Engineering a Therapeutic Lectin by Uncoupling Mitogenicity from Antiviral Activity

Abstract

Summary A key effector route of the Sugar Code involves lectins that exert crucial regulatory controls by targeting distinct cellular glycans. We demonstrate that a single amino acid substitution in a banana lectin, replacing histidine 84 with a threonine, significantly reduces its mitogenicity while preserving its broad-spectrum antiviral potency. X-ray crystallography, NMR spectroscopy, and glycocluster assays reveal that loss of mitogenicity is strongly correlated with loss of pi-pi stacking between aromatic amino acids H84 and Y83, which removes a wall separating two carbohydrate binding sites, thus diminishing multivalent interactions. On the other hand, monovalent interactions and antiviral activity are preserved by retaining other wild-type conformational features and possibly through unique contacts involving the T84 side chain. Through such fine-tuning, target selection and downstream effects of a lectin can be modulated so as to knock down one activity while preserving another, thus providing tools for therapeutics and for understanding the Sugar Code.

The Tetrameric Plant Lectin BanLec Neutralizes HIV through Bidentate Binding to Specific Viral Glycans

“…The assembly of BanLec in previous X-ray crystal structures is a dimer and/or an asymmetric tetramer different to that observed for artocarpin and jacalin (Meagher et al, 2005; Sharma and Vijayan, 2011; Singh et al, 2004; Swanson et al, 2015). Using our preparation of BanLec, which we knew to be tetrameric in solution, we solved a crystal structure of BanLec in a new crystallization condition.…”

“…Native MS revealed BanLec is a tetrameric protein (with no other stoichiometries observed), which was unexpected as previous reports suggested it is a dimeric protein (Meagher et al, 2005; Sharma and Vijayan, 2011; Singh et al, 2005; Swanson et al, 2015). To validate our observation, we examined BanLec by size-exclusion chromatography/multi-angle light scattering (SEC-MALS).…”

“…Although BanLec has been previously reported to be a dimeric protein (Meagher et al, 2005; Sharma and Vijayan, 2011; Singh et al, 2005; Swanson et al, 2015), we found it to exist as a tetramer in the mass spectrum (Figure 2A) (discussed further below). Upon incubation with Man9 glycans, we observed BanLec-glycan complexes with between one and four Man9 bound (Figure 1C).…”

“…Crystallographic and molecular dynamics (MD) studies have reported BanLec-binding specificity to disaccharides with α-1,3-mannosyl linkages in both binding sites (Sharma and Vijayan, 2011), and BanLec-HIV recognition is expected via mannose epitopes on high-mannose N-glycans. A binding model was proposed recently whereby each binding site acts independently to engage viral carbohydrates (Swanson et al, 2015), but was not tested with high-mannose N-glycans. Protein-carbohydrate interactions are typically weak, with milli- or micromolar affinity (Weis and Drickamer, 1996), but can be strengthened through multivalent interactions and glycan clustering (e.g., the Env glycan shield).…”

Why Do Membranes of Some Unhealthy Cells Adopt a Cubic Architecture?

“…The H84T variant showed lower multivalent activity but retained monovalent activity and antiviral potency. 48 Until this study, neither BanLec had been used in agglutination experiments with a suitable model of a biological membrane.…”