Production of glycoprotein therapeutics in Chinese hamster ovary (CHO) cells is limited by the cells' generic capacity for N-glycosylation, and production of glycoproteins with desirable homogeneous glycoforms remains a challenge. We conducted a comprehensive knockout screen of glycosyltransferase genes controlling N-glycosylation in CHO cells and constructed a design matrix that facilitates the generation of desired glycosylation, such as human-like α2,6-linked sialic acid capping. This engineering approach will aid the production of glycoproteins with improved properties and therapeutic potential.

Lysosomal replacement enzymes are essential therapeutic options for rare congenital lysosomal enzyme deficiencies, but enzymes in clinical use are only partially effective due to short circulatory half-life and inefficient biodistribution. Replacement enzymes are primarily taken up by cell surface glycan receptors, and glycan structures influence uptake, biodistribution, and circulation time. It has not been possible to design and systematically study effects of different glycan features. Here we present a comprehensive gene engineering screen in Chinese hamster ovary cells that enables production of lysosomal enzymes with N-glycans custom designed to affect key glycan features guiding cellular uptake and circulation. We demonstrate distinct circulation time and organ distribution of selected glycoforms of α-galactosidase A in a Fabry disease mouse model, and find that an α2-3 sialylated glycoform designed to eliminate uptake by the mannose 6-phosphate and mannose receptors exhibits improved circulation time and targeting to hard-to-reach organs such as heart. The developed design matrix and engineered CHO cell lines enables systematic studies towards improving enzyme replacement therapeutics.

The Chinese hamster ovary cell (CHO) is the major host cell factory for recombinant production of biological therapeutics primarily because of its "human-like" glycosylation features. CHO is used for production of several O-glycoprotein therapeutics including erythropoietin, coagulation factors, and chimeric receptor IgG1-Fc-fusion proteins, however, some O-glycoproteins are not produced efficiently in CHO. We have previously shown that the capacity for O-glycosylation of proteins can be one limiting parameter for production of active proteins in CHO.

Precise gene editing technologies are providing new opportunities to stably engineer host cells for recombinant production of therapeutic glycoproteins with different glycan structures. The glycosylation of recombinant therapeutics has long been a focus for both quality and consistency of products and for optimizing and improving pharmacokinetic properties as well as bioactivity. Structures of glycans on therapeutic glycoproteins are important for circulation, biodistribution and bioactivity. In particular, the latter has been demonstrated for therapeutic IgG1 antibodies where the core α1,6Fucose on the conserved N-glycan at Asn297 have remarkable dampening effects on antibody effector functions. We previously explored precise gene engineering and design options for N-glycosylation in CHO cells, and here we focus on engineering options possible for N-glycans on human IgG1. We demonstrate stable precise gene engineering of rather homogenous biantennary N-glycans with and without galactose (G0F, G2F) as well as the α2,6-linked monosialylated (G2FS1) glycoform. We were unable to introduce substantial disialylated glycoforms. Instead we engineered a novel monoantennary homogeneous N-glycan design with complete α2,6-linked sialic acid capping. All N-glycoforms may be engineered with and without core α1,6Fucose. The stably engineered design options enable production of human IgG antibodies with an array of distinct glycoforms for testing and selection of optimal design for different therapeutic applications.

No abstract