Multiplexed engineering glycosyltransferase genes in CHO cells via targeted integration for producing antibodies with diverse complex-type N-glycans

Abstract

Therapeutic antibodies are decorated with complex-type N-glycans that significantly affect their biodistribution and bioactivity. The N-glycan structures on antibodies are incompletely processed in wild-type CHO cells due to their limited glycosylation capacity. To improve N-glycan processing, glycosyltransferase genes have been traditionally overexpressed in CHO cells to engineer the cellular N-glycosylation pathway by using random integration, which is often associated with large clonal variations in gene expression levels. In order to minimize the clonal variations, we used recombinase-mediated-cassette-exchange (RMCE) technology to overexpress a panel of 42 human glycosyltransferase genes to screen their impact on antibody N-linked glycosylation. The bottlenecks in the N-glycosylation pathway were identified and then released by overexpressing single or multiple critical genes. Overexpressing B4GalT1 gene alone in the CHO cells produced antibodies with more than 80% galactosylated bi-antennary N-glycans. Combinatorial overexpression of B4GalT1 and ST6Gal1 produced antibodies containing more than 70% sialylated bi-antennary N-glycans. In addition, antibodies with various tri-antennary N-glycans were obtained for the first time by overexpressing MGAT5 alone or in combination with B4GalT1 and ST6Gal1. The various N-glycan structures and the method for producing them in this work provide opportunities to study the glycan structure-and-function and develop novel recombinant antibodies for addressing different therapeutic applications.

Deep learning explains the biology of branched glycans from single-cell sequencing data

“…The same was found for B4galt1 (beta-1,4-galactosyltransferase 1), the primary enzyme transferring β1,4-galactose to glycans ( Bydlinski et al., 2018 ). B4GALT1 is indispensable for efficient α2,6-sialylation ( Figure 6 A), potentially explaining the decreasing SHAP values upon B4galt1 expression ( Khoder-Agha et al., 2019 ; Nguyen et al., 2021 ).…”

Deep Learning Explains the Biology of Branched Glycans from Single-Cell Sequencing Data

“…The same was found for B4galt1 (beta-1,4-galactosyltransferase 1), the primary enzyme transferring β1,4-galactose to glycans (Bydlinski et al, 2018). B4GALT1 is indispensable for efficient α2,6-sialylation (Figure 6A), potentially explaining the decreasing SHAP values upon B4galt1 expression (Khoder-Agha et al, 2019; Nguyen et al, 2021).…”

Sialylation as an Important Regulator of Antibody Function

“…To increase sialylation, different strategies have been developed, including overexpression of the CMP-sialic acid transporter, overexpression of sialyltransferases, and knock out of genes ( 168 – 171 ). Production cell lines engineered to overexpress trans-Golgi enzymes B4GALT1 and ST6GAL1 have been described to generate IgG with enriched ~70% biantennary sialylated glycoforms ( 169 ). Co-expression of B4GALT1 and ST6GAL1 with IgG increased the amount of sialylation to 32% ( 170 ).…”