Elevated glutathione production by adding precursor amino acids coupled with ATP in high cell density cultivation of Candida utilis

Abstract

Aims:  Three precursor amino acids and adenosine triphosphate (ATP) are necessary for fermentative production of glutathione. In this study, our aims were to develop a strategy to enhance glutathione production by adding three precursor amino acids coupled with ATP in high cell density (HCD) cultivation of Candida utilis.

Methods and Results:  A high-glutathione yeast strain, C. utilis WSH 02-08, was used in this study. Whole fermentative process for glutathione production was divided into two phases of cell growth and glutathione synthesis. Cells concentration was increased by HCD cultivation. Meanwhile, intracellular glutathione content was enhanced by the addition of three precursor amino acids. Concentrations of three precursor amino acids added at stationary phase were optimized by response surface methodology. Moreover, the addition of ATP 15 h after the addition of the three amino acids can further enhance glutathione production. Based on aforementioned phenomenon, a strategy of adding three precursor amino acids coupled with ATP was developed to enhance glutathione production.

Conclusion:  Without the addition of three precursor amino acids and the ATP, a total glutathione of 1123 mg l⁻¹ was achieved after 60-h cultivation. In comparison, addition of three precursor amino acid counterparts resulted in a total glutathione of 1841 mg l⁻¹. Moreover, by adding amino acids combined with ATP, a total glutathione of 2043 mg l⁻¹ was achieved after 72-h cultivation, increased by 81·9% and 11%, respectively, as compared with the control and the one without ATP addition.

Significance and Impact of the Study:  This is the first report on investigating changes of the intracellular three precursor amino acids and ATP, and γ-glutamylcysteine synthase activity in HCD cultivation of C. utilis for glutathione production. A strategy of combining addition of three precursor amino acids with ATP was developed to enhance glutathione production in C. utilis.

Introduction

Glutathione, as a tripeptide of glutamate, cysteine and glycine, is widely distributed in living organisms and is involved in enzyme activity regulation, protein expression and leukotrienes synthesis (Pastore et al. 2003).

Besides being extracted from some active tissues, glutathione can be produced by chemical method (Douglas 1989), enzymatic reaction (Kumagai et al. 1989), and microbial fermentation (Sakato and Tanaka 1992; Alfafara et al. 1993). Among these methods, biotechnological synthesis of glutathione has been widely exploited. To date, however, enzymatic production of glutathione (GSH) has not been commercialized because of relatively high production cost. However, alternative production of glutathione by yeast fermentation is very efficient and practical.

Some yeast strains, such as Saccharomyces cerevisiae and Candida utilis, are currently used for fermentative glutathione production on industrial scales. Being an intracellular product in yeast, a combination of high cell density (HCD) and high intracellular glutathione content can lead to a high yield of glutathione. However, an increased biomass can inevitably result in lowered intracellular glutathione content. However, the addition of precursor amino acids required for glutathione synthesis is an easy approach.

Cysteine was confirmed as a key amino acid for increasing glutathione production, but it inhibited cell growth notably at the same time in S. cerevisiae (Alfafara et al. 1992a). Therefore, a suitable cysteine addition strategy is central to enhancing GSH production without causing growth inhibition. Optimal time for cysteine addition was at stationary stage of cell growth and the single-point addition of cysteine was better than continuous or other addition methods for the enhancement of glutathione production (Alfafara et al. 1992b). Moreover, cysteine addition coupled with glycine and glutamic acid can bring a better effect on glutathione production (Wen et al. 2004).

Biosynthesis of glutathione in S. cerevisiae occurs in two adenosine triphosphate (ATP)-dependent steps (Meister 1988). First, γ-glutamylcysteine synthase (γ-GCS) catalyses the formation of γ-glutamylcysteine (γ-GC): glutamic acid+ cysteine + ATP → γ-GC + ADP + Pi; then glutathione synthase (GS) catalyses the formation of glutathione: γ-GC + glycine + ATP → glutathione + ADP + Pi.

Evidently, these enzyme catalysing reactions of glutathione synthesis will be inhibited if ATP is not sufficiently supplied, which means ATP is indispensable for efficient production of glutathione besides the three precursor amino acids.

Glucose, as an important fuel molecule in most micro-organisms, can be split and converted to three-carbon unit of pyruvate with small amounts of ATP and nicotinamide adenine dinucleotide reduced (NADH) captured during glycolytic reactions. Moreover, in the presence of molecular oxygen (O₂), pyruvate is further channelled through citric acid cycle and degraded to carbon dioxide (CO₂) with substantial amount of ATP generated. Glycolytic pathway followed by citric acid cycle in S. cerevisiae was considered the major way to generate sufficient ATP for fermentative glutathione biosynthesis (Murata et al. 1981).

In conclusion, both ATP and three precursor amino acids are closely related to GSH production. However, the effects of the addition of three precursor amino acids coupled with ATP on glutathione production have never been reported in HCD cultivation of C. utilis.

We previously observed that cysteine addition combined with dissolved oxygen (DO) controlling can achieve an efficient production of glutathione in HCD cultivation of C. utilis (Liang et al. 2008a). In this work, addition of cysteine accompanied by glutamic acid and glycine was further applied to fed-batch culture in C. utilis. Moreover, ATP was added 15 h after addition of a mixture of three precursor amino acids as glutathione-specific production rate stopped increasing. By adopting this strategy of imposing three precursor amino acid addition with ATP in HCD cultivation of C. utilis, a high glutathione production can be achieved.

Materials and methods

Micro-organism and culture media

A high-GSH yeast strain, C. utilis WSH 02-08, was used in this study. The seed medium contained (g l⁻¹): glucose 20, peptone 20 and yeast extract 10 at pH 6·0. The seed culture was prepared in a flask on a reciprocal shaker at 200 rev min⁻¹ and 30°C for 20 h. The medium for batch fermentation contained (g l⁻¹): glucose 15, ammonium sulfate 8, KH₂PO₄ 3 and MgSO₄ 0·25.

HCD cultivation of Candida utilis

The HCD cultivation of C. utilis was done as described in our previous method (Liang et al. 2008a). After 45-h cultivation with three-stage operation mode (batch culture, exponential glucose feeding and constant glucose feeding), cell concentration reaches 102 g l⁻¹ and the glutathione yield is 981 mg l⁻¹.

Control of DO and pH

During cell rapid growth stage, aeration was maintained at 1 v/v/m and agitation was operated at 400 rev min⁻¹. In glutathione synthesis stage after the addition of a mixture of three precursor amino acids, DO was controlled at 5% in former 3 h, and 20% in later 24 h by maintaining aeration at 1 vvm and adjusting agitation in the range of 50–200 rev min⁻¹. The pH was controlled automatically at 5·5 by adding 3 mol l⁻¹ H₂SO₄ or 3 mol l⁻¹ NaOH solutions.

Design of response surface methodology (RSM)

As cell growth stopped after 45-h cultivation in 7·0-l fermenter, culture broth was transferred to several 250-ml flasks each containing 50-ml medium to carry out RSM experiment. Concentrations of three amino acids added at 45 h were optimized by using a central composite design. A series of mixture of three precursor amino acids with different concentrations were added at 45 h, and glutathione synthesis was stopped at 60 h. Culture conditions of temperature and agitation rate were fixed at 30°C and 200 rev min⁻¹, respectively. Three factors were: glutamic acid (X₁), glycine (X₂) and cysteine (X₃). The ranges of the three factors to be evaluated were: 20 mmol l⁻¹ ≦ X₁, X₂, X₃ ≦ 40 mmol l⁻¹. Evaluated response Y was intracellular GSH yield (mg l⁻¹).

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 glutathione assay. Glutathione concentration was determined according to the method described by Tietze (1969). Dry cell weight (DCW) was determined after drying the cells at 105°C to a constant weight. The concentration of intracellular ATP was determined by high performance liquid chromatography (HPLC) according to the method described by Veciana-Nogues et al. (1997). HPLC was equipped with a Hypersil ODS column (4·6 × 200 mm) packed with 5-μm particle size C18 packing material and a UV detector (254 nm). Solvent A was CH₃OH, and solvent B was 0·1 mol l⁻¹ KH₂PO₄ at pH 7·0 adjusted with 0·1 mmol l⁻¹ KOH solutions. The gradient programme was 100% of solvent B from 0 to 5 min with the flow rate of 1 ml min⁻¹, and 90% solvent B from 5 to 25 min with the flow rate of 1·5 ml min⁻¹. Three precursor amino acids were detected according to the method described by Wang et al. (2006). The γ-GCS activity was assayed as described by Kenchappaa et al. (2003). All experiments were performed in triplicate and repeated twice, and the average of the results was used.

Results

Optimization of precursor amino acid composition by RSM

Statistical experimental design, as an efficient way to improve experimental works, has been widely used in chemistry, chemical engineering and biotechnology (Fang 1980; Atkinson 1995; Morgan 1995). Among these design methods, RSM is considered as a powerful technique for testing multiple process variables and identifying interactions between these variables, and a combination of factors generating an optimal response can be identified by this technique (Li et al. 2002). Moreover, application of RSM to fermentation process can result in an increased product yield and reduced process variability, development time and overall cost (Elibol 2003).

In this study, HCD can be obtained by fed-batch culture, but intracellular glutathione content is low after 45-h cultivation. Addition of three precursor amino acids required for glutathione synthesis is an easy approach to increase intracellular glutathione content. To determine optimal concentrations of three amino acids added at stationary stage of cell growth, the RSM experiment was applied. Based on cysteine added (data not shown) at stationary phase in recent research, the addition concentrations of three precursor amino acids were set in the range of 20–40 mmol l⁻¹

Variable design and experiment results are shown in Tables 1 and 2. Each amino acid has five levels and corresponding 20 experiments were designed. Each experiment was carried out in triplicate and repeated twice, and the average of the experimental results was used. After 60-h cultivation, highest experimental value of glutathione yield (1823 mg l⁻¹) was obtained in flasks as addition concentrations of the three precursor amino acids were all at 30 mmol l⁻¹.

Table 1.   Design of variables

Variables	Symbols	Coding
		−1·682	−1	0	1	1·682
Glutamic acid (mmol l^−1)	X1	20	24	30	36	40
Glycine (mmol l^−1)	X2	20	24	30	36	40
Cysteine (mmol l^−1)	X3	20	24	30	36	40

Table 2.   Design and results of response surface methodology (RSM) experiment

No.	X1	X2	X3	Y
	Glutamic acid	Glycine	Cysteine	GSH yield (mg l^−1)
1	0·000	0·000	0·000	1823·5
2	1·000	−1·000	−1·000	1723·5
3	0·000	0·000	1·682	1732·6
4	−1·000	−1·000	−1·000	1731·1
5	1·000	1·000	1·000	1791·1
6	0·000	1·682	0·000	1773·2
7	0·000	0·000	0·000	1823·5
8	1·000	−1·000	1·000	1773·5
9	1·000	1·000	−1·000	1783·5
10	0·000	0·000	0·000	1823·5
11	1·682	0·000	0·000	1805·0
12	0·000	1·682	0·000	1734·2
13	−1·000	−1·000	1·000	1752·1
14	−1·000	1·000	−1·000	1766·4
15	0·000	0·000	0·000	1823·5
16	0·000	0·000	0·000	1823·5
17	−1·000	1·000	1·000	1773·5
18	0·000	0·000	0·000	1823·5
19	0·000	0·000	−1·682	1669·2
20	−1·682	0·000	0·000	1791·3

Regression coefficients for glutathione yield are shown in Table 3, and regression equations of glutathione yield can be achieved according to these coefficients. Correlation coefficient of regression models of glutathione yields was 96·7%, which meant a high fitting accuracy. The parameters of SE coefficient, T and P, represent coefficients of standard errors, statistical calculation value of test and value of coefficient test, respectively.

• 

In the aforementioned equation, Y represents glutathione yield (mg l⁻¹); X₁, X₂ and X₃ represent concentrations of glutamic acid, glycine and cysteine, respectively (mmol l⁻¹).

Table 3.   Regression coefficients for glutathione yield

Term	Coefficient	SE coef.	T	P
Constant	1109·47	76·33	14·54	0·000
X1	2·06	2·38	0·87	0·407
X2	15·64	2·38	6·58	0·000
X3	46·47	2·38	11·13	0·000
X1 × X1	−0·06	0·03	−2·0	0·073
X2 × X2	−0·21	0·03	−7·5	0·000
X3 × X3	−0·40	0·03	−14·1	0·000
X1 × X2	−0·03	0·04	−0·68	0·511
X1 × X3	−0·04	0·04	0·96	0·358
X2 × X3	−0·07	0·04	−1·84	0·096
R^2	96·7%

Based on achieving maximum glutathione yield, optimal concentrations of three precursor amino acids (glutamic acid 37 mmol l⁻¹, glycine 35 mmol l⁻¹, cysteine 32 mmol l⁻¹) were calculated based on the aforementioned equation.

As indicated in Fig. 1, among three tested amino acids, both cysteine and glycine influenced GSH production significantly, but glutamic acid showed relatively less significant effect, which was consistent with the conclusion of Wen et al. (2005). Glutamic acid can be easily biosynthesized from intermediates via glycolysis and action of glutamate dehydrogenase (Eikmanns 2005; Yokota and Lindley 2005). As a primary metabolite, glutamic acid can be easily produced and can satisfy other substances’ synthesis requirements, such as glutathione production. Multi-metabolic pathways for glutamic acid synthesis in cell itself can explain why glutathione production was not significantly affected by glutamic acid.

Figure 1.  The surface and contour plots of response surface methodology.

Effects of the three optimized amino acids on glutathione production

The HCD cultivation in 7·0-l fermenter was implemented by applying three-stage operation mode (batch fermentation stage, exponential glucose feeding and constant glucose feeding). Cell concentration reaches 102 g l⁻¹ and glutathione yield is 981 mg l⁻¹ after 45-h cultivation (Fig. 2). Moreover, without the three amino acid addition, maximum glutathione yield of 1123 mg l⁻¹was achieved after 60-h cultivation.

Figure 2.  Time course of cell growth, glucose consumption and glutathione production. (□) Glucose; () dry cell weights (DCW); (⋄) intracellular glutathione content; (◆) glutathione concentration.

In comparison, a mixture of optimal concentrations of three precursor amino acids (glutamic acid 37 mmol l⁻¹, glycine 35 mmol l⁻¹, cysteine 32 mmol l⁻¹) calculated by RSM in flasks were further added into the broth in 7·0-l fermenter to enhance glutathione production at 45 h as glucose feeding ceased. As shown in Fig. 3, glutathione concentration increased immediately after glucose feeding cessation and addition of three amino acids at stationary stage of cell growth. Moreover, 15 h after the three precursor amino acid addition, glutathione-specific production rate decreased greatly and GSH concentration stopped increasing. At this point, glutathione production reached the maximal value of 1841 mg l⁻¹ and corresponding intracellular GSH content was 1·83% (w/w).

Figure 3.  Effects of the three amino acid addition on glutathione production. () Glutathione concentration; (◆) glutathione-specific production rate.

Glutathione is synthesized by two consecutive reactions catalysed by enzymes in the presence of ATP and three precursor amino acids. These reactions will be inhibited if ATP or precursor amino acids are not sufficiently abundant or if γ-GCS activity was completely inhibited. Based on the aforementioned status of glutathione synthesis, we reasonably concluded that cessation of glutathione production can be attributed to three factors: lack of ATP or/and three precursor amino acids or inhibition of γ-GCS activity.

Changes of the intracellular three amino acids and ATP concentrations and γ-GCS activity

To investigate whether the cessation of glutathione production was caused by insufficient ATP or shortage of three precursor amino acids, or the inhibition of γ-GCS activity concentrations of intracellular ATP and three amino acids and γ-GCS activity were determined during the period of 45–60 h.

As shown in Fig. 4, intracellular concentrations of three precursor amino acids reached the highest 3 h after addition and then decreased continuously with extending of cultivation time. After 60-h cultivation, three intracellular precursor amino acids was not exhausted, which suggested that inhibition of glutathione synthesis was not caused by the shortage of these amino acids.

Figure 4.  Dynamic changes of intracellular concentrations of three constituent amino acids and adenosine triphosphate (ATP), and γ-glutamylcysteine synthase (γ-GCS) activity. (□) Intracellular concentration of glycine; (⋄) intracellular concentration of glutamic acid; (△) intracellular concentration of cysteine; (○) intracellular concentration of ATP; (◆) γ-GCS activity.

Meanwhile, intracellular ATP was measured and results indicated that intracellular ATP concentration decreased rapidly after addition of three amino acids at 45 h. With prolonging of glutathione synthesis, ATP concentration was lowered continuously and almost consumed after 54-h cultivation.

As it is known, two molecule of ATP are needed for synthesizing one molecule of glutathione. For fermentative production of glutathione, ATP is mainly generated from glycolysis and citric acid cycle system in C. utilis by splitting and converting glucose to CO₂. It is evident that the amounts of ATP produced by cells themselves are strongly dependent on glucose availability. In our present study, with the cessation of glucose feeding and the addition of three precursor amino acids at 45 h, ATP was exhausted rapidly with extension of cultivation time and almost exhausted at 54-h cultivation.

However, at this time period, glutathione production increased continuously until at 60 h. Under physiological conditions, surplus of energy produced by cells themselves was stored in several forms, such as phosphates, inositol and ATP. As intracellular ATP was not enough to meet the needs in the synthesis of some metabolites, such as glutathione, energy in other forms was immediately transferred into ATP to continue those reactions. Hence, we can reasonably conclude that other forms of energy stored in cells were released and transformed as ATP to continue glutathione production at 54 h as intracellular ATP were depleted. This means the ATP needed for glutathione production at 54–60 h were mainly coming from the transfer of energy in other forms.

The γ-GCS is a critical enzyme in glutathione synthesis (Ruiz and Blumwald 2002), and its activity is feedback-inhibited by glutathione (Richman and Meister 1975). The activity of γ-GCS during 45- to 60-h period was assayed, and the results indicated that γ-GCS decreased gradually with an increase in intracellular glutathione content, which suggested that cessation of glutathione production was partially attributed to the decrease in γ-GCS.

Based on earlier results, It was suggested that cessation of glutathione production at 60 h was caused by ATP shortage and decrease in γ-GCS, but not the lack of three precursor amino acids. However, compared with the decrease in γ-GCS activity, we convincingly believed that the shortage of ATP may account for a lion’s share in inhibiting glutathione production. Hence, a strategy of combining the addition of three amino acids with ATP was developed to further enhance glutathione yield as its production stopped at 60-h cultivation.

Addition of ATP to further enhance glutathione production

To further enhance glutathione production, ATP with different concentrations was added into the broth at 60 h as glutathione-specific production rate stopped increasing. Results shown in Fig. 5 indicated that glutathione-specific production rate and glutathione concentration increased immediately after ATP addition. With an increase in ATP addition concentration, glutathione-specific production rate was enhanced accordingly. However, there was no distinct difference between glutathione production with ATP addition at 3 and 4 g l⁻¹. Moreover, after 9 h of ATP addition, both glutathione-specific production rate and its concentration stopped increasing. Therefore, 3 g l⁻¹ of ATP was added into the broth to enhance glutathione production in the following work. After 72-h cultivation, GSH yield reaches 2043 mg l⁻¹ and the corresponding intracellular GSH content is 2·03% (w/w).

Figure 5.  Effect of adenosine triphosphate (ATP) addition on glutathione production. () 1 g l ⁻¹; (◆) 2 g l⁻¹; () 3 g l⁻¹; (×) 4 g l⁻¹.

Discussion

Micro-organisms, such as S. cerevisiae and C. utilis, have been applied in fermentative production of glutathione on industrial scales. The ultimate aim for glutathione fermentation was to achieve a high glutathione yield which can be obtained by increasing cell density and intracellular glutathione content. However, an increased cell density is often offset by a decrease in intracellular glutathione content.

Glutathione production by C. utilis was closely related to cell growth and intracellular amino acid metabolism. We previously showed that a high glutathione yield can be achieved through HCD cultivation and cysteine addition.

Based on a previous study, we further enhance glutathione production by optimizing the three precursor amino acid compositions by RSM in this work. Results suggested that both cysteine and glycine showed an competent effect on enhancing glutathione production. Glutamic acid did not show a significant effect. After 60-h cultivation, a total glutathione of 1841 mg l⁻¹ was obtained and the corresponding intracellular glutathione content was increased from 0·96% to 1·83% (w/w).

Moreover, it was observed that glutathione production ceased increasing 15 h after the addition of three precursor amino acids. By further investigating the dynamic changes of intracellular three precursor amino acids and ATP, and γ-GCS activity, we demonstrated that cessation of glutathione production was caused by shortage of ATP and a decrease in γ-GCS, but not the lack of three precursor amino acids.

According to the earlier results, a strategy of addition of three precursor amino acids combined with ATP was developed intending to further improve glutathione production, and the results verified the effectiveness of this proposed strategy.

By adopting this strategy, glutathione-specific production rate increased immediately after addition of ATP and a final glutathione yield of 2043 mg l⁻¹ was achieved at 72-h cultivation, which was 11% higher than that without ATP addition.

For optimal production of glutathione by fermentation in C. utilis, an efficient ATP regeneration system was indispensable besides three precursor amino acids. Glucose, as a carbon source, is converted to ATP which in turn is utilized for cell growth and glutathione production.

To achieve a high glutathione yield, the fermentative process for GSH production was divided into two stages of cell growth followed by glutathione synthesis in this study. At cell growth stage, ATP, produced by utilizing glucose, can satisfy cell growth and glutathione synthesis sufficiently (data not shown). However, during the stage of glutathione synthesis, ATP-generation system was stopped with the cessation of glucose feeding. Clearly, enhancement of glutathione production to a higher degree can be achieved by adding ATP at 60 h as intracellular ATP was exhausted.

Meanwhile, overproduction of glutathione will inevitably cause inhibition of γ-GCS; hence, taking some measures to release this inhibition must be considered as intracellular glutathione reaches to a relatively higher level. Overexpression of γ-GCS was believed as an efficient way to release this inhibition. For example, Noctor et al. (1998) reported that overexpression of γ-GCS in chloroplast of poplar can markedly increase the γ-GC and GSH levels, which was further explained by the metabolic pathway by Mendoza-Co′zatl and Moreno-Sa′nchez (2006). Moreover, we previously observed that low pH stress can alleviate the feedback inhibition of intracellular glutathione on γ-GCS activity by secreting glutathione into the medium (Liang et al. 2008b).

Furthermore, owing to the fact that glutathione yield can be effectively enhanced by adding three precursor amino acids coupled with ATP, it is suggested that the strategy developed in this study is a feasible method in practical application of glutathione fermentation. To further minimize production cost, ATP regeneration, not through addition of ATP, but by conversion from glucose, may be another promising way for enhancing glutathione production in HCD cultivation of C. utilis in future studies.

Acknowledgements

This project was financially supported by Chinese National 863 Project (2006AA10Z313), and the Innovation Fund of Medium or Small Science and Technology Enterprise (06C26213201074).