Outer Membrane Vesicles: A Challenging Yet Promising Platform for COVID-19 Vaccines

Abstract

The outer membrane vesicles (OMVs) are vesicles released from Gram-negative bacteria, which present a range of biological applications, such as vaccine adjuvants. OMVs present several pathogen-associated molecular patterns, being immunogenic and capable of triggering different arms of the immune response. Thus, they are suitable for mucosal and parenteral delivery, feasible to obtain and have been used in licensed-vaccines previously. However, the extraction protocols and manipulations can modify OMVs cargo and, consequentially, the immunization results. Therefore, this chapter will review OMVs use as adjuvant and discuss results from COVID-19 vaccines which employed this technique.

1. Introduction

Outer membrane vesicles (OMVs) are naturally released vesicles from the outer membrane of Gram-negative bacteria. These vesicles are composed by several antigens, such as lipopolysaccharide (LPS), phospholipids, and specific proteins, according to the bacterium species [1], hence, being important for bacteria evolution and survival [2]. The antigens of OMVs also act as pathogen-associated molecular patterns (PAMPs), and their use as vaccine adjuvants have been discussed [3]. OMVs not only activate innate immunity but also improve humoral and cellular adaptative responses, and they are affordable to obtain and suitable for different immunization routes—parenteral and mucosal. OMVs vaccines against Neisseria meningitidis, a Gram-negative pathogen, have already been approved for human use [4].

The COVID-19 pandemic led to an urge for vaccines. To address this issue, different vaccine platforms have been tested and developed in record time [5, 6]. It is crucial that all countries have access to vaccines to control the disease [7]. Local production supports this universal access, provided that it diminishes the costs of vaccine manufacturing compared to commercial values; thus, it facilitates the distribution [8]. However, the local production should be suitable to infrastructure, expertise, and other particularities of each country [9].

Adjuvants are used to improve vaccine efficacy—they increase the antigen’s immunogenicity, support dose sparing, and modulate the overall immunity toward an adequate pattern of response [10]. This chapter aims to describe OMVs structure, explain their adjuvant effect, and discuss how they could be used for COVID-19 vaccines.

2. Outer membrane vesicles: biology, isolation, and purification

OMVs are negative-charged circular structures, ranging from 20 to 250 nm, produced through the blebbing of the outer membrane of Gram-negative bacteria [3]. This outer membrane is composed of a phospholipid bilayer, expressing lipopolysaccharide (LPS, or endotoxin) and several other antigens, depending on the bacterium species and strain. Below the outer membrane, there is a layer of peptidoglycan and the inner membrane [11].

To release the vesicles, the outer membrane is detached from the peptidoglycan; however, bacterium integrity must be preserved. This process is mediated by cross-linking proteins and lipids: such molecules are expressed in the outer membrane and linked with peptidoglycan, so the modulation of these molecules might increase or decrease the membranes’ attachment [11]. The outer membrane protein A (OmpA) protein, the Tol-Pal complex, and the LPS are examples of cross-linking molecules related to OMV release [12, 13].

The bacterium modulates OMV release to cover different functions: communication between bacteria; transportation of protein, DNA or RNA; metabolites excretion; nutrient acquisition; protection against bacteriophages and biofilm production, among others [1]. Figure 1 shows the organization of a Gram-negative bacterium envelope, the blebbing of the outer membrane, and summarizes the functions of these structures.

Figure 1.

Structure, release, and content of OMVs: Gram-negative bacteria present a complex cell wall, formed by a cytoplasmic membrane, the periplasm, which has a thin peptidoglycan layer and the outer membrane. The membranes are composed of a phospholipid bilayer and present LPS, among other antigens (according to the bacterium strain). The vesicles originate from the blebbing of the outer membrane and carry various structures, such as the LPS, proteins, DNA, and toxins, which are capable of activating the immune system. OMVs: outer membrane vesicles, LPS: lipopolysaccharide, PG: peptidoglycan. (Figure created with BioRender).

Even though the OMVs are produced physiologically by bacterium, the isolation of these structures following natural release provides a limited number of vesicles. Hence, laboratory protocols were standardized to improve OMVs induction [2]. The first techniques consisted of chemical protocols to put the cells under stressful conditions, modulating pH and molarity through detergents and salts [14]. With the popularization of genetic engineering, it became possible to manipulate the bacterium strains to release more vesicles [15]. Moreover, the molecular approach allows to overexpress antigens of interest in the OMVs content [2].

Another application of genetically modified OMVs was developed by GlaxoSmithKline. The generalized modules for membrane antigens (GMMAs) are OMVs obtained by Gram-negative bacteria engineered to improve vesiculation [16]. A Shigella vaccine developed using this strategy entered clinical trials [17]. Nevertheless, this strategy is versatile, since antigens from different pathogens can be coupled to the OMVs—Mycobacterium tuberculosis and Chlamydia trachomatis were proven to work—exploring vesicles as delivery and adjuvant system combined [16, 18].

Given that Gram-negative bacteria express LPS, which is highly reactogenic, this antigen should be purified from the OMVs, respecting the acceptable levels for preclinical and clinical trials [19]. This is the reason that led to the popularization of deoxycholate extraction—the detergent stimulates OMV release and reduces the LPS content concomitantly [1]. When different protocols are used to produce OMVs, the LPS detoxification can be performed using Sepharose-4B-Polymyxin or monoclonal antibodies columns [20]. Recently, a removal protocol using Cationic amphiphilic peptides was proposed [21]. Another option is to use bioengineered strains, modified to express a nontoxic LPS [22].

Finally, it is important to understand that different protocols result in different OMV content. The ideal OMV composition for the planned application should be aligned with the extraction method [23].

3. Outer membrane vesicles and immune activation

OMVs were firstly used for pathogen-specific vaccines against Neisseria meningitidis from serogroup B [4]. However, antigens present in OMVs cargo, regardless of bacteria species, act as pathogen-associated molecular patterns (PAMPs). With the increasing understanding about the interplay between innate and adaptative immunity, PAMPs have been suggested as promising adjuvants [9].

Several pattern recognition receptors (PRRs) recognize OMVs structures. Considering the Toll-like receptors (TLRs) family, Flagellin activates TLR-5, CpG DNA is recognized by TLR-9, ribosomal RNA activates TLR-13, and other lipoproteins can interact with TLRs-1, -2, and -6 [3]. LPS binds to TLR-4, although, as discussed previously, its concentration should be under acceptable levels for clinical use [19]. In addition, OMVs from bacteria like Aggregatibacter actinomycetemcomitans and Escherichia coli activated NOD-(NOD) like receptors in cell culture [24, 25]. Bordetella pertussis vesicles upregulated the NLRP3 inflammasome pathway [26]. The outer membrane protein A (OmpA) from Klebsiella pneumonia was described as an activator of innate and humoral responses interacting with scavenger receptors LOX-1 and SREC, TLR-2, and long pentraxin PTX-3 [27].

Not only PRRs are activated, but different immune pathways could be explored using OMVs as adjuvants. The vesicles of Moxarella catarrhalis and Haemophilus influenzae activated B-cells via IgD receptor [28, 29]. B-cell-independent activation is not the main goal of vaccination, since a T-cell-dependent response is needed for immunologic memory. However, OMVs also support T-cell activation and B-cell proliferation [30]. OMVs from B. pertussis supported cellular and humoral immune response of the mucosa, upregulating genes related to Th17 response and IgA secretion [31].

Importantly, it was verified that preexisting immunity against the vesicles did not affect the adjuvanticity potential [32]. Besides that, the 20–250 nm size of OMVs is ideal for uptake by antigen-presenting cells (APCs) and OMVs upregulate genes related to antigen presentation molecules, such as CD-80, CD-86, and MHC-II in macrophages and dendritic cells [3, 30]. All that considered, the immune activation conferred by OMVs makes them an interesting adjuvant [2].

Noteworthy, activating PRRs triggers cellular signaling that culminates in an inflammatory response [9]. Because of that, the reactogenicity of OMVs should be studied. Rossi et al. [33] reviewed the tests available (such as rabbit pyrogenicity and monocyte activation tests) and their advantages and limitations.

To note, Gram-positive bacteria also release membrane vesicles; however, only recently the scientific community has started to investigate them. Vesicles of Mycobacterium genus are known to activate TLR-2, and extracellular vesicles of Streptococcus pneumoniae and Bacillus anthracis were effective antigens for immunization against these pathogens [34]. Vesicles isolated from Streptococcus pneumoniae were internalized by dendritic cells and enhanced tumor necrosis factor (TNF)-α release, without toxic effects [35]. The use of such vesicles as adjuvants might be a promising field of study too.

4. Previous experience using OMVs in vaccines

The OMVs were first studied as a technology to develop vaccines against N. meningitidis from serogroup B. During the 80s, three OMV-based vaccines were used to control meningococcal disease epidemics: Va-Mengoc-BC (Finlay Institute) in Cuba, MENZB (Novartis) in New Zealand and MenBvac (Norwegian Institute of Public Health) in Norway [4].

Recent efforts to develop an OMV vaccine against Shigella sonnei were undertaken, as well as to use N. meningitidis OMV vaccines to prevent N. gonorrhoeae infection [36, 37]. Supported by the success of the S. sonnei OMV vaccine, GlaxoSmithKline is about to start clinical trials using the same platform for Salmonella vaccines [16]. OMVs seem an interesting platform for B. pertussis vaccines as well, although no candidates went to clinical trials yet [2]. Table 1 summarizes the OMVs-based vaccines which have been studied in clinical trials. The Soberana 01, a COVID-19 vaccine which uses OMVs as adjuvant, will be discussed in more detail in the next section.

Vaccine	Pathogen	Phase	Reference(s)
Va-Mengoc-BC (Finlay Institute)	N. meningitidis	Licensed	[4]
MENZB (Novartis)	N. meningitidis	Licensed	[4]
MenBvac (Norwegian Institute of Public Health)	N. meningitidis	Licensed	[4]
S. sonnei GMMA (1790GAHB) (GlaxoSmithKline)	S. sonnei	Phase I	[17]
S. Typhimurium GMMA (GlaxoSmithKline)	S. Typhimurium	Entering Phase I/II	[16]
S. Enteritidis GMMA (GlaxoSmithKline)	S. Enteritidis	Entering Phase I/II	[16]
Soberana 01 (Finlay Institute)	SARS-CoV-2	Phase II clinical trial	[38, 39]

Table 1.

OMV-based vaccines licensed or in clinical trials against different pathogens.

Nonetheless, OMVs are immunogenic, stable, and feasible to isolate, making them an affordable vaccine platform and adjuvant option [9].

5. COVID-19 vaccines and outer membrane vesicles: promising results

The importance of affordable, accessible vaccines has been highlighted since the beginning of the COVID-19 pandemic [40]. As discussed so far, the vesicles are feasible to obtain, are immunogenic, improve the humoral and cellular arms of the immune response, and are suitable for both parenteral and mucosal delivery [2]. Thus, OMVs have been licensed for human use before [4].

Before SARS-CoV-2, OMVs were explored as vaccine adjuvants for other viruses. A fusion of Zika virus and N. meningitidis OMVs resulted in a promising nanoparticle, which enhanced IgG titers, cytokine markers of Th1/Th2 responses, and immunologic memory, such as IL-2/IL-4 and TGF-β. Moreover, the nanoparticles induced neutralizing antibodies against the Zika virus when administrated alone or combined with mesoporous silica as adjuvant [41].

Different investigations proposed OMVs of E. coli as Influenza adjuvant. In Shim et al. Alum and an oil-in-water emulsion were more effective to enhance the antibody titers, but OMVs modulated the immune response toward a Th1 profile, with IFN-γ and higher IgG2c/IgG1 ratio; thus, the hemagglutinin inhibition titers were higher using OMVs as adjuvants. All that resulted in protection of ferrets after a lethal viral challenge [42]. Watkins et al. also used OMVs as Influenza adjuvant. Similarly, the immunization resulted in the survival of mice from different genetic backgrounds after lethal viral challenges using H1N1 and H3N2 viruses. Thus, the humoral response presented a mixed Th1/Th2 profile. Interestingly, the authors observed that the OMVs promoted DCs activation and maturation, even though they expressed a modified LPS with reduced reactogenicity [43].

A multivalent Influenza/Middle East Respiratory Syndrome Coronavirus (MERS-CoV) vaccine adjuvanted was recently proposed. OMVs from E. coli presenting the hemagglutinin of Influenza H1N1 and the receptor-binding domain (RBD) of MERS-CoV induced neutralizing antibodies against both viruses and immunized animals survived an H1N1 challenge [44]. Taken together, these results show that OMVs are a versatile tool to adjuvant heterologous antigens, even in multivalent strategies.

The COVID-19 vaccine database of the World Health Organization (WHO) (updated on July 22, 2022) describes two vaccines using OMVs in preclinical trials. Plus, Soberana 01, from Finlay Institute, uses OMVs as adjuvants and is going into Phase II trials [38, 45]. Even though this is an unexpressive number, promising results were obtained in preclinical experiments.

The receptor binding domain (RBD) of SARS-CoV-2, adjuvanted by OMVs from N. meningitidis (from serogroups B or C) and Alum, improved IgA and IgG titers, thus, IL-1 and IL-17 secretion following a mixed intramuscular/intranasal delivery [46]. Besides that, the adjuvants contributed to increasing avidity and neutralization of antibodies [47], which are relevant parameters for COVID-19 humoral response [48]. Similarly, the Spike protein adjuvanted by OMVs from N. meningitidis induced a robust humoral response, with neutralizing antibodies and prevention of viral replication after challenge [49]. To note, the intranasal delivery of the antigenic preparation induced IgA in the nose and the lungs, conferring mucosal protection [49].

Vibrio cholerae and E. coli OMVs were compared as adjuvants for intranasal RBD immunization [50]. Both preparations elicited a robust systemic and mucosal immune response, supported by neutralizing antibodies. The most promising results were obtained when OMVs from each bacterium were administrated in turn, as a heterogeneous prime-booster approach [50].

An intranasal vaccine composed of the Spike protein and OMVs from Salmonella typhimurium induced not only systemic and mucosal immunity but also neutralizing antibodies against the delta variant. Upon viral challenge, only hamsters immunized with the adjuvanted preparation presented less severe lung pathology and lower virus titers in bronchoalveolar lavage [51].

The Finlay Institute responded to the pandemic developing Soberana 01, which uses RBD as antigen and vesicles from N. meningitidis from serogroup B and alum as adjuvants. In mice, they observed that this formulation was more immunogenic than RBD only adsorbed in Alum: it induced robust B-cell memory and neutralizing antibodies; thus, the antibodies recognized mutant RBD from SARS-CoV-2 variants [39]. Considering clinical use, the group described that the vaccine was safe and immunogenic, and the OMVs contributed to inducing functional, neutralizing antibodies [45].

6. Conclusion

In conclusion, OMVs are versatile tools with several biological applications, including vaccine development. The adjuvanticity of these vesicles relies on the innate immunity activation conferred by PAMPs; however, OMVs expression changes according to bacterium species and extraction protocol. OMVs are stable, safe, and suitable for mucosal and parenteral delivery. In addition, having a history of previous use in vaccines makes them a promising adjuvant candidate for translating to clinical use. Preclinical data showed that OMVs supported functional antibodies and protection against SARS-CoV-2; thus, there are COVID-19 vaccines adjuvanted by OMVs undergoing clinical trials. Nonetheless, OMVs prove how biomolecules produced by microorganisms are remarkable tools for the biomedical area.

Acknowledgments

This chapter was supported by São Paulo Research Foundation (FAPESP) (grant number 18/04202-0) and Coordination for the Improvement of Higher Education Personnel (grant number 88887.661236/2022-00).

Conflict of interest

The authors declare no conflict of interest.