Добавил:
Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз: Предмет: Файл:

Tools and Applications of Biochemical Engineering Science

.pdf
Скачиваний:
29
Добавлен:
15.08.2013
Размер:
3.28 Mб
Скачать

Cell Retention Devices for Suspended-Cell Perfusion Cultures

169

133.Huang Y, Wang X-B, Becker FF, Gascoyne PRC (1997) Biophys J 73:1118

134.Doscolis A, Kalogerakis N, Behie LA, Kaler KVIS (1997) Biotechnol Bioeng 54:239

135.Doscolis A, Kalogerakis N, Behie LA (1999) Cytotechnology 30:133

136.Kalogerakis N, Doscolis A, Behie LA (1998) In Merten OW, Perrin P, Griffiths B (eds) New developments and new applications in animal cell technology, Kluwer Academic Pub, Dordrecht, p 369

137.Svarovsky L (1985) Solid-liquid separation processes and technology. Elsevier Science Pub, Amsterdam

138.Rebsamen E, Goldinger W, Scheirer W, Merton OW, Palfi GE (1987) In: Spier RE, Griffiths JB (eds) Modern approaches to animal cell technology. Butterworth, London, p 548

139.Rebsamen E, Goldinger W, Scheirer W, Merton OW, Palfi GE (1987) Develop Biol Standard 66:273

140.Hawrylik SJ, Wasilko DJ, Pillar JS, Cheng JB, Lee SE (1994) Cytotechnology 15:253

141.Mercille S, Johnson M, Lemieux R, Massie B (1994) Biotechnol Bioeng 43:833

Received: April 2001

Fed-Batch Cultures of Escherichia coli Cells with Oxygen-Dependent nar Promoter Systems

Ho Nam Chang*, Se Jong Han1, Seong-Chun Yim, Mu-ri Han2, Jongwon Lee3

* Department of Chemical Engineering, KAIST, Taejon 305-701, Korea

E-mail: hnchang@mail.kaist.ac.kr

1 Fine Chemical Research Institute, Samchully Pharmaceutical, Shiheung, Kyunggi 429-450, Korea

2 R&D Center of Coreana Cosmetics, Cheonan, Chungnam 330-830, Korea

3Department of Biochemistry, School of Medicine, Catholic University of Daegu, Taegu 705-034, Korea

Dedicated to Prof. Dr. Wolf-Dieter Deckwer on the occasion of his 60th birthday

The recombinant proteins produced from Escherichia coli as a host cell need to be made at as low a cost as possible because of the end of the monopoly following expiry of the patent on early pharmaceutical proteins, and thus expanding applications to non-pharmaceutical large-scale products. We review in this article how the various promoters used in recombinant E. coli could affect its protein products, especially with emphasis on relatively new oxy- gen-dependent nar promoters for b-galactosidase production. Several studies carried out in the authors’ laboratory show that the nar promoter does not require any chemicals except 1% nitrate and oxygen for protein production. And according to recent work with the modified strains it is possible to produce the enzyme (b-galactosidase) even without the nitrate ions at 45% of its total protein content when its cell density reached OD = 176.

Keywords. Nar promoter, Oxygen-dependent, Inducible promoter, Fed-batch, Recombinant

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

172

2

E. coli Promoters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

172

2.1

Thermal and Chemical Promoters . . . . . . . . . . . . . . . . . . . .

172

2.2

VHb Promoter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

173

2.3

The nar Promoter . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

173

3

Fed-Batch with nar Promoter System Requiring Nitrate . . . . . . . .

174

3.1

The nar Promoter as an Inducible Promoter . . . . . . . . . . . . . .

174

3.2

Fed-Batch Culture of pMW61/RK5265 . . . . . . . . . . . . . . . . . .

175

4

Fed-Batch with nar Promoter System Requiring No Nitrate . . . . . .

177

4.1

Characterization of the nar Promoter System with Various

 

 

Combinations of Plasmids and Mutant Host E. coli Strains . . . . . .

177

4.2

Strain Effect on the Induction of the nar Promoter in Fed-Batch

 

 

Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

178

4.3

Effects of Cell Density, DO Level, and Nitrate on the

 

 

pMW618/W3110narLSystem in Fed-Batch Cultures . . . . . . . . . .

178

5

Comparison with Other Promoter Systems . . . . . . . . . . . . . . .

179

6

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180

 

Advances in Biochemical Engineering/

 

Biotechnology, Vol. 74

 

 

Managing Editor: Th. Scheper

 

 

© Springer-Verlag Berlin Heidelberg 2002

172 H. N. Chang et al.

List of Abbreviations

DO

Dissolved oxygen

OD600

Optical density at 600 nm

1 Introduction

The early stages of the genetic engineering era focused on the production of high-value products such as insulin, interferons, etc. Attention had to be paid to the bioreactor operation and downstream processing to meet regulatory standards. However, the progress of genetic engineering and bioprocessing techniques made it possible to diversify its products from high-value to mediumand low-value ones. Late participants in the recombinant DNA product markets always came up with new production technologies and, furthermore, expiration of patents forced manufacturers to lower the production cost of existing products already on the market. Optimization of recombinant protein production activities was focused on increasing cell densities and protein content in each cell to as high a level as possible.

Several excellent review articles have addressed this issue from molecular biology to high density cell cultures of Escherichia coli cells [1–4]. We will briefly review E. coli promoter system with emphasis on oxygen-dependent VHb and nar promoters and discuss fed-batch cultures of the nar promoter system requiring nitrates or no nitrates. These new systems will be compared with other existing promoter systems.

2

E. coli Promoters

A promoter to be used in E. coli should have certain characteristics to render it suitable for high-level production of recombinant protein. First, it must be strong, resulting in the accumulation of protein making up to 10–30% or more of the total cellular protein. Second, it should exhibit a minimal level of basal transcriptional activity. Third, promoters should be capable of induction in a simple and cost-effective manner.

2.1

Thermal and Chemical Promoters

The most widely used promoters for large-scale protein production are thermal or chemical inducible ones. Temperature-regulated promoters are based on the bacteriophage l promoters PL and PR . These promoters are regulated by cI gene and induced by elevating culture temperature from approximately 30°C to about 40°C [5, 6]. General disadvantages of using temperature induction include induction of the heat-shock system, denaturation of overexpressed protein product, and expenditure of high energy [7, 8]. Induction of the lac, the tac, and the

Fed-Batch Cultures of Escherichia coli Cells with Oxygen-Dependent nar Promoter Systems

173

phage T7 promoters are usually performed by addition of the lactose analog, iso- propylthio-b-D-galactoside (IPTG). This is not very practical in large-scale production processes because IPTG is very expensive and furthermore the use of IPTG for the production of human therapeutic proteins is undesirable because of its toxicity [9]. Induction of the trp promoter is accomplished by starvation of tryptophan, a procedure that is well suited for large-scale cultures of recombinant E. coli. One disadvantage of trp promoter is that the best medium, including tryptophan for the growth, cannot be used for induction. Several other expression vectors can be modulated by simple cultivation parameters such as concentration of phosphate [10,11] or oxygen [12–14]. Phosphate inducible systems are also of interest for large-scale processes, since the induction is easily accomplished by phosphate-limited growth conditions. The induction ratios of most phosphate promoters were reported to be more than 100-fold [15].

2.2

VHb Promoter

Promoters having an activity that can be modulated by varying the dissolved oxygen of the culture medium offer several favorable advantages for the design of vectors to be used in the industrial production of recombinant proteins [16]. Since ensuring adequate aeration in high-density cultures is a problem in any case in fermentors, the use of oxygen-regulated promoters provides an inexpensive means of controlling product synthesis. One of the oxygen-regulated promoters, VHb promoter from an obligate aerobe Vitreoscilla, was tested as an inducible promoter [17–20]. In a two-stage fed-batch fermentation Khosla et al. [19] overproduced b-galactosidase to levels approaching 10% of the soluble cellular proteins by growing recombinant E. coli to OD600 = 10 – 20 at a dissolved oxygen concentration of 20% for growth and then by reducing the oxygen concentration to below 5% air saturation for induction. An overall 30-fold increase in promoter activity and OD600 = 40 – 50 of cell density was achieved [19]. Buddenhagen et al. reported that oxygen-limited fermentation resulted in about a sixfold increase of recombinant a-amylase production in E. coli containing the Vitreoscilla hemoglobin gene [21].

2.3

The nar Promoter

Escherichia coli is a facultative bacterium which grows by a respiratory process under aerobic conditions or by a fermentative process under anaerobic conditions. Under aerobic conditions, oxygen serves primarily as a final electron acceptor in the electron transport chain. Reduction in oxygen supply can affect electron transport because the depletion of oxygen and substitutes of another terminal electron acceptor results in a smaller difference in the redox potential between the dehydrogenated couple and the final acceptor. In the absence of oxygen, the electron from the reduced carrier is transferred to other organic or inorganic compounds, thereby shortening the electron transport chain and resulting in a further reduction in ATP production [22]. When E. coli is grown

174

H. N. Chang et al.

under anaerobic conditions, several genes not normally expressed under aerobic conditions are expressed [23]. The nar operon encoding nitrate reductase is one of them. Nitrate reductase plays an important role in the anaerobic energy metabolism of E. coli [24]. Nitrate reductase can utilize nitrate (NO3) as an electron acceptor to convert it to nitrite (NO2) in the absence of oxygen, changing the fermentative process to a respiratory process [25]. Therefore, the nar operon not normally expressed under aerobic conditions is maximally induced under anaerobic conditions in the presence of nitrate to use nitrate as an electron acceptor.

The nar operon (narGHJI) has been cloned and sequenced and the genetic basis of its regulation has been extensively characterized [26–28]. Regulation of expression of the nar operon is at the level of transcription and is dependent on three well-characterized transacting factors, FNR, NARL, and IHF (integration host factor) [29]. FNR is required for the transcription of the nar promoter under anaerobic conditions, while NARL and IHF mediate a dramatic stimulation of anaerobic transcription in the presence of nitrate. When growth conditions change from aerobic conditions to anaerobic conditions, the nar operon is partially induced by binding of the regulatory protein, FNR, to the nar promoter. Then it is maximally induced in the presence of nitrate by binding of another regulatory protein, NARL, to the nar promoter together with FNR [30]. IHF is a small, heterodimeric DNA-binding protein that is required for a number of different functions in E. coli, including DNA replication, chromosomal integration of l-phage DNA, and regulation of expression of diverse genes and operons [31]. IHF is required along with NARL for the induction of transcription from the nar promoter by nitrate [29]. Zhang and DeMoss demonstrated that binding of IHF induces a sharp bend in the narG promoter, centered in the IHF binding site [32]. These results suggest that NARL binds at the upstream site and the FNR-RNA polymerase complex at the transcription start site [33]. Initiation of transcription of the nar operon increases when these three transacting proteins bind to the specific sites on the nar promoter, resulting in the expression of the structural genes, narG, narH, and narI [27, 34]. The nar promoter was cloned, and the lacZ gene was downstream to the nar promoter instead of the structural genes to make the study of the mechanism of the induction of the nar promoter easy [30].

3

Fed-Batch with nar Promoter System Requiring Nitrate

3.1

The nar Promoter as an Inducible Promoter

The characterization of the nar promoter in E. coli with the intact nar operon on the chromosome was carried out. Expression of b-galactosidase was maxi-

mal when the nar promoter was induced at OD600 = 1.7 under anaerobic conditions in the presence of 1% nitrate. At this optimal condition the induction

ratio was approximately 250 and the specific b-galactosidase activity was approximately 7500 Miller units at OD600 = 2.7 [35]. In this study, we used plasmid

Fed-Batch Cultures of Escherichia coli Cells with Oxygen-Dependent nar Promoter Systems

175

pXR8971 and E. coli strain Mv1190 that has the intact nar operon on the chromosome, which enables it to produce nitrate reductase. Lowering the cellular redox potential by nitrate reductase produced from chromosome can suppress further induction of the nar operon, resulting in the autoregulation of expression of nitrate reductase [36, 37].

Lee reported on the characterization of pMW616/ES2001 system [38]. The used strain, E. coli ES2001, was made by insertion of Tn10 on fnr gene affecting expression of the nar promoter according to the dissolved oxygen level [28]. The plasmid, pMW616, has the modified nar promoter, by which several site-di- rected mutagenesis was performed [37]. The maximum specific b-galactosidase activity was obtained when E. coli was grown under aerobic conditions, and then the modified nar promoter was induced at OD600 = 2.2 under microaerobic conditions (DO = 1 – 2%). Under these conditions, the maximum specific b-galactosidase activity was 13,000 Miller units. However, the induction ratio was very low, approximately 2.

Another nar promoter system, pMW61/RK5265, was characterized by Lee et al. using flask and batch cultures [39]. The E. coli RK5265 has the mutant nar operon on the chromosome, the narG gene encoding one subunit of nitrate reductase was mutated, and thus the autoregulation of expression of nitrate reductase was eliminated [40]. The plasmid pMW61 having the nar promoter was constructed by introduction of KpnI site into pSL8RB2, a derivative of pSL800 [37]. In batch experiments, it was found that induction of pMW61/RK5265 system was suppressed during the growth stage when E. coli was grown at high dissolved oxygen (DO) level (80%) in the presence of 1% nitrate. Then the nar promoter was fully induced when the culture was shifted to microaerobic con-

ditions at OD600 = 1.7 by lowering the DO level to 1–2%. The maximum specific b-galactosidase activity expressed from the reporter lacZ gene was 36,000

Miller units, equivalent to approximately 35% of the total cellular proteins at OD600 = 3.2, and the induction ratio was approximately 300.

3.2

Fed-Batch Culture of pMW61/RK5265

As mentioned in the previous section, so far three different nar promoter systems (pXR8971/Mv1190, pMW616/ES2001, and pMW61/RK5265) were characterized using batch cultures. Of these systems, pMW61/RK5265 provided the most promising properties as a new inducible promoter system (Table 1). Therefore we carried out fed-batch culture of the pMW61/RK5265 system to investigate the feasibility of utilizing the nar promoter system for the industrial-

Table 1. Comparison of different nar promoter systems

System

Induction OD600

Max units

Induction ratio

Reference

pXR8971/Mv1190

1.7

7,500

250

35

pMW616/ES2001

2.2

13,000

2

38

pMW61/RK5265

1.7

36,000

300

39

 

 

 

 

 

176

H. N. Chang et al.

scale fermentation to produce recombinant proteins. Fed-batch cultures were performed in a 2.5-l jar fermentor at 37 °C using modified R medium. Dissolved oxygen level was adjusted by changing the stirring speed and the air supply [41].

As the growth state of cells may affect induction of the nar promoter, the effect of cell density on the expression of b-galactosidase was investigated. First, cells were grown to different cell densities (OD600 = 17, 35, 78, respectively) at high DO level (80%) before the nar promoter was induced by lowering the DO level to 1–2%. In the first attempt of induction, the nar promoter was induced after cells were grown at high DO level (80%) simply by lowering the DO levels to 1–2%, and the DO levels were kept at this low level continuously throughout the whole induction period (Table 2, row A). Under these induction conditions, however, the acetic acid concentration increased continuously to 37 g/l, and the cells stopped growing at an OD600 lower than 100 [41]. In an attempt to avoid accumulation of acetic acid, alternating aerobic and microaerobic conditions were applied until the cell density became high enough (OD600 > 80) so that acetic acid concentration did not have any significant effect on the growth of the cells by the end of the operation. When this operating strategy was adopted, the final acetic acid concentration became only 7 g/l, and the cell density reached

OD600 = 160 [41]. When the nar promoter was induced too early (Table 2, row C) or too late (Table 2, row D), the maximal specific b-galactosidase activities were

less than 30,000 Miller units. However, when the nar promoter was induced by lowering the DO level at OD600 = 35 after 15 h of cultivation, the maximum specific b-galactosidase activity reached 40,000 Miller units (Table 2, row B). As the specific b-galactosidase activity just before induction was 1000 Miller units, the induction ratio was 40 [41].

The maximum specific b-galactosidase activities were also strongly dependent on the DO levels before and after induction. Maintenance of high DO levels before induction and low DO levels during the induction period was important to maximize the expression of b-galactosidase (Table 2, rows C, E, F, and G).

In the absence of nitrate, the maximum specific b-galactosidase activity became only 4000 Miller units (Table 2, row H). Therefore, addition of nitrate into the growth medium before induction is necessary to maximize the expression of b-galactosidase.

Table 2. Results of fed-batch cultures of pMW61/RK5265 system

No.

Induction

DO before

DO after

NO3

Max. units

 

OD600

induction

induction

 

 

A

35

80%

1–2%

1%

23,000

B

17

80%

Microaerobic/aerobic

1%

25,000

C

35

80%

Microaerobic/aerobic

1%

40,000

D

78

80%

Microaerobic/aerobic

1%

26,000

E

35

50%

Microaerobic/aerobic

1%

26,000

F

35

20%

Microaerobic/aerobic

1%

25,000

G

35

80%

10%

1%

11,000

H

35

80%

Microaerobic/aerobic

0%

4,000

 

 

 

 

 

 

Fed-Batch Cultures of Escherichia coli Cells with Oxygen-Dependent nar Promoter Systems

177

4

Fed-Batch with nar Promoter System Requiring No Nitrate

Although wild type nar promoter system, pMW61/RK5265, has promising properties to be used as an inducible promoter, it has one disadvantage: addition of nitrate as an inducer in addition to low DO level is necessary for the maximal induction, which not only increases induction cost but also causes water contamination after discharge of waste culture medium.We characterized several different combinations of the nar promoter (pMW61, pMW618), mutations on the chromosome of host E. coli strains (narL, narG), and host E. coli strains (MC4100, JM107, W3110) to find the best nar promoter system which can be used as an oxygen-dependent inducible system even without addition of nitrate. Plasmid pMW618, a derivative of pMW61, was made by site-directed mutagenesis on the –10 and –35 regions of the wild type nar promoter to eliminate dependence of the nar promoter on nitrate [37].

4.1

Characterization of the nar Promoter System with Various Combinations of Plasmids and Mutant Host E. coli Strains

Cells were grown in a jar fermentor in LB medium at high DO level (DO > 80%), and then the nar promoter was induced by keeping the DO level low (DO = 1 – 2%) after cell density reached OD600 = 1.5. Table 3 shows the maximum specific b-galactosidase activities and induction ratios of the E. coli strain RK5265(narG), RK5278(narL), and RK5278narG(narLnarG) having plasmid pMW61 or pMW618. This table shows that very low expression levels were obtained in cases of pMW61/RK5278 and pMW61/RK5278narG. The E. coli strain RK5278 was constructed by mutating the narL gene on the chromosome of MC4100 strain; thus NARL, one of the regulatory proteins of nar operon, could not be synthesized in this strain. Because NARL could not be produced in this strain, the nar promoter was only induced by FNR, another regulatory protein. In case of pMW618/RK5265, the maximum specific b-galactosidase activity was 35,000 Miller units, but the induction ratio was very low because the expression levels before induction were high. In the previous study it has already been shown that pMW61/RK5265(narG) cannot be used as an oxygen-

Table 3. Effect of combination of plasmids and mutant host E. coli on the induction of the nar promoters in batch cultures

System

Description

Max units

Induction ratio

 

 

 

 

pMW61/RK5265

narG

20,000

260

pMW618/RK5265

narG

34,000

23

pMW61/RK5278

narL

3,600

260

pMW618/RK5278

narL

29,000

112

pMW61/RK5278narG

narLnarG

2,600

213

pMW618/RK5278narG

narLnarG

15,000

78

 

 

 

 

178

H. N. Chang et al.

Fig. 1. Fed batch fermentation of E. coli W3110narLharboring plasmid pMW618. Cells were grown in a jar fermentor in modified R medium at high DO level (DO > 80%) for 3.5 h before the feeding began. When cell density reached OD600 = 35, the nar promoter was induced by lowering DO level to 1–2% with alternating microaerobic and aerobic conditions. OD600 open circles, DO level filled diamonds, the specific b-galactosidase activity filled squares, and acetic acid concentration filled triangles were measured. Error bars represent the standard deviation of at least three samples taken from a single run

dependent inducible promoter in the absence of nitrate because the expression level of b-galactosidase was low (Table 2, row H). From this study, it was concluded that pMW618/narLmutant was the best combination for the induction of the nar promoter in the absence of nitrate. This conclusion was supported when the E. coli W3110 was used [43]. Therefore, pMW618 and narLmutant were used as a plasmid and E. coli host, respectively, for further characterization.

4.2

Strain Effect on the Induction of the nar Promoter in Fed-Batch Cultures

The E. coli strains RK5278 (narL), JM107narL, and W3110narLharboring plasmid pMW618 were cultivated at the optimal conditions obtained in the previous study for the pMW61/RK5265 (Table 4, rows A, B, and C). Host E. coli strains have strong effects on the induction of the modified nar promoter. Of the three E. coli strains, the pMW618/W3110narLsystem had the highest expression level of b-galactosidase (58,000 Miller units) and the highest cell density (OD600 of 176) (Fig. 1). Therefore, W3110narLwas chosen as the strain for the further studies.

4.3

Effects of Cell Density, DO Level, and Nitrate on the pMW618/W3110narLSystem in Fed-Batch Cultures

Of three cases of different cell densities (OD600 = 18, 35, 80) at the induction time, the highest value of specific b-galactosidase activity was obtained when

Fed-Batch Cultures of Escherichia coli Cells with Oxygen-Dependent nar Promoter Systems

179

Table 4. Results of fed-batch cultures of pMW618/narLmutant system

 

 

 

 

 

 

 

 

 

No.

Strain

Induction

DO before

DO after

NO3

Max.

 

 

OD600

induction

induction

 

units

A

RK5278

35

80%

Microaerobic/aerobic

0%

8,000

B

JM107narL

35

80%

Microaerobic/aerobic

0%

52,000

C

W3110narL

35

80%

Microaerobic/aerobic

0%

58,000

D

W3110narL

18

80%

Microaerobic/aerobic

0%

54,000

E

W3110narL

80

80%

Microaerobic/aerobic

0%

51,000

F

W3110narL

35

50%

Microaerobic/aerobic

0%

47,000

G

W3110narL

35

80%

10%

0%

30,000

H

W3110narL

35

80%

Microaerobic/aerobic

1%

28,000

 

 

 

 

 

 

 

the nar promoter was induced at OD600 = 35 (Table 4, rows C, D, and E). It was also necessary to grow cells at as high a DO level as possible for maximal

expression and to keep DO level low during induction period for the nar promoter to be fully induced (Table 4, rows C, F, and G). In case of pMW618/W3110narL–, the addition of nitrate resulted in negative effect on the expression of b-galactosidase (Table 4, row H).

5

Comparison with Other Promoter Systems

Rinas and her associates at GBF, Germany compared the production of human basic fibroblast growth factor (bFGF) using E. coli TG1 [44]. The cells with thermal induction promoter resulted in a final protein titer of 4.88 g/l with a cell mass of 61 g/l and 80 mg bFGF/g cell, and those with IPTG yielded 1.09 g protein/l with a cell mass of 135 g/l and 8.1 mg bFGF/g cell. Insulin fusion protein was produced in a final concentration of 4.5 g/l using E. coli with thermal promoter [45]. This corresponds to concentrations of insulin B-chain, 800 mg/l and A-chain, 600 mg/l, respectively. Interferon a (IFN-a) was produced in a concentration of around 4 g/l with a total biomass of 40 g/l containing 93 mg/g cell [46]. For b-galactosidase using tac promoter, Labes et al. [47] reported 11,000 Miller units/g cell with the induction at an OD of 0.2–0.6. The maximum specific activity of the enzyme in the present study was 58,000 Miller units/g cell at an OD of 176, which amounts to approximately 45% of total cellular protein. Lee JW and associates characterized the nar promoter system from metabolic engineering viewpoint [48].

The hmp gene isolated from Bacillus subtilis can be induced to produce b-ga- lactosidase under anaerobic conditions. The cells are grown in the presence of nitrate to OD600 = 0.2 at high DO level (80%), and then induced by lowering the DO level to 1–2%. The maximum specific b-galactosidase activity increased to the maximal value of 1750 Miller units at OD600 = 2.5 [49].

Соседние файлы в предмете Химия