Junctional and allele-specific residues are critical for MERS-CoV neutralization by an exceptionally potent germline-like antibody

Abstract

The MERS-CoV is an emerging virus, which already infected more than 1,300 humans with high (∼36%) mortality. Here, we show that m336, an exceptionally potent human anti-MERS-CoV antibody, is almost germline with only one somatic mutation in the heavy chain. The structure of Fab m336 in complex with the MERS-CoV receptor-binding domain reveals that its IGHV1-69-derived heavy chain provides more than 85% binding surface and that its epitope almost completely overlaps with the receptor-binding site. Analysis of antibodies from 69 healthy humans suggests an important role of the V(D)J recombination-generated junctional and allele-specific residues for achieving high affinity of binding at such low levels of somatic hypermutation. Our results also have important implications for development of vaccine immunogens based on the newly identified m336 epitope as well as for elucidation of mechanisms of neutralization by m336-like antibodies and their elicitation in vivo.

Figure 1. Crystal structures of Fab m336 in complex with MERS-CoV RBD.

(a) Complex structure in cartoon representation. Heavy and light chains of m336 are coloured light green and light blue, respectively. The receptor-binding subdomain on MERS-CoV RBD is coloured orange with Trp535 shown in shticks representation. (b) Sequences of m336 heavy chain and light chain with residues that interact with the MERS-CoV RBD highlighted with orange dots. Antibody residues are numbered according to the Kabat nomenclature. (c) Interaction between CDRs of m336 and the MERS-CoV RBD. m336 CDRs are shown in cartoon representation and the MERS-CoV RBD is depicted in semitransparent surface and cartoon. (d) Detailed interactions between m336 and patches 1 and 2 of the MERS-CoV RBD. Key residues are highlighted in stick representation. Hydrogen bonds and salt bridges are displayed with dotted lines connecting interacting atoms. The non-canonical disulphide bond between HCDR3 residues Cys98 and Cys100c is in yellow sticks. Both panels (c,d) are shown as different 90° views from panel (a).

Figure 2. m336 precisely targets the DPP4-binding site on MERS-CoV RBD.

(a) Mapping of the m336 epitope and DPP4-binding site on MERS-CoV RBD sequence. Residues interacting with DPP4 and m336 are marked with purple and green dots, respectively. MERS-CoV residue numbers are labelled above the sequence. (b) Plot of buried surface areas of MERS-CoV RBD residues that interact with both m336 and DPP4. Residue numbers of MERS-CoV RBD residues are labelled for each data point. (c) Epitope of m336 (green) overlaps with DPP4-binding site (purple) on the MERS-CoV RBD (orange). (d) Comparison of the electrostatic potential surfaces of the MERS-CoV RBD-contacting regions on m336 and DPP4. Molecular surfaces are coloured red for acidic areas and blue for basic areas with the RBD footprints highlighted in orange and olive boundary lines. (e) When superposed on MERS-CoV RBD, the volume of m336 overlaps with that of RBD-bound DPP4 indicating antibody m336 exploits similar angle of approach as receptor DPP4 to engage the MERS-CoV and achieves neutralization by blocking DPP4 binding to MERS-CoV.

Figure 3. Immunogenetic analysis showed a germline heavy chain with a unique junctional motif.

(a) m336 V_H, V_L sequence alignments with germline gene segments are shown. Identical residues are denoted by dots and junction residues at HCDR3 are boxed. Antibody residues are numbered according to the Kabat nomenclature. (b) Close-up view of interactions between the junctional amino acids in the HCDR3 of m336 with MERS-CoV RBD. (c) Amino acid sequences of HCDR3s that were found similar to that of m336 from 454 sequencing analysis of IgM libraries derived from 69 healthy human subjects (denoted by five-letter codes), and from the report by Tang et al. (3B11 and 3B12). Red boxes show the junction amino acid residues in HCDR3s. (d) Germline-rooted circular phylogenetic tree of m336 V_L-like antibody sequences (denoted by five-letter codes, m336 in red and KV1-17*01 in green) found in IgM libraries derived from 69 healthy human subjects and two newborn babies along with the anti-rabies antibody light chain (SC4098, in purple) from the report by Kramer et al..

Figure 4. The binding and neutralization activities.

(a) Representation of the amino acids in m336 that are divergent from those in the predicted germline precursor. None of the five residues in the light chain has any direct contact with the MERS-CoV RBD. (b) Binding of IgG1s m336, m336-gH, m336-gL-FR, m336-gL and a negative control IgG1 m102.4 to MERS-CoV S1 protein measured by ELISA. (c) Neutralization of viruses pseudotyped with the MERS-CoV S glycoprotein. (d) A summary of the binding and neutralization results of m336 and the germline revertants. (e) Binding of Fab m336 allele-specific mutants (F54L and K73E) to MERS-CoV S1 protein measured by ELISA. (f) Binding of Fab m336 junctional mutants (V95A, N100dA, R100eA and G100fA) to MERS-CoV S1 protein measured by ELISA.