Integrative structural analysis of lacto-N-fucopentaose I recognition by antibodies R-17F and R-13E using NMR, docking, and molecular dynamics simulations.

Anti-glycan antibodies have become essential tools in biochemical assays, tissue staining, and diagnostic applications; however, the structural basis of glycan-antibody interactions is poorly understood, limiting rational design of antibodies with enhanced affinity and specificity. Key challenges include the crystallization of glycan-antibody complexes and the acquisition of complete electron density maps of bound glycans. To gain insight into the mechanisms of interaction between antibodies and glycans, here we constructed three-dimensional models of glycan-antibody complexes by integrating NMR spectroscopy with computational simulations. As a representative system, we selected lacto-N-fucopentaose I (LNFP I), which is expressed on undifferentiated induced pluripotent stem cells, and two anti-glycan monoclonal antibodies, R-17F and R-13E. Saturation transfer difference NMR of antibody-glycan solutions indicated that the fucose and N-acetylglucosamine moieties of LNFP I lie in close spatial proximity to the antibody and likely form the interaction surface, with R-17F exhibiting broader contact and more stable binding than R-13E. These data were used as experimental constraints for constructing antibody-glycan complex models through molecular docking and molecular dynamics simulations based on an LNFP I crystal structure and AlphaFold models of the antibody Fv fragment. The resulting models revealed that the non-reducing end of LNFP I is specifically recognized by both R-17F and R-13E, while the reducing end is exposed to the solvent, suggesting compatibility with its native glycolipid anchor. This study provides new structural insights into glycan-antibody recognition and highlights the utility of combining NMR and molecular simulations for the rational design of glycan-targeting antibodies.