Periodic Reporting for period 1 - SUGARSmart (SUGARSmart: Smart design of recombinant antibody fragments specific for carbohydrate molecules)
Periodo di rendicontazione: 2018-07-10 al 2020-07-09
Glycosylation is one of the most frequently used post-translational modifications (PTM), where proteins are altered with various glycan molecules. Such glycans are sometimes found only in particular groups, species, or organisms. For example, blood group antigens, Galα1,3Gal motif, and N-glycolylneuraminic acid (NeuGc) are not found in humans, and exposure to these non-self glycans can trigger severe allergic reactions in humans. Thus, it is essential to prevent the presence of any antigenic carbohydrate in biopharmaceuticals, and a majority of them are produced in animal cell lines. Therefore, it is crucial to design molecules that can be used to detect the presence of such antigenic carbohydrates in food or pharmaceuticals. Such biomolecules can be particular proteins called lectins, antibodies, and aptamers. The project “SUGARSmart” looks at the development of computational workflow to engineer glycan recognition molecules and enhance their binding affinity and specificity.
The objective of this Marie Skłodowska Curie Action (MSCA) has been to (1) improve computational methodologies for modeling antibody-glycan recognition and (2) develop a computation workflow for the rational design of glycan recognition molecules. A parallel goal of this MSCA Fellowship is to foster the development of the researcher’s career in the glycosciences.
The objective of this Marie Skłodowska Curie Action (MSCA) has been to (1) improve computational methodologies for modeling antibody-glycan recognition and (2) develop a computation workflow for the rational design of glycan recognition molecules. A parallel goal of this MSCA Fellowship is to foster the development of the researcher’s career in the glycosciences.
During this project, work toward fulfilling these objectives has been carried out to address some of the proposed project's key aims. A computational pipeline to model structures of glycan-binding antibodies using their protein sequences has been established. It further evaluates performance docking programs in terms of their binding affinity prediction capabilities for antibody-glycan complexes. We found that predicting antibody-glycan binding modes and their binding affinity may disagree with state-of-art, high-throughput experimental techniques such as microarray and ELISA, as the glycans are attached to a surface in such experiments, and only the terminal end is exposed for binding. In contrast, in docking computations, glycans are free to adopt any confirmation. This project highlights limitations of the computational approaches in exploring structure, dynamics, and binding energetics of antibody-glycan complexes and discusses future direction to mitigate these limitations.
The scFv antibodies, which do not have a well-defined CDR region described by standard definitions, can be modeled using the proposed workflow. This work has addressed computational approaches' limits in predicting the experimental scFv-glycan binding data where glycans are conjugated to a surface. Important key observations have been made, and we can now approach a computational solution for mimicking experiment like conditions in predicting antibody-glycan binding.