Scale-up of microbial fermentation using recombinant ...

Jason Brown, Paula Decaria, Nephi Jones, and Tommy Smith

Ap plica t ion note CO291 8 0

Scale-up of microbial fermentation using recombinant E. coli to produce an 80 kDa protein in Thermo Scientific HyPerforma 30 L and 300 L Single-Use Fermentors

Introduction Fermentation production facilities can utilize the Thermo Scientific™ HyPerforma™ Single-Use Fermentor (S.U.F.) instead of traditional stainless steel cleaning-in-place/ sterilization-in-place (CIP/SIP) fermentor vessels without modifying their existing procedures. As expression systems are moved to production volumes, they can easily be scaled from a benchtop fermentor (or flask) to the S.U.F. Typically, with other single-use bioreactors used as fermentors, culture procedures have to be modified substantially and risk negatively affecting the organism or plasmid because of the performance constraints of those bioreactors. The HyPerforma S.U.F. (Figure 1) was designed to meet the unique requirements of microbial fermentation instead of being modified from a cell culture bioreactor. The S.U.F. utilizes three Rushton-type impellers along with vessel geometric proportions, spacing, baffle configurations, and agitation rates common in rigidwalled industry CIP/SIP fermentors and research fermentors (Figure 2). The motor and direct drive system of the vertically centered drive agitator has been designed and tested to help ensure reliability. The proprietary single-use sterile rotary seal bearing port technology has been proven to be the most reliable and widely used in the biotechnology industry and requires zero maintenance, unlike traditional mechanical seals. The agitator can sustain an agitation rate of 375–600 rpm and delivers 11 hp/1,000 gal (2.27 W/L) of mixing power, offering capacity beyond systems that use a magnetically coupled impeller. In addition, the modular design of the HyPerforma S.U.F. impeller drive train and tank baffles can be customized to meet specific culture or facility needs. Available in 30 L and 300 L sizes, the HyPerforma S.U.F. offers superior mass transfer performance and meets the process demands of dense and rapidly growing microbial cultures.

Figure 1. Thermo Scientific HyPerforma 30 L and 300 L SingleUse Fermentors with condensers.

Figure 2. 30 L and 300 L S.U.F. bioprocess containers (BPCs) with condenser and foam sensor.

2

A special high-flow version of the drilled hole sparge is provided as an integral component of the single-use fermentor. To better handle the high sparge rates of microbial systems, the S.U.F. condenser system was designed for airflow well beyond 2 vvm. It serves to both protect the filters from fouling and significantly reduce liquid evaporation loss during the process. All polymer components of the condenser system are fully integrated as single-use contact materials. Extensive applicationbased stress tests have demonstrated that S.U.F. high-flow filter options can sustain moist E. coli culture air at 2 vvm for continuous durations exceeding two weeks.

Materials and methods For E. coli testing, K12 medium was used with an exponential feed solution containing 75% glucose and 2.5% MgSO4. K12 medium consisted of yeast extract (5 g/L), dextrose (25 g/L), KH2PO4 (2 g/L), K2HPO4 (3.13 g/L), (NH4)2SO4 (5 g/L), MgSO4 (1.22 g/L), thiamine (0.25 g/L), K12 trace element solution (1 mL/L), and kanamycin sulfate (0.025 g/L). K12 trace element solution was composed of (final concentrations given): FeCl3·6H2O (6.42 mg/L), NaCl (5 mg/L), MnCl2·4H2O (4 mg/L), ZnSO4·7H2O (1 mg/L), CuSO2·5H2O (0.4 mg/L), Na2MoO4·2H2O (0.5 mg/L), H3BO3 (0.5 mg/L).

The pre-sterile, gamma-irradiated S.U.F. eliminates the need for cleaning. Common cleaning procedures can take 1–2 days depending on stainless steel CIP treatments and sterilization procedures. The S.U.F. can be set up, media can be hydrated and sterile filtered, and probes can be calibrated in about 2 hours. The time, labor, and cleanup savings are substantial, and the S.U.F. also offers the benefit of eliminating cross-contamination risk. As proven with other single-use technologies, the S.U.F. offers the benefits of reduced initial capital investment, greatly simplified facility design, and much faster installation and process qualification.

For the traditional stainless steel fermentor, 50 L of K12 medium was hydrated in the vessel and sterilized at 121˚C for 20 min. Dextrose and MgSO4 were sterilized in an autoclave and added to the medium after vessel sterilization. Trace elements were sterile filtered and added aseptically to sterile medium.

The S.U.F. system also provides single-use options preconfigured into the single-use container that facilitate process control monitoring of head-space gas pressure, foam generation, pH, dissolved oxygen (DO), and temperature. The S.U.F. vessel is designed to be retrofitted and integrated into existing fermentation controllers. End users should take note of options for measuring process liquid volume via load cells. In addition, pH, DO, or any other conventional 12 mm internal diameter (ID) SIP sensors can be safely integrated via the proven Thermo Scientific™ Single-Use Bioreactor (S.U.B.) aseptic-type probe connector kits. The S.U.F. provides simple scale-up and robust transfer of microbial expression production procedures; even unique processes used in stainless steel CIP/SIP fermentors can be readily adapted to single-use technologies. In this application note, we compare the HyPerforma 30 L and 300 L S.U.F. to a traditional stainless steel fermentor for the production of an 80 kDa protein in E. coli.

The K12 medium for the S.U.F. was hydrated in 25 L or 250 L batches with a Thermo Scientific™ HyPerforma™ Single-Use Mixer DS 300. Hydrated medium was filtered through a 0.45 µm filter followed by a sterile 2" 0.2 µm filter for the 30 L S.U.F. or a sterile 5" 0.2 µm filter for the 300 L S.U.F. A sterile media filter was connected via tube-welded C-Flex™ tubing (aseptic technique using quick connects is an option). E. coli exponential fed-batch cultures Exponential fed-batch cultures of E. coli strain BL21(DE3) expressing an 80 kDa protein were undertaken with 30 L and 300 L working volumes in HyPerforma Single-Use Fermentors (42 L and 420 L total volume) and in a traditional CIP/SIP fermentor with a 100 L working volume (New Brunswick BioFlo™ 610, 125 L total volume). The critical process parameters of the S.U.F. including pH, DO, temperature, liquid mass, foam level, and overlay pressure were controlled using Applikon™ S.U.F. bioprocess controllers configured with two 1 vvm mass flow controllers. The 316 stainless steel 100 L vessel has aspect ratios, impeller type, and impeller spacing similar to the S.U.F. The E. coli seed cultures were prepared the day before protein production culture. High-density seed medium was filter sterilized, and 50 µg/mL of sterile kanamycin stock was added prior to inoculation. High-density seed medium was inoculated at 0.03% with freezer stock 12–16 hours prior to fermentor inoculation. Seed volume was 1% of initial production volume. The cultures were grown at 37°C and 250 rpm overnight. Optical density (OD) at 600 nm was recorded, and seed culture was transferred aseptically through quick connects to 25 L, 250 L, or 50 L cultures in their respective vessels (30 L S.U.F., 300 L S.U.F., or 100 L traditional fermentor).

Results Air flow rates in the traditional 100 L SIP fermentor reached above two times the liquid volume during various points of the exponential fed-batch E. coli culture. The culture OD at 600 nm reached 90 ±10 by 10.5 hours. Gas flow rates for the traditional SIP fermentor were similar to those used in the S.U.F (Figure 3). Two experiments each were performed for the 30 L S.U.F. and the 300 L S.U.F. The first 30 L S.U.F. culture reached an OD600 of 140 with a final DCW of 47 g/L by 10 hours. The second 30 L S.U.F. culture reached an OD600 of 121 with a DCW of 45 g/L by 14 hours. Air flow rates of the 30 L S.U.F. culture reached above 2 vvm with the first culture and 1.7 vvm with the second culture. The first 300 L culture reached an OD600 of 130 with a DCW of 39 g/L in 14 hours. The second 300 L culture reached an OD600 of 140 with a DCW of 36 g/L in 15 hours. From every 9.46 L centrifuged, about 1.35 kg of wet cell pellet was collected. The total wet cell mass was about 43 kg per 300 L culture. Air flow rates of the 300 L culture reached 1.6 vvm with the first culture and 1.4 vvm with the second culture.

120

400 Stirrer (rpm)

100 300

80 60

200

40 100 0

20 0

2

4

6

8

10

Gas Flow (slpm), % O2 supplemented

140

0

Time (hr) rpm

Gas (slpm)

% O2 sup

30 L E.coli Glucose Exp Fed

B 500

80

400

60

300 40

200

20

100 0

0 0

2

4

6

8

Gas Flow (slpm), % O2 supplemented

100

600

10

Time (hr) rpm

Gas (slpm)

% O2 sup

300 L E.coli Glucose Exp Fed

C

500

350 300

400

250 300

200 150

200

100 100

50 0

Gas Flow (slpm)

After the maximum OD was achieved, the temperature was set to 10˚C. Samples of all cultures were taken at various time points. Part or all of culture pellets were collected, and protein was extracted and analyzed for protein production scale-up. Cells were collected for protein extraction via centrifugation at 21,000 rpm (Cepa Z41 high-speed centrifuge, 3" bowl ID). The protein was further processed for use as prototype products. At the completion of experiments, the single-use seed flasks and S.U.F. were sterilized in a small autoclave prior to disposal (Steris Amsco Centry V120, autoclave inside diameter of 22.5" and length of 36", with 19.5" x 19.5" door).

500

Stirrer (rpm)

Glucose was maintained at 20 ±10 g/L until an OD600 of 130 ±10 was achieved. Appropriate PharMed™ tubing for pump and feed rate was built into the feed line set prior to irradiation and was connected to a feed bag of sterile glucose, aseptically, prior to initiating feed program. Exponential feed program or manual set points were ramped up to maintain the optimal 20 ±10 g/L concentration.

3

SIP 100 L E.coli Glucose Exp Fed

A

Stirrer (rpm), % flow O2 supplemented

E. coli cultures were operated to optimize protein production. The pH was controlled at 7 ± 0.1 with 28% ammonium hydroxide and 5 M H2SO4. Dissolved oxygen was set to 40% and controlled via cascade settings, with stirrer rpm increased first, then air, followed by oxygen. The temperature was controlled at 37˚C until the OD600 reached 30 ±2, at which point the temperature was set to 30˚C. When OD600 reached 40, IPTG was added to induce protein expression. Dry cell weight (DCW) samples were taken periodically for analysis.

0

2

4

6

8

10

0

Time (hr) rpm

% O2 sup

Gas (slpm)

Figure 3. Exponential fed-batch E. coli cultures for 80 kDa protein production. Plots depict agitation, gas flow, and percent oxygen supplemented into gas flow. (A) Plot of 100 L traditional SIP fermentor culture. Initial volume was 50 L and final volume of the culture was about 75 L. (B) Plot of 30 L S.U.F. culture; final culture volume was about 30 L. (C) Plot of 300 L S.U.F. culture; final culture volume was about 300 L.

4

The OD600 in S.U.F. experiments averaged 136 and DCW averaged 42 g/L (n = 4). As noted, the growth rates were similar to that of stainless steel SIP fermentors, though the SIP fermentor feed ended at an OD600 of 91 around 11 hours (Figure 6). The protein yields were similar in all runs (Figure 7). E.coli SUF vs. Stainless Steel SIP 140 40.00

120 100 80

20.00

60

OD600

DCW (g/L)

30.00

40 10.00 20 0

0.00 0

2

4

6

8

10

12

14

Time (hr) SUF avg DCW

SUF avg OD

SIP OD

Figure 6. Culture growth rates. Graph depicts OD600 and DCW over time for E. coli cultures (n = 4) during 80 kDa protein expression in K12 medium with 2% glucose maintained through an exponential feed. S.U.F. data were taken from an average of two 30 L cultures and two 300 L cultures.

Lane 1

2

3

4

5

6

7

8

9 10

Figure 7. Western blot of expressed 80 kDa protein using anti6xHis antibody. Lanes 2–6 are before induction, and lanes 7 and 8 are post-induction. Lane 1 is a protein ladder and lane 10 is empty. Similar gel images and protein yield were seen in SIP and S.U.F. cultures.

Discussion On average (n = 4), E. coli cultures at an OD600 of 136 with a DCW of 42 g/L were achieved in the S.U.F by 13 hours of culture. Equivalent expression of the 80 kDa protein was observed when compared to the traditional 100 L SIP vessel. The set points of the cultures were maintained throughout the culture. Pressure was maintained below the BPC operating optimal maximum of 0.5 psi (35 mbar). Agitation range and gas flow rates were sufficient at producing DO levels to maintain exponential growth and achieve the desired density and expression. Temperature control of the vessel met expectations. A small auxiliary heat exchanger was added between the temperature control unit (TCU) and vessel, which allowed a small flow of cold house water to rapidly cool the 300 L vessel prior to induction during highdensity (OD600 of 30) exponential growth, when process was shifted from 37˚C to 30˚C. The single-use condenser system prevented condensate from fouling the exhaust filter in these and previous cultures. Foam was monitored by the single-use foam sensor in these tests, allowing each run to be completed using only a single filter. TCUs for condensers were adequate for these cultures. Conclusions The HyPerforma S.U.F. is revolutionary in being the first truly scalable fermentor capable of delivering productionlevel working volumes up to 300 L. Because each S.U.F. system offers a compact footprint and can be set up in just a few hours, now even industrial-scale output can be achieved using 3 or 4 of the 300 L S.U.F. vessels operated as a parallel batch (yielding 1,000–1,200 L of harvest). This technology allows cultures to be quickly scaled up with less effort, lower contamination risk, and more process flexibility. These experiments confirm that very common technical transfer operations and process optimization methods are robust across both system working volumes (30 L and 300 L) and can also be directly applied when legacy processes are transferred to single-use technologies from conventional fermentation vessels. The critical performance parameters of heat transfer, liquid mixing, two-phase mass transfer, and process monitoring have been met in order to ensure that minimal effort is required to convert an existing process to a single-use production platform. The high performance demonstrated with this S.U.F. technology creates a new benchmark for features that should be important when choosing a system for microbial fermentation applications.

5

Ordering information Product

Cat. No.

Notes

30 L Single-Use Fermentor BPC

SH31035.01 customized with single-use sensors

CX5-14 film, single-use Mettler-Toledo™ pH and DO sensors, foam sensor, 2 x 5" hydrophobic PE high-flow filter, condenser

30 L Single-Use Fermentor Hardware

SUF0030.9002

Optional, also used SV50177.306 (load cell display), SV50988.04 (load cells), SV51006.01 (cable/tube management system), SV50992.10 (bottle holder), two SV51006.03 (feed bag hook)

Single-Use Condenser Hardware (30 L fermentor)

SV51016.02

Includes ThermoFlex™ 900 chiller (air cooled)

ThermoFlex 2,500 (heater/chiller for vessel)

SV50239.27

Air-cooled unit, A/D remote control (liquid-cooled option available: SV50239.05)

300 L Single-Use Fermentor BPC

SH31030.02 customized with single-use sensors

CX5-14 film, single-use Mettler-Toledo pH and DO sensors, foam sensor, 2 x 10" hydrophobic PE exhaust high-flow filters, condenser

300 L Single-Use Fermentor Hardware

SUF0300.9002

Optional, also selected SV50177.306 (load cell display), SV50988.03 (load cells), SV51006.02 (cable/tube management system), SV50992.10 (bottle holder), two SV51006.03 (feed bag hook)

Single-Use Condenser Hardware (300 L fermentor)

SV51016.01

Includes ThermoFlex™ 2500 chiller (air cooled)

ThermoFlex 10,000 (heater/chiller for vessel)

SV50239.29

Air-cooled unit, A/D remote control (liquid-cooled option available: SV50239.07)

Thermo Scientific™ Masterflex™ LS Peristaltic Pump

SV50241.02

Two per 300 L S.U.F.; one used as integrated nutrient feed pump, one used as condensate return

Exhaust Filter Heater 120 VAC

SV50191.41

10" style with integrated thermostat controller

Exhaust Filter Pinch Clamp for 30 L system

SV50177E.16

Preserves second filter as backup filter

Exhaust Filter Pinch Clamp for 300 L system

SV50177E.15

Preserves second filter as backup filter

Exhaust Filter Pinch Clamp for 30 L System

SV50177E.16

Preserves second filter as backup filter

Exhaust Filter Pinch Clamp for 300 L System

SV50177E.15

Preserves second filter as backup filter

S.U.B./S.U.F. Probe Kits

SH30720.01 SV50177.01

Disposables and tray for autoclaving of 12 x 200 mm SIP probes for sterile connection

SUMDS0300.9003

S.U.M. with adjustable head and 5–300 L volume capabilities

SV50177.77

Reusable HUB for S.U.M.

SV50177.49

300 L shaft

SV30959.04

300 L drum

SV50959.03

50 L shaft

SV50517.38

50 L drum

SH30647.06

300 L open-top liner

SH30749.08

300 L impeller and drive shaft sleeve

SH30399.01

50 L open-top liner

SH30749.11

50 L impeller and drive shaft sleeve

Single-Use Mixer (S.U.M.) DS 300 300 L S.U.M. Shaft and Drum 50 L S.U.M. Shaft and Drum Liner for 300 L S.U.M. Liner for 50 L S.U.M.

Ap plica t ion note CO291 8 0

Thermo Scientific™ single-use technologies and Life Technologies™ bioprocessing portfolios are now working together to provide you with innovative solutions and exceptional customer service from drug discovery through large-scale commercial production. Under Thermo Fisher Scientific, this unity represents decades of expertise and support, offering you a broad range of products for single-use technologies, cell culture, purification, and analytics. We are dedicated to providing you with the quality and innovation you need and deserve. Let’s move forward together.

Thermo Fisher Scientific, San Jose, CA USA is ISO Certified.

Find out more at lifetechnologies.com/sut © 2014 Thermo Fisher Scientific Inc. All rights reserved. Life Technologies and Thermo Scientific are brands of Thermo Fisher Scientific. All trademarks are the property of Thermo Fisher Scientific a nd its subsidiaries unless otherwise specified. Applikon is a trademark of Applikon, B.V. BioFlo is a trademark of Eppendorf, Inc. Mettler-Toledo is a trademark of Mettler-Toledo AG. PharMed and C-Flex are trademarks of Saint-Gobain Performance Plastics Corporation.

Africa-Other +27 11 570 1840 Australia +61 2 8844 9500 Austria +43 1 333 50 34 0 Belgium +32 53 73 42 41 Canada +1 800 530 8447 China +86 10 8419 3588 Denmark +45 70 23 62 60 Europe-Other +43 1 333 50 34 0 CO29180 1014

Finland/Norway/Sweden +46 8 556 468 00 France +33 1 60 92 48 00 Germany +49 6103 408 1014 India +91 22 6742 9434 Italy +39 02 950 591 Japan +81 45 453 9100 Latin America +1 608 276 5659

Middle East +43 1 333 50 34 0 Netherlands +31 76 579 55 55 South Africa +27 11 570 1840 Spain +34 914 845 965 Switzerland +41 61 716 77 00 UK +44 1442 233555 USA +1 800 532 4752