The importance of cross-neutralizing antibodies to the recently emerged variants of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is evident in light of the fact that these variants are highly transmissible as well as being able to evade immune responses to earlier variants of this virus.
A new preprint research paper posted to the bioRxiv* server reports on the ability of a set of neutralizing mouse antibodies to protect mice against multiple lineages of SARS-CoV-2.
Antibodies and virus neutralization
About twenty vaccines have been authorized for use worldwide, in addition to several monoclonal antibodies, as combinations, to prevent and treat SARS-CoV-2 infection, respectively.
This virus enters host cells to establish successful infection by binding via its spike protein to the host cell receptor, the angiotensin-converting enzyme 2 (ACE2). The part of the spike that binds to the receptor is called the receptor-binding domain (RBD) and forms an attractive target for antibodies, both therapeutic and preventive.
Neutralizing antibody titers are highly correlated with protection against reinfection, but since vaccination will take a while to cover the global population, there is an urgent need for therapeutic antibodies and drugs to treat and prevent severe illness.
The currently available monoclonal antibodies (mAbs) have been raised against the spike of the original Wuhan variant of the virus. However, the emergence of escape mutations, especially in the RBD region of the spike, threatens the usefulness of these mAbs.
Since most neutralizing antibodies act by blocking viral entry, targeting the RBD, mutations at the RBD may reduce their efficiency of the mAbs.
The N-terminal domain (NTD) is another region that is frequently targeted in newer variants, as it is neutralized by many antibodies. These variants may thus be capable of evading vaccine-induced antibodies.
Study aims and details
The researchers from the Krammer Laboratory, Icahn School of Medicine at Mount Sinai, aimed to evaluate the binding and neutralizing ability of 14 mouse mAbs raised against the viral RBD to both the RBD and the spike protein. They also assessed the potential for non-neutralizing mAbs to reduce the viral load.
Thirdly, the study examined the binding of these mAbs to the RBD mutants containing point and grouped mutations that define the three major variants of concern (VOCs) – the UK (B1.1.7), the South African (SA) variant (B.1.351), and the Brazil variant (P.1).
Finally, the mAbs were tested for neutralization against the UK and SA variants.
What were the results?
The murine immunoglobulin G (IgG) antibodies were raised in hybridoma cells. All 14 were found to bind to the RBD with high affinity, as shown by the very low minimal binding concentrations (MBC).
When tested against the full spike protein, most of them bound well, but three had higher MBCs relative to the RBD. The researchers postulate that this could be due to the covering of the RBD-binding epitopes on the full spike as compared to the RBD expressed in isolation.
Ten of the 14 mAbs were specific for the SARS-CoV-2 RBD, but four showed cross-reactivity to SARS-CoV RBD as well.
When tested with the live virus in a microneutralization assay, where the antibody's ability to block viral infection of the culture cells is determined as the reciprocal of the dilution at which the cells remain completely uninfected, six showed high neutralizing ability.
In all six cases, the concentration at which 50% of cells remained uninfected (IC50) by the live virus was very low, at 0.1-1 ug/ml. This indicates that these can prevent viral entry and replication at high dilutions, especially two which had the lowest IC50.
Reduction of viral titers in mice
The researchers tested the mAbs for their ability to block the entry of the virus and reduce viral loads in engineered mice expressing the human ACE2 gene. The mAbs were administered two hours before challenging the mice with SARS-CoV-2.
Plaque reduction neutralization tests showed that only the six mAbs with neutralizing capacity were able to protect the animal against viral entry and replication. Three of the six reduced the viral load by two logs on day 3, while another two caused a markedly more significant reduction in viral titer.
On day 5, none of the animals which received a neutralizing mAb showed signs of significant viral presence in their lungs, though two showed a residual virus
Lung pathology reduced
Lung tissue was harvested on day 4 for histopathology and immunohistochemistry (IHC). This showed some evidence of interstitial pneumonia, perhaps because of the high dose of the virus.
Some mild inflammation was found to be the result of injecting the adenoviral hACE2 vector, but the inflammation scores in mice that received the vector and the virus were higher. There was no sign of antibody-dependent enhancement of disease (ADE).
Nucleoproteins were almost undetectable, showing that the neutralizing mAbs all prevented viral entry into the cell, and thus lowered the viral titer in the lung.
Binding to RBD variants
The study also shows that some variant RBDs like K487R and N487R caused one mAb each to lose complete binding. The mutations E484K, F486A, and F490K, also reduced binding by a third mAb. The grouped mutations on the SA variant RBD also caused loss of binding by the same mAb.
However, five of the mAbs maintained 50% binding affinity at least to all mutant RBDs, and some had higher binding than for the wild-type RBD. The researchers comment, “The ability to bind all RBDs could be a function of antibody affinity which, when high, can allow the antibody to maintain its footprint.”
Overall, neutralizing mAbs bound wildtype and mutant RBDs at somewhat equivalent levels.
Of particular interest, four of the six neutralizing mAbs bound both wildtype and the SA RBD variant, while all six neutralized the UK RBD variant. The latter has only one RBD mutation, N501Y.
Of the four which maintained binding and neutralization of the SA variant, the IC50 increased from 2.5-fold to 5-fold that for the wildtype virus or the UK variant.
The reasons for the loss of neutralization of the SA variant could have been the lower binding to the E484-containing RBD, for one mAb, and an altered epitope presentation on the spike compared to the RBD for the other.
Inhibition of ACE2-RBD binding
The researchers also found by structural analysis of the six neutralizing mAbs that two of them failed to form stable complexes with the RBD, and could not be visualized. Three of them showed an overlap with the ACE2 binding site that could explain how they prevented ACE2-RBD binding.
These bind at various angles to the RBD, indicating that they bind to different epitopes.
What are the implications?
The study shows that the spike RBD can undergo multiple mutations while still retaining its ability to bind to the ACE2 receptor and cause infection of the host cell.
“The plasticity of the RBD is alarming because extensive changes in the RBD could reduce the efficacy of current vaccines and additional booster vaccinations with updated vaccines may be needed for protection in the future.”
The lack of protection by non-neutralizing mAbs is attributable to their IgG1 isotype, which, in mice, precludes interactions mediated by effector Fc receptors (FcRs). Thus, antibodies that do not activate FcRs and do not have neutralizing capacity lack protective efficacy.
The two antibodies that protected against viral entry the best are IgG2a, a subtype associated with Fc-Fc interactions in mice, supporting the postulate that antibody binding to activating effector FcRs are key to the protective effects of antibodies.
In humans, IgG2 antibodies do interact strongly with FcRs, and most antibodies to this virus belong to this isotype.
The study suggests the potential for humanized forms of these neutralizing mAbs to be developed as therapeutics with strong activity against the wildtype virus, as well as the UK and SA variants.
It’s all in the long but excellent title:
“Murine monoclonal antibodies against RBD of SARS-CoV-2 neutralize authentic wild type SARS-CoV-2 as well as B.1.1.7 and B.1.351 viruses and protect in vivo in a mouse model in a neutralization dependent manner” https://t.co/Lbw4seLU4D- Dr Kevin Purcell (@kevinpurcell) April 6, 2021
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.