Available online at www.sciencedirect.

com

Enzyme and Microbial Technology 42 (2008) 284289

A novel strategy of enhanced glutathione production in high cell density cultivation of Candida utilisCysteine addition combined with dissolved oxygen controlling
Guo-Bin Liang a,c , Guo-Cheng Du a,b, , Jian Chen a,b,
School of Biotechnology, Jiangnan University, Wuxi 214122, China Key Laboratory of Industrial Biotechnology, Ministry of Education, Jiangnan University, Wuxi 214122, China c School of Chemistry and Chemical Engineering, Jiangsu Teachers University of Technology, Changzhou 213001, China
b a

Received 19 September 2007; received in revised form 9 October 2007; accepted 9 October 2007

Abstract Effects of dissolved oxygen (DO) concentration on glutathione (GSH) production and cysteine oxidation were investigated in high cell density cultivation of Candida utilis. Lower DO concentration favors cysteine absorption but retards GSH production. Higher DO promotes GSH production but accelerates cysteine oxidation in the broth. A two-step DO control strategy was developed and compared for the potential in enhancing GSH production and cysteine absorption. By using the two-step DO control strategy, a 40% decrease in cysteine addition and a 13% increase in GSH production are observed as compared with that at constant DO of 40%. 2007 Published by Elsevier Inc.
Keywords: Candida utilis; Glutathione (GSH); High cell density cultivation; Cysteine addition; Dissolved oxygen (DO)

1. Introduction Glutathione (GSH), a low molecular-mass thiol, functions in many cellular processes including protection of cells against oxidation and is of increasing interest in medical treatment, foods and cosmetics industry [1]. GSH production by yeast fermentation is efcient and practical. As an intracellular product in yeast, the ultimate goal of the fermentative production of GSH is to achieve a high total GSH yield by increasing cell density and intracellular GSH content. Enhancement of cell concentration can be achieved by process optimization. For instance, Sakato and Tanaka [2] developed a feeding control system to achieve a high cell density for maximizing GSH yield. Meanwhile, addition of precursor amino acids required for GSH is another easy approach to increase spe-

Corresponding author at: Key Laboratory of Industrial Biotechnology, Ministry of Education, Jiangnan University, Wuxi 214122, China. Tel.: +86 510 85918309; fax: +86 510 85918309. Corresponding author at: School of Biotechnology, Jiangnan University, Wuxi 214122, China. Tel.: +86 510 85913661; fax: +86 510 85910799. E-mail addresses: gcdu@jiangnan.edu.cn (G.-C. Du), jchen@jiangnan.edu.cn (J. Chen). 0141-0229/$ see front matter 2007 Published by Elsevier Inc. doi:10.1016/j.enzmictec.2007.10.008

cic GSH production rate (PG ) by enhancing intracellular GSH content. Alfafala et al. [3] investigated the effects of cysteine and related compounds on GSH production. Their results showed that cysteine is a key amino acid for increasing the specic GSH production rate. In Saccharomyces cerevisiae cysteine for GSH synthesis can be supplied by cells to synthesize and to take up from the medium. Recent advancement in biochemical studies has enabled researchers to use various approaches to obtain cysteine in fermentation, but high-level microbial production of cysteine has not yet been achieved [4]. For fermentative production of GSH, cysteine synthesized by cells is not enough as a precursor for synthesis of GSH. Therefore, a suitable cysteine addition strategy is needed to enhance GSH production. It was demonstrated that single-shot addition of cysteine is better than continuous or several additions [5]. And the increase in specic GSH production rate (PG ) by single shot method can be achieved without growth inhibition if cysteine dose is maintained at 0.7 mmol/g cell or less. However, cysteine in the medium was oxidized rapidly after its addition. How to enhance the transporting degree of cys-

G.-B. Liang et al. / Enzyme and Microbial Technology 42 (2008) 284289

285

teine from the medium into cells in reduced form is central to promoting cysteine utilization rate for GSH production. S. cerevisiae or Candida utilis employed in fermentative production of GSH are aerobic. Dissolved oxygen (DO) concentration, a very important factor, affects cell growth and GSH production markedly. Autoxidation of cysteine to cystine by dissolved oxygen in phosphate buffer with pH 7.4 was reported and the result indicated that cysteine is completely oxidized after incubation at 37 C for 2 h [6]. However, in the literature there are no details on oxidation degree of cysteine addition in S. cerevisiae or C. utilis cultivation for GSH production. In this paper, a detailed research about effects of DO concentrations on cysteine oxidation and GSH biosynthesis were carried out. The results showed that lower DO benets cysteine uptake but retards GSH synthesis and vice versa. Two-step DO control strategy is applied to enhance cysteine uptake degree for GSH production and can efciently enhance GSH yield in high cell density cultivation of C. utilis.
2. Materials and methods 2.1. Microorganism and culture media
A high-GSH yeast strain, C. utilis WSH 02-08, was used in this study. The seed medium contained (g/l): glucose 20, peptone 20, and yeast extract 10 at pH 6.0. The seed culture was prepared in a ask on a reciprocal shaker at 200 rpm and 30 C for 20 h. The medium for batch fermentation contained (g/l): glucose 15, ammonium sulfate 8, KH2 PO4 3 and MgSO4 0.25.

3. Results and discussion 3.1. High cell density cultivation of C. utilis The whole process of GSH fermentation is composed of three phases. It began with a short batch phase. After the carbon source was exhausted, a fed-batch culture with glucose feeding was followed to obtain high cell density. Then, cysteine, a precursor of GSH, was added into the broth in the transformation phase to enhance intracellular GSH content. Riesenberg [9] reported that E. coli growth was inhibited when initial glucose concentration exceeded 50 g/l. Our previous study showed that the suitable initial glucose for batch culture of C. utilis was 15 g/l and after 8 h cultivation the glucose was consumed. It is, therefore, important to develop a fed-batch culture to achieve high cell density. In recent years, direct substrate feedback control and indirect feedback control schemes [1012] have been applied in high cell density cultivation of S. cerevisiae. However, the mentioned feeding strategies are sophisticated and difcult to scale up because accurate and expensive instruments are needed. Li et al. [13] compared the effects of different glucose feeding methods on GSH fermentation by E. coli and found that exponential feeding can greatly improve cell concentration, cell productivity and total GSH concentration. In our study, exponential feeding strategy was applied after 8 h batch fermentation. Derived from a mass balance with an assumption of constant cell yield on substrate, the following equation of feeding rate was derived from and has been applied to the production of bakers yeast [14]. F= (VX)0 exp(t ) YX/S (SF S ) (1)

2.2. Batch and fed-batch cultivation

Initial glucose was 15 g/l for batch culture in a 7 l fermentor and 500 g/l for the feed medium, including (per litre) 1 g MgSO4 , 30 g (NH4 )2 SO4 and 6 g KH2 PO4 . Ten percent (v/v) of seed culture was inoculated into a 7 l fermentor with a working volume of 5 l. A total amount of 450 g glucose was fed with an exponential feeding after 8 h of batch fermentation. The exponential feeding was stopped at 20 h cultivation, followed by feeding at a constant rate of 5.5 g/(l h) to 45 h.

2.3. Control of DO and pH

During the cell growth phase, the aeration was maintained at 1 vvm and the agitation was operated at 400 rpm. In GSH synthesis after cysteine addition, DO was controlled at 5% in former 3 h, and 20% in later 12 h by maintaining aeration at 1 vvm and adjusting agitation in the range of 50200 rpm. The pH was controlled automatically at 5.5 by adding 3 mol/l H2 SO4 or 3 mol/l NaOH solutions. Under the same condition, each experiment was carried out at least three times.

where F is the feeding rate (ml/h); is the specic growth rate (h1 ); V0 , X0 and S are the initial volume of medium (ml), concentration of biomass (g/l) and residual glucose concentration (g/l) at feeding time 0 h, respectively; YX/S is the yield coefcient of biomass on glucose (g/g); and SF is the glucose concentration in the feeding solution (g/l). Based on batch fermentation analysis, the values of V0 (3.5), YX/S (0.7), and X0 (10.5) were measured experimentally, SF was set at 500 g/l, if the value of specic growth rate was set, the feeding rate can be determined. By setting (h1 ) at 0.15, 0.2 and 0.25, a set of ow rate values can be calculated from Eq. (1). Since continuously changing
Table 1 Effects of different specic growth rates () on cell growth and GSH production Parameters Results 0.15 55.4 509.7 0.92 0.132 0.56 6.45 0.2 81.2 763.2 0.94 0.198 0.63 6.74 0.25 64.7 580.3 0.90 0.143 0.17 1.84

2.4. Analytical methods

A culture broth of 25 ml was centrifuged at 3500 g for 15 min and the cells were washed twice with ice-cold saline (0.85% NaCl, w/v). The wet cells were extracted with 40% (v/v) ethanol at 30 C for 2 h, and centrifuged at 5000 g for 20 min, and the supernatant was used for GSH assay and intracellular total thiols determination. Glutathione concentration was determined according to the method described by Tietze [7]. Total thiols concentration was determined based on the method described by Ellman [8]. Dry cell weight (DCW) was determined after drying the cells at 105 C to a constant weight.

Nominal specic growth rate (h1 ) Maximum Dry cell weight (g/l) Maximum GSH production (mg/l) Maximum Intracellular GSH content (%) Actual average specic growth rate (h1 ) Biomass yield on glucose (g/g) GSH yield on glucose (mg/g)

286

G.-B. Liang et al. / Enzyme and Microbial Technology 42 (2008) 284289

Table 2 Effects of different glucose feeding at constant rates on cell growth and GSH production Parameters Mode of cultivation Fed-batch I Glucose feeding rate (g/(l h)) Culture time (h) Maximum DCW (g/l) Maximum GSH (mg/l) GSH content (%) Glucose consumption rate (g/(l h)) Average specic growth rate (h1 ) Cell yield on glucose (g/g) GSH yield on glucose (mg/g) Cell productivity (g/(l h)) GSH productivity (mg/(l h)) 4 45 93.3 891 0.95 3.82 0.071 0.13 1.31 0.49 5.2 Fed-batch II 5.5 45 102.1 981 0.96 5.31 0.084 0.15 1.60 0.84 8.8 Fed-batch III 7 45 96.1 908 0.91 6.72 0.077 0.08 0.85 0.60 5.9

The glucose feeding rates of Fed-batch I, Fed-batch II and Fed-batch III are 4, 5.5 and 7 g/(l h), respectively.

the ow rate was difcult to achieve, ow rate was changed each hour to approximate exponential feeding. The effects of on the cell density and GSH production were shown in Table 1. The experimental results showed that among three setting values of specic growth rate, with at 0.25, the nal cell density was not the highest. The reason is that both glucose and by-products (data did not show) are accumulated in the broth, which will turn the metabolic pathway from aerobic cell growth to anaerobic fermentation. To limit the accumulation of glucose and by-products in the broth, the value was set at 0.2, and the feeding of glucose was stopped when the glucose concentration in broth reached 1.6 g/l. It is obvious that both dry cell weight and GSH yield are increased greatly. However, further decrease in the set value of to 0.15 will cause a lower biomass yield and GSH yield on glucose. After 12 h exponential feeding, both DCW and GSH yield under nominal at 0.2 are higher than that under other two cases. Moreover, under at 0.2, the main kinetic parameters of the biomass yield and the GSH yield on glucose reached the highest value. Therefore, the specic growth rate at 0.2 was adopted in the following work. After exponential feeding phase, to avoid glucose and by products accumulation, a suitable constant feeding rate of glucose should be determined for cell growth and supplying energy required for biosynthesis of GSH. The feeding rate of glucose

was changed from exponential feeding to constant feeding after exponential feeding for 12 h. At this point, the cell concentration is about 81.2 g/l and the GSH yield is 760 mg/l. The comparison of the results obtained between three kinds of constant feeding is shown in Table 2. With an increase in feeding rate of glucose, glucose consumption rate was enhanced accordingly. As a result, the average specic growth rate and GSH production rate are lowered which result in a decrease in biomass yield and GSH yield on glucose. By comprehensively comparing the main kinetic parameters of both biomass and GSH yields on glucose, a suitable feeding rate is 5.5 g/(l h). By applying three-stage operation mode (batch fermentation stage, exponential glucose feeding and constant glucose feeding), the cell concentration reaches 102 g/l and the GSH yield is 981 mg/l after 45 h cultivation (Fig. 1). 3.2. Effects of cysteine addition coupled with high DO on GSH production Fed-batch fermentation is preferable for reaching high cell density, but the intracellular GSH content decreased with an increase in cell density. Addition of precursor amino acids required for GSH is an easy approach to increase GSH content. Cysteine is a key amino acid to increase GSH production, but it retards growth notably in S. cerevisiae. Therefore, a suit-

Fig. 1. Time-course of glucose consumption, cell growth and GSH production: ( ) glucose; ( ) dry cell weights (DCW); () intracellular GSH content; ( ) GSH concentration.

G.-B. Liang et al. / Enzyme and Microbial Technology 42 (2008) 284289

287

Fig. 2. Effects of cysteine addition on GSH production and intracellular total thiols concentration: () 0 mmol/l (control); ( ) 30 mmol/l; ( ) 40 mmol/l; () 50 mmol/l; ( ) 60 mmol/l.

able cysteine addition strategy should be developed to enhance GSH production without causing growth inhibition. In our study, single-shot addition of cysteine was utilized to increase GSH production at 45 h as the glucose feeding stopped. It is observed that DO changes from 10% to about 40% immediately after glucose feeding cessation and cysteine addition. The results shown in Fig. 2 indicate that with increasing concentration of cysteine addition, GSH production was enhanced correspondingly. As cysteine concentration reaches 50 mmol/l, GSH concentration, intracellular GSH content and GSH specic production rate are the highest. After 15 h of cysteine addition, the GSH specic production rate decreased rapidly and its concentration stopped increasing. Therefore, in subsequent experiments, 50 mmol/l of cysteine was added into the broth. After 60 h cultivation, the GSH production reaches 1534 mg/l and its specic production yield is 1.50% (w/w). Intracellular thiol-groups, existing in the form of cysteine residues, are mainly from cysteine, GSH and glutamylcysteine. Their concentrations can indirectly reect the amounts of cysteine in the cells. As shown in Fig. 2, most of cysteine added in the broth was not absorbed into cells at 40% DO, just because most of cysteine was oxidized to cystine (data not shown) in the presence of DO. 3.3. Effects of DO on cysteine absorption and GSH production To explore the inuence of different DO concentration on cysteine oxidation and GSH production, DO at constant concentrations of 5, 10, 20 and 40% by controlling agitation speeds were developed and compared for their potential in improving GSH production and affecting cysteine oxidation. First, 50 mmol/l of

cysteine was added at 45 h with different DO levels to determine the total intracellular reduced thiols concentration and GSH production. As indicated in Fig. 3, the intracellular thiol concentration increases rapidly in 3 h after the single addition of cysteine. It is observed that most cysteine is uptaken in 3 h after its addition. Moreover, the thiols concentration is negatively correlated with the DO level. For example, with DO at 5% of air saturation the intracellular thiols concentration reaches 11.2 mmol/l, which increased by 50% as compared with that under 40% of DO. Meanwhile, GSH production with different DO levels were compared and the results indicated that although lower DO could result in more absorption of cysteine into the cells, the nal GSH yield was not enhanced in contrast to that at higher DO. For example, the GSH yield at DO 5% is a slight lower than that at DO 40%. Moreover, there is no distinct difference for GSH production between DO 20 and 40% of air saturation. It can be reasonably suggested that DO can signicantly affects the GSH production as its concentration lies in the range of 520% of air saturation. In conclusion, it is evident that, at 5% of DO, the cysteine added in the broth can be uptaken effectively and 20% of DO can sufciently satisfy GSH production. It is well known that the ultimate aim of optimizing process in fermentation was to obtain the maximum amount of product with the least cost. Cysteine, a key amino acid to enhance GSH production in fermentation, is easily oxidized in the broth by oxygen. However, oxygen not only affects the cysteine oxidation and its absorption but also inuences the GSH production. How to control DO concentration, on one side, to save cysteine addition, and on the other side, to maximize GSH production, was investigated in the following experiment.

288

G.-B. Liang et al. / Enzyme and Microbial Technology 42 (2008) 284289

Fig. 3. Effects of different DO on GSH production and total intracellular thiols concentration: () 5% DO; ( ) 10% DO; ( ) 20% DO; ( ) 40% DO.

Fig. 4. Effect of cysteine addition on GSH production with two stage DO control: ( ) glucose; ( ) dry cell weights; ( ) intracellular GSH concentration; ( ) intracellular GSH content; ( ) residual thiols; ( ) total thiol. Table 3 Effects of cysteine addition at DO of 5% on total intracellular thiols concentration Cysteine (mmol/l) Total thiols (mmol/l) 20 6.52 25 7.12 30 7.43 35 7.96 40 8.91

duction yield of GSH increases from 1.50 to 1.66% (w/w). More importantly, a 40% of cysteine addition is saved. 4. Conclusions To achieve a high GSH concentration by fermentation, enhancement of cell density and intracellular GSH content have been widely investigated. For example, Patoomporn et al. [15] pointed out that an optimal solution to maximize GSH production can be realized by a two-step fermentation process of cell growth followed by GSH production. During GSH synthesis, addition of the precursor amino acids was an easy approach. Cysteine was conrmed to be a key amino acid for increasing the specic GSH production rate. However, it is indicated in this study that different levels of DO have remarkable effects on cysteine oxidation (data not shown) and GSH production. Higher DO promotes GSH production but accelerates cysteine oxidation in the broth. Lower DO retards GSH production, but reduces the oxidation degree of cysteine signicantly. We develop a two-step oxygen control strategy intending to reduce cysteine oxidation and enhance GSH production. The results are proved to be a better policy for enhancing GSH yield and decreasing cysteine addition in the broth. By adopting the above strategy, same intracellular cysteine residues can be obtained by adding 30 mmol/l cysteine, as compared with that of 50 mmol/l cysteine added at 40%

3.4. GSH production with two-step DO supply strategy Based on above experimental results, a two-step DO control strategy was proposed to enhance cysteine uptake for GSH production. It is demonstrated that, with DO at 40% of air saturation, GSH yields are the highest as cysteine added in the broth reaches 50 mmol/l. At this point, the intracellular thiols concentration is 7.52 mmol/l in 3 h after cysteine addition. To obtain the same intracellular thiols at DO of 5%, different amounts of cysteine were added into the broth to determine the optimal cysteine concentration. Table 3 indicates that the total intracellular thiols concentration reaches 7.43 mmol/l after 3 h with 30 mmol/l cysteine addition. Compared to the cysteine quantity added at 40% of air saturation, a 40% decrease in cysteine addition is achieved but almost with the same intracellular thiols concentration. Therefore, 30 mmol/l of cysteine was added into the broth in the later two-step DO control experiments. By applying the two-step DO control strategy after addition of cysteine, in which DO was controlled at 5% of air saturation in the rst 3 h and then at 20% in the following 12 h, GSH yield reaches 1767 mg/l (Fig. 4), increased by 13%. The specic pro-

G.-B. Liang et al. / Enzyme and Microbial Technology 42 (2008) 284289

289

air saturation. Moreover, the nal GSH yield is increased by 13%. The toxicity of cysteine to various cell types has been previously reported [1618]. In this study, reduction in cysteine addition at least has two functions for GSH production. First, the cost is greatly decreased. The aim of optimizing process in fermentation is to maximize the product but with the least cost. Secondly, decrease in cysteine addition avoids inhibition to the cells that can explain why a 13% increase in GSH production can be observed with less cysteine addition. By applying the two-step oxygen control strategy, on one hand, the quantity of cysteine addition is greatly reduced; on the other hand, less cysteine causes lower inhibition to the cells and in turn leads to higher GSH production. It is suggested that two-step DO control strategy is a feasible method in practical application of GSH fermentation. Acknowledgements This project was nancially supported by Chinese National 863 Project (2006AA10Z313), and the Innovation Fund of Medium or Small Science and Technology Enterprise (06C26213201074). References
[1] Meister A. Selective modication of glutathione metabolism. Science 1983;220:4727. [2] Sakato K, Tanaka H. Advanced control of glutathione cultivation process. Biotechnol Bioeng 1992;40:90412. [3] Alfafala CG, Kanda A, Shioi T, Shimizu H, Shioya S, Suga K. Effect of amino acids on glutathione production by Saccharomyces cerevisiae. Appl Microbiol Biotechnol 1992;36:53840.

[4] Wada M, Takagi H. Metabolic pathways and biotechnological production of l-cysteine. Appl Microbiol Biotechnol 2006;73:4854. [5] Alfafala CG, Miura K, Shimizu H, Shioya S, Suga K. Cysteine addition strategy for maximum glutathione production in fed-batch culture of Saccharomyces cerevisiae. Appl Microbiol Biotechnol 1992;37:1416. [6] Hogg N. The effect of cyst(e)ine on the auto-oxidation of homocysteine. Free Radic Bio Med 1999;27:2833. [7] Tietze F. Enzymatic method for quantitative determination of monogram amounts of total and oxidized glutathione: application to mammalian blood and other tissue. Anal Biochem 1969;29:50222. [8] Ellman GL. Tissue sulfhydryl groups. Arch Biochem Biophys 1959;82:707. [9] Riesenberg D. High cell density fermentation of Escherchia coli. Curr Opin Biotechnol 1991;2:3804. [10] Tan TW, Zhang M, Gao H. Ergosterol production by fed-batch fermentation of Saccharomyces cerevisiae. Enzyme Microb Technol 2003;33:36670. [11] Tong Q, Wang G, Du GC, Chen J. Effect of feeding method on production of glutathione by Saccharomyces cerevisiae. Ind Microbiol 2003;33: 1922. [12] Chen J, Zhang S, Chu J, Zhuang Y, Luo J, Bai H. Ethanol evolution rate: a new parameter to determine the feeding rate for the production of avermectins by Streptomyces avermitilis. Biotechnol Lett 2004;26: 10913. [13] Li Y, Chen J, Mao YY, Lun SY. Effect of additives and fed-batch culture strategies on the production of glutathione by recombinant Escherichia coli. Process Biochem 1998;33:70914. [14] Li Y, Chen J, Song Q, Katakura Y, Lun SY. Fed-batch culture strategy for high yield of bakers yeast with high fermentative activity. Chin J Biotechnol 1997;13:10513. [15] Patoomporn C, Shioya S, Suga K. Maximum histidine production by fedbatch culture of Brevibacterium avum. J Ferment Bioeng 1991;71:18690. [16] Nishiuch Y, Sasaki M, Nakayasu M, Oikawa A. Cytotoxicity of cysteine in culture media. In Vitro 1976;12:6358. [17] Nath KA, Salahudeen AK. Autoxidation of cysteine generates hydrogen peroxide: cytotoxicity and attenuation by pyruvate. Am J Physiol 1993;264:30614. [18] Vina J, Saez GT, Wiggins D, Roberts AF, Hems R, Krebs HA. The effects of cysteine oxidation on isolated hepatocytes. Biochem J 1983;313:3944.