Improved Ethanol Production from Xylose by Candida shehatae Induced by Dielectric Barrier Discharge Air Plasma

Plasma Science and Technology, Vol.16, No.6, Jun. 2014 Improved Ethanol Production from Xylose by Candida shehatae Induced by Dielectric Barrier Discharge Air Plasma CHEN Huixia ( ) 1,2, XIU Zhilong ( ) 2, BAI Fengwu ( ) 2 1 School of Fisheries and Life Science, Dalian Ocean University, Dalian 116023, China 2 School of Life Science and Biotechnology, Dalian University of Technology, Dalian 116024, China Abstract Xylose fermentation is essential for ethanol production from lignocellulosic biomass. Exposure of the xylose-fermenting yeast Candida shehatae (C. shehatae) CICC1766 to atmospheric pressure dielectric barrier discharge (DBD) air plasma yields a clone (designated as C81015) with stability, which exhibits a higher ethanol fermentation rate from xylose, giving a maximal enhancement in ethanol production of 36.2% compared to the control (untreated). However, the biomass production of C81015 is lower than that of the control. Analysis of the NADH (nicotinamide adenine dinucleotide)- and NADPH (nicotinamide adenine dinucleotide phosphate)- linked xylose reductases and NAD + -linked xylitol dehydrogenase indicates that their activities are enhanced by 34.1%, 61.5% and 66.3%, respectively, suggesting that the activities of these three enzymes are responsible for improving ethanol fermentation in C81015 with xylose as a substrate. The results of this study show that DBD air plasma could serve as a novel and effective means of generating microbial strains that can better use xylose for ethanol fermentation. Keywords: dielectric barrier discharge air plasma, Candida shehatae, ethanol fermentation, xylose, xylose reductase, xylitol dehydrogenase PACS: 52.77. j DOI: 10.1088/1009-0630/16/6/12 (Some figures may appear in colour only in the online journal) 1 Introduction With the recognition that the global crude oil reserve is finite and the ecological environment is deteriorating as a result of the overconsumption of petroleum-based transportation fuels, ethanol that is both renewable and environmentally friendly has garnered global attention [1]. Currently, ethanol is mainly produced from sugar- and grain-based feedstocks such as sugarcane in Brazil and corn in the United States [2,3]. Sugarcane, a specific crop to temperate and tropical regions, is not universally available, while grains are traditionally produced for food and animal feed. Thus sugarcane and grains are not sustainable for the large scale production of fuel ethanol as the use of these crops for fuel production could endanger the security of the grain supply, particularly in developing countries [4]. Vast quantities of sugars are naturally present as the structural polysaccharides cellulose and hemicellulose in lignocellulosic biomass, like residues of grains and forest products and crops grown specifically for fuel production, which are sustainable for the production of biofuels and bio-based chemicals to alleviate dependence on crude oil and to benefit the rural economy [5 7]. However, the major sugars derived from lignocellulosic biomass are glucose and xylose, and ethanologen species such as Zymomonas mobilis and Saccharomyces cerevisiae (S. cerevisiae) have evolved with glucose but not xylose metabolic pathways [8]. Some yeasts can utilize xylose. To date, the most extensively studied xylose-fermenting yeasts include Candida shehatae, Pachysolen tannophilus and Pichia stipitis. The different ethanol producing abilities of xylose consuming strains are most probably explained by differences in the nature of the initial steps of xylose metabolism of yeasts [9]. In yeast, xylose is first reduced to xylitol by xylose reductase (XR) that uses either nicotinamide adenine dinucleotide (NADH) or nicotinamide adenine dinucleotide phosphate (NADPH) as a coenzyme, but with a preference towards NADPH. Xylitol is then oxidized to xylulose with NAD + by xylitol dehydrogenase (XDH) [10]. The different cofactor specificities would lead to cofactor imbalance. Before entering the pentose phosphate pathway (PPP), xylulose is phosphorylated to xylulose 5-phosphate by xylulokinase (XK). Although yeasts can utilize xylose, these species exhibit very poor ethanol yields, low ethanol tolerance and fermentation rate compared with the glucose-fermenting yeast S. cerevisiae [11], which present many challenges in developing robust strains for supported by National Natural Science Foundation of China (No. 20576018) 602

CHEN Huixia et al.: Improved Ethanol Production from Xylose by Candida shehatae Induced by DBD ethanol production from lignocellulosic biomass. Over the past several years, S. cerevisiae has been the subject of various research efforts to engineer it with the metabolic pathway of xylose-fermenting yeast. Despite successful expression of the three-enzyme encoding genes XR/XDH/XK in S. cerevisiae, the rate of xylose utilization and ethanol yield are still very low due to the imbalance of co-factors [11]. To date, various sources of mutagens, such as X-rays, gamma-rays, ultraviolet rays, laser, neutron and ion implantation have been developed and used for breeding mutants [12 17], and efforts are still being invested in developing new sources of mutagens to improve the mutational spectrum and increase mutation rate. Atmospheric pressure dielectric barrier discharge (DBD) air plasma, one of the electrical discharge plasma technologies, generates a mixture of reactive species, UV radiation, energetic ions and charged particles [18,19]. It has been exploited in sterilization to effectively kill microorganismsv [20,21]. DBD plasma has the advantages of easy operation, safety, low cost and wide application. A plasma jet was successfully employed to generate mutations in Streptomyces avermitilis [22]. However, the use of DBD air plasma as a method to induce mutants is rarely reported. In our previous work, it was found that DBD air plasma can change the metabolic pathway of Klebsiella pneumoniae as determined from a mixed population of cells, which contained both damaged and undamaged cells after DBD air plasma treatment under certain conditions, leading to an enhanced production of 1,3- Propanediol [23]. Isolation of a single cell with a stable functional change will be vital to verifying the mutagenic effect of DBD air plasma. In this study, DBD air plasma was used to induce mutation with a stable metabolic change in the yeast C. shehatae, resulting in improved ethanol production from xylose. 2 Materials and methods 2.1 Microorganism, media, culture C. shehatae CICC 1766, which was obtained from the China Center of Industrial Culture Collection, was used as the wild-type strain. It was routinely maintained in a basic medium, which consists of the followings (g L 1 ): yeast exacts 3, peptone 3, D-xylose 20, with/without agar 20. The fermentation medium used to evaluate the ethanol fermentation performance of the mutant(s) was composed of 50 g L 1 xylose, 100 g L 1 glucose or 30 g L 1 xylose + 55 g L 1 glucose, supplemented with 3.75 g L 1 each of yeast extract and peptone. All media were sterilized at 115 o C for 20 min. 2.2 DBD plasma exposure system and operation The DBD plasma system is schematically shown in Fig. 1. Two quartz dielectric barriers were connected to the electrodes with a diameter of 6 cm. The voltage and its frequency were set at 12 kv and 20 khz, respectively. The corona discharge was generated in the air between the two quartz dielectric barriers with a clearance of 4 mm. Fig.1 Schematic representation of the DBD air plasma apparatus C. shehatae was incubated in a basic medium to exponential phase (cultured for 15 h) at 30 o C and 150 rpm. The culture was then kept at 4 o C for 2 h [24], followed by centrifugation at 4000 g for 5 min to collect the cells. The cells were re-suspended in sterilized de-ionized water to a density of 4 10 7 /ml, and 1 ml of this cell suspension was spread onto the lower quartz dielectric barrier and exposed to DBD air plasma. Multiple samples were treated for different time periods ranging from 10 s to 50 s, with each treatment carried out in triplicate. 2.3 Survival rate determination DBD air plasma-treated and control samples were collected and diluted 10,000-fold for the plate count analysis, whereby 0.1 ml aliquots were spread onto the plates and incubated at 30 o C for 48 h. Colonies appearing on the plates were counted and the average number per plate was calculated. The survival rates for the DBD air plasma-treated cells for different exposure times were expressed as the percentage of emerging colonies relative to those of the control. 2.4 Screening of C. shehatae for enhanced ethanol production Since the intracellular dehydrogenase of live cells can reduce colorless 2,3,5-triphenyl tetrazolium chloride (TTC) to red. The activity of the dehydrogenase is proportional to the color intensity [25], which in turn correlates with ethanol production ability. TTC was thus used as an indicator for the preliminary screening to identify clones with enhanced ethanol production abilities. Samples were spread onto plates containing a basic agar medium and incubated at 30 o C for 48 h for the viable cells to grow. The plates were then overlaid with a TTC medium consisting of xylose (10 g L 1 ), agar (15 g L 1 ) and TTC (0.05 g L 1 ), and 603

Plasma Science and Technology, Vol.16, No.6, Jun. 2014 incubated for 3 h to enable the reduction of TTC to occur. Clones that showed a more intense red color on the plate were selected and cultured in flasks containing a xylose medium for 36 h to measure their ethanol production. The clones that showed an improvement in ethanol production of 5% or more over the control were regarded as positive clones. The stabilities of the enhanced ethanol production trait in these positive clones were evaluated by 15 consecutive subculturings that alternated between TTC plate and flask cultures. The clone that showed no deterioration in color and ethanol production was selected for further study by ethanol fermentation and enzyme activity assays. glucose was determined by using a glucose analysis instrument (SBA-50B bio-sensor, China). All the experiments were performed in triplicate. The t-test (onesided) was applied to evaluate the statistical significant differences between clones derived from plasma-treated cells and control. Probability value (P ) was calculated to express significant differences, and * means P < 0.05, ** means P < 0.01 and *** means P < 0.001. 2.5 The survival rate of C. shehatae decreased with prolonged exposure to DBD air plasma, with no viable cells remaining after an exposure of 40 s. The survival rates were about 92%, 68% and 8% for 10 s, 20 s and 30 s, respectively. Cells exposed for 20 s and 30 s were screened for enhanced ethanol production with respect to the control as detected by the TTC plate method. Hundreds of clones were screened by the TTC plate. The clones that showed a more intense red color on the TTC plate were isolated and further validated by their enhanced ethanol production in flask cultures. After 15 consecutive subcultures, these two clones, designated as C80828 and C81015, were found to be stable. Both clones showed increased ethanol production over the control, with C81015 yielding the highest production of 27.1% more than the control in the basic culture. Fig. 2 illustrates the color change of some clones derived from DBD air plasma-treated cells and from control cells when the TTC medium was overlaid for 10 min. C81015 changed its color faster and more obviously than others and showed the most intense red color (indicated by arrow), a feature which correlated well with its highest increase in ethanol production ability. C81015 was therefore chosen for further analysis by NADH- and NADPH-linked XR and NAD+ -linked XDH activity assays and by fermentation with either xylose, glucose or a mixture of glucose and xylose as substrates. 3 3.1 Assays of intracellular xylose reductase and xylitol dehydrogenase activities Cells were incubated in a basic medium for 36 h and then harvested by centrifugation at 4000 g for 5 min followed by washing (three times) in phosphate buffer saline (PBS, ph=7.4). Cell lysis was performed according to Ref. [26]. In brief, equal numbers of cells from the control of the selected positive clone were resuspended in sterile PBS (ph=7.4). Equal volumes of 0.5 mm glass beads were added to the cell suspensions and vortexed at maximum speed for 10 40 s with 20 s cooling in between. Cellular debris and glass beads were removed by centrifugation at 8000 g and 4 o C for 20 min. The supernatants were collected and kept at 70 o C for enzyme activity assays. The activities of the NADH- and NADPH-linking XR and NAD+ -linking XDH were measured by using a spectrophotometer (JASCO V-560, Japan) at 30 o C according to the protocol reported by Verduyn et al. [27] and Neuhauser et al [28]. One unit of enzyme activity was defined as the amount of enzyme that consumed or produced 1 µmol of NAD(P)H per min in the reaction. The protein concentration in the extract was determined by Bradford assay [29]. 2.6 Ethanol fermentation performance Results Identification of C. shehatae with improved ethanol production Cells were grown to the exponential phase and inoculated into flasks containing 100 ml different fermentation media with an inoculum of 5%. Fermentation was carried out at 30 o C and 150 rpm for 132 h, and samples were withdrawn at an interval of 12 h for biomass, ethanol and residual sugar concentration determinations. 2.7 Analysis method Biomass was determined in dry cell weight (DCW) after the samples were dried at 80 o C for 8 h. Ethanol was determined by gas chromatography (SHIMADZU GC-14B, Japan) with an FID detector and n-butanol as an internal standard [30]. Xylose was analyzed by utilizing the 3, 5-dinitrosalicylic acid method [31], while Fig.2 Changes in the functional characteristic of C. shehatae as detected by the TTC plate. The clone corresponding to C81015 is indicated by an arrow 604

CHEN Huixia et al.: Improved Ethanol Production from Xylose by Candida shehatae Induced by DBD 3.2 Activities of XR and XDH Significant improvements in the specific activities of NADH- and NADPH-linked XR and NAD+ -linked XDH were observed in the cell extract of C81015 compared to control, representing an increase of 33.9% (P < 0.01), 61.5% (P < 0.05) and 66.5% (P < 0.01) for NADH- and NADPH-linked XR and NAD+ -linked XDH, respectively (Fig. 3). The increased levels of these enzyme activities correlated well with the improved ethanol production observed in the flask cultures. Fig.3 Comparison of XR and XDH activities between C81015 and control C. shehatae, * means P > 0.05 and ** means P < 0.01 3.3 Ethanol fermentation Fig.4 Fermentation of xylose (a), glucose (b), and xyloseglucose mixture (c) by C81015 and control C. shehatae plate. Shaded symbols, controls; open symbols, C81015. N and : residual xylose, H and : residual glucose, and : ethanol concentration, and : biomass. * means P < 0.05, ** means P < 0.01, and *** means P < 0.001 Fig. 4(a) illustrates the ethanol fermentation profiles of C81015 with a xylose medium. The ethanol fermentation rate of C81015 was improved significantly from 24 h compared to the control (P < 0.05). While the highest ethanol concentration achieved by the control at 108 h was only 7.39 g L 1, the ethanol concentration achieved by C81015 was 9.71 g L 1, and continued to increase slightly to 9.84 g L 1 at 120 h, representing a total increase in ethanol production of 36.2% (P < 0.001) over the control. The highest molar yield achieved by the control was 0.54 and that achieved by C81015 was 0.71. The ethanol yield per cell was also calculated. As shown in Fig. 5, the ethanol yield per cell of C81015 was higher than that of the control during the whole fermentation time (P < 0.05). It increased during the first 100 h and got the highest value of 1.43 for C81015. However, its highest value achieved by the control was only 0.87. The profiles of xylose consumption were similar for both C81015 and the control, but the biomass production of C81015 was significantly lower from 48 h, reaching 6.79 g(dcw)/l at 108 h compared with 9.20 g(dcw)/l for the control. The fermentation was repeated five times, and the ethanol production of C81015 was significantly higher than that of the control. Fig.5 Comparison of ethanol yield per cell between C81015 and control C. shehatae. : C81015, : control. * means P < 0.05, ** means P < 0.01 and *** means P < 0.001 605

Plasma Science and Technology, Vol.16, No.6, Jun. 2014 Ethanol fermentations with glucose and a mixture of glucose and xylose were further investigated for C81015, and the results are illustrated in Fig. 4(b) and Fig. 4(c), respectively. Compared to the control, C81015 yielded a higher biomass concentration and slightly slower substrate consumption during the fermentation of glucose. However, there was no significant difference in ethanol production between C81015 and the control. As for the ethanol fermentation with mixed sugars, the ethanol production profiles were almost the same for C81015 and the control during the first 50 h when glucose was consumed, but the ethanol production by C81015 proceeded at a faster rate than the control from 60 h to 120 h, when xylose was being consumed. The maximum ethanol production by C81015 was significantly higher (P < 0.05) than that attained by the control. 4 Discussion The results presented in this paper point to a potential usage of DBD air plasma as a method to improve the ability of C. shehatae to produce ethanol from xylose. As reported previously, different reactive oxygen species (ROS) such as OH, O, O 2, H 2O 2, and O 3 can be generated by DBD air plasma [18,32], and ROS are considered to be the most important agents contributing to the killing effect of nonequilibrium plasmas, with minor contributions from heat, charged particles, electric field, and UV radiation during plasma discharges at atmospheric pressure [33]. In our previous work, it was found that S. cerevisiae intracellular ROS was induced to increase significantly by DBD air plasma in atmospheric pressure treatment [34]. Since it is well documented that ROS can accelerate the mutation of DNA [35 37], which might be responsible for inducing characteristic changes, possibly including mutations, in DBD plasma-treated C. shehatae. In this study, a clone of C. shehatae CICC 1766 with a stably enhanced ethanol production trait was selected after screening a large sample of clones derived from cells treated with DBD plasma. Compared to the control, C81015 could ferment xylose to ethanol more effectively, achieve higher ethanol production as the fermentation proceeded, and reach a maximum of 36.2% at the end of fermentation. Unexpectedly, less biomass was accumulated by C81015 during the ethanol fermentation that used xylose, which together with the higher ethanol production contributed to the significant improvement in the conversion rate of xylose to ethanol (P < 0.001). The enhanced ethanol production ability of C81015 was maintained in repeated experiments, which suggests that DBD air plasma could be an effective method to obtain variants of C. shehatae with stable characteristic changes. It is well known that C. shehata is able to consume glucose and xylose to produce ethanol [38]. When both xylose and glucose exist as substrates, it will consume glucose first because of the competitive inhibition of xylose transport by glucose [39]. An increase in ethanol production by C81015 was observed only when xylose and not glucose was the sole carbon source. The use of mixed sugars for fermentation (Fig. 4(c)) also supports this notion. The xylose consumption rate increases only after the glucose concentration has decreased to around 10 g L 1. C. shehatae consumed glucose mainly during the first 50 h, so there was no significant increase in ethanol production by C81015 during this period. It was only when the glucose was depleted after 50 h that ethanol production began to increase at a significant rate relative to the control as a result of xylose consumption. Based on the metabolic network of recombinant S. cerevisiae [40,41], we deduced the metabolic network of C. shehatae on xylose and a glucose substrate (Fig. 6). According to Fig. 5, it was indicated that this metabolic change might exist in the step of xylose to xylulose conversion in this study. In xylose-fermenting yeasts, xylose is first reduced to xylitol by XR that uses either NADH or NADPH as a co-factor, but with a preference towards NADPH in the case of C. shehatae [42]. When xylose is the only carbon source, NADPH is consumed in the PPP to support the growth of C. shehatae cells. Since the activity of the NADPH-linked XR was enhanced in C81015, more NADPH was consumed in the pathway of xylose to xylitol, and xylitol was further converted to xylulose via the enhanced NAD + -linked XDH. Thus ethanol production was accelerated, but growth of the C. shehatae was compromised because of the competitiveness for NADPH between the two metabolic pathways. When glucose was presented in the medium, the production of NADPH from the conversion of glucose-6-phosphate to ribulose-5-phosphate in PPP was more than the consumption of reduced cofactors in xylose reduction, therefore allowing excess NADPH to be used for biosynthesis. As far as the conversion of xylose to ethanol is concerned, at least, these results indicated that change in the activities of both NADH- and NADPH-dependent XR and NAD + - linked XDH was an important factor induced by DBD air plasma to bring about an enhancement in ethanol production from xylose in C. shehatae. The analysis of the two key enzymes supported this assumption, as both of them were significantly enhanced in C81015. Fig.6 The possible model of xylose and glucose metabolism in C. shehatae 606

CHEN Huixia et al.: Improved Ethanol Production from Xylose by Candida shehatae Induced by DBD In conclusion, the present study reports for the first time the mutation effect of DBD air plasma on yeast cells. This study suggested that DBD air plasma could be used as a novel and effective method to develop C. shehatae strains with improved xylose-fermenting ability. However, despite its potential applications, further studies should be conducted to examine the specific change of XR and XDH at the DNA level to determine whether the enhanced enzyme activities were due to altered enzymes or increased expression levels of the unmodified enzymes via changes in the regulatory elements of the genes coding for these two enzymes. In addition, the production of the ethanol was still low and more mutation work needs to be done. Acknowledgments The authors thank Dr. Alan K Chang for his assistance in the revision of the manuscript. References 1 Farrell A E, Plevin R J, Turner B T, et al. 2006, Science, 311: 506 2 Bai F W, Anderson W A, Moo-Young M. 2008, Biotechnol. Adv., 26: 89 3 Balat M, Balat H. 2009, Appl. Energ., 86: 2273 4 Makenete A, Lemmer W, Kupka J. 2008, Int. Food Agribusiness Manage. Rev., 11: 101 5 Lynd L R, Cushman J H, Nichols R J, et al. 1991, Science, 251: 1318 6 Dodds D R, Gross R A. 2007, Science, 318: 1250 7 Jordan N, Boody G, Broussard W, et al. 2007, Science, 316: 1570 8 Aristidou A, Penttilä M. 2000, Curr. Opin. Biotech., 11: 187 9 Lee W J, Ryu Y W, Seo J H. 2000, Process Biochem., 35: 1199 10 Bruinenberg P M, Bot H P M, Dijken J P, et al. 1984, Appl. Microbiol. Biot., 19: 256 11 Chu B C H, Lee H. 2007, Biotechnol. Adv., 25: 425 12 Gregory W C. 1955, Argon. J., 47: 396 13 Kayhart M. 1956, Radiat. Res., 4: 65 14 Predieri S, Magli M, Zimmerman R H. 1997, Euphytica, 93: 227 15 Mazza C A, Battista D, Zima A M, et al. 1999, Plant Cell Environ., 22: 61 16 Chen Y P, Yue M, Wang X L. 2005, Plant Sci., 168: 601 17 Gu S B, Yao J M, Yuan Q P, et al. 2006, Appl. Microbiol. Biot., 72: 456 18 Laroussi M, Richardsonand J P, Dobbs F C. 2002, Appl. Phys. Lett., 81: 772 19 Gaunt L F, Beggs C B, Georghiou G E. 2006, IEEE Trans. Plasma Sci., 34: 1257 20 Yu H, Xiu Z L, Ren C S, et al. 2005, IEEE Trans. Plasma Sci., 33: 1405 21 Wang C H, Wu Y, Li G F. 2008, J. Electrostat., 66: 71 22 Wang L Y, Huang Z L, Li G, et al. 2010, J. Appl. Microbiol., 108: 851 23 Dong X Y, Xiu Z L, Hou Y M, et al. 2009, IEEE Trans. Plasma Sci., 37: 920 24 Carratore R D, Croce C D, Simili M, et al. 2002, Mut. Res.-Gen. Tox. En., 513: 183 25 Ghaly A E, Ben-Hassan R M. 1993, Appl. Biochem. Biotech., 43: 77 26 Favre C, Aguilar P S, Carrillo M C. 2008, Free. Radical. Bio. Med., 45: 1446 27 Verduyn C, Van Kleef R, Frank J, et al. 1985, Biochem. J., 226: 669 28 Neuhauser W, Haltrich D, Kulbe K D, et al. 1997, Biochem. J., 326: 683 29 Bradford M M. 1976, Anal. Biochem., 72: 248 30 Sun L H, Jiang B, Xiu Z L. 2009, Biotechnol. Lett., 31: 371 31 Miller G L. 1959, Anal. Chem., 31: 426 32 Moreau M, Orange N, Feuilloley M G J. 2008, Biotechnol. Adv., 26: 610 33 Deng X, Shi J, Kong M G. 2006, IEEE Trans. Plasma Sci., 34: 1310 34 Chen H X, Bai F W, Xiu Z L. 2010, IEEE Trans. Plasma. Sci., 38: 1885 35 Møller P, Wallin H. 1998, Mutat. Res.-Rev. Mutat., 410: 271 36 Díaz-Llera S, Podlutsky A, Österholm A M, et al. 2000, Mut. Res.-Gen. Tox. En., 469: 51 37 Ralser M, Wamelink M M, Kowald A, et al. 2007, J. Biol., 6: 10 38 Lebeau T, Jouenne T, Junter G A. 1997, Enzyme Microb. Tech., 21: 265 39 Meinander N Q, Hahn-Hägerdal B. 1997, Appl. Environ. Microb., 63: 1959 40 Meinander N Q, Boels I, Hahn-Hägerdal B. 1999, Bioresour. Technol., 68: 79 41 Grotkjær T, Christakopoulos P, Nielsen J, et al. 2005, Metab. Eng., 7: 437 42 Gírio F M, Peito M A, Amaral-Collaço M T. 1989, Appl. Microbiol. Biot., 32: 199 (Manuscript received 25 March 2013) (Manuscript accepted 30 July 2013) E-mail address of corresponding author XIU Zhilong: zhlxiu@dlut.edu.cn 607