Traditional Expression Systems
For decades, expression of recombinant proteins has been done using non-proprietary cell cultures. Commonly used microbial systems are the bacterium Escherichia coli and the yeast Saccharomyces cerevisiae. For expression in mammalian cells, Chinese hamster ovary (CHO) cultures are often used. Microbial cells, being unicellular, are robust in cell culture, and protein production tends to be faster and cheaper in microbial versus mammalian systems. Mammalian cells, although more difficult to culture, have inherent mechanisms for secreting functionally active proteins, which can simplify protein recovery and purification.
Specific proteins can be better suited to one type of expression system over another. The choice of an expression system is influenced by the size, complexity, and glycosylation of an individual protein. Eukaryotic cells (mammalian and yeast) can support expression of larger proteins than bacteria can. Mammalian cells can express and assemble multi-domain proteins, such as monoclonal antibodies and fab fragments, which are increasingly popular as therapeutics. Glycosylation is the addition of sugar motifs to proteins, and it profoundly affects protein function and antigenicity. Patterns of glycosylation are highly species-specific, and mammalian cell lines have glycosylation repertoires most similar to humans.
Protein expression is an empirical process. Even with the most seemingly appropriate cell system, success is not guaranteed. Potential complications lie at every step of the process of protein synthesis and purification. Because proteins share common structural motifs, certain problems, such as the ones list below, are common.
Protein aggregation: Suboptimal conditions during protein expression, purification, or storage can alter the native structure of a protein and result in aggregation. In bacterial cells, which do not have endogenous secretory pathways, expressed protein can accumulate internally and form inclusion body aggregates. Sometimes aggregation can be reversed, but sometimes it cannot.
Proteolytic degredation: Proteases native to the host cell can degrade recombinant proteins. Genetic engineering to remove the host cell’s most active proteases can sometimes minimize the extent of proteolysis.
Inadequate disulfide bond generation: Disulfide bridges stabilize protein structure but can be a challenge to correctly form and maintain. Disulfide bonds are often addressed through trial and error, such as by targeting a specific extracellular excretion pathway or overexpression of chaperones.
Sometimes technical challenges cannot be worked around, such as when expression fails or yield is inadequate. Even when technical pitfalls can be worked around, they increase Resources needed for production. Addressing them often means lengthy purification and refolding regimens, which decrease overall yield.
Proprietary Expression Systems
Proprietary expression systems have been designed to eliminate or minimize some of the common hurdles with traditional microbial and mammalian systems. Several proprietary systems have been engineered with the goals of boosting yield, maintaining protein functionality, and streamlining isolation and purification. The following are examples of proprietary microbial systems.
Pichia Yeast (Research Corp. Technologies): Pichia is a genus of yeast that produces proteins with disulfide bonds and glycosylation. Pichia can be grown cheaply, with methanol as its only source of energy. It can also grow to very high cell densities, which increases the yield per a given volume of cell suspension.
Research Corp. Technologies has two Pichia pastoris systems. Pichia Expression is used for producing non-glycosylated products, or for products in which yeast glycosylation is acceptable. Pichia GlycoSwitch™ is used for producing human-like glycosylated products.
ESETEC E. coli (Wacher Biotech): ESETEC secretes recombinant protein products into the culture broth, which obviates isolation from the periplasm. Extracellular transportation of properly folded proteins occurs via the Secretory (Sec) pathway.
XS Sugar-Inducible E. coli (Lonza): These systems, which are induced by either rhamnose or melibiose, are capable of expression over 10 g/L. Expressed proteins can be recovered from the periplasm or released extracellularly. Correctly folded antibody fragments (including Fabs and scFv) have been produced.
Pseudomonas Bacteria (Pfenex): Pseudomonas fluorescens is a gram-negative bacterium that can be grown inexpensively in a mineral salts medium. In the Pfenex expression system, strain development is done by using a high-throughput, parallel processing model.
Construction and testing of thousands of unique expression strains combining novel gene expression strategies and host cell phenotypes can be done within five weeks. The strategy enables rapid production strains capable of expressing complex proteins at high titers, with yields sometimes more than 20 g/L.
Corynex Cornynebacterium (Ajinomoto): Corynebacterium glutamicum is a gram-positive bacterium, meaning it lacks endotoxins. Endotoxins can be antigenic in humans, and their separation from protein product adds complexity to the purification process. For the Corynex system, C. glutamicum was engineered to express correctly folded, active recombinant proteins directly into the extracellular fermentation broth. Proteins can be targeted for secretion via either the Sec or Twin Arginine Translocase (Tat) pathways.
An example of a proprietary mammalian expression system is PER.C6 (Crucell), an immortalized human cell line. PER.C6 cells grow to high density, which means more product can be harvested from smaller reactors. Product from the cell line has been used in preclinical and clinical studies. The Lonza Group and Boehringer Ingelheim also provide manufacturing of proteins through mammalian cell culture.
Outsourcing and Licensing Traditional Vs. Proprietary
Myriad companies around the globe can express proteins quickly and relatively cheaply in traditional systems such as E. coli. Because generic methodology is applied without optimization, there is no guarantee of expression of an individual construct. In addition, the final product may not be pure. And maintaining confidentiality of the product may be difficult, especially in certain countries.
Expression through a proprietary system can be accomplished by either licensing the technology to use in house or outsourcing production of the protein through the company that developed the system. Each approach has potential advantages and drawbacks.
Outsourcing is typically faster than licensing, as negotiating a license is often a long process. However, contract manufacturing organizations (CMOs) can have ongoing contracts with multiple clients, which can lead to capacity constraints and substantial lead times for production. One advantage of outsourcing is that costs associated with production failures are borne by the company manufacturing the protein. However, a drawback of outsourcing is that it can be difficult to maintain confidentiality of intellectual property. As in any industry, outsourcing can lead to inefficiencies in communication and decision-making.
Licensing typically allows for more control in the production process. Having control is an advantage when the platform is going to be used for multiple projects. Licensing also provides flexibility and facilitates troubleshooting during development and production.
But licensing means tapping in-house resources such as equipment and employee time. At many organizations, budget cuts and layoffs have resulted in a shortage of manpower. In addition, licensing can be costly and involves a substantial investment of time to set contracts in place, and there are no guarantees a platform will produce desired results.
The following questions can help guide the process of deciding between in-house licensing and outsourcing:
- Where are you in the production process?
- How much control do you want to have over production?
- Do you want to devote in-house resources to production and optimization?
- What is your budget and timeframe?
- How much protein product do you need?
Outsourcing is more often done when recombinant proteins are to be used either as reagents or for exploratory research. In these cases, the quantity of protein needed may be relatively low. And when a protein is to be used in exploratory research, control over its production may not be as important as when there is an expectation of profitability.
Proprietary expression systems can fill a niche in producing recombinant proteins that have failed with more traditional systems. Insight into production needs can aid in determining whether to license or outsource alternate expression systems.
Joel White is business manager of Biotechnology at Ajinomoto North America, Inc. He can be reached at email@example.com.