Researchers develop new tools to optimize CHO cell lines for making biologic drugs
Chinese hamster ovary (CHO) cells are the most commonly used cells to produce biologics—protein-based drugs for treating cancers, autoimmune diseases and much more. CHO cells are the workhorses behind more than half of the top-selling biologics on the market today, including Humira, Avastin and Rituxan, to name a few.
Despite their wide use, it has been challenging for researchers to optimize production of biologics from CHO cells. For example, the protein yields from CHO cells are sometimes low—a factor that contributes to the high costs of these pharmaceuticals.
Researchers at the CHO Systems Biology Center at the University of California San Diego are pioneering various efforts to gain an in-depth understanding of CHO cells and to advance cell engineering research. The Center brings together an interdisciplinary team of researchers from the UC San Diego Jacobs School of Engineering, UC San Diego School of Medicine and Sanford Burnham Prebys Medical Discovery Institute.
Boosting product yields
In a Cell Systems paper published last year, Center researchers collaborated with several teams around the world and developed a comprehensive genome-scale model of CHO cell metabolism that identifies specific pathways to maximize protein production. The project was led by Center co-director Nathan Lewis, a professor in the Department of Pediatrics at UC San Diego.
Researchers used the model to predict how much protein CHO cells can actually make when subjected to two treatments commonly used in industry to enhance protein production. One treatment involves lowering the culture temperature, the other involves adding a salt called sodium butyrate to the culture medium.
The model predicted that these treatments barely boost protein production—and do so at the expense of cell growth. "This tradeoff is very inefficient," Lewis said, because the decrease in cell growth is only mildly compensated by a small increase in protein production.
Researchers used the model to simulate other changes to CHO cells. In particular, they studied the efficacy of genetic changes that enhance the flow of the secretory pathway, which is a route by which therapeutic proteins are made and released outside the cell. The model predicted that these cellular changes could increase protein production three times more than the commonly used industrial treatments.
"This finding demonstrates that the secretory pathway is an important pipeline in the cell machinery that we can engineer to make more protein," Lewis said.
In a study published early this year in Scientific Reports, Lewis and colleagues demonstrated a different method to increase protein production—and also increase cell growth rate. The method involved mapping the activity of all the ribosomes in CHO cells as they are producing a therapeutic antibody. In the process, researchers found that silencing one gene not only improved cellular growth, but also resulted in an 18 percent increase in antibody production.
Glycosylation, which is the attachment of sugar chains to proteins, is another cellular process Center researchers are investigating in depth. Controlling glycosylation—which sugars, and how many of them are added to the desired protein—is critical to producing high quality pharmaceutical products, but it is extremely challenging.
In a study published in Biotechnology Journal, researchers developed an algorithm that predicts how researchers can modify CHO cells to obtain desired glycosylation patterns when making biologics and their generic versions, called biosimilars. This work can facilitate efforts to significantly drive down the cost of leading protein drugs, Lewis said.
Center researchers are developing and refining additional state-of-the-art resources to rationally engineer and optimize CHO cell lines for drug development. These include high quality genome sequences of the CHO cell line, next-generation genome editing technologies, a growing library of engineered CHO cell lines, enhanced "clean" CHO cell lines that are free of contaminants, maps of "safe harbor" integration sites—sites in the CHO genome where human genes can be safely inserted to improve protein expression—and sophisticated methods for analyzing and interpreting omic data.
Hooman Hefzi et al. A Consensus Genome-scale Reconstruction of Chinese Hamster Ovary Cell Metabolism, Cell Systems (2016). DOI: 10.1016/j.cels.2016.10.020
Philipp N. Spahn et al. Predictive glycoengineering of biosimilars using a Markov chain glycosylation model, Biotechnology Journal (2017). DOI: 10.1002/biot.201600489