The Route Matters

by Gertrud U. Rey

There are currently 315 therapeutic drugs and 210 vaccine candidates in development to treat or prevent SARS-CoV-2 infection. Many of these vaccines are designed to be administered by injection into the muscle. 

Intramuscular injection of a vaccine antigen typically induces a systemic (serum) immune response that involves the action of IgM and IgG antibodies. IgM antibodies appear first and typically bind very strongly to antigens, to the extent that they often cross-react with other, non-specific antigens. IgG antibodies arise later, are a lot more specific than IgM, and provide the majority of antibody-based immunity against invading pathogens. Intramuscular immunization usually does not induce very high levels of serum IgA, a type of antibody that is more prevalent in mucosal surfaces and represents a first line of defense against invasion by inhaled and ingested pathogens. The role of IgA in the serum is mostly secondary to IgG, in that IgA mediates elimination of pathogens that have breached the mucosal surface.    

Several of the SARS-CoV-2 vaccine candidates currently in clinical trials consist of a replication-deficient adenovirus with an inserted gene that encodes a SARS-CoV-2 antigen. The suitability of adenoviruses as vectors for delivering foreign genes into cells was discussed in a previous post, which summarized preliminary phase I/II clinical trials assessing the safety and efficacy of a chimpanzee adenovirus-vectored replication-deficient SARS-CoV-2 vaccine candidate encoding the full-length SARS-CoV-2 spike protein (AZD1222). The spike protein has been the primary antigenic choice for a number of SARS-CoV-2 vaccine candidates because it mediates binding of the virus to the ACE2 host cell receptor via its receptor-binding domain (RBD), and it also mediates fusion of the viral particle with the host cell membrane via its fusion domain. Both of these spike domains are highly immunogenic and are targeted by neutralizing antibodies, which bind viral antigens, inactivating virus and preventing infection of new cells. However, preliminary results suggest that AZD1222 only protects against SARS-CoV-2 lung infection and pneumonia but doesn’t appear to prevent upper respiratory tract infection and viral shedding.

To mediate fusion of the virus particle to the host cell membrane, the SARS-CoV-2 spike protein undergoes a structural rearrangement from its pre-fusion conformation. Because the pre-fusion form is more immunogenic, vaccines encoding the spike protein often contain a mutation that locks the translated spike protein into this pre-fusion structure. In a recent publication, virologist Michael Diamond and colleagues analyzed the efficacy of an adenovirus-vectored SARS-CoV-2 vaccine candidate and compared its protective effects after intramuscular injection to those after administration by the intranasal route. The vaccine, named ChAd-SARS-CoV-2-S, is similar to AZD1222 except that its spike gene encodes the pre-fusion stabilized spike protein. To assess the antibody responses induced by intramuscular vaccination with ChAd-SARS-CoV-2-S, the authors injected mice with 10 billion viral particles of either ChAd-SARS-CoV-2-S or a control vaccine consisting of the same adenovirus shell, but lacking the spike protein gene insert. They found that one dose of ChAd-SARS-CoV-2-S induced strong serum IgG responses against both the entire spike protein and the RBD, but no IgA responses in the serum or in mucosal lung cells.

While antibodies are an important part of the adaptive immune response, cell-mediated immunity is just as important and at the very least results in activation of white blood cells that destroy ingested microbes and also produces cytotoxic T cells that directly kill infected target cells. During a first exposure to a pathogen, T helper cells typically sense the presence of antigens on the surface of the invading pathogen and release a variety of signals that ultimately stimulate B cells to secrete antibodies to those antigens and also stimulate cytotoxic T cells to kill infected target cells. Analysis of these T cells in mice immunized with one or two doses of intramuscularly administered ChAd-SARS-CoV-2-S revealed that two vaccine doses induced both T helper and cytotoxic T cell responses against the whole spike protein. Collectively, these results suggest that although intramuscular vaccination produces strong systemic adaptive immune responses against SARS-CoV-2, it induces little, if any, mucosal immunity.

To determine whether intramuscular immunization with ChAd-SARS-CoV-2-S protects mice from infection, the authors intentionally infected (“challenged”) immunized mice with SARS-CoV-2. Although a single vaccine dose protected the mice from SARS-CoV-2 infection and lung inflammation, the mice still had high levels of viral RNA in the lung after infection, suggesting that intramuscular administration of the vaccine does not lead to complete protection from infection.  

In an effort to see whether vaccination by the intranasal route provides more complete protection, the authors inoculated mice with a single dose of ChAd-SARS-CoV-2-S or control vaccine through the nose. Analysis of serum samples and mucosal lung cells four weeks after vaccination revealed that recipients of ChAd-SARS-CoV-2-S had high spike- and RBD-specific levels of neutralizing IgG and IgA in both the serum and the lung mucosa, and that the number of B cells producing IgA was about five-fold higher than that of B cells producing IgG. Interestingly, the neutralizing antibodies were also able to inactivate SARS-CoV-2 viruses containing a D614G change in the spike protein, suggesting that ChAd-SARS-CoV-2-S can effectively protect against other circulating SARS-CoV-2 viruses. Intranasal vaccination also induced SARS-CoV-2-specific cytotoxic T cells in the lung mucosa, specifically T cells that produce interferon gamma, an important activator of macrophages and inhibitor of viral replication.

The ideal immune response is “sterilizing” – meaning that it completely protects against a new infection and does not allow the virus to replicate at all. To evaluate the ability of a single intranasal dose of ChAd-SARS-CoV-2-S to induce sterilizing immunity, the authors analyzed immunized and infected mice for serum antibodies produced against the viral NP protein. Because the vaccine does not encode the NP protein, any antibodies produced against this protein would be induced by translation of the NP gene from the challenge virus and active replication of the virus. All of the mice immunized with a single dose of intranasally administered ChAd-SARS-CoV-2-S had very low levels of anti-NP antibodies compared to recipients of the control vaccine, suggesting that ChAd-SARS-CoV-2-S induced strong mucosal immunity that prevented SARS-CoV-2 infection in both the upper and lower respiratory tract. This means that if intranasally immunized mice were to be exposed to SARS-CoV-2, they would not be able to replicate the virus or transmit it to others. 

The study has some notable limitations. First, it is well known that mice can be poor predictors of human disease outcomes. Second, because the mouse ACE2 receptor doesn’t easily bind SARS-CoV-2, the mice were engineered to express the human ACE2 receptor, which added a further artificial variable to an already imperfect model system. Third, it is presently unknown how long the observed immune responses would last. That being said, studies with influenza virus have shown that mucosal immunization through the nose can elicit strong local protective IgA-mediated immune responses. Further, there are clear advantages to intranasal vaccine administration: inoculation is simple, painless, and does not require trained professionals. The adequacy of a single dose would also lead to more widespread compliance. Lastly, a vaccine that prevents viral shedding would be ideal, because in addition to preventing disease in the exposed individual, it would prevent transmission to others. 

None of the SARS-CoV-2 vaccine candidates currently in clinical trials are delivered by the intranasal route. If the results observed in these mouse experiments can be duplicated in humans, ChAd-SARS-CoV-2-S would clearly be superior to other SARS-CoV-2 vaccine candidates. 

[The material in this blog post is also covered in this video.]