Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an emerging virus that is highly pathogenic and has caused the recent worldwide pandemic officially named coronavirus disease (COVID-19). Currently, considerable efforts have been put into developing effective and safe drugs and vaccines against SARS-CoV-2. Vaccines, such as inactivated vaccines, nucleic acid-based vaccines, and vector vaccines, have already entered clinical trials. In this review, we provide an overview of the experimental and clinical data obtained from recent SARS-CoV-2 vaccines trials, and highlight certain potential safety issues that require consideration when developing vaccines. Furthermore, we summarize several strategies utilized in the development of vaccines against other infectious viruses, such as severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV), with the aim of aiding in the design of effective therapeutic approaches against SARS-CoV-2.
The coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has posed a serious threat to public health.1,2,3 SARS-CoV-2 belongs to the Betacoronavirus of the family Coronaviridae, and commonly induces respiratory symptoms, such as fever, unproductive cough, myalgia, and fatigue.4,5,6 To better understand the virus, numerous studies have been performed, and strategies have been established with the aim to prevent further spread of COVID-19, and to develop efficient and safe drugs and vaccines.7 For example, the structures of viral proteins, such as the spike protein (S protein), main protease (Mpro), and RNA-dependent RNA polymerase (RdRp), have been uncovered,8,9,10 providing information for the design of drugs against SARS-CoV-2. In addition, elucidating the immune responses induced by SARS-CoV-2 is accelerating the development of therapeutic approaches. In essence, diverse small molecule drugs and vaccines are being developed to treat COVID-19. According to the World Health Organization (WHO), as of September 17, 2020, 36 vaccine candidates were under clinical evaluation to treat COVID-19, and 146 candidate vaccines were in preclinical evaluation. Given that vaccines can be applied for prophylaxis and the treatment for SARS-CoV-2 infection, in this review, we introduce the recent progress of therapeutic vaccines candidates against SARS-CoV-2. Furthermore, we summarize the safety issues that researchers may be confronted with during the development of vaccines. We also describe some effective strategies to improve the vaccine safety and efficacy that were employed in the development of vaccines against other pathogenic agents, with the hope that this review will aid in the development of therapeutic methods against COVID-19.
Target antigen for SARS-CoV-2 vaccines
Coronaviruses (CoVs), including SARS-CoV, MERS-CoV, and SARS-CoV-2, are cytoplasmically replicating, positive-sense, single-stranded RNA viruses with four structural proteins (namely S protein, envelope (E) protein, membrane (M) protein, and nucleocapsid (N) protein).11 Generally, the S protein plays a crucial role in eliciting the immune response during disease progression.12
SARS-CoV-2 enters host cells via the same receptor, angiotensin-converting enzyme 2 (ACE2), as SARS-CoV, and the S protein is required for cell entry.13,14,15 The trimeric S protein contains two subunits, S1 and S2, which mediate receptor binding and membrane fusion, respectively. The S1 subunit contains a fragment called the receptor-binding domain (RBD) that is able to bind ACE2.16,17 Binding of the S protein to the ACE2 receptor triggers complex conformational changes, driving the S protein from a prefusion conformation to a postfusion conformation. The decoration of the postfusion conformation with N-linked glycans was suggested as a potential strategy for the virus to evade the host immune response.18 Previous studies reported that vaccines encoding SARS-CoV S protein elicited potent cellular and humoral immune responses in murine challenge models and in clinical trials.19,20,21 Similarly, the S gene is regarded as a key target for SARS-CoV-2 vaccines.22 The S protein of CoVs, especially the RBD, is able to induce neutralizing antibodies (NAbs) and T-cell immune responses.23,24,25,26 An animal study demonstrated that SARS-CoV-2 RBD-specific IgG accounted for half of the S protein-induced antibody responses.27 RBD-specific antibodies and T cells were also detected in the sera of discharged SARS-CoV-2-infected patients.28 Moreover, NAb titers were significantly correlated with the levels of anti-RBD IgG, and RBD-specific IgG titers were suggested as a surrogate of neutralization potency against SARS-CoV-2 infection.26,28 Furthermore, immunization with RBD was initially successful in eliciting NAbs in rodents without mediating antibody-dependent enhancemnt.29 Thus, RBD is a promising target for SARS-CoV-2 vaccines and previous knowledge from using RBD-based vaccines against SARS-CoV and MERS-CoV could inform the design of RBD-based SARS-CoV-2 vaccines.
Apart from the S protein, other proteins, such as the N protein, M protein, non-structural proteins (nsps), and accessory proteins, may have the potential to serve as antigens. Indeed, viral proteins and their interactions with host factors were associated with imbalanced host immune responses, such as low type I interferons (IFN-I) and IFN-III levels, and elevated pro-inflammatory cytokine levels (Fig. 1a).30,31 Recent studies found that nsp13 of SARS-CoV-2 targeted the IFN pathway by associating with TBK1, and nsp15 interfered with this pathway by associating with RNF41. The open reading frame 6 (ORF6) protein interacted with the mRNA export factor NUP98-Rae1. ORF9b indirectly interacted with the mitochondrial antiviral signaling (MAVS) protein via its interaction with translocase of outer membrane 70 (Tom70).32 Moreover, ORF8 was shown to significantly downregulate the major histocompatibility complex class I (MHC-I) expression in diverse cell types via lysosomal degradation, thereby disrupting antigen presentation and impairing the cytotoxic T lymphocytes (CTLs)-mediated killing of virus-infected cells.33 Previous reports demonstrated that the CoV N protein induced protective specific CTLs.34,35,36,37 Moreover, NAbs titers significantly correlated with the number of N protein-specific T cells, suggesting that the production of NAbs might be linked with the activation of antiviral T cells.28,38 Another study reported that antisera to M proteins exhibited high neutralizing titers toward SARS-CoV infection, indicative of the importance of M protein for developing an effective protein-based vaccine.39 Recently, Grifoni et al. noticed that cluster of differentiation 4 (CD4)+ T-cell responses were primarily directed against the S, M, and N proteins and partially against nsp3, nsp4, and ORF840 (Fig. 1b). Regarding CD8+ T-cell responses, the SARS-CoV-2 M and S proteins were strongly recognized, and significant reactivity was observed for other antigens, such as nsp6, ORF3a, and the N protein (Fig. 1b).40 The data suggests that beyond the S protein, the CD8+ T-cell response to SARS-CoV-2 elicited by an optimal vaccine may benefit from additional class I epitopes, such as those derived from the M, nsp6, ORF3a, and/or N protein. However, whether they can be used as the target antigen requires further investigation.
The development of SARS-CoV-2 vaccines
Inactivated vaccines and live-attenuated vaccines
Due to the urgent need to combat COVID-19, diverse SARS-CoV-2 vaccine types are currently under development, including inactivated vaccines, nucleic acid vaccines, adenovirus-based vector vaccines, and recombinant subunits vaccines (Fig. 2). Inactivated viruses are made non-infectious via physical or chemical approaches and are attractive because they present multiple viral proteins for immune recognition, have stable expression of conformation-dependent antigenic epitopes, and can be easily produced in large quantities.41 Purified inactivated viruses have been traditionally used for vaccine development and have been found to be effective in preventing viral diseases, such as influenza. The inactivated SARS-CoV-2 vaccine candidate, BBIBP-CorV, demonstrated potency and safety in animal models; thus, is expected to undergo further testing in clinical trials.42 Another study evaluating a purified inactivated SARS-CoV-2 virus vaccine candidate, PiCoVacc, showed the induction of NAbs against SARS-CoV-2 in mice, rats, and rhesus macaques with no notable cytokine changes or pathology observed in the macaques.27 The inactivated SARS-CoV-2 vaccine containing aluminum hydroxide developed by Sinovac has entered phase 3 clinical trials, with results from the phase 2 trial demonstrating that two doses of 6 μg/0.5 mL or 3 μg/0.5 mL of the vaccine were well-tolerated and immunogenic in healthy adults (Table 1).43 Phase 2 trial results of the inactivated SARS-CoV-2 vaccine, constructed by Wuhan Institute of Biological Products and Sinopharm, reported that the geometric mean titers (GMT) of NAbs were 121 and 247 at day 14 after 2 injections in participants receiving vaccine on days 0 and 14 and on days 0 and 21, respectively, displaying only transient and self-limiting adverse reactions.44
Table 1 The development of vaccine candidates in phase 3 clinical stageFull size table
Live-attenuated vaccines have demonstrated success in treating infections such as smallpox and poliomyelitis.45 Three SARS-CoV-2 live-attenuated vaccines that utilize a weakened virus as the antigen are under preclinical evaluation. However, such vaccines may revert to virulence in some cases. Although the virus itself can be used to develop vaccines, concerns have been raised that the inclusion of epitopes that do not induce NAbs or confer protection may skew the immune response, thereby requiring further investigation.
Nucleic acid vaccines
Nucleic acid vaccines, such as mRNA vaccines and DNA vaccines, are other popular vaccine forms. These vaccines are delivered into human cells, where they will then be transcribed into viral proteins. Among the CoV proteins, S protein has been the most common candidate. mRNA vaccines represent a promising alternative compared to conventional vaccines due to their high potency, ability for rapid development, and cost-efficient production.46,47 However, the physiochemical properties of mRNA may influence its cellular delivery and organ distribution, and the safety and efficacy of mRNA vaccine use in humans remain unknown. Phase 1/2 studies investigating RNA vaccines (BNT162b1) targeting the RBD of the S protein, developed by Pfizer and BioNTech, reported that the vaccine caused mild to moderate local and systematic symptoms in most vaccinators, with the vaccine eliciting higher neutralizing titers after the second dose compared to the COVID-19 convalescent sera panel (Table 1).48 Phase 1 trial assessing mRNA-1273 that encoded the stabilized prefusion SARS-CoV-2 S protein demonstrated that the two-dose vaccine series did not cause severe adverse events and could elicit neutralization and Th1-biased CD4+ T-cell responses (Table 1).49 The lipid nanoparticles (LNP)-encapsulated mRNA vaccine encoding SARS-CoV-2 RBD called ARCoV conferred potent protection against SARS-CoV-2 in mice and non-human primates after two immunization doses. Moreover, it could be stored at room temperature, which would be more convenient for transportation and storage.50
DNA vaccines also have great therapeutic potential due to their ability to enhance T-cell induction and antibody production, the excellent biocompatibility of plasmid DNA, low-cost manufacturing, and their long shelf life.51 However, their disadvantage is that the DNA molecules must cross the nuclear membrane to be transcribed, and they generally have low immunogenicity. A study of various DNA vaccine candidates encoding different forms of the SARS-CoV-2 S protein discovered that vaccinated rhesus macaques were able to develop humoral and cellular immune responses and that vaccine-induced NAb titers were associated with protective efficacy.52 Notably, DNA vaccines induced type I helper T cells (Th1) instead of type II helper T cells (Th2) responses with no observed enhancement of clinical disease in rhesus macaques. However, a report concerning a MERS-CoV DNA vaccine observed NAbs in just half of all subjects and titers noticeably waned during the course of the study follow-up.53 Future studies should explore whether DNA vaccines are effective in inducing long-term NAbs and whether non-neutralizing antibody responses can confer protection or cause more severe disease.
Vector vaccines are generally constructed from a carrier virus, such as an adeno or pox virus, and are engineered to carry a relevant gene from the virus, usually the S gene for CoVs. The key advantage of vector vaccines is that the immunogen is expressed in the context of a heterologous viral infection, which induces the innate immune responses required for the adaptive immune responses.54 Nevertheless, this strategy may induce prior immunity to the vector and are limited to presenting only a small number of CoV antigens to the host immune system. Clinical trials regarding an adenovirus type 5 (Ad5) vector vaccine carrying recombinant SARS-CoV-2, developed by CanSino Biological Inc. and Beijing Institute of Biotechnology, revealed that the vaccine at a dose of 5 × 1010 viral particles per mL was safer than the vaccine at 1 × 1011 viral particles, and elicited comparable immune response to it55 (Table 1). However, high pre-existing Ad5 immunity and increasing age reduced NAbs response and the pre-existing immunity might also influence T-cell immune response post-vaccination.56 Thus, further investigation is required to address these problems influencing vaccine efficacy. Phase 1/2 studies of a heterologous COVID-19 vaccine comprising a recombinant adenovirus type 26 (rAd26) vector and a recombinant adenovirus type 5 (rAd5) vector, both carrying the S gene of SARS-CoV-2, demonstrated that the pre-existing immune response to the vectors rAd26 and rAd5 did not influence the titre of RBD-specific antibodies (Table 1). Therefore, heterologous vaccination may be a good option to antagonize the negative impacts of immune response to vaccine vectors.57 Moreover, a phase 3 study was performed to determine the efficacy, safety, and immunogenicity of a chimpanzee adeno (ChAd)-vectored vaccine platform encoding a codon-optimized full-length SARS-CoV-2 S protein (ChAdOx1 nCoV-19). In a preclinical trial, SARS-CoV-2 genomic RNA was detected in nasal swabs from all rhesus macaques, with no discrepancy in viral load between nasal swabs on any day between ChAdOx1 nCoV-19-vaccinated and control animals, despite the lack of pneumonia and absence of immune-enhanced disease following viral challenge in vaccinated animals.58 However, in the phase 1/2 trial, ChAdOx1 nCoV-19 was shown to be safe, tolerated, and immunogenic. Moreover, local and systemic reactions, including pain, fever, and muscle ache, could be reduced by taking paracetamol59 (Table 1). Notably, safety is a crucial issue in vaccine development; therefore, greater emphasis on improving safety should be placed when testing the SARS-CoV-2 vaccines. Ad26COVS1 designed by Janssen Pharmaceutical Companies also entered the phase 3 clinical stage and its preclinical study showed that a single immunization with an Ad26 vector encoding a prefusion stabilized S protein triggered potent NAb responses and well protected the vaccinated rhesus macaques60 (Table 1).
Subunit vaccines and virus-like particles vaccines
Subunit vaccines in which viral proteins are injected into the host have the potential to exhibit efficacy in protecting animals or human from viral infection. However, given that only a few viral components are included which do not display the full antigenic complexity of the virus, their protective efficacy may be limited and, in some cases, they may cause unbalanced immune responses.61 Yang et al. constructed a subunit vaccine composed of residues 319–545 of the SARS-CoV-2 RBD and produced it through the baculovirus expression system. The preclinical study reported that the vaccine could protect the non-human primates from SARS-CoV-2 infection with little toxicity62 (Table 2). Virus-like particles (VLPs) constitute another type of protein-based vaccines that are composed of proteins from the viral capsid.63 VLPs stimulate high immune responses due to their repetitive structures and are safer than several other vaccine platforms because they lack genetic material. The construction of VLPs similar to the authentic virus is a significant step in the development of an effective vaccine against infection. Several teams are currently working on engineering protein-based vaccines; however, the clinical results have not been published to date. Despite the fact that vaccine development is a lengthy and expensive process that typically involves multiple candidates and requires a lot of time to produce a licensed vaccine, it is vital to continue developing vaccines for the prevention and treatment of COVID-19.Table 2 The development of vaccine candidates in phase 1 or phase 2 clinical stageFull size table
Neutralizing antibodies against SARS-CoV-2
NAbs play a critical role in controlling viral infection.64 The most commonly used antibody formats include monoclonal antibodies (mAbs), single-domain antibodies, single-chain variable fragments (scFvs), and functional antigen-binding fragments (Fabs) (Fig. 3a). Neutralizing monoclonal Abs can be isolated from recovered people previously infected with virus (Fig. 4a) or immunized transgenic animal models (Fig. 4b). NAbs, particularly those targeting the RBD of SARS-CoV-2, may serve as a promising therapeutic approach to viral infection65,66 (Table 3). Recently, three non-competing epitopes for the RBD (namely RBD-A, RBD-B, and RBD-C) have been identified, with RBD-A considered as the preferred target. RBD-A-directed NAb CC12.1 was shown to potently neutralize the pseudovirus.67 A cohort of NAbs were also shown to be able to bind the RBD and perturb the RBD-ACE2 interaction, such as BD-368–2, B38, H4, B5, CB6, and CV30.68,69,70,71,72,73,74 However, 47D11 and H2 did not compromise the spike-receptor interaction, although it was capable of binding to the epitope of the RBD of SARS-CoV-2.69,75 A study found that ACE2 competitor antibodies neutralized the viral infection by blocking ACE2 binding and inducing S1 dissociation, as well as demonstrating a weak association between antibodies potency and their binding affinity.66 However, a separate report revealed the correlation between serum RBD binding and virus neutralization.67 Additional efforts are required to characterize the factors that influence the neutralizing activities of NAbs. In light of the close relationship between SARS-CoV and SARS-CoV-2, scientists have attempted to identify SARS-CoV NAbs that cross-reacted with SARS-CoV-2. Antibodies derived from previously SARS-CoV-infected patients, such as S309, ADI-55689, and ADI-56046, were shown to cross-neutralize SARS-CoV-2.66,76 S309, which targeted a conserved glycan-containing epitope within the S protein, also displayed fragment crystallizable (Fc)-dependent effector mechanisms, such as antibody-dependent cell cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP).76 Moreover, a few NAbs targeted non-RBD regions (Table 3). For instance, CV1 and its clonal variant CV35 bound to an epitope distinct from the RBD and both exhibited lesser potency than CV30 that targeted the RBD region.74 Further efforts should focus on the identification of potent NAbs from recovered patients. Moreover, structural analysis using the Reverse Vaccinology 2.0 approach is expected to uncover the exact epitopes of NAbs in order to promote immunogen design and guide vaccine strategies (Fig. 3b).77
Table 3 Potential neutralizing antibodies targeting SARS-CoV-2Full size table
Apart from conventional antibodies, camelids generate heavy chain antibodies (HCAbs) composed of only two heavy chains with a single variable domain (VHH or nanobody) and two constant regions per chain. Nanobodies can be constructed based on sequences of the camelid immunized with viral proteins (Fig. 4c) or on human sequences. Compared to traditional antibodies, nanobodies have several unique characteristics due to their small size, including access to more epitopes, low production expense, and the possibility for large-scale production in prokaryotic expression systems.78 Moreover, nanobodies can be administered via an inhaler directly to the site of infection, which is particularly beneficial for the treatment of respiratory diseases.79 The disadvantages of utilizing nanobodies as therapeutics could be that they may show immunogenicity due to their camel derivation and lack Fc-mediated effector functions. However, humanization and the development of fully human antibodies could improve the nanobodies.80 Recently, the SARS-CoV RBD-directed single-domain antibody VHH-72 displayed cross-reactivity with the SARS-CoV-2 RBD and was capable of disrupting RBD-receptor-binding dynamics. Furthermore, a bivalent VHH-72-Fc construct exhibited neutralizing activity against SARS-CoV-2 S pseudoviruses.81 Analysis on fully human single-domain antibodies identified from an antibody library using SARS-CoV-2 S1 as panning antigens revealed that the antibodies n3088 and n3130 were able to neutralize SARS-CoV-2 by targeting a cryptic epitope situated in the spike trimeric interface, even though they are not able to compete with ACE2 for SARS-CoV-2 RBD binding.82 These two antibodies may serve as promising alternatives that may be less immunogenic than camelid or humanized nanobodies, given that they are entirely derived from human sequences.
In addition to the antibodies mentioned above, scFvs and Fabs hold promise for treating COVID-19, and have already demonstrated benefits in the context of fighting against SARS-CoV and MERS-CoV. The scFv 80R was shown to compete with ACE2 for interaction with the S1 subunit, and efficiently neutralized SARS-CoV in vitro.83 Recently, the RBD-specific scFv-human Fc 5C2 was found to effectively neutralize the SARS-CoV-2 S protein and inhibit ACE2 from binding to the S protein.84 Moreover, previous studies revealed that human mAbs or Fabs, such as MERS-27 and m336, could recognize epitopes on the RBD of MERS-CoV that overlapped with the dipeptidyl peptidase 4 (DPP4)-binding site and neutralized pseudotyped and/or live MERS-CoVs in vitro.85,86 The scFv and Fabs have short generation time, high antigen affinity, and structural stability.87 However, whether scFv and Fab are effective against SARS-CoV-2 requires further investigation.
Since there is a lack of effective therapies for treating a cohort of SARS-CoV-2-infected patients, further development of NAbs specifically targeted against SARS-CoV-2 may be worthwhile, as well as the continued investigation of NAbs against SARS-CoV and MERS-CoV that can cross-react with SARS-CoV-2. A SARS-CoV-2 variant carrying the Spike D614G mutation, which has greater infectivity, has become the dominant form in many geographic regions.88 It is noteworthy that CoVs have high mutation rates and NAbs have several limitations. As such, the use of NAbs that can synergistically recognize different epitopes warrants further research. The combination of REGN10987 and REGN10933 NAbs, which bound to two non-overlapping epitopes of the RBD, did not generate escape mutants.89,90 Antibody 553–15 identified by Wan et al. could substantially improve the neutralizing capacity of other NAbs they discovered.91 Nevertheless, the cocktail therapy approach is costly and may not induce long-term immune responses. Thus, continued efforts are required to improve the efficacy of cocktail therapy, and to assess whether it is practical and safe for clinical use.
Safety concerns regarding vaccine development
The most important criterion of vaccines is safety. Previous experience from the development of SARS-CoV and MERS-CoV vaccines has raised concerns of pulmonary immunopathology correlating with Th2 responses65 (Fig. 5b). Th2 is a subgroup of T cells that can secrete Th2-type cytokines, such as interleukin 4 (IL-4), IL-5, IL-10, and IL-13, and aberrant levels of Th2 cytokines can cause immune reactions that lead to eosinophil infiltrations. In murine models, four different SARS-CoV vaccines led to the occurrence of Th2-type immunopathology with high eosinophil infiltration, which served as a marker for Th2-type hypersensitivity.92 This was also observed in mice vaccinated with inactivated MERS-CoV vaccines which had eosinophil infiltrations, with the levels of IL-5 and IL-13 higher than those before vaccination.93 Moreover, it is proposed that the immunopathologic reaction following vaccination may be partially attributed to the presence of the N protein in the vaccine, but this requires further validation.94,95 Recent studies on cytokine changes in patients infected with SARS-CoV-2 also observed increased secretion of Th2 cytokines, which might contribute to the lung immunopathology.96,97,98 Thus, controlling the T-cell response must be considered when designing vaccines against SARS-CoV-2.
While the humoral immune response induced by vaccines may represent a potent approach of conferring protection against CoV infection, an abnormal antibody response may also result in physical deterioration of patients (Fig. 5b). In SARS-CoV-infected macaque models, vaccine-induced S-specific IgG resulted in severe acute lung injury (ALI) because IgG disturbed the inflammation-resolving response of macrophages and the blockade of Fc gamma receptor (FcγR) reduced such influence.99 Moreover, deceased patients displayed higher titers of NAbs and faster NAb responses which dropped more quickly than in recovered patients during the acute infection, potentially triggering a systematic breakdown of the immune system and exerting the immunopathologic effects on the lung and spleen.99,100 Consistently, patients severely infected with SARS-CoV-2 frequently exhibited more robust IgG responses and increased antibodies titers, which linked with the worst clinical condition and suggested antibody-dependent enhancement (ADE) of SARS-CoV-2 infection.101,102 Whether SARS-CoV-2 vaccines will cause abnormal antibody responses is currently unknown and additional research is required to address the potential lung damage caused by SARS-CoV-2 vaccine candidates.
Age is known to influence vaccine immunity. Vaccinated aged animals that were challenging to immunize also displayed eosinophilic immune pathology in the lungs. Worse still, neutralizing titers were significantly reduced in aged vaccinated groups compared to young groups.95,103 In essence, elderly populations with underlying diseases including diabetes, hypertension, and cardiovascular disease are at high risk for infection by SARS-CoV-2.52,104 Given the severity of disease in elderly people, aged animal models are essential for the preclinical validation of vaccines.
Download pdf: https://www.nature.com/articles/s41392-020-00352-y.pdf