Mass Production - Drug Discovery and Development

Bioreactors purposely designed for pharmaceutical protein production in mammalian cell lines sustain higher culture volumes, cell type variation, and more to meet demands of BioPharma.

Bioreactors in an Amgen Process Development lab. (Source: Amgen)

Ever hear the term sour grapes?

Well, the use of this term in a colloquial sense is quite common. But its true meaning is hidden in the fermenter’s lexicon. Fermentation theory really started in the 1850s when Louis Pasteur found that for brewer’s yeast to biochemically convert grape extract to ethanol, the mixture must be free of other microorganisms including lactobacilli, which if present, can convert the alcohol to lactic acid. In other words, the industrial product ethanol had to be prepared in a homogeneous cell culture under sterile conditions.

Much about the manufacturing of industrial products using biological systems has not changed. Homogeneous cultures are still necessary and production still must be conducted under sterile conditions. Nowadays, the products are produced by cells grown in a bioreactor (which provides the sterile conditions) and the mammalian cell is one of the major producers of pharmaceutical products.

“The current state of the art [for the science of mammalian cell-containing bioreactors, in general] is basically doing it in suspension culture or in a stirrer tank bioreactor,” says Shang-Tian Yang, PhD, director, Ohio Bioprocessing Research Consortium, and professor, Department of Chemical and Biomolecular Engineering, Department of Biomedical Engineering, and Department of Food Science and Technology Ohio State Biochemistry Program at The Ohio State University, Columbus, Ohio. “The hydrodynamic environment [inside the bioreactor] can have a profound effect on cell physiology, and therefore, the protein produced in the process. And for many of the protein drugs, glycosylation and other properties that are critical for the drug’s efficacy … can change when the hydrodynamic environment changes.” And that environment may change with the type of bioreactor used. For example, Yang has used the fibrous bed bioreactor (FBB) using a non-woven fiber as the support for a number of different anchorage-dependent cell lines including Chinese hamster ovary (CHO) cells and hybridomas, achieving cell densities of around 1x 108 per liter. This cell density has resulted in high levels of protein productivity for a number of different recombinant proteins and monoclonal antibodies. But these may be the properties of this bioreactor only, and might not be safely generalized to other types of bioreactors.

click to enlarge An SEM picture of CHO cells grown in a fibrous matrix (sample taken from a FBB). In this photo, one can see the high cell density in the matrix. (Source: S.T. Yang, PhD)

Yang is trying to develop a system for producing recombinant proteins, but not for commercial purposes. In fact, he produces recombinant proteins on a laboratory scale in bioreactor culture volumes up to a maximum of 200 ml, which is a very small volume compared to those used for commercial scale production. And although large variation makes it difficult to compare protein quantities yielded from one type of bioreactor to another, Yang says that “in general, with the same cell line in my bioreactor system, I can produce 10 times more [protein] in terms of the productivity and the titer is also much higher than that obtained from the conventional system of suspension culture in a stirrer tank or microcarrier.”

Tipping the scale
Amgen, Seattle, Wash., produces myriad therapeutic proteins for the various phases of development, with the scale of production determined by the project stage and end use. For example, “for discovery and early drug development, [Amgen] mostly uses 10-liter to 100-liter scale fed batch bioreactors and disposable bag bioreactor systems for production,” says Pranhitha Reddy, PhD, scientific director of Cell Sciences & Technology at Amgen. Reddy’s work involves the development and optimization of a stable cell expression system for the production of therapeutic proteins. Specifically, she and her team engineer CHO cell lines to stably express these proteins at high yields and high quality in fed batch bioreactors. Using bioreactors purposely designed for mammalian cell systems enables Amgen to meet the “various demands for therapeutic protein candidates from early development through clinical trials and commercial production.”

Storage tank (in front) and highly equipped 2500-L stirred tank cell culture bioreactor at Rentschler Biotechnologie GmbH. The vessels extend about two levels. Here, the upper part with the cover is seen whereas the major part reaches down to below stairs. (Source: Rentschler Biotechnologie GmbH)

Amgen uses fed batch bioreactors for different levels of protein production. “Usually, one-liter scale fed batch bioreactors are used to develop and optimize our cell culture processes and up to 500-liter scale bioreactors are used to provide material for downstream purification and formulation process development,” says Reddy, who adds that “early pilot scale runs are usually performed at the 2000-liter scale and clinical manufacturing scale can range from 400-liter to 15,000-liter.”

Roland Wagner, PhD, is no stranger to large scale manufacturing of recombinant proteins. In addition to being a professor for biotechnology, Wagner also serves as the senior vice president of development at Rentschler Biotechnologie GmbH, a contract manufacturing organization located in Laupheim, Germany. This company produces recombinant pharmaceutical proteins primarily with CHO cell cultures in pilot scale bioreactors up to the 2500-liter scale, depending on the needs of their client, the pharmaceutical company. Wagner commented on this production process by saying “normally, the [protein] producers are very conservative. They rely on stirrer tank bioreactors (normally made from stainless steel) and on bubble-aerated big tanks, at particular height-to-diameter ratios in the range of three to one. The routinely used equipment allows you to rely on the sterility, such that you have no unexpected events occurring during the day compared to a facility or a bioreactor process that is more exotic.”

Growing Up Bioreactor
In the laboratory of Walid A Farhat, MD, FRCSC, associate professor in the Department of Surgery, Division of Urology, University of Toronto and an associate scientist at the Institute of Medical Sciences, a bioreactor is used to cultivate cells for urinary bladder tissue engineering. Specifically, his research aims to develop methods for seeding cells on a matrix or a scaffold and then to measure the cellular response. Farhat developed a bioreactor that he hopes will simulate normal human urinary bladder function. In this model, cells are placed on a matrix or scaffold where they are subjected to filling pressures similar to those occurring in the bladder. The pressures are created in order to prime the cells in vitro prior to inserting them into an animal model. The objective of the priming is to see if the cells will differentiate, which, according to Farhat, will be “a necessity for tissue engineering purposes in the future.” The research is currently in animal studies, where porcine bladders are being used as a model for development of a system for engineering human tissue. However, cells from human bladders might be available for use in the near future, pending Institutional Review Board approval.

A more recent project in Farhat’s lab aims to directly tackle the issue of dedifferentiation of human cells. In this case, Farhat has observed that when cells that are functioning normally in intact human bladder in vivo are removed and grown in cell culture, they lose their ability to dedifferentiate adequately. Along the same lines, Farhat has also been subjecting bone marrow progenitor cells and stem cells to filling pressures similar to those found in the urinary bladder to see if, when stretched, these stem cells will be stimulated to differentiate. Farhat has performed gene expression analysis using real-time PCR to show that the undifferentiated cells have expressed new bladder-specific cell surface receptors in response to their new environment; these data have been published.

Limitations, challenges
According to Wagner, a major limitation of the mammalian cell-based bioreactor process is its level of protein production. Indeed, the specific quantity of pharmaceutical protein a cell can produce depends on its genetic makeup. “And, [for this reason], there is a demand on the geneticist or cell biologist to improve the cell lines,” he says. Another limitation Wagner points out is that the amount of nutrients and dissolved oxygen one can reach in a given bioreactor limits the maximum cell concentration for that bioreactor, which he says is a limitation of all fermentation processes.

Reddy agrees that the level of protein production is a limiting factor, especially for hard-to-express proteins. But, she says, development of high yielding, robust bioreactor processes for such proteins and “starting with a good host cell line, using stable expression vectors, and isolating recombinant cell lines that grow and perform consistently in bioreactors” have improved Amgen’s success rate.

Finally, among the strengths of his bioreactor system, Yang cites lower cost and higher productivity and efficiency compared to other types of bioreactors such as the hollow fiber. However, he also explains that a major weakness of his system is that it has not been subject to regulatory review by agencies such as the US Food and Drug Administration (FDA). The materials used in constructing the FBB need to be certified as biocompatible and safe before the FBB can be used to produce protein drugs for clinical applications. He explains that complex hydrodynamics of a stirrer-tank bioreactor may change when a manufacturing process is scaled up. So FDA requires a protein pharmaceutical prepared in a bioreactor be prepared the same way every time, allowing consistency in quality from one batch to the next. To ensure consistency, he says, FDA requires continuous quality assurance testing of the protein product. And any change in that manufacturing process must be approved by FDA.

Regardless of cell type, bioreactor type, or scale, the mammalian cell is an important tool for manufacturing many types of pharmaceutical proteins. And as long as they are genetically-endowed to make crucial biological modifications to such proteins, the mammalian cell-based bioreactor is likely to remain a tool of the pharmaceutical industry for years to come.

This article was published in Drug Discovery & Development magazine: Vol. 11, No. 8, August, 2008, pp. 36-38.