Scaling Up Cell Culture

Courtesy of New Brunswick Scientific WAY BIGGER THAN A T-225: The CelliGen Plus bioreactor with packed-bed basket option. The packed-bed system is fully scaleable, from 500 mL to 150 L. Scientists routinely press eukaryotic cells into service as organic factories, cranking out everything from antibodies to viruses. How much biomass these researchers need to conduct their research, however, varies. Individual researchers can generally get what they need to coat the wells of an ELISA plate, for instance, from a few large flasks' worth of cells. But larger-scale, production efforts require considerably more cells. Because cell culture is so labor-intensive, simply scaling up the number of growth chambers is not really a viable option for these investigators; instead they must look to entirely different cell growth options. The term "large-scale bioreactor" brings to mind visions of enormous bomb-like stainless steel structures, affixed with valves and tubes and industrial-scale controls. Such tanks have been used for cell culture for nearly 50 years, notes chemical engineer Daniella Kranjac, director of sales and marketing for Wave Biotech of Bridgewater, NJ. But these days, biotechnologists have other options, too: everything from roller bottles to spin flasks to slurries of tiny beads. GETTING THE RIGHT FLASK To some labs, moving from a couple of T75s (tissue culture flasks with 75 cm2 growth surface area) to an incubator full of T225s is considered large-scale. Several companies capitalize on that viewpoint, offering products that emulate much of the T flask's technology and methodology. They provide an easy path to scaling up operations, and, best of all, require no additional capital investment. Nalge Nunc International of Rochester, NY, for example, offers a 500-cm2 triple flask, a three-tiered chamber inoculated and fed like a traditional T flask, with the same footprint as a T175. But for even greater growth capacity, the company offers the Cell Factory, a series of 632-cm2 stackable chambers welded together as a single unit, with common fill and vent ports connected by channels to allow liquid and air flow between them. "We have people that scale up from flasks, and then will go to a one-layer cell factory, and then go into a 10-layer," says Heidi McIntosh, product manager for cell culture and cryoware. "And then we have people who are purely doing production and use four 40s [40-stack] at a whack." (Each 40-stack provides more than 25,000 cm2 of growth area.) Corning Life Sciences of Acton, Mass., meanwhile, has taken advantage of Cell Factory patent expirations to introduce a nearly identical product. Corning CELLStack's trays are available in identical configurations and footprint: 636-cm2 cell growth area per tray, in layers of one, two, 10, and 40 trays. The company also will make the CELLStack to custom configurations, says Christopher Hartofelis, product and account manager for bioprocess. He says the principal differences between the two brands are that Corning's product offers screw caps on the ports (rather than a nonventable seal) and has larger openings between layers for better gas exchange. CAPITAL INVESTMENTS For larger-scale culture, some researchers use roller bottles, essentially T flasks whose entire inner surface is available for cell growth. As a special apparatus slowly rotates the bottles, the cells are alternately bathed in medium and exposed to the atmosphere, which in turn allows for a larger usable surface area and increased oxygenation. The associated mixing also prevents gradients from forming in the medium. Most roller apparatuses, whether benchtop, cart-mounted, or automated, can accommodate standard-sized (approximately 1,050 cm2) bottles. Among the largest of these is the R2P Production Roller Apparatus, offered by Wheaton Science Products of Millville, NJ. The R2P can handle 88 double-length bottles, or about 150,000 cm2 of surface area per rack. Some manufacturers offer "expanded surface" roller bottles, which use pleats or other patterns to obtain even more growth area--4,200 cm2 per double-length bottle in the case of Nunc's 5X PETG product, or 369,600 cm2 per rack. These racks can be housed in warm rooms, roll-in incubators, or in units such as Wheaton's Breeder, a fold-up incubator that fits over the R2P. As an alternative to roller bottles, Corning offers its CellCube system, in which 10, 25, 50, or 100 plates are stacked a little over a millimeter apart, and held vertically while media is continuously pumped through. A complete bioreactor system, the CellCube system includes a circulation pump, media pump, controller, oxygenator, and associated tubing and hardware necessary to operate a module containing as many as 100 stacks. The "highly compact, automated system [is] used for mass culturing for mostly anchorage-dependent cells," Hartofelis explains. Corning's Production Cart will run as many as four 100-stack modules, providing 340,000 cm2 of growth area. Courtesy of Wheaton Science Products The Wheaton one deck, R2P Roller Rack System used for growing adherent cells. READY FOR A SPIN The oxygen requirement for animal and plant cells is the greatest impediment to simply letting these cells grow in giant vats. To achieve sufficient air exchange, tanks are constantly agitated, and are either kept small enough for the headspace to provide ample oxygen, or equipped with an aeration device such as a sparger. The simplest tank is a spinner flask, often little more than a wide cylinder with ports or sidearms, in which cells are agitated by a magnetic stir bar. Wheaton's one-liter MantaRay is a sterile, disposable version of the concept. But these systems can get fairly sophisticated, too. Wheaton's Cell Optimizer System, for example, comes complete with a digital control system, pump, heater, stir pad, and magnetic impeller. The Cell Optimizer can monitor and control the culture's pH, dissolved O2, and temperature, and allow as many as four gases to be mixed and delivered to the headspace or a sparger. The system is available in working volumes from 250 ml to three liters. The term "stirred tank reactors" is generally reserved for larger, stand-alone systems with integrated impellers and heating units, and fairly sophisticated controls. Pierre Guerin Technologies Biolafitte, based in Mauze-Sur-Le-Mignon, France, recently installed a 28,000-liter mammalian cell bioreactor in Oss, Netherlands, for pharmaceutical manufacturer Diosynth Biotechnology. But, Pierre Guerin's US technical services manager, Jeff Guertin, points out that the distinction between spinner flask and stirred tank reactor is becoming a semantic one. The company's new Tryton system, marketed as a stirred tank, is available in sizes as small as three liters. At the same time, Wheaton markets the 45-liter Mini-Pilot Plant as a spinner flask; product manager for life sciences Jack Silver considers the Mini-Pilot to be essentially a larger version of the sophisticated Cell Optimizer. These systems differ from bacterial fermenters mainly in the geometry of the vessel and the type of impeller used, Guertin explains. Both animal and plant cells require oxygenation, but animal cells are relatively fragile, so great care must be taken to reduce shear. Some manufacturers solve this problem by rocking the culture instead of stirring it, which is the principle behind Wave Biotech's bioreactor. Cells are incubated in disposable bags, available in working volumes ranging from 100 ml to 500 liters. Inventor and company founder Vijay Singh notes that perhaps the greatest advantage of the Wave Bioreactor is the savings realized in labor costs, as the Wave's disposability obviates the many hours per run typically devoted to cleaning, sterilizing, pumping, and maintaining a stirred tank. In addition he points out revalidation is not necessary in the event of contamination. A STICKY SITUATION A gentle stirring or rocking is ideal for cells that grow in suspension, but many cells can grow only on a substrate. Microcarrier beads and their kin provide one solution the small, sometimes porous particles to which the cells adhere vastly expand the culture's carrying capacity. A new take on this old idea is Nunc's 2-D MicroHex, a "virtually two-dimensional microcarrier for large-scale production of anchorage-dependent cells in bioreactors," according to product literature. These nonporous hexagons are composed of polystyrene, whereas the new Fibra-Cel Disks from New Brunswick Scientific of Edison, NJ, are made from nonwoven polyester fiber and polypropylene, allowing the cells to become trapped in the fiber system. Not just an add-on to a stirred tank, a large number of bioreactors have been designed specifically to take advantage of microcarrier technology. These generally fall into one of two main categories: packed-bed and fluidized-bed reactors. In a packed-bed system, microcarriers are immobilized so they"cannot move with the culture medium. "Cells are shielded from turbulence, allowing higher stirring rates and increased oxygen transfer, resulting in extremely high yields of cells," boasts NBS's Web site about the fibrous-bed basket option of their CelliGen Plus stirred-tank bioreactor. The basket offers the added benefit of keeping the media cell-free, for simplified protein harvesting. Suspension cells, too, are protected from turbulence by the beds. Fluidized-bed reactors come in several varieties, the common thread being that the carriers are free to move with the culture medium. (The simplest of these would be a spinner flask with microcarriers.) One example is the Cytopilot from Vogelbusch of Vienna, Austria, available in laboratory (2-liter) through production (80-liter) scales. The system comprises upper and lower cylindrical chambers separated by a porous plate. A magnetic stirrer drives the medium through both cylinders, with the carriers remaining in the lower chamber. Cell densities up to 2x1011 cells per liter of carrier have been observed, with 1 ml of carrier roughly corresponding to the surface area of a standard roller bottle. CARRIER FIBERS AND MEMBRANES Several systems use porous fibers and membranes to trap cells and larger macromolecules such as secreted proteins, while granting media and its nutrients free access; toxic metabolites are simultaneously diluted or eliminated as well. Among these systems are hollow fiber reactors, available in a variety of configurations, ranging from the diminutive CellMax system (Spectrum Laboratories, Rancho Dominguez, Calif.) and FiberCell Solo (FiberCell Systems, Frederick, Md.), to the several-thousand-liter, multicartridge, custom-built bioreactors. These perfused systems allow "a 3-D culturing," notes Jason Partee, sales manager for CellMax. "A regular cartridge has 2,200 cm2 of surface area for cells to grow on, but you're growing three-to-four deep. You're getting the equivalent of maybe 30-plus T75 flasks in that little cartridge." Such perfused systems, whether the media is recycled, replaced, or both, work for both adherent and suspension cells, "because anything that they excrete is collected inside the small cartridge in a small volume," Partee notes. Integra Biosciences of Chur, Switzerland, manufactures two unique products along similar lines. The Tecnomouse consists of up to five replaceable flat- and hollow-fiber CultureCassettes in an apparatus that independently circulates media and gases through the cassettes. The system, which can produce more than 10 mg of antibody per day, is housed in an incubator (the control unit stays outside), or can be purchased with an optional thermohood. In Integra's CELLine 1000, cells are cultured between a gas-permeable polypropylene-polyethylene membrane and a cellulose membrane having a 10,000-kD molecular weight cutoff, which separates the 15-ml culture chamber from the one-liter nutrient compartment (350-ml and six-well systems are also available). Product literature claims that this static system, which resembles a triple flask, can yield antibody concentrations as high as 5 mg/ml. DOWNSTREAM PROCESSING If the supernatant is already cell-free and highly concentrated with product (as it often is with membrane systems), half of the downstream processing battle has already been won. But for suspension cells cultured in larger batches, supernatants must be pumped out and clarified before they can be subjected to most chromatographic purification procedures. "Both mass and heat transfer are scale-dependent," Singh emphasizes. "Time-dependent phenomena such as degradation, proteolysis, clipping, cleaving--which you would not see in small scale because you can process things so rapidly--become the major problems in large-scale downstream processing." It is thus imperative to accomplish at least these first steps as efficiently as possible. This is usually done by either continuous-flow centrifugation or cross-flow filtration, explains Marc Von Keitz, director or the University of Minnesota's Biotechnology Resource Center. Continuous-flow centrifugation is capable of delivering a cell pellet, he notes, whereas cross-flow filtration is well suited for concentrating cells and recovering supernatant or a protein in it. For truly large-scale operations, however, "the only real solution is a cross-flow filter," says Singh, who until four years ago ran Schering-Plough's biotech operations. He explains that large-scale systems need capacities of 100 liters per minute, a far cry from the 10-15 liters per minute a continuous flow centrifuge might deliver. Researchers considering large-scale cell culture should answer some usage questions, the most crucial of which is whether the cells grow in suspension or are anchorage-dependent, according to Margaret Smith Briggs, former codirector of Core Bio Services at the University of California, San Diego. Investigators should also determine whether the goal is to obtain a secreted product or the cells themselves. Other issues to consider include ease of sterilization and use, need for automation, expertise required, capital and ongoing expenses, physical space required, time and speed of operation, downstream processing needs, scalability of process and validation concerns, single or multiple vessels, and batch or continuous product. Researchers can discuss these issues with any of the many contract manufacturing organizations and university core facilities, or the National Institutes of Health-funded National Cell Culture Center in Minneapolis (www.nccc.com). Josh P. Roberts (tcwriter@msn.com) is a freelance writer in Minneapolis, Minn. function sendData() { document.frm.pathName.value = location.pathname; result = false if (document.frm.score[0].checked) result = true; if (document.frm.score[1].checked) result = true; if (document.frm.score[2].checked) result = true; if (document.frm.score[3].checked) result = true; if (document.frm.score[4].checked) result = true; if (!result) alert("Please select") return result; } .myradio{background-color : #94bee5} Please indicate on a 1 - 5 scale how strongly you would recommend this article to your colleagues? Not recommended 1 2 3 4 5 Highly recommended Please register your vote

Scaling Up Cell Culture

Interested in reading more?

Become a Member of

Courtesy of New Brunswick Scientific
WAY BIGGER THAN A T-225: The CelliGen Plus bioreactor with packed-bed basket option. The packed-bed system is fully scaleable, from 500 mL to 150 L.