High-yield production of biologically active recombinant protein in shake flask culture by combination of enzyme-based glucose delivery and increased oxygen transfer
© Ukkonen et al; licensee BioMed Central Ltd. 2011
Received: 5 October 2011
Accepted: 12 December 2011
Published: 12 December 2011
This report describes the combined use of an enzyme-based glucose release system (EnBase®) and high-aeration shake flask (Ultra Yield Flask™). The benefit of this combination is demonstrated by over 100-fold improvement in the active yield of recombinant alcohol dehydrogenase expressed in E. coli. Compared to Terrific Broth and ZYM-5052 autoinduction medium, the EnBase system improved yield mainly through increased productivity per cell. Four-fold increase in oxygen transfer by the Ultra Yield Flask contributed to higher cell density with EnBase but not with the other tested media, and consequently the product yield per ml of EnBase culture was further improved.
Shake flasks are the most commonly applied laboratory-scale cultivation vessels for production of recombinant proteins due to their very simple set-up and operation. However, on the downside shake flasks do not provide an ideal environment for recombinant protein expression due to relatively low aeration capacity and the batch mode of cultivation. Therefore, the yield of heterologously expressed proteins in shake flasks is usually much lower than in bioreactors where high oxygen transfer rates can be provided and growth rate can be regulated by fed-batch mode of operation based on growth rate limiting glucose feeding.
While it is virtually impossible to apply glucose feeding in a shake flask by an external pump like in a bioreactor, fed-batch-like conditions can be created in simple shaken cultures by an internal glucose release system, called EnBase® technology [1, 2]. The EnBase system is based on a soluble polysaccharide in the medium, from which glucose is released through the action of a specific enzyme. In this way, concentration of the enzyme controls the rate at which glucose becomes available in the medium. Enhancement of recombinant protein yield in shaken microbial cultures has been demonstrated with the first generation EnBase system , based on a solid starch gel at the bottom of the cultivation vessel, as well as the second generation system [2–5] comprising no solid matrix but a soluble polysaccharide in the liquid medium. The second generation EnBase includes supplementation with complex nutrients, and is therefore not strictly a fed-batch with glucose as the sole available carbon source. The controlled glucose feeding improves recombinant protein expression, especially regarding protein solubility, compared to conventional complex media . The glucose control also enables growth to higher cell densities, which contributes to increased volumetric productivity.
The controlled glucose feeding partly addresses the problem of limited oxygen transfer rate in shake flasks. Growth rate can be adjusted according to the aeration capacity of the vessel, thus preventing the detrimental effects that oxygen limitation may have on cell growth and protein productivity. However, even if EnBase enables aerobic growth at higher cell densities than batch cultures, oxygen will eventually also become limiting when cell density reaches a certain limit. Therefore, growth and protein production in EnBase cultures are expected to be further improved by increased oxygen transfer capacity of the vessel.
In this study we demonstrate the combined use of the controlled enzymatic glucose release technology, EnBase, and a high aeration capacity vessel, the Ultra Yield Flask, for high yield production of active recombinant protein. Oxygen transfer coefficients were experimentally determined for the Ultra Yield Flask and a conical glass shake flask, and the influence of the different oxygen transfer capacities on production of correctly folded recombinant protein was evaluated in the EnBase medium, Terrific Broth and ZYM-5052 lactose autoinduction medium.
Results and Discussion
Determination of OTR and KLa
Oxygen transfer rates (OTR) and oxygen transfer coefficients (KLa) in 500 ml Ultra Yield Flask and 1000 ml Erlenmeyer flask.
OTR [mmol l-1 h-1]
500 ml Ultra Yield Flask
118.1 ± 7.3
474.1 ± 29.2
1000 ml Erlenmeyer Flask
30.5 ± 4.5
122.4 ± 18.1
Use of the EnBase medium and Ultra Yield Flask for recombinant protein production
Heterologous expression of Lactobacillus alcohol dehydrogenase (Adh) in E. coli RB791 was used as a model system to study the effects of culture medium and flask type on recombinant protein production. Terrific Broth and ZYM-5052 lactose autoinduction medium were used in parallel with the EnBase medium, and cultivations in all three media were performed in both 1000 ml Erlenmeyer flask and 500 ml Ultra Yield (UY) Flask with a broth volume of 100 ml.
Surprisingly, in TB and ZYM-5052 the flask type did not affect either cell density or medium pH (Figure 2, 3). Final OD600 of 30 was obtained in both media, corresponding to CDW of 8.1 g l-1. This is relatively high compared to cell densities typically obtained in shake flask batch cultures, but might be explained by the choice of media and relatively good aeration also in the Erlenmeyer flasks due to low flask filling volume (10%) and use of the special AirOtop membrane seals. pH in ZYM-5052 was stably maintained at neutral range, while in TB pH increased up to 8.5, indicating high ammonia production due to utilization of amino acids as carbon source.
A notable difference between Erlenmeyer and UY flasks was that in Erlenmeyer the specific productivity in EnBase increased slightly from 6 h sample to 24 h sample, while in the EnBase UY flasks specific productivity dropped from 6 h to 24 h by approximately 10-30%. As a result, activity per ml in UY flasks slightly decreased during the extended expression period despite increase in cell density. A possible explanation for the decrease in specific productivity could be exhaustion of ampicillin and consequent overgrowth by plasmid-free non-producing cells in the late phases of EnBase cultivation. The cultivation period in EnBase medium is relatively long (41 h), cell densities are high and no ampicillin was added after the initial dose at inoculation. As the mechanism of ampicillin resistance is secretion of β-lactamase into the extracellular space to degrade ampicillin, high cell densities result in high β-lactamase activity in the medium and the selective pressure will be lost . Therefore, the higher cell density in the UY flasks could result in faster removal of the selective pressure and, consequently, larger fraction of the population being plasmid-free in the end of cultivation. If this is the reason for decreasing activity per cell, the yield in the UY flask might be increased simply by addition of more ampicillin at induction, or by using a vector that applies a more stable antibiotic for selection .
The high increase in production of Adh in the soluble and active form by EnBase compared to TB and ZYM-5052 may be at least partly explained by growth rate. In TB and ZYM-5052 Adh expression took place in the exponential growth phase; OD600 increased by 10-fold during the 5 h period after induction of TB cultures, corresponding to a doubling time of approximately 1.5 h. In ZYM-5052 growth rate was similar during the same time period. In EnBase cultures, by contrast, doubling time during the first 6 h after induction was approximately 4.5 to 5 h. As suggested previously , it is likely that the fast growth in batch media such as TB and ZYM-5052 is associated with a high protein synthesis rate that could result in improper folding and product aggregation, while in EnBase cultures the slower growth allows for slower protein synthesis that may better match the capacity of cellular protein folding machinery. However, a more detailed study on the influence of growth rate would be needed to confirm this. Apart from the growth rate, protein synthesis in TB is likely impeded by the apparently non-optimal ratio of nitrogen compounds and the primary carbon source (glycerol) that results in unfavorably high medium pH through the bacterial metabolism.
In terms of the effect of improved oxygen availability on cell density and volumetric protein yield in the EnBase system, our findings are in good agreement with an earlier report by Pilarek et al.  who demonstrated the use of oxygen-saturated liquid perfluorodecalin to enhance oxygen transfer into ml-scale multiwell plate cultures. Introduction of the liquid oxygen carrier into EnBase cultures resulted in 40% higher cell density and correspondingly increased volumetric protein yield in the same Adh-producing clone as used in our study. Therefore, multiwell plates with EnBase and perfluorodecalin constitute an efficient tool for high-throughput screening applications, whereas in shake flask scale the EnBase and Ultra Yield Flask system provides a more cost-efficient alternative to the use of perfluorodecalin or related chemicals. The fed-batch-like glucose control of EnBase provides easy scale-up from shake flask to fed-batch bioreactors [13–15], and the improved oxygen transfer by the Ultra Yield Flask brings the system even closer to the conditions of a stirred bioreactor with high oxygen transfer rate.
The combination of controlled glucose feeding and a cultivation vessel with highly increased oxygen transfer capacity is a very powerful tool for enhancement of recombinant protein production in simple shaken cultures. Compared to Terrific Broth and ZYM-5052 autoinduction medium, the EnBase medium with the enzymatic glucose release system enhanced the yield of active recombinant Adh by 113- and 77-fold, respectively. Further 1.6-fold improvement in total volumetric yield of EnBase cultures was achieved by the use of Ultra Yield Flask providing 4-fold higher oxygen transfer rate. In this case the higher product yield was due to increased cell density. Due to its very easy operation, the Ultra Yield Flask with the EnBase glucose feeding technology might represent a convenient alternative to laboratory scale bioreactors for high cell density growth and high-yield recombinant protein production, especially in high throughput applications. The fed-batch-like conditions and high oxygen transfer capacity also make the system ideal for process scale-up.
Determination of oxygen transfer coefficients
Where tstandard [h] is the time needed for complete oxidation in the beaker without prior shake flask incubation
tbeaker [h] is the time needed for complete oxidation in the beaker after shake flask incubation
tshake flask [h] is the incubation time in shake flask
OTRstandard [mmol l-1 h-1] is oxygen transfer rate in the beaker.
Where c(Na2SO3) [mmol l-1] is sodium sulfite concentration in the reaction solution
vO2 = 0.5, volumetric coefficient for oxygen in the reaction SO32- + 0.5 O2 → SO42-
tR [h] is the reaction time needed for complete oxidation.
Where cO2* [mmol l-1] is oxygen saturation concentration in the solution
cO2 [mmol l-1] is oxygen concentration in the liquid phase boundary.
where PO2 is the partial pressure of oxygen (0.21 atm)
H is Henry's law constant; at 30°C, H = 844.32 L · atm · mol-1 .
Lactobacillus alcohol dehydrogenase (Adh) was heterologously expressed in Escherichia coli RB791 [F-, IN(rrnD-rrnE1), λ-, lacIqL8] transformed with plasmid pQE30:adh.
EnBase medium was constituted by dissolving four EnPresso® medium tablets (BioSilta, Finland) into 100 ml of sterile water. The EnBase medium is composed of a mineral salt medium and phosphate buffer base (for detailed composition of the mineral salt medium see ) supplemented with some complex nutrients, trace elements solution and the soluble glucose polysaccharide. The enzyme (EnZ I'm; BioSilta) for glucose release from the soluble polysaccharide was added to concentration 0.6 U l-1 shortly before inoculation. At induction the culture was supplemented to a higher concentration of complex nutrients (peptone and yeast extract) by adding the EnPresso Booster tablet.
Terrific Broth (TB) medium contained (per liter): tryptone 12 g; yeast extract 24 g; K2HPO4 9.4 g; KH2PO4 2.2 g; 87% glycerol 4 ml.
ZYM-5052 autoinduction medium  contained (per liter): tryptone 10 g; yeast extract 5 g; Na2HPO4 3.56 g; KH2PO4 3.40 g; NH4Cl 2.68 g; Na2SO4 0.71 g; 87% glycerol 4 ml; glucose 0.5 g; lactose 2 g; trace elements solution 2 ml.
To maintain selective pressure, all media were supplemented with 100 μg ml-1 ampicillin. As high degree of foaming was observed in the UY flasks, Antifoam 204 (Sigma Aldrich) was added to all media in UY flasks at the time of induction.
E. coli RB791[pQE30:adh] was cultivated overnight on Luria-Bertani agar plates with 2 g l-1 glucose and 100 μg ml-1 ampicillin, harvested and stored as a glycerol stock at -70°C. All expression cultures were inoculated with the glycerol stock to OD600 of 0.10-0.15. Cultivations were always performed at 30°C in an orbital shaker with 25 mm offset and shaking speed of 250 rpm. Initial broth volume was 100 ml. The cultivation vessels were 500 ml Ultra Yield Flask (Thomson Instrument Company, USA) and 1000 ml conical (Erlenmeyer) glass flask. Both flask types were closed with air-permeable membranes, the AirOtop Enhanced Seals (Thomson Instrument Company, USA). A fresh membrane seal was changed every time the flask was opened for induction or sampling.
The EnBase cultures were induced after overnight cultivation (17 h) with 0.4 mM IPTG. At the same time, two EnPresso Booster tablets (BioSilta) were added to each 100 ml culture together with an additional dose of the EnZ I'm (0.6-6 U l-1). Samples were harvested for measurement of cell density, pH and product activity at 6 h and 24 h after induction.
Terrific Broth cultures were induced after 3 h incubation at OD600 = 1.2-1.5 with 0.4 mM IPTG. Samples were harvested at 5 h and 21 h from induction. Autoinduction cultures were incubated for a total of 24 h, and samples were harvested at 8 h and 24 h.
We have previously observed the optimal time of IPTG addition to be after overnight cultivation (15-18 h) for EnPresso, and the exponential growth phase (OD600 = 0.8-1.5) for TB (data not shown). This was the rationale for the different induction times used in these two media. Cultivations were performed in duplicates.
Cell density was recorded by measurement of optical density in 1 ml cuvettes at 600 nm. Cell dry weight (CDW) was determined at the end of cultivation in triplicate samples. The correlation between OD600 and CDW was CDW (g l-1) = 0.27 · OD600.
For pH measurement, 0.2 ml samples were harvested and the cells were spun down. pH was measured from the supernatant with IQ2400 pH probe (IQ Scientific).
For analysis of recombinant protein yield, 100 μl broth samples were centrifuged at 13300 rpm for 4 min at 4°C. Supernatants were discarded and the pellets were frozen at -20°C. After thawing on ice, pellets were resuspended in 100 μl of BugBuster (Novagen). 2 μl of Lysonase Bioprocessing Reagent (Novagen) was added to each sample to lyse the cells. Samples were then centrifuged at 13300 rpm for 4 min at 4°C to remove cell debris.
Proteins in the cell lysate were visualized on reducing SDS-PAGE gels stained with Coomassie Brilliant Blue. Total proteins (insoluble and soluble fraction) were analyzed from the lysates before centrifugation, and soluble protein fractions were analyzed from the lysate supernatant after removal of debris and insolubles by centrifugation.
where X is the dilution factor
Vtotal [ml] is the total volume of reaction mixture in the well
εNADPH is extinction coefficient for NADPH (6.22 mM-1 cm-1)
Vsample [ml] is sample volume
d [cm] is the light path (sample height)
One unit (U) of Adh activity is here defined as the amount of Adh required for conversion of 1 mM substrate (ethyl-4-chloroacetoacetate) min-1 at 20°C and pH 7.0.
This work has been supported by a fund from the Finnish Funding Agency for Technology and Innovation (Tekes) #1695/31/2010 to BioSilta Oy. The funding source had no role in study design, collection, analysis or interpretation of data, writing of the report or decision to submit the article for publication.
- Panula-Perälä J, Šiurkus J, Vasala A, Wilmanowski R, Casteleijn MG, Neubauer P: Enzyme controlled glucose auto-delivery for high cell density cultivations in microplates and shake flasks. Microb Cell Fact. 2008, 7: 31- 10.1186/1475-2859-7-31View ArticleGoogle Scholar
- Krause M, Ukkonen K, Haataja T, Ruottinen M, Glumoff T, Neubauer A, Neubauer P, Vasala A: A novel fed-batch based cultivation method provides high cell-density and improves yield of soluble recombinant proteins in shaken cultures. Microb Cell Fact. 2010, 9: 11- 10.1186/1475-2859-9-11View ArticleGoogle Scholar
- Nguyen V, Hatahet F, Salo K, Enlund E, Zhang C, Ruddock L: Pre-expression of a sulfhydryl oxidase significantly increases the yields of eukaryotic disulfide bond containing proteins expressed in the cytoplasm of E.coli. Microb Cell Fact. 2011, 10: 1- 10.1186/1475-2859-10-1View ArticleGoogle Scholar
- Ehrmann A, Richter K, Busch F, Reimann J, Albers SV, Sterner R: Ligand-induced formation of a transient tryptophan synthase complex with αββ subunit stoichiometry. Biochemistry. 2010, 49: 10842-10853. 10.1021/bi1016815View ArticleGoogle Scholar
- Tegel H, Yderland L, Boström T, Eriksson C, Ukkonen K, Vasala A, Neubauer P, Ottosson J, Hober S: Parallel production and verification of protein products using a novel high-throughput screening method. Biotechnol J. 2011, 6: 1018-1025. 10.1002/biot.201000430View ArticleGoogle Scholar
- Brodsky O, Cronin C: Economical parallel protein expression screening and scale-up in Escherichia coli. J Struct Funct Genomics. 2006, 7: 101-108.View ArticleGoogle Scholar
- Peti W, Page R: Strategies to maximize heterologous protein expression in Escherichia coli with minimal cost. Protein Expr Purif. 2007, 51: 1-10. 10.1016/j.pep.2006.06.024View ArticleGoogle Scholar
- Glazyrina J, Materne E, Hillig F, Neubauer P, Junne S: Two-compartment method for determination of the oxygen transfer rate with electrochemical sensors based on sulfite oxidation. Biotechnol J. 2011, 6: 1003-1008. 10.1002/biot.201100281View ArticleGoogle Scholar
- Glazyrina J, Materne E, Dreher T, Storm D, Junne S, Adams T, Greller G, Neubauer P: High cell density cultivation and recombinant protein production with Escherichia coli in a rocking-motion-type bioreactor. Microb Cell Fact. 2010, 9: 42- 10.1186/1475-2859-9-42View ArticleGoogle Scholar
- Pierce J, Gutteridge S: Large-scale preparation of ribulosebisphosphate carboxylase from a recombinant system in Escherichia coli characterized by extreme plasmid instability. Appl Environ Microbiol. 1985, 49: 1094-1100.Google Scholar
- Fiedler M, Skerra A: proBA complementation of an auxotrophic E. coli strain improves plasmid stability and expression yield during fermenter production of a recombinant antibody fragment. Gene. 2001, 274: 111-118. 10.1016/S0378-1119(01)00629-1View ArticleGoogle Scholar
- Pilarek M, Glazyrina J, Naubauer P: Enhanced growth and recombinant protein production of Escherichia coli by a perfluorinated oxygen carrier in miniaturized fed-batch cultures. Microb Cell Fact. 2011, 10: 50- 10.1186/1475-2859-10-50View ArticleGoogle Scholar
- Šiurkus J, Neubauer P: Heterologous production of active ribonuclease inhibitor in Escherichia coli by redox state control and chaperonin coexpression. Microb Cell Fact. 2011, 10: 65- 10.1186/1475-2859-10-65View ArticleGoogle Scholar
- Šiurkus J, Neubauer P: Reducing conditions are the key for efficient production of active ribonuclease inhibitor in Escherichia coli. Microb Cell Fact. 2011, 10: 31- 10.1186/1475-2859-10-31View ArticleGoogle Scholar
- Šiurkus J, Panula-Perälä J, Horn U, Kraft M, Rimseliene R, Neubauer P: Novel approach of high cell density recombinant bioprocess development: Optimisation and scale-up from microlitre to pilot scales while maintaining the fed-batch cultivation mode of E. coli cultures. Microb Cell Fact. 2010, 9: 35- 10.1186/1475-2859-9-35View ArticleGoogle Scholar
- Rettich TR, Battino R, Wilhelm E: Solubility of gases in liquids. 22. High-precision determination of Henry's law constants of oxygen in liquid water from T = 274 K to T = 328 K. J Chem Thermodyn. 2000, 32: 1145-1156. 10.1006/jcht.1999.0581. 10.1006/jcht.1999.0581View ArticleGoogle Scholar
- Studier F: Protein production by auto-induction in high-density shaking cultures. Protein Expr Purif. 2005, 41: 207-234. 10.1016/j.pep.2005.01.016View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.