Cross-reactive antibodies after SARS-CoV-2 infection and vaccination

Decision letter after peer review:

Thank you for submitting your article "Cross-reactive antibodies after SARS-CoV-2 infection and vaccination" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Sara Sawyer as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Andrew T McGuire (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

In this study, Grobben et al. examines whether binding antibodies that cross-react with the spikes of diverse coronaviruses are elicited by SARS-CoV-2 infection. The manuscript is well written, and the figures are laid out in an easy to interpret manner. This study will be of interest to those who are interested in developing pan coronavirus vaccines. However, the following issues should be addressed before acceptance:

1. What are the relative contributions to the SARS-CoV-2 neutralizing activity in the sera upon depletion of S1 or S2 subdomains? It would be very interesting to assess whether there a reduction of serum neutralizing activity following S2 mAb depletion as would imply that the cross binding S2 antibodies might have neutralizing activity against diverse CoVs.

2. Do the authors have access to human post-vaccine sera? It would be interesting to compare the results of natural infection to vaccination.

3. Could any of the differences in cross reactive binding be due to differences in the stability of the various spike proteins? Presumably all of the spike-ectodomain designs will stabilize the proteins in the pre-fusion state, but it is not demonstrated that they are. Misfolding could lead to the destruction of relevant conformation-dependent cross reactive epitopes.

4. Related to the point 4 above, is the S2 used in the depletion studies in the pre-fusion or post-fusion conformation? Since the 2 conformations are quite different structurally, it's likely that they are antigenically distinct, and may therefore fail to deplete out relevant cross-reactive serum antibodies.

If neither of the points (3 and 4) can be easily addressed experimentally these caveats should be pointed out in the discussion.

5. The authors suggest that the macaques are likely CoV naïve, however the data suggest that this may not be true. If the animals are truly naive, then all of the cross-reactive antibodies should have been elicited by the SARS-CoV-2 spike protein and similarly depleted by the SARS-CoV-2 spike protein. For example, in Figure 4b all of the SARS-CoV-2 spike binding antibodies are depleted by the SARS-CoV-2 spike, but only about 70% of the SARS and OC43 binding antibodies were depleted. Do the authors have an alternative explanation for this?

Reviewer #1:

In this study, Grobben et al. evaluated the cross-reactivity of anti-SARS-CoV-2 S antibodies against the S proteins of a variety of coronaviruses. The results shown by the authors are important and should be published for the post-vaccination era. However, the cross-reactivity is mostly evaluated by only the binding affinity to the S proteins.

The results shown by Grobben et al. are basically important and should be published for the post-vaccination era. However, the cross-reactivity is mostly evaluated by only the binding affinity to the S proteins. The authors should evaluate the neutralizing activity of these antibodies using live viruses and/or pseudoviruses. Using all coronaviruses might be difficult, but some (e.g., seasonal coronaviruses such as 229E and NL63) should be used.

Evaluation of neutralizing activity would be important for the authors' conclusion. Evaluating the binding affinity to a variety of coronavirus S protein is very important and insightful. However, to consider its biological activity, neutralizing effect is much more important, because binding affinity and neutralizing affinity are not necessarily correlated.

Reviewer #3:

Here the authors were seeking to assess whether SARS-CoV-2 infection elicited antibodies that cross react with diverse human CoVs using a Luminex platform. The major conclusions are that cross-binding antibodies are elicited by SARS-CoV-2 infection, and that there is a positive association between the magnitude of the response and the sequence similarity to SARS-CoV-2. As a result, most cross-reactive antibodies target the S2 region which is the most conserved among diverse CoVs. The study also shows that immunization with a SARS-CoV-2 spike derived nanoparticle vaccine similarly elicits cross reactive CoV antibodies in non-human primates. Overall, the data suggests that the S2 domain might be important for the development of pan CoV vaccines.

The study is largely observational, and conclusions are supported by the data. The major limitation is that it is not clear what role the measured cross-reactive antibodies might play in protective immunity. If a function could be assigned to the cross-reactive antibodies ie neutralization or ADCC, then these findings might be more impactful for guiding the design of pan CoV vaccines.

What are the relative contributions to the SARS-CoV-2 neutralizing activity in the sera upon depletion of S1 or S2 subdomains? It would be very interesting to assess whether there a reduction of serum neutralizing activity following S2 mAb depletion as would imply that the cross binding S2 antibodies might have neutralizing activity against diverse CoVs.

Do the authors have access to human post-vaccine sera? It would be interesting to compare the results of natural infection to vaccination.

Could any of the differences in cross reactive binding be due to differences in the stability of the various spike proteins? Presumably all of the spike-ectodomain designs will stabilize the proteins in the pre-fusion state, but it is not demonstrated that they are. Misfolding could lead to the destruction of relevant conformation-dependent cross reactive epitopes.

Related to the above point, is the S2 used in the depletion studies in the pre-fusion or post-fusion conformation? Since the 2 conformations are quite different structurally, it's likely that they are antigenically distinct, and may therefore fail to deplete out relevant cross-reactive serum antibodies.

If neither of the points can be easily addressed experimentally these caveats should be pointed out in the discussion.

The authors suggest that the macaques are likely CoV naïve, however the data suggest that this may not be true. If the animals are truly naive, then all of the cross-reactive antibodies should have been elicited by the SARS-CoV-2 spike protein and similarly depleted by the SARS-CoV-2 spike protein. For example, in Figure 4b all of the SARS-CoV-2 spike binding antibodies are depleted by the SARS-CoV-2 spike, but only about 70% of the SARS and OC43 binding antibodies were depleted. Do the authors have an alternative explanation for this?