Summary: The JN.1-sublineage KP.3.1.1 recently emerged as the globally prevalent SARS-CoV-2 variant, demonstrating increased infectivity and antibody escape. We investigated how mutations and a deletion in the KP.3.1.1 spike protein (S) affect ACE2 binding and antibody escape. Mass spectrometry revealed a new glycan site at residue N30 and altered glycoforms at neighboring N61. Cryo-EM structures showed that the N30 glycan and rearrangement of adjacent residues did not significantly change the overall spike structure, up-down ratio of the receptor-binding domains (RBDs), or ACE2 binding. Furthermore, a KP.3.1.1 S structure with hACE2 further confirmed an epistatic effect between F456L and Q493E on ACE2 binding. Our analysis shows SARS-CoV-2 variants that emerged after late 2023 are now incorporating reversions to residues found in other sarbecoviruses, including the N30 glycan, Q493E, and others. Overall, these results inform on the structural and functional consequences of the KP.3.1.1 mutations, the current SARS-CoV-2 evolutionary trajectory, and immune evasion.
Antibodies are crucial therapeutics, comprising a substantial portion of approved drugs due to their safety and clinical efficacy. Traditional antibody discovery methods are labor-intensive, limiting scalability and high-throughput analysis. Here, we improved upon our streamlined approach combining structural analysis and bioinformatics to infer heavy and light chain sequences from cryo-EM (cryo–electron microscopy) maps of serum-derived polyclonal antibodies (pAbs) bound to antigens. Using ModelAngelo, an automated structure-building tool, we accelerated pAb sequence determination and identified sequence matches in B cell repertoires via ModelAngelo-derived hidden Markov models (HMMs) associated with pAb structures. Benchmarking against results from a nonhuman primate HIV vaccine trial, our pipeline reduced analysis time from weeks to under a day with higher precision. Validation with murine immune sera from influenza vaccination revealed multiple protective antibodies. This workflow enhances antibody discovery, enabling faster, more accurate mapping of polyclonal responses with broad applications in vaccine development and therapeutic antibody discovery.
Antibodies are crucial therapeutics, comprising a significant portion of approved drugs due to their safety and clinical efficacy. Traditional antibody discovery methods are labor-intensive, limiting scalability and high-throughput analysis. Here, we improved upon our streamlined approach combining structural analysis and bioinformatics to infer heavy and light chain sequences from electron potential maps of serum-derived polyclonal antibodies (pAbs) bound to antigens. Using ModelAngelo, an automated structure-building tool, we accelerated pAb sequence determination and identified sequence matches in B cell repertoires via ModelAngelo derived Hidden Markov Models (HMMs) associated with pAb structures. Benchmarking against results from a non-human primate HIV vaccine trial, our pipeline reduced analysis time from weeks to under a day with higher precision. Validation with murine immune sera from influenza vaccination revealed multiple protective antibodies. This workflow enhances antibody discovery, enabling faster, more accurate mapping of polyclonal responses with broad applications in vaccine development and therapeutic antibody discovery.
Human metapneumovirus is a leading cause of viral respiratory infection in children, and can cause severe lower respiratory infection in infants, the elderly, and immunocompromised patients. However, there remain no licensed vaccines or specific treatments for hMPV infection. Although the hMPV fusion (F) protein is the sole target of neutralizing antibodies, the immunological properties of hMPV F are still poorly understood. To further define the humoral immune response to the hMPV F protein, we isolated two new human monoclonal antibodies (mAbs), MPV458 and MPV465. Both mAbs are neutralizing in vitro and target a unique antigenic site harbored within the trimeric interface of the hMPV F protein. We determined both MPV458 and MPV465 have higher affinity for monomeric hMPV F than trimeric hMPV F. MPV458 was co-crystallized with hMPV F, and the mAb primarily interacts with an alpha helix on the F2 region of the hMPV F protein. Surprisingly, the major epitope for MPV458 lies within the trimeric interface of the hMPV F protein, suggesting significant breathing of the hMPV F protein must occur for hMPV F protein recognition of the novel epitope. In addition, significant glycan interactions were observed with a somatically mutated light chain framework residue. The data presented identifies a novel epitope on the hMPV F protein for structure-based vaccine design, and provides a new mechanism for human antibody neutralization of viral glycoproteins.
Influenza viruses evolve rapidly, driving seasonal epidemics and posing global pandemic threats. While neuraminidase (NA) has emerged as a vaccine target, shared molecular features of NA antibody responses are still not well understood. Here, we describe cryo-electron microscopy structures of the broadly protective human antibody DA03E17, which was previously identified from an H1N1-infected donor, in complex with NA from A/H1N1, A/H3N2, and B/Victoria-lineage viruses. DA03E17 targets the highly conserved NA active site using its long CDR H3, which features a DR (Asp-Arg) motif that engages catalytic residues and mimics sialic acid interactions. We further demonstrate that this motif is conserved among several NA active site-targeting antibodies, indicating a common receptor mimicry strategy. We also identified potential antibody precursors containing this DR motif in all donors of a healthy human donor BCR database, highlighting the prevalence of this motif and its potential as vaccine targeting. Our findings reveal shared molecular features in NA active site-targeting antibodies, offering insights for NA-based universal influenza vaccine design.