key: cord-0861602-goeimpqd
authors: Lee, Mi-Hyun; Kim, Bum-Joon
title: COVID-19 vaccine development based on recombinant viral and bacterial vector systems: combinatorial effect of adaptive and trained immunity
date: 2022-02-14
journal: J Microbiol
DOI: 10.1007/s12275-022-1621-2
sha: 02accd8087d92f205dc8cc85f838e0ec3d368b48
doc_id: 861602
cord_uid: goeimpqd

Severe acute respiratory syndrome coronavirus 2 virus (SARS-CoV-2) infection, which causes coronavirus disease 2019 (COVID-19), has led to many cases and deaths worldwide. Therefore, a number of vaccine candidates have been developed to control the COVID-19 pandemic. Of these, to date, 21 vaccines have received emergency approval for human use in at least one country. However, the recent global emergence of SARS-CoV-2 variants has compromised the efficacy of the currently available vaccines. To protect against these variants, the use of vaccines that modulate T cell-mediated immune responses or innate immune cell memory function, termed trained immunity, is needed. The major advantage of a vaccine that uses bacteria or viral systems for the delivery of COVID-19 antigens is the ability to induce both T cell-mediated and humoral immune responses. In addition, such vaccine systems can also exert off-target effects via the vector itself, mediated partly through trained immunity; compared to other vaccine platforms, suggesting that this approach can provide better protection against even vaccine escape variants. This review presents the current status of the development of COVID-19 vaccines based on recombinant viral and bacterial delivery systems. We also discuss the current status of the use of licensed live vaccines for other infections, including BCG, oral polio and MMR vaccines, to prevent COVID-19 infections.

The COVID-19 outbreak began at the end of 2019 due to the sudden appearance of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). As of November 2021, there have been over 250 million cases with over 5 million deaths globally (Worldometer, 2021) . To control this pandemic caused by COVID-19, effective vaccines for preventing SARS-CoV-2 infection are urgently needed.

Similar to SARS-CoV, SARS-CoV-2 consists of four major structural proteins: the spike glycoprotein, nucleocapsid, membrane and envelope proteins. Of these, the spike protein contains S1 including the receptor-binding domain (RBD) and S2 subunits, which mediate the entry of the virus into the host cell by binding to the human angiotensin-converting enzyme 2 (hACE2) receptor (Zhu et al., 2020a) . Therefore, the S protein plays a pivotal role in eliciting immune responses against SARS-CoV-2 and is a major target for neutralizing antibodies in humans (Huang et al., 2020; Yang and Du, 2021) . Moreover, the amino acid sequences of the spike protein were observed to contain a number of CD4 + and CD8 + T-cell epitopes, highlighting their potential roles in eliciting adaptive immune responses (Grifoni et al., 2020 (Grifoni et al., , 2021 . Therefore, most COVID-19 vaccine candidates have been designed to provide the S protein or RBD as the target antigen, which is responsible for inducing immune responses. To date, a number of COVID-19 vaccine candidates have been developed on the basis of different platforms, such as non-replicating or replicating viral vectors, protein subunits, conventional whole inactivated or live-attenuated virus, mRNA and DNA (Nagy and Alhatlani, 2021) . Of these, a total of 21 vaccines are currently authorized for human use worldwide (Tracker, 2021) . Although no significant side effects have been reported for currently used vaccines, the emergence of variants of SARS-CoV-2, including lineages B.1.1.7 and B.1.617.2, and the spread of these variants worldwide pose a serious threat to public health because they compromise the effectiveness of currently developed vaccines (Rambaut et al., 2020; Wibmer et al., 2021; Zhou et al., 2021) . Therefore, to overcome the risk posed by variants of SARS-CoV-2 that are less susceptible to protective antibodies, vaccine strategies har-nessing T cell-mediated immune responses or innate cellbased heterologous effects that less affected by these variants should be developed (Sauer and Harris, 2020) . Compared to other platforms, vaccines that use viral or bacterial delivery systems have a distinct benefit in that they can induce off-target effects to defend against unrelated pathogens via an innate immune memory system termed trained immunity. In epidemiological studies, some live viral or bacterial vaccines have been reported to induce heterologous vaccine effects via trained innate immune cells, resulting in a reduction in all-cause mortality from infectious agents (Agrawal, 2019; Nascimento et al., 2020; Marín-Hernández et al., 2021) . The enhanced nonspecific immune response of trained immunity can further enhance both specific humoral and cell-mediated immune responses to defend against SARS-CoV-2, providing better protection against even vaccine escape variants (Covián et al., 2021) .

In this review article, we will first review the ability of licensed live vaccines for other infections, including the BCG, oral polio and measles-mumps-rubella (MMR) vaccines, to prevent COVID-19 infections via heterologous vaccine effects. Second, we will discuss the current COVID-19 vaccine platforms based on viral and bacterial delivery systems and discuss the advantages and disadvantages of the different systems used for vaccine delivery. 

A number of lines of evidence have indicated that childhood vaccination with some live-attenuated vaccines, such as the tuberculosis vaccine, bacilli Calmette-Guerin (BCG), smallpox and polio vaccines, can induce beneficial, nonspecific, heterologous vaccine effects against unrelated pathogens by eliciting innate cell-mediated immune responses (Nascimento et al., 2020; Pasco and Anguita, 2020) . These innate cell-based heterologous effects are referred to as trained immunity. For example, BCG vaccination of neonates has been reported to reduce neonatal and infant mortality independent of its effect on tuberculosis (Thysen et al., 2020) . In addition, vaccination against smallpox can lead to partial protection against measles, pertussis and scarlet fever, suggesting a long-lasting nonspecific protective effect associated with the vaccine, independent of adaptive immune responses based on T and B lymphocytes (Sánchez-Ramón et al., 2018) . Innate immune cells, such as monocytes and natural killer (NK) cells that are stimulated by inducers imprint memory ability to prepare for a secondary infection through epigenetic modification, metabolic changes and production of pro-inflammatory cytokines, such as IL-6 TNF-α and IL-1β (Netea et al., , 2020a . Since trained immunity provides longterm immunological memory against unrelated pathogens, specifically about three months duration in a mouse, it can be used for the development of effective vaccines to promote host resistance against a broad range of pathogens by training innate immune cells (Sánchez-Ramón et al., 2018; Gyssens and Netea, 2019; Netea et al., 2020b) . Accordingly, various clinical trials have been performed to test the protective effects of licensed live vaccines that are already being used for other infections against COVID-19.

BCG is one of the most widely used live vaccines against Mycobacterium tuberculosis infections globally. It was generated as an attenuated strain via serial subcultures of Mycobacterium bovis to prevent dissemination in the body. BCG has been shown to provide protection against not only tuberculosis but also different infections unrelated to tuberculosis, including leprosy, Buruli ulcer, malaria, respiratory viral infections and yellow fever virus infections (Basak et al., 2021) . Given the potential of BCG to prevent infections caused by unrelated pathogens, as described above, the use of BCG for the prevention of SARS-CoV-2 infection has attracted increasing attention. Several epidemiologic studies have demonstrated that nations with active BCG vaccination programs, including Japan and South Korea have lower incidence rates and reduced mortality associated with COVID-19 (Kumar et al., 2020; Jakhmola et al., 2021) . Consistently, the data from the WHO immunization monitoring program have also shown an inverse relationship between BCG vaccination and COVID-19 incidence and mortality, suggesting the potential of BCG vaccination to control the COVID-19 pandemic via heterologous vaccine effects (Marín-Hernández et al., 2021) . Currently, a total of 21 randomized controlled clinical trials are in progress to evaluate the protective effects of BCG vaccination against COVID-19 at various ages in different countries, including the Netherlands, Brazil, Denmark, Australia, the US, Germany and France (Table 1 ). In addition, there are three phase 3 clinical trials for a recombinant BCG termed VPM1002 to test its protective effects against COVID-19 (NCT04439045, NCT04387409, and NCT-04435379); this recombinant BCG was originally designed to further potentiate the vaccine effect of BCG against tuberculosis infections (Nieuwenhuizen et al., 2017) . However, there is currently no clear evidence demonstrating that vaccination with BCG or the rBCG VPM1002 can protect against COVID-19 infections. In Taiwan, whether neonatal BCG vaccination could alleviate severe COVID-19 symptoms in the 4-24 year age group was investigated. There was no significant difference in COVID-19 symptoms between the groups that received and did not receive BCG vaccination (Su et al., 2021) . Nonetheless, the potential of BCG as a COVID-19 vaccine has not been completely eliminated. More in-depth studies with more cases with reliable vaccination records grouped by different types of BCG strains and different periods after vaccination might be needed.

Some live-attenuated viruses could be strong inducers of trained immunity. Respiratory adenovirus infection has been reported to induce trained immunity in the lungs by generating long-lasting memory alveolar macrophages, resulting in protective effects against bacterial infection (Yao et al., 2018) . This suggests the potential of adenovirus to be effectively used as a new vaccine strategy for protecting against respiratory disease. The MMR vaccine has been used globally since 2001 and can provide effective protection against measles, mumps and rubella. It has been hypothesized that the MMR vaccine could confer cross-protection against or reduce the severity of COVID-19 infection (Anbarasu et al., 2020) . Consistently, a recent study showed that there is a significant inverse correlation between mumps IgG titers and COVID-19 severity in individuals who had received the MMR vaccine in childhood (Gold et al., 2020) . A negative case-control study using a recent measles outbreak with MMR vaccination was conducted among healthcare workers in Sweden to investigate the potential protective effect of the MMR vaccine against SARS-CoV-2. The results indicated that while no substantial protective effect of the MMR vaccine was observed in the whole study population, significant effectiveness in preventing symptomatic disease was seen in men, suggesting that there may be a protective effect of the MMR vaccine against SARS-CoV-2 in males but not females (Lundberg et al., 2021) . In addition, the MMR vaccine has been reported to exert protective effects against COVID-19 in adults in a retrospective cohort study in Turkey (Yengil et al., 2021) . In line with these studies, two clinical trials (NCT04333732 and NCT0464623) are currently underway to evaluate the efficacy of the vaccine in preventing COVID-19 infection based on MMR-induced trained immunity (Table 2) . Oral polio vaccination (OPV) has also been reported to exert beneficial nonspecific effects, particularly against respiratory infections. Of note, it has recently been reported that OPV can lead to a 62% reduction in deaths caused by respiratory infections during the post-neonatal period (1-35 months) in Bangladesh (Andersen et al., 2018; Nielsen et al., 2021a Nielsen et al., , 2021b . Hence, there are two phase 4 clinical trials (NCT-04445428 and NCT04639375) to evaluate whether OPV can ameliorate COVID-19 severity (Table 2) . Furthermore, a retrospective study provided additional evidence for COVID-19 protection via cross-protective humoral immunity induced by OPV. Sera from a series of vaccinated people also inhibited SARS-CoV-2 infection in vitro via the production of cross-protective antibodies that were induced by OPV and capable of binding the RNA-dependent RNA polymerase (Comunale et al., 2021) . There is a clinical trial to investigate the protective effects of the Shingrix vaccine against COVID-19 (NCT-04523246); this vaccine is composed of recombinant varicella zoster virus and the adjuvant AS01B (Table 2) . Its safety has been already proved, as the herpes zoster vaccine was approved by the U.S. FDA in 2017 (FDA, 2021) and the clinical trial focuses on the induction of protection against SARS-CoV-2 by training the innate immune system.

Bactofection, using bacterial vectors to deliver foreign genes into host cells, is a promising vaccine platform that allows the introduction of antigens into specific target cells, including antigen-presenting cells (Motin and Torres, 2009; Chamekh, 2015) . There are several advantages of using recombinant bacterial vector systems for vaccine development. First, most bacterial vectors can easily incorporate large target sequences via plasmid or phage systems, and their generation requires relatively little labor and inexpensive production procedures (Lin et al., 2015) . Second, live bacterial vectors themselves can act as vaccine adjuvants via innate cell activation, resulting in the induction of both cell-mediated immune responses and humoral immune responses against the delivered target antigens, unlike protein-based subunit vaccines, which can induce mainly humoral immune responses (Detmer and Glenting, 2006; Silva et al., 2014) . Third, bacterial vectors can lead to further enhanced vaccine effects by inducing trained immunity, as described above (Goodridge et al., 2016; Covián et al., 2019) . Despite the overt advantages of this approach in vaccine development, there is a safety concern that these bacteria might become activated and cause disease in the body, which makes it difficult to recruit participants for clinical trials (WHO, 2021b). Herein, we have introduced the current status of COVID-19 vaccines based on recombinant bacteria, focusing on the published literature (Table 5) .

Francisella tularensis is a Gram-negative aerobic bacterium that can infect both invertebrates and mammals, causing sepsis, fever, pneumonia and possibly death in humans (Feldman et al., 2001; Steiner et al., 2014) . Although the F. tularensis live vaccine strain has been reported to provide some protection against tularemia, it shows high toxicity and reduced protective effects against aerosolized F. tularensis infections (Jia et al., 2009) . Hence, its virulence has been attenuated for application as a reliable vaccine platform mainly through deletion of the capsule synthesis gene (capB) (Jia et al., 2010) . In a preclinical study, plasmids for overexpressing the pathogenicity island proteins of F. tularensis were modified to develop the COVID-19 vaccine by incorporating different regions of SARS-CoV-2. Each recombinant F. tularensis strain was administered intradermally or intranasally with a prime-boost vaccination regimen. Consequently, coexpression of the SARS-CoV-2 membrane and nucleocapsid protein ameliorated the severity of lung pathology by SARS-CoV-2 infection in a golden Syrian hamster model by in-ducing anti-nucleocapsid antibody production (Jia et al., 2021) (Table 5) .

There are more than 150 species within the genus Mycobacterium, including strict human pathogens such as M. tuberculosis and M. leprae, and nontuberculous mycobacteria (NTM) with lower pathogenic potential, most of which exist in the environment (Pereira et al., 2020) . Mycobacteria have a thick lipid-rich cell wall, and the cell wall components can act as strong vaccine adjuvants to elicit innate cell activation. Mycobacterium paragordonae (Mpg) is a slow-growing NTM that exists in the environment and rarely causes diseases in humans and animals. Mpg is temperature sensitive and shows an optimal growth rate at 30°C but cannot grow at 37°C, indicating its potential as a safer vaccine for tuberculosis infection than BCG due to its inability to survive in the human body (Kim et al., 2019) . Indeed, it has been reported that compared to BCG, Mpg showed higher safety in both in vivo and in vitro studies and elicited a stronger protective effect against both M. tuberculosis and M. abscessus infections (Kim et al., 2017) . Moreover, recombinant Mpg expressing HIV-1 p24 (rMpg-p24) can induce enhanced p24 specific immune responses in vaccinated mice as evidenced by increased p24specific T lymphocyte proliferation, gamma interferon induction, antibody production and cytotoxic T lymphocyte (CTL) responses (Kim et al., 2019) demonstrating the potential of recombinant Mpg (rMpg) as a recombinant bacteria-based vaccine platform. Hence, in a recent study, SARS-CoV-2 receptor-binding domain (RBD)-expressing rMpg (rMpg-RBD-7) was generated to elicit RBD-specific immune responses in a mouse model with single-or two-dose vaccination regimens. Moreover, rMpg-RBD-7 led to enhanced cell-mediated immune responses as well as humoral immune responses, supporting its feasibility as a COVID-19 vaccine candidate (Kim et al., 2021) (Table 5) .

Salmonella is one of the most common causes of food poisoning and can infect various types of cells, including epithelial cells, macrophages and dendritic cells. Salmonella typhimurium is a Gram-negative bacterium that is usually found in the intestinal lumen and has been intensively investigated as a vaccine delivery system (Chin'ombe, 2013; Roland and Brenneman, 2013; Clark-Curtiss and Curtiss, 2018) . Deletion of the virulence transcriptional regulatory protein phoP enabled the generation of an attenuated S. typhimurium vaccine platform (Groisman et al., 1989; Methner et al., 2004) . In a recent study, oral administration of recombinant attenuated S. typhimurium, which contains the fulllength spike gene of SARS-CoV-2, exerted a protective effect against SARS-CoV-2 infection, mainly by inducing SARS-CoV-2-specific humoral immunity, in a rat model, suggesting the feasibility of a COVID-19 oral vaccine (Zhu et al., 2020b) (Table 5) .

Since SARS-CoV-2 is a respiratory disease that is transmitted through the upper respiratory mucosa, mucosal immunity could be the first line of defense against the virus (Gallo et al., 2021) . Probiotics could be a potent delivery system for the COVID-19 vaccine because they can interact with antigenpresenting cells to induce robust mucosal innate immune responses in the gut and respiratory tract (Moradi-Kalbolandi et al., 2021) . Lactiplantibacillus plantarum has been proposed as a potent COVID-19 vaccine. Previously, L. plantarum has been studied as a modulator of mucosal antiviral immunity in the context of oral vaccines to prevent influenza and Newcastle disease (Yang et al., 2017; Ho et al., 2020) . Although further in vivo studies are needed, a recent in vitro study indicated that recombinant L. plantarum with the optimized spike gene of SARS-CoV-2 may induce mucosal immune responses against SARS-CoV-2 (Wang et al., 2020) (Table 5) .

Viral vectors have been considered reliable vehicles for gene delivery into host cells. Since the initial attempt in the 1970s, the application of this approach has been widely extended, mainly for vaccine development and gene therapy (Ura et al., 2014; Bull et al., 2019) . Although viral vectors originate from pathogenic viruses, they have typically been engineered to be attenuated by deleting genes that are necessary for replication or pathogenicity (Bull et al., 2019) . Since viral vectorbased vaccines can generate endogenous antigens in a broad range of host cells, they can effectively induce both humoral and cell-mediated immune responses (Vrba et al., 2020) . In Table 3 addition, large-scale manufacturing is allowed, and they do not require freezing for transport or storage (Li et al., 2021b) . However, pre-existing immunity to the viral vector can impair the use of the vector as a vaccine delivery platform (Shirley et al., 2020) . Herein, we discuss two categories of viral vectorbased vaccines: non-replicating and replicating viral vector vaccines. Based on published research articles, 9 types of viral vectors have been developed for the COVID-19 vaccine, including 4 non-replicating viral vectors (Tables 2 and 3) , namely, the adenovirus vector (AdV), modified vaccinia virus Ankara (MVA) vector, parainfluenza virus vector (PIV), and lentivirus (LV), and 5 replication competent viral vectors (Table  4) , namely, the single-cycle adenovirus (SC-Ad), the vesicular stomatitis virus (VSV) vector, the measles virus vector (MeV), influenza virus (IV), and Newcastle disease virus (NDV) (Bezbaruah et al., 2021) .

Replication-deficient viral vector vaccines had not been approved before the COVID-19 pandemic. However, a total of 5 types of AdV-based vaccines, AZD1222 (Oxford/Astra-Zeneca), JNJ-78436735 (Johnson & Johnson), Ad5-nCoV (CanSino Biologics of Chinese), Sputnik V and Sputnik Light (Gamaleya Research Institute of Russia), currently have emergency approval and are in clinical use in one or more nations (Abdulla et al., 2021; Francis et al., 2021) (Table 6 ). Non-replicating viral vectors are typically generated by genetic deletion of replication ability. Through genome engineering, a larger space is created in the genome as many genes are removed, and longer inserts can be incorporated (Choi and Chang, 2013) . Although high doses can be administered to elicit sufficient immune responses due to the lack of replication capacity, this is considered a safer vaccine strategy than replicating viral vector vaccines (Robert-Guroff, 2007) .

For the generation of non-replicating viral vector vaccines, adenovirus (AdV) vectors have been used most frequently. The AdV genome is 26-45 kb of linear double-stranded DNA, and the virus can be classified into different groups and serotypes. AdV vectors can be modified to eliminate replication ability by deleting the E1 genes that are necessary for replication. In addition, E3 region genes can also be deleted to further create space for larger insert genes (Ura et al., 2014) . However, although AdV vectors can provide an efficient gene delivery system in host cells, they are not appropriate for repeated vaccination, especially in the case of human AdV. Since more than 80% of people have been exposed to human AdV, they may already possess preexisting neutralizing antibodies that bind to injected AdV vectors, interfering with entry into target cells. To avoid preexisting neutralizing antibodies, rare types of human AdV vectors or nonhuman vectors, such as chimpanzee adeno- Sun et al. (2020b) viruses, are being used for vaccine delivery (Li et al., 2021a; Mendonça et al., 2021) . Among the 5 vaccines approved for emergency use, AZD-1222 (ChAdOx1 nCoV-19, NCT04516746, NCT04760132) is based on a chimpanzee adenovirus vector that contains the SARS-CoV-2 spike protein (Table 2) . It showed 70.4% vaccine efficacy after clinical testing in 23 countries and received approval in over 120 countries (Falsey et al., 2021) . JNJ-78436735 (NCT04505722) is an adenovirus serotype 26based vaccine that showed 66% efficacy after 4 weeks with a single dose (WHO, 2021a). In addition, a two-vector vaccine termed Gam-COVID-Vac (SputnikV, NCT04741061, NCT-04530396, NCT04954092, NCT04713488, and NCT04564716) has been developed and are currently authorized in 72 countries (Table 2) . This vaccine is a combination of recombinant human Ad26 and Ad5 vectors containing the same spike gene of SARS-CoV-2 incorporated (Jones and Roy, 2021) . According to recent reports, Gam-COVID-Vac showed 91.6% vaccine efficacy, with 45 severe adverse events in 16,427 patients (Logunov et al., 2021) . Unlike Sputnik V, Sputnik Light is composed of only the Ad26 vector and has been applied in a single-dose regimen (NCT04741061, NCT04713488) ( Table  2) . It showed 75.28% vaccine efficacy against delta variants in the group between the ages of 18 and 59 years, suggesting that it has even higher efficacy than other two-shot vaccines (Dolzhikova et al., 2021) . Additionally, an oral tablet COVID-19 vaccine termed VXA-COV2-1.1-S expressing two different SARS-CoV-2 proteins, spike (S) and nucleoprotein (N) was designed to protect against both prevalent and emerging strains. A phase I study revealed that VXA-COV2-1 is generally well tolerated and could have broader activity against variants with low coverage by firstgeneration vaccines incorporated (Mascellino et al., 2021) .

Currently, its safety, immunogenicity and efficacy are being evaluated in a phase 2 clinical trial (NCT05067933) ( Table 3) . Two clinical trials for AdCLD-CoV19, a novel COVID-19 vaccine based on the Ad5/35 chimeric adenoviral vector, are underway to evaluate their immunogenicity and protective effects against COVID-19 in healthy volunteers (NCT04666012, NCT05047692) (Table 3) . Moreover, a preclinical study of a gorilla adenovirus-based COVID-19 vaccine called COVITAR (GRAd-COV2) in both mouse and macaque models indicated that GRAd-COV2 could induce neutralization of SARS-CoV-2 infection and elicit a robust cell-mediated immune response (Capone et al., 2021) . Therefore, a phase 2/3 trial is also underway for COVITAR to assess its safety and efficacy in protecting against COVID-19 (NCT04791423) ( Table 3) .

MVA is an attenuated vaccina virus that has a replication-incompetent phenotype due to loss of approximately 15% of the vaccinia virus genome. Since its biological safety in vivo has been proven, recombinant MVA viruses can be handled at biosafety level 1 (Altenburg et al., 2014) . Despite its high safety, this type of vector can also lead to elicit immune responses comparable to those induced by replication competent vaccinia virus vectors. However, there are some environmental risks of dissemination of the recombinant MVA vector through excreta and blood from the treated patient when high-risk foreign genes are inserted (Verheust et al., 2012) . COH04S1 is a synthetic MVA vector that contains both the spike and nucleocapsid genes of SARS-CoV-2. A phase 1 clinical trial is ongoing to evaluate the safety and optimal dose of COH04S1, and the next phase of the clinical trial is planned to examine vaccine efficacy by comparison with vaccines that have been given an Emergency Use Authorization (Table 3) . Furthermore, it has been reported that mice injected with a single or two-dose of MVA vectorbased vaccine incorporating the SARS-CoV-2 spike gene are all survived after challenging with SARS-CoV-2 virus (García-Arriaza et al., 2021) . Another preclinical study showed that MVA-SARS-2-S vaccination led to reduction of viral loads in lung against SARS-CoV-2 infection in human ACE2-transduced mice (Tscherne et al., 2021) (Table 5) .

Parainfluenza virus belongs to the Paramyxoviridae family, which consists of negative single-stranded RNA viruses. It has been reported as the second most common pathogen that causes respiratory diseases in children under 5 years of age (Álvarez-Argüelles et al., 2018) . Parainfluenza virus type 5 (PIV5) has been effectively used as an efficient viral vector for protection against respiratory infections, including influenza virus, influenza A H5N1 virus, rabies virus, respiratory syncytial virus and Mycobacterium tuberculosis (Chen et al., 2015; Xiao et al., 2021) . Unlike positive single-stranded RNA viruses, the PIV5 vector is stable and has a low frequency of mutation in host cells. Furthermore, the serum of only approximately 30% of people was found to have neutralizing antibodies at low titers prior to exposure to PIV5 . A COVID-19 vaccine, CVXGA1 (PIV5-SARS CoV-2), based on a non-replicating PIV5 vector has been developed, and it showed no weight loss and 100% survival rate by inhibiting SARS-CoV-2 replication in the upper respiratory tract in a mouse model (Table 5) . A phase 1 clinical study is also investigating its safety and immunogenicity in humans (NCT04954287) ( Table 3) .

Since the non-replicative lentiviral vector (LV) can elicit powerful adaptive immunity, it represents a promising vaccine platform for delivering target antigens (Ku et al., 2021b) . In an HIV-1 vaccine trial, the safety of this vector has also been proven in humans (2011006260-52 EN) , and it has been used for gene therapy studies. Furthermore, LV has a broad host cell range and a low risk of reduced vaccine efficacy due to pre-existing immunity when it is enveloped with vesicular stomatitis virus G glycoprotein (VSV-G) (Ku et al., 2021a) . In a recent study, it was reported that intranasal vaccination with an LV vector encoding SARS-CoV-2 spike protein reduces viral loads in lungs by inducing mucosal immunity in rodents, suggesting the feasibility of LV-based intranasal vaccination against SARS-CoV-2. Additionally, non-integrative version of this vector for human clinical trials also showed less severe pulmonary lesions with low copies of SARS-CoV-2 RNA in lungs (Ku et al., 2021b) (Table 5) .

Unlike non-replicating viral vector vaccines, replicating viral vectors are already being manufactured for worldwide use. rVSV-ZEBOV was the first approved vaccine for Ebola based on a replication-competent viral vector in 2019 (Ku et al., 2021a) . Replication competent vectors can induce robust and persistent immune responses by producing many copies of antigen at low doses in host cells. Due to its strong immune induction ability, this system is also considered an effective mucosal delivery platform (Robert-Guroff, 2007) . However, compared to replication-deficient viral vectors, these vectors have limited space for inserted genes and still have some risk of genotoxic events caused by excessive or mutated antigen production (Choi and Chang, 2013) .

A "single-cycle" Ad (SC-Ad) vector has been used for the development of some vaccines, including vaccines against influenza A, HIV-1 or Clostridium difficile (Barry et al., 2020) . In the best currently available SC-Ad format, since key late genes of SC are deleted, the virus can replicate its genome but cannot produce progeny adenovirus virions. It has been reported that SC-Ad can elicit higher and more persistent transgene-specific IgA production than non-replicating Ad after a single intranasal immunization in hamsters (Crosby et al., 2015) . Therefore, a COVID-19 vaccine using an "SC-Ad6 vector" is also being evaluated in a phase I clinical study (NCT04839042) ( Table 3 ).

VSV contains a negative single-stranded RNA genome and belongs to the Rhabdoviridae family with rabies virus. It can infect a broad range of hosts and often causes mild illnesses in humans. VSV vectors have been examined as efficient delivery platforms in vaccine development studies (Tober et al., 2014) . Since humans are rarely exposed to this vector, there are low titers of preexisting antibodies, and the vector is less pathogenic due to modification of the VSV vector via replacement of its glycoprotein with other proteins. Additionally, the authorization of recombinant VSV for the Ebola vaccine in 2019 demonstrated its safety (Heppner et al., 2017; Bache et al., 2020) . A research group in Israel showed that the replacement of VSV glycoprotein with the spike protein of SARS-CoV-2 (VSV-ΔG-spike), termed IIBR-100, leads to the production of SARS-CoV-2 neutralizing antibodies after single-dose vaccination in a hamster model (Yahalom-Ronen et al., 2020) . Furthermore, a phase 2/3 clinical trial was performed in 2021 to evaluate the protective efficacy of this approach after massive immunization (NCT04990466) ( Table 4 ). It has been reported that another VSV vector-based COVID-19 vaccine from a Washington University research group, which is also generated by replacing the native glycoprotein gene with the SARS-CoV-2 spike gene, could elicit production of antibodies against the RBD associated with human angiotensin-converting enzyme 2 (ACE2) at high titers. Two doses of VSV-eGFP-SARS-CoV-2 vaccination showed a reduction of lung inflammation and transferring of the sera from vaccinated mice to SARS-CoV-2-challenged mice reduced viral burdens, supporting its protective effect (Case et al., 2020) (Table 5) .

Measles is one of the most contagious diseases, and it undergoes aerosol transmission. It has a negative single-stranded RNA genome and belongs to the Paramyxoviridae family. Live-attenuated MeV vaccines have been proven to be safe and effective since the 1960s, and MeV vectors have been a prominent delivery platform in studies of vaccines against HIV, HBV, HCV, influenza virus, and dengue virus (Zuniga et al., 2007; Frantz et al., 2018) . In recent studies, MeV vectors were engineered with different SARS-CoV-2 spike genes and RBD genes to compare their vaccine efficacy in vivo. Notably, it was found that recombinant MeV with the target antigen inserted into the pre-fusion S (PreS) region had the strongest protective effects with reduced severity of lung pathology by preventing the cytokine storm and viral replication in the lungs (Lu et al., 2021) . When the full-length spike gene was incorporated, recombinant MeV elicited Th1-biased immune responses, inducing not only a neutralizing antibody response but also S protein-specific clearance ability (Hörner et al., 2020) (Table 5) . Furthermore, a novel MeV-based COVID-19 vaccine termed TMV-083/ V-591 was tested for safety and immunogenicity in a phase 1 clinical trial with 90 participants divided into high-and lowdose groups (NCT04497298) ( Table 4) .

Influenza vaccines are updated frequently because of the fast evolution of the virus, and most authorized vaccines use inactivated or attenuated forms. A quadrivalent inactivated influenza vaccine that contains killed H1N1, H3N2 and influenza B virus strains was developed in 2012 and is often modified with the annual circulating strains (Nuwarda et al., 2021) . Influenza virus contains a negative single-stranded RNA genome large enough to express long foreign gene sequences and leads to efficient induction of mucosal immune responses. Influenza A, one of the most common types of influenza viruses, has been developed as a vaccine platform to protect against respiratory diseases (Barría et al., 2013; Pérez-Girón et al., 2014) . Replacement of the neuraminidase (NA)-coding region of IV with the SARS-CoV-2 spike RBD sequence was assessed for its protective effect in a mouse model. It could generate a neutralizing antibody against SARS-CoV-2 with a single intranasal immunization in a mouse model (Loes et al., 2020) . Another preclinical study used a recombinant influenza A virus, which expresses the SARS-CoV-2 spike RBD incorporated into the hemagglutinin ORF. It can generate effective neutralizing antibodies and provide protection against SARS-CoV-2, especially after boosting post-immunization (Koonpaew et al., 2021) (Table 5) . Furthermore, a clinical trial of live-attenuated recombinant IV (DelNS1-nCoV-RBD LAIV), based on a nonstructural NS1 protein-deficient influenza A virus vector, was undertaken to evaluate its safety and immunogenicity after two doses administered intranasally (NCT04809389) ( Table 4 ).

Newcastle disease virus (NDV) belongs to the Paramyxoviridae family and was first identified in Indonesia in 1926. Although its zoonoses are associated with birds, this virus rarely causes influenza-like symptoms in humans (Xiao et al., 2021) . Since there is no preexisting immunity against NDV in humans, the NDV vector Fig. 1. COVID-19 vaccines based on recombinant viral and bacterial delivery systems can provide a better protection via combinatorial effect of adaptive and trained immunity. First, they possess various pattern-associated molecular patterns (PAMPs) leading to adjuvant free vaccination. Second, they can induce adaptive T cell-mediated and humoral immune responses specific to SARS-CoV-2 infection. Third, they can induce broad-spectrum off-target vaccine effects via trained immunity, providing a better protective effect even against vaccine escape variants. has been considered a safe gene delivery platform due to host range restriction. The first attempt to generate recombinant NDV virus was made in 2000, and an NDV vector vaccine expressing the hemagglutinin (HA) gene of influenza virus was tested in mammals (Hu et al., 2020) . Recently, a SARS-CoV-2 vaccine based on NDV virus expressing the spike protein of SARS-CoV-2 was introduced as a live virus vaccine candidate. NDV vector vaccines elicit high levels of neutralizing antibodies and show no detectible viral load in the lung when the vaccine is given intramuscularly in rodents (Sun et al., 2020a (Sun et al., , 2020b (Table 5 ). The safety and immunogenicity of a SARS-CoV-2 vaccine based on NDV virus has been evaluated in a phase 1 clinical trial (NCT04871737) ( Table 4) .

The recent emergence of new variants of SARS-CoV-2 and their rapid worldwide expansion may compromise the efficacy of COVID-19 vaccines currently authorized for human application by interfering with currently used vaccines. To minimize the risk posed by SARS-CoV-2 variants, vaccine strategies that modulate T cell-mediated immune responses or induce innate cell-based heterologous effects that are less affected by mutated variants are needed. Compared to other platforms, vaccines using viral or bacterial delivery systems have several distinct merits. First, since the delivery vector itself possesses various pattern-associated molecular patterns (PAMPs), adjuvant therapy is not needed to activate innate cells. Second, these vaccines can induce both T cell-mediated and humoral immune responses against SARS-CoV-2 infection. Third, they can induce both broad-spectrum off-target vaccine effects via the unrelated vector itself, which are mediated partly by trained immunity, and specific humoral and cell-mediated immune responses against delivered COVID-19 antigens, which can provide a better protective effect, even against vaccine escape variants (Fig. 1) . Since more promising new vaccines using viral or bacterial delivery systems are under development, these approaches are expected to broaden the repertoire of COVID-19 vaccine regimens to potentiate the efficacy of current vaccines via combination with other currently available vaccine platforms.

A summary of the SARS-CoV-2 vaccines and technologies available or under development

Heterologous immunity: role in natural and vaccine-induced resistance to infections

Modified vaccinia virus ankara (MVA) as production platform for vaccines against influenza and other viral respiratory diseases

New clinical and seasonal evidence of infections by Human Parainfluenzavirus

Protection of K18-hACE2 mice and ferrets against SARS-CoV-2 challenge by a single-dose mucosal immunization with a parainfluenza virus 5-based COVID-19 vaccine

Vaccine repurposing approach for preventing COVID 19: can MMR vaccines reduce morbidity and mortality?

National immunization campaigns with oral polio vaccine reduce all-cause mortality: a natural experiment within seven randomized trials

Safety, immunogenicity and risk-benefit analysis of rVSV-ΔG-ZEBOV-GP (V920) Ebola vaccine in Phase I-III clinical trials across regions

Localized mucosal response to intranasal live attenuated influenza vaccine in adults

Single-cycle adenoviruses (SC-Ads) as platforms for immunostimulation

Can BCG vaccine protect against COVID-19 via trained immunity and tolerogenesis?

Developmental landscape of potential vaccine candidates based on viral vector for prophylaxis of COVID-19

Recombinant vector vaccine evolution

Immunogenicity of a new gorilla adenovirus vaccine candidate for COVID-19

Replication-competent vesicular stomatitis virus vaccine vector protects against SARS-CoV-2-mediated pathogenesis in mice

Genetically engineered bacteria in gene therapy-hopes and challenges

Efficacy of parainfluenza virus 5 (PIV5)-based tuberculosis vaccines in mice

Recombinant Salmonella enterica serovar Typhimurium as a vaccine vector for HIV-1 Gag

Viral vectors for vaccine applications

Salmonella vaccines: conduits for protective antigens

Poliovirus vaccination induces a humoral immune response that cross reacts with SARS-CoV-2. Front

BCG-induced cross-protection and development of trained immunity: implication for vaccine design

Induction of trained immunity by recombinant vaccines

Amplified and persistent immune responses generated by singlecycle replicating adenovirus vaccines

Live bacterial vaccines-a review and identification of potential hazards

One-shot immunization with Sputnik Light (the first component of Sputnik V vaccine) is effective against SARS-CoV-2 Delta variant: efficacy data on the use of the vaccine in civil circulation in Moscow

Phase 3 safety and efficacy of AZD1222 (ChAdOx1 nCoV-19) COVID-19 vaccine

An outbreak of primary pneumonic tularemia on Martha's Vineyard. N. Eng

Review of COVID-19 vaccine subtypes, efficacy and geographical distributions

Measles-derived vaccines to prevent emerging viral diseases

The central role of the nasal microenvironment in the transmission, modulation, and clinical progression of SARS-CoV-2 infection

COVID-19 vaccine candidates based on modified vaccinia virus Ankara expressing the SARS-CoV-2 spike protein induce robust T-and B-cell immune responses and full efficacy in mice

Analysis of measles-mumps-rubella (MMR) titers of recovered COVID-19 patients

Harnessing the beneficial heterologous effects of vaccination

SARS-CoV-2 human T cell epitopes: adaptive immune response against COVID-19

Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals

Salmonella typhimurium phoP virulence gene is a transcriptional regulator

Heterologous effects of vaccination and trained immunity

Safety and immunogenicity of the rVSVΔ G-ZEBOV-GP Ebola virus vaccine candidate in healthy adults: a phase 1b randomised, multicentre, double-blind, placebo-controlled, doseresponse study

Oral administration of the probiotic bacterium Lactobacillus acidophilus strain L-55 modulates the immunological parameters of the laying hen inoculated with a Newcastle disease virus-based live attenuated vaccine

A highly immunogenic and effective measles virus-based Th1-biased COVID-19 vaccine

Newcastle disease virus as a vaccine vector for 20 years: a focus on maternally derived antibody interference

Structural and functional properties of SARS-CoV-2 spike protein: potential antivirus drug development for COVID-19

A comparative analysis of COVID-19 outbreak on age groups and both the sexes of population from India and other countries

Replicating bacterium-vectored vaccine expressing SARS-CoV-2 Membrane and Nucleocapsid proteins protects against severe COVID-19-like disease in hamsters

A Francisella tularensis live vaccine strain (LVS) mutant with a deletion in capB, encoding a putative capsular biosynthesis protein, is significantly more attenuated than LVS yet induces potent protective immunity in mice against F. tularensis challenge

Recombinant attenuated Listeria monocytogenes vaccine expressing Francisella tularensis IglC induces protection in mice against aerosolized Type A F. tularensis

Sputnik V COVID-19 vaccine candidate appears safe and effective

Recombinant Mycobacterium paragordonae expressing SARS-CoV-2 receptor-binding domain as a vaccine candidate against SARS-CoV-2 infections

Adenoviral vector vaccine platforms in the SARS-CoV-2 pandemic

Intestinal colonisation-inhibition and virulence of Salmonella phoP, rpoS and ompC deletion mutants in chickens

The role of mucosal immunity and recombinant probiotics in SARS-CoV2 vaccine development

Vaccines for biodefense and emerging and neglected diseases

An overview of current COVID-19 vaccine platforms

Characterisation of alternative expression vectors for recombinant Bacillus Calmette-Guérin as live bacterial delivery systems

Defining trained immunity and its role in health and disease

Trained immunity: a tool for reducing susceptibility to and the severity of SARS-CoV-2 infection

Trained immunity: a program of innate immune memory in health and disease

National immunisation campaigns with oral polio vaccine may reduce all-cause mortality: analysis of 2004-2019 demographic surveillance data in rural Bangladesh

The recombinant Bacille Calmette-Guérin vaccine VPM1002: ready for clinical efficacy testing

An overview of influenza viruses and vaccines

Lessons from Bacillus Calmette-Guérin: harnessing trained immunity for vaccine development

Nontuberculous mycobacteria: Molecular and physiological bases of virulence and adaptation to ecological niches

Mucosal polyinosinic-polycytidylic acid improves protection elicited by replicating influenza vaccines via enhanced dendritic cell function and T cell immunity

Preliminary genomic characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a novel set of spike mutations

Replicating and non-replicating viral vectors for vaccine development

Salmonella as a vaccine delivery vehicle

Trained immunity-based vaccines: a new paradigm for the development of broad-spectrum anti-infectious formulations

An effective COVID-19 vaccine needs to engage T cells

Immune responses to viral gene therapy vectors

Live bacterial vaccine vectors: an overview

Host-pathogen interactions and immune evasion strategies in Francisella tularensis pathogenicity

Newcastle disease virus (NDV) expressing the spike protein of SARS-CoV-2 as a live virus vaccine candidate

A Newcastle disease virus (NDV) expressing a membrane-anchored spike as a cost-effective inactivated SARS-CoV-2 vaccine

Neonatal BCG vaccination and child survival in TB-exposed and TB-unexposed children: a prospective cohort study

VSV-GP: a potent viral vaccine vector that boosts the immune response upon repeated applications

COVID19 vaccine traker

Immunogenicity and efficacy of the COVID-19 candidate vector vaccine MVA SARS 2 S in preclinical vaccination

Developments in viral vector-based vaccines

Biosafety aspects of modified vaccinia virus Ankara (MVA)-based vectors used for gene therapy or vaccination

Development and applications of viral vectored vaccines to combat zoonotic and emerging public health threats

A recombinant Lactobacillus plantarum strain expressing the spike protein of SARS-CoV-2

A single-dose recombinant parainfluenza virus 5-vectored vaccine expressing respiratory syncytial virus (RSV) F or G protein protected cotton rats and African green monkeys from RSV challenge

WHO, World Health Organizaion. 2021a. The Janssen Ad26.COV2.S COVID-19 vaccine: What you need to know

Live Attenuated Vaccines (LAV)

SARS-CoV-2 501Y. V2 escapes neutralization by South African COVID-19 donor plasma

Worldometer. 2021. COVID-19 Coronavirus Pandemic

Parainfluenza Virus 5 (PIV5) priming followed by SIV/HIV viruslike-particle boosting induces potent and durable immune responses in nonhuman primates

A single dose of recombinant VSV-ΔG-spike vaccine provides protection against SARS-CoV-2 challenge

SARS-CoV-2 spike protein: a key target for eliciting persistent neutralizing antibodies

Recombinant Lactobacillus plantarum expressing HA2 antigen elicits protective immunity against H9N2 avian influenza virus in chickens

Induction of autonomous memory alveolar macrophages requires T cell help and is critical to trained immunity

Effectiveness of booster measles-mumps-rubella vaccination in lower COVID-19 infection rates: a retrospective cohort study in Turkish adults

Evidence of escape of SARS-CoV-2 variant B. 1.351 from natural and vaccine-induced sera

Transplantation of mesenchymal stem cells: a potential adjuvant therapy for COVID-19

Oral delivery of SARS-CoV-2 DNA vaccines using attenuated Salmonella typhimurium as a carrier in rat

Attenuated measles virus as a vaccine vector

This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (Grant No. HV20C0147). The funder was not involved in the analysis, interpretation of data, the writing of this article or the decision for publication.

The authors declare that the article have no potential conflicts of interest to disclose.