The COVID-19 pandemic is threatening to come back in greater intensity during the winter and fall. Faced with this threat, vaccine development efforts are entering a phase of renewed focus, attempting to identify antigens capable of eliciting protective levels of neutralizing antibodies. A new study published on the preprint server bioRxiv* in October 2020 reports on a novel mRNA vaccine candidate's performance in two animal models.
The Spike Protein and Immunity
SARS-CoV-2 is an RNA virus with numerous spike proteins studding its lipid bilayered envelope. The spike glycoprotein, which exists in trimeric form, is a significant target of the immune response. It binds to the host cell receptor, the angiotensin-converting enzyme 2 (ACE2), to facilitate viral entry into and infection of the host.
Most current vaccine candidates bind to a pre-fusion stabilized spike protein, whether a recombinant protein combined with an adjuvant, or inserted by viral vectors, or delivered as DNA or mRNA molecules that are transcribed to the corresponding protein.
The spike protein has an S1 and S2 subunit that is required at different stages of viral entry. The S1 is responsible for attachment to ACE2 through its receptor-binding domain (RBD). The RBD is known to have an ‘up’ and ‘down’ conformation, which affects the virus's binding to the receptor. Only in the ‘up’ state does RBD-ACE2 interaction occur.
Stabilizing the Prefusion Spike Protein
In the prefusion state, the spike protein must be stabilized, which is typically by two proline substitutions (S2P). These do not significantly alter the spike structure from the native protein, which is essential since many neutralizing antibodies act by binding to the former. This focus is based on previous research on the coronaviruses implicated in the earlier SARS and MERS outbreaks and other seasonal coronaviruses. The prefusion form of the spike protein was used.
The spike protein of SARS-CoV-2 has a polybasic furin cleavage site at the interface of S1 and S2 subunits, a unique feature in this virus that may be traced to the leap across the species barrier from animals to human hosts. It is thought to be central to the infectivity of the virus and lung pathology.
The cleavage site's presence has been postulated to cause slight changes in the conformation of the trimeric spike, which could promote its engagement with ACE2. However, in cell culture, the spike protein has been found in both cleaved and uncleaved form simultaneously. In fact, almost half of the spike protein within intact viral particles is in the uncleaved form, making the former premise doubtful.
The neutralizing antibody response can be based on either blocking the RBD-ACE2 interaction or preventing the spike protein's transition from pre- to post-fusion form. The contribution of the furin cleavage site is still to be established beyond doubt.
GSAS Variants Prevent Cleavage
The researchers constructed different spike forms to further explore these areas, including the 2P and the cleavage site (with the polybasic RRAR residues), which has undergone GSAS mutations. They synthesized mRNA to produce four types of spike protein, the wildtype (WT), the stabilized prefusion mutant 2P, furin cleavage site mutant (GSAS), or a double mutant (2P/GSAS) version.
They found that both the WT and the S2P proteins showed endogenous cleavage. The other two remained intact. They also performed ELISA testing of the antibody activity in the serum sample. They found that binding antibody activity was similar for all four protein variants after one dose and increased after the second.
They then examined the neutralizing capacity of immune sera on fluorescent pseudoviruses expressing the spike protein. They found that neutralizing antibodies appeared only after the second dose and were spread over a broader range than the binding antibodies with no significant difference in titer but a trend towards a higher neutralizing activity for the GSAS variant.
MRT5500 Elicits Potent Neutralizing Antibodies
They examined these constructs' ability to elicit an immune response in mice when administered in the form of a lipid nanoparticle (LNP)-encapsulated 2P/GSAS S mRNA formulation. This is currently named MRT5500.
This was given in two intramuscular doses at intervals of 3 weeks in mice and animal models. Using doses over a range covering a tenfold variation, they found that in mice, anti-spike neutralizing antibodies are induced by MRT5500 in a dose-dependent fashion.
However, there is no significant increase in PsV neutralization titers on day 35 between the low and high-dose groups, which may indicate that a dose of 1 µg produces saturation. The peak PsV titers in mice at this point were much higher than those found in convalescent sera from COVID-19 patients.
In NHPs, they tested three higher dosages, namely, 15, 45, or 135 µg per dose. A dose-dependent rise in neutralizing antibody titer was demonstrated, but the difference between titers at each dosage was not significant. In almost all animals, one dose produced antibodies detected by the ELISA, with rising titers after the second dose. They observed high nAb titers at day 35.
Overall, whatever the dose, they found that the neutralizing antibody titer went up ~130-fold from the titer found in animals prior to immunization. They also found a Th1-biased T cell response following immunization, which indicates a low risk of vaccine-associated enhanced respiratory disease (VAERD).
Implications and Future Directions
The researchers comment, “mRNA-based vaccine development provides a rapid pathway for effective evaluation of multiple vaccine construct designs which we employed for our initial evaluation of S antigen mRNA vaccine candidates against SARS-CoV-2.”
The attempt to provide a vaccine antigen with the right conformation to elicit neutralizing responses was successful since otherwise, immunogenicity will remain unrelated to neutralizing potency.
The researchers identified mutations that resulted in a stabilized form of the prefusion spike protein and made this mRNA vaccine unique among current spike mRNA candidate vaccines by stabilizing it in the prefusion form. This was to allow for the possibility that the double proline mutation alone might not accomplish this, and also because it is not clear whether the spike cleavage is essential to its change from the pre-fusion to the post-fusion form. Thus, they chose to block the furin cleavage site and thus prevent this transition.
They found that the spike variants containing the GSAS mutations had higher nAb titers in both cases, perhaps because of the difference in the efficiency with which the spike protein is trafficked from the ER to the surface of the cell after the alteration in the furin cleavage site. Another reason could be that the selected dose was not capable of differentiating the neutralizing response correctly. Nonetheless, the experiment suggests the potential neutralizing capacity of the MRT5500 vaccine in clinical situations.
This formulation's advantages include the ability to induce efficient antigen expression de novo and thus elicit high immune responses. More research is required to establish its capacity to produce effective expansion and maturation of memory B cells for durable immunity.
This double mutant vaccine candidate locks the spike protein in its prefusion conformation and elicits potent neutralizing antibodies, as well as a Th1-biased cellular response, which makes it ideal for further development as a potential COVID-19 vaccine.
bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.