Over the past few years, mammalian cell expression systems have gained attention in the biopharmaceutical industry for producing biologics. The current status of glycosylation mechanisms in these systems, coupled with the prevalence of monoclonal antibodies as today’s predominant therapeutic proteins, has significantly shaped the trajectory of biologics development. Therapeutic recombinant glycoproteins, including monoclonal antibodies, exhibit different biological properties due to their varied glycan profiles. Therefore, the development of cell genetic modification strategies aims not only to enhance the cell-specific productivity but also to optimize glycan profile distribution to increase therapeutic activity. In addition, advancements in “omics” technologies provide new possibilities for improving the expression of these aspects, which is of great help to the development of new strategies, especially CHO cells.

Recent estimates value the biologics market at $140 billion, comprising over 200 therapies, with a substantial portion constituting recombinant glycoproteins, increasingly produced in cell-expressed products. This trend underscores a heightened emphasis on post-translational modifications, particularly glycosylation, in biologics production. In fact, Extensive efforts in recent years have been directed towards understanding how glycosylation affects the therapeutic biological activity. Studies indicate that appropriate glycosylation can enhance recombinant protein properties, such as increasing the stability and prolonging blood circulation half-life, while mitigating immunogenicity.

Among mammalian cell expression systems, CHO cells are by far the most commonly used cell line, contributing to over 70% of recombinant biological protein production, primarily monoclonal antibodies. This review focuses on recent advancements in mammalian cell glycoprotein production, with a focus on CHO cells. We outline the various expression systems currently used for therapeutic glycoprotein production and discuss cell engineering strategies aimed at enhancing biologics production and quality optimization. Finally, we highlight recent trials employing diverse “omics” approaches to improve glycoprotein production and glycosylation.

Cell Host

  • CHO Cells

Because the recombinant glycoprotein produced by CHO cells has the same sugar chain as humans, the resulting product is more likely to be compatible and biologically active in human host cells. Additionally, these cells are impervious to interference from human viruses, minimizing biosafety risks for commercial production. This resilience stems from the absence of genes facilitating viral entry into CHO cells. Different gene amplification systems have been developed in CHO cells with high titers and yields. Currently, many biotherapeutic glycoproteins approved by regulatory bodies like EMA are manufactured using CHO cells. These include well-known monoclonal antibodies such as Siltuximab (SYLVANT®), Pertuzumab (PERJETA®) and Rituximab (RITUXAN®), as well as other tissue plasminogen activators (tPa, ACTILYSE®, ACTIVASE®) and human deoxyribonuclease (PULMOZYME®). In fact, more than half of the 13 new biologics approved in 2015 were recombinant glycoproteins produced in CHO cells. Notably, several monoclonal antibodies approved in recent years, such as daratumumab (DARZALEX®), and mepolizumab (NUCALA®), Evolocumab, used to treat conditions like multiple myeloma, asthma, and hypercholesterolemia, are also produced in CHO cells. Despite these advantages, CHO cells do not replicate all human glycosylation types, such as α-2,6-sialylation and α-1,3/4-fucosylation. In addition, polysaccharides produced by CHO cells do not occur in human cells, with certain structures like N-glycolylneuraminic acid (Neu5Gc) and galactose-α-1,3-galactose (α-gal) being either absent or present at very low levels. The human immune system can produce antibodies, and these structures can contribute to the neutralization of the immunogenicity of relevant biotherapeutics. While efforts in metabolic engineering have shown some progress, CHO cells still face limitations in producing certain γ-carboxylate recombinant proteins such as coagulation factors (Kumar 2015). Moreover, proteins requiring proteolysis for maturation may not undergo complete cleavage and activation in CHO cells. For example, industrial-scale co-expression of Furin in CHO cells has demonstrated fully cleaved and active von Willebrand factor and coagulation factor VIII B. Similarly, co-expression of protein precursor convertase and human osteogenic protein-7 has been achieved in CHO cells.

  • Human Cell Lines

One way to produce human-analogous glycosylation is employing human cell lines for recombinant glycoprotein production. While this strategy may not ensure the ideal glycosylation mode, it does yield non-immunogenic polysaccharides. Among the human cell lines commonly used for therapeutic glycoprotein production are HEK293 cells and HT-1080 cells, derived from human kidney and embryonic fibroma, respectively. Drotrecogin alfa (Xigris®), the first therapeutic glycoprotein produced in human cells, gained the EMA approval in 2001 and 2002. However, it was withdrawn from the market in 2011 due to insufficient beneficial therapeutic efficacy. Subsequently, four glycoproteins were approved for treatment, namely glucosidase alpha, erythropoietin (DYNEPO®), Idursulfase (ELAPRASE®) and Velaglucerase alfa (VPRIV®), all produced in HT-1080 cells using gene activation techniques. Notably, Epoeitin delta, produced by HT-1080 cells, exhibited superior homogeneity of tetra-antennary glycans, higher sialic acid, and no tetra-antennary glycans compared to its CHO cell counterpart. However, commercial reasons led to its withdrawal from the market. Velaglucerase alfa displayed a glycan profile akin to products from in other CHO and carrot cells of the same species (CEREZYME® and ELELYSO®). Despite differing glycan profiles, these products demonstrated similar in vitro enzymatic activity, stability, and efficacy, although 24% of patients developed neutralizing anti-imiglucerase antibodies impacting protein activity.

In 2014, a large number of therapeutic proteins produced in human cell lines were approved, including four new glycoproteins approved by the EMA. rFVIIIFc (ALPROLIX®) and rFIXFc (ELOCTATE®) stand out, designed to mitigate bleeding in patients afflicted with hemophilia A and B. They include FVIII and FIX protein domains fused to the Fc portion of IgG1. rFVIIIFc has six tyrosine sulfation positions vital for its functionality, while the carboxylation of the first dodeca-carboxyglutamic acid residue in rFIXFc is pivotal for its activity. Expression of these glycoproteins in HEK293 cells provides better tyrosine sulfation compared to CHO cells, with the added advantage of producing the product without any α-gal and Neu5Gc. Another noteworthy mention is TRULICITY®, a Fc fusion protein approved in 2014 for type 2 diabetes treatment, manufactured by HEK293-EBNA1 cells. Human-clrhFVIII (NUWIQ®), serving as a clotting factor replacement for disabled hemophilia A, garnered approval from the EMA in 2014, respectively. It is synthesized in the HEK293-F cell line, exhibiting a protein profile akin to factor VIII but without α-gal and Neu5Gc.

Currently, various human cell lines are under utilization for recombinant glycoprotein production, undergoing pre-clinical or clinical development phases. PER.C6 cells, containing transformed type 5 E1A and E1B-encoding sequences derived from human embryogenic retinal cells, demonstrate the capability to yield high titers of IgG without the necessity for gene amplification. MOR103, a colony-stimulating factor of monoclonal antibodies targeting granulocyte-macrophages, is being developed to treat rheumatoid arthritis and multiple sclerosis. CL184, a monoclonal antibody used for rabies virus, is undergoing clinical trials phase 1/2 after being produced by PER.C6 cells. The HKB-11 cell line, a fusion of HEK293S and the human B cell line, recently exhibited elevated levels of protein production with α-2,3 and α-2, 6-sialic acid conjugates. Additionally, human amniotic fluid cells and CAP (CEVEC’s Amniocyte Production) cells originating from human liver cancer HuH-7 cells are currently undergoing preclinical evaluation, demonstrating promising results akin to human glycosylation patterns.

  • Other Mammalian Cell Lines (Non-human)

Baby hamster kidney (BHK) cells are mainly used for vaccine production, with only two recombinant proteins currently produced using these cells. Notably, clotting factors such as Factor VIIa (NovoSeven®) and Factor VIII (Kogenate® and Kovaltry®) are among them. These large glycoproteins, rich in sugar groups and disulfide bonds pose significant challenges during production.

Mouse myeloma cells (NS0 and Sp2/0), generated from tumor cells, are not used to produce endogenous immunoglobulins but are harnessed for monoclonal antibody production. Examples include cetuximab (ERBITUX®) and Palivir Co., Ltd. (SYNAGIS®).

  • Non-Mammalian Cell Lines and Other Expression Systems

In the past decade, a notable trend has been the increasing use of mammalian cells for producing recombinant glycoproteins. However, it’s essential to remember that a substantial number of recombinant proteins continue to be generated using other expression systems. These systems lack sufficient glycosylation capacity due to the absence of requisite enzymes, thereby mainly facilitating non-glycosylated expression. Bacterial expression systems boast rapid growth and high yields, yet encounter protein accumulation as there’s no accompanying mechanism for extracting proteins from inclusion bodies before in vitro replication. Notably, commercially available unglycosylated enzymes like asparaginase and collagenase are produced in bacterial expression systems. Yeast expression systems, known for their rapid division and high yields, are also utilized. Therapeutic proteins produced via yeast expression systems include ocriplasmin (JETREA®) and catridecagog (TRETTEN®). Insect and plant cells have the capability to produce recombinant proteins with complex sugar chain structures, albeit different from those found in humans. Plant cell production involves alpha-1, 3-fructose and β-1, 2-xylose, absent in human cells, thereby potentially triggering immune responses. Insect cells undergo modification to yield high-mannose or oligomannose structures before N-glycochain production. Notably, glycoproteins produced in plant and insect cells lack sugar chains on sialic acid residues. Notable treatments produced via insect cell expression systems are the human papillomavirus vaccine (CERVARIX®), prostate cancer immunotherapy (PROVENGE®), and influenza vaccine (FLUBLOK®). Additionally, some therapeutic proteins are produced in transgenic animals, which, like other mammalian cell expression systems, often exhibit glycosylation patterns differing from native human proteins. Notable examples include the production of human antithrombin from genetically modified goats as the first therapeutic drug on the market utilizing genetically modified animals. The C1-esterase inhibitor (Ruconest®) produced from rabbit milk, gained approval from the EMA in 2011. The third product is a genetically modified egg produced with recombinant human lysosomal acidifying lipase approved by the EMA in 2015.

The services and products provided by Creative Biolabs are as follows:

Product Name Type
Tetraglucoside (CAT#: GOS0205S) Oligosaccharides
Sodium mannuronate (CAT#: GMS0389S) Monosaccharides
Rhamnolipid (CAT#: GCS0115S) Carbohydrate-based surfactants

 

Reference: Marie-Eve Lalonde., et al. “Therapeutic glycoprotein production in mammalian cells.” Journal of biotechnology. 2017, 251: 128-140.