Additional production systems 5231 Yeast

Attention has also focused upon a variety of additional production systems for recombinant biopharmaceuticals. Yeast cells (particularly Saccharomyces cerevisiae) display a number of characteristics that make them attractive in this regard. These characteristics include:

• their molecular biology has been studied in detail, facilitating their genetic manipulation;

• most are GRAS-listed organisms ('generally regarded as safe'), and they have a long history of industrial application (e.g. in brewing and baking);

• they grow relatively quickly in relatively inexpensive media, and their tough outer wall protects them from physical damage;

• suitable industrial-scale fermentation equipment/technology is already available;

• they possess the ability to carry out post-translational modifications of proteins.

The practical potential of yeast-based production systems has been confirmed by the successful expression of a whole range of proteins of therapeutic interest in such systems. However, a number of disadvantages relating to heterologous protein production in yeast have been recognized. These include:

• Although capable of glycosylating heterologous human proteins, the glycosylation pattern usually varies from the pattern observed on the native glycoprotein (when isolated from its natural source, or when expressed in recombinant animal cell culture systems).

• In most instances, expression levels of heterologous proteins remain less than 5 per cent of total cellular protein. This is significantly lower than expression levels typically achieved in recombinant E. coli systems.

Despite such potential disadvantages, several recombinant biopharmaceuticals now approved for general medical use are produced in yeast (S. cerevisiae) -based systems (Table 5.5). Interestingly, most such products are not glycosylated. The oligosaccharide component of glycoproteins produced in yeasts generally contains high levels of mannose. Such high mannose-

Table 5.5 Recombinant therapeutic proteins approved for general medical use that are produced in S. cerevisiae. All are subsequently discussed in the chapter indicated

Trade name





Engineered short-acting insulin

Diabetes mellitus



Granulocyte macrophage CSF (GM-CSF)

Bone marrow transplantation


Recombivax, Comvax,

All vaccine preparations



Engerix B, Tritanrix-

containing rHBsAg as

HB, Infanrmix,

one component

Twinrix, Primavax,


Revasc, Refludan





Urate oxidase





Diabetic ulcers


type glycosylation patterns generally trigger their rapid clearance from the blood stream. Such products, therefore, would be expected to display a short half-life when parenterally administered to humans, and some yeast sugar motifs can be immunogenic in humans. Fungal production systems

Fungi have elicited interest as heterologous protein producers, as many have a long history of use in the production of various industrial enzymes such as a-amylase and glucoamylase. Suitable fermentation technology, therefore, already exists. In general, fungi are capable of high-level expression of various proteins, many of which they secrete into their extracellular media. The extracellular production of a biopharmaceutical would be distinctly advantageous in terms of subsequent downstream processing. Fungi also possess the ability to carry out post-translational modifications. Patterns of glycosylation achieved can, however, differ from typical patterns obtained when a glycoprotein is expressed in a mammalian cell line. Again, this can trigger a reduction is serum half-life or immunological complications in humans.

Most fungal host strains also naturally produce significant quantities of extracellular proteases, which can potentially degrade the recombinant product. This difficulty can be partially overcome by using mutant fungal strains secreting greatly reduced levels of proteases. Although researchers have produced a number of potential therapeutic proteins in recombinant fungal systems, no biop-harmaceutical produced by such means has thus far sought/gained marketing approval. Transgenic animals

The production of heterologous proteins in transgenic animals has gained much attention in the recent past. The generation of transgenic animals is most often undertaken by directly microinject-ing exogenous DNA into an egg cell. In some instances, this DNA will be stably integrated into

Table 5.6 Proteins of actual/potential therapeutic use that have been produced in the milk of transgenic animals


Animal species

Expression level in milk (mg l 1)







Factor VIII



Factor IX





20 000







Antithrombin III


14 000

Human a-lactalbumin






Protein C






the genetic complement of the cell. After fertilization, the ova may be implanted into a surrogate mother. Each cell of the resultant transgenic animal will harbour a copy of the transferred DNA. As this includes the animal's germ cells, the novel genetic information introduced can be passed on from one generation to the next.

A transgenic animal harbouring a gene coding for a pharmaceutical^ useful protein could become a live bioreactor-producing the protein of interest on an ongoing basis. In order to render such a system practically useful, the recombinant protein must be easily removable from the animal, in a manner which would not be injurious to the animal (or the protein). A simple way of achieving this is to target protein production to the mammary gland. Harvesting of the protein thus simply requires the animal to be milked.

Mammary-specific expression can be achieved by fusing the gene of interest with the promoter-containing regulatory sequence of a gene coding for a milk-specific protein. Regulatory sequences of the whey acid protein (WAP), P-casein and a- and P-lactoglobulin genes have all been used to date to promote production of various pharmaceutical proteins in the milk of transgenic animals (Table 5.6).

One of the earliest successes in this regard entailed the production of human tPA in the milk of transgenic mice. The tPA gene was fused to the upstream regulatory sequence of the mouse WAP, the most abundant protein found in mouse milk. More practical from a production point of view, was the subsequent production of tPA in the milk of transgenic goats, again using the murine WAP gene regulatory sequence to drive expression (Figure 5.2). Goats and sheep have proven to be the most attractive host systems, as they exhibit a combination of attractive characteristics. These include:

• high milk production capacities (Table 5.7);

• ease of handling and breeding, coupled to well-established animal husbandry techniques.

A number of additional general characteristics may be cited that render attractive the production of pharmaceutical proteins in the milk of transgenic farm animals. These include:

• Ease of harvesting of crude product, which simply requires the animal to be milked.

Mammary cell

Mammary cell

V Upstream / processing

V Upstream / processing

Downstream processing

Figure 5.2 The production and purification of tPA from the milk of transgenic goats. (WAP promoter: murine whey acid promoter). The downstream processing procedure yielded in excess of an 8000-fold purification factor with an overall product yield of 25 per cent. The product was greater than 98 per cent pure, as judged by sodium dodecyl sulfate (SDS) electrophoresis

• Pre-availability of commercial milking systems, already designed with maximum process hygiene in mind.

• Low capital investments (i.e. relatively low-cost animals replace high-cost traditional fermentation equipment) and low running costs.

• High expression levels of proteins are potentially attained. In many instances, the level of expression exceeds 1 g protein/litre milk. In one case, initial expression levels of 60 g L1 were observed, which stabilized at 35 g L1 as lactation continued (the expression of the a1-antitrypsin gene, under the influence of the ovine P-lactoglobulin promoter, in a transgenic sheep). Even at expression levels of 1 g L1, one transgenic goat would produce a similar quantity of product in 1 day as would be likely recoverable from a 50-100 l bioreactor system.

Table 5.7 Typical annual milk yields as well as time lapse between generation of the transgene embryo and first product harvest (first lactation) of indicated species


Annual milk yield (l)

Time to first production batch (months)
















• Ongoing supply of product is guaranteed (by breeding).

• Milk is biochemically well characterized, and the physicochemical properties of the major native milk proteins of various species are well known. This helps rational development of appropriate downstream processing protocols (Table 5.8).

Despite the attractiveness of this system, a number of issues remain to be resolved before it is broadly accepted by the industry. These include:

• Variability of expression levels. Although in many cases the expression levels of heterologous proteins exceed 1 g T1, expression levels as low as 1.0 mg T1 have been obtained in some instances.

• Characterization of the exact nature of the post-translational modifications the mammary system is capable of undertaking. For example, the carbohydrate composition of tPA produced in this system differs from the recombinant enzyme produced in murine cell culture systems.

• Significant time lag between the generation of a transgenic embryo and commencement of routine product manufacture. Once a viable embryo containing the inserted desired gene is generated, it must first be brought to term. This gestation period ranges from 1 month for rabbits to 9 months for cows. The transgenic animal must then reach sexual maturity before breeding (5 months for rabbits, 15 months for cows). Before they begin to lactate (i.e. produce the recombinant product), they must breed successfully and bring their offspring to term. The overall time

Table 5.8 Some physicochemical properties of the major (bovine) milk proteins





Serum albumin



25 (total)





(g n

Mass (kDa)












Isoelectric point







K-casein only





lag to routine manufacture can, therefore, be almost 3 years in the case of cows or 7 months in the case of rabbits. Furthermore, if the original transgenic embryo turns out to be male, a further delay is encountered, as this male must breed in order to pass on the desired gene to daughter animals, who will then eventually produce the desired product in their milk.

Another general disadvantage of this approach relates to the use of the microinjection technique to introduce the desired gene into the pronucleus of the fertilized egg. This approach is inefficient and time consuming. There is no control over issues such as if/where in the host genomes the injected gene will integrate. Overall, only a modest proportion of manipulated embryos will culminate in the generation of a healthy biopharmaceutical-producing animal.

A number of alternative approaches are being developed that may overcome some of these issues. Replication-defective retroviral vectors are available that will more consistently (a) deliver a chosen gene into cells and (b) ensure chromosomal integration of the gene. A second innovation is the application of nuclear transfer technology.

Nuclear transfer entails substituting the genetic information present in an unfertilized egg with donor genetic information. The best-known product of this technology is 'Dolly' the sheep, produced by substituting the nucleus of a sheep egg with a nucleus obtained from an adult sheep cell. (Genetically, therefore, Dolly was a clone of the original 'donor' sheep.) An extension of this technology applicable to biopharmaceutical manufacture entails using a donor cell nucleus previously genetically manipulated so as to harbour a gene coding for the biopharmaceutical of choice. The technical viability of this approach was proven in the late 1990s upon the birth of two transgenic sheep, 'Polly' and 'Molly'. The donor nucleus used to generate these sheep harboured an inserted (human) blood factor IX gene under the control of a milk protein promoter. Both produced significant quantities of human factor IX in their milk. The first (and only, at the time of writing) such product to gain approval anywhere in the world is 'Atryn', a recombinant human antithrombin that is produced in the milk of transgenic goats. Atryn is used to treat thromboembolism in surgery of people with congenital antithrombin deficiency.

In addition to milk, a range of recombinant proteins have been expressed in various other targeted tissues/fluids of transgenic animals. Antibodies and other proteins have been produced in the blood of transgenic pigs and rabbits. This mode of production, however, is unlikely to be pursued industrially for a number of reasons:

• Only relatively low volumes of blood can be harvested from the animal at any given time point.

• Serum is a complex fluid, containing a variety of native proteins. This renders purification of the recombinant product more complex.

• Many proteins are poorly stable in serum.

• The recombinant protein could have negative physiological side effects on the producer animal.

Therapeutic proteins have also been successfully expressed in the urine and seminal fluid of various transgenic animals. Again, issues of sample collection, volume of collected fluid and the appropriateness of these systems render unlikely their industrial-scale adoption. One system that does show i ndustrial promise, however, is the targeted production of recombinant proteins in the egg white of transgenic birds. Targeted production is achieved by choice of an appropriate egg-white protein promoter sequence. A single egg white typically contains 4 g protein, of which approximately half is derived from the ovalbumin gene. Using an ovalbumin promoter-based expression system, a single hen could produce as much as 300 g recombinant product annually. Large quantities of product can, therefore, potentially accumulate in the egg, which can then be collected and processed with relative ease. The traditional use of eggs as incubation systems in the production of some viral vaccines would also render regulatory and some manufacturing issues more straightforward. Many of both the potential strengths and weaknesses of this system are encapsulated in recent findings relating to egg-produced antibodies.

Fully assembled and functional antibodies have recently been produced in eggs at levels of several milligrams per egg. When compared with the same antibodies produced by traditional CHO-based cell culture, some differences in product glycosylation detail were evident, although antigen binding affinity was not altered. The glycocomponent plays a number of roles, including influencing the ability of the antibody to interact with various immune effector cells, thereby triggering antibody-dependent cell cytoxicity (ADCC), which is important for effective therapeutic functioning. In this particular case, the altered glycosylation detail actually appeared to enhance ADCC. However, tests in mice showed the egg-derived antibody to have a much-reduced serum half-life (reduced from around 200 h to 100 h) and issues of antigenicity still remain to be assessed. Transgenic plants

The production of pharmaceutical proteins using transgenic plants has also gained some attention over the last decade. The introduction of foreign genes into plant species can be undertaken by a number of means, of which Agrobacterium-based vector-mediated gene transfer is most commonly employed. Agrobacterium tumefaciens and Agrobacterium rhizogenes are soil-based plant pathogens. Upon injection, a portion of Agrobacterium Ti plasmid is translocated to the plant cell and is integrated into the plant cell genome. Using such approaches, a whole range of therapeutic proteins have been expressed in plant tissue (Table 5.9). Depending upon the specific promoters used, expression can be achieved uniformly throughout the whole plant or can be limited to, for example, expression in plant seeds.

Plants are regarded as potentially attractive recombinant protein producers for a number of reasons, including:

Table 5.9 Some proteins of potential/actual therapeutic interest that have been expressed (at laboratory level) in transgenic plants


Expressed in

Production levels achieved



0.003% of total soluble plant




0.02% of soluble leaf protein



0.1% of leaf weight



Not listed



0.000 02% of fresh weight



250 ng ml-1 extract



1.0% of seed weight

Hepatitis B surface antigen


0.007% of soluble leaf protein

Antibodies/antibody fragment



• cost of plant cultivation is low;

• harvest equipment/methodologies are inexpensive and well established;

• proteins expressed in seeds are generally stable in the seed for prolonged periods of time (often years);

• plant-based systems are free of human pathogens (e.g. HIV).

However, a number of potential disadvantages are also associated with the use of plant-based expression systems, including:

• variable/low expression levels sometimes achieved;

• potential occurrence of post-translational gene silencing (a sequence-specific mRNA degradation mechanism);

• glycosylation patterns achieved differ significantly from native human protein glycosylation patterns, and plant glycoforms are invariably immunogenic in humans;

• the potential presence of biologically active plant metabolites (e.g. alkaloids) that would 'contaminate' the crude product;

• environmental/public concerns relating to potential environmental escape of genetically altered plants;

• seasonal/geographical nature of plant growth.

For these reasons, as well as the fact that additional tried-and-tested expression systems are already available, production of recombinant therapeutic proteins in transgenic plant systems has not as yet impacted significantly on the industry.

The most likely focus of future industry interest in this area concerns the production of oral vaccines in edible plants/fruit, such as tomatoes and bananas. Animal studies have clearly shown that ingestion of transgenic plant tissue expressing recombinant subunit vaccines (see Chapter 13 for a discussion of subunit vaccines) induces the production of antigen-specific antibody responses not only in mucosal secretions, but also in the serum. The approach is elegant, in that direct consumption of the plant material provides an inexpensive, efficient and technically straightforward mode of large-scale vaccine delivery, particularly in poorer world regions. However, several hurdles hindering the widespread application of this technology include:

• the immunogenicity of orally administered vaccines can vary widely;

• the stability of antigens in the digestive tract varies widely;

• the genetics of many potential systems remain poorly characterized, leading to inefficient transformation systems and low expression levels.

CaroRX and Merispace are (at the time of writing) two of the lead plant-produced biopharmaceu-ticals. Both are in phase II clinical trials. Neither is destined for parenteral administration. CaroRX is a recombinant antibody that targets Streptococcus mutans, a major causative agent of bacterial tooth decay. Binding prevents bacterial adherence to teeth, and the product is being developed for regular topical administration. Merispace is a recombinant mammalian gastric lipase enzyme produced in transgenic corn. It is intended to be used orally to counteract lipid malabsorption relating to exocrine pancreatic insufficiency caused by conditions such as cystic fibrosis and chronic pancreatitis.

In order to overcome environmental concerns in particular, some companies are investigating the use of engineered plant cell lines as opposed to intact transgenic plants in the context of biopharmaceutical production. One company (DowAgroSciences) gained approval in 2006 for a veterinary subunit vaccine against Newcastle disease in poultry produced by such means. Insect cell-based systems

A wide range of proteins have been produced at laboratory scale in recombinant insect cell culture systems. The approach generally entails the infection of cultured insect cells with an engineered baculovirus (viral family that naturally infect insects) carrying the gene coding for the desired protein placed under the influence of a powerful viral promoter. Amongst the systems most commonly employed are:

• the silkworm virus Bombyx mori nuclear polyhedrovirus (BmNPV) in conjunction with cultured silkworm cells (i.e. Bombyx mori cells);

• the virus Autographa californica nuclear polyhedrovirus (AcNPV), in conjunction with cultured armyworm cells (Spodoptera frugiperda cells).

Baculovirus/insect cell-based systems are cited as having a number of advantages, including:

• High-level intracellular recombinant protein expression. The use of powerful viral promoters, such as promoters derived from the viral polyhedrin or P10 genes, can drive recombinant protein expression levels to 30-50 per cent of total intracellular protein.

• Insect cells can be cultured more rapidly and using less expensive media compared with mammalian cell lines.

• Human pathogens (e.g. HIV) do not generally infect insect cell lines.

However, a number of disadvantages are also associated with this production system, including:

• Targeted extracellular recombinant production generally results in low-level extracellular accumulation of the desired protein (often in the milligram per litre range). Extracellular production simplifies subsequent downstream processing, as discussed later in this chapter.

• Post-translational modifications, in particular glycosylation patterns, can be incomplete and/or can differ very significantly from patterns associated with native human glycoproteins.

Therapeutic proteins successfully produced on a laboratory scale in insect cell lines include hepatitis B surface antigen, IFN-y and tPA. To date, no therapeutic product produced by such means has been approved for human use. Two veterinary vaccines, however, have: 'Bayovac CSF E2' and 'porcilis pesti' are both subunnit vaccines (Chapter 13) containing the E2 surface antigen protein of classical swine fever virus as active ingredient. The vaccines are administered to pigs in order to immunize against classical swine fever; an overview of their manufacture is provided in Figure 5.3.

Figure 5.3 Generalized overview of the industrial-scale manufacture of recombinant E2 classical swine-fever-based vaccine, using insect cell culture production systems. Clean (uninfected) cells are initially cultured in 500-1000 l bioreactors for several days, followed by viral addition. Upon product recovery, viral inactivating agents such as beta propiolactone or 2-bromoethyl-imminebromide are added in order to destroy any free viral particles in the product stream. No chromatographic purification is generally undertaken, as the product is substantially pure; the cell culture medium is protein free and the recombinant product is the only protein exported in any quantity by the producer cells. Excipients added can include liquid paraffin and polysorbate 80 (required to generate an emulsion). Thiomersal may also be added as a preservative. The final product generally displays a shelf life of 18 months when stored refrigerated

Figure 5.3 Generalized overview of the industrial-scale manufacture of recombinant E2 classical swine-fever-based vaccine, using insect cell culture production systems. Clean (uninfected) cells are initially cultured in 500-1000 l bioreactors for several days, followed by viral addition. Upon product recovery, viral inactivating agents such as beta propiolactone or 2-bromoethyl-imminebromide are added in order to destroy any free viral particles in the product stream. No chromatographic purification is generally undertaken, as the product is substantially pure; the cell culture medium is protein free and the recombinant product is the only protein exported in any quantity by the producer cells. Excipients added can include liquid paraffin and polysorbate 80 (required to generate an emulsion). Thiomersal may also be added as a preservative. The final product generally displays a shelf life of 18 months when stored refrigerated

Figure 5.4 Overview of the industrial manufacture of the IFN-m product 'Vibragen Omega'. Refer to text for details

An alternative insect cell-based system used to achieve recombinant protein production entails the use of live insects. Most commonly, live caterpillars or silkworms are injected with the engineered baculovirus vector, effectively turning the whole insect into a live bioreactor. One veterinary biopharmaceutical, Vibragen Omega, is manufactured using this approach, and an overview of its manufacture is outlined in Figure 5.4. Briefly, whole, live silkworms are introduced into pre-sterile cabinets and reared on laboratory media. After 2 days, each silkworm is inoculated with engineered virus using an automatic microdispenser. This engineered silkworm polyhedrosis virus harbours a copy of cDNA coding for feline IFN-ro. During the subsequent 5 days of rearing, a viral infection is established and, hence, recombinant protein synthesis occurs within the silkworms. After acid extraction, neutralization and clarification, the recombinant product is purified chromatographically. A two-step affinity procedure using blue sepharose dye affinity and copper chelate sepharose chromatography is employed. After a gel filtration step, excipients (sorbitol and gelatin) are added and the product is freeze-dried after filling into glass vials.

Diabetes 2

Diabetes 2

Diabetes is a disease that affects the way your body uses food. Normally, your body converts sugars, starches and other foods into a form of sugar called glucose. Your body uses glucose for fuel. The cells receive the glucose through the bloodstream. They then use insulin a hormone made by the pancreas to absorb the glucose, convert it into energy, and either use it or store it for later use. Learn more...

Get My Free Ebook

Post a comment