Back

Structural and mutational analyses define distinct molecular routes to broad SARS-CoV-2 receptor-binding domain recognition

Abernathy, M. E.; Foreman, W. B.; Lopez, J. A.; Baharani, V. A.; Vahdat, D.; Lee, Y. E.; Eso, M. R.; Wang, Z.; Bieniasz, P. D.; Nussenzweig, M. C.; Starr, T. N.; Barnes, C. O.

2026-08-21 biophysics
10.64898/2026.08.17.745277 bioRxiv
Show abstract

Broadly reactive antibodies elicited by SARS-CoV-2 infection or vaccination can reveal conserved viral vulnerabilities and inform vaccines with broad coronavirus coverage. Here, we characterize two human-derived monoclonal antibodies, B2014 and C5078, that recognize conserved epitopes on the SARS-CoV-2 RBD and retain activity across antigenically distinct variants. Notably, C5078 also recognizes diverse sarbecoviruses and remains active against currently circulating variants, including XFG and NB.1.8.1. Cryo-EM structures reveal that B2014 recognizes an epitope adjacent to the class 3 antibody site, whereas C5078 targets the highly conserved, cryptic site V epitope. Structural analysis defines how C5078 uses affinity-matured interactions to engage conserved RBD residues, providing a molecular basis for its exceptional breadth. Deep mutational scanning across multiple SARS-CoV-2 variant backgrounds further defines potential pathways of antibody escape, explaining the loss of B2014 activity against antigenically evolved variants while revealing a high barrier to escape from C5078. Together, these findings define distinct structural solutions for broad RBD recognition and highlight conserved, mutationally constrained epitopes that may serve as targets for vaccines designed to elicit antibody responses resilient to ongoing SARS-CoV-2 evolution and future sarbecovirus emergence.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.