WO1993003135A1 - Culture and fermentation method and apparatus - Google Patents

Abstract

The present invention comprises a novel culture process in which cells or subcellular biocatalysts are confined within a chamber which is capable of being pressurized. The cells are immersed in a nutrient medium with no gas phase in contact with the medium. Cells or subcellular biocatalysts which are cultured according to the present invention show significantly increased yields of cellular products when compared to conventional culturing techniques.

Description

CULTURE AND FERMENTATION METHOD AND APPARATUS

Field of the Invention The present invention relates to an improved method and apparatus for the continuous culture of biocatalysts. More particularly, the present invention relates to a method and apparatus for culturing biocatalysts, such as cells or enzyme systems, under high pressure conditions thereby allowing for the maintenance of biocatalysts at high density with significantly increased yields of cellular products.

Background of the Invention

The term "fermentation" as used herein means any of a group of chemical reactions induced by living or nonliving biocatalysts. The term "culture" as used herein means the suspension of any such biocatalyst in a liquid medium for the purpose of maintaining chemical reactions. The term "biocatalysts" as used herein, includes enzymes, vitamins, enzyme groups, immobilized enzymes, subcellular component, prokaryotic cells and eukaryotic cells. The term "hyperbaric pressure" means any hydraulic pressure greater than atmospheric pressure.

The culture of microbial cells (fermentation) or animal and plant cells (tissue culture) are central to a multiplicity of commercially-important chemical and biochemical production processes. Living cells are employed for such purposes as a result of the fact that living cells, using generally easily obtainable starting materials, can economically synthesize commercially- valuable chemicals. For example, yeast cells can produce large quantities of ethanol (useful for human consumption as beer, wine, or other spirits) if fed solutions of agriculturally-produced sugars under the appropriate conditions, while, in contrast, the de novo synthesis of ethanol by organic synthetic methods is quite expensive. Similarly, starting from simple nutrient molecules, living cells can produce protein molecules of immense commercial value which could not be produced at all by synthetic methods.

Fermentation involves the growth or maintenance of living cells in a nutrient liquid media. In a typical batch fermentation process, the desired microorganism or eukaryotic cell is placed in a defined medium composed of water (usually at least 1000 times the volume of the cells), nutrient chemicals and dissolved gases, and allowed to grow (or multiply) to a desired culture density. The liquid medium must contain all the chemicals which the cells require for their life processes and also should provide the optimal environmental conditions for their continued growth and/or replication.

Currently, a representative microbial cell culture process might utilize either a continuous stirred-tank reactor or a gas-fluidized bed reactor in which the microbe population is suspended in circulating nutrient media. Similarly, in vitro mammalian cell culture employs a suspended culture of cells in roller flasks or, for cells requiring surface attachment, cultures grown to confluence in tissue culture flasks containing nutrient medium above the attached cells. The living cells, so maintained, then metabolically produce the desired product(s) from precursor chemicals introduced into the nutrient mixture. The desired product(s) are either purified from the liquid medium or are extracted from the cells themselves.

Examples of methods employing fermentations of cells growing in either agitated aqueous suspension or with surface attachment are described, for example, in U.S. Patent Nos. 3,450,598; 3,843,454; 4,059,485; 4,166,768; 4,178,209; 4,184,916; 4,413,058; and 4,463,019. Further reference to these and other such conventional cell culturing techniques may be found in such standard texts as Kruse and Patterson, Tissue Culture Methods and Applications. Academic Press, New York, 1977; and Collins and Lyne's, Microbiological Methods. Butterworths, Boston, 1989.

There are a number of disadvantages inherent in such typical fermentation processes. On a commercial scale, such processes require expensive energy expenditures to maintain the large volumes of aqueous solution at the proper temperature for optimal cell viability. In addition, because the metabolic activity of the growing cell population causes decreases in the optimal levels of nutrients in the culture media as well as causing changes in the media pH, the process must be continuously monitored and additions made to control nutrient concentration and pH at optimal levels.

In addition, the optimal conditions under which the desired cell type may be cultured is usually near the optimal conditions for the growth of many other undesirable cells or microorganisms. Extreme care and expense must be taken to initially sterilize and to subsequently exclude undesired cell types from gaining access to the culture medium. Next, such fermentation methods, particularly those employing aerobic organisms, are limited to low yields of product or low rates of product formation as a result of the inability to deliver adequate quantities of dissolved oxygen to the metabolizing organism. Finally, such batch processes can only be operated for a finite time period before the buildup of excreted wastes in the fermentation media require process shutdown followed by system cleanup and resterilization.

The high costs associated with the preparation, sterilization, and temperature control of the large volumes of aqueous nutrient media needed for such cultures has led to the development of a number of processes whereby the desired cell type or enzyme can be immobilized in a much smaller volume through which smaller quantities of nutrient media can be passed. Cell immobilization also allows for a much greater effective density of cell growth and, in theory, results in a much reduced loss of productive cells to output product streams. Thus, methods and processes for the immobilization of living cells are of considerable interest in the development of commercially valuable biotechnologies.

An early method for the immobilization of cells or enzymes involved the entrapment of such biocatalysts on or within dextran, polyacrylamide, nylon, polystyrene, calcium alginate, or agar gel structures. Similarly, the ability of many animal cells to tenaciously adhere to the external surface of spherical polymeric "microcarrier beads" has likewise been exploited for the immobilization of such cells. These gel- or bead-immobilization methods effectively increase the density of the catalyst-containing fraction, thereby effectively trapping these structures in the lower levels of relatively slow-flowing bioreactor chambers. Such gel-entrapment or microcarrier- immobilized methods are taught, for example, in U.S. Patent Nos. 3,717,551; 4,036,693; 4,148,689; 4,189,534; 4,203,801; 4,237,033; 4,237,218; 4,266,032; 4,289,854; 4,293,654; 4,335,215; and 4,898,718. More background information on cell immobilization techniques is discussed in Chibata, et al., "Immobilized Cells in the Preparation of Fine Chemicals", Advances in Biotechnological Processes, Vol. I, A.R. Liss, Inc., New York,

1983. See also Clark and Hirtenstein, Ann. N.Y. Acad. Sci. 369, 33-45 (1981), for more background information on microcarrier culture techniques.

These immobilization methods suffer from a number of drawbacks. First, such entrapment of cells within gels has been shown to interfere with the diffusion of gases (particularly oxygen and carbon dioxide) into and out of the cell environment, resulting in either low cell growth (reduced oxygen input) or gel breakage (high internal CO2 pressure). In addition, the poor mechanical properties and high compressibility of gel-entrapment media lead to unacceptably high pressure problems in packed bed bioreactors. Similarly, the crushing of microcarrier beads and the destruction of attached cells by hydraulic shear forces in agitated tank reactors (necessary to increase gas exchange) leads to reduced viability and productivity.

Another method for the immobilization of living cells or enzymes currently in use involves the use of packed-bed reactors. In these methods, free cells or cells bound to microcarrier beads are suspended in a rigid or semi-rigid matrix which is placed within a culture bioreactor. The matrix possesses interstitial passages for the transport of liquid nutrient media into the reactor, similarly disposed passages for the outflow of liquid media and product chemicals, and similar interstitial passages through which input and output gases may flow. Bioreactors of this type include the vat type, the packed- column type, and the porous ceramic-matrix type bioreactor. Such methods are taught, for example, in U.S. Patent Nos. 4,203,801; 4,220,725; 4,279,753; 4,391,912; 4,442,206; 4,537,860; 4,603,109; 4,693,983; 4,833,083; 4,898,718; and 4,931,401. These methods of immobilization all suffer from a number of problems, particularly when scaled-up to production size. First of all, such bioreactors are subject to concentration gradients. That is, the biocatalysts nearer the input nutrient liquid feed see higher substrate levels than those further downstream. Conversely, those biocatalysts further from the input liquid stream (and closer to the exit liquid port) see increased concentrations of waste products and often suffer suboptimal environmental conditions, such as a changed pH and/or lowered dissolved oxygen tension. Next, such bioreactors are particularly susceptible to the "bleeding" of biocatalysts detached from the matrix (or released by cell division), with the result that output ports become clogged with cells and/or debris. The result is an unacceptable pressure drop across the bioreactor which causes further deterioration of production. Finally, such vertical packed-bed bioreactors in which glass or other microcarrier beads are packed subject the lower portion of the bed to the weight of those beads above, with the inevitable result that both beads and cells are crushed by the sheer weight and number of beads need for production-scale columns. A more recently-developed class of methods for cell immobilization involves the confinement of the desired cells between two synthetic membranes. Typically, one membrane is microporous and hydrophilic and in contact with the aqueous nutrient media, while the opposing membrane is ultraporous and hydrophobic and in contact with a flow of air or an oxygen-enriched gas. Such processes thus provide the cells with an environment in which nutrient liquid input and waste liquid output can occur through channels separate from the cell-containing space and similarly provide gaseous input and output through similarly disposed channels, again separate from the cell-containing space. Embodiments of methods of this class have utilized stacks of many flat membranes forming a multiplicity of cell compartments, have utilized series of synthetic membrane bags, one within the other, and have utilized spirally-wound membrane configurations. Such methods are taught, for example, in U.S. Patent Nos. 3,580,840; 3,843,454; 3,941,662; 3,948,732; 4,225,671; 4,661,455; 4,748,124; 4,764,471; 4,839,292; 4,895,806; and 4,937,196.

Unfortunately, there are a number of problems with such methods, particularly for any commercial, large-scale usage. First, such devices in which a multiplicity of membranes are stacked in series are quite costly to manufacture and are extremely difficult to correctly assemble. Next, the requirement that the membrane which separates the nutrient channels from the immobilized cells be hydrophilic necessarily results in cell attachment across pores and/or pore clogging by insolubles in the nutrient feed or waste output liquids which wet this membrane. The result is the development over time of "dead pockets" where cell growth cannot occur. This situation reduces greatly the effective cell concentration and lowers product yield. Finally, these methods involve devices with a large number of inlet and outlet ports and external fittings which substantially increase both cost and the probability that leakage and contamination will occur.

A final class of methods for cell immobilization involves the employment of capillary hollow fibers (usually configured in elongated bundles of many fibers) having micropores in the fiber walls. Typically, cells are cultured in a closed chamber into which the fiber bundles are placed. Nutrient aqueous solutions flow freely through the capillary lumena and the hydrostatic pressure of this flow results in an outward radial perfusion of the nutrient liquid into the extracapillary space in a gradient beginning at the entry port Similarly, this pressure differential drives an outward flow of "spent" media from the cell chamber back into the capillary lumena by which wastes are removed. Cells grow in the extracapillary space either in free solution or by attachment to the extracapillary walls of the fibers. Typically, oxygen is dissolved into the liquid fraction of the extracapillary space by means of an external reservoir connected to this space via a pump mechanism. Waste products in the intracapillary space may be removed by reverse osmosis in fluid circulated outside of the cell chamber. Such methods are taught, for example, by U.S. Patent Nos. 3,821,087; 3,883,393; 3,997,396; 4,087,327; 4,184,922; 4,201,845; 4,220,725; 4,442,206; 4,722,902; 4,804,628; and 4,894,342. There are a number of difficulties with the use of methods based on capillary hollow fiber cell immobilization methods.

Cracauer et al. (U.S. Patent No. 4,804,628) have extensively documented these difficulties. Among them are: (1) An excessive pressure drop through the fiber assembly. The fragile nature of the fibers results in complete breakdown if fiber of production-scale length is required; (2) Adverse chemical gradients occur within the cell compartment. Gradients of nutrients and waste products often occur in such chambers; (3) Formation of anoxic pockets and discrete microenvironments occur within the cell chamber. Because of the inaccessibility of liquids, gases, and cells to all portions of the fiber bundle as a result of their design, not all areas of the cell chamber are equally effective in cell production; and (4) Mass transfer limitations in nutrient feed and product output increase with time. As cells grow to higher densities, they tend to self-limit the capacities of the hollow fiber chambers (see Col. 1, lines 53-66). The prior art demonstrates that while cell immobilization is a greatly desired method for increasing the productivity of living cells in culture, there are a number of drawbacks associated with each class of method. A central problem of all such culture methods is, as Wrasidlo et al. (U.S. Patent No. 4,937,196) assert, "adequate oxygenation of the cultured cells and removal of carbon dioxide has been a limiting factor in the development of more efficient and economical designs" (see Col. 1, lines 63-65, of U.S. Patent No.

4,937,196).

Living cells are unable to derive any benefit from gaseous oxygen. Living cells derive benefit solely from oxygen dissolved within the aqueous media which surrounds the cells. In batch fermentations which are common for microbial production, the sparging of air or oxygen-enriched gases through the aqueous nutrient media is intended to replace the dissolved oxygen consumed by the metabolizing cells. In this method, most of the gas exits unused while dissolved oxygen levels are maintained. Similarly, the sparging of air (or oxygen) into the nutrient media prior to its use in animal cell culture is intended to maintain a level of dissolved oxygen in the media. While the normal concentration of oxygen in water varies from about 0.2 to 0.3 mM (depending on such factors as pH and ionic strength), it is possible to increase this concentration to as much as 0.5 mM by applying approximately two atmospheres of oxygen pressure over a water solution. To maintain adequate oxygen concentrations in fermentation media, most of the prior art has focused on increasing the contact between gas and liquid by (1) producing a very small bubble size (a function of the sparging frit pore size), (2) by using high speed agitation to increase the rate of oxygen entrance into the hquid phase, or (3) by using a gaseous overpressure of one or two atmospheres above the culture medium to increase dissolved oxygen levels.

In the case of animal cell culture, the typical design of animal cell culture chambers has heretofore made it difficult to consider using overpressures greater than a fraction of an atmosphere. Thus, the most common method for increasing oxygen levels employs gas permeable membranes or fibers in contact with flowing nutrient hquid to maintain dissolved oxygen levels. Such methods are taught, for example, by U.S. Patent Nos. 3,968,035; 4,001,090; 4,169,010; 4,774,187; 4,837,390; 4,833,089; and 4,897,359.

There are a number of problems associated with these methods of increasing the concentration of dissolved oxygen in nutrient media. First and foremost, nearly all of these methods are unable to increase dissolved oxygen concentrations above that obtainable at atmospheric pressure due to the generally fragile nature of other components of the cell culture process. Next, methods which involve vigorous agitation of the liquid-gas mixture to effect increased rates of oxygen dissolution are not applicable to animal cells, which are quite fragile and can easily be damaged by hydraulic shear forces. Finally, those methods which do apply an increased gaseous overpressure above the culture media to increase dissolved oxygen concentrations cannot be scaled up much more than to approximately 1-2 atmospheres of overpressure before it becomes impossible to access the cell-containing liquid media for cell harvest or product isolation without destroying the cultured cells. Nevertheless, the teachings of each of the above methods warrant individual discussion.

U.S. Patent No.4,897,359 (granted to Oakley et al) discloses a method for oxygenating animal cell culture media for subsequent introduction into cell culture vessels in which an oxygenated gas, at an indeterminate pressure, is passed through a multiplicity of gas permeable tubes surrounded by the liquid medium to be oxygenated. While the pressure of the input gas may be above atmospheric pressure, the pressure of the oxygenated exit liquid can be no more than atmospheric pressure. If the oxygenated exit liquid were above atmospheric pressure, it would result in outgassing of the liquid medium when the medium was introduced into the typical cell culture vessel. Such outgassing would also result in bubble formation within the media, which would be extremely deleterious to animal cell viability. Thus, the method of the invention of Oakley et al. is useful only in assuring that the cell culture media possesses the maximum dissolved oxygen concentration obtainable at atmospheric pressure. U.S. Patent No. 4,837,390 (granted to Reneau) discloses a method of preservation of living organs (for subsequent transplant) in which hyperbaric conditions (2 to 15 bars or 29 to 218 pounds per square inch (psi)) are maintained. In the Reneau method, a living organ is placed in a chamber capable of withstanding pressure and a perfusion liquid containing nutrients is pumped into and out of the chamber while a gaseous oxygen overpressure is also applied to the chamber. The method does not discuss cell culture or fermentation.

U.S. Patent No. 4,833,089 (granted to Kojima et al.) discloses a cell culture method in which a gaseous overpressure of oxygen or air is applied over a stirred liquid media in which cells are cultured. In this method, the pressure limitations of the apparatus (which include peristaltic pumps, flexible low-pressure pump tubing, and low pressure filter apparati) necessarily limit the method to overpressures of 0.3 - 0.7 kg/cm^ (approximately 4.3 - 10 psi). Thus, the concentration of dissolved oxygen in the media bathing the cells is limited to values only slightly greater than that obtainable at atmospheric pressure. (Col. 4, lines 15-17)

U.S. Patent No. 4,774,187 (granted to Lehmann) discloses a method for the culture of microbial cells in which a gaseous overpressure is applied over stirred liquid media in which cells are cultured. In this method, the gaseous overpressure makes it impossible to access the interior of the culture compartment without depressurization and cell destruction. The inventor overcomes this problem by raising an overflow line from the media-containing bioreactor to a height such that the liquid pressure of this overflow line equals the gas overpressure. By the establishment of a siphon originating in the elevated overflow vessel connected to the overflow line, one may withdraw liquid or cells from the culture chamber without depressurizing the chamber.

Because the typical culture medium is essentially an aqueous solution, the system pressure is limited to the height of a column of water which would balance the system pressure. Thus, for example, at a system pressure of 37 psi (gauge), a column of water approximately 50 feet in height would be required. Thus, the method from a practical standpoint, is limited to dissolved oxygen levels obtainable at 1 - 2 atmospheres of overpressure.

U.S. Patent No. 4,169,010 (granted to Marwil) discloses a method for improved oxygen utilization during the fermentation of single cell protein in which a gaseous overpressure above a stirred nutrient liquid in a bioreactor containing the growing cells is utilized to increase oxygen delivery to the growing cells. In this method, the recirculation of cell-free media (lean ferment) obtained by centrifugation of the bioreactor contents is passed back into the bioreactor through an absorber section containing a gas contacting zone. The gaseous overpressure is maintained by a gas pressure regulator device which blocks pressure release or vents the gas in response to a desired dissolved oxygen sensor setting. The patent discloses overpressures of about 0.1 to 100 atmospheres (approximately 16.2 to 1485 psi) (Col. 7, lines 28-30, of U.S. Patent No. 4,169,010). The inventor states that a maximum desirable gaseous overpressure of 1 to 2 atmospheres is preferable. Presumably, the reason why a maximum desirable gaseous overpressure of 1 to 2 atmospheres is preferable in the Marwil method, and would be difficult to exceed, arises from the fact that the metabolizing cells also release carbon dioxide, a metabolite which must be removed from the nutrient media by gas evolution if cell viability is to be maintained. Gas overpressures greater than 1 to 2 atmospheres utilized to increase dissolved oxygen content would necessarily result in very large dissolved carbon dioxide levels retained within the nutrient media which could not be removed until the gaseous overpressure was released. It should be noted that carbon dioxide solubility in aqueous solution is approximately an order of magnimde greater than that of oxygen. The inability to remove dissolved carbon dioxide from the media while still delivering increased oxygen to the media would cause a decrease in aqueous pH, a serious problem of the method of this patent. In addition, the method of Marwil is designed solely for the continuous harvest of cells; the method cannot be applied to the continuous harvest of the aqueous solution which might contain an excreted cellular product chemical. U.S. Patent No. 4,001,090 (granted to Kalina) discloses a method for microbial cell culture which incorporates a process for improved oxygen utilization very similar to that outlined above for Marwil (U.S. Patent No. 4,169,010). The method of Kalina directly addresses the problem of carbon dioxide removal mentioned earlier in connection with the method of Marwil. This problem is eliminated by the inclusion of a gas-liquid separator in the fermenter circuit. In the method of Kalina, an oxygenated gas at an unspecified pressure greater than atmospheric is released into the fermentation chamber at its bottom (common sparging). However, by means of a backpressure device, the media is maintained at an overpressure of as much as 3 to 3.5 atmospheres (44.1 to 51.5 psi) to provide both a motive force for the media recirculation as well as to aid in the removal of excess gas distal to the fermentation zone (Col. 4, lines 35-37). The Kalina process relies heavily on the presence of gas bubbles for the agitation of the media and is suitable solely for use in microbial cell fermentation; the method could not be applied to animal cell culture because such cells are extremely sensitive to hydraulic shear forces and are damaged or destroyed by contact with air-water interfaces such as those encountered in gas bubble-containing media,

U.S. Patent No. 3,968,035 (granted to Howe) discloses a method for the "super-oxygenation" of microbial fermentation media in which the common sparging of an oxygen-containing gas into the fermentation media is replaced by the introduction of this gas into an "oxidator" vessel in which high-shear agitation is used to reduce the average size of the gas bubbles, thus increasing the available surface area for gas-liquid contact with the result that maximal dissolved oxygen concentration is maintained. The fermentation media which has thus been treated is pumped into the fermentation reactor while exhausted media from this same source provides the input to the "oxidator" vessel. The process of this invention thus provides a combined liquid and oxygen-enriched gaseous mixture to the culture chamber, a situation which is inapplicable to animal cell culture for the previously-mentioned reasons.

Finally, because the immobilization of cells or microorganisms requires that a cell confinement chamber is part of the process system, the recent literature has been examined for comparison. There are a number of cell culture chambers in existence. Many of these chambers provide for the input and output of a liquid stream, several have viewing ports, and all provide a surface upon which cells may attach or a chamber in which suspended cells may be cultured. Such methods are taught, for example, in U.S. Patent Nos.

3,871,961; 3,753,731; 3,865,695; 3,928,142; 4,195,131; 4,308,351; 4,546,085; 4,667,504; 4,734,372; 4,851,354; and 4,908,319.

In all cases, the operating pressure of these confinement chambers is one atmosphere (or less) and thus these chambers are unsuitable for processes in which increased dissolved oxygen levels are desired and are necessarily limited to those dissolved oxygen levels obtainable at atmospheric pressure.

Summary of the Invention The present invention comprises a novel culture process in which cells or subcellular biocatalysts are confined within a chamber which is capable of being pressurized. The cells are immersed in a nutrient medium with no gas phase in contact with the medium. The chamber has an input port and an exit port through which the nutrient medium is circulated. The exit port in the chamber is preferably blocked by a porous, liquid-permeable structure of defined pore size. According to the present invention, the pores of this structure are smaller than the physical dimensions of the cells or catalysts, yet are large enough to allow free passage of liquids and dissolved nutrient and product molecules. According to the present invention, the cells or biocatalysts are completely confined within this chamber with no gas phase above the medium. Only liquids are passed into and out of the chamber. To cause nutrient liquids to flow through the dense bed of cells or catalysts in the chamber, high pressure pumps are employed to force the nutrient liquid to flow through the chamber. The confined cells or biocatalysts are unexpectedly unaffected by the resultant increase in hydraulic pressure as long as high frequency pressure fluctuations are not present. Thus, fresh, optimal liquid nutrient media is presented to the confined cells or biocatalysts at all times during the process flow while the desired cellular products are immediately accessible at the output of the confinement chamber. It is also contemplated as part of the present invention to supply high concentrations of dissolved gas to the biocatalysts. It has unexpectedly been found that dissolving high concentrations of oxygen in the nutrient medium that bathes the cells in the chamber causes the biocatalysts to produce significantly greater quantities of desired products which can be easily collected as the medium exits the chamber. It is important to note that the gas must be dissolved in the medium and that no gas phase can be present above the medium.

The present invention can be used to produce high yields of industrial chemicals or pharmaceutical products from biocatalysts such as bacteria, yeasts, fungi, and eukaryotic cells or from subcellular organelles, such as mitochondria. These cells can be either naturally occurring or can be genetically manipulated to produce the desired product In addition, the present invention can be used to remove or destroy harmful toxic products via bioremediation in which the toxic chemical is converted into an environmentally benign product

Accordingly, it is an object of the present invention to provide a method and apparatus by which biocatalysts are immobilized within a containment chamber under hyperbaric conditions while nutrient liquids are fed into the chamber and effluent liquids containing desired metabolic product(s) exit the chamber.

It is yet another object of the present invention to provide a method and apparatus by which biocatalysts are immobilized within a containment chamber under hyperbaric conditions while media with toxic chemicals are fed into the chamber and the biocatalysts in the chamber neutralize the toxic chemicals thereby converting them into an environmentally benign product It is a further object of the present invention to provide a method and apparatus by which biocatalysts, including living cell populations, may be immobilized under hyperbaric conditions and either aerobic or anaerobic fermentations performed in which liquid nutrient and substrate nutrients are converted to product-containing output liquid streams.

It is a further object of the present invention to provide a method and apparatus by which bacterial cell populations may be immobilized under hyperbaric conditions and fermentations performed in which hquid nutrient and substrate media are converted to product-containing output liquid streams. It is a further object of the present invention to provide a method and apparatus by which fungal cell populations may be immobilized under hyperbaric conditions and fermentations performed in which hquid nutrient and substrate nutrients are converted to product-containing output hquid streams.

It is a further object of the present invention to provide a method and apparatus by which yeast cell populations may be immobilized under hyperbaric conditions and fermentations performed in which hquid nutrient and substrate nutrients are converted to product-containing output hquid streams.

It is a further object of the present invention to provide a method and apparatus by which eukaryotic animal cell populations may be immobilized under hyperbaric conditions and fermentations performed in which liquid nutrient and substrate nutrients are converted to product-containing output hquid streams.

It is a further object of the present invention to provide a method and apparatus by which either prokaryotic or eukaryotic plant cell populations may be immobihzed under hyperbaric conditions and fermentations performed in which liquid nutrient and substrate nutrients are converted to product- containing output hquid streams.

It is a further object of the present invention to provide a method and apparatus by which enzymes or enzyme systems immobilized on solid supports or catalysts immobilized on solid supports or cell components immobilized on solid supports may be confined and catalyzed chemical conversions be effected under hyperbaric conditions in which liquid substrate nutrients are converted to product-containing output hquid streams.

.Another object of the present invention is to provide a method and apparatus by which dissolved oxygen concentrations (or other dissolved gases) in the nutrient Hquid flow directed into a cell confinement chamber may be raised to any desired level, depending on die applied hydrostatic pressure.

Another object of the present invention is to provide a method and apparatus by .which either a nutrient gaseous substrate (such as oxygen) in the nutrient input liquid flow directed into a cell confinement chamber or an excreted respiratory gas (such as, for example, carbon dioxide) in the output liquid flow may be maintained in the dissolved state until liquid-gas disengagement is desired, generally far downstream of the cell immobihzation compartments). .Another object of the present invention is to provide a method and apparatus by which the conversion of an available chemical substrate into a desired product may be effected by a series of stepwise biocatalyst-mediated conversions in which each chemical conversion step is effected by one of a series of hyperbaric immobilization chambers inserted serially or in parallel into the flow stream.

Mother object of the present invention is to provide a non¬ specific, general method and apparatus for cell culture or fermentation which can be applied to any cell type without significant variation.

Another object of the present invention is to provide a method and apparatus for cell culture or fermentation which significantly reduces both the capital and labor costs of production and production facilities.

Another object of the present invention is to provide a method and apparatus for cell culture or fermentation which is much less susceptible to contamination by opportunistic organisms. other object of the present invention is to provide a method and apparams for cell culture or fermentation in which the liquid environment bathing the desired biocatalyst is essentially invariant in time, i.e., the pH, ionic strength, nutrient concentrations, waste concentrations, or temperature do not vary as a function of time in the biocatalyst's environment Mother object of the present invention is to provide a continuous fermentative or cell culture method.

Another object of the present invention is to provide a method and apparatus for cell culture or fermentation in which cycles of proliferation, growth, or product formation can be accomplished simply by varying the input nutrient feed composition. Another object of the present invention is to provide a method and apparams for cell culture or fermentation which can continue for the lifetime(s) of the immobilized microorganism or cell type.

Mother object of the present invention is to provide a method and apparatus for culturing biocatalysts under hyperbaric conditions which thereby significantly increases the yield of products from the biocatalyst.

Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which increases the conversion efficiency (of substrate to product) of the culture process. Another object of the present invention is to provide a method and apparatus for cell culture or fermentation which significantly reduces the volume of water required to support the culture process.

Another object of the present invention is to provide a method and apparams for cell culture or fermentation which significantly reduces the cost of heating or cooling the aqueous media required to support the culture process.

Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as antibiotics from bacterial fermentations. Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as enzymes or other proteins from fungal fermentations.

Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as ethanol or other short-chain alcohols and acids from the fermentation of microorganisms.

Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as protein hormones from genetically-transformed microorganisms.

Another object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as protein hormones from eukaryotic cells.

Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as amino acids, nitrogenous bases, or alkaloids from the fermentation of microorganisms.

Mother object of the present invention is to provide a method and apparams for cell culture or fermentation which results in higher yields of products such as fuel-grade ethanol from the fermentation by yeasts of sugar- containing agricdtural material.

Mother object of the present invention is to provide a method and apparams which would reduce the fermentation time required to produce alcoholic beverages such as beer and wine. Mother object of the present invention is to provide an easily scaled-up method and apparams for cell culture or fermentation which can be commercially employed.

These and other objects, features and advantages of the present invention will become apparent after a review of the following detailed description of the disclosed embodiment and the appended claims.

Brief Description of the Figures

The accompanying figures, which are incorporated and form a part of the specification, illustrate several embodiments of the present invention and, together with the description, serve to explain the principles of the invention.

Fig. 1 is an illustration in block form the basic hyperbaric process of the present invention.

Fig. 2 is an illustration in block form one embodiment of the basic aerobic hyperbaric process.

Fig. 3 is an illustration in pictorial form an embodiment of the basic aerobic hyperbaric process.

Fig. 4 is an illustration one embodiment of the hyperbaric confinement chamber. Fig 5. is a view of the inside of the two halves of the hyperbaric confinement chamber shown in Fig.4.

Fig. 6 is an illustration of an alternative embodiment of the hyperbaric confinement chamber employing hollow-fiber technology.

Fig.7 illustrates in block form an alternative embodiment of the hyperbaric method which includes auxiliary control and sensing mechanisms as well as two serially-connected hyperbaric confinement chambers. Fig. 8 is a representation of the data obtained in the experiments of Example I.

Fig. 9 is an illustration of one embodiment of the hyperbaric confinement chamber.

Detailed Description

The present invention comprises a novel culture process comprising cells or subcellular biocatalysts which are confined within a chamber which is capable of being pressurized. According to the present invention, there is no gas phase in contact with the cells or biocatalysts. The exit port of the chamber is blocked by a porous, liquid-permeable structure of defined pore size. Optionally, the input port may also be blocked by a porous, liquid-permeable structure of defined pore size. According to the present invention, the pores of this structure are smaller than the physical dimensions of the cells or catalysts, yet are large enough to allow free passage of liquids and dissolved nutrient and product molecules.

The cells or catalysts are completely confined within the pressurized chamber while liquids may be passed into and out of the chamber. To cause nutrient liquids to flow through a dense bed of cells or catalysts in the chamber, high pressure pumps are employed to force the nutrient liquid to flow through the chamber. The confined cells or biocatalysts are unaffected by the resultant increase in hydrauhc pressure as long as high frequency pressure fluctuations are not present Thus, fresh liquid medium with optimal nutrients is presented to the confined cells or biocatalysts at all times during the process flow while the desired cell products are immediately accessible at the output of the confinement chamber.

There are a number of advantages of this method of culturing cells or biocatalysts in comparison to those taught by the prior art. The range of application of the present invention is quite broad. It can be employed without significant change to any type of biocatalyst or cell culture regimen, ranging from the culture of microbial cells, to the culture of animal and plant cells, subcellular organelles or components, or bead-immobilized enzymes. Heretofore production facilities have had to be tailored to the metabolic requirements of each microorganism or cell type. Furthermore, the current inability to continuously culture delicate mammalian cells in commercially- significant quantities has resulted in the need to genetically manipulate more amenable microorganisms into producing mammalian protein products.

The fermentation and culture apparams of the present invention is significantly simpler and less expensive than that of the prior art The ability to confine cells or other biocatalysts to a chamber in which virtually all of the available space is occupied by the cells or biocatalysts results in a considerable economic advantage. The volumes of hquid required are reduced to as little as 1% of the water required in prior art fermentation apparati.

The overall cost of the fermentation process machinery is greatly reduced because the process is essentially a pumped liquid line with the biocatalyst chamber interposed in the flow stream. The production costs associated with the normal heating and cooling of the large volumes of water used in typical fermentation processes (often 50,000 gallons or more) are not required in the process and apparams according to the present invention. All that is necessary in the method of the present invention is to maintain temperature control of the immobilization chamber itself .

Because the system is a closed liquid flow line under pressure, there is no place in the process stream where a contaminant could be introduced. The process system can be easily sterilized by conventional means and thus the only source of contamination is the input hquid reservoir. The input liquid reservoir can be sterilized by conventional means. Such a reduction in the possible sources of contamination conveys a large economic advantage.

The process of the present invention reduces the costs associated with the maintenance of system variables such as pH, nutrient concentration, temperature control, and the buildup of waste products in the environment of the working biocatalysts. According to the present invention, the confined biocatalysts "see" only an optimal hquid environment introduced via the input pump. In conventional batch fermentation, the working cells quite often are limited to short batch run times because of the inability to remove waste products which cause changes in the system variables.

The method and apparams according to the present invention is a continuous fermentation or culture process. That is, once the process is begun, there is never a need to shut down the process until the practical lifespan of the immobilized microorganism or cell is exceeded. In the case of microorganisms capable of self-replication, changes in the input nutrient flow composition can be used to trigger alternate cycles of growth or production, thereby reducing the need to shut down the process flow. In the case of mammalian cell culture, the process flow can continue for the lifetime of the immobihzed cells. Prior art methods of fermentation are believed to have hfe spans which are considerably shorter than the present invention. It is to be understood that the present invention also includes the

"chaining together" of pressurized chambers containing several microorganisms or cell types to arrive at a desired chemical product through several intermediate conversion steps. Thus, for example, one might utilize a particular fungus confined in one chamber to convert compound A to compound B, feed the output of this chamber to a second chamber where a yeast converts compound

B to compound C, and finally, feed the output of this second chamber to a third chamber where a bacterium converts compound C to compound D, the final desired product The chambers can be arranged either in parallel or in series.

The fact that the cells or catalysts are bathed in a liquid at pressures greater than one atmosphere according to the present invention allows for the complete control of dissolved gas concentration. This is quite important in the case of aerobic organisms (those requiring oxygen for viability). Such cells perform oxidative chemical conversions to produce a desired product. The efficiency of product formation is inhibited in prior art fermentation systems simply by the inability of currently available fermentation apparati to deliver adequately dissolved oxygen to the cells to allow fully efficient function. This is a drawback to prior art aerobic fermentation methods.

In the hyperbaric fermentation process of the present invention, the amount of dissolved oxygen (or any other gas) which can be delivered to the cells is a function of the applied pressure. A corollary of this effect is that the carbon dioxide (or other excretory gases) liberated by die respiring cells remains in solution under hyperbaric conditions and is removed r om the environment of the immobihzed biocatalysts by the process liquid flow. Other gases that may be used in accordance with the present invention include, but are not limited to, air, 02, NH3, NO2, He, Ar, N2 and H2 or a mixture thereof. The gases helium, argon and nitrogen are particularly important in anaerobic systems in that these gases are generally inert and do not effect cell function and can be used to replace oxygen in the medium. The selection of the gas to be used in accordance with the present invention will depend upon the particular biocatalyst being used in the chamber. The present invention significantly increases the product yield of microbial fermentations or cell cultures. As noted earlier, because the method is a continuous one, production can continue for the effective lifetime of the immobilized biocatalyst This fact alone accounts for a sizable increase in productivity. In addition, it has been discovered that in the case of aerobic organisms, an increase of more than 100% in the rate of product formation may be observed when dissolved oxygen levels are raised above normal. This effect can be observed even when there is no apparent direct chemical connection between oxygen availabiHty and product formation. It is believed that this effect is related to basic metabolic rate increases.

The employment of an apparatus incorporating the method of the present invention as a research tool for the investigation of, for example, the optimal conditions for antibiotic production by a particular microorgamsm, is a great aid in microbiological or mammalian cell research. Currently, the process of determining the optimal conditions for microbial, plant or animal cell expression of a particular product is very laborious. Variously composed formulations are tried, one by one, either in small-scale batch fermentations or in petri dish trials, to arrive at some optimal formulation which maximizes product formation. Using an apparams incorporating the method of the present invention, one can instantly perturb the input hquid feed to the immobihzed microorganism or cell (by injecting the test chemical into the hquid flow) and instantly observe the effect of this perturbation in the output liquid flow. A rapid series of such perturbations and analyses greatly speeds such an optimization procedure. A list of the specialty chemicals and pharmaceuticals which are produced by microbial or cellular fermentations is quite large, and includes such examples as the penicillin drugs, a host of oti er antibiotic drugs, fuel alcohol, and such foodstuffs as cheeses, yogurts, and alcoholic beverages. The recent explosion of research into the molecular biology of the gene has resulted in the development of a new field, called "genetic engineering", in which genes coding for a desired product are located, excised, and introduced into a well- behaved microorganism which then produces the desired product in high yield. Examples of such a strategy include the economic production of insulin, growth hormone, and interferon for medical applications. A microbial fermentation or a culture of eukaryotic cells (termed

"cell or tissue culture") are quite similar. The desired organism or cell is placed in a defined medium composed of water (usually approximately 1000 times the volume of the cells) and nutrient chemicals, and allowed to grow (or multiply) to some culture density. The living cells then produce d e desired product from precursor chemicals introduced into die nutrient mixture. The desired product is then purified from of the hquid medium.

There are a number of disadvantages inherent in these types of prior art fermentations. First of all, on any scale other than that of the laboratory, the heating or cooling of large volumes of water (necessary to maintain cell viabihty), the need to constandy adjust its pH (also important for cell viabihty), and the need to constantly adjust the concentration of nutrients (as they are consumed by the cells) adds significantly to the cost of production. Furthermore, the generally dilute nature of the desired product witiiin such large volumes of aqueous media requires considerable work and cost for its isolation, concentration, and purification. Most importantly, any attempt to adapt fermentation processes into continuous production results in the loss of the cell population to the out-flowing product stream. The inability to inexpensively confine the working cells in a production vessel while the nutrient medium is passed into the vessel and die product is passed out of the vessel has generally necessitated tfiat such fermentations are typically performed in a "batch" mode. A fermentation is set up, run for a prescribed period, the process stopped, d e contents fractionated and die product(s) isolated, and die production facility cleaned and set up for the next batch. Generally, batch fermentation is a more expensive method man a continuous process.

The high costs associated with typical microbial batch fermentations as outlined hereinabove has long been recognized by biochemical and microbiological researchers. Over the past 15 years, a number of methods for the immobilization of cells have been developed. Cells (i.e., animal or microbial cells) have been encapsulated in various types of gels, have been grown in porous beads, and most recendy, have been cultured on hollow fibers which are immersed in the nutrient mixture. These strategies are designed to immobilize the cells of interest while allowing flows of nutrients to pass over the immobile structures and die desired products leach out of die cells into the output liquid flow. Unfortunately all of these earlier attempts at cell immobilization have inherent flaws which severely limit their commercial application. The present invention is a novel cell immobilization meti od which is named for me purposes herein "hyperbaric fermentation". In me process according to die present invention, typically cells are confined in a chamber which is capable of being pressurized. The chamber with the cells is gradually pressurized d ereby raising the hydrostatic pressure witiiin the cell bed until an optimal pressure level is reached. Because hydrostatic pressures are rapidly transmitted tiirough cell membranes, die internal cell pressure equals the externally-applied pressure, and die cells "see" no pressure gradient across tiieir boundary membranes.

The cell confinement chamber possesses an exit port which is blocked by a porous, liquid-permeable structure of defined pore size. Hereinafter, the porous, hquid-permeable structure of defined pore size is called the exit frit, although it is to be understood that in some embodiments it incorporates a membrane above a support grid. The pores of this structure are necessarily smaller than the dimensions of the confined cells, yet are large enough to allow free passage of liquids and dissolved nutrient and product molecules. To cause a flow of nutrient media to occur through die bed of cells above die exit frit the cell chamber is placed under a positive hydrauhc pressure of 10 to 20,000 psi or more. The preferred hydraulic pressure is between 100 to 5000 psi with die most preferred hydrauhc pressure of approximately 1500 psi. Nutrient liquids are pumped tiirough the chamber at constant flow rates, and products are immediately accessible at the output of me cell chamber. It is also contemplated as part of die present invention tiiat die flow rates of liquid medium can vary according to die needs of die particular biocatalysts.

Now with particular reference to the embodiments shown in

Figs. 1 through 5, the hyperbaric fermentation process of the present invention begins witii the supply of a cell culture medium from a medium resevoir to the input of a high pressure pump as illustrated by die flow chart in Fig. 1. The media reservoir is maintained at ambient room pressure although it can be heated to any desired temperature. In one embodiment of the present invention, this high pressure pump is a high pressure hquid chromatography (HPLC) pump capable of delivering constant flow rates of 0.1 to 10 mL min of aqueous media at pressures from approximately 100 to 5000 psi. Otiier embodiments of the present invention may employ different high pressure pumps with different pressure and flow rate characteristics. The type of pump used is not critical to the present invention. The liquid flow is next optionally passed tiirough a pulse dampening device similar to those found in many high pressure pumping systems to minimize pressure fluctuations downstream. Then the hquid flow is passed tiirough a flow-tiirough pressure sensor. In the case of an anaerobic fermentation process such as that depicted in Fig. 1, air may be excluded from the media reservoir by normal methods and the maintenance of anaerobic conditions throughout die process is easily accomplished because the pressure- resistant connecting piping precludes the entrance of oxygen into the process flow at any point downstream of die media reservoir.

The pressurized flow of hquid nutrient media is next apphed to a two-position high pressure switching valve 60 (Rheodyne, Inc., Cotati, CA) This valve is suitable for flow rates of 0 to 10 mL/min. Referring now to Fig.

1, the hquid flow applied at port number 1 can be switched internally to connect to either ports number 2 or 6. When port number 1 is connected to port number. 6 (dotted path), me hquid flow padi leads from port number 6 to port number 5 via a stainless steel bypass loop. In this switch position, port number 5 is in internal communication witii port number 4. Thus, the hquid flow passes from port number 1 to port number 4, the final output port. A ball valve attached to port number 4 is normally closed and is only open during cell chamber pressurization. The hquid flow from port number 4 is directed into the system pressure regulator (see Fig. 1). The system pressure regulator 12 is, in one embodiment, a flow controller such as tiiose available from Varian Associates (Sunnyvale, CA), which maintains die hydraulic pressure of the entire liquid system at a preset value. In this case, the whole system may be maintained at pressures of approximately 0 to 5000 psi, aldiough higher pressures can be maintained. Note that, as a consequence of die ability to maintain the whole of the process system at hyperbaric levels, gas remains dissolved in die liquid until die pressure drops to ambient values after exit of the fluid from the system pressure regulator. Thus, the process flow is strictly a liquid flow. There is no gas phase in the system. Note further tiiat thermostatting of die system pressure regulator may be required as a result of the endodiermic nature of the liquid-gas disengagement process.

When the switching valve 60 is in the position as described above, the cell chamber can then be opened and loaded witii die desired cells or biocatalysts. After loading, the cell chamber 100 can be gradually pressurized by opening the ball valve connected to port number 4 (Fig. 1). By gradually opening die pressure equalizing valve, pressure is allowed to build on botii the input and output sides of die cell chamber. The pressure inside die loaded cell chamber can be monitored by d e sensor attached to d e chamber output. The chamber is thereby brought to a pressure equaling the system pressure.

In this manner, the cell chamber 100 can be gradually brought from ambient pressure to the desired system pressure without applying a large pressure pulse to the cell compartment and its contents. After the cell chamber 100 is pressurized, the ball valve connected to port number 4 is closed, tiius isolating tiie cell chamber. The switching valve position in which the flow is directed to bypass die cell chamber, as oudined above, is employed when loading and pressurizing the confinement chamber and is also employed when calibrating the system as to gas and hquid composition, flow rate, etc. prior to actual use (see Example 1).

In the case when switch port number 1 is internally connected to port number 2 (as shown in Fig. 1 witii solid lines), the liquid flow patii leads from port number 2 to port number 3 via a loop in which are mounted die cell confinement chamber and a pressure sensor. Normally, the ball valve connected to port number 4 is closed, allowing hquid flow only tiirough the ceU confinement chamber 100. In this switching valve position, port number 3 is in internal communication with port number 4. Thus, the hquid flow passes from port number 1 through the cell chamber loop and exits tiirough ports number 3 and 4. This is the normal operating flow path in which the hquid nutrient medium is directed to flow tiirough d e cell confinement chamber 100.

For an aerobic fermentation or cell culture, the embodiment of the present invention takes a different form (see Figs. 2 and 3). To dissolve a desired gas or gases (typically oxygen or air) into the nutrient medium, a volume of die aqueous media is pumped from the nutrient feed pump 10 into an absorption reservoir 20 wherein the gas and hquid medium are mixed and die gas is dissolved in the hquid medium. At the same time, a source of gas is also apphed to the absorption reservoir 20 through a line equipped witii a pressure regulator 23 and a check valve 17. Because die concentration of gas which is dissolvable in the hquid witii which the gas comes in contact is directly related to die pressure of the gas-liquid system, it is possible to establish a desired dissolved gas concentration in the nutrient hquid simply by varying the gas pressure. The driving force for the entrance of the applied gas(es) into the aqueous hquid flow is the pressure differential between the hquid and gas streams. Dissolving the gas in d e hquid medium is accomplished by adjusting the gas pressure to values greater tiian that of the liquid pressure. Thus, the concentration of dissolved gas(es) in the net hquid flow is a function of: (1) the pressure differential between the liquid and gas flows; (2) die flow rates of the gas(es) and hquids; and (3) die kinetics of d e gas dissolution process. For example, it has been found that at liquid flow rates of approximately 1.0 m/min, that a dissolved oxygen concentration of 11.3 mM may be achieved in an aqueous solution at 30°C by the application of approximately 200 psi of oxygen gas pressure, whereas the apphcation of approximately 1550 psi of oxygen gas pressure to a slightly different aqueous solution at 25°C results in a dissolved oxygen concentration of approximately 250 mM.

To assure that a gradient of dissolved gas concentration within the liquid in d e absorption reservoir 20 does not form and to speed the development of a uniform concentration of dissolved gas, an absorber recirculation pump 30 (high pressure pump number 2 in Fig. 2) is used to recirculate the gas-hquid mixture from the bottom of the absorption reservoir 20 to the top. Note tiiat both of these high pressure pumps (the nutrient feed pump 10 and die absorber recirculation pump 30) are internally equipped witii output check valves to be sure the hquid only flows in the direction indicated by die arrows. The absorption reservoir 20 in one embodiment of the present invention is a stainless steel container whose volume is a function of the desired flow rates of gas and hquid. Thus, the absorption reservoir volume 20 must be adjusted such tiiat die physically-dictated time required for dissolution of the gas into a liquid state can be accomphshed. In die case of liquid flows of less tiian

10 mL min, this volume can be as little as lOOmL; in the case of larger flows a gas absorption reservoir of 1 to 2 L or more may be required.

As shown in Figs. 2 and 3, the aqueous nutrient media feed circuit and die absorption recirculation pump circuit can be separated from the rest of the process by means of the isolation valve 40 connected to the input of the cell chamber feed pump 50 (high pressure pump number 3 in Fig. 2). Thus, it is possible to establish a replenishable volume of nutrient media containing the desired concentration of gas(es) in these circuits separately from the rest of the process flow if this isolation valve is closed. Note that the fluid circulating in the recirculation pump circuit is composed of both liquid containing the desired concentration of dissolved gas, as well as excess undissolved gas.

It is undesirable to allow contact between the cultured biocatalysts and* undissolved gas. For this reason, the process components downstream of die isolation valve, which together form the basic hyperbaric process as depicted in Fig. 1, must be operated at a system pressure greater than that of the gas mixing circuits. It has been determined that the operation of the cell chamber feed circuit (see Fig.3) at pressures approximately 500 psi greater than the pressure developed in the gas mixing circuits is sufficient to insure that any gas bubbles in the output of the gas-hquid mixing circuits will dissolve into die flowing hquid which exits the cell chamber feed pump 50 (Fig. 3). These additional quanta of dissolved gas add incrementally to the total dissolved gas concentration. Thus, it is desirable tiiat a determination of die gas concentration fed into the cell confinement chamber be made downstream of die cell chamber feed pump 50 (high pressure pump number 3 in Fig. 2).

The balance of the aerobic high pressure process depicted in Figs. 2 or 3 downstream of the gas-liquid mixing circuits is identical to the basic process stream of Fig. 1, and is operated in an identical fashion. Note that because it is necessary to establish a system hydrauhc pressure greater than that in the gas-hquid mixing circuits prior to opening the isolation valve which connects these circuits to the cell chamber feed circuit die apparatus is equipped with a priming line input 70 to which an aqueous hquid reservoir is apphed at system startup. This is used to establish a suitable system hydraulic pressure by adjustment of the system pressure regulator. As depicted in Fig. 3, the overall process system may be optionally equipped widi pressure gauges and/or microprocessor control loops monitoring flow indicating controllers (FIC) as well as pressure indicators (PI) and the like common to a person of ordinary skill in the art.

M alternative method for the mixing of hquid nutrient media widi a desired gas involves mixing of these two components in a hquid pump headspace. Varian Instruments Model 5020 HPLC (Sunnyvale, CA), which possesses solenoid-operated inlet valves, can be used to accomphsh mixing of hquid media with gas in this manner. In this system, it is possible to connect the liquid media supply to one inlet valve and connect a desired gas supply to anotiier inlet valve and allow the electronic controller to sample each inlet for an optimal gas-liquid mix. In this method, die supply gas pressure must be carefully maintained at input pressures greater than the pump backpressure. Gas dissolution into the liquid flow and the mixing of the two liquid components then occurs in the pump head oudet chamber.

Another alternative method for the precise mixing of liquid nutrient media with a desired gas can be accomphshed by applying the gas supply to die input side of a high pressure positive displacement pump at pressures close to tiiat required for dissolution of the gas into the pressurized liquid flow and utilizing this high pressure pump to meter die gas flow delivery rate to the gas liquid absorption reservoir. Note that gas delivery by this metiiod eliminates the need for gas flow restrictors and check valves in the process flow system.

Figs. 4 and 5 show one embodiment of the cell confinement chamber. The chamber is manufactured of stainless steel which is machined in two halves which are bolted togetiier with six machine bolts. With reference to Fig. 4, the upper half-chamber 105 is internally machined to provide an input to the chamber interior 110, flow distribution channels 115, an inner O-ring channel 120, and an internal biocatalyst confinement chamber 125 of approximately 3.0 mL volume. The bottom half-chamber 130 is machined to allow die insetting of a support lip for a porous metal frit (the exit frit) 135 as well as an outer O-ring channel 140 and an solvent output channel 145. Thus, when the two half-chambers 105, 130 are fitted widi chemically-resistant inner and outer 0-rings and are bolted togetiier, the cell chamber forms a pressure- resistant vessel 100 with a single input port 110 above the cell chamber and a single exit port 145 beneath the exit frit 135. The loading of die hollow internal chamber above the exit frit

135 widi either bead- immobihzed enzymes or microbial, plant, or animal cells is accomphshed by addition of a hquid or slurry suspension of the beads or cells prior to connection of the upper chamber-half input port 110 to the process flow system. The choice of the porosity of die metal exit frit 135 is dictated by the size of die cell or enzyme, or cell immobilization bead one desires to retain within the compartment Thus, for example, a 0.2 micron (μM) frit porosity is optimal for tiiose fermentations in which bacterial cells are to be confined witiiin the chamber, while a frit porosity of 5 μm is adequate for die retention of yeast cells. In another embodiment of die present invention, the porous metal exit frit may be replaced by a honeycombed ceramic frit of defined porosity. In yet another embodiment of the present invention, the exit frit may be a coarser metal frit which is overlaid widi a filter membrane of defined pore size (Micro Filtration Systems, Dublin, CA).

It is to be understood tiiat die geometry of the cell chamber is not critical except for its ability to withstand die apphed internal pressure. CeU chambers of cylindrical, spherical, or otiier geometrical configurations may replace the embodiment specified in Figs.4 and 5.

Another embodiment of die present invention utilizes porous fiber bundles to deliver nutrients to the biocatalysts in die pressurized chamber. Referring now to Fig. 6, the chamber 200 is manufactured of stainless steel.

The chamber has a high pressure fiber input end cap 205 into which the nutrient medium is fed under high pressure and a fiber output end cap 220.

The chamber 200 also has an inner fiber capillary bundle 210 in the cell culture compartment 215 as shown in the cutaway view of the chamber. The inner fiber bundle 210, such as those available from .Miicon Inc., Beverly, MA, is connected to die input end cap 205 and output cap 220 so that the flow of nutrient medium flows through the interior of the fibers in the fiber capillary bundle 210. Medium is circulated into the input flow cap 225 through the culture compartment 215 and out the output flow cap 230. In this way, the inner fiber bundle 210 is bathed by the medium that is circulated in the compartment 215.

In operation, the appropriate cells suspended in media are loaded into the culture compartment 215 via media input flow port 225. Fresh media is circulated tiirough the fibers 210 under hyperbaric pressure. This allows the cells access to nutrients which diffuse through the fiber walls. To collect cell products, the cell medium is injected tiirough media input flow port 225 and circulated through the culture compartment 215 and out through cell media output flow port 230. Products can also be collected in the medium that is collected tiirough the cell media output flow port 230. ^N Note that in the cell chamber embodiment of Fig. 6, a high pressure flow of liquid tiirough and around die fragile hollow fiber capillaries will not cause fiber rupture as long as the pressure differential across the fiber wall is held near zero. Small pressure differentials across the fiber wall may be used to drive nutrient or product molecules through the fiber walls. In yet another embodiment of the present invention, the biocatalyst cell chamber may take the form of an elongated cylinder in which tubular frit "fingers" of the appropriate porosity are mounted in an axial configuration within die columnar body of die chamber to increase die available exit frit surface area without appreciably increasing the physical size of the biocatalyst cell chamber. In the case of cell chambers not containing hollow fiber bundles, there is no limitation to the size of the biocatalyst cell chamber. For analytical purposes, a chamber size of 1 to 10 mL is sufficient for larger scale commercial apphcations, chamber sizes of 1 to 10 L or more may be required, witii a similar scale-up of pump capacity to assure adequate perfusion of the chamber contents. In practice, it has been determined mat a disc-shaped cell bed is a preferred configuration in which the problems of concentration gradients, pH changes, and anoxic pockets within the cell bed can be eliminated with the appropriate choice of flow rate and disc thickness.

Referring now to Fig. 7, the present invention comprises optional features including, but not limited to, sensing electrodes for dissolved nutrients and gases, sampling ports in various configurations, in-line gas and/or air filtering apparati, and die like features common to microbial fermentations and/or cell culture. Similarly, the cell confinement chambers) can optionally be thermostatted to provide optimal cell growth or productivity conditions, can be serially connected into a common flow process configuration to provide for the step-wise conversion of a substrate material into a final desired product by several different microorganisms or cell types. In an analytical embodiment, various outputs can be entrained with analytical instruments such as HPLC separation columns, spectrophotometers, etc. The general sequence of operations in the preferred embodiment of me present invention in the case of an aerobic fermentation is as follows: (1) the system is configured for aerobic operation by making the appropriate connections between the gas and hquid supply systems (see Fig. 2 and 3); (2) d e gas-liquid mixing circuits are disengaged from the cell chamber feed circuit by closing die isolation valve. (3) The cell chamber system priming valve is opened and a sterile aqueous reservoir apphed at this input; (4) die high pressure switching valve is turned to die confinement chamber bypass position, the cell chamber feed pump is started at die desired flow rate, and die desired system pressure set by the adjustment of die system pressure regulator, (5) the appropriate nutrient media for the organism to be cultured is prepared, sterilized, and apphed at die low pressure inlet to the nutrient feed pump and approximately 100 mL of nutrient is pumped into die absorption reservoir whereupon this pump is stopped; (6) the gas supply is opened and set at the desired operating pressure and the recirculation pump is flowed at an appropriate rate for approximately 30 minutes; (7) the cell chamber system priming valve is closed, die isolation valve is opened, die nutrient feed pump is started at the same flow rate as the cell chamber feed pump, and die flow of hquid is calibrated as to system pressure, system temperature, hquid flow rate, and die concentration of liquids and gas(es), pH, etc. in the output flow are determined. (8) The cell chamber is opened and die desired microorganism, cell type, or bead-immobihzed biocatalyst is inoculated into the cell chamber

(typically as a slurry in the maintenance nutrient media) and the chamber is sealed; (9) die ball valve connected between switch port number 1 and the pressure equalizing valve (Fig. 3) is opened so tiiat the cell chamber is pressurized to the process system pressure; (10) a ball valve is closed and die switching valve is turned to direct die hquid flow tiirough the cell chamber loop.

This effectively directs die hquid flow in its entirety through the cell chamber under normal operating conditions.

Mother embodiment of the present invention is shown in Fig. 9. In this embodiment medium is introduced from eitiier end of the chamber at 320 or 325. How into and out of the chamber can be easily reversed. The biocatalyst can be introduced via die inoculation port at 305 and die medium is introduced under pressure from ether end of die chamber. The walls of mbes 320 are manufacture from either metallic or ceramic frit material which is permeable to the hquid medium but is impermeable to the biocatalysts. If cells are the biocatalyst a pore size of approximately 0.2 μ is adequate. The mbes 320 and 325 are shown in Fig. 9 as straight mbes but they can be any shape in any configuration. The chamber 300 has a blind flange 330 and a screw-on flange 335 at both ends of die chamber. In operation, the medium passes through the walls of the mbes 320 and batiies the biocatalyst. the medium then passes through the walls of the exit tube and is removed from the chamber. As stated before, the flow can be easily reversed if the exit tube becomes obstructed witii cells or other debris. By reversing the flow, the exit tube becomes die inlet tube and die debris is forced away from die tube.

System shut down (when it is desired tiiat the confined cells be recoverable) requires that first the gas supply is closed; die absorption reservoir relief valve 15 (Fig. 3) is opened to release any gaseous overpressure; and, d e system is allowed to flow until no gas is present in the output liquid, an indication that the only dissolved gases remaining in the system are at atmospheric pressure. Whereupon the remainder of d e system can be shut down by a reversal of die startup procedure. System sterilization between runs 5 may be accomplished by either passage of steam or a sterilizing solution through the process system.

The microbial organisms which may be used in the present invention include, but are not limited to, dried cells, wet cells harvested from broth by centrifugation or filtration, or from the cultured broth itself. These

10 microbial cells are classified into die following groups: bacteria, actinomycetes, fungi, yeast, and algae. Bacteria of the first group, belonging to Class Shizomycetes taxonomically, are Genera Pseudomonas, Acetobacter, Gluconobacter, Bacillus, Corynebacterium, Lactobacillus, Leuconostoc, Streptococcus, Clostridium, Brevibacterium, Arthrobacter, or Erwinia, etc.

15 (see R.E. Buchran and N.E. Gibbons, Bergev's Manual of Determinative

Bacteriology. 8th ed., (1974), Wilhams and Wilkins Co.). Actinomycetes of the second group, belonging to Class Shizomycetes taxonomically, are Genera Streptomyces, Nocardia, or Mycobacterium, etc. (see R.E. Buchran and N.E. Gibbons, Bergev's Manual of Determinative Bacteriology. 8th ed., (1974),

20 Williams and Wilkins Co.). Fungi of the third group, belonging to Classes

Phy corny cetes, Ascomycetes, Fungi imperfecti, and Bacidiomycetes taxonomically, are Genera Mucor, Rhizopus, Aspergillus, Penicillium, Monascus, or Neurosporium, etc. (see J.A. von Mk, "The Genera of Fungi Sporulating in Pure Culture", in Illustrated Genera of Imperfect Fungi.3rd ed.,

25 V.von J. Cramer, H.L. Barnett, and B.B. Hunter, eds. (1970), Burgess Co.).

Yeasts of the fourth group, belonging to Class Ascomycetes taxonomically, are Genera Saccharomyces, Zygosaccharomyces, Pichia, Hansenula, Candida, Torulopsis, Rhodotorula, Kloechera, etc. (see J. Lodder, The Yeasts: A Taxonomic Study. 2nd ed., (1970), North-Holland). Algae of the fifth group i- include green algae belonging to Genera Chlorella and Scedesmus and blue- green algae belonging to Genus Spirulina (see H. Ta iya, Studies on Mieroalgae and Photosvnthetic Bacteria. (1963) Univ. Tokyo Press). It is to be understood that the foregoing listing of microorganisms is meant to be merely representative of the types of microorganisms that can be used in the

35 fermentation process according to the present invention. The culture process of the present invention is also adaptable to eukaryotic plant or animal cells which can be grown either in monolayers or in suspension culture. The ceh types include, but are not limited to, primary and secondary cell cultures, and diploid or heteroploid cell lines. Otiier cells which can be employed for the purpose of virus propagation and harvest are also suitable. CeUs such as hybridomas, neoplastic cells, and transformed and untransformed cell lines are also suitable. Primary cultures taken from embryonic, adult, or tumorous tissues, as well as cells of established cell lines can be employed. Examples of typical such cells include, but are not limited to, primary rhesus monkey kidney cells (MK-2), baby hamster kidney cells

(BHK21), pig kidney cells (IBRS2), embryonic rabbit kidney cells, mouse embryo fibroblasts, mouse renal adenocarcinoma cells (RAG), mouse medullary tumor cells (MPC-11), mouse-mouse hybridoma cells (1-15 2F9), human diploid fibroblast cells (FS-4 or AG 1523), human hver adenocarcinoma cells (SK-HEP-1), normal human lymphocytic cells, normal human lung embryo fibroblasts (HEL 299), WI 38 or WI 26 human embryonic lung fibroblasts, HEP No. 2 human epidermoϊd carcinoma cells, HeLa cervical carcinoma cells, primary and secondary chick fibroblasts, and various ceh types transformed widi, for example, SV-40 or polyoma viruses (WI 38 VA 13, WI 26 VA 4, TCMK-1, SV3T3, ___.). Other suitable established cell lines employable in the method of die present invention will be apparent to d e person of ordinary skill in the art

The products tiiat can be obtained by practicing the present invention are any metabolic product that is the result of the culturin of a cell, either eukaryotic or prokaryotic; a ceh subcellular organelle or component, such as mitochondria, nuclei, lysozomes, endoplasmic reticulum, golgi bodies, peroxisomes, or plasma membranes or combinations thereof; or an enzyme complex, either a natural complex or a synthetic complex, i.e., a plurality of enzymes complexed togetiier to obtain a desired product One of die advantages of the present invention is being able to produce a desired chemical from a cell without having to go through the laborious process of isolating the gene for the chemical and then inserting the gene into a suitable host ceh so tiiat the chemical can be produced in corαmercial quantities. By practicing the present invention, a mammalian cell that is known to produce a desired chemical can be directly cultured according to die present invention to produce large quantities of the desired chemical Products tiiat can be produced according the present invention include, but are not limited to, immunomodulators, such as interferons, interleukins, growth factors, such as erythropoietin; monoclonal antibodies; antibiotics from microorganisms; coagulation proteins, such as Factor VHI; fibrinolytic proteins, such as tissue plasminogen activator and plasminogen activator inhibitors; angiogenic proteins; and hormones, such as growth hormone, prolactin, glucagon, and insulin.

The term "culture medium""includes any medium for die optimal growth of microbial, plant or animal cells or any medium for enzyme reactions including, but not limited to, enzyme substrates, cof actors, buffers, and die like necessary for the optimal reaction of the enzyme or enzyme system of choice. Suitable culmre media for cell growth will contain assimilable sources of nitrogen, carbon, and inorganic salts, and may also contain buffers, indicators, or antibiotics. Any culture medium known to be optimal for the culture of microorganisms, cells, or biocatalysts may be used in the present invention. While such media are generally aqueous in nature for the culture of living organisms, organic solvents or miscible combinations of water and organic solvents, such as dimethylformamide, methanol, dietiiyl ether and die like, may be employed in those processes for which they are proved efficacious, such as those bioconversions in which immobihzed biocatalysts are employed. Passage of the liquid media through the process system may be either one-pass or the liquid flow may be recycled tiirough the system for higher efficiency of conversion of substrate to product Desired nutrients and stimulatory chemicals may be introduced into die process flow, eitiier via the low pressure nutrient supply or via an injection valve in the process flow upstream of the cell chamber.

It will be appreciated that the present invention is adaptable to any of the well-known tissue culture media including, but not limited to, Basal Medium Eagle's (BME), Eagle's Minimum Essential Medium (MEM),

Dulbecco's Modified Eagle's Medium (DMEM), Ventrex Medium, Roswell Park Medium (RPMI 1640), Medium 199, Ham's F-10, Iscove's Modified Dulbecco Medium, phosphate buffered salts medium (PBS), and Earle's or Hank's Balanced Salt Solution (BSS) fortified with various nutrients. These are commercially-available tissue culture media and are described in detail by

HJ. Morton (1970) In Vitro 6, 89-108. These conventional culture media contain known essential amino acids, mineral salts, vitamins, and carbohydrates. They are also frequently fortified with hormones such as insulin, and mammalian sera, including, but not limited to, bovine calf serum as well as bacteriostatic and fungistatic antibiotics. Although ceh growth or ceh respiration within the cell chamber cannot be directly visualized, such metabolism may readily be monitored by die chemical sensing of substrate depletion, dissolved oxygen content, carbon dioxide production, or the like. Thus, for example in the case of a fermentation of a species of Saccharomyces cerevisiae, inoculation of the ceh chamber with a small starter population of ceUs can be followed by an aerobic fermentation regime in which glucose depletion, dissolved oxygen depletion, and carbon dioxide production across the cell confinement chamber are measured either chemically or via appropriate sensing electrodes. Thus, cell replication can be allowed to proceed until an optimal cell bed size is reached. Withdrawal of dissolved oxygen input at this time causes the immobilized yeast cells to shift into anaerobic fermentation of glucose with a resultant production of ethanol, a process which can likewise be monitored chemically (see Example IH).

Similarly, without any process modification, the process of the present invention can be utilized as a bioreactor for immobihzed chemical catalysts, enzymes or enzyme systems. In such a process, a catalyst, an enzyme or an enzyme system is chemically immobilized on a sohd support including, but not hmited to, diatomaceous earth, silica, alumina, ceramic beads, charcoal, or polymeric or glass beads which are then introduced into the cell chamber into which has been mounted an exit frit of a pore size larger than the solid support. The reaction medium, either aqueous, organic, or mixed aqueous and organic solvents, flows tiirough the process system and tiirough the packed bed within die cell chamber. The catalyst, enzyme, or enzyme system converts a reactant in the process flow medium into die desired product or products. Similarly, in other apphcations, either cells or ceh components including, but not limited to, vectors, plasmids, or nucleic acid sequences

(RNA or DNA) or the like may be immobilized on a sohd support matrix and confined under hyperbaric conditions for similar utilization in converting an introduced reactant into a desired product

Commercial application of the present invention can be in the production of medicaUy-relevant ceUularly-derived molecules including, but not limited to, anti-tumor factors, hormones, therapeutic enzymes, viral antigens, antibiotics and interferons. Examples of possible product molecules which might be advantageously prepared using the method of the present invention include, but are not limited to, bovine growth hormone, prolactin, and human growtii hormone from pituitary ceUs, plasminogen activator from kidney ceUs, hepatitis-A antigen from cultured hver cells, viral vaccines and antibodies from hybridoma cells, insulin, angiogenisis factors, fibronectin, HCG, lymphokines, IgG, etc. Otiier products will be apparent to a person of ordinary skill in the art.

While the foregoing description illustrates various embodiments and apphcations of the hyperbaric culture process of the present invention, the foUowing examples are presented by way of Ulustration only to demonstrate die operation and effectiveness of the process and are not to be construed as limiting the invention to the specific examples described or to the precise mode of operation.

Example I To demonstrate me capability of the hyperbaric culture process to markedly increase die concentration of dissolved oxygen in a typical aerobic microbial ceU culture medium, an apparams was set up which comprised all d e portions of the present invention iUustrated in Figs. 2 or 3. The nutrient media was composed of: yeast extract (6 gm/L), peptone (6 gm. ), glucose (10 gm L), ethanol (4 % [v/v]), and acetic acid (2 % [v/v]). The ceU chamber feed pump circuit was isolated from the gas mixing circuits by closing die isolation valve connecting this circuit to the gas mixing circuits and opening the cell chamber priming line valve to which a media reservoir was attached. The switching valve in the ceU chamber circuit was placed in die position in which the cell chamber would be bypassed, since no ceU inoculation was to be performed.

The cell chamber feed pump was started at a flow rate of 1.0 mL/min and a hydrauhc system pressure of 2000 psi was set for the liquid media flow. A system temperature of 25°C was established. A 200 mL stainless steel cylinder was used as die gas-liquid mixing/absorption chamber. The nutrient feed pump, also equipped with a media reservoir, was started after opening the absorption reservoir relief valve (Fig. 3) and 100 mL of media was pumped into the absorption reservoir at 10 mL/min, whereupon the nutrient pump was stopped and die relief valve closed. A standard bottle of oxygen gas, obtained commerciaUy at a tank pressure of 2200 psi, was equipped with a tank regulator set for 1550 psi, and die gas line connected to the absorption reservoir via a check valve (17 in Fig. 3).

The recirculation pump was started at a flow rate of 15 mL min and absorption of gas by the recirculated media aUowed to proceed for 30 minutes. Next the ceU chamber priming line valve was closed and die adjacent circuit isolation valve opened to connect die gas absorption circuits to the chamber feed circuit and die nutrient feed pump was restarted at a flow rate of 1.0 mL/min. The process was run for 30 minutes to allow steady-state conditions to be established through the process. In contrast to the clear, golden appearance of die input nutrient hquid media, die output stream was a milky-white fluid composed of both gas and liquid which had begun to disengage in the output tubing distal to die system pressure regulator. Aliquots of the input and output liquid media were sampled for dissolved oxygen content. The results are presented in Fig. 8. The input nutrient media was found to have a dissolved oxygen content of 0.284 mM, a value which is typical for continuous stirred tank reactors (CSTR). This value is displayed as the leftmost bar of Fig. 8. The concentration of oxygen found in the output stream for this experiment, in contrast, was determined to be approximately 250 mM (rightmost bar, Fig. 8). The two middle bars in Fig. 8 are representative dissolved oxygen values obtained from the literature for typical continuous stirred tank reactors utilizing either one atmosphere (14.7 psi) or two atmospheres of oxygen gas overpressure above hquid fermentation media.

Example II

Using an identical experimental setup as that detailed in Example I, and using the same nutrient media supply composition, an experiment was run in which Acetobacter aceti (ATCC No.23746) was inoculated into the ceU chamber. The bacteria, which had been grown to log phase in a conventional fermentation apparams, were introduced into the chamber above a 0.2 μM metal exit frit as an approximately 2.0 mL slurry obtained by low speed centrifugation of a 300 mL aliquot of the batch ferment The process system was initiated as described earlier, and a process flow rate of 0.3 mL at 1500 psi established, widi oxygen gas apphed to the mixing/absorption chamber at a pressure of 1000 psi. The isolation valve between ports number 1 and 2 of d e switching valve

(see Fig. 3) was opened and die pressure equalizing valve in this circuit graduaUy opened until the pressure in the ceU chamber equaled die system hydrauhc pressure, whereupon die shutoff valve was closed. After samples of me nutrient supply medium were taken for later analysis, the process was allowed to flow, for approximately 15 hours, whereupon a sample of the product output hquid was taken.

Etiianol and acetic acid were analyzed by gas chromatography on a Tracor Model MT-220 chromatograph equipped witii an AUtech Model PRP X300 column held at 220°C with nitrogen as die carrier gas flowing at 4.0 mL/min. Detection of separated peaks was accomplished with a flame ionization detector. Etiianol was detected near die column void volume, while acetic acid was retained approximately 2.5 minutes. Gas chromatographic analysis of the starting nutrient media was compared to mat of the product effluent at T=15 hours and die results are presented in Table I.

Table I

Comparative data taken from U.S. Patent No.4,463,019 (granted to Okuhara, et al., Col. 6, lines 49-68, Col. 7, lines 1-22, and Fig. 2), witii the assumption that the ferment of these data contained at least 2.0 mL of active bacteria.

Note that the dissolved oxygen content of die ferment employing the method of the present invention in this example was approximately 160 mM, while the dissolved oxygen content for die comparative data (marked wid an asterisk) was in the range of 20-30% (v/v), equivalent to approximately 0.350 mM dissolved oxygen. Note further that the fermentation according to the present invention resulted in the production of acetic acid at a rate approximately twice as large as that of the comparative prior art method. The estabhshment of a process temperature of 22°C, which is approximately eight degrees lower than that known to be optimal for these bacteria, was necessary because the exothermic nature of the chemical oxdation of ethanol to acetic acid by the enzymes of these bacteria results in the production of a considerable amount of excess heat which must be carried away by die hquid flow. Experiments in which the initial process temperature was set at the optimal temperature of approximately 30°C resulted in loss of the microbial population as a result of excessive heating. Thus, when processes requiring increased dissolved oxygen are desired, utilization of the method of the present invention requires that consideration be taken of the amount of heat which wiU be produced by oxidative biochemical processes. Either flow rates must be adjusted to carry away this heat or die steady-state process temperature must be lowered to preserve the microbial or ceUular populations.

Example HI

To demonstrate the capability of the hyperbaric culture process of the present invention to (1) aerobicaUy grow a proliferative microorganism and, (2) subsequently anaerobically to cause the production of etiianol, an apparams was set up according to Figs.2 or 3. A system temperature of 30°C was set a flow rate of 0.3 παL/min and a system pressure of 1500 psi was chosen for the hquid media flow. The nutrient media was composed of: yeast extract (3 gm L), peptone (3 gm/L), and glucose (1 % [w v]). A 200 mL stainless steel cylinder was used as the gas-hquid mixing absorption chamber.

A standard bottle of oxygen, obtained commercially at a tank pressure of 2200 psi, was equipped witii a tank regulator set for 1000 psi. The high pressure switching valve shown in Fig. 2 was set in an intermediate position, the absorption reservoir relief valve opened, and 100 mL of media was pumped into the absorption reservoir at 10 mL min, whereupon die nutrient feed pump was stopped, die absorption reservoir relief valve closed, die gas line opened to pressurize the absorption reservoir, and die switching valve was set in the position in which the ceU chamber would be bypassed. Both the nutrient feed pump and die ceU chamber feed pumps were set at flow rates of 0.3 rcd_/min, while die recirculation pump was set at 15 mL/min.

The cell chamber was opened and a 1.0 mL slurry of Saccharomyces cerevisiae (ATCC 4126) suspended in the above medium was loaded into the chamber over a 5 μM metal exit frit and die chamber closed. The isolation valve between ports number 1 and 2 of the switching valve (see Fig. 3) was opened and die pressure equalizing valve in this circuit gradually opened until the pressure in the cell chamber equaled die system hydraulic pressure, whereupon the shutoff valve was closed. The switching valve was then set into the normal operating position (see Figs.2 or 3). The process was aUowed to run for approximately 15 hours, whereupon the supply of oxygen to the process system was shut off and die absorption reservoir rehef valve opened to remove any oxygen overpressure. The process was allowed to run for an additional 18 hours and die output hquid flow analyzed for glucose content by the ø-toluidine colorometric assay (Sigma Chemical Company, St. Louis, MO) and etiianol (by the previously-mentioned gas chromatographic method). At the end of the experiment, the ceU chamber was opened and tiie ceU mass washed out and pelleted by low speed centrifugation. The recovered ceU mass occupied 3.1 mL, approximately a three-fold increase in ceU volume. The results of the chemical analysis are summarized in Table II.

Table II

Comparative data taken from R.P. Jones, et al. (Process Biochem. April/May,

(1981), pp.42-49, Table 6), with the assumption tiiat the ferment of these data contained at least 1.0 mL wet cells per hter of ferment. The yeast strain of the comparative data was S. cerevisiae NCYC 239.

The data of Example HI demonstrate several unique aspects of die fermentation process according to d e present invention. First of ah, the shift from a proliferative growth phase to an anaerobic production phase was accomplished simply by withdrawing the gaseous oxygen input; there was no additional change in the process flow. The growth phase resulted in a tripling of cell volume in approximately 15 hours. Next, at a flow rate of 0.3 mL and an immobihzed volume of yeast cells of 3.1 mL, efficiency in glucose conversion to ethanol was greatly increased.

Example IV To demonstrate die suitability of the process of the present invention for use with a more fragtie ceU type, as weU as to demonstrate d e abUity of d e process to increase ceUular production of a clinicaUy-significant protein by the raising of dissolved oxygen levels, two separate experiments were run.

In Experiment A, an apparams was set up which comprised aU die portions of the present invention Ulustrated in Figs. 2 or 3. A system temperature of 30°C was set and a flow rate of 1.0 mL/min was chosen for the hquid media flow input to the pump circuit containing the ceU chamber. A system pressure of 1000 psi was established. The nutrient media was composed of Dulbecco's modified Eagle's medium (DMEM) with 4.5 gm/L glucose and 4.0 mM L-glutamine added. No fetal bovine serum was added to die DMEM. The sterile media was introduced into the ceU chamber circuit at the ceU chamber system priming input A recombinant clone of the AtT-20 mouse pituitary ceU line (AtT-20-GH13/5/4) which expresses human growth hormone

(hGH) was chosen for these experiments. This well-studied animal ceU line has been successfuUy transfected witii plasmids encoding eiti er insulin or hGH (see H.-P. H. Moore, et al, (1983) Cell 35, 531 and H.-P. H. Moore and R. B.

KeUy, (1985) / . CellBiol. 101, 1773).

The ceUs had been grown in spinner flasks in DMEM supplemented with 10% fetal calf serum until "tumor spheroids" composed of self-aggregated AtT-20 ceUs had formed. M aliquot of 0.4 mL of sedimented spheroids in approximately 2.5 ml of DMEM was introduced into the ceU chamber and the chamber pressurized as noted earlier. The high pressure switching valve was turned to aUow the flow of DMEM from the ceU chamber circuit pump to flow through the chamber. After a single-pass sample was taken at 10 minutes for hGH assay, the system output was redirected to recirculate back into the media input reservoir. The total media volume in the system was 75 mL. The system was aUowed to run for 4 hours, whereupon an additional sample was taken for hGH assay. Quantitation of secreted hGH in the hquid output was accomphshed by liquid phase radioimmunoassay using a kit obtained from Cambridge Medical Research, Inc. (BUlerica, MA) that employed a rabbit anti-hGH antiserum. Note that the only oxygen available to the cells during this experiment was that dissolved in the media reservoir at atmospheric pressure (approximately 0.250 mM).

In Experiment B, the same apparams and conditions were employed as in Experiment A, except that after the introduction of approximately 100 mL of DMEM into the gas mixing absorption chamber via the nutrient pump input the oxygen tank regulator was set to 150 psi and d e absorption reservoir thus pressurized to 150 psi with oxygen gas. The recirculation pump was set to flow at 15 mL/min and die media recirculated for approximately 30 minutes until the dissolved oxygen content of d e media stabihzed at 11.3 mM. The media reservoir was connected to the input of the nutrient feed pump, the cell chamber priming line valve closed, and die adjacent isolation valve connecting the gas-hquid mixing absorption circuits to the cell chamber circuit opened.

The nutrient feed pump and d e ceU confinement chamber feed pump were allowed to flow at 1.0 mL/min, while the recirculation pump continued to flow at 15 mL min. M aliquot of 0.4 mL of sedimented spheroids in approximately 2.5 ml of DMEM was introduced into the cell confinement chamber and the chamber pressurized as noted earlier. The high pressure switching valve was turned to ahow the flow of DMEM from the ceh chamber circuit pump to flow through the chamber. After a single-pass sample was taken at 10 minutes for hGH assay, the system output was redirected to recirculate back into die media input reservoir. The total media volume in the system was 150 mL. The system was aUowed to run for 4 hours, whereupon an additional sample was taken for hGH assay. Quantitation of secreted hGH in the liquid output was accomplished as noted above. The results of tiiese experiments are summarized in Table HI.

Table in

* Comparative values for the same cells maintained by conventional culture techniques (see A. Sambanis, G. Stephanopoulos, and H. F. Lodish (1990), Cytotech. 4, 111.

As can be seen from die data summarized in Table HI, hGH production rates were increased almost 10 fold initially in Experiment B over that of Experiment A. Even after four hours of culmre, the rate of hGH production was stiU approximately 5 fold die initial rate in the absence of increased 02. The rate of hGH production after four hours of culture was also approximately 5 fold higher than that reported in die hterature.

The foregoing description of the preferred embodiments of the process of the present invention and die examples have been presented for purposes of iUustration and description and to teach a better understanding of die invention. It is not intended to be exhaustive or to limit die invention to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. The particular embodiments described and die examples chosen are intended to best explain the principles of the invention and its practical application to thereby enable others skilled in the relevant art to best utilize the present invention in various embodiments and with various modifications as are suited to the particular use contemplated. It is intended to cover in die appended claims aU such modifications, variations, and changes as faU within die scope of the process of this invention.

Claims

CLAIMSWe claim:

1. A method of culturing a biocatalyst comprising the step of incubating the biocatalyst in a liquid medium under hyperbaric conditions with no gas phase in contact with the medium.

2. The method of Claim 1, wherein die medium contains a dissolved gas.

3. The metiiod of Claim 2, wherein the dissolved gas is selected from die group consisting of air, 02, NH3, NO2, M, He, N2 and H2 or a mixture thereof.

4. The method of Claim 3, wherein the medium contains oxygen dissolved ti erein.

5. The method of Claim 1 , wherein the biocatalyst is a cell.

6. The method of Claim 5, wherein the cell is a prokaryotic cell.

7. The method of Claim 6, wherein the prokaryotic cell is a bacteria.

8. The method of Claim 5, wherein the cell is a eukaryotic cell.

9. The method of Claim 8, wherein the eukaryotic cell is selected from the group consisting of algae cells, yeast cells, fungus cells, insect cells, reptile cells and mammalian ceUs.

10. The method of Claim 9, wherein the eukaryotic cell is selected from the group consisting of mammalian ceUs.

11. The method of Claim 1, wherein the biocatalyst is a subceUular component

12. The method of Claim 11, wherein the biocatalyst is an enzyme complex.

13. The metiiod of Claim 12, wherein die enzyme complex is immobihzed on a sohd support

14. The method of Claim 1, wherein the hquid medium is a nutrient medium.

15. The method of Claim 1, wherein the hquid medium is a balanced salt solution.

16. The method of Claim 1 , wherein the medium contains an organic solvent.

17. The method of Claim 1, wherein the biocatalyst is incubated anaerobicaUy.

18. The method of Claim 1, wherein the biocatalyst is incubated aerobicaUy.

19. A method of producing a product from a biocatalyst comprising die steps of: a. incubating the biocatalyst in a hquid medium under hyperbaric conditions with no gas phase in contact with die medium; and b . coUecting the medium with die product therein.

20. The method of Claim 19, comprising the additional step of isolating the product from the medium,

21. The method of Claim 19, wherein the medium contains a dissolved gas.

22. The metiiod of Claim 21, wherein the dissolved gas is selected from the group consisting of air, 02, NH3, NO2, M, He, N2 and H2 or a mixture thereof.

23. The method of Claim 22, wherein the medium contains oxygen dissolved tiierein.

24. The method of Claim 19, wherein the product is produced anaerobicaUy.

25. The method of Claim 19, wherein the biocatalyst is a ceU.

26. The method of Claim 25, wherein the cell is a prokaryotic cell.

27. The method of Claim 26, wherein the ceU is a bacteria.

28. The method of Claim 26, wherein the ceU is a eukaryotic ceU.

29. The method of Claim 28, wherein the eukaryotic cell is selected from die group consisting of algae cells, yeast cells, fungus cells, insect cells, reptile cells and mammalian cells.

30. The metiiod of Claim 29, wherein the eukaryotic cell is a mammalian ceU.

31. The method of Claim 19, wherein the biocatalyst is a subceUular component.

32. The method of Claim 19, wherein die biocatalyst is an enzyme complex.

33. The method of Claim 32, wherein the enzyme complex is immobilized on a sohd support.

34. The method of Claim 19, wherein the medium contains organic solvents.

35. The method of Claim 19, wherein the medium is a nutrient medium.

36. The method of Claim 19, wherein the medium is a balanced salt solution.

37. A fermentation and culturing apparatus comprising a means for containing a biocatalyst in a liquid medium under hyperbaric conditions widi no gas phase in contact with the medium.

38. The apparams of Claim 37, further comprising an imput means for supplying the Hquid medium.

39. The apparams of Claim 37, further comprising an exit means for coUecting the hquid medium.

40. The apparams of Claim 37, further comprising a means for dissolving a gas in the medium.

41. The apparams of Claim 40, wherein the gas is selected from the group consisting of air, 02, NH3, NO2, M, He, N2 andH2 or a mixture thereof.

42. The apparams of Claim 41, wherein the gas is 02-

43. The apparams of Claim 37, wherein the means for containing biocatalysts under pressure is a chamber.

44. The apparatus of Claim 43, wherein the chamber contains an inlet port and an exit port.

45. The apparams of Claim 37, wherein the chamber has a means for controlling d e temperature of the chamber.

46. The apparams of Claim 44, wherein the exit port contains a frit capable of allowing die medium to pass through the frit but not aUowing die biocatalysts to pass through the frit.

47. The apparams of Claim 46, wherein the frit is selected from the group consisting of membranes of defined pore sizes, porous foamed metal, ceramic honeycomb monohdi and coarser frits overlaid with membranes of defined pore sizes.

48. The apparams of Claim 37, wherein the means for containing a biocatalyst under pressure with no gas phase comprises a plurality of the biocatalyst containing means.

49. The apparams of Claim 48, wherein the plurality of the biocatalyst containing means are arranged in paraUel.

50. The apparams of Claim 37, wherein the plurality of the biocatalyst containing means are arranged in series.

51. The apparams of Claim 37, wherein the means for containing a biocatalyst contains a means for separating the medium from the biocatalyst.

52. The apparams of Claim 51, wherein the means for separating the medium from die biocatalyst is a hoUow fiber capable of allowing only the medium to pass tiirough the waUs of d e fibers.

53. The apparatus of Claim 43, wherein the chamber contains a means for immobilizing cells.

54. The apparams of Claim 53, where the means for immobilizing cells is selected from the group consisting of dextran, polyacrylamide, nylon, polystyrene, calcium alginate, or agar gel.