GMMA Is a Versatile Platform to Design Effective Multivalent Combination Vaccines

Technology platforms are an important strategy to facilitate the design, development and implementation of vaccines to combat high-burden diseases that are still a threat for human populations, especially in low- and middle-income countries, and to address the increasing number and global distribution of pathogens resistant to antimicrobial drugs. Generalized Modules for Membrane Antigens (GMMA), outer membrane vesicles derived from engineered Gram-negative bacteria, represent an attractive technology to design affordable vaccines. Here, we show that GMMA, decorated with heterologous polysaccharide or protein antigens, leads to a strong and effective antigen-specific humoral immune response in mice. Importantly, GMMA promote enhanced immunogenicity compared to traditional formulations (e.g., recombinant proteins and glycoconjugate vaccines), without negative impact to the anti-GMMA immune response. Our findings support the use of GMMA as a “plug and play” technology for the development of effective combination vaccines targeting different bugs at the same time.


Introduction
GMMA (Generalized Modules for Membrane Antigens) is a technology platform with great potential for affordable and effective vaccines [1]. Gram-negative bacteria spontaneously release blebs of the outer membrane, also called Outer Membrane Vesicles (OMV), that have been considered attractive for vaccine design [2,3]. OMV, in fact, are basically components of Gram-negative bacteria containing surface-exposed antigens in native conformation and orientation, together with immunostimulatory molecules, such as lipopolysaccharide (LPS), lipoproteins or peptidoglycans. GMMA are OMV derived from bacteria genetically engineered to enhance OMV release [4,5]. Mutations are also introduced to modify the lipid A structure of LPS to minimize the capacity of GMMA to promote reactogenic response once injected, still maintaining an immunopotentiator effect of the Toll-like receptor 4, which is triggered by lipid A [6,7]. GMMA can be produced at high yields using a simple and robust process of manufacturing, potentially leading to affordable vaccines [1,4]. An advanced GMMA-based vaccine to prevent Shigella sonnei infection has been shown to be well-tolerated and immunogenic in clinical trials in healthy adults and endogenous population [8][9][10]. reaction mixture. The mixture was incubated at 25 • C for 30 min, then activated GMMA were purified by ultracentrifugation (110,000 rpm, 16 min, 4 • C). Resulting GMMA (70% recovery by micro BCA) had 43.8% of NH 2 groups derivatized with the BS3 linker, according to the TNBS colorimetric method [20].
Purified activated GMMA were added to FdeC in PBS buffer with a w/w ratio of GMMA to protein antigen of 1:1 and a GMMA concentration of 5.75 mg/mL. After gently mixing overnight at room temperature, the conjugate was purified by ultracentrifuge (110,000 rpm, 4 • C, 1 h) and resuspended in PBS.

Synthesis of the Bivalent Conjugate
Purified GMMA activated with the BS3 linker were added to FdeC protein antigen in PBS buffer with a w/w ratio of GMMA to FdeC of 1:1 and a GMMA concentration of 6.45 mg/mL. After gently mixing overnight at room temperature, the conjugate (S. sonnei GMMA BS3-FdeC) was purified by ultracentrifuge (110,000 rpm, 4 • C, 1 h) and resuspended in NaPi 100 mM, pH 6.5, for a further conjugation step. After having verified conjugate formation by SDS-PAGE/Western blot, the S. sonnei GMMA BS3-FdeC conjugate at the concentration of 2.1 mg/mL was incubated with NaIO 4 5 mM for 30 min at a 25 • C controlled temperature, in the dark. Excess of NaIO 4 was quenched with Na 2 SO 3 at a final concentration of 10 mM, for 15 min at room temperature. SslE protein antigen (w/w ratio of GMMA conjugate to SslE 1:1 at a GMMA concentration of 1.18 mg/mL) and NaBH 3 CN (1:1 w/w ratio with GMMA) were directly added to the reaction mixture. After gently mixing overnight at room temperature, the conjugate ((S. sonnei GMMA-BS3-FdeC)ox-SslE) was purified by ultracentrifuge (110,000 rpm, 4 • C, 1 h) and resuspended in PBS.

Linkage of Heterologous Saccharides to GMMA
Conjugation via SIDEA (Adipic Acid Bis(N-hydroxysuccinimmide)) Chemistry MenA, MenC or Hib oligosaccharides terminally activated with SIDEA as previously described [21] were added to a suspension of GMMA in NaPi 50 mM, pH 7.2. The mixture was stirred overnight at room temperature. Different conjugation conditions were used according to the polysaccharide linked and the GMMA used, as detailed in Table S1. Conjugates were purified by ultracentrifugation (110,000 rpm, 4 • C, 1 h) and recovered in PBS. For the synthesis of the bivalent conjugate, the same procedure was used by simultaneously adding MenA and MenC oligosaccharides to GMMA.

Conjugate Characterization
Conjugates were characterized by micro BCA for total protein recovery and SDS-PAGE/Western blot analysis to confirm conjugate formation ( Figure S1). To quantify the amount of linked protein antigen, amino acid quantification was used. Amount of saccharide antigen linked was determined by HPAEC-PAD after performing acid hydrolysis directly on GMMA, as previously described [22][23][24][25][26][27]. It was verified that there was no interference from GMMA in the quantification of each saccharide. Nanoparticle Tracking Analysis (NTA) was used to count the number of GMMA particles in solution [19] and estimate the number of polysaccharide chains per GMMA (Table S1).
Absence of free proteins was estimated by analysis of the purified conjugates by HPLC-SEC by using a TSK gel 4000-6000 PW columns (Tosoh Bioscience LCC, King of Prussia, PA, USA) in series and eluting with PBS for 60 min at a flow rate of 0.5 mL/min [19]. Percentage of free saccharide was calculated by solid phase extraction (SPE) using a C4 cartridge (BGB Analytik, Böckten, Switzerland) followed by HPAEC-PAD analysis [25,26].
A detailed characterization of all GMMA conjugates tested in this study is reported in Table S1.

Amino Acid Analysis
Gas phase hydrolysis of the samples (GMMA conjugates and corresponding GMMA alone and unconjugated protein antigen) was performed. All samples were prepared in triplicate, as follows: a volume corresponding to 20 µg total protein as estimated by micro BCA was pipetted into a glass test tube (Waters, Milford, CT, USA) and evaporated to dryness using a centrifugal evaporator (ThermoFisher Scientific, Waltham, MA, USA). Once dried, samples were placed inside the hydrolysis vessel and 200 µL of a 6 M HCl solution containing 0.1% (w/v) phenol was added inside the vessel outside the sample tubes. After performing 3-4 alternating vacuum-nitrogen flushing steps to remove oxygen from the hydrolysis vessel, the sealed vessel was placed in an oven at 114 • C for the required time (usually 16, 24 or 48 h). Then, sample tubes were dried under vacuum in a SpeedVac.
Hydrolyzed samples were resuspended in 100 µL of 100 mM HCl, vortexed, then 10 µL were derivatized with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) and analyzed by Ultra-Performance Liquid Chromatography (UPLC) following Waters AccQTag Ultra kit instructions. A calibration curve (in the range 15-250 nmol/mL for each amino acid) was built by diluting the Amino Acid Hydrolysate Standard (2.5 mM) from the Waters kit with Milli-Q water. The A260 of the derivatized amino acids were measured with a Photometric Diode Array (PDA) detector. Empower 3 (Waters) software was used for system control and data acquisition.
The proportion of amino acids in the conjugate coming from the antigen (β a ) and from the GMMA (β G ) was determined by finding the values of β a and β G that minimized the sum of errors squared (Equation (1)).
where α ci is the proportion of the ith amino acid in the conjugate hydrolysate, α ai is the proportion of the ith amino acid in the antigen hydrolysate and α Gi is the proportion of the ith amino acid in the GMMA hydrolysate, and this is summed over 15 amino acids (20 natural amino acids excluding Asn and Gln that are converted to Asp and Glu respectively, and Trp, Cys and Met, that are totally or partially destroyed during hydrolysis). Note that β a + β G = 1.

SDS-PAGE/Western Blot
GMMA conjugates were analyzed by SDS-PAGE/Western blot in comparison to free protein antigens and GMMA to verify conjugate formation. Samples (5-20 µL, with a protein content of 2-10 µg) were mixed with NuPAGE SDS sample buffer (1/5 v/v) and loaded on the gel (3-8% or 7% Tris-acetate NuPAGE, Invitrogen, Carlsbad, CA, USA). The gel was electrophoresed at 45 mA in NuPAGE Tris-Acetate SDS running buffer (20x, Invitrogen). The gel was trans blotted (iBlot, Invitrogen) to nitrocellulose membrane (Invitrogen). The membrane was blocked with 3% BSA-PBS Tween-20 0.05% and immuno-stained with primary antibody (2 h at room temperature, RT) followed by anti-mouse or anti-rabbit IgG (1 h at RT) conjugated to alkaline phosphatase or to Horseradish Peroxidase respectively, and developed with the alkaline phosphatase substrate kit (Sigma-Aldrich, St. Louis, MO, USA) or Opti-4CN (Bio-Rad, Hercules, CA, USA). Both primary and secondary antibodies were diluted in 0.1% BSA-PBS with 0.05% Tween-20 (BSA-PBS Tween buffer). CD1 5-week-old outbred female mice were immunized subcutaneously (s.c.) or intramuscularly (i.m.) at days 0 and 28. Adult rats were immunized i.m. at days 0 and 28. For all formulations with aluminum hydroxide (Alhydrogel), it was verified that the antigens (GMMA, free antigens and conjugates) were almost completely adsorbed (<10% free antigens as verified by SDS-PAGE silver staining).

Immunogenicity Studies in Mice and Rats
Anti-antigen-specific IgG levels were measured at days 1, 27 and 42 by enzyme-linked immunosorbent assay (ELISA) [28]. Purified CSP, Pfs25, fHbp v1, SslE and FdeC proteins were used for ELISA plate-coating at 1 µg/mL in carbonate buffer or PBS buffer (the latter for SslE and FdeC). Purified O-antigen (OAg) from S. Typhimurium was used for ELISA plate-coating at 5 µg/mL in carbonate buffer, purified MenA and MenC capsular polysaccharides were used at 5 µg/mL in PBS pH 8.2, purified Shigella sonnei LPS was used at 0.5 µg/mL in PBS and Hib polysaccharide conjugated to Human Serum Albumin (HSA) was used at 2 µg/mL in PBS. ELISA units were expressed relative to a mouse antigen-specific antibody standard serum curve, with the best five-parameter fit determined by a modified Hill plot. One ELISA unit is defined as the reciprocal of the dilution of the standard serum that gives an absorbance value equal to 1 in this assay. Each mouse serum was run in triplicate.

Statistical Analysis
Datasets were analyzed using a two-tailed nonparametric Mann-Whitney test with Prism (GraphPad Software, San Diego, CA, USA). p-values lower than 0.05 were considered statistically significant, and p-values were rounded to the nearest larger number.

GMMA as a Carrier for Protein Antigens
We started by testing the concept of GMMA as a carrier for protein antigens. We used P. falciparum Pfs25 and CSP as model antigens and displayed them on S. Typhimurium GMMA, with the potential to cover both invasive nontyphoidal salmonellosis and malaria. Since both diseases affect mainly young children in Africa [31][32][33], there is a rationale for a combination vaccine.
To determine whether conjugation to GMMA led to an increased immunogenicity of the protein antigens, we compared the humoral immune response of mice immunized with protein antigen-GMMA conjugates with corresponding recombinant proteins alone or physically mixed with GMMA, at the same doses of antigen and GMMA.
We observed that at low immunogenic doses of Pfs25 or CSP antigens [34][35][36][37], the conjugation to S. Typhimurium GMMA strongly enhanced the anti-antigen-specific IgG response compared not only to the antigen alone but also to the antigen physically mixed to GMMA ( Figure 1A,B). The enhancement of the antigen-specific antibody response was observed after just one immunization. Importantly, we also observed that the presence of the malaria protein antigens on GMMA surface did not impact on the serum IgG response to Salmonella OAg, considered the major protective antigen for this pathogen ( Figure 1D,E).
Subsequently, we immunized mice with MenB fHbp chemically conjugated to S. Typhimurium GMMA or with the corresponding physical mixture of fHbp and GMMA at the same doses or GMMA and fHbp alone. There is a clinical rationale in developing a vaccine that targets both nontyphoidal Salmonella and Neisseria meningitidis, both prevalent in Sub-Saharan Africa [38][39][40]. This Neisseria meningitidis antigen allowed us to measure not only the antigen-specific IgG production, but, also the functionality of the antigen-specific antibody response. For Neisseria meningitis, SBA is considered a correlate of protection in humans and consequently, it is an optimal functional assay to generate a proof of concept of the effectiveness of the humoral immune response in mice [41,42].
The GMMA conjugate strongly enhanced the anti-fHbp IgG response compared to the physical mixture or the antigen alone after one immunization, but not after the second ( Figure 1C). Although the magnitude of the responses after the second immunization was similar for the fHbp-GMMA conjugate and for fHbp, there was a marked difference in the quality of the response as judged by two criteria: (1) The SBA elicited when tested on MenB with an homologous sequence of fHbp was substantially higher with fHbp-GMMA than with unconjugated fHbp alone or mixed with GMMA ( Figure 2), and (2) the specificity of the response: fHbp-GMMA, but not unconjugated fHbp alone or the mixture of fHbp and GMMA, elicited a broad specificity, reacting strongly with MenA and MenW strains with heterologous fHbp sequences ( Figure 2). substantially higher with fHbp-GMMA than with unconjugated fHbp alone or mixed with GMMA ( Figure 2), and (2) the specificity of the response: fHbp-GMMA, but not unconjugated fHbp alone or the mixture of fHbp and GMMA, elicited a broad specificity, reacting strongly with MenA and MenW strains with heterologous fHbp sequences ( Figure 2). Importantly, the functionality of the sera, as measured by SBA using the S. Typhimurium invasive Malawian clinical isolate D23580, was not decreased by chemical conjugation of fHbp to GMMA ( Figure 2). This result is consistent with the lack of impact on the serum IgG response to Salmonella OAg ( Figure 1F), as already observed with malaria antigens (Figure 1D,E). . Summary graphs of anti-antigen-specific IgG geometric mean units (bars) and individual antibody levels (dots) are reported. In graphs (D-F), no significant differences in ELISA units were observed. Datasets were analyzed using a two-tailed nonparametric Mann-Whitney test. p-values (rounded to the nearest larger number) less than 0.05 were considered statistically significant.  . Summary graphs of anti-antigen-specific IgG geometric mean units (bars) and individual antibody levels (dots) are reported. In graphs (D-F), no significant differences in ELISA units were observed. Datasets were analyzed using a two-tailed nonparametric Mann-Whitney test. p-values (rounded to the nearest larger number) less than 0.05 were considered statistically significant. substantially higher with fHbp-GMMA than with unconjugated fHbp alone or mixed with GMMA ( Figure 2), and (2) the specificity of the response: fHbp-GMMA, but not unconjugated fHbp alone or the mixture of fHbp and GMMA, elicited a broad specificity, reacting strongly with MenA and MenW strains with heterologous fHbp sequences ( Figure 2). Importantly, the functionality of the sera, as measured by SBA using the S. Typhimurium invasive Malawian clinical isolate D23580, was not decreased by chemical conjugation of fHbp to GMMA (Figure 2). This result is consistent with the lack of impact on the serum IgG response to Salmonella OAg ( Figure 1F), as already observed with malaria antigens (Figure 1D,E).   Importantly, the functionality of the sera, as measured by SBA using the S. Typhimurium invasive Malawian clinical isolate D23580, was not decreased by chemical conjugation of fHbp to GMMA (Figure 2). This result is consistent with the lack of impact on the serum IgG response to Salmonella OAg ( Figure 1F), as already observed with malaria antigens (Figure 1D,E).

GMMA as an Effective Carrier for Polysaccharides
Having verified that GMMA used as a carrier for protein antigen generated a significant enhanced immunogenicity for the antigen, we asked whether GMMA could be an effective carrier for polysaccharides.
MenC and MenA oligosaccharides were conjugated to MenB GMMA. To evaluate the capacity of saccharide-GMMA conjugates to elicit an anti-saccharide humoral immune response, we compared the saccharide-specific antibody response induced in mice by saccharide-GMMA conjugates, the corresponding physical mixtures of saccharide and GMMA or by CRM197 glycoconjugates as gold standards. In the presence of Alhydrogel, both saccharide-GMMA conjugates induced anti-saccharide-specific serum IgG levels comparable to corresponding CRM197 glycoconjugates, whereas the physical mixtures of GMMA and saccharides were poorly or non-immunogenic ( Figure 3A,B). We also measured the capacity of sera derived from MenA-or MenC-immunized mice to promote the rabbit complement-mediated lysis of Neisseria meningitidis, by using a previously described in vitro bactericidal assay [30,41,42]. Immunization with saccharide-GMMA conjugates resulted in a stronger functional response ( Figure 3C,D) compared to CRM197 glycoconjugates. Indeed, two weeks after the last immunization, the SBA measured in mice immunized with MenA-or MenC-GMMA conjugates was much higher than that induced by corresponding CRM197 glycoconjugates. The same result was not obtained by simply physically mixing oligosaccharides and GMMA, even if SBA titers against MenA and MenC strains were measurable, due to the immune response induced by GMMA per se. We also investigated persistence of the response induced and found that SBA titers remained higher in mice immunized with GMMA conjugates compared to CRM197 conjugates six months after the last immunization, despite a decrease in activity ( Figure 3C,D).
To understand whether the results obtained with MenA and MenC oligosaccharides were independent from the species involved in the GMMA production, we generated similar results using S. Typhimurium GMMA as a carrier. The data further confirmed the generally superior carrier effect of GMMA compared to CRM197 ( Figure S2).
When administered in absence of Alhydrogel, unlike the response elicited by the CRM197 glycoconjugate, MenC-GMMA (MenB) conjugate generated an anti-saccharide-specific IgG response after a single injection. Similar results were obtained by immunizing mice with Hib oligosaccharides conjugated to MenB GMMA ( Figure 4). The superior functionality of saccharide-specific antibody response generated when immunizing with MenC-GMMA conjugate was also observed in absence of Alhydrogel formulation (Figure 4).

Chemical Linkage of Two Different Proteins or Saccharides on the Same GMMA Did Not Generate Immunointerference
Through chemical conjugation, we can also display multiple antigens on the same GMMA particle. We verified this by linking two different E. coli antigens, SslE [43,44] and FdeC [45], on the same S. sonnei GMMA particle [4], with the potential to protect against two major diarrheagenic pathogens, Enterotoxigenic E. coli (ETEC) and Shigella [46]. The conjugate presenting SslE and FdeC on the same GMMA particle (Table S1) induced anti-E. coli antigen-specific IgG response and anti-S. sonnei LPS IgG response, not significantly different from the anti-antigen-specific IgG response induced by corresponding monovalent conjugates ( Figure 5).
specific IgG response. Summary graphs of anti-PS IgG geometric mean units (bars) and individual antibody levels (dots) are reported. SBA titers of pooled sera collected at day 42 from each group against MenC strain are reported in the table in panel A. Datasets were analyzed using a two-tailed nonparametric Mann-Whitney test. p-values (rounded to the nearest larger number) less than 0.05 were considered statistically significant.

Chemical Linkage of Two Different Proteins or Saccharides on the Same GMMA Did Not Generate Immunointerference
Through chemical conjugation, we can also display multiple antigens on the same GMMA particle. We verified this by linking two different E. coli antigens, SslE [43,44] and FdeC [45], on the same S. sonnei GMMA particle [4], with the potential to protect against two major diarrheagenic pathogens, Enterotoxigenic E. coli (ETEC) and Shigella [46]. The conjugate presenting SslE and FdeC on the same GMMA particle (Table S1) induced anti-E. coli antigen-specific IgG response and anti-S. sonnei LPS IgG response, not significantly different from the anti-antigen-specific IgG response induced by corresponding monovalent conjugates ( Figure 5). Figure 5. SslE and FdeC presented on the same S. sonnei GMMA particle compared to corresponding monovalent conjugates. Eight CD1 mice per group were i.m. immunized at days 0 and 28, with 5 µg total protein, in the presence of Alhydrogel. Sera were collected at days -1, 27 and 42 and analyzed for anti-E. coli antigen-specific (A,B) and anti-S. sonnei LPS (C) IgG response. Summary graphs of IgG geometric mean units (bars) and individual antibody levels (dots) are reported. Datasets were analyzed using a two-tailed nonparametric Mann-Whitney test. p-values (rounded to the nearest larger number) less than 0.05 were considered statistically significant.
We also linked similar amounts of MenA and MenC oligosaccharides (Table S1) on S. Typhimurium GMMA. Also, in this case, no immune interference was detected by linking the two different saccharides on the same GMMA: the resulting conjugate elicited antibodies with strong Figure 5. SslE and FdeC presented on the same S. sonnei GMMA particle compared to corresponding monovalent conjugates. Eight CD1 mice per group were i.m. immunized at days 0 and 28, with 5 µg total protein, in the presence of Alhydrogel. Sera were collected at days 1, 27 and 42 and analyzed for anti-E. coli antigen-specific (A,B) and anti-S. sonnei LPS (C) IgG response. Summary graphs of IgG geometric mean units (bars) and individual antibody levels (dots) are reported. Datasets were analyzed using a two-tailed nonparametric Mann-Whitney test. p-values (rounded to the nearest larger number) less than 0.05 were considered statistically significant.
We also linked similar amounts of MenA and MenC oligosaccharides (Table S1) on S. Typhimurium GMMA. Also, in this case, no immune interference was detected by linking the two different saccharides on the same GMMA: the resulting conjugate elicited antibodies with strong bactericidal titers against MenA and MenC strains, similar to that induced by immunization with the monovalent MenA or MenC-GMMA conjugates ( Figure 6). Again, the functionality of antibody response to S. Typhimurium was not negatively affected ( Figure 6). bactericidal titers against MenA and MenC strains, similar to that induced by immunization with the monovalent MenA or MenC-GMMA conjugates ( Figure 6). Again, the functionality of antibody response to S. Typhimurium was not negatively affected ( Figure 6).

Discussion
There is still a huge global burden of infectious diseases, causing 91% of deaths in low-and middle-income countries [47], where implementation of existing vaccines is also difficult [15]. In the absence of commercial incentives, simple affordable technologies to target multiple diseases represent a preferred route for vaccine development. Use of technology platforms appear also beneficial to accelerate design, development and implementation of vaccines to fight antimicrobial resistance [48]. GMMA is a simple and effective platform to fight bacterial pathogens. Preclinical and clinical data have confirmed the potential of GMMA to induce strong immune responses [4,5,[8][9][10]49,50]. Here, we have documented the use of GMMA as a "carrier" for antigens from other pathogens, as a general way to design multivalent combination vaccines against serious infectious diseases.
Expression of protein or saccharide antigens on OMV surface induces a strong antigen-specific immune response [3,51]. Recently, nanoparticle systems, including virus-like particles, gold nanoparticles and liposomes, have been explored for the display of carbohydrate and protein antigens, combining the multivalent presentation with the special physico-chemical properties of nano-sized particles [52][53][54]. Outer membrane protein complex (OMPC) from Neisseria meningitidis has been used as a carrier in the licensed Haemophilus influenzae type b conjugate vaccine [55,56] and also as a carrier for protein antigens, such as Pfs25H and Pfs230 malaria antigens, whose linking to OMPC resulted in an increased immunogenicity in mice compared to Pfs25H alone or physically mixed to OMPC and Pfs230-EPA, respectively [11,57].
Here, we selected chemical conjugation to link different types of antigen, even derived from phylogenetically distant pathogens, to GMMA surface. Chemical conjugation allows to play with antigens' density, length and attachment site, facilitating the design of optimal vaccines. We are currently investigating these aspects. Once the optimal design for a specific vaccine combination has

Discussion
There is still a huge global burden of infectious diseases, causing 91% of deaths in low-and middle-income countries [47], where implementation of existing vaccines is also difficult [15]. In the absence of commercial incentives, simple affordable technologies to target multiple diseases represent a preferred route for vaccine development. Use of technology platforms appear also beneficial to accelerate design, development and implementation of vaccines to fight antimicrobial resistance [48]. GMMA is a simple and effective platform to fight bacterial pathogens. Preclinical and clinical data have confirmed the potential of GMMA to induce strong immune responses [4,5,[8][9][10]49,50]. Here, we have documented the use of GMMA as a "carrier" for antigens from other pathogens, as a general way to design multivalent combination vaccines against serious infectious diseases.
Expression of protein or saccharide antigens on OMV surface induces a strong antigen-specific immune response [3,51]. Recently, nanoparticle systems, including virus-like particles, gold nanoparticles and liposomes, have been explored for the display of carbohydrate and protein antigens, combining the multivalent presentation with the special physico-chemical properties of nano-sized particles [52][53][54]. Outer membrane protein complex (OMPC) from Neisseria meningitidis has been used as a carrier in the licensed Haemophilus influenzae type b conjugate vaccine [55,56] and also as a carrier for protein antigens, such as Pfs25H and Pfs230 malaria antigens, whose linking to OMPC resulted in an increased immunogenicity in mice compared to Pfs25H alone or physically mixed to OMPC and Pfs230-EPA, respectively [11,57].
Here, we selected chemical conjugation to link different types of antigen, even derived from phylogenetically distant pathogens, to GMMA surface. Chemical conjugation allows to play with antigens' density, length and attachment site, facilitating the design of optimal vaccines. We are currently investigating these aspects. Once the optimal design for a specific vaccine combination has been identified, genetic tools could be used for antigen expression on GMMA surface, further reducing costs for vaccine manufacture.
In the cases investigated, GMMA significantly improved the antigen-specific humoral immune response. When immunizing with GMMA-antigen conjugates, compared to antigen alone or physically mixed with GMMA, the antigen-specific IgG production was prevalently enhanced, but more interestingly, the functionality of the antibody was generally increased. Thus, GMMA used as an antigen carrier may generate a more effective immunization. Consistently, in our studies, we found immunization with GMMA-antigen conjugates as the best treatment to obtain a good antibody response after one administration. Also, by working with saccharide antigens, we demonstrated that the ability of GMMA conjugates to induce total polysaccharide-specific IgG titers was not inferior to more traditional CRM197 conjugates, but with stronger functionality, confirming what was recently found for nontyphoidal Salmonella GMMA compared with established glycoconjugates [49].
No enhanced immunogenicity was seen in mixtures of antigen and GMMA compared to antigen alone, the antigen needs to be displayed on GMMA surface to result in increased immunogenicity. This suggests that GMMA are not simply acting as an adjuvant. We believe that the immunogenicity of GMMA used as an antigen "carrier" could be the result of efficient antigen presentation to antigen-specific lymphocytes. Certainly, the mode of action of this effect of GMMA used as an antigen carrier is worthy of deep investigation because immunological mechanisms might be unveiled that could be exploited to also design more effective GMMA-based combination vaccines. Studies are ongoing to investigate the ability of GMMA to induce antigen-specific T-cell responses and the proportion of memory T-helper cells.
The generation of an effective immunization after one injection is remarkable. Indeed, reducing the number of injections would further decrease the cost of vaccination campaigns as well as improve the affordability of GMMA-based vaccines. In addition, less injections would facilitate the acceptability of the vaccine and the completion of vaccination schedules, consequently improving the immunization coverage. Finally, reducing the number of injections would minimize the adverse events following vaccination due to immunization errors, which is not a trivial question in large immunization campaigns. All these aspects are critical for immunization campaigns in low-income countries.

Conclusions
GMMA are characterized by a simpler manufacturing process compared to other traditional carriers [1,4] and our results demonstrate that GMMA can be used as a carrier for heterologous antigens to improve the immune response to the antigen without affecting the immune response to GMMA. This means that GMMA can play a dual role of antigen and carrier [58], facilitating the development of vaccines that are socially justified but that will require an innovative business case to be sustainable [15]. Importantly, we have shown that different antigens can be linked to the same GMMA particle with no immune interference. These features support the use of the GMMA technology for the development of simple multivalent vaccines, covering multiple diseases at the same time. By using GMMA and antigens (both proteins and polysaccharides) from different pathogens as models, we have demonstrated the broad applicability of this technology.
This technology could be extended to other molecules, including viral proteins and small molecules, and represent a rapid tool to test the potentiality of the GMMA platform beyond its application to Gram-negative bacterial pathogens.
Based on the WHO recommendation to promote innovation in vaccine development to satisfy critical medical needs [18], our preclinical studies offer a proof of concept that GMMA exploited as a carrier might be a successful "plug and play" technology to design effective and affordable multivalent combination vaccines targeting different bugs at the same time.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-393X/8/3/540/s1, Figure S1: GMMA conjugates formation is verified by SDS-PAGE/Western blot analysis. Figure S2: MenA and MenC oligosaccharides were conjugated to S. Typhimurium GMMA and compared in mice to corresponding MenA-or MenC-CRM197 conjugates, GMMA alone or physically mixed to Men oligosaccharides. Table S1: Conjugation conditions used and main characteristics of the GMMA conjugates tested in this study.