD-Wuhan after (a) one, (b) two, or (c) three doses, in sera collected 15 d after each dose and at 1/100 dilution. (d) After three doses, IgG recognized the RBD of omicron BA.2 variant in a 1/25 dilution. The dotted line shows the cutoff value.

The binding strength of antibodies to the RBD-Wuhan and Omicron shows that RBD alone induced antibodies of low avidity to the Wuhan variant and, as this group presented antibody titer below the cutoff value for the Omicron strain, we could not assess its avidity. The mean avidity of the 0.5 μg group was intermediary, probably due variability between mice (suppl. Figure S5), and groups immunized using 1.0 or 1.5 μg of RBD per dose showed high avidity, statistically superior to RBD alone (p < .001 and p < .01, respectively) (Table 1). When the adjuvanted groups were compared, the only difference was between the groups that received 0.5 and 1.0 μg of RBD (p < .01). Meanwhile, the avidity to the Omicron variant was intermediary and showed no difference between the groups.

Table 1. Avidity index (mean ± SD) (%) of IgG to RBD-Wuhan and Omicron variants after 3 immunization doses.

	RBD 0.5+OMV+AH	RBD 1.0+OMV+AH	RBD 1.5+OMV+AH	RBD
RBD-Wuhan	46.25 (± 26.93)	85.42 (± 11.60)	71.05 (± 15.38)	24.85 (± 10.01)
RBD-Omicron	44.22 (± 24.85)	46.93 (± 23.14)	44.11 (± 16.82)	–

IgG isotypes vary according to their Fc function¹⁹ and reflect the cytokine microenvironment.²⁰ The immune response to RBD (Figure 3) was mediated by IgG1 antibodies, suggesting a Th2 response, although some individuals showed levels of IgG2a, IgG2b, and IgG3 above the cutoff.

Figure 3. Comparison of the (a) IgG1, (b) IgG2a, (c) IgG2b, and (d) IgG3 recognizing RBD-Wuhan in each group, using sera diluted at 1/100. The dotted line shows the cutoff value.

OMVs as adjuvant induced specific IgG of high avidity and mixed isytopes

The immunization induced anti-OMVs antibodies (Figure 4). Interestingly, the group immunized with 1.5 μg RBD showed IgG slightly above the cutoff after the first immunization; two immunization doses increased IgG levels but with some variability among individuals and, after three doses, the response was homogeneous, and no difference was assessed between the groups that received different RBD doses. Moreover, the group immunized with OMV+AH presented similar IgG levels, suggesting that mixing the vesicles with other antigens does not impair the specific OMV-response.

Figure 4. IgG response to OMVs from N. meningitidis C:2a:P1.5 strain after (a) one, (b) two, or (c) three doses, in sera diluted at 1/100. The dotted line shows the cutoff value.

Regarding the functionality of anti-OMVs antibodies, all groups presented high avidity and no statistical difference was observed (Table 2). Like IgG levels, the use of OMV+AH in combination with RBD resulted in a similar response to OMV+AH, showing no compromise to OMV’s own immunogenicity. Individual AI is on supplementary material (suppl. Figure S5).

Table 2. Avidity index (mean ± SD) (%) of IgG to N. meningitidis C:2a:P1.5 OMVs after 3 immunization doses.

	RBD 0.5+OMV+AH	RBD 1.0+OMV+AH	RBD 1.5+OMV+AH	OMV+AH
OMVs	81.98 (± 5.8)	80.62 (± 13.12)	74.95 (± 13.15)	69.74 (± 21.2)

OMVs induced more variable IgG isotypes than RBD, suggesting a mixed response (Figure 5). IgG1 was homogenous, but IgG2a and IgG2b were detected above the cutoff too. The same pattern was observed for mice immunized with OMV+AH alone.

Figure 5. Comparison of the (a) IgG1, (b) IgG2a, (c) IgG2b, and (d) IgG3 recognizing C:2a:P1.5 OMVs in each group, using sera diluted at 1/100. The dotted line shows the cutoff value.

RBD and OMVs stimulated cytokine secretion by splenocytes

Both RBD and OMVs stimuli induced cytokine secretion, showing the presence of memory-cells for both antigens. IL-4 had more spots, corroborating data of IgG1 isotype. However, IFN-γ was present too, agreeing with the mixed-OMV pattern and suggesting a Th1/Th2 profile for SARS-CoV-2. Naïve splenocytes secreted less cytokines and RBD alone presented some cellular response, but lower than the adjuvanted groups. Considering the three immunization doses studied, the group immunized with 0.5 μg of RBD had the best cellular response (Figure 6).

Figure 6. The panel above shows the cellular response evaluated by ELISpot, showing the IFN-γ secretion after (a) RBD or (b) OMVs stimuli, and IL-4 secretion after (c) RBD or (d) OMVs stimuli. The quantifications are relative to pooled splenocytes (n = 2) of each group, assayed in triplicates.

The offspring presented IgG of similar levels and avidity of their mothers

The passive transference of IgG was assessed in adjuvanted groups (RBD 0.5, 1.0 or 1.5 μg+OMV+AH), which showed a robust humoral response, and in naïve mice as control. Sera of the offspring were titrated in parallel with their mothers, all collected 21 days after birth, and, even though the offspring showed slightly lower levels, the curves were quite similar (suppl. Figures S2–S4). IgG specific to RBD-Wuhan and OMVs was transferred in all groups; IgG also recognized RBD-Omicron (Figure 7).

Figure 7. The offspring of immunized mothers had IgG that recognized (a) RBD-Wuhan, (b) RBD-Omicron, and (c) N. meningitidis C:2a:P1.5 OMVs. The dotted line shows the cutoff value.

Overall, AI of all groups agreed with their mothers, being high for RBD-Wuhan and OMVs and intermediary for RBD-Omicron (Table 3). Individual AI is at supplementary material, OMV-avidity was higher only for group RBD 1.0 compared with RBD 0.5 (p < .01) and RBD 1.5 (p < .05) (suppl. Figure S6).

Table 3. Avidity index (mean ± SD) of IgG transferred to the offspring, for each antigen studied.

	RBD 0.5+OMV+AH	RBD 1.0+OMV+AH	RBD 1.5+OMV+AH
RBD-Wuhan	93.68 (± 4.93)	96.6 (± 6.71)	90.30 (± 5.02)
RBD-Omicron	48.52 (± 3.35)	47.05 (± 2.28)	46.55 (± 3.88)
OMVs	79.56 (± 4.11)	95.64 (± 7.33)	84.40 (± 1.10)

Discussion

Our group observed that N. meningitidis C:2a:P1.5 OMVs were promising adjuvants for RBD protein before¹⁰ but, in this study, we used a different immunization approach, a higher experimental number and verified if this strategy would elicit antibodies that recognize the Omicron variant. As our main goal is to study OMVs’ contribution for future antigenic preparations, we characterized the immune response not only to RBD but also to OMVs, checking if the vesicles could offer a combined response against the virus and N. meningitidis.

It is well known that adjuvants support the immune response to subunit vaccines, which are safer but less immunogenic than inactivated and attenuated ones.^21–23 OMVs are useful as modulating-adjuvants, because they present several components that interact with the immune system.²⁴ Here, we observed increased IgG levels in mice that received the RBD mixed with OMV+AH (Figure 2), agreeing with previous manuscripts that used OMVs from different Gram-negative bacteria as adjuvants for SARS-CoV-2^25–27 and other viruses.^28,29 The mice immunized with 1.0 or 1.5 μg of RBD presented detectable IgG after two doses; however, there was no difference between dosages when the third dose was administered. In other dose–response studies, the double of the RBD protein (20 μg vs 10 μg) formulated in liposomes enhanced IgG production³⁰; whereas in Yang et al.³¹ RBD plus Alum showed a dose-dependent gain of antibodies only if 0.1 or 1 μg were compared to 10 μg, which are much different dosages.

Affinity is a relevant functional parameter for COVID-19 because high-avidity antibodies are required to impair the binding of RBD to ACE-2 receptor, neutralizing SARS-CoV-2.³² Here, the immunization with RBD alone resulted in low avidity-antibodies, whereas RBD plus adjuvants had better avidity to RBD-Wuhan (Table 1). It is worth mentioning that the avidity of the group that received 0.5 μg of RBD per dose was intermediary, while the 1.0 and 1.5 μg were high, probably due to mice variability, since the individual analysis showed that a few mice had low avidity (suppl. Figure S5). In humans, a third dose of vaccine resulted in enhanced avidity against wild-type and variants of SARS-CoV-2,^33,34 and affinity maturation was suggested as a good prognostic factor for re-infections, provided that patients presenting high avidity had a mild course of COVID-19.³⁵ Moreover, avidity correlated well with neutralizing antibodies, important for COVID-19 protection.^34,36 From IgG subclasses, it is known that mouse IgG1, predominant for the RBD response (Figure 3) interact with macrophages, neutrophils, and natural killer cells via Fc-γRIII binding.³⁷ Fc-mediated functions limited COVID-19 pathogenesis in K18-hACE2 mice.³⁸ In summary, a broad humoral response, with neutralizing antibodies and Fc-functions seems beneficial for COVID-19 control.³⁹

The IgG from groups immunized with adjuvants also recognized the Omicron variant (Figure 2). Likewise, in Jian et al.²⁵ a preparation of Salmonella-OMVs decorated with SARS-CoV-2 RBD induced antibodies that reacted with Delta variant. Our results show an intermediary avidity for RBD-Omicron, suggesting some biological function. Importantly, a neutralization assay would be required to fully support the assumption of protection against Omicron and Wuhan variants.

Given the OMVs inner immunogenicity,⁷ it is expected that its use, even as adjuvant, would elicit antibodies specific to bacterium antigens. We observed a robust IgG response for the OMVs (Figure 4), agreeing with the immunogenicity of Neisseria vesicles.^40–42 IgG specific to bacterium antigens in RBD+OMV+AH groups was not reduced if compared to the group that received only OMV+AH, suggesting that vesicles maintain their immunogenicity when combined with heterologous antigens. Comparing the separated and combined administration of a 9-valent N. meningitidis OMV and the 13-valent pneumococci conjugated vaccine in mice, the combined administration had no effect in the OMV response and even enhanced anti-pneumococci titers.⁴³

As presented, the immunization with RBD+OMV+AH induced high avidity IgG-anti OMVs (Table 2). The literature established, from mouse and human studies, a good correlation between avidity of antibodies and bactericidal activity for N. meningitidis.^44–46

Considering the isotypes, IgG1, IgG2a, and IgG2b were detected; on the other hand, IgG3 was much closer to the cutoff (Figure 5). Vaccination using the Class 1 outer membrane protein of meningococci achieved higher bactericidal dilution, while mice mainly presented IgG1 and IgG2a⁴⁵; in humans, a N. meningitidis B OMV vaccine induced bactericidal antibodies while dominated by IgG1 isotype, which would be similar to mouse IgG2a.¹⁶ Besides the IgG isotype, antibody abundance and multiple epitopes-binding were pointed as key to opsonize Neisseria efficiently, favoring complement activation and bactericidal activity.⁴⁷ In this aspect, OMVs are interesting options for meningococcal vaccines. Even though the humoral response data (IgG titer, avidity, and isotypes) suggests functionality, the serum-bactericidal assay would be required to prove it, as the gold standard for meningococcal disease protection.⁴⁸

It is desirable that new SARS-CoV-2 vaccines provide humoral and cellular responses for a broad immunity.³ The PAMPs in OMVs are likely to activate Th1 and Th2 signaling pathways, thus inducing a mixed type of response.⁴⁹ Our results suggested Th2-polarized response for SARS-CoV-2, mediated by IgG1, and mixed Th1/Th2 response for N. meningitidis, given the presence of IgG1, IgG2a, and IgG2b (Figures 3 and 5).²⁰ ELISpot corroborated such results by IL-4 secretion but also showed IFN-γ-secreting cells, irrespective of the stimuli (Figure 6). Taken together, these results suggest an overall mixed Th1/Th2 response for both antigens, what agrees with several investigations: Escherichia coli OMVs resulted in balanced IgG1/IgG2a, IFN-γ, IL-12 and IL-6 secretion,²⁸ OMVs from Helicobacter pylori elicited IgG1 and IgG2c, along with IL-4 and IFN-γ cytokines,⁵⁰ and the N. meningitidis OMV vaccines approved for human use led to IFN-γ and IL-4 secretion.⁴¹ Apart from cellular response, ELISpot also suggests the presence of memory cells activated upon antigenic stimulation, which is extremely important for protection and supported by adjuvant use.²

The dose–response data suggest that different RBD concentrations (0.5, 1.0, or 1.5 μg) would not impair the overall humoral response if three SC doses were administered, with similar IgG levels and avidity after the last dose and no remarkable difference in isotypes, but 0.5 μg of RBD resulted in superior cellular response. It has been discussed that the affinity between the antigen and the T cell receptor is more important than antigenic dosage to elicit memory cells, the hypothesis being that high-affinity cells overgrow the low-affinity ones when there is limited antigen available, resulting in the selection of a superior cellular response if lower antigenic doses are used.^51–53

The offspring presented specific IgG for RBD-Wuhan, RBD-OMVs and cross-reactive IgG for RBD-Omicron, at similar levels and avidity of their mothers.¹⁵ Although gestation period and placenta constitution limit the comparison between mouse model to humans,⁵⁴ such results are still interesting and raise a critical point of discussion, provided that it is important to prevent infectious disease in newborns. Although COVID-19 was less frequent in the youngest, there is a risk for developing multi-inflammatory system syndrome and large studies showed considerable pediatric hospitalizations after the emergence of Omicron variant⁵; thus, the main risk group for developing meningococcal disease are infants younger than 1 year of age.⁵⁵ Maternal antibodies seem to be protective against COVID-19, as IgG from cord-blood showed neutralization against the Wuhan strain and, in less extent, to Alpha and Delta variants.⁵⁶ Furthermore, protecting mothers is relevant for reducing COVID-19 impact in the gestation, as the infection was associated with enhanced risk of hypertension disorders, preterm birth, ICU admission, and even maternal mortality.⁵⁷ Bactericidal antibodies against N. meningitidis were also detected in sera of children up to 3 months old, possibly due to maternal transmission.⁵⁸

Taken all these data together, a preparation adjuvanted by N. meningitidis OMVs has not only the potential to enhance immunogenicity of a recombinant SARS-CoV-2 antigen but also to induce an immune response to the bacterium. Such a combined response might be a useful tool, since viral infections propitiate bacterial adhesion and invasion on the respiratory surface.⁵⁹ Some SARS-CoV-2 and N. meningitidis co-infections were published during the pandemic,^60,61 and cases of meningococcal disease by the unsual serogroup X too.⁶² After the pandemic, increased trends of meningococcal disease have been observed.^63–65

Even though this study does not impact current COVID-19 vaccines, it opens some perspectives for future research: it would be interesting to test a preparation without AH to verify the sole adjuvancity of the vesicles, and intranasal delivery could be performed to evaluate mucosal-immunity. From a broad standpoint, OMVs are immunogenic, safe, and an accessible option which could be explored for multivalent vaccines.

Strengths and limitations

OMVs are widely tested as adjuvant, but little is discussed about the mixed-response they might trigger, directed to the antigen of interest and to the bacterial ones. Our manuscript described this aspect in detail, using OMV antigens in parallel with SARS-CoV-2 RBD for all assays; the antigenic preparation proposed here is easy to formulate, and it was evaluated in different dosages. Maternal–fetal transference of antibodies and newborn protection have not been widely studied in COVID-19, so both experimental and human investigations are warranted. Our study has limitations as well: we did not verify if the immune response would last until mice were middle-aged or elderly; we could not proceed the neutralization of SARS-CoV-2 and serum-bactericidal assay for N. meningitidis, two very important assays to predict protection; we used isogenic mice as a model, which does not reflect human immunology perfectly, including the impact of maternal–fetal transference of IgG and if there is any protection of the offspring; and only two mice per group were assayed for cellular response.

Conclusions

Our results led to the conclusion that OMV+AH was a promising adjuvant option for RBD; three SC doses induced comparable IgG levels and avidity, regardless of the protein dosage, but a lower protein concentration conferred the best cellular response. Antibodies recognized RBD from Omicron variant, although with only intermediary avidity, and the offspring received IgG of similar aspects of their mothers. Finally, using OMVs as adjuvant elicited a specific N. meningitidis immune response, so its use may support bivalent vaccines to confer protection from meningococcal disease too.

Acknowledgments

The authors thank the isolation of N. meningitidis C:2a:P1.5 strain by Dr Ana Paula Lemos (Bacteriology Center, Adolfo Lutz Institute, São Paulo, SP, Brazil), and the plasmids donations for RBD production, from Dr Florian Krammer (Icahn School of Medicine of Mount Sinai Hospital, New York, NY, USA) (RBD-Wuhan) and from Dr Silvia Beatriz Boscardin (Parasitology Department of Institute of Biomedical Sciences, University of São Paulo, São Paulo, SP, Brazil) (RBD-Omicron) to Dr Carlos Roberto Prudencio.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2346963.

Additional information

Funding

This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo/FAPESP [grant numbers 2018/04202-0, 2021/11936-3, 2022/05566-1], Conselho Nacional de Desenvolvimento Científico e Tecnológico/CNPq [grant number 305301/2022-5], Coordenação de Aperfeiçoamento de Pessoal de Nível Superior/CAPES (finance code 001) and São Paulo Secretary of Health/FESIMA.