By Julia Cino, PhD
Over the last several decades, geneticists have learned how to manipulate DNA to identify, excise, move and place genes into a variety of organisms that are quite different genetically from the source organism. A major use for many of these recombinant organisms is to produce proteins. Since many proteins are of immense commercial value, numerous studies have focused on finding ways to produce them inexpensively, easily and in a fully functional form.
The production of a functional protein is intimately related to the cellular machinery of the organism producing the protein. E. coli has been the factory of choice for the expression of many proteins because its genome has been fully mapped and the organism is easy to handle; grows rapidly; requires an inexpensive, easy-to-prepare medium for growth; and secretes protein into the medium which facilitates recovery of the protein. However, E. coli is a prokaryote and lacks intracellular organelles, such as the endoplasmic reticulum and the golgi apparatus that are present in eukaryotes, which are responsible for modifications of the proteins being produced. Many eukaryotic proteins can be produced in E. coli but are produced in a nonfunctional, unfinished form, since glycosylation or post-translational modifications do not occur. Therefore, researchers have recently turned to eukaryotic yeast and mammalian expression systems for protein production.
Pichia pastoris Expression System
One such eukaryotic yeast is the methanoltrophic Pichia pastoris. Pichia pastoris has been developed to be an outstanding host for the production of foreign proteins since its alcohol oxidase promoter was isolated and cloned; its transformation was first reported in 1985 [1,2]. Compared to other eukary otic expression systems, Pichia offers many advantages, because it does not have the endotoxin problem associated with bacteria nor the viral contamination problem of proteins produced in animal cell culture. Furthermore, P. pastoris can utilize methanol as a carbon source in the absence of glucose. The P. pastoris expression system uses the methanol-induced alcohol oxidase (AOX1) promoter, which controls the gene that codes for the expression of alcohol oxidase, the enzyme which catalyzes the first step in the metabolism of methanol. This promoter has been characterized and incorporated into a series of P. pastoris expression vectors. Since the proteins produced in P. pastoris are typically folded correctly and secreted into the medium, the fermentation of genetically engineered P. pastoris provides an excellent alternative to E. coli expression systems. A number of proteins have been produced using this system, including tetanus toxin fragment, Bordatella pertussis pertactin, human serum albumin and lysozyme. (3 - 7).
Minimizing Growth Limiting Factors
Another advantage of Pichia pastoris is a prolific growth rate. Therefore, it would seem easy enough to culture it in a shake flask. This seeming advantage, however, can pose a host of problems, including pH control, oxygen limitation, nutrient limitation and temperature fluctuation. Researchers at New Brunswick Scientific (Edison, NJ) found that by switching from a shaker to a fermentor, protein production in Pichia could be increased by over 140% (3). The fermentor enables dissolved oxygen (DO) levels to be raised, not just by increasing agitation, but by increasing air flow, by supplementing the air stream with pure oxygen, or by doing all three either in series or in parallel. Nutrient limitation can also be minimized, since fermentors can be run in fed-batch mode, where fresh media or grow th limiting nutrients can be pumped into the vessel at a rate that is capable of replenishing the nutrients that are depleted. Shakers can only run in a batch mode, meaning that the growth of the cells is limited by the nutrients present in the medium at the time of inoculation. The fermentors fed-batch mode further enables methanol flow rates to be controlled to condition the cells to the presence of the methanol, as well as provide methanol at the proper rate to allow addition of just enough methanol for protein synthesis while preventing excess methanol addition which can cause toxicity.
Researchers have found that optimum protein production in P. pastoris occurs at 30C, and that all protein expression ceases at 32C. However, high heat loads occur when P. pastoris is actively growing or expressing high levels of protein. In actively growing shake flask cultures, it is not uncommon for the temperature to increase 25C if left uncontrolled. Therefore, it is imperative that all aspects of the fermentor, including temperature controller, heat exchanger, vessel and piping, must be designed to regulate temperature for optimum protein production. In addition to the fermentors internal controller, an external bioprocessing software (BioCommand, New Brunswick Scientific) is routinely used to supervise the process, as well as to provide optimal nutrient feed rates, based on either the current status of the culture or to actuate pre-determined control scenarios.
Research was conducted in BioFlo 3000 benchtop fermentors (New Brunswick Scientific) (Figure 1) with interchangeable, autoclavable vessels of 1.25 to 10 L working volume, as well as in a BioFlo 4500 fermentors with sterilizable-in-place vessels of 15 L and 20 L working volume, (New Brunswick Scientific ). However, these procedures can be adapted to other size fermentors thereby making the protocols scalable. In the authors laboratories, P. pastoris fermentations are run as multi-stage fed-batch processes with oxygen supplementation (Table 1). Here, oxygen is supplied automatically to meet the dissolved oxygen requirements for highdensity cell growth.
Method for a Typical Culture
A frozen vial of 1 ml P. pastoris sample was inoculated into a 1 L shake flask with 150 mL Yeast Nitrogen Base (YNB)-glycerol medium. A variety of genetically engineered P. pastoris strains were used, many of which are slow growing on methanol (muts) and engineered to produce proteins of interest. The culture was incubated at 30C, 240 rpm, for 14 hours in an environmental incubator shaker (New Brunswick Scientific). The entire 150 mL volume of inoculum was transferred to a 3.3 L fermentor vessel (total volume) containing 1.5 L of basal salts medium (see media components, Table 2) plus 4.4 mL/L trace metal solution (4). The temperature was controlled at 30C. The dissolved oxygen was set at 30% and pH is at 5.0. Ammonium hydroxide solution (30%) was used as the base solution to adjust the pH. After 20 hours of batch culture, the optical density (OD) reaches 42. The glycerol fed-batch process was then initiated. The feeding medium consisted of 50% glycerol and 12 mL/L of trace metal solution. The feed rate was 24 mL/L/h, which was adjusted automatically based on the DO reading. DO control was maintained by the proportional integral derivitive (PID) cascade controller, which changes the speed of agitation. Pure oxygen was automatically supplied to the fermentor to keep the DO level at the setpoint after the agitation speed reached the maximum allowable setpoint. After the growth phase, a half-hour carbon-source starvation period was establish ed before the culture was switched to the production phase.
The production phase (methanol feeding) was started after 43 hours of cell growth. The production feed medium consists of 100% methanol and 12 mL/L trace metal solution. Feeding rates were divided into three stages: 6 hr induction, 48 hr in a high-feed-rate stage and 44 hr in a low-feed-rate stage. The feeding rate of the induction stage was ramped from 1 to 10.9 mL/L/hr, which was controlled by the computer program. Feeding rates in the high and low rate stages were 15 and 2 mL/L/hr respectively. The total volume of feed was 2 L. During the fermentation, oxygen demand can be quite high and oxygen was added to the air stream automatically. (Figure 2)
pH is usually adjusted to inhibit the activities of proteinases existing in the culture broth during the production phase. Furthermore, since a host strain that is protease deficient was used, it was not necessary to change the pH level when culture was shifted from cell growth phase to production phase. It has been found that pH 5 is the optimal for cell metabolism and cell growth and that the oxygen consumption rate is higher at that pH. Using this protocol, optical densities of up to 630 can be obtained. Protein expression, of course, varies with the particular protein being expressed.
Although not without problems related to its culture, P. pastoris culture protocols are scalable and have become a powerful tool for the production of commercially valuable proteins.
More information on P. pastoris culture (expression vectors, protocols, etc.) can be obtained from Invitrogen, Inc., Carlsbad, CA and from Pichia Protocols published by Humana Press and edited by David R. Higgins and James M. Cregg
Figure 1: BioFlo 3000 fermentor with supervisory control system.
Figure 2: Pure oxygen s upplementation profile of a P. pastoris culture showing the increased oxygen requirement of the culture.
MEDIUM COMPONENTS AND FEED SOLUTIONS
H3PO4 - 27 ml/L
CaSO4. 2 H2O - 0.9 g/L
K2SO4 - 18 g/L
MgSO4. H2O -15 g/L
KOH - 4.13 g/L
Trace Metals Solution - 4.4 ml/L
Glycerol - 40 g/L
Trace Metals Solution
Cupric sulfate . 5 H2O - 6.0 g/L
Sodium iodide - 0.08 g/L
Manganese sulfate . H2O - 3.0 g/L
Sodium molybdate - 0.2 g/L
Boric acid - 0.02 g/L
Cobalt chloride - 0.5 g/L
Zinc chloride - 20 g/L
Ferrous sulfate . 7 H2O - 65.0 g/L
Biotin - 0.2 g/L
Sulfuric acid - 5.0 ml
Water - to 1.0 liter
Glycerol Feed Solution:
50% glycerol with 12 ml/L trace metals solution
Methanol Feed Solution:
100% glycerol with 12 ml/L trace metals solution
Dr. Cino is Product Manager, New Brunswick Scientific, PO Box 4005, Edison, NJ 08818-4005. Phone 800-631-5417. Fax: 732-287-4222. Web: www.nbsc.com. E-mail: firstname.lastname@example.org. The author acknowledges Yinliang Chen, Jeffrey Krol, Victor Sterkin of New Brunswick Scientific
1. Cregg, J.M., J. F. Tschopp, C. Stillman, R. Siegel, M. Akong, W. S. Craig, R. G. Buckholz, L. R. Madden, P. A. Kellaris, G. R. Davis, B. L. Smiley, J. Cruze, R. Torregrossa, G. Velicelebi and G. P. Thill. 1987. High-level expression and efficient assembly of hepatitis B surface antigen in the methylotrophic yeast, Pichia pastoris. BIO /TECHNOLOGY, Vol 5, 479-485
2. Brierley, R.A., C. Bussineau, R. Kosson, A. Melton and R. S. Siegel. 1992. in Fermentation Development of Recombinant Pichia pastoris Expression the Heterologous Gene: Bovine Lysozyme, Annals New York Academy of Sciences, 350-362
3. Chen, Y., Krol, J., Cino, J., Freedman, D., White, C., and Komives, E., 1996. Continuous Production of Thrombomodulin from a Pichia pastoris Fermentation . Journal Chem. Tech. Biotechnol. Vol. 67, 143-148
4. Clare, J.J., F.B. Rayment, S.P. Ballantine, K. Sreekrishna and M.A.Romanos, 1991. Highlevel expression of tetanus toxin fragment C in Pichia pastoris strains containing multiple tandem integrations of the gene. BIO/TECHNOLOGY, Vol 9, 455-460
5. Cregg, J.M., T.S. Vedvick and W.C. Raschke. 1993. Recent advances in the expression of foreign genes in Pichia pastoris. BIO/TECHNOLOGY, Vol.11, 905
6. Digan, M. E., S. V. Lair, R. A. Brierley, R. S. Siegel, M. E. Williams, S. B. Ellis, P. A. Kellaris, S. A. Provow, W. S. Craig, G. Velicelebi and M. M. Harpold. 1989. Continuous production of a novel lysozyme via secretion from the yeast, Pichia pastoris. Vol 7, 160-164
7. Tschopp, J. F., G. Sverlow, R. Kosson, W. Craig and L. Grinna. 1987. High-level secretion of glycosylated invertase in the methylotrophic yeast, Pichia pastoris . BI0 /TECHNOLOGY, Vol 5, 1305
Agrawal, P., G. Koshy and M. Ramseier. 1989. An algorithm for operation a fed-batch fermentor at optimum specific-growth rate. Biot echnol. Bioeng., 33, 115-125
Aiba, S., S. Nagai and Y. Nishizaqa. 1976. Fed-batch culture of Saccharomyces cerevisiae : a perspective of computer control to enhance the productivity in bakers yeast cultivation. Biotechnol Bioeng., 18, 1001-1011
Bentley, W. E. and D. S. Kompala. 1989. A novel structured kinetic modeling approach for the analysis of plasmid instability in recombinant bacterial cultures. Biotechnol. Bioeng., 33, 49-61
Komives, E. A. 1994. Expression of Highly disulfide Bonded Proteins in Pichia pastoris, Structure. 2,1003-1005
Modak, J. M. and H. C. Lim. 1989. Simple nonsingular control approach to fed- batch fermentation optimization. Biotechnol. Bioeng., 33, 11-15
Shimizu, N., S. Fukuzono, K. Fujimori, N. Nishimura and Y. Odawara. 1988. Fed-batch cultures of reconbinant Escherichia coli with inhibitory subtance concentration monitoring. J. Ferment. Technol., 66, 2, 187-191
Wei, D., S. J. Paurulekar and W. A. Weigandin. 1990. Multivariable control of continuous and fed-batch bioreactors. Biotechnical Engineering VI, Ann. N.Y. Acad. Sci. Vol 589, 508-528
Wu, W., K. Chen and H. Chiou. 1985. On-line optimal control for fed-batch culture of Bakers yeast production. Biotechnol. Bioeng., 27, 756-760
Yamane, T., T. Kume, E. Sada and T. Takamatsu. 1977. A simple optimization technique for fed-batch culture. J. Ferment. Technol., 55,587-593
Yang, X. 1992. Optimization of a cultivation process for recombinant protein production by Escherichia coli . J. Biotechnol. 23, 271-289