Hamster rules

1 Mar 2011

Chinese hamster ovary and baby hamster kidney cell lines are the best choices for producing therapeutic glycoprotein that stays in the bloodstream longer

Reconstructed fluorescence microscopy image of baby hamster kidney  cells, showing the nucleus (blue), GnT I (red) and sialyl transferase I  (green).

Reconstructed fluorescence microscopy image of baby hamster kidney cells, showing the nucleus (blue), GnT I (red) and sialyl transferase I (green).

Photo by P. Zhang

Glycosylation—the attachment of sugar chains, or glycans, to proteins—is a common chemical modification that cells use to produce functionally important glycoproteins. The glycoprotein erythropoietin (EPO) is of particular therapeutic interest as it regulates red blood cell production, which makes it useful for treating patients with blood disorders.

A glycoprotein may have one or more sialic acids attached to its N-glycans, and the extent of sialylation can have a tremendous impact on the pharmacokinetics of the glycoprotein. Highly sialylated EPO, for example, tends to stay in the bloodstream much longer and have greater efficacy, so scientists have been looking for ways to enhance EPO sialylation. One approach to achieve this is to modify the genetics of cells that produce EPO.

The conventional method for analyzing glycoprotein sialylation is costly and time-consuming. Zhiwei Song and co-workers at the A*STAR Bioprocessing Technology Institute recently adopted a technique called isoelectric focusing (IEF), which allows different molecules to be imaged separately based on differences in electrical charge, to assess the extent of EPO sialylation quickly and accurately. They have used IEF to study how CMP-sialic acid transporter, one of the best-studied nucleotide-sugar transporters, affects EPO sialylation in Chinese hamster ovary (CHO) cells1. They have also used IEF to show that the overexpression of GnT I—the gene that encodes the transferase enzyme GnT I—enhances EPO sialylation in a mutant CHO cell line2. These findings are important and will help pharmaceutical companies develop the strategies to produce highly sialylated EPO. However, the effectiveness of using other genes and other cell lines still remained unclear.

Song and his co-workers have now used IEF to perform a systematic functional analysis on 31 glycosylation-related genes in six cell lines3. They found that CHO and baby hamster kidney cells (pictured) were the most effective in enhancing EPO sialylation. None of the 31 genes, individually or in combination, was able to improve EPO sialylation in these two cell lines. The researchers also tested other cell lines including human embryonic kidney, monkey kidney fibroblast, mouse embryonic fibroblast and mouse myeloma cells, although none of these provided a comparable sialylation enhancement.

“Researchers in biotechnology have tried for many years to improve sialylation by expressing glycosylation-related genes,” says Song. “However, each time only one or two genes have been studied. Our work represents the first systematic analysis of many genes in six commonly used mammalian cell lines, providing a general guide for engineering the glycosylation pathway in these cells.”

The A*STAR-affiliated researchers contributing to this research are from the Bioprocessing Technology Institute.

Want to stay up-to-date with A*STAR’s breakthroughs? Follow us on Twitter and LinkedIn!


  1. Chan, K. F., Zhang, P. & Song, Z. Identification of essential amino acid residues in the hydrophilic loop regions of the CMP-sialic acid transporter and UDP-galactose transporter. Glycobiology 20, 689–701 (2010). | article
  2. Goh, J. S. Y. et al. RCA-I-resistant CHO mutant cells have dysfunctional GnT I and expression of normal GnT I in these mutants enhances sialylation of recombinant erythropoietin. Metabolic Engineering 12, 360–368 (2010). | article | Metabolic engineering: Clearing a hurdle
  3. Zhang, P. et al. A functional analysis of N-glycosylation-related genes on sialylation of recombinant erythropoietin in six commonly used mammalian cell lines. Metabolic Engineering 12, 526–536 (2010). | article

This article was made for A*STAR Research by Nature Research Custom Media, part of Springer Nature