Mucin O-glycan microarrays

The study of mucin O-glycan recognition by glycan-binding proteins has been a challenging area of research at the interface of chemistry and biology. Compared with N-glycans, the development of methods for mucin O-glycans has lagged behind and underrepresentation of O-glycans in any of the current microarray libraries have hampered their application in O-glycan recognition studies. A major reason is that, thus far, there has not been a universal O-glycanase for enzymatic release of O-glycans from mucins. Methods of chemical release result in degradation or modification of the core regions. Therefore, isolated O-glycans have been very limited while chemical/enzymatic synthesis has been slow. As for other types of glycans, a variety of approaches have been developed for construction of arrays using different strategies to overcome the limitation of direct immobilization of glycans onto solid matrices. In this presentation, we overview the current state of play in the construction of O-glycan libraries obtained after their release from mucin glycoproteins and from chemical and chemoenzymatic synthesis for microarray construction using non-covalent and covalent immobilization, and highlight their applications.

Introduction

Mucins are high molecular mass glycoproteins present in mucous secretions and expressed at the epithelial surfaces of gastrointestinal, genitourinary, and respiratory and reproductive tracts, where they shield the surface against chemical and physical damages and protect against infection by pathogens [1]. The mucus layer presents a myriad of potential binding sites for commensal and pathogenic microbes, and also for leucocyte adhesion molecules [2,3^•].

Mucins are heavily glycosylated and contain large amounts (>50% of the total molecular mass) of glycan chains linked via their core sugar residue GalNAc O-linked to Ser/Thr of the peptide backbone. Mucins are rich in Ser and/or Thr and their O-glycans are often present in clustered state along the polypeptide chains [4]. Although the common monosaccharide core is simple, there are eight different mucin O-glycan core types that comprises di-saccharides or trisaccharides (Figure 1). The core structures are extended by linear or branched backbone chains (type 1 or type 2), and are capped with different peripheral sequences which are often variously fucosylated, sialylated and sulphated, comprising a variety of recognition motifs (e.g. the blood-group related antigens) recognized by glycan binding proteins [5].

Understanding of glycan–protein interactions at the molecular level has been of prime interest to glycobiologists and to researchers in diverse fields, including immunology, infection and oncology, and is central to development of glycomic approaches for glycan-based therapies and diagnostics. Because of the weak glycan–protein interactions, their detection and characterization have been difficult. Glycan microarrays have become essential tools for high-sensitivity detection and specificity assignment of glycan–protein interactions and have revolutionized the unravelling of these interactions in health and disease processes. Glycan microarrays have the advantages of multivalent or clustered presentation of glycan probes, and a variety of immobilization strategies have been described for generation of the microarrays [6^••,7,8].

For arraying, glycan probes are immobilized on glass slides by either non-covalent interactions [6^••] or by covalent chemical reactions of the functional groups on the glycan probes with the modified slide surfaces [7]. These two forms of microarrays have been used by the two microarray resources which have the largest glycan libraries and serve the scientific community. The US Consortium for Functional Glycomics (CFG), now the National Center for Functional Glycomics, uses the covalent strategy [7] and the Glycosciences Laboratory at Imperial College (abbreviated here to GL-IC) uses the non-covalent immobilization based on the neoglycolipid (NGL) technology [6^••].