Production of heterologous proteins by microbial and animal cells Monoclonal antibodies produced by animal cells
Fermenters developed in stages 3 and 4. Animal cell reactors developed
Control and sensors developed in stages 3 and 4
Batch, fed-batch or continuous Continuous perfusion developed for animal cell processes
Introduction of foreign genes into microbial and animal cell hosts. In vitro recombinant DNA techniques used in the improvement of stage 3 products in 1801. By the mid-1800s the role of yeasts in alcoholic fermentation had been demonstrated independently by Cagniard-Latour, Schwann and Kutzing but it was Pasteur who eventually convinced the scientific world of the obligatory role of these micro-organisms in the process. During the late 1800s Hansen started his pioneering work at the Carlsberg brewery and developed methods for isolating and propagating single yeast cells to produce pure cultures and established sophisticated techniques for the production of starter cultures. However, use of pure cultures did not spread to the British ale breweries and it is true to say that many of the small, traditional, ale-producing breweries still use mixed yeast cultures at the present time but, nevertheless, succeed in producing high quality products.
Vinegar was originally produced by leaving wine in shallow bowls or partially filled barrels where it was slowly oxidized to vinegar by the development of a natural flora. The appreciation of the importance of air in the process eventually led to the development of the 'generator' which consisted of a vessel packed with an inert material (such as coke, charcoal and various types of wood shavings) over which the wine or beer was allowed to trickle. The vinegar generator may be considered as the first 'aerobic' fermenter to be developed. By the late 1800s to early 1900s the initial medium was being pasteurized and inoculated with 10% good vinegar to make it acidic, and therefore resistant to contamination, as well as providing a good inoculum (Bio-letti, 1921). Thus, by the beginning of the twentieth century the concepts of process control were well established in both the brewing and vinegar industries.
Between the years 1900 and 1940 the main new products were yeast biomass, glycerol, citric acid, lactic acid, acetone and butanol. Probably the most important advances during this period were the developments in the bakers' yeast and solvent fermentations. The production of bakers' yeast is an aerobic process and it was soon recognized that the rapid growth of yeast cells in a rich wort led to oxygen depletion in the medium which, in turn, resulted in ethanol production at the expense of biomass formation. The problem was minimized by restricting the initial wort concentration such that the growth of the cells was limited by the availability of the carbon source rather than oxygen. Subsequent growth of the culture was then controlled by adding further wort in small amounts. This technique is now called fed-batch culture and is widely used m the fermentation industry to avoid conditions of oxygen limitation. The aeration of these early yeast cultures was also improved by the introduction of air through sparging tubes which could be steam cleaned (de Becze and Liebmann, 1944).
The development of the acetone-butanol fermentation during the First World War by the pioneering efforts of Weizmann led to the establishment of the first truly aseptic fermentation. All the processes discussed so far could be conducted with relatively little contamination provided that a good inoculum was used and reasonable standards of hygiene employed. However, the anaerobic butanol fermentation was susceptible to contamination in the early stages by aerobic bacteria, and by acid-producing anaerobic ones once anaerobic conditions had been established in the later stages of the process. The fermenters employed were vertical cylinders with hemispherical tops and bottoms constructed from mild steel. They could be steam sterilized under pressure and were constructed to minimize the possibility of contamination. Two-thousand-hecta-litre fermenters were commissioned which presented the problems of inoculum development and the maintenance of aseptic conditions during the inoculation procedure. The techniques developed for the production of these organic solvents were a major advance in fermentation technology and paved the way for the successful introduction of aseptic aerobic processes in the 1940s.
The third stage of the development of the fermentation industry arose as a result of the wartime need to produce penicillin in submerged culture under aseptic conditions. The production of penicillin is an aerobic process which is very vulnerable to contamination. Thus, although the knowledge gained from the solvent fermentations was exceptionally valuable, the problems of sparging a culture with large volumes of sterile air and mixing a highly viscous broth had to be overcome. Also, unlike the solvent fermentations, penicillin was synthesized in very small quantities by the initial isolates and this resulted in the establishment of strain-improvement programmes which became a dominant feature of the industry in subsequent years. Process development was also aided by the introduction of pilot-plant facilities which enabled the testing of new techniques on a semi-production scale. The development of a large-scale extraction process for the recovery of penicillin was another major advance at this time. The technology established for penicillin fermentation provided the basis for the development of a wide range of new processes. This was probably the stage when the most significant changes in fermentation technology took place resulting in the establishment of many new processes over the period, including other antibiotics, vitamins, gibberellin, amino acids, enzymes and steroid transformations. From the 1960s onwards microbial products were screened for activities other than simply antimicrobial properties and screens became more and more sophisticated. These screens have evolved into those operating today utilizing miniaturized culture systems, robotic automation and elegant assays.
In the early 1960s the decisions of several multi-national companies to investigate the production of microbial biomass as a source of feed protein led to a number of developments which may be regarded as the fourth stage in the progress of the industry. The largest mechanically stirred fermentation vessels developed during stage 3 were in the range 80,000 to 150,000 dm3. However, the relatively low selling price of microbial biomass necessitated its production in much larger quantities than other fermentation products in order for the process to be profitable. Also, hydrocarbons were considered as potential carbon sources which would result in increased oxygen demands and high heat outputs by these fermentations (see Chapters 4 and 9). These requirements led to the development of the pressure jet and pressure cycle fermenters which eliminated the need for mechanical stirring (see Chapter 7). Another feature of these potential processes was that they would have to be operated continuously if they were to be economic. At this time batch and fed-batch processes were common in the industry but the technique of growing an organism continuously by adding fresh medium to the vessel and removing culture fluid had been applied only to a very limited extent on a large scale. The brewers were also investigating the potential of continuous culture at this time, but its application in that industry was short-lived. Several companies persevered in the biomass field and a few processes came to fruition, of which the most long-lived was the ICI Pruteen animal feed process which utilized a continuous 3,000,000-dm3 pressure cycle fermenter for the culture of Methylophilus meth-ylotrophus with methanol as carbon source (Smith, 1981; Sharp, 1989). The operation of an extremely large continuous fermenter for time periods in excess of 100 days presented a considerable aseptic operation problem, far greater than that faced by the antibiotic industry in the 1940s. The aseptic operation of fermenters of this type was achieved as a result of the high standards of fermenter construction, the continuous sterilization of feed streams and the utilization of computer systems to control the sterilization and operation cycles, thus minimizing the possibility of human error. However, although the Pruteen process was a technological triumph it became an economic failure because the product was out-priced by soybean and fishmeal. Eventually, in 1989, the plant was demolished, marking the end of a short, but very exciting, era in the fermentation industry.
Whilst biomass is a very low-value, high-volume product, the fifth stage in the progress of the industry resulted in the establishment of very high-value, low-volume products. The developments in in vitro genetic manipulation, commonly known as genetic engineering, enabled the expression of human and mammalian genes in micro-organisms, thereby enabling the large scale production of human proteins which could then be used therapeutically. According to Dykes (1993) it was the small, venture-capital biotechnology companies that pioneered the development of heterologous proteins for therapeutic use. The established pharmaceutical companies used the new genetic engineering techniques to help in the discovery of natural products and in the rational design of drugs; for example, mammalian receptor proteins have been cloned and used in in vitro detection systems.
Table 3.8 (Chapter 3) lists the recombinant proteins licensed for therapeutic use. According to Dykes (1993), insulin and human growth hormone have been the two most successful products but other products have far greater potential. Erythropoietin and the myeloid colony stimulating factors (CSFs) control the production of blood cells by stimulating the proliferation, differentiation and activation of specific cell types. Erythropoietin has been used to treat renal-failure anaemia and may have application in the treatment of the platelet deficiency associated with cancer chemotherapy; it is expected to become the top-selling therapeutic protein by the mid-1990s with annual sales of around $1200 million (Dykes, 1993). Granulocyte-colony stimulating factor (G-CSF), which is used during cancer chemotherapy, generated sales of over $230 million in 1991 and, due to other uses, its sales could reach $1000 million by 1996. A number of different growth factors are involved in wound healing and recombinant forms of these proteins would be expected to yield significant returns during the 1990s.
The commercial exploitation of recombinant proteins has necessitated the design of contained production facilities. Thus, these processes are drawing on the experience of vaccine fermentations where pathogenic organisms have been grown on relatively large scales. Also, recombinant proteins have been classified as bio-logicals, not as drugs, and thus come under the same regulatory authorities as do vaccines. The major difference between the approval of drugs and biologicals s that the process for the production of a biological iihim' be precisely specified and carried out in a facility that has been inspected and licensed by the regulatory -luthority. which is not the case for the production of drugs (antibiotics, for example) (Bader, 1992). Thus, any changes which a manufacturer wishes to incorporate into a licensed process must receive regulatory approval. For drugs, only major changes require approval prior to implementation. The result of these containment and regulatory requirements is that the cost of developing a recombinant protein process is extremely high. Buckland (1992) illustrated this point in his claim that "It now costs as much to build a 3000 dm3 scale facility for Biologies as for a 200,000 dm3 scale facility for an antibiotic. Also, even though titres are now reasonable for a recombinant protein (1 g dnT3), the cost of manufacture kg"1 of bulk drug is about two orders of magnitude higher than that of an antibiotic at 10 g dm ~3 titre". Also, the development time for a recombinant protein is considerably longer than that for an antibiotic. For example, Bader (1992) claimed that it is feasible for an antibiotic plant to begin production four years after the initiation of the plant design whereas seven years would be required before a recombinant protein could be produced.
The exploitation of genetic engineering approximately coincided with another major development in biotechnology which influenced the progress of the fermentation industry — the production of monoclonal antibodies. The availability of monoclonal antibodies opened the door to sophisticated analytical techniques and raised hopes for their use as therapeutic agents. Although the promise of therapeutic agents has yet to be realized (only one monoclonal antibody has been licensed for clinical use, OKT3, used in the treatment of acute renal allograft rejection (Webb, 1993)), their use as tools in biological research has increased exponentially. Thus, animal cell culture processes were established to produce monoclonals on a commercial scale. Subsequently, animal cells were also used as hosts for the production of some human proteins, especially where post-translational modification was essential for protein activity. Although these animal cell processes were based on microbial fermentation technology a number of novel problems had to be solved — animal cells are extremely fragile compared with microbial cells, the achievable cell density is very much less than in a microbial process and the media are very complex. These aspects are considered in Chapters 4 and 7.
The outstanding developments in recombinant fermentations (stage 5) have tended to overshadow the progress which has been made in recent years in establishing new fermentations based on conventional microbial products (the continuing development of stage 4). However, the appreciation by the pharmaceutical industry that the activity of microbial metabolites extended well beyond antibacterials has resulted in a number of new microbial products reaching the marketplace in the late 1980s and early 1990s. Buckland (1992) listed four secondary metabolites which were launched in the 1980s: cyclosporin, an immunoregulant used to control rejection of transplanted organs; imipenem, a modified carbapenem which has the widest antimicrobial spectrum of any antibiotic; lovastatin, a drug used for reducing cholesterol levels and ivermectin, an anti-parasitic drug which has been used to prevent 'African River Blindness' as well as in veterinary practice. Buckland summarized these developments succinctly, "One of the best kept secrets (unintentionally kept as a secret) in the 1980s in Biochemical Engineering was that working on secondary metabolites was a fascinating, important and rewarding experience. Furthermore the four products listed added together have higher sales than all of the recombinant products added together". Thus, it is still relevant to heed Foster's warning (1949) "never underestimate the power of the microbe".
Was this article helpful?