Encoding and Estimating the Remarkable Diversity of Possible Sialyltrisaccharides in Nature

Abstract

Glycan microarray development is a critical need for high throughput analysis of glycan‐protein interactions. Over last few years we have collaborated with several groups, and studied the binding of various protein molecules using this powerful technique, with a focus on non‐reducing terminal sialyltrisaccharides found in nature. Our current >200‐glycan microarray library includes a variety of sialoglycans terminated with common sialic acids (Neu5Ac, Neu5Gc, Kdn, and their modified derivatives) which were synthesized chemo‐enzymatically by our collaborators. Printing of these glycans on functionalized microarray slides gives a powerful approach to study glycan‐protein interactions.Considering the ever‐increasing size of our glycan library, microarray data sorting and analysis poses a major hurdle in any high throughput binding study. This situation called for a numerical bar‐coding system that assigns a unique code for individual glycans. Besides allowing for motif searches, one purpose of this system is to easily reorder the glycans in various logical ways during spreadsheet analyses, and another is to plan for further optimization of the printing process. The coding system encompasses the diverse linkage and stereochemistry of glycosyl bonds. Since most of our biological interest is in sialic acids, the first three digits of the code are assigned to describe the non‐reducing terminal sialic acid, its modifications, and the linkage. Underlying monosaccharides are assigned successively from the non‐reducing to the reducing end, each with three digits assigned to describe the monosaccharide, its modifications, and the linkage. Branching in glycan structures may be addressed by grouping digits in parentheses. The number sequence ends with a digit that represents the terminal amine linkers. With this method, we can accommodate all the structural variety that we have so far in our library, and many more. In addition, this system may be capable of creating an output of glycan symbol structures (using the Symbol Nomenclature for Glycans) via relevant programming.Using our coding system, we are able to consider new combinations of biologically‐possible defined terminal sialoglycan trisaccharides on O‐glycans, N‐glycans and gangliosides. In the course of setting up this coding system, we noted that the theoretical population of sialoglycan trisaccharide sequences is more than 205 million. Filtering out the impossible combinations, the number tentatively dropped two orders of magnitude to 1,359,709 (~106 possible combinations) of possible sialyltrisaccharides in nature. While we developed this system for linear trisaccharides, it has not escaped our notice that simply amplifying the calculation to a biantennary N‐glycan with two terminal sialoglycan trisaccharide sequences would result in squaring of the number of possibilities, giving >1012 potential combinations.Support or Funding InformationNIH U01 CA199792This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.