Discussion

In this report, we present data on antibody modulation of Spike-monocyte interactions. To the best of our knowledge, only one previously published report showed Spike-bead phagocytosis after opsonization with 50% heat-inactivated serum and a 16 hours incubation of beads with THP-1 cells (22). We believe that the data from our experiments are more representative of the first few events after Spike-monocyte contact. Phagocytosis of small particles such as virions is a process that takes minutes, not hours (27). That is why we use shorter incubation times (30 min) and perform dose-response analysis across varying plasma concentrations. The dose-response analysis we performed exposed an antibody-mediated modulation of the Spike-monocyte interactions. It is important to elaborate on the concentrations we utilized, be it for plasma or monoclonal antibodies. For plasma, as tissues have a lower concentration of plasma proteins than whole blood, we used 1% as the highest concentration. As for the monoclonal antibody concentrations, we used 100 µg/ml as the highest concentration because it is roughly 1% of the antibody concentration present in plasma (10 mg/ml). Even though most of the experiments in this study are performed in vitro, we believe the effects we observe on phagocytosis could also be relevant in vivo, as the modulation effects occur already at relatively low antibody concentrations and would thus cover many physiological niches and scenarios.

That an increased binding of antibodies to a prey results in reduced or blocked phagocytosis is in stark contrast to what is typically expected (28). It cannot be explained by specific monoclonal interactions, as it is seen across diverse monoclonals as well as in convalescent polyclonal samples. This block in phagocytosis is only related to Spike protein. A combination of known mechanisms could potentially explain how SARS-CoV-2 could avoid phagocytosis using Spike protein. Bivalent trans-binding of antibodies is known to promote virion phagocytosis (29), where antigens are cross-linked depending on their density at the surface. Spike protein density on SARS-CoV-2 varies (30), and is increased with the D614G mutation (31). An increase in Spike anti-body levels would lead to a competition of epitope binding, ultimately favoring the switch from bivalent trans-binding to monovalent binding, potentially leading to a reduction in phagocytosis. A synergistic mechanism could further aid SARS-CoV-2. Antigen height (especially below 10 nm) is important for efficient phagocytosis (32), and most likely, a consistent antigen height is beneficial as well. Besides altering its density, SARS-CoV-2 also appears to be able to dramatically change the Spike protein conformation, where some proteins stand up vertically from the surface, and others are tilted down horizontally (30). At high anti-Spike levels, this would present an approaching phagocyte with a monovalently opsonized, irregular surface with variable antigen height (15-25 nm), making the interaction difficult. In contrast, at low anti-Spike levels, the antibodies would be able to clasp Spike proteins in a bivalent, upright manner, presenting the phagocyte with a coherently opsonized surface at an effectively low antigen height. Careful mechanistic studies are needed to test this hypothesis…

https://www.biorxiv.org/content/10.1101/2021.10.14.464464v1.full

>>14853428

…

How does the concentration-dependent phagocytosis modulation translate into an infection model? We utilized our anti-Spike antibodies as a therapeutic regimen in our animal infection. This creates a scenario where the mouse is already fighting the infection before therapy gets administered. The requirements, therefore, on therapeutic antibodies are higher than on prophylactic ones. When comparing the 100-µg dose of Ab59 and Ab94 (equivalent to 2.5 times lower binding affinity), we noticed that Ab59 showed better protection than the latter antibody. This could be attributed to either the neutralizing activity of Ab59, which Ab94 lacks, or the lower effective dose, given the lower Spike binding affinity of Ab94. Consistent with our data on phagocytosis (Fig. 5), animals treated with the high dose of Ab59 (500 µg) fared worse than the animals that got the 100-µg dose. Interestingly, the animals that got the affinity-corrected dose for Ab94 (250 µg, equivalent to 100 µg of Ab59 in terms of binding affinity) fared the best in our cohort. The animal infection data on Ab94 protecting the mice is, most importantly, congruent with the phagocytosis data, indicating a role in infection management for non-neutralizing antibodies. In future studies, it remains to be established how infections would proceed in animals treated with an excessive dose of non-neutralizing antibodies. Our animal experiments show that monoclonal antibodies could be viable therapeutics even if they lack neutralizing potential. This is an interesting discovery that, to the best of our knowledge, has not been shown before. Additionally, the dose variation data from Ab59-treated groups reflects the data we have presented in the concentration-dependent modulation of Spike phagocytosis section. These results may explain the clinical findings seen in antibody therapy trials in the sense that having low or excessive dosage of antibodies offers no clinical benefit (33).

It is widely accepted that a strong positive correlation exists between COVID19 disease severity and antibody titers (34, 35, 36, 37). High antibody titers are generally associated with severe disease and hospitalization. The high titers are thought to be a consequence of the severe infection. Our results pose a new question: could the high anti-Spike titers seen in hospitalized patients instead (at least partially) contribute to the immune dysregulation and worsening patient outcomes? These questions are relevant in the light of the FDA’s recommendation not to use monoclonal antibodies in hospitalized COVID19 patients (and who are seropositive) due to possible worsening of symptoms (3, 5). At the same time, it is important to note that convalescent plasma treatment for COVID19 was shown to be neither beneficial nor detrimental (38). Overall, the results presented in this study highlight a concentration-dependent modulation of phagocytosis by anti-Spike antibodies. This modulation phenomenon might help explain the unclear clinical benefit seen with monoclonal antibody treatment for COVID19. This modulation is seen in patient material and translates well to animal infection experiments. This opens several avenues for further experimentation. The biophysical mechanism underlying the antibody-mediated phagocytic modulation is an exciting topic to pursue, as are the bridging immune steps between phagocytosis and animal protection.