Engineering Glycosyltransferases into Glycan-binding Proteins Using a Mammalian Surface Display Platform

On July 18, 2025, Sriram Neelamegham and team at the State University of New York, Buffalo published a paper titled "Engineering glycosyltransferases into glycan binding proteins using a mammalian surface display platform" in Nature Communications. This paper successfully engineered a mammalian glycosyltransferase (GT) to function as a highly specific glycan-binding protein (GBP). This work demonstrates a powerful, predictive approach for generating customized tools capable of analyzing Cell-surface Glycosylation patterns with unprecedented specificity.

Overview

Glycans, or complex carbohydrates, are critical components of the cell surface, mediating a vast array of biological processes, including cell-to-cell signaling, immune response, and pathogen recognition. However, the study of these structures has long been hampered by the lack of tools that can specifically distinguish between the enormous structural diversity of glycans (the "sugar code"). Traditional glycan-binding proteins (lectins) often exhibit broad or undefined specificity, leading to ambiguous results. Furthermore, the complexity and low immunogenicity of glycan epitopes make the generation of highly specific anti-glycan antibodies technically challenging. There was a clear need within the synthetic glycobiology community for a rational, modular, and predictable method to design customized high-specificity glycan recognition agents. This paper addresses that need by proposing an enzyme-to-binder engineering principle.

Research Results

• Repurposing GTs via Rational Design

The first key step involved transforming an enzyme, porcine ST3Gal1 (pST3Gal1), which normally transfers sialic acid to a specific sugar chain, into a pure binding agent. GTs possess a natural binding pocket for their glycan substrate, making them ideal scaffolds. Through rational protein design, the team introduced an H302A mutation into the pST3Gal1 active site. Crucially, this single substitution abolished the enzyme's transferase activity while retaining its capacity to bind the target sialylated core-2 O-linked glycan structure (Neu5Acα2-3Galβ1-3[GlcNAc(β1-6)]GalNAcα). This initial engineering effort validated the hypothesis that glycoenzymes could be converted into specific glycan probes.

• Directed Evolution via Mammalian Surface Display

To enhance the binding affinity and specificity of the engineered protein, the researchers developed a novel Mammalian Cell-Surface Display Platform. This platform allowed for high-throughput screening of numerous mutant variants directly on the cell surface, mimicking the native cellular environment where glycan recognition naturally occurs. Using this directed evolution approach, they identified an optimized variant, termed sCore2, which contained three key mutations (H302A/A312I/F313S). The sCore2 mutant demonstrated significantly improved and enhanced binding specificity compared to the initial single-mutant construct, confirming the utility of the display platform for rapid optimization.

Fig.1 H302A displays strict binding specificity for the sialylated core-2 motif. (Hombu, et al., 2025)

• High-Resolution Glycan Mapping in Human Tissues

The functional utility of the newly engineered sCore2 probe was validated across diverse biological samples. Using advanced techniques like spectral flow cytometry on human blood cells and tissue microarray analysis on paraffin-embedded tissue sections, the sCore2 probe revealed distinct and complex cell- and tissue-specific staining patterns. This ability to precisely map the distribution of sialyl core-2 O-linked glycans demonstrates that engineered GT-derived probes can serve as highly valuable reagents for single-cell glycosylation pathway analysis. These results open the door to defining the roles of specific glycan structures in cellular function and disease at a high resolution previously unattainable with traditional tools.

Conclusion

This work definitively establishes a novel and rational strategy for generating specific glycan-binding proteins by protein engineering of GTs. The successful creation of the sCore2 probe, capable of defining sialyl core-2 distribution in human tissues, is a significant advancement. This methodology offers a predictive, template-based route to expand the toolkit of glycan-specific reagents dramatically. The innovative application of enzyme structure as a blueprint for binding specificity suggests that a similar engineering approach can be extended to the large family of other glycoenzymes. Such designer probes will be invaluable in future diagnostics, therapeutic targeting (especially in oncology and immunology), and for fundamentally advancing our understanding of the 'glycome' in the age of single-cell omics.