With a long standing track record of success, GS-CHO based expression systems have become the most widely used platform for developing and manufacturing recombinant therapeutic proteins today. Ongoing innovation is creating next-generation GS-CHO systems to meet the evolving needs of the biologics industry.
Glutamine synthetase knockout Chinese Hamster Ovary (GS-CHO) cells are one of the leading mammalian cell lines used to produce monoclonal antibodies (mAbs) and other recombinant therapeutic proteins, such as bispecific and Fc fusion proteins.
Many of these therapeutic proteins have shown longstanding success in treating autoimmune diseases and cancer. Even in challenging therapy spaces, such as neurodegenerative disease, therapeutic antibodies have recently been achieving success. As populations age, the demand for broad and effective treatment offerings will only increase. In response, the biomanufacturing industry must build its manufacturing capacity and innovate upon existing platform technologies.
To meet the rising demands on therapeutic protein production, GS-CHO expression system cell lines need to be robust and able to generate higher product yields at manufacturing scales. At the same time, the quality of the final product needs to be strictly controlled.
Table of contents
Approaching the demands of the biologics industry with innovation
Today, drug developers in the biologics industry face unique challenges in developing and manufacturing recombinant proteins. These include working with an increasingly wide range of therapeutic protein formats, while at the same time optimizing titers for reduced cost of goods (COGs).
A key component in addressing these challenges is optimizing the expression vector DNA. Innovative approaches to expression vector design, including the use of novel engineered components, are needed to meet industry challenges head on.
“When GS-CHO expression platforms were developed years ago, it was a simpler therapeutic protein landscape, mostly dominated by mAbs or single-chain proteins;”
“Today, we have to manufacture a much wider range of ever more diverse protein types, such as asymmetric, multi-chain and bispecific antibodies, across all our manufacturing processes, from small scale to commercialization,” O’Callaghan continues.
Developing such diverse modalities often requires GS-CHO cell lines to co-express, fold together, and then secrete multiple protein chains as complex conglomerates. These proteins also require post-translational modifications, such as the addition of extensive carbohydrate structures. Additionally, recombinant protein products need to be expressed at high titers at multiple scales throughout the development journey – from early-stage microplates through to large scale bioreactor production runs during later stages.
Historically, CHO production titers of such proteins were optimized by improving the production media and overall bioprocess, combined with large-scale screening of producer clones. Today, novel, innovative approaches hold potential for further improving the efficacy and control of the GS-CHO platform for synthesizing diverse proteins.
Optimizing expression vectors in novel ways
“A fundamental part of the GS-CHO platform is the expression vector – a set of genetic instructions we give to the CHO cell telling it how to synthesize the recombinant protein of interest,” O’Callaghan elaborates. “When it comes to expression vectors, there’s a lot of optimization we can do, especially considering the constituent genetic elements and their many combinations.”
A key way in which Lonza is improving vector design is through the development of novel promoters. These can allow the expression levels of product genes to be better controlled within the host line for improved titer and product quality. In this approach, differential product gene expression levels are achieved through the respective promoter strengths and individual chain arrangements in the vector.
“Last year, we recently updated our design criteria for making double gene vectors to express mAbs,” states O’Callaghan. “We found that placing the heavy chain in front of the light chain in the vector backbone gives us pretty consistent increases in stable pool titer.”
“For more complex products, such as bispecific antibodies with two heavy and two light chain genes in the vector backbone, we advise customers to screen for the optimal vector design on a per-product basis, by varying the relative gene orders in the vector backbone,” O’Callaghan adds.
“Our approach to vector design optimization is based on empirical testing of variants. So, if we want to optimize expression of a bispecific antibody that has two heavy chains and two light chains, we would directly test vectors containing a range of different gene orders. We screen those variants to find the best combination for that particular protein,” explains O’Callaghan. “The results of this screening work can have a profound influence on titer and product quality.”
“For products that have three constituent protein chains, we may consider changing the gene copy number ratio by duplicating one of the genes,” continues O’Callaghan. “We may also look at different gene orders in the vector, in combination with gene duplication, to see which variant is optimal.”
A subsequent screening analysis identifies the expression vector design that correlates with the highest producing stable pool, along with optimal product assembly.
“We will screen vector variants using CHO-based expression workflows according to the therapeutic product’s position in the drug development pipeline and customers’ needs,” explains O’Callaghan.
“For example, you can screen expression vector variants in parallel in a high-throughput transient gene expression format to get an idea of which vector variant will give you the best titer for a given product,” O’Callaghan elaborates. “In the early stages of a product’s journey, where speed is critical and only small quantities of product are required, this can help save time.”
“On the other hand, if we are looking to support customers who need increased quantities of a therapeutic protein candidate to support toxicology studies, we may perform the vector optimization work on stable pools,” O’Callaghan adds.
“We will screen vector variants in different stable pools, using different relative gene orders. We then look at factors like the final product titer and quality, the assembly characteristics of the material, and the growth of the cells during a fed-batch process,” explains O’Callaghan.
“Extensive analysis of the data can then indicate which vector design is likely to better support commercially viable manufacturing of the product further down the line,” concludes O’Callaghan.
To meet the development needs of diverse protein formats, Lonza’s GS piggyBac® transposase technology can accommodate a large DNA cargo and targets transcriptionally active and genetically stable parts of the host CHO cell genome. With this technology, tailored vector constructs can achieve higher yields compared to conventional vector designs based on random integration, and support faster post-transfection recovery of stable pools.
Cell lines with enhanced performance capabilities
For mAbs that are targeted for killing tumor cells, a cellular mechanism called antibody dependent cell-mediated cytotoxicity (ADCC) can enhance an antibody’s ability to destroy cancer cells.
“We know that antibodies which lack the core N-glycan fucose sugar residue have a stronger affinity with Fcγ receptors of immune system effector cells. This increases the tumor killing effect,” explains O’Callaghan. “And that’s why we developed the GS Effex® cell line – using targeted gene edits, we created a host cell line with the ability to create afucosylated antibodies at high titers.”
Data science in process development
Working with GS-CHO cell lines involves managing multiple processes with efficiency, from vector design to scalability. Data science allows developers to make more informed decisions.
“Machine learning and AI tools can potentially help us de-risk the entire protein expression workflow, and facilitate precise predictions on the output of an intricate biomanufacturing process,” says O’Callaghan.
“Everything from product sequence and expression vector design, through to clone selection and media optimization, has the potential to be further optimized within an agile, platform process informed by the use of prescriptive in silico tools,” O’Callaghan explains. “This will require improved data collection and management coupled to real-time analysis for rapid decision making across the whole manufacturing process.”
The next generation of GS-CHO
“Creating the next generation of GS–CHO cell lines will require an even deeper, robust understanding of current manufacturing processes.”
“At Lonza, we’ve been building out an industry-leading assembly for our GS Xceed® CHOK1SV GS-KO® genome with the addition of further layers of omics data. These include 3D interactome datasets, epigenetic markers, and other genome-wide readouts on expression, using methods such as RNA-Seq and STARR-seq,” O’Callaghan continues.
“So, we can really begin to see how our genome is structured in 3D genome space, and more importantly how it may be regulated,” explains O’Callaghan. “This kind of deeper level understanding of our CHO cell factory will help us design improved expression solutions going forward.”
Leading global innovation
As a global leader in GS-CHO expression systems, Lonza is paving the way for future innovation in the development and manufacturing of biotherapeutics.
Lonza In Your Lab® enables drug developers to take Lonza’s technology into their own lab under license. Alternatively, customers can outsource their therapeutic protein development and manufacturing to Lonza at any stage in the process. This offers clients a broad range of options to develop the biologic treatments of tomorrow.
See how Lonza’s GS® expression systems can benefit your program from drug discovery to commercialization.
Image Courtesy: Lonza Group