Send to printer »

Feature Articles : Oct 1, 2010 (Vol. 30, No. 17)

Protein Production Continues to Post Gains

Latest Advances Demonstrate that Biological Limits Have Yet to Be Achieved
  • K. John Morrow Jr., Ph.D.

There are a variety of ways in which the problem of low protein yields in bioprocessing platforms can be addressed. These include vector design, cell-line selection, and media optimization. In addition, the properties of the product must be taken into account, including folding and secondary modifications, especially glycosylation. Oxford Global Conferences upcoming “Proteins Congress” will profile various improvements in performance, demonstrating that the biological limits of productivity have yet to be reached.

Shire Human Genetic Therapies is developing enzyme-replacement therapies for human genetic disorders and currently has four products  on the market, according to Paolo Martini, Ph.D., director of protein expression and purification research. The company has targeted lysosomal storage diseases.

Shire has two approved therapies in the U.S. for Gaucher and Hunter disease and four therapies, including ones for Fabry disease and hereditary angioedema, in the non-U.S. market.

“In order to optimize protein production, we carried out an analysis of over 500 plasmids to search out the most productive candidates,” Dr. Martini states. “We narrowed our search down to three or four promoters on seven different plasmids that were strong overproducers.” 

“We have a completely natural therapy; the enzymes are produced in human cells,” Dr. Martini continues. “We have found that it is unnecessary to use modifications of the protein molecules in order to obtain optimal performance. The enzymes function in the lysosomes, and modification of their structures appears not to be required.”

Given the plethora of lysosomal storage diseases, it appears that a number of therapies will be forthcoming. “The simplicity of our approach means that this straightforward technology will have broad application for the category of systemic diseases,” Dr. Martini concludes.

Polyclonal Recombinant Antibodies

Symphogen is developing an alternative to monoclonal antibody therapies based on a mixture of polyclonal recombinant antibodies. Such combinations directed against different epitopes on the same antigenic molecule can work together in a synergistic fashion, resulting in an apparent high affinity for the mixture.

“One of our lead projects is Sym001 (rozrolimupab), a polyclonal antibody mixture consisting of 25 different recombinant anti-Rhesus D (RhD) antibodies,” explains Anne Bondgaard Tolstrup, Ph.D., director of antibody expression.

“This is a complex product, one that we are doing in partnership with Swedish Orphan Biovitrum. But in comparison to traditional hyperimmune immunoglobulins (raised in lab animals), it is more consistent and well-defined.”

Sym001 represents an alternative to the conventional anti-RhD hyperimmune immunoglobulins used as therapy for hemolytic disease of the newborn. The development of Sym001 and other products required the preparation of a two-tiered polyclonal master and a working cell bank, by which the chosen antibody clones were mixed, aliquoted, and frozen in a number of ampoules.

For GMP manufacturing, working cell bank ampoules were thawed and expanded for production of the antibody mixture in a single reactor.

Regulatory approval of these complex products has required the development of analytical antibody chemistry procedures to ensure the consistency and quality of the mixtures. Currently Sym001 and Sym004 (an anti-EGFR antibody mixture) are in Phase II and Phase I trials, respectively.

“While some of Symphogen’s early products are highly complex, the more recent efforts tend toward a much simpler antibody mixture,” Dr. Tolstrup explains. “For example, Sym004, designed to inhibit ligand binding, activation, and subsequent receptor signaling, consists of just two antibodies against the epidermal growth factor receptor.”

Sym004 induces rapid and efficient internalization and degradation of the EGF receptor, bringing about the immune-mediated killing of the cancer cells.

Currently, there are two FDA-approved monoclonal antibodies targeting EGFR—cetuximab (Erbitux) and panitumumab (Vectibix). These are used for the treatment of metastatic colorectal cancer, but their costs are extraordinary, running to $10,000 or more per month, and clinical trials have shown only limited extension of life.

“Our preclinical data demonstrate that Sym004 exhibits superior anticancer efficacy in vivo compared to currently marketed anti-EGFR monoclonal antibodies,” Dr. Tolstrup concludes.

Intestinal-Growth Promoting Peptide

Nycomed focuses on pharmacological agents for the treatment of gastroenterological, respiratory, and inflammatory diseases as well as pain, osteoporosis, and tissue management. Dan Bach-Kristensen, Ph.D., a senior pharmaceutical scientist, will discuss his work with an intestinal growth promoting peptide at the meeting.

“Teduglutide is a 33 amino acid long peptide analog of the naturally occurring glucagon-like peptide-2 that we are co-developing with NPS Pharmaceuticals for the treatment of small bowel syndrome,” Dr. Bach-Kristensen states. “The molecule has been modified by replacing an alanine at position 2 with a glycine, which considerably lengthened its in vivo half-life. It is a simple entity with no post-translational modifications.”

According to Dr. Bach-Kristensen, an ordinary adult small intestine has the total surface area of a tennis court, whereas this is restricted to an area the size of a ping pong table in patients suffering from short bowel syndrome. This is usually the result of surgery for infection, inflammation, or as a result of cancer or trauma. The peptide stimulates the elongation of the intestinal villi, thereby increasing the overall surface area of the truncated bowel.

Alternative treatments for short bowel syndrome are limited with no approved products available in the EU. “The FDA has approved the use of human growth hormone for this indication,” notes Dr. Bach-Kristensen, “but the effect is time-limited, and in the majority of trials patients rapidly return to their base-line state.”

Research on teduglutide, which began in the late 1990s, included evaluation of different production processes. Originally the peptide was produced in a secreting mammalian cell line, but Dr. Bach-Kristensen says that production of the peptide by bacterial inclusion bodies is a viable alternative. The main purification challenge is the existence of isomers that can be separated by liquid chromatography and mass spectrometry. Electron-capture dissociation can be used to produce a set of fragments for characterization. In this platform, free radicals are produced that donate electrons to a series of fragments.

“In clinical trials, patients experience increased absorption of nutrients and electrolytes and can see marked improvement in their quality of life,” Dr. Bach-Kristensen explains. “This means that they may be less dependent on parenteral nutrition and may in some cases be moved off of it entirely.”
The studies on teduglutide have advanced to late Phase III trials.

Bispecific Antibodies Against Cancer

According to Melvyn Little, Ph.D., CSO of Affimed Therapeutics, the company is getting ready for clinical trials with its tetravalent, bispecific antibodies. “An IND application was recently filed and approved with the FDA. Patient recruitment will start immediately.”

These developments are a positive sign in the history of the tetravalent, bispecific therapeutic antibody technology (TandAb technology). Although investigators were designing these molecules as far back as the mid 1990s, it has taken years to move to the point at which Affimed is now positioned.

TandAbs are constructed by joining together four antibody variable domains comprising the binding sites of two different antibodies. This four domain chain dimerizes with an identical chain in a head-to-tail fashion. They are expressed as a single polypeptide comprising four variable domains connected via amino-acid linkers varying in length. After translation in the endoplasmatic reticulum, the monomeric polypeptide pairs head-to-tail with another monomer, forming a functional homodimer TandAb® molecule.

In one case, two of the sites are directed against CD3, part of the T-cell receptor, and the other ends of the molecules are directed against a tumor cell antigen. The result is a product that binds extremely tightly to both the T cell and the tumor cell, bringing about a rapid lysis of the target.

In another permutation, the company has developed a product that is highly specific for the CD16A receptor on natural killer (NK) cells; two of its binding sites recruit NK cells and the other two binding sites target a tumor cell. The tight binding between the NK and tumor cells again leads to the demise of the malignant cells.

The two foremost targets under scrutiny are Hodgkin lymphoma, using an anti-CD-30 antibody, and non-Hodgkin lymphoma, in which the antigenic target is CD-19. Affimed also has a daughter company, AbCheck, which is responsible for screening antibody libraries and developing antibodies against new antigens. 

Part of the challenge for Affimed was to develop a robust GMP production process. The investigators began production with a secreting mammalian cell line. “Fortunately, there are no secondary post-translational modifications of the molecules to worry about, and we were able to develop the upstream and downstream purification process with little loss of product,” Dr. Little remarks. Now the firm is interested in producing future products in E. coli by purification from inclusion bodies.

The lead product for clinical use is a lyophilizate in a low-salt formulation, which was found to be extremely stable, retaining activity for at least one year at 25° and even for six months at 40°C. These results are important since they establish that in the future such antibody therapeutics may be shipped at room temperature (or even sit in hot airplane shipping facilities) without loss of potency.

Glycosylation Issues in Purification

The Oxford Protein Production Facility (OPPF) is a structural proteomics facility at Oxford University, funded by the Medical Research Council. The group’s work in dealing with glycosylation issues in protein purification will be reviewed by Joanne Nettleship, Ph.D., senior scientist, at the meeting.

“We’re involved in high-throughput technologies using 96-well plates to clone, express, and purify proteins for U.K. academics,” says Dr. Nettleship, summarizing the overall mission of the facility.

The lab’s small-scale expression protocols allow for fast, effective assessment of expression level from a large number of plasmids in parallel. The level of secreted and intracellular protein expression can be assessed within a week of cloning the gene of interest. “These diverse needs call on the use of human HEK293 cells, E. coli, and baculovirus expression systems.”

Glycosylation is necessary for property folding of many eukaryotic proteins, so these situations require the use of the HEK293 cell line. “For some of our more challenging proteins, we may look at the performance of more than one expression system,” she explains.

Dr. Nettleship favors transient expression, since in a high-throughput laboratory where rapid screening of many constructs is required and repeated expression experiments are not needed, the transient system is less time consuming.

Given that many investigators need homogeneous proteins in order to make crystals for x-ray crystallographic analysis, Dr. Nettleship and her colleagues have optimized approaches that ensure that glycosylation will be uniform through the application of kifunensine, the a-mannosidase I inhibitor. This leaves the proteins in a high mannose state, after which they can be treated with mannosidase to achieve homogeneity of the product.

Dr. Nettleship and her colleagues on the OPPF team set in motion and manage on an ongoing basis the high-throughput crystallization facility. “Using laboratory automation, we have established a protein-production pipeline capable of processing a thousand input sequences per year,” she concludes.

A Changing Landscape

Protein-purification technologies are evolving rapidly both downstream and upstream as a result of a broad mix of new products. Given their inherent nature, biologics will never reach the level of simplicity and consistency of small molecules. Nonetheless, we can anticipate that continuing progress in this dynamic area will  result in many improvements that will make these molecules more economical and more accessible to the public.