Sodium benzoate stimulates xylitol production by Candida mogii

Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

Contents lists available at ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice

Sodium benzoate stimulates xylitol production by Candida mogii Sarote Sirisansaneeyakul a,b,*, Ben Kop a, Worasit Tochampa c, Siwaporn Wannawilai a,b, Ravipim Chaveesuk d, Wen-Chien Lee e a

Department of Biotechnology, Faculty of Agro-Industry, Kasetsart University, Bangkok 10900, Thailand Center for Advanced Studies in Tropical Natural Resources (CASTNAR), NRU-KU, Kasetsart University, Bangkok 10900, Thailand c Department of Agro-Industry, Faculty of Agriculture, Natural Resources and Environment, Naresuan University, Phitsanulok 65000, Thailand d Department of Agro-Industrial Technology, Faculty of Agro-Industry, Kasetsart University, Bangkok 10900, Thailand e Department of Chemical Engineering, National Chung Cheng University, Minhsiung 621, Taiwan b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 23 April 2013 Received in revised form 10 September 2013 Accepted 15 September 2013 Available online 18 October 2013

Xylitol is an important commercial sweetener that can be produced by fermentation. Previous studies of xylitol production have not been able to combine high average productivity and yield in a single process. Benzoate is a highly chaotropic stressor and has been found to stimulate the fermentation metabolism of yeasts at low concentrations. Therefore, in the current work, it was hypothesized that benzoate increases xylitol production, because it is a concurrent kosmotropic/compatible solute under aerobic conditions. Shake flask experiments without control of dissolved oxygen revealed that sodium benzoate added at a concentration of up to 150 ppm stimulated the fermentation of xylose to xylitol by Candida mogii. Sodium benzoate at a concentration of >200 ppm had a clear inhibitory effect on the xylose metabolism. Controlled batch fermentations carried out in bioreactors with and without sodium benzoate (150 ppm) were used to further assess its potential for improving the xylitol production. Under highly aerobic conditions (dissolved oxygen concentration >75% of air saturation), the presence of sodium benzoate (150 ppm) increased both specific xylitol productivity and yield. The specific xylitol productivity increased by >2-fold and the yield increased by 30%, relative to control. However, benzoate increased the xylitol yield only slightly under oxygen limiting conditions. The volumetric xylitol productivity previously obtained for most potential strains was highest under microaerobic conditions. However, in the current study, the xylose consumption was remarkably enhanced under aerobic conditions in the presence of sodium benzoate, which makes an increase in the volumetric productivity of xylitol feasible for industrial applications. This method is readily applicable to previously developed xylitol processes by simply adding a suitable amount of sodium benzoate. The findings of this study devise new interventions for microbial processes in industrial reactors to expand the microbial tolerance of chaotropic stressors and, hence, the biotic windows for such processes. ß 2013 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Keywords: Xylitol Xylose Sodium benzoate Dissolved oxygen Candida mogii

1. Introduction Xylitol is a natural five-carbon sugar alcohol that is widely used as a sweetener in the food industry [1]. Xylitol for commercial use is produced in an expensive chemical process [1,2]. An alternative is to convert D-xylose derived from plant material to xylitol by yeast fermentation. Such processes have been extensively studied [3–5], but remain expensive. Earlier studies revealed that the yeast Candida mogii is a promising producer of xylitol from xylose [6]. In

* Corresponding author at: Department of Biotechnology, Faculty of AgroIndustry, Kasetsart University, Bangkok 10900, Thailand. Tel.: +66 2 5625086; fax: +66 2 5794096. E-mail addresses: [email protected], [email protected] (S. Sirisansaneeyakul).

general, the yeast converts xylose to xylitol, which is then consumed for cell growth and maintenance. Accumulation of xylitol occurs if its consumption is reduced or prevented. Benzoate, a well-known growth inhibitor, offers the possibility of enhancing xylitol accumulation by suppressing its consumption and xylose fermentation. Sodium benzoate is a highly chaotropic stressor [7,8], similar to other aromatics such as phenol and benzyl alcohol, which have a comparable log P and chaotropicity-mediated mode of action, inhibit cellular systems, and induce stress responses to protect macromolecular systems against the macromolecule-disordering effects of the aromatic solute [9,10]. These include the synthesis of kosmotropic/stabilizing compatible solutes such as polyols, including xylitol [7], which can reorder membranes, proteins, and other structures in the presence of a kosmotropic substance or compatible solute, as demonstrated in various studies [10–13].

1876-1070/$ – see front matter ß 2013 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jtice.2013.09.007

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

Nomenclature C O2 CP CNa qO2 qP qS YP/S YX/S

oxygen concentration (mmol/L) maximum xylitol concentration (g/L) sodium benzoate concentration (ppm) specific oxygen uptake rate (mmol/(g h)) specific xylitol production rate (g/(g h)) specific xylitol consumption rate (g/(g h)) xylitol yield (g/g) biomass yield (g/g)

Greek letters m specific growth rate (d1)

This has also been demonstrated via studies of the protective effects of compatible solutes, including polyols, on the activity of enzymes and other macromolecular systems [10,12,14,15]. Benzoate is a widely used food preservative. At low concentrations, benzoate has been found to stimulate the fermentation metabolism of certain yeasts [16–19]. Warth [19] showed that a low concentration (0.5 mM) of benzoic acid reduced the growth but stimulated the fermentation rate of Saccharomyces cerevisiae. The increased fermentation rate was explained by the energy required by the yeast to transport benzoic acid out of the cell [19]. Up to a benzoic acid concentration of 0.5 mM, the ATP levels and the intracellular pH remained relatively high. At higher concentrations of benzoic acid, the fermentation was inhibited, leading to a decrease in ATP levels and intracellular pH. In further studies, benzoic acid was found to inhibit glycolysis at the levels of pyruvate kinase and phosphoglycerate kinase [19]. A similar protective mechanism reduces the energy available for growth in the presence of sorbic acid in S. cerevisiae cultures [16]. Verduyn et al. [17] showed that low concentrations of benzoic acid stimulated yeast respiration. Low amounts of benzoate decreased biomass yield but increased the specific oxygen uptake rate [17]. High concentrations of benzoate reduced the specific oxygen uptake rate and promoted alcoholic fermentation [17]. Francois et al. [20] reported increased concentrations of glucose-6phosphate and fructose-6-phosphate in yeast cells in a medium supplemented with benzoate. A decreased concentration of fructose-1,6-diphosphate indicates an inhibition of 6-phosphofructo-1-kinase. Krebs et al. [21] found that phosphofructokinase was inhibited to a greater extent than hexokinase at acidic pH. Benzoate was found to reduce the ATP levels [21]. No information is available on the effect of benzoate on xylitol production. In view of the above mentioned effects of benzoic acid on yeast metabolism, we hypothesized that supplementation of the culture medium with benzoic acid has the potential to enhance the xylitol yield and productivity by affecting the metabolism in two possible ways (Fig. 1). Firstly, a possible inhibition of phosphofructokinase could divert the metabolic flux through the pentose phosphate pathway to trigger a greater regeneration of NADPH (Fig. 1). This, in turn, may lead to an increased activity of the enzyme xylose reductase and, therefore, an enhanced specific xylitol production rate. Secondly, an increased ATP demand may lead to an increased xylose uptake rate with a consequential increase in the production of NADH and increased production of ATP through respiration. Therefore, the addition of benzoic acid is expected to force the cells to produce more ATP to transport the benzoate out of the cell. An increase in the ATP demand should initiate more xylose uptake, more NADH production, and an increased productivity. An excess of NADH may contribute to an increased xylitol yield by inhibiting the xylitol dehydrogenase. The

735

effect of sodium benzoate on xylitol production has not been examined, yet. This work focused on assessing the hypothesized effects.

2. Materials and methods 2.1. Microorganism, culture media, and inoculum preparation Candida mogii ATCC 18364 (TISTR 5892) was maintained on YM agar slants at 4 8C. The agar slant medium contained (per L) 3 g yeast extract, 3 g malt extract, 5 g peptone, 10 g glucose, and 20 g agar. A newly sub-cultured slant was incubated at 37 8C for 24 h. A loopful of this culture was inoculated into a minimal medium to produce the preculture for the experiments. The minimal medium had the following composition (per L of medium): 18.75 g KH2PO4, 6 g (NH4)2HPO4, 1.13 g MgSO47H2O, 0.1 g CaCl2, 36.5 mg myoinositol, 18.2 mg calcium pantothenate, 3.66 mg thiamine-HCl, 0.9 mg pyridoxal-HCl, 0.018 mg biotin, 9.1 mg FeCl3, 6.4 mg MnSO4 H2O, 5.46 mg ZnSO47H2O, 1.46 mg CuSO45H2O, and 20 g glucose [6,22]. For preparing the inoculum, 1 mL of the yeast suspension with an optical density of 0.8 at 620 nm (0.4 g/L dry cell weight) was transferred to each of two 250-mL Erlenmeyer flasks. Each flask contained 20 mL of the minimal medium described earlier. The yeast was grown aerobically at 250 rpm, 30 8C, for 24 h. The contents of each flask were then transferred to 500-mL Erlenmeyer flasks, each containing 180 mL of the minimal medium. These flasks were cultivated under the conditions specified above. 2.2. Shake flask fermentations Inocula were grown on a rotary shaker at 250 rpm for 24 h. The first pre-cultures were grown in test tubes containing 10 mL of the minimal medium. This preculture (2.5 mL) was used to inoculate a 250-mL Erlenmeyer flask containing 22.5 mL of the minimal medium (10% inoculum). Xylitol production was carried out in 500-mL Erlenmeyer flasks containing 250 mL of the minimal medium without glucose and initially 10 g/L of xylose. Eight shake flasks were simultaneously cultured on a rotary shaker (250 rpm, 30 8C). The flasks had different concentrations of sodium benzoate: 0, 100, 150, 200, 300, 400, 500, and 600 ppm. The pH was manually controlled at pH 6.0 by adjusting every 2 h as necessary. The flasks were sampled every 2 h. The samples were analyzed for concentrations of biomass, xylose, and xylitol. 2.3. Bioreactor batch fermentations 2.3.1. Effect of sodium benzoate under aerobic conditions A laboratory stirred-tank bioreactor (B.E. Marubishi, Japan) was used for aerobic cultures. The yeast was grown for 24 h in 3.7 L of minimal medium initially containing 10 g/L glucose and 5 g/L xylose as carbon sources. The dissolved oxygen (DO) concentration was kept above 75% of air saturation by operating the bioreactor at an aeration rate of 1 vvm and an agitation speed of 600 rpm. The pH was maintained at 4.5 by automatic addition of 6 M NaOH, as needed. The temperature was controlled at 30 8C. The xylitol production phase was started by adding 300 mL of a solution of Dxylose to achieve a xylose concentration of 10 g/L in the culture medium. The pH was then manually adjusted to 6.0 by adding 6 M NaOH. Once all xylose and xylitol had been consumed, the next experiment began by adding 250 mL of a xylose solution and 50 mL of a sodium benzoate solution so that the initial concentrations were 10 g/L for xylose and 150 ppm for sodium benzoate. All other conditions remained as specified above. The fermentation was run until the carbon source was depleted.

736

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

Fig. 1. Effect of benzoic acid on xylose catabolism: benzoic acid has an inhibitory effect on the enzyme phosphofructokinase (PFK). This is proposed to increase the metabolic flux along the thickened arrows and reduce the flux for the dashed arrows. Enzyme and metabolite abbreviations: XR, xylose reductase; XDH, xylitol dehydrogenase; XuK, xylulokinase; G6PDH, glucose-6-phosphate dehydrogenase; 6PGDH, 6-phosphogluconate dehydrogenase; PGI, phosphoglucose isomerase; PFK, phosphofructokinase; TPI, triose phosphate isomerase; TK, transketolase; TA, transaldolase; PDC, pyruvate decarboxylase; PDH, pyruvate dehydrogenase; ADH, alcohol dehydrogenase; ALDH, aldehyde dehydrogenase; Ru5P, ribulose-5-phosphate; Ri5P, ribose-5-phosphate.

2.3.2. Effect of sodium benzoate at a dissolved oxygen concentration of 20% A high cell density was used at inoculation as this experiment intended to measure the intracellular levels of glycerol and acetate. A 5-L stirred fermentor (Biostat B, B. Braun Biotech International, Germany) was used. For the growth phase, the yeast was grown in 3.5 L of minimal medium initially containing 40 g/L glucose and 10 g/L of xylose. After the carbon sources were depleted, the fermentation was switched to the xylitol production phase. DXylose solution (500 mL) was added, to achieve an initial concentration of 30 g/L. The DO concentration was controlled at

20% of air saturation by automatic control of the stirring speed. Once the carbon source was depleted, the benzoate-supplemented xylitol production phase was started by adding 500 mL of a solution of D-xylose and 100 mL of a solution of sodium benzoate to achieve initial concentrations of 30 g/L and 150 ppm, respectively. 2.3.3. Effect of sodium benzoate at 1% dissolved oxygen The fermentations were run in a 5-L stirred fermentor (Biostat B). The yeast was grown in 3.5 L of minimal medium containing initially 10 g/L glucose and 5 g/L xylose. Once the carbon sources had been depleted, the fermentation was switched to the xylitol

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

production phase. The DO was controlled at 1% of air saturation before adding the xylose solution in the xylitol production phase. For the experiment without sodium benzoate, a 500-mL solution of D-xylose was added to achieve an initial concentration of 30 g/L in the fermentor. For the experiment with sodium benzoate, a 450mL solution of D-xylose and a 50-mL solution of sodium benzoate were added so that the initial concentrations of D-xylose and sodium benzoate were 30 g/L and 150 ppm, respectively. The DO was controlled at 1% of air saturation by automatic adjustments of the stirring speed (380–450 rpm). The aeration rate remained in the range of 0.1–0.3 vvm. 2.4. Analytical methods Broth samples (duplicates) were centrifuged (1700  g, 10 min; Sigma 203 centrifuge, B. Braun Biotech International) and the supernatants were kept for further analysis. The cell pellet was washed twice, each time with 5 mL of deionized water, and recovered by centrifugation as above. The cells were then dried to a constant weight (24 h, 105 8C), cooled in a desiccator, and weighed to calculate the dry cell concentration. The culture supernatant (duplicates) was analyzed for xylose and xylitol. The xylose concentration was determined spectrophotometrically according to the method of Deschatelets and Yu [23]. The principle was based on the formation of furfural from pentose in acetic acid/thiourea solution at 70 8C, which further reacts with p-bromoaniline acetate to form a pink-colored product, whose absorbance was measured at 520 nm. The determination of xylitol was performed with the method of Adler and Gustafsson [24]. The reaction was based on the oxidation of xylitol to formaldehyde by periodate in acid solution, interrupted by addition of 2,3-butanediol, and further reacted with 2,4-pentanedione/ammonium acetate to form a yellow-colored product, whose concentration was determined spectrophotometrically at 410 nm. Duplicate samples taken for analysis of the intracellular glycerol and acetate content were immediately placed in a water bath at 100 8C for 7 min to stop the metabolic reactions and then rapidly cooled in an ice bath. The cells were recovered by centrifugation and frozen at 18 8C for later analysis. Subsequently, the cells were disrupted by agitation with glass beads. For this, an Eppendorf tube was filled with 0.4 g of wet cells, 0.6 g of glass beads (0.3 mm in diameter), and 0.5 mL of deionized water. The tube was shaken vigorously (10 pulses of 10 s each with cooling on an ice batch after each pulse) on a mini-Beadbeater (Biospec Products, Wakenyaku Co., Ltd., Japan). The supernatant was separated from the cell debris and glass beads by centrifugation (1700  g, 10 min). Glycerol and acetate in the supernatant were measured enzymatically (Boehringer test kits cat. no. 10148270035 and 10148261035; Boehringer Mannheim R-Biopharm, Germany). All analyses were conducted in duplicate and average data were shown. 2.5. Calculations of the fermentation parameters To obtain the highest amount of xylitol from a batch process, the fermentation should be terminated at the point where the xylitol concentration is at its maximum. To obtain comparable results among the different experiments, the different parameters are calculated from the beginning (t1) to the specified time (t2) where the xylitol concentration (CP) is maximized during batch fermentation. 2.5.1. Specific growth rate In general, the specific growth rate (m) is obtained from an exponential growth phase in the fermentation process. It can be calculated from the slope of the curve of ln CX versus the fermentation time, t.

737

2.5.2. Yield coefficients The amount of biomass and xylitol formed per unit of substrate consumed are specified as the biomass (YX/S) and xylitol (YP/S) yields. These parameters are calculated as follows: Y X=S ¼

C X;2  C X;1 C S;1  C S;2

(1)

Y P=S ¼

C P;2  C P;1 C S;1  C S;2

(2)

where CX,1, CX,2 CS,1, CS,2 CP,1, and CP,2 are the concentrations of biomass (X), xylose (S), and xylitol (P) at the beginning (time = t1) and at the specified time of the fermentation process (time = t2), respectively. 2.5.3. Volumetric rates of substrate consumption and xylitol production Volumetric rates of substrate consumption (Qs) and xylitol production (QP) were calculated as follows: QS ¼

C S;2  C S;1 t2  t1

(3)

QP ¼

C P;2  C P;1 t2  t1

(4)

2.5.4. Specific rates of xylose consumption and xylitol production The specific rates of xylose consumption (qS) and xylitol production (qP) were calculated from the relevant volumetric rates (Qs, QP), as follows: qS ¼

1 Q ððC X;1 þ C X;2 Þ=2Þ S

(5)

qP ¼

1 Q ððC X;1 þ C X;2 Þ=2Þ P

(6)

3. Results and discussion 3.1. Xylose fermentations in shake flasks Preliminary experiments in shake flasks were used to identify the optimum concentration of sodium benzoate required for stimulating the production of xylitol. As shown in Fig. 2, the specific growth rate decreased gradually with an increase in the added sodium benzoate. Compared with the yeast growth rate in the absence of sodium benzoate (0.36 d1), the specific growth rate decreased to 0.24 and 0.072 d1 at the initially present amounts of sodium benzoate (100 and 600 ppm, respectively). On the contrary, the specific substrate consumption rate (qS) and specific xylitol production rate (qP) did not decline significantly even at sodium benzoate concentrations of 100 or 150 ppm. As shown in Fig. 2, sodium benzoate added up to a concentration of 150 ppm had a clear stimulatory effect on the biomass specific rate of the xylitol production. At stimulatory concentrations of sodium benzoate, the specific xylose uptake rate was also higher than in the control flask. In this concentration range of sodium benzoate, xylose consumption was stimulated likely because of the need to produce more ATP to provide energy for transporting the benzoate out of the cells. Under the oxygen-limited conditions of the shake flasks, addition of sodium benzoate reduced the specific growth rate compared to that of the control and some of the energy in the form of ATP that would otherwise be used for growth and normal maintenance was diverted to excreting the toxic benzoate from the cells. This diversion of ATP also reduced the biomass yield on substrate (YX/S) (Fig. 2) and, therefore, more of the xylose taken up

738

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

Fig. 2. Effect of sodium benzoate on growth and xylitol production in shake flasks with an initial xylose concentration of 10 g/L. (All samples were analyzed in duplicate, average data are shown.)

by the cell was converted to xylitol. Thus, the xylitol yield on substrate increased. As the concentration of the sodium benzoate added increased from 200 to 600 ppm, sodium benzoate inhibited growth (m and qS) but enhanced the xylitol yield, as shown in Fig. 2. At these high concentrations of sodium benzoate, the yeast could no longer remove the benzoate sufficiently and rapidly; thus, the benzoate accumulated in the cells, which likely lowered the intracellular pH [18,19]. The increased xylitol yield (YP/S) and productivity (qP) could be indirectly attributed to an inhibition of the enzyme phosphofructokinase, as shown in Fig. 1. In studies with Zygosaccharomyces bailii, Warth [18] showed that an increased concentration of benzoic acid in the culture medium led to increased concentrations of fructose-6-phosphate and glucose-6phosphate, due to a possible inhibition of phosphofructokinase. This suggests a possible enhancement of the regeneration of NADPH by diversion of the metabolic flux to the part of the pentose phosphate pathway where NADPH is regenerated. That is, the steps from glucose-6-phosphate to 6-phosphogluconate and on to ribulose-5-phosphate (Fig. 1). A portion of the ribulose-5phosphate ultimately feeds the production of glycerylaldehyde3-phosphate and the other portion reenters the pentose phosphate pathway as fructose-6-phosphate (Fig. 1). The regeneration of

NADPH met the requirement of a cofactor for the conversion of xylose to xylitol. The production of xylitol by C. mogii has been confirmed in shake flask cultures under optimal conditions obtained above (with/without 150 ppm sodium benzoate, 30 g/L xylose, 250 rpm, 30 8C) (Fig. 3). The presence of sodium benzoate (150 ppm) gave remarkable specific rates of xylose consumption/xylitol production and xylitol yield, i.e., 0.058  0.002 g/(g h), 0.018  0.001 g/ (g h), and 0.302  0.011 g/g, which were higher than those attained from the culture without sodium benzoate (0 ppm; approximately 7, 21, and 13%, respectively) (Fig. 3). Clearly, the specific growth rate of C. mogii decreased (8%) with an inhibitory amount of sodium benzoate (150 ppm), resulting in less biomass yield obtained (13%) (Fig. 3). Sodium benzoate is clearly stressful/inhibitory to yeast cells due to its chaotropic activity [7–15]. However, at low levels, it is a well-known phenomenon that stress boosts metabolism and growth. Similar results were found in this work, when C. mogii was cultured with 150 ppm sodium benzoate (Figs. 2 and 3). Aromatic chaotropic stressors are typically highly inhibitory at concentrations in the range of 5–100 mM [8,9]. This is consistent with the low concentrations of sodium benzoate (<1 mM, <144 ppm) reported in the current study, which stimulated energy generation

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

3.2. Bioreactor culture under aerobic conditions The flask experiments suggested an optimal concentration of sodium benzoate of 150 ppm. Thus, subsequent batch fermentations were carried out in bioreactors with a sodium benzoate concentration of 150 ppm. The DO concentration was controlled at various levels during the xylitol production phase. Figs. 4–6 show the time courses of biomass, xylose, and xylitol concentrations during the fermentation at three different DO concentrations with and without the addition of sodium benzoate to the medium.

35 With sodium benzoate DCW Xylose Xylitol

Concentration (g/L)

30

25

Fig. 7a shows the effect of sodium benzoate on xylitol production under aerobic conditions (non-limiting DO concentration of >75% of air saturation) of the bioreactor. The fermentation profiles under these aerobic conditions are shown in Fig. 4. In the absence of oxygen limitation, the specific growth rate (m) was higher in the presence of sodium benzoate than in its absence (1.056 and 1.152 d1, respectively). The specific xylose uptake rate (qS) was also stimulated by the presence of sodium benzoate. Under highly aerobic conditions, the specific growth rate of the control culture (no sodium benzoate) was similar to the values previously reported [22,27,28] for aerobic conditions. The stimulated specific xylose consumption rate in the presence of sodium benzoate allowed more ATP to be generated, but the biomass yield on the substrate YX/S was lower than in the absence of sodium benzoate (Fig. 7a). Therefore, more energy went into the maintenance metabolism to transport the benzoate out of the cell. As in the shake flask experiments, the xylitol yield and specific xylitol production rate were increased in the presence of 150 ppm of sodium benzoate (Fig. 7a). Relative to the fermentation without sodium benzoate (control), the specific xylitol productivity increased from 0.068 to 0.14 g/(g h) and the yield increased from 0.38 to 0.50 g/g. For the benzoate-supplemented culture, we hypothesized an increased ATP demand linked to the energy requirements for

14 12 11 10 9 8 7 6 5

Biomass Xylose Xylitol

4 3 2

Without sodium benzoate DCW Xylose Xylitol

20

(a) With Na-benzoate

13

Concentration (g/L)

(increased specific xylose uptake rate, qS), metabolic activity (increased specific xylitol formation rate, qP), and retarded growth (a decreased specific growth rate, m, and biomass yield, YX/S) rather than eliminated activity altogether. Importantly, studies of chaotrope-induced stress on microbial cells show that microbes respond by boosting energy generation [8,9] and this is consistent with the findings presented here. Furthermore, evidence from other microbial species demonstrates that cells can preferentially synthesize chaotropic and kosmotropic metabolites, including polyols, in order to enhance tolerance to chaotropic stressors or other conditions that have an impact in the macromolecular order, including temperature extremes [10,13,25]. This finding is also consistent with the increase in kosmotropic metabolites/xylitol in the presence of chaotropic metabolites/sodium benzoate [7]. Whereas weak chaotropic stressors such as ethanol can also cause a significant reduction of water activity at physiological concentrations [25,26], aromatic chaotropes, e.g., sodium benzoate, phenol, benzyl alcohol, and cresol, have chaotropic activity that is an order of magnitude higher (with limited solubility). As a result, the water activity is the inhibitory stress parameter, while the chaotropic activity remains constant [7–9,14].

739

1 0 0

1

2

3

4

5

15

6

7

8

9

10

11

12

13

Time (h)

10

14

(b) Without Na-benzoate

13

5

12 11

0 8

16

24

32

40

48

56

64

72

80

10

Time (h) Sodium benzoate

YX/S

YP/S

qS

μ -1

qP -1

-1

-1

-1

Concentration (g/L)

0

9 8 7

(ppm)

(g/g)

(g/g)

(d )

(g·g ·h )

(g·g ·h )

0

0.274 ±

0.267 ±

0.360 ±

0.054 ±

0.015 ±

0.007

0.006

0.000

0.000

0.000

0.238 ±

0.302 ±

0.330 ±

0.058 ±

0.018 ±

2

0.005

0.011

0.008

0.002

0.001

1

Biomass Xylose Xylitol

6 5 4 3

150

0 Fig. 3. The xylitol production confirmed in shake flask cultures from Candida mogii with/without sodium benzoate (150 ppm) with an initial xylose concentration of 30 g/L. The biomass/xylitol yields (YX/S, YP/S) and specific rates (m, qS, qP) of growth, xylose uptake, and xylitol production were calculated from 0 to 56 h. Values expressed as mean  SE and derived from duplicates have a statistically significant difference between with/without sodium benzoate at p < 0.1

0

1

2

3

4

5

6

7

8

9

10

11

12

13

Time (h) Fig. 4. Fermentation profiles of Candida mogii with (a) and without (b) sodium benzoate (150 ppm) under aerobic conditions (dissolved oxygen level of >75% of air saturation). (All samples were analyzed in duplicate, average data are shown.)

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

740

35

35

(a) With Na-benzoate

(a) With Na-benzoate

Biomass Xylose Xylitol

30

25

Concentration (g/L)

Concentration (g/L)

25 20 15 10

20 15 10 5

5

0

0 0

2

4

6

8

10

12

14

16

0

18

5

10

15

20

25

30

35

40

Time (h)

Time (h) 35

35

(b) Without Na-benzoate

Biomass Xylose Xylitol

(b) Without Na-benzoate

30

30

25

25

Concentration (g/L)

Concentration (g/L)

Biomass Xylose Xylitol

30

Biomass Xylose Xylitol

20 15 10

20 15 10 5

5

0

0 0

2

4

6

8

10

12

14

16

18

Time (h)

0

5

10

15

20

25

30

35

40

Time (h)

Fig. 5. Fermentation profiles of Candida mogii with (a) and without (b) sodium benzoate (150 ppm) at a controlled dissolved oxygen concentration of 20% of air saturation and a high biomass concentration. (All samples were analyzed in duplicate.)

Fig. 6. Xylitol production profile with (a) and without (b) sodium benzoate (150 ppm) at a controlled dissolved oxygen concentration of 1% of air saturation. (All samples were analyzed in duplicate, average data are shown.)

transporting the benzoate out of the cells. An increased need for ATP is expected to increase the rate of xylose uptake, the sole source of energy in the culture medium. An elevated ATP production in the respiratory chain, in turn, implies an increased rate of respiration (Fig. 1). The increased respiration rate in the presence of sodium benzoate, as observed in the present work, suggests that the yeast was able to re-oxidize excess cytosolic NADH to NAD+ via mitochondrial NADH-dehydrogenase and glycerol-3-phosphate shuttle in aerobic conditions [29,30]. Physiological responses of different yeasts to external stimuli can be quite different, as observed for the intracellular accumulation of NADH by Candida guilliermondii and Candida tropicalis [31– 33]. For example, the xylose reductase of C. guilliermondii depends exclusively on NADPH and acetate production regenerates NADPH [31]. The xylose reductase of C. tropicalis has a dual dependency on both NADPH and NADH [31]. The different dependency on cofactors leads to differences in physiological responses in these yeasts. In C. tropicalis, ethanol and glycerol accumulate together as the xylose consumption rate is increased by feeding formate as a cosubstrate [31]. In C. guilliermondii, formate feeding results in accumulation of glycerol and acetate with only a slight increase in the specific xylose consumption [31]. With the production of ethanol and glycerol in C. tropicalis, NADH is re-oxidized to NAD+.

The production of acetate in C. guilliermondii indicates the regeneration of NADPH. Acetate accumulation suggests that NADPH becomes a limiting factor in the conversion of xylose in C. guilliermondii. If the xylose conversion is assumed to be influenced only by coenzymes, C. mogii is expected to respond in a manner similar to that of C. guilliermondii, where NADPH is mainly used as coenzyme for converting xylose to xylitol [6]. 3.3. Batch bioreactor culture under oxygen-limited conditions (20% DO) In the absence of other nutrient limitations, the specific growth rate (m) of the yeast is expected to approach the maximum specific growth rate (mmax) under non-limiting levels of DO. In a complex medium, a high value of 1.56 d1 has been reported for the specific growth rate of C. mogii ATCC 18364 [6]. In a minimal medium, the m-value was reduced to 0.96 d1 [6]. The m-value of 1.056 d1 obtained in this study at 75% of air saturation level in a minimal medium (Fig. 7a) is quite consistent with published data. At a DO level of 20% of air saturation, however, the m-value was only 0.672 d1 (Fig. 7b), or 64% of the value observed at the 75% DO level. Therefore, a 20% DO level at the high biomass concentration used (>15 g/L initial concentration; Fig. 5b) appears to be too low to support growth at the maximum possible rate. Nevertheless, a

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

0.7

Without sodium benzoate With sodium benzoate

(a) 0.6

3.5 3.0 2.5

0.4

2.0

0.3

1.5

0.2

1.0

0.1

.5

00.0

0.0 3.0

0.5 0.4

2.5

(b)

2.0

0.3

1.5

0.2

1.0

0.1

.5

00.0

0.0

0.6

YX/S, YP/S, qP and qS

(b)

(C)

3.0

0.5

2.5

0.4

2.0

0.3

1.5

0.2

1.0

0.1

.5

00.0 Yx/s

µ, QS and Qp

YX/S, YP/S, qP and qS

0.6

µ, QS and Qp

0.5

Yp/s

qp

qs

µ mu

µ, QS and Qp

YX/S, YP/S, qP and qS

75%w1

0.0 Qs

Qp

Fig. 7. Effect of sodium benzoate on growth and xylitol production under aerobic conditions, (a) dissolved oxygen level of >75% of air saturation, (b) at 20% dissolved oxygen and (c) at 1% dissolved oxygen. (All samples were analyzed in duplicate, average data are shown.)

controlled 20% level of DO implies that the oxygen supply rate is sufficient to satisfy the oxygen demand of the metabolic processes. If the oxygen supply rate was insufficient, the DO concentration could not have been stabilized at a value of >0% of air saturation. Potentially, the hypothesis relating to the action of benzoate can be tested by comparing the intracellular levels of glycerol and acetate in cells grown under a somewhat limiting oxygen concentration in the presence of sodium benzoate, with the cells grown under the same conditions but in the absence of benzoate. Therefore, bioreactor fermentation was carried out with the DO level controlled at 20% of air saturation. To obtain a high final concentration of the biomass for the extraction of glycerol and acetate, a high initial cell concentration of 16 g/L was used. All other culture conditions were selected to be close to the optimal conditions previously identified by Tochampa [34]. The results are shown in Fig. 7b. As previously noted, the specific growth rate (Fig. 7b) was found to be much lower compared to values seen for the highly aerobic conditions (DO level of >75%) shown in Fig. 7a. The addition of sodium benzoate significantly reduced the specific growth rate as

741

compared to the control (no benzoate added) owing to the increased ATP demand. This increased diversion of energy to cell maintenance decreased the biomass yield on substrate (Fig. 7b). The data shown in Figs. 5 and 7b were for a broth with a biomass concentration that was nearly 3-fold greater compared to the broth used for the experiments shown in Fig. 7a. A controlled DO concentration of 20% of air saturation apparently influenced the metabolism to favor fermentation over growth. A high level of aeration is detrimental to xylitol production, but promotes growth [35]. Xylitol is produced under conditions of reduced oxygen availability, but the production rate is reduced if the oxygen supply is severely limited [35]. Although a DO concentration of 20% has been previously reported not to be growth limiting for C. mogii [22], the present data do not support this. Under oxygen limiting conditions, sodium benzoate lacks the stimulatory effect (Fig. 7b), as previously discussed, and the qS value is not increased. Inhibition of the enzyme 6-phosphofructo-1-kinase by sodium benzoate led to an increased yield of xylitol (Fig. 7b) from xylose as the metabolic flux was diverted into the pentose phosphate pathway (Fig. 1), as observed also in the shake flask experiments. The metabolic intermediates of glycerol and acetate were not detected in the cells and in the extracellular culture fluid of the samples taken from bioreactor cultures run at the DO level of 20% of air saturation, irrespective of whether sodium benzoate was added or not. It seems, therefore, that a 20% DO level is growth limiting, but does provide sufficient oxygen for the reoxidation of NADH. In studies with other Candida yeasts, accumulation of glycerol and acetate was detected only at quite low DO concentrations [31–33]. 3.4. Batch cultivation under severe oxygen limitations (1% dissolved oxygen level) The fermentation profile at severe oxygen limitations is shown in Fig. 6. In batch cultures in which the DO level was controlled at 1% of air saturation, oxygen was clearly a limiting substrate, as evidenced by the decreased specific growth rate upon addition of sodium benzoate (Fig. 7c). As the oxygen was limiting, there was no stimulation of the specific xylose consumption rate, as was seen in shake flask cultures. However, the specific rates of xylose consumption and xylitol production were slightly higher at the 1% DO level (Fig. 7c) compared to the values at the 20% DO level (Fig. 7b) in sodium benzoate-supplemented cultures. The higher specific growth rate at the 1% DO level (Fig. 7c, control culture) compared to the value at the 20% DO level (Fig. 7b, control) is likely a consequence of large differences in the biomass concentration of the cultures being compared. At the low oxygen level (Fig. 7c), the addition of sodium benzoate had little effect on the biomass yield coefficient and the xylitol yield coefficient. Thus, under oxygen limiting conditions, the yeast could still effectively transport the sodium benzoate out of the cells at the expense of a reduced specific growth rate. Compared to the conditions in which oxygen did not limit growth, the addition of sodium benzoate did not stimulate xylitol fermentation under oxygen limiting conditions. Effects of low level oxygen concentrations on xylose fermentation by C. mogii have been reported in the literature [6]. Oxygen was found to influence the key fermentation parameters, as shown in Table 1. The oxygen solubility in water at 30 8C is about 7.5 mg/L, or 234 mmol/L. A 20% DO level is equivalent to a concentration of about 47 mmol/L and a 1% oxygen saturation level is equivalent to a concentration of about 2 mmol/L. Thus, the 20% oxygen saturation level used in this work falls in between the 8 and 100 mmol/L DO concentrations shown in Table 1. Therefore, for the fermentation conducted at 20% DO, the specific oxygen uptake rate is expected to be in the range of 1.05 and 1.65 mmol/(g h) as shown in Table 1.

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743

742

Table 1 Effect of oxygen concentration on yields and specific rates of D-xylose uptake and xylitol production reported in the literature [6]. C O2 (mmol/L)

YX/S (g/g)

YP/S (g/g)

qS (g/(g h))

qP (g/(g h))

m (d1)

qO2 (mmol/(g h))

100 8 2

0.22 0.19 0.05

0.48 0.54 0.63

0.25 0.13 0.08

0.12 0.07 0.05

0.960 0.552 0.072

1.65 1.05 0.50

m, qP, qS = specific rates of growth, xylitol production, and xylose consumption; YP/S, YX/S = yields of xylitol and biomass, respectively.

Over the oxygen concentration range of 8–100 mmol/L, the biomass yield and the xylitol yield are not much affected by changes in the oxygen concentration; however, the specific substrate consumption rate and the specific xylitol production rate are substantially reduced by a decrease in the oxygen level (Table 1). A further small change in the DO concentration from 8 to 2 mmol/L significantly reduced the specific growth rate and the biomass yield, but increased the xylitol yield (Table 1). In the fermentor run at the 1% DO level, there was still sufficient oxygen available for an efficient reoxidation of NADH. This resulted in a higher specific growth rate (Fig. 7c) than that obtained at the 2mmol/L DO level (Table 1). In addition, in the fermentor (Fig. 7c), the rates of xylose consumption and xylitol production were high compared to the values shown in Table 1, but the xylitol yield was low. At the lowest controlled DO level used, the conditions in the fermentor were still not optimal for attaining a high xylitol yield. Under conditions where both oxygen and xylose were not limiting, the addition of benzoic acid significantly improved both xylitol yield and productivity. Under conditions where oxygen was a limiting substrate, the yeast metabolism was repressed and only a small increase in the xylitol yield was observed upon addition of sodium benzoate. When oxygen was not limiting, the increased ATP demand for transporting benzoate out of the cells increased the xylose consumption rate and the yeast metabolism was stimulated. Similar results have been reported for S. cerevisiae. Warth [19] reported that benzoate added at low levels led to an increased fermentation rate in S. cerevisiae, apparently to provide the energy needed for eliminating benzoate from the cell. In order to obtain the stimulatory benefit of sodium benzoate in the fermentation of xylitol, sufficient amounts of both oxygen and xylose must be present in the culture medium. However, a DO concentration of 1% of air saturation is not low enough for attaining a high xylitol yield [6]. This finding suggests that monitoring the microaerobic xylitol fermentation using the DO concentration may not be a satisfactory strategy as it is difficult to measure low levels of DO reliably. Instead, the specific oxygen uptake rate of the fermentation should be monitored. Unfortunately, this would require more sophisticated instruments for accurately measuring the air mass flow rate, the oxygen concentrations in the aeration, and exhaust gas streams [6]. Sometimes, adding glucose to the medium coupled with the oxidation-reduction potential-stat to control glucose-feeding can be useful for enhancing xylitol conversion from xylose during oxygen-limited fermentation [36]. 4. Conclusions Batch fermentations were used to examine the effect of sodium benzoate on xylitol production from xylose. The addition of sodium benzoate up to a concentration of 150 ppm had a clear stimulatory effect on the biomass specific rate of xylitol production. In fermentations that were not limited by both oxygen (DO concentration >75% of air saturation) and xylose, the addition of 150 ppm sodium benzoate improved both the xylitol yield on substrate and the biomass specific xylitol productivity. The specific xylitol productivity increased by >2-fold and the yield increased by 30% relative to control (fermentation without sodium

benzoate). To date, the volumetric xylitol productivity of most fermenter strains was found to be highest under microaerobic conditions. We demonstrated here that the xylose consumption by yeasts is remarkably enhanced under aerobic conditions in the presence of sodium benzoate, which causes an increase in the volumetric productivity of xylitol that is feasible for industrial applications. This method is applicable to previously developed xylitol manufacturing processes by simply adding a suitable amount of sodium benzoate. Interestingly, chaotropicity has been shown to limit microbial processes as well as render potential yeast fermentations containing chaotropic salts [7,14,25,37]. This was confirmed in the current work for xylitol in the xylose fermentation by C. mogii in the presence of sodium benzoate. This is a highly relevant and up-todate report because chaotropic and kosmotropic activities are utilized in a wide range of biotechnological processes [38–40]. Conflict of interest The authors declare no conflict of interest. Acknowledgements This work was supported by the Asian–European Master of Science and Technology Program co-funded by the European Commission, European Union, under the Asia Link Program and the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission. Prof. Yusuf Chisti, from School of Engineering, Massey University, New Zealand is grateful for the manuscript proof reading, reversion, and valuable scientific suggestions. References [1] Parajo´ JC, Dominguez H, Dominguez JM. Biotechnological production of xylitol. Part 1. Interest of xylitol and fundamentals of its biosynthesis. Bioresour Technol 1998;65:191–201. [2] Affleck RP. Recovery of xylitol from fermentation of model hemicellulose hydrolysates using membrane technology. USA: Virginia Polytechnic Institute and State University; 2000 [MSc Thesis]. [3] Granstro¨m TB, Izumori K, Leisola M. A rare sugar xylitol. Part I: the biochemistry and biosynthesis of xylitol. Appl Microbiol Biotechnol 2007;74:277–81. [4] Granstro¨m TB, Izumori K, Leisola M. A rare sugar xylitol. Part II: biotechnological production and future applications of xylitol. Appl Microbiol Biotechnol 2007;74:273–6. [5] Sirisansaneeyakul S, Chainoy R, Vanichsriratana W, Srinophakun T, Chisti Y. Xylitol production by liquid emulsion membrane encapsulated yeast cells. J Chem Technol Biotechnol 2009;84:1218–28. [6] Sirisansaneeyakul S, Staniszewski M, Rizzi M. Screening of yeasts for production of xylitol from D-xylose. J Ferment Bioeng 1995;80:565–70. [7] Cray JA, Russell JT, Timson DJ, Singhal RS, Hallsworth JE. A universal measure of chaotropicity and kosmotropicity. Environ Microbiol 2013;15:287–96. [8] Cray JA, Bell ANW, Bhaganna P, Mswaka AY, Timson DJ, Hallsworth JE. The biology of habitat dominance; can microbes behave as weeds? Microb Biotechnol 2013;6:453–92. [9] Hallsworth JE, Heim S, Timmis KN. Chaotropic solutes cause water stress in Pseudomonas putida. Environ Microbiol 2003;5:1270–80. [10] Bhaganna P, Volkers RJM, Bell ANW, Kluge K, Timson DJ, McGrath JW, et al. Hydrophobic substances induce water stress in microbial cells. Microb Biotechnol 2010;3:701–16. [11] Hallsworth JE, Prior BA, Nomura Y, Iwahara M, Timmis KN. Compatible solutes protect against chaotrope (ethanol)-induced, nonosmotic water stress. Appl Environ Microbiol 2003;69:7032–4.

S. Sirisansaneeyakul et al. / Journal of the Taiwan Institute of Chemical Engineers 45 (2014) 734–743 [12] Williams JP, Hallsworth JE. Limits of life in hostile environments: no barriers to biosphere function? Environ Microbiol 2009;11:3292–308. [13] Chin JP, Megaw J, Magill CL, Nowotarski K, Williams JP, Bhaganna P, et al. Solutes determine the temperature windows for microbial survival and growth. PNAS 2010;107:7835–40. [14] Hallsworth JE, Yakimov MM, Golyshin PN, Gillion JLM, D’Auria G, de Lima Alves F, et al. Limits of life in MgCl2-containing environments: chaotropicity defines the window. Environ Microbiol 2007;9:801–13. [15] Bell ANW, Magill E, Hallsworth JE, Timson DJ. Effects of alcohols and compatible solutes on the activity of b-galactosidase. Appl Biochem Biotechnol 2013;169:786–94. [16] Bracey D, Holyoak CD, Coote PJ. Comparison of the inhibitory effect of sorbic acid and amphotericin B on Saccharomyces cerevisiae: is growth inhibition dependent on reduced intracellular pH. J Appl Microbiol 1998;85:1056–66. [17] Verduyn C, Postma E, Scheffers WA, Van Dijken JP. Effect of benzoic acid on metabolic fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermentation. Yeast 1992;8:501–17. [18] Warth AD. Mechanism of action of benzoic acid on Zygosaccharomyces bailii: effects on glycolytic metabolite levels, energy production, and intracellular pH. Appl Environ Microbiol 1991;57:3410–4. [19] Warth AD. Effect of benzoic acid on glycolytic metabolite levels and intracellular pH in Saccharomyces cerevisiae. Appl Environ Microbiol 1991;57:3415–7. [20] Francois J, Schaftingen EV, Hers HG. Effect of benzoate on the metabolism of fructose 2,6-bisphosphate in yeast. Eur J Biochem 1986;154:141–5. [21] Krebs HA, Wiggins D, Stubbs M, Sols A, Bedoya F. Studies on the mechanism of the antifungal action of benzoate. Biochem J 1983;214:657–63. [22] Tochampa W, Sirisansaneeyakul S, Vanichsriratana W, Srinophakun P, Bakker HH, Chisti Y. A model of xylitol production by the yeast Candida mogii. Bioproc Biosyst Eng 2005;28:175–83. [23] Deschatelets L, Yu EKC. A simple pentose assay for biomass conversion studies. Appl Microbiol Biotechnol 1986;24:379–97. [24] Adler L, Gustafsson L. Polyhydric alcohol production and intracellular amino acid pool in relation to halotolerance of the yeast Debaromyces hansenii. Arch Microbiol 1980;124:123–30. [25] Hallsworth JE. Ethanol-induced water stress in yeast. J Ferment Bioeng 1998;85:125–37. [26] Hallsworth JE, Nomura Y, Iwahara M. Ethanol-induced water stress and fungal growth. J Ferment Bioeng 1998;86:451–6.

743

[27] Sirisansaneeyakul S, Tochampa W, Bashir I, Rizzi M, Bhuwapathanapun S. Kinetic modeling of pH affecting xylitol production by Candida mogii. Thai J Agric Sci 2000;33:159–66. [28] Sirisansaneeyakul S, Tochampa W, Bashir I, Rizzi M, Bhuwapathanapun S. Continuous production of xylitol by cell recycling system. Thai J Agric Sci 2000;33:99–106. [29] Bakker BM, Overkamp KM, van Maris AJ, Ko¨tter P, Luttik MA, van Dijken JP, et al. Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae. FEMS Microbiol Rev 2001;25:15–37. [30] Rigoulet M, Aguilaniu H, Ave´ret N, Bunoust O, Camougrand N, GrandierVazeille X, et al. Organization and regulation of the cytosolic NADH metabolism in the yeast Saccharomyces cerevisiae. Mol Cell Biochem 2004;256:73–81. [31] Granstro¨m T, Leisola M. Controlled transient changes reveal differences in metabolite production in two Candida yeasts. Appl Microbiol Biotechnol 2002;58:511–6. [32] Granstro¨m T, Ojamo H, Leisola M. Chemostat study of xylitol production by Candida guilliermondii. Appl Microbiol Biotechnol 2000;55:36–42. [33] Granstro¨m T, Arisstidou AA, Leisola M. Metabolic flux analysis of Candida tropicalis growing on xylose in an oxygen-limited chemostat. Metab Eng 2002;4:248–56. [34] Tochampa W. Optimal fed-batch control of xylitol production. Thailand: Kasetsart University; 2005 [dissertation]. [35] Aguiar Jr WB, Faria LFF, Couto MAPG, Araujo OQF, Pereira Jr N. Growth model and prediction of oxygen transfer rate for xylitol production from D-xylose by C. guilliermondii. Biochem Eng J 2002;12:49–59. [36] Sheu DC, Duan KJ, Chen YC, Chen CW. Production of xylitol from Candida tropicalis by using an oxidation-reduction potential-stat controlled fermentation. Biotechnol Lett 2003;25:2065–9. [37] Bodaker I, Sharon I, Suzuki MT, Feingersch R, Shmoish M, Andreishcheva E, et al. Comparative community genomics in the Dead Sea: an increasingly extreme environment. ISME J 2010;4:399–407. [38] Dennison C, Lovrient R. Three-phase partitioning: concentration and purification of proteins. Protein Expr Purif 1997;11:149–61. [39] Gaur R, Sharma A, Khare SK, Gupta MN. A novel process for extraction of edible oils Enzyme assisted three phase partitioning (EATPP). Bioresour Technol 2007;98:696–9. ¨ nal S. Purification of a-galactosidase from [40] Sen A, Eryilmaz M, Bayyaktar H, O pepino (Solanum muricatum) by three-phase partitioning. Sep Purif Technol 2011;83:130–6.