Cell Culture and Upstream Processing

Michael Butler

• Publisher: Taylor & Francis
• Year: 2007
• Print ISBN: 9780415399692
• eBook ISBN: 9781134148295
• Topic: Biological Sciences
• Subtopic: Biochemistry
• Index: Biological Sciences

Upstream processing refers to the production of proteins by cells genetically engineered to contain the human gene which will express the protein of interest. The demand for large quantities of specific proteins is increasing the pressure to boost cell culture productivity, and optimizing bioreactor output has become a primary concern for most pharmaceutical companies. Each chapter in Cell Culture and Upstream Processing is taken from presentations at the highly acclaimed IBC conferences as well as meetings of the European Society for Animal Cell Technology (ESACT) and Protein Expression in Animal Cells (PEACe) and describes how to improve yield and optimize the cell culture production process for biopharmaceuticals, by focusing on safety, quality, economics and operability and productivity issues.

Cell Culture and Upstream Processing will appeal to a wide scientific audience, both professional practitioners of animal cell technology as well as students of biochemical engineering or biotechnology in graduate or high level undergraduate courses at university.

Glycosylated proteins

6
Post-translational modifications of recombinant antibody proteins

Roy Jefferis

6.1 Introduction

Following the success of the human genome project, the transcriptome, proteome, glycoproteome, glycome etc. have become foci of interest (Mann and Jenson, 2003). An understanding of human complexity is being sought not in the total number of genes but in the protein products of those genes; and other molecules synthesized through the action of protein products. The transcriptome/proteome exceeds the genome due to differential splicing of nuclear RNA, protein post-translational modifications (PTMs) and protein products generated and/or released in cascade reactions (e.g. coagulation, complement activation etc.). It has been estimated that human identity/integrity depends on the action of 10⁶ individual molecules (http://us.expasy.org/sprot/hpi/; Bauman and Meri, 2004; O’Donovan et al., 2001). Possibly, the most frequent and diverse PTM is glycosylation since it is estimated that ~50% of genes encode for proteins with the potential to bear N-linked glycans, i.e. they express the Asn-X-Ser/Thr motif, where X may be any amino acid except proline (Wong, 2005). Additionally, O-linked glycans add to glycoprotein complexity; however, their potential presence cannot be predicted from gene or protein sequences. Defects in genes contributing to N- and O-linked glycosylation pathways result in congenital disorders of glycosylation (CDG) having serious medical consequences (Butler et al., 2003; Freeze, 2002). Changes in the glycosylation profiles of specific proteins may serve as disease markers (Axford et al., 2003; Gu et al., 1994; Holland et al., 2002, 2006; Ito et al., 1993; Parekh et al., 1985; Takahashi et al., 1987; Youings et al., 1996) whilst the significance of other disease related changes is yet to be elucidated (Poland et al., 2005).

It will be evident that a recombinant protein should, ideally, exhibit the same PTMs as the endogenous protein product. However, it is important to recognize that the structure determined for an endogenous protein is that of a molecule that has had a residence time in a body compartment/fluid prior to being subject to multiple isolation and purification protocols. The structure of this purified product could differ from that of the nascent molecule secreted from its tissue of origin. Similarly, recombinant proteins are synthesized in an ‘alien’ tissue (CHO, NS0 cells etc.), are exposed to the culture medium, products of the host cell line and subject to rigorous downstream and formulation processes. Lack of structural fidelity can impact on function, stability and immunogenicity; an immune response may impact therapeutic efficacy and/or result in harmful reactions (side-effects) (Sinclair and Elliott, 2005; Smalling et al., 2004). Glycosylation and other PTMs have been shown to be species, tissue and gender specific (Davies et al., 2001; Gomord et al., 2005; Hadley et al., 1995; Jefferis, 2005; Raju et al., 2000; Shields et al., 2002; Shinkawa et al., 2003; Umana et al., 1999; van den Nieuwenhof et al., 2000). Currently, protein therapeutics have a shelf life of 18–24 months, which is testament to a lack of structural integrity or suboptimal formulation.

Prokaryotic systems (e.g. E. coli) are unable to produce glycosylated proteins and the aglycosylated protein product may form an inclusion body that has to be extracted, solublized, and refolded in vitro. In contrast, yeast systems add very high mannose structures whilst insect cells add paucimannose structures. Plants may differentially glycosylate proteins but consistently add α-(1-3) fucose and β-(1-2) xylose sugars that are reported to be immunogenic/allergenic in humans (Gomord et al., 2005). Cellular productivity and PTMs are influenced by cell culture conditions, for example temperature, growth rate, media composition (Andersen et al., 2000; Mimura et al., 2001a; Rodriguez et al., 2005), and the addition of butyrate etc. has been shown to increase production and influence the glycoform profile of glycoprotein products (Mimura et al., 2001a). Considerable success has been reported for increased productivity of antibody in CHO cell lines, with levels of 5 g L⁻¹ being achieved and 10 g L⁻¹ being set as a goal (Birch, 2005). However, high production levels may overwhelm the PTM machinery resulting in poor product quality; it is essential, therefore, to characterize a product at an early stage in clone selection to optimize both productivity and quality. Essential nutrients may also compromise product quality, e.g. glycation through the nonenzymatic addition of glucose (Harris, 2005; Lapolla, 2001) or oxidation of methionine side chains (Harris, 2005). The mammalian CHO, NS0 and Sp2/0 cell lines produce an endogenous carboxypeptidase-b that differentially cleaves the C-terminal lysine residues from antibody heavy chains, adding structural and charge heterogeneity (Harris, 2005). Thus, CHO cells may be particularly inappropriate for the production of the complement proteins C3a, C4a etc. that bear functionally significant C-terminal arginine residues; the desArg forms of these proteins having a different profile of activities.

6.2 Common post-translational modifications

Proteins can display an extraordinary range of PTMs (>340) (Walsh and Jefferis, 2006; http://us.expasy.org/sprot/hpi/). The recombinant protein therapeutics currently licensed are either soluble proteins, normally secreted from the cells in which they are biosynthesized, or soluble extracellular domains of membrane proteins. The normal proteolytic processing of pre- and pro-proteins, e.g. removal of N-terminal methionine, signal peptide and fidelity of disulfide bond formation has not presented a problem when produced in mammalian cells. The PTM that has been a major focus of interest has been glycosylation. Whilst it is estimated that ~60% of eukaryotic proteins are phosphorylated and acetylation, these PTMs are mostly confined to proteins mediating intracellular signalling, trafficking and control functions. Such molecules have not been developed as recombinant protein therapeutics, to-date.

6.3 Recombinant antibody therapeutics

The last two sentences of the ground-breaking Kohler and Milstein paper read: ‘Such cells can be grown in vitro in massive cultures to provide specific antibody. Such cultures could be valuable for medical and industrial use.’ (Kohler and Milstein, 1975). Recombinant antibody therapeutics (rMAbs) are predicted to become the largest family of disease-modifying drugs available to clinicians (Tanner, 2005) and the 2006 market value for therapeutic rMAb alone is predicted at $15 billion! Their efficacy results from specificity for a target antigen and biological activities (effector functions) activated by the immune complexes formed. Eighteen rMAbs are currently licensed and hundreds are in clinical trials or under development. The biopharmaceutical industry has met the challenge to produce rMAbs, although productivity, cost, and potency remain to be optimized. All antibody therapeutics currently licensed are produced by mammalian cell culture, utilizing Chinese hamster ovary (CHO), mouse NSO or mouse Sp2/0 cell lines; other systems under development and evaluation, include transgenic animals, yeasts, fungi, plants etc. All rMAbs have shown a potential for immunogenicity whether presented as mouse, chimeric, humanized or fully human sequences. These responses are referred to as human anti-mouse antibody (HAMA), human anti-chimeric antibody (HACA) or human anti-human antibody (HAHA) (Mirik et al., 2004); the promise that fully human antibodies may not be immunogenic has not been realized for Humira (Adalimumab), generated by phage display from a human heavy- and light-chain library, since 12% of patients have been shown to produce anti-Humira antibodies. Such antibody responses will prejudice treatment if they are neutralizing, lead to clearance of the therapeutic or sensitize the patient for severe reactions on re-exposure (Smalling et al., 2004).

The effectiveness of rMAb in oncology depends on sensitizing target cells for subsequent killing by the mechanisms of antibody dependent cellular cytotoxicity (ADCC) and/or complement dependent cytotox...

Table of contents

• Contents
• Contributors
• Abbreviations
• Preface
• Overview on mammalian cell culture
• The producer cell line
• Media development
• Glycosylated proteins
• The Bioprocess
• Index