The keys to successful SARS-CoV-2 vaccination are efficient antigen processing presentation by APC cells, the induction of SARS-CoV-2-neutralizing antibodies, and the activation of T cell responses and lasting immune memory. Although many SARS-CoV-2 vaccines satisfy these prerequisites and have entered clinical trials [6, 33], their production costs and outputs may limit the widespread application of these vaccines in the global pandemic. E. coli is the most productive and least costly protein expression system. Vaccines against SARS , dengue fever , Middle East respiratory syndrome , hepatitis E , influenza , and human papillomavirus  have been successfully expressed as nanostructures in E. coli. Therefore, the correct folding or assembly of viral antigens into nanostructures in E. coli could elicit potential immune protection, thus enabling the successful development of an inexpensive vaccine in E. coli.
The RBD of SARS-CoV-2 is the main mediator of the binding between viruses and the host cell receptor ACE2 . It is also the main target of the SARS-CoV-2 vaccine [8, 40–42]. In its natural conformation, the RBD of SARS-CoV-2 shows a trimeric structure [16, 43]. The host immune response caused by this trimeric structure may not be identical to the host immune response caused by the monomeric structure. RBD dimers  and multimers [10, 11] have better immunoprotective ability than RBD monomers. Therefore, multimerization of RBD may be a key step in the design of SARS-CoV-2 vaccines. This study also found that RBD-PP has a stronger ability to bind hACE2 than monomers, suggesting the importance of RBD polymers (Figure 1I).
Because of their scale characteristics, nanovaccines can increase the effectiveness of lymph node targeting and antigen presentation . Their unique structural features can enhance immunogenicity by presenting polyvalent antigens, can stabilize antigens, and can induce adjuvant activities . A variety of organic and inorganic nanoparticles have been used in nanovaccines. For example, antigen-loaded nanomaterials, such as gold [46–48], silica [49, 50], magnetosomes , poly(lactic-co‐glycolic acid) , and polyethyleneimine , have been successfully used in vaccines against viruses, tumors, bacteria, and inflammatory and immunoregulatory disorders . However, the application of nanovaccines is often limited by their toxicity, bioincompatibility, instability, scale-up shortcomings, and low interbatch reproducibility. Biopolymerizable nanoparticles that rely entirely on biosynthesis are gradually emerging. The available techniques mainly utilize trimer tags, the C-terminal domain of T4 fibritin, three-helix bundles , self-assembling proteins (ferritin , lumazine synthase [57, 58], encapsulin ), and self-assembling peptides [60–62] to carry large proteins or peptides to form regular polymers. Vaccines based on these polymeric structures have induced excellent immunogenic effects in terms of different aspects [57, 63]. Therefore, it is highly important to continue to develop nanovaccine delivery systems with antigen polymerization.
The site of ACE2 binding to the RBD does not have a glycosylation modification, so it is theoretically possible to design RBD-based vaccines in prokaryotic systems . Here, we used the polymerization function of ClyA to achieve the regular polymerization of RBD in E. coli, which seemed to simulate an RBD polymeric structure conformation close to that of the natural virus, thus avoiding the incorrect folding of E. coli–expressed RBD. ClyA protein naturally polymerizes in E. coli OMVs [18, 23]. The ClyA protein is also widely used in genetic engineering for the translocation of heterologous proteins to modify OMVs [64, 65]. OMVs have many toxic proteins, so there is a safety issue with using OMVs as a vaccine vector. If this type of self-assembled ClyA nanopore can directly carry heterologous proteins and form nanometer-scale structures, the self-assembled ClyA nanopore (approximately 14 nm) would be an ideal vaccine platform because it can easily target lymph nodes and reduce toxicity to other organs (Figure 2A) [44, 66], stimulating stronger antigen presentation. More importantly, the polymerization of ClyA is particularly suitable for the presentation of antigens that need to be polymerized, especially the RBD of SARS-CoV-2. No technology utilizing ClyA nanopores as a vaccine vector has been developed before.
We developed ClyA-RBD, a protein formed by the fusion of ClyA, a porin that is polymerized in bacteria, with full-length RBD. The fusion protein ClyA-RBD is abundantly expressed in the cytoplasm of E. coli and can migrate to the periplasm and the outer membrane , where the oxidative environment favors its proper folding . ClyA undergoes a conformational rearrangement triggered by the lipid membrane and assembles into pores on the bacterial outer membrane [20, 21]. RBD forms a polymeric structure with ClyA, and ClyA-RBD nanopores can be obtained from the outer membrane of E. coli using surfactants . This is because lipid membranes and surfactants can trigger and maintain the conformational rearrangement of ClyA monomers [21, 30, 68]. We utilized this feature to successfully induce the self-reassembly of the ClyA-RBD monomers from membrane in surfactant.
ClyA is a toxic protein that can punch holes in red blood cells to dissolve red blood cells. Therefore, the use of ClyA as a vaccine delivery system also presents safety concerns. However, it is encouraging that ClyA that has formed a nanopore structure on OMV, lacks hemolytic activity and loses the function of further attacking host cells [69–71]. In addition, the toxicity of the ClyA monomer with high hemolytic activity can also be removed through gene mutation  or truncation of the C-terminus . The PP for vaccine delivery here was designed to address its safety issues. We allowed the ClyA monomer to form a complete nanopore on the bacterial membrane, thereby losing hemolytic activity. Afterwards, the bacterial membrane was removed with the help of detergents, thus ensuring that the PP loses hemolytic activity without destroying the nanostructure. In our preliminary safety experiments, we did not see obvious safety concerns, nor did we see obvious hemolytic toxicity in major organs (Figure 5).
The keys to the success of this nanopore technique are that (1) the ClyA and RBD components of the ClyA-RBD monomers separately undergo correct folding without forming inclusion bodies, and then the correctly folded oligomers are translocated to the bacterial outer membrane due to the specific secretion of ClyA to the outer membrane, thus ensuring that the correct structures of ClyA and RBD exist on the bacterial outer membrane. (2) The ClyA-RBD polymers on the bacterial outer membrane switch from membrane-triggered to surfactant-triggered[19, 30]. Therefore, stable RBD-PP can be obtained in vitro. (3) The production process of RBD-PP is subject to strict temperature and time constraints. For example, it is critical to control the process of expression in bacteria overnight at 30°C, slow surfactant replacement at 15°C and polymerize into RBD-PP at 37°C.
The RBD-PP vaccine delivery system developed here has the following advantages: (1) ClyA simulates the supporting effect of SARS-CoV-2 S2 on RBD in S1, and the fusion of ClyA and RBD does not interfere with either protein conformation (Figure 1B). Therefore, correct ClyA polymerization also realizes correct RBD polymerization (Figure 1F). (2) Compared with monomeric RBD, the ability of regularly polymerized RBD to bind hACE2 is significantly enhanced, which illustrates the importance of RBD-PP for the efficient polymerization and exposure of RBD (Figure 2I). (3) RBD-PP formed a stable nanostructure of approximately 14 nm (Figure 1F). This size is very conducive to the delivery of antigens to lymph nodes, thus greatly enhancing antigen uptake, processing and presentation (Figure 2). (4) The most important point is that RBD-PP may cause this nonglycosylated modified RBD to display a spatial conformation partially similar to the virus spike structure. (5) RBD-PP has the triple function of inducing humoral immunity, cellular immunity, and immune memory. Neutralizing antibodies play a major role in resisting SARS-CoV-2. It has been proven that neutralizing antibody titers above 100 can effectively resist SARS-CoV-2 infection[28, 74]. However, increasing evidence has also proven the important role of the T cell response and immune memory in the fight against COVID-19 [32, 75]. (6) Our special preparation process not only enables ClyA not only to form polymerized nanostructures but also o inactivates its hemolytic activity, thereby preventing the safety problems of ClyA as a vaccine carrier.
In short, the advantages of RBD-PP are due to its polymerization and display of RBD nanostructures, enabling RBD-PP to exhibit effective immunological activation capabilities, including lymph node targeting (Figure 2A), DC targeting (Figure 2C), induction of Tfh cells (Figure 2G) and GC B cells (Figure 2H). Therefore, the ability of RBD-PP to induce humoral immunity (Figure 3), cellular immunity and immune memory (Figure 4) was stronger than that of the RBD monomer. However, the most important point is that RBD-PP may cause this nonglycosylated modified RBD to adopt a spatial conformation similar to that of the viral spike structure, which is shown by the following evidence: (1) The binding capacity of RBD-PP and hACE2 increased (Figure 1I). (2) Four neutralizing antibodies recognized RBD-PP (Figure 1J). (3) The ratio of neutralizing potency to binding potency increased significantly (Figure 4I). In summary, we have developed a RBD-PP vaccine that achieves high RBD polymerization in E. coli, which will provide a design for an inexpensive vaccine against SARS-CoV-2 (Figure 6).
Bacterial PPs have not previously been utilized as vaccine vectors. We developed this PP-based vaccine vector platform and conducted a complete in vitro and in vivo evaluation. This also suggests that other proteins that can assemble into pores may be modified in the same way to form vaccine vector platforms for wider application. Because of the urgency of COVID-19 vaccine development, we tested the PP platform to make a vaccine against SARS-CoV-2, but different nanopores are likely to be developed in the future and applied as nanoplatforms in this and other fields, including viral vaccines, bacterial vaccines, tumor vaccines, drug delivery, and disease diagnosis.