Effects of lignin-derived phenolic compounds on xylitol production and key enzyme activities by a xylose utilizing yeast Candida athensensis SB18

Bioresource Technology 121 (2012) 369–378

Contents lists available at SciVerse ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Effects of lignin-derived phenolic compounds on xylitol production and key enzyme activities by a xylose utilizing yeast Candida athensensis SB18 Jinming Zhang a,b, Anli Geng b,⇑, Chuanyi Yao a, Yinghua Lu a,c, Qingbiao Li a,c a

Department of Chemical and Biochemical Engineering, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, PR China School of Life Sciences and Chemical Technology, Ngee Ann Polytechnic, Singapore c The Key Lab for Chemical Biology of Fujian Province, Xiamen 361005, PR China b

h i g h l i g h t s " Xylitol producer Candida athensensis SB18 was tested for phenolic compound inhibition. " The inhibition follows phenol > syringaldehyde > 4-hydroxylbenzaldehyde > vanillin. " Inhibition was insignificant when the total inhibitor content was below 1.0 g/L. " The inhibitors affected more xylose reductase than xylitol dehydrogenase activity. " The inhibitory effects strongly correlated to their in vivo assimilation.

a r t i c l e

i n f o

Article history: Received 29 March 2012 Received in revised form 3 July 2012 Accepted 5 July 2012 Available online 14 July 2012 Keywords: Candida athensensis SB18 Phenolic compound Inhibitors Xylose reductase Xylitol dehydrogenase

a b s t r a c t Candida athensensis SB18 is potential xylitol producing yeast isolated in Singapore. It has excellent xylose tolerance and is able to produce xylitol in high titer and yield. However, by-products, such as phenolic compounds, derived in lignocellulosic biomass hydrolysate might negatively influence the performance of this strain for xylitol production. In this work, four potential phenolic inhibitors, such as vanillin, syringaldehyde, 4-hydroxybenzaldehyde and phenol, were evaluated for their inhibitory effects on xylitol production by C. athensensis SB18. Phenol was shown to be the most toxic molecule on this microorganism followed by syringaldehyde. Vanillin and 4-hydroxylbenzaldehyde was less toxic than phenol and syringaldehyde, with vanillin being the least toxic. Inhibition was insignificant when the total content of inhibitors was below 1.0 g/L. The presence of phenolic compounds affected the activity of xylose reductase, however not on that of xylitol dehydrogenase. C. athensensis SB18 is therefore a potential xylitol producer from hemicellulosic hydrolysate due to its assimilation of such phenolic inhibitors. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction As a natural five carbon sugar alcohol with sweetness, xylitol has been increasingly used in food and pharmaceutical industries due to several advantages. Recently, the demand for xylitol in food industries as an alternative sweetener has created a strong market for the development of low-cost xylitol production processes. With the increased interests in alternative energy and biochemical sources, lignocellulosic materials are becoming attractive as a potential low-cost feedstock for the production of biofuels and value-added chemicals such as bioethanol and xylitol. Lignocellu-

⇑ Corresponding author. Address: School of Life Sciences and Chemical Technology, Ngee Ann Polytechnic, 535 Clementi Road, Singapore 599489, Singapore. Tel.: +(65) 64608617; fax: +(65) 64679109. E-mail address: [email protected] (A. Geng). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.07.020

losic materials contain polysaccharides such as celluloses and hemicelluloses. The hydrolysate of the hemicellulosic fraction containing mainly D-xylose can be used as the substrate for xylitol production by chemical or biotechnological means (Nigam and Singh, 1995). Currently, the most promising path for xylitol production from xylose rich hemicellulosic feedstock is microbial fermentation (Converti et al., 2000; Carvalho et al., 2003; Rodrigues et al., 2006). Hemicellulosic hydrolysate can be obtained by chemical pretreatment of the lignocellulosic biomass. Organosolv and dilute acid pretreatment methods are among the most commonly used and effective pretreatment means (Geng et al., 2003; Villarreal et al., 2006). Such methods usually involve the operation under high temperature and pressure. Under such conditions, several by-products are derived from sugars and lignin; they are liberated in the hemicellulosic hydrolysate. These compounds include three major groups: (1) furan derivatives (furfural and 5-hydroxymethyl

370

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

furfural (HMF)); (2) weak acids (mainly acetic acid, formic acid and levulinic acid); (3) phenolic compounds such as vanillin, syringaldehyde, 4-hydroxybenzaldehyde, and phenol (Buchert et al., 1990; Almeida et al., 2007). As reported, furan derivatives are released from the further degradation of pentose (furfural) and hexose (HMF), while weak acids are generated when furfural and HMF are further degraded (Ulbricht et al., 1984), and phenolic compounds on the other hand are produced from the partial breakdown of lignin (Sears et al., 1971). The above by-products derived from sugars and lignin negatively affect the fermentation efficiency due to their toxic influence on the fermentative microorganisms (Mussatto and Roberto, 2004). The type and concentration of the compounds existing in the hydrolysates depend on the type of raw materials and the pretreatment conditions. Their inhibitory effects vary with their concentration and the microorganisms (Martin and Jönsson, 2003). In order to improve the fermentative efficiency and to enhance the metabolic activity of the microorganisms, a greater understanding of the inhibitory mechanisms of the individual toxic compounds, their interactive effects, and the in vivo degradation possibility of these compounds is required. Among these toxic compounds liberated during the pretreatment process, weak acids were proven to inhibit cell growth of microorganisms due to the inflow of undissociated acid into the cytosol (Stouthamer, 1979; Verduyn et al., 1992; Russell, 1992). Furan derivatives were shown to reduce the specific growth rate and the cell biomass yield, and they can be converted to their corresponding alcohols (Palmqvist et al., 1999; Palmqvist and Hägerdal, 2000). While phenolic compounds have been suggested to exert a considerable inhibition in the fermentation process due to its ability to affect cell membranes to serve as selective barriers and enzyme matrices (Heipieper et al., 1994), their inhibition mechanisms on microorganisms have not yet been completely elucidated, largely due to the heterogeneity of the group and the lack of accurate qualitative analyses. 4-hydroybenzoic acid, vanillin, and catechol were identified as the major constituents in the willow hemicellulose hydrolysate by Jönsson et al. (1998) and their toxicity can be reduced by laccase treatment. Vanillin constitutes a major fraction of the phenolic monomers in the hemicellulosic hydrolysate of softwood, such as spruce, pine and willow (Palmqvist and Hägerdal, 2000) and its presence (1 g/L) decreased the ethanol yield by 25% (Ando et al., 1986). A variety of aromatic monomeric compounds such as vanillin, syringaldehyde, 4-hydroxybenzaldehyde and 5-hydroxylmethylfurfural were identified to be present in the hemicellulosic hydrolysate of birchwood (Buchert et al., 1990) and the total concentration of such phenolics could reach about 2 g/L (Clark and Mackie, 1984). The presence of such phenolic compounds was shown to inhibit yeast growth and therefore reduce xylitol production to different degrees depending on their concentration (Kelly et al., 2008; Cortez and Roberto, 2010b). Candida athensensis SB18 is potential xylose-producing yeast isolated in Singapore with excellent xylose tolerance and xylitol production capabilities. In order to investigate the potential of this strain in converting of hemicellulosic sugars to xylitol, the present work evaluates the inhibitory effects of the most often detected phenolic by-products, such as vanillin, syringaldehyde, 4-hydroxybenzaldehyde and phenol on the fermentative behavior and enzyme activities of the new isolate C. athensensis SB18. The effects of both individual and combination of these inhibitors on cell growth, xylitol production and the key enzymes activities are studied. In addition, phenolic compound degradation is performed in order to better understand of the inhibitory mechanism of such phenolic compounds. Such knowledge will be useful for the development and optimization of a lignocellulosic biomass pretreatment method and for the application of C. athensensis SB18 in industrial xylitol production.

2. Methods 2.1. Microorganism and inoculum cultivation The xylitol producing yeast C. athensensis SB18 was isolated in Singapore soil samples (Zhang et al., 2012) and was maintained on YPX-agar plates containing (g/L): xylose (Merck, Germany), 20; yeast extract (Merck, Germany), 10; peptone (Difco, USA); and agar (Merck, Germany), 20. Inoculum was prepared by transferring the cells from the 24 h YPX agar plates to 250-mL shaking flasks containing 100 mL seed culture medium consisting of (g/L): xylose, 50; yeast extract, 10; and peptone, 20. Flasks were incubated at 150 rpm and 30 °C for 24 h. Cells were harvested by centrifugation at 8000g and 4 °C for 4 min. They were then washed twice using sterile water and stored at 4 °C until use as the inoculum. 2.2. Medium and fermentation condition Experiments were carried out in 100-mL Erlenmeyer’s flasks containing 40 mL of the fermentation medium composed of (g/L): xylose, 50; yeast nitrogen base (YNB), 6.76; yeast extract, 1.0; and urea, 2.0. All the nutrients were prepared separately and were autoclaved at 121 °C for 15 min. YNB and urea were sterilized through 0.2 lm filter. Flasks were inoculated with an inoculum size of 1.5 g/L (unless otherwise stated) and incubated at 30 °C for 4–7 days. Fermentation experiments were conducted in the presence of different inhibitors at varied concentrations. Samples were withdrawn periodically to determine the cell density at 600 nm (OD600) and the concentration of residual substrates and products. All experiments were conducted in duplicate. 2.3. Inhibitor cocktail The stock solution of all the inhibitors including syringaldehyde (Syr), 4-hydroxybenzaldehyde (Hba), vanillin (Van) and phenol (Phe), was prepared by dissolving the desired mass of each phenolic compound in the sterile fermentation medium. They were then re-sterilized through 0.2 lm filter. The initial inhibitor concentration in the stock solution was (g/L): Syr, 2.0; Van, 2.0; Hba, 1.0; and Phe, 1.0, individually. The inhibitor cocktail were freshly prepared by mixing the individual inhibitor stock solution to the desired concentration in the fermentation medium. 2.4. Crude enzyme extraction Samples were withdrawn periodically during fermentation process and cells were harvested by centrifugation at 8000g and 4 °C for 4 min. Cell pellets were washed twice with cold sterile distilled water and re-suspended with 0.25 M potassium phosphate buffer (pH 7.0) supplemented with 0.1 M 2-mercaptoethanol. The final cell suspension (about 3 g/L) was mechanically disrupted in 2 mL eppendorf tubes under vortex mixing using glass beads with a particle size of 0.5 mm in a volumetric ratio of 1:1 (1 mL of cell suspension to 1 mL of glass beads). Cell disruption was conducted for 5 min with 20 s of disruption followed by 30 s intervals on ice (Gurpilhares et al., 2006). The resulted solution was then centrifuged at 8000g and 4 °C for 6 min and the supernatants were assayed for the activities of xylose reductase (XR) and xylitol dehydrogenase (XDH). 2.5. Enzyme activity assay Xylose reductase (XR) and xylitol dehydrogenase (XDH) activities were determined by measuring the absorbance at 340 nm

371

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378 12

Blank 0.05 g/L 0.10 g/L 0.50 g/L 1.00 gL

8

4

6 4

2

2

0

0 0

10

Blank 0.10 g/L 0.50 g/L 1.00 g/L 2.00 gL

8

6

-10

B: Vanillin

10

Biomass (g/L)

10

Biomass (g/L)

12

A: 4-Hydroxybenzaldehyde

20

30

40

50

60

70

-10

0

10

C: Phenol

12

10

Blank 0.05 g/L 0.10 g/L 0.50 g/L 1.00 gL

6

4

D: Syringaldehyde

10

Blank 0.10 g/L 0.50 g/L 1.00 g/L 2.00 gL

0

0 10

20 30 40 50 Incubation time (h)

50

60

70

4 2

0

40

6

2

-10

30

12

8 Biomass (g/L)

Biomass (g/L)

8

20

Incubation time (h)

Incubation time (h)

60

70

-10

0

10

20

30 40 50 Incubation time (h)

60

70

Fig. 1. Effects of individual inhibitor on the growth of Candida athensensis SB18.

Table 1 Effect of different inhibitors on cell growth (at 57 h) and xylitol production (at 45 h). Inhibitor

Inhibitor (g/L)

Biomass X (g/L)

Inhibition (%)

Specific growth rate lX (1/h)

Xylitol P (g/L)

Yield YP/S (g/g)

Xylitol productivity QP (g/L/h)

Hba

0 0.05 0.1 0.5 1 0 0.1 0.5 1 2

10.32 9.83 9.98 8.95 8.66 10.32 10.40 9.80 8.89 0.04

0.00 4.80 3.29 13.24 16.11 0.00 0.79 5.01 13.81 99.63

0.036 0.033 0.033 0.038 0.041 0.036 0.035 0.040 0.045 0.000

30.05 29.93 31.11 30.03 30.17 30.05 30.90 31.52 32.92 0.00

0.60 0.60 0.62 0.60 0.60 0.60 0.62 0.63 0.67 0

0.67 0.67 0.69 0.67 0.67 0.67 0.69 0.70 0.73 0

Phe

0 0.05 0.1 0.5 1

11.08 10.22 9.18 7.45 3.69

0.00 7.80 17.13 32.73 66.73

0.038 0.034 0.031 0.026 0.012

31.34 32.07 32.12 30.68 15.55

0.63 0.64 0.64 0.61 0.48

0.70 0.71 0.71 0.68 0.35

Syr

0 0.1 0.5 1 2

11.08 8.93 7.54 6.48 4.61

0.00 19.40 32.00 41.55 58.40

0.023 0.023 0.020 0.018 0.022

31.34 31.75 30.91 27.68 19.66

0.63 0.64 0.64 0.61 0.51

0.70 0.71 0.69 0.62 0.44

Van

Note: Hba: 4-hydroxylbenzaldehyde; Van: vanillin; Phe: phenol; Syr: syringaldehyde.

and 30 °C using a spectrophotometer (UVmini-1240, Shimadzu, Japan). Enzyme activity assay protocols from earlier reports were slightly modified (Chiang and Knight, 1966; Smiley and Bolen, 1982). For each XR assay, the 1.5 mL cuvette contained 100 lL de-ionized water, 0.5 mL potassium phosphate buffer (250 mM, pH7.0), 100 lL 2-mercaptoehanol (100 mM), 100 lL 3.4 mM nicotinamide adenine dinucleotide phosphate (NADPH) (Sigma), and 100 lL xylose (0.5 M, Merck, Germany). XR activity was then recorded after the addition of 100 lL crude enzyme extract. The

oxidation of NADPH to NADP+ was measured by following the decrease in absorbance at 340 nm (A340). One unit (U) of XR activity was defined as the amount of the enzyme that catalyzes the oxidation of 1 lmol of NADPH per minute. For XDH activity assay, the procedure was the same as described above except that the cofactor NADPH was replaced by 3.4 mM nicotinamide adenine dinucleotide (NAD+) (Sigma) and the substrate xylose was replaced with 0.5 M xylitol (Applichem, USA). One unit (U) of XDH was defined as the amount of the enzyme that catalyzes the reduction of

372

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378 12

B Blank Van + Hba Hba + Syr Syr + Van Syr + Phe Van + Phe Hba + Phe

10

Biomass (g/L)

8 6

12

8

4

6 4

2

2

0

0 -10

0

10

20

Blank Syr + Hba + Van Syr + Hba + Phe Hba + Van + Phe All

10

Biomass (g/L)

A

30

40

50

60

70

80

-10

0

Incubation time (h)

10

20

30

40

50

60

70

80

Incubation time (h)

Fig. 2. Combined effects of inhibitors on the growth of Candida athensensis SB18. The inhibitor cocktail contained (g/L): 4-hydroxylbenzaldehyde 0.5; vanillin, 0.5; phenol, 0.25 and syringaldehyde, 0.5.

Table 2 Combined effects of inhibitors. Inhibitors

Biomass X (g/L)

Inhibition (%)

Specific growth rate (1/h)

Xylitol (g/L)

Xylitol yield YP/X (g/g)

Xylitol productivity QP (g/L/h)

Time (h)

Blank VHa HSa SVa SPa VPa HPa SHVb SHPb HVPb ALLc

10.82 8.71 7.45 7.69 8.92 8.47 8.14 6.07 5.41 6.85 4.64

0.00 19.47 31.08 28.89 17.55 21.72 24.73 43.90 49.95 36.66 57.14

0.031 0.044 0.033 0.037 0.033 0.027 0.030 0.055 0.038 0.055 0.080

29.57 29.76 28.32 28.42 29.31 32.36 29.60 28.02 26.91 29.24 24.82

0.59 0.60 0.58 0.59 0.59 0.65 0.64 0.56 0.57 0.59 0.53

0.64 0.65 0.62 0.62 0.64 0.70 0.87 0.59 0.59 0.64 0.43

46 46 46 46 46 46 34 46 46 46 58

Note: In all the mixtures, the initial concentration of each inhibitor were constant as follows (g/L): 4-hydroxylbenzaldehyde 0.5; vanillin, 0.5; phenol, 0.25 and syringaldehyde, 0.5. a mixture two inhibitors. VH: 0.5 g/L vanillin + 0.5 g/L 4-hydroxylbenzaldehyde; HS: 0.5 g/L 4-hydroxylbenzaldehyde + 0.5 g/L syringaldehyde; SV: 0.5 g/L syringaldehyde + 0.5 g/Lvanillin; SP: 0.5 g/L syringaldehyde + 0.25 g/L phenol; VP: 0.5 g/L vanillin + 0.25 g/L phenol; HP: 0.5 g/L 4-hydroxylbenzaldehyde + 0.25 g/L phenol. b mixture of three inhibitors. SHV: 0.5 g/L syringaldehyde + 0.5 g/L 4-hydroxylbenaldehyde + 0.5 g/L vanillin; SHP: 0.5 g/L syringaldehyde + 0.5 g/L 4-hydroxylbenzaldehyde + 0.25 g/L phenol; HVP: 0.5 g/L 4-hydroxylbenzaldehyde + 0.5 g/L vanillin + 0.25 g/L phenol. c mixture of all inhibitors.

1 lmol of NAD+ per minute. Reaction mixture without the enzyme extract was used as the blank. Enzymatic activity was presented as the specific activity expressed as units per mg of protein in the crude enzyme extract (U/mg protein). The protein concentration in the crude enzyme extract was assayed using the Quik Start Bradford Protein Assay Kit (Bio-Rad, USA) and bovine serum albumin (BSA) was used as the standard. 2.6. Analytical methods All samples were withdrawn and filtered through 0.2 lm filters and properly diluted prior to high performance liquid chromatography (HPLC) analysis. The analysis was performed using an Agilent 1200 series HPLC system (Agilent Technologies Inc.). Xylose and xylitol were separated on an Aminex HPX-87H column (Bio-Rad, USA) operating at 75 °C using 5 mM sulphuric acid as the mobile phase at a flow rate of 0.6 mL/min. They were then detected using an online Refractive Index Detector. Analysis of the phenolic compounds, such as syringaldehyde, vanillin, 4-hydroxybenaldehyde and phenol, were performed using the same HPLC system equipped with a Eclipse XDB-C18 column (Bio-Rad, USA) operated at 37 °C and a ultraviolet (UV) detector set at 280 nm. The mobile phase in this case consisted of methanol and 1% acetic acid with the volumetric ratio of 85:15, supplied at a flow rate of 0.6 mL/min. Cell biomass was monitored spectrophotometrically by measuring absorbance at 600 nm. The measurement was made such that

the optical density at 600 nm (OD600) of the samples was smaller than 0.70, as obtained by sample dilution. This is to ensure that the Beer–Lambert law applies. Forty-milliliter samples of whole culture were filtered through 0.45 lm pre-dried, preweighed glass fiber membrane filters and dried at 105 °C till constant weight was obtained. Biomass dry weight was calculated as the difference between the membrane dry weight before and after the culture broth filtration. A calibration curve was prepared between OD600 and biomass dry weight. The OD value was then converted to biomass dry weight using such calibration curve. Biomass concentration (grams per liter) was found to follow the regression equation: X (g/L) = 0.314  (OD600). 3. Results and discussion 3.1. Individual and combined effects of phenolic compounds on cell growth The effects of 4-hydroxybenzaldehyde (Hba: 0–1.0 g/L), vanillin (Van: 0–2.0 g/L), phenol (Phe: 0–1.0 g/L) and syringaldehyde (Syr: 0–2.0 g/L) as individual and combined inhibitors on cell growth of C. athensensis SB18 were investigated. Experiments were conducted in 100-mL flasks containing 40 mL medium under the same cultivation conditions and results are displayed in Fig. 1. Apparently, the intensity of inhibition increased with the rise of initial concentration of the phenolic compounds by showing the prolonged lag phase for cell growth and the reduced cell biomass for-

373

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

B: Vanillin

A: 4-Hydroxybenzaldehyde Blank 0.05 g/L 0.10 g/L 0.50 g/L 1.00 g/L

20

20

10

10

0

0

0 20

30

40

50

60

Xylitol (g/L )

70

40

50

40

40

20

10

10

0

0 60

Blank 0.10 g/L 0.50 g/L 1.00 g/L 2.00 g/L

D: Syringaldehyde

50

20

50

10

50

30

20 30 40 Incubation time (h)

0 0

30

10

20

60

40

30 Blank 0.10 g/L 20 0.50 g/L 1.00 g/L 10 2.00 g/L

Incubation time (h)

Blank 0.05 g/L 0.10 g/L 0.50 g/L 1.00 g/L

0

30

60

C: Phenol

50

Xylose (g/L)

50

40

70

30

30

20

20

10

10

0

0 0

10

20

30

40

50

60

Xylitol (g/L)

20 30 40 Incubation time (h)

40

10

Xylose (g/L)

60

10

Xylose (g/L )

30

30

0

50

40

Xylitol (g/L)

Xylose (g/L )

40

50 50

Xylitol (g/L )

50

70

Incubation time (h)

Fig. 3. Effects of individual inhibitor on xylitol production using Candida athensensis SB18. Solid symbols, xylose; empty symbols, xylitol.

mation. In addition, phenol and syringaldehyde were obviously more toxic than 4-hydroxylbenzaldehye and vanillin evidenced by their more significant effects on cell growth. C. athensensis SB18 growth was more sensitive to phenol than to syringaldehyde, as shown in Fig. 1C and D. Under the conditions investigated, the inhibition of 4-hydroxylbenzaldehyde and vanillin on cell growth was insignificant when their concentration was below 0.5 g/L demonstrated by the slight decrease in biomass production (Table 1). However, with increase of vanillin and 4-hydroxylbenzaldehyde (up to 1.0 g/L), cell specific growth rate moderately increased, indicating the in vivo detoxification of these two compounds by diverting more energy for cell synthesis. Interestingly, in the presence of 0.1 g/L vanillin, a slight increase in cell biomass was observed compared with the blank, suggesting the assimilation of vanillin at low concentration (Table 1). This is in accordance with the report by Kelly et al. (2008). When vanillin concentration was further increased to 2.0 g/L, cell growth was completely inhibited; this was consistent with the reports by Delgenes et al. (1996). However, recently, it was reported that vanillin at 2.0 g/L enhanced xylitol production by Candida guilliermondii (Cortez and Roberto, 2010a), different from the results in this study, revealing that the inhibitory effects of phenolic compounds vary among different microbial species (Delgenes et al., 1996). As shown in Table 1, at the initial concentration of 1.0 g/L, cell growth inhibition was 66.73% by phenol, and 41.55% by syringaldehyde, whereas the inhibition by 4-hydroxylbenzyldehye and vanillin were 16.11% and 13.81%, respectively. This confirmed that phenol and syringaldehyde are more toxic than 4-hydroxybenzaldehyde and vanillin. In addition, phenol was more toxic than syringaldehyde by showing a more notable decrease in the specific growth rate, suggesting the incapability of in vivo phenol

detoxification by C. athensensis SB18. Interestingly, although 58.4% inhibition was observed for syringaldehyde at 2.0 g/L, the specific growth rate was slightly increased to 0.022 L/h at this concentration. This suggests that C. athensensis SB18 might be able to in vivo assimilate syringaldehyde though it is very toxic to cell growth. Cortez and Roberto reported that syringaldehyde was generally more toxic than vanillin at the concentration of 2.0 g/L in C. guilliermondii FTI 20037 (Cortez and Roberto, 2010b). However, in this study, at 2.0 g/L, inhibition of cell growth for C. athensensis SB18 was almost 100% by vanillin, whereas that was only 58.4% by syringaldehyde. This suggests that at high concentration (2.0 g/L and above), C. athensensis SB18 is more sensitive to vanillin than to syringaldehyde. Again it confirmed that the inhibitory effects varied among the inhibitors and the microorganism species (Delgenes et al., 1996). The combined effects of these phenolic compounds on the cell activities are presented in Fig. 2 and Table 2. Cell growth was slightly affected by the presence of two phenolic compounds (Fig. 2A), while the presence of the combination of three and four of the phenolic compounds caused greater inhibition on cell mass production demonstrated by the longer lag phase (Fig. 2B). According to the analysis of the inhibitory effects of each individual phenolic compound, we concluded that at 1.0 g/L and below, the toxicity level of these compounds was in the following order: vanillin < 4-hydroxylbenzaldehyde < syringaldehyde < phenol. For the binary mixture of these phenolic compounds without phenol, the order for the combined inhibitory effects was: vanillin with 4-hydroxylbenzaldehyde < syringaldehyde with vanillin < 4-hydroxylbenzaldehyde with syringaldehyde, which was consistent with their individual inhibitory effects (Table 1). However when phenol was present even at a lower concentration (0.25 g/L) than others, the order of the

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

60

A

60

Blank

60

Van+Hba

50

B

50

50

40

40

40

40

30

30

30

30

20

20

20

20

10

10

10

10

0

0

0

0

Syr+Hba

10

20

30 40 50 Incubation time (h)

60

60

70

Xylose (g/L )

80

0

20

40

60 Blank Syr+HDB+Van Syr+HDB+Phe 50 HDB+Van+Phe All 40

30

30

20

20

10

10

0

0

C

50

Xylose (g/L)

10

0

10

20

30 40 50 Incubation time (h)

60

70

30 40 50 Incubation time (h)

60

50

70

80

Xylitol (g/L)

0

Xylitol (g/L )

Syr+Van Xylose (g/L )

60

Blank Syr+Phe Van+Phe Hba+Phe

Xylitol (g/L)

374

80

Fig. 4. Combined effects of inhibitors on xylitol production using Candida athensensis SB18. The inhibitor contents were (g/L): 4-hydroxylbenzaldehyde, 0.5; vanillin, 0.5; phenol, 0.25 and syringaldehyde, 0.5, when all inhibitors were present. Solid symbols, xylose; empty symbols, xylitol.

inhibitory effects was changed. Surprisingly, the combination of the two most toxic inhibitors (phenol and syringaldehyde) caused the least inhibition of cell growth, while the combination of 4-hydroxylbenzaldhyde and phenol inhibited the cell growth most significantly (Table 2). The above results indicate that in the presence of phenol, the inhibitory effects of these phenolic compounds are not independent. On the other hand, the interaction between phenol and other phenolic compounds might have altered the inhibitory effects. As to the case of three compound mixture, the most toxic was the combination of syringaldehyde, 4-hydroxylbenzaldehyde and phenol by showing the highest inhibition degree (49.95%) even though phenol concentration was lower than the rest. The longest lag phase of 22 h was detected for the medium containing all the inhibitors and 57.14% of cell growth inhibition was obtained. Interestingly, higher specific growth rate was observed for the combinations of VH, SHV, HVP and ALL (Table 2), suggesting the trigger of in vivo detoxification by C. athensensis SB18 under such conditions. From the above analysis, we can conclude that in general the phenolic compounds investigated in this study inhibited the growth of C. athensensis SB18. However the degree of inhibition largely depends on the type of phenolic compound, its concentration and their combinations. It was reported that the mechanisms of such inhibitory effects might be attributed to their negative effects on cell membrane and intracellular metabolisms (Zaldivar et al., 1999; Fitzgerald et al., 2004). 3.2. Individual and combined effects of phenolic compounds on xylitol production The inhibitory effects on xylose consumption and xylitol production by individual inhibitors are shown in Fig. 3. As can be seen, the presence of less than 1.0 g/L 4-hydroxybenzaldehyde,

syringaldehyde and vanillin, or less than 0.5 g/L phenol almost had no effects on xylitol production and xylose comsumption compared with the control. Longer lag phase was observed for 4hydroxylbenzaldehyde, vanillin syringaldehyde at 1.0 g/L, phenol at 0.5 g/L and above. In addition, xylitol production and xylose consumption were more sensitive to the presence of syringaldehyde indicated by the apparent changes in rates of xylose up-take and xylitol accumulation (Fig. 3D and Table 1). However, such changes were insignificant for the rest of phenolic compounds except phenol at 1.0 g/L and vanillin at 2.0 g/L. It seemed that there was a threshold concentration for each phenolic compound, e.g. 1.0 g/L vanillin, 1.0 g/L 4-hydroxylbenzaldehyde, 1.0 g/L syringaldehyde, and 0.5 g/L phenol. The maximal xylitol concentration, xylitol yield and xylitol productivity were almost unaffected when inhibitor concentration was at or below such threshold concentrations (Table 1). Such results suggest that efforts should be made to reduce the concentration of phenolic inhibitors under the inhibiting threshold in order to enhance the fermentation efficiency using C. athensensis SB18. Sampaio et al. reported that although the addition of up to 3.0 g/L vanillin led to a strong inhibition on the metabolic activity of Debaryomyces hansenii UFV-170, whereas no appreciable effect was evident on final xylitol concentration (Sampaio et al., 2007). This phenomenon was partially supported in our study using C. athensensis SB 18. Interestingly, 1.0 g/L vanillin appeared to enhance xylitol production indicated by the slightly higher xylitol concentration, concurring with the recent results reported by Cortez and Roberto (2010a). However, in this case, such enhancement effect of vanillin happened at a lower concentration (1.0 g/L) compared to the results reported in the literature. This might be due to the lower vanillin tolerance of C. athensensis SB18 as evidenced by the complete inhibition of xylitol production

375

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

12.5

50

40

10.0

30

7.5

20

5.0

10

2.5

0

0.0 -10

0

10

20

30

40

50

60

70

10.0

30

7.5

20

5.0

10

2.5

0

0.0 0

10

20

Xylose and xylitol (g/L)

50

15.0

60

12.5

50

70

12.5

10

2.5

0.0

0

0.0

0 60

15.0

Xylose Xylitol Biomass

2.5

10

50

D: 3.0 g/L initital biomass

5.0

5.0

40

80

20

20

30

70

7.5

7.5

20

60

30

30

10

50

10.0

10.0

0

40

40

40

-10

Xylose and xylitol (g/L)

Xylose Xylitol Biomass

Biomass (g/L)

C: 1.0 g/L initital biomass

30

Incubation time (h)

Incubation time (h) 60

12.5

40

-10

80

15.0

Xylose Xylitol Biomass

Biomass (g/L)

60

-10

80

0

10

20

30

40

50

60

70

Biomass (g/L)

Xylose and xylitol (g/L)

50

B: 0.1 g/L initital biomass

15.0

Xylose and xylitol (g/L)

Xylose Xylitol Biomass

Biomass (g/L)

A: Blank, 0.1 g/L initital biomass

60

80

Incubation time (h)

Incubation time (h)

Fig. 5. Effect of inoculum size on cell growth and xylitol production. The inhibitor contents were (g/L): 4-hydroxylbenzaldehyde 0.5; vanillin, 0.5; phenol, 0.25 and syringaldehyde, 0.5.

Table 3 Effect of inoculum size. Inoculum size (g/ L)

Time (h)

Biomass X (g/ L)

Inhibition (%)

Specific growth rate (1/ h)

Xylose (g/ L)

Xylitol (g/ L)

Xylitol Yield YP/X (g/ g)

Xylitol productivity QP (g/ L/h)

0 0.1 1 3

39 63 51 39

10.22 5.43 5.74 6.72

0.00 46.87 43.84 34.25

0.037 0.030 0.017 0.016

1.37 0.00 0.54 3.07

31.44 27.33 25.78 27.01

0.65 0.55 0.52 0.58

0.81 0.43 0.51 0.69

and cell growth at 2.0 g/L vanillin. Among all the phenolic compounds tested, phenol was the most toxic by showing the lowest xylitol production (15.55 g/L), the lowest yield of 0.48 g/g and the slowest productivity (0.35 g/L/h) at 1.0 g/L concentration (Table 1). According to Table 1, the inhibitory effects of these compounds were in the following order: vanillin < 4-hydroxylbenzaldehyde < syringaldehyde < phenol, consistent with their effects on cell growth. As shown in Fig. 4, the combination of multiple phenolic inhibitors had almost no effects on the maximal xylitol concentration except the case when all inhibitor were present. This was consistent with the results obtained when each individual inhibitor was present below the threshold concentration. For the binary mixture without phenol, the combination of two inhibitors resulted in longer lag phase; however the difference between the different combinations was insignificant (Fig. 4A). The combination of phenol with each of the rest phenolic compounds almost did not influence lag phase for xylose consumption and xylitol production (Fig. 4B); however, it enhanced xylitol production, xylitol yield and productivity (Table 2). This is interesting because as an individual inhibitor, phenol was the most toxic compound on the metabolic

activity of C. athensensis SB18 as discussed earlier; however, its interaction with other phenolic compounds seems to have reduced its toxicity indicated by their enhancement of xylitol production, in particular, for the combinations of phenol with vanillin and with 4hydroxylbenzaldehyde. Similar conclusion was reported by Kelly and her coworkers that the combinations of inhibitors were slightly less inhibitory to cell growth and xylitol production than single inhibitors by C. guilliermondii (Kelly et al., 2008). On the other hand, the combination of three phenolic compounds caused prolonged lag phase; however, the difference among different combinations were insignificant (Fig. 4C). When all inhibitors were present, much longer lag phase (34 h) and lower xylitol productivity (0.43 g/L/h) were observed (Table 2), indicating the increased level of the inhibitory effects. It is noticeable that xylitol yield, and productivity were less influenced by the combinations of the inhibitors except the combination of phenol with vanillin or with 4-hydroxylbenzaldehyde (Table 2). This is in agreement with the earlier reports (Rodrigues et al., 2003; Cortez and Roberto, 2010b) that the xylose-to-xylitol bioconversion by C. guilliermondii was less affected by the addition of inhibitors compared with cell growth. Based on the above analysis we can conclude that the

376

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

A: Xylose reductase (XR)

Blank Inhibitor

Blank Inhibitor 0.025

Enzyme activity (U/mg)

Enzyme activity (U/mg)

0.6

B: Xylitol dehydrogenase (XDH)

0.030

0.4

0.2

0.020

0.015

0.010

0.005

0.0

0.000

0

12

24

36

48

60

72

84

0

12

24

36

48

60

72

84

Incubation time (h)

Incubation time (h) 40

C: XR/XDH

Blank Inhibitor

Enzyme activity ratio

30

20

10

0 0

12

24

36 48 60 Incubation time (h)

72

84

Fig. 6. Time profiles of key enzyme activities in the presence of phenolic compounds. The inhibitor contents were (g/L): 4-hydroxylbenzaldehyde 0.5; vanillin, 0.5; phenol, 0.25 and syringaldehyde, 0.5.

inhibition of phenolic compounds on C. athensensis SB18 largely affects the biosynthesis of cell biomass, while it has less influences on xylitol formation and xylose assimilation.

3.3. Inhibition of phenolic compounds on xylose fermentation under different inoculum size Three inoculum levels were introduced to evaluate the effect of initial cell density on the inhibition effects of the phenolic compound mixture containing 1.0 g/L 4-hydroxylbenzaldehyde, 1.0 g/L vanillin, 0.5 g/L phenol and1.0 g/L syringaldehyde. As can be seen in Fig. 5, the increase of initial cell density from 0.1 to 3.0 g/L made negligible contribution to final xylitol concentration; however, it significantly reduced the lag phase. The increase of inoculum size moderately decreased the inhibition degree of cell growth with little effects on the final xylitol production (Table 3). In addition, with such increase of inoculum size, cell growth rate was reduced, whereas xylitol productivity was increased. The above results generally agree with the report by Cortez and Roberto that the inhibition of phenolic compounds on C. guilliermondii could be alleviated by the increase of cellular concentration and the inhibition degree on cellular growth decreased with the increase of initial cellular density (Cortez and Roberto, 2010b). Again the xylitol yield was almost uninfluenced with the change of inoculum size. However, higher inoculum size resulted in shorter fermentation time for the maximal xylitol production. With an inoculum size of 3.0 g/L, optimal xylitol production was obtained at 39 h, same as the control; however, its maximal xylitol production was lower than that of the control (31.44 g/L), with approximately 14% decrease. Correspondingly, the maximal cell

biomass production was only 6.72 g/L, whereas that for the control was 10.22 g/L, with almost 35% decrease. Likewise, xylitol productivity was reduced to 0.69 g/L/h, whereas that for the control was 0.81 g/L/h, around 14.8% decrease. The above results suggest that the increase of inoculum size could not completely diminish the inhibitory effects of these phenolic compounds on final xylitol production and cell biomass growth although it greatly reduced the time for xylitol production and moderately decreased the inhibition degree for cell growth. Much higher cell density can be applied to further alleviate the inhibitory effects. However, at the industrial scale, very high inoculum size is impractical. The above analysis proves that a moderately high cell biomass density could serve as an effective mean to improve xylose-to-xylitol bioconversion efficiency.

3.4. Effect of phenolic compounds on xylose reductase (XR) and xylitol dehydrogenase (XDH) activities It was reported that the presence of limited amount of acetic acid in the fermentative medium had positive effect on the activi-

Table 4 Assimilation of phenolic compounds by Candida athensensis SB18. Phenolic compounds

Vanillin 4-hydroxybenzaldehyde Syringaldehyde Phenol

Concentration (g/L) 0h

39 h

75 h

0.5 0.5 0.5 0.25

0.002 ± 0.000 0.003 ± 0.000 0.364 ± 0.005 0.232 ± 0.001

0 0 0.227 ± 0.002 0.224 ± 0.001

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

ties of XR and XDH in C. guilliermondii (Lima et al., 2004), and furans such as hydroxylmethyl furfural (HMF) and furfural were also observed to act as strong inhibitors to many soluble enzymes (Modig et al., 2002). This experiment was therefore conducted to evaluate the impact of phenolic inhibitors on the activities of XR and XDH during fermentation. As illustrated in Fig. 6A, XR was much more inhibited in the earlier stage of fermentation (<50 h). On the other hand, throughout the fermentation process, the activity of XDH was almost unaffected by these phenolic compounds (Fig. 6B). Due to the more significant inhibition on XR at the earlier stage of fermentation, the ratio of XR to XDH was reduced greatly in the earlier stage of fermentation and it then resumed to the comparable level to the control (Fig. 6C). It was elucidated that phenolic compound may act on the biological membranes to affect their ability to serve as selective barriers and enzyme matrices (Heipieper et al., 1994; Fitzgerald et al., 2004). XR is one of the key enzymes in the bioconversion of xylose to xylitol and xylose is the best inducer for XR. The lower XR activities obtained at the initial stage of fermentation indicate the lower xylose concentration available in the intracellular spaces possibly due to the negatively affected cell membranes. In addition, similar to the case of most xylitol producing yeasts, in C. athensensis SB18, XR was NADPH dependent and XDH was NAD+ dependent (Winkelhausen and Kuzmanova, 1998). The lower XR activity at the beginning of fermentation might be due to the low level of NADPH, which was being consumed by the phenolic compound degrading enzymes for in vivo detoxification (Vilímková et al., 2008). After 50 h of fermentation, the XR activity was fully recaptured indicating the completion of in vivo detoxification and the recovery of cell membranes and the NADPH level necessary for XR. On the other hand, the relative constant XDH activity obtained in this study is expected as C. athensensis SB18 is an efficient xylitol producer; its XDH activity is consistently low allowing sufficient xylitol accumulation. 3.5. Assimilation of phenolic compounds by C. athensensis SB 18 It was reported that aromatic aldehydes can be converted to their respective acids or alcohols by C. guilliermondii (Tanaka and hirokane, 2000; Cortez and Roberto, 2010a). The assimilation of the phenolic compounds was explained as the most important mechanisms for in vivo detoxification (Palmqvist and Hägerdal, 2000). In this part of the study, the assimilation the four phenolic compounds by C. athensensis SB 18 was investigated. Table 4 presents the phenolic compound assimilation results. It can be seen that vanillin and 4-hydroxybenzaldehyde was completely metabolized in 39 h of incubation with vanillin being consumed slightly faster. On the other hand, syringaldehyde was consumed at a lower rate with 0.364 and 0.227 g/L unconsumed at 39 and 75 h, respectively, while phenol almost could not be consumed by C. athensensis SB 18 with 0.224 g/L unconsumed even at 75 h with an initial concentration of 0.25 g/L. This accords perfectly well with our previous conclusion that the inhibitory effects of the phenolic compounds investigated in this study was: vanillin < 4-hydroxylbenzaldehyde < syringaldehyde < phenol. Such results strongly suggest that there is a close correlation between the inhibitory effects of these phenolic compounds and their in vivo degradation by the microorganism. The higher tolerance of C. athensensis SB 18 to vanillin and 4-hydroxylbenzaldehyde was clearly due to the rapid assimilation of these two compounds. On the other hand, the sensitive response to syringaldehyde by C. athensensis SB 18 was because of the much slower assimilation rate of this compound by this microorganism. In addition, the stronger inhibitory effect of phenol was due to the failure of in vivo phenol degradation by C. athensensis SB 18. The above results and analysis is in full agreement with the enzyme activity analysis that the lower XR activity

377

at the earlier stage of the fermentation was due to the in vivo detoxification activities. For the efficient bioconversion of hemicellulosic hydrolysate to xylitol, it is therefore important to keep such phenolic compounds below their in vivo detoxification level. 4. Conclusion The presence of lignin-derived phenolic compounds significantly affected xylitol production by C. athensensis SB18. The toxicity level follows the order: vanillin < 4-hydroxylbenzaldehyde < syringaldehyde < phenol; this closely correlated to their in vivo assimilation. Below an overall content of 1.0 g/L, the inhibitory effects were insignificant. Phenolic compounds mainly affected the lag phase, final cell density and sometimes xylitol productivity rather than xylitol final concentration and xylitol yield. They influenced more xylose reductase activity than xylitol dehydrogenase activity. Increase of the inoculum size alleviated such inhibitory effects. Acknowledgements The authors are grateful for the financial support to this work from Singapore Totalisation Board and Ngee Ann Kongsi. The authors also would like to thank Ngee Ann Polytechnic for providing the internship opportunities. References Ando, S., Arai, I., Kiyoto, K., Hanai, S., 1986. Identification of aromatic monomers in steam-exploded poplar and their influence on ethanol fermentation. J. Ferment. Technol. 64, 567–570. Almeida, J.R.M., Modig, T., Petersson, A., Hägerdal, B.H., Liden, G., Grauslund, F.G., 2007. Increased tolerance and conversion of inhibitors in lignocellulosic hydrolysates by Saccharomyces cerevisiae. J. Chem. Technol. Biotechnol. 82, 340–349. Buchert, J., Niemela, K., Puls, J., Poutanen, K., 1990. Improvement in the fermentability of steamed hemicellulose hydrolysate by ion exclusion. Proc. Biochem. 25, 176–180. Carvalho, W., Santos, J.C., Canilha, L., Almeida e Silva, J.B., Felipe, M.G.A., Mancilha, I.M., Silva, S.S., 2003. A study on xylitol production from sugarcane bagasse hemicellulosic hydrolysate by Ca-alginate entrapped cells in a stirred tank reactor. Proc. Biochem. 39, 2135–2141. Chiang, C., Knight, S.G., 1966. D-xylose reductase and D-xylitol dehydrogenase from Penicillium chrysogenum. Methods Enzymol. 9, 188–193. Clark, T.A., Mackie, K.L., 1984. Fermentation inhibitors in wood hydrolysates derived from the softwood Pinus radiata. J. Chem. Technol. Biotechnol. 34, 101–110. Converti, A., Dominguez, J.M., Perego, P., Silva, S.S., Zilli, M., 2000. Wood hydrolysis and hydrolysate detoxification for subsequent xylitol production. Chem. Eng. Technol. 23, 1013–1020. Cortez, D.V., Roberto, I.C., 2010a. Improved xylitol production in media containing phenolic aldehydes: application of response surface methodology for optimization and modeling of bioprocess. J. Chem. Technol. Biotechnol. 85, 33–38. Cortez, D.V., Roberto, I.C., 2010b. Individual and interaction effects of vanillin and syringaldehyde on the xylitol formation by Candida guilliermondii. Bioresour. Technol. 101, 1858–1865. Delgenes, J.P., Moletta, R., Navarro, J.M., 1996. Effects of lignocelluloses degradation products on ethanol fermentations of glucose and xylose by Saccharomyces cerevisiae, Zymomonas mobilis, Pichia stipitis, and Candida shehatae. Enzyme Microb. Technol. 19, 220–225. Fitzgerald, D.J., Stratford, M., Gasson, M.J., Ueckert, J., Bos, A., Narbad, A., 2004. Mode of antimicrobial action of vanillin against Escherichia coli, Lactobacillus plantarum and Listeria innocua. J. Appl. Microbiol. 97, 104–113. Geng, Z.C., Sun, R.C., Sun, X.F., Lu, Q., 2003. Comparative study of hemicelluloses released during two-stage treatment with acidic organosolv and alkaline peroxide from Caligonum monogoliacum and Tamarix spp.. Polym. Degrad. Stab. 80, 315–325. Gurpilhares, D.B., Hasmann, F.A., Pessoa Jr, A., Roberto, I.C., 2006. Optimization of glucose-6-phosphate dehydrogenase releasing from Candida guilliermondii by disruption with glass beads. Enzyme Microb. Technol. 39, 591–595. Heipieper, H.J., Weber, G.J., Sikkema, J., Kewelo, H., de Bont, J.A.M., 1994. Mechanism of resistance of whole cells to toxic organic solvents. Trend Biotechnol. 12, 409– 415. Jönsson, L.J., Palmqvist, E., Nilvebrant, N.O., Hahn-Hägerdal, B., 1998. Detoxification of wood hydrolysates with laccase and peroxidase from the white-rot fungus Trametes versicolor. Appl. Microbiol. Biotechnol. 49, 691–697. Kelly, C., Jones, O., Barnhart, C., Lajoie, C., 2008. Effect of furfural, vanillin and syringaldehyde on Candida guilliermondii growth and xylitol biosynthesis. Appl. Biochem. Biotechnol. 148, 97–108.

378

J. Zhang et al. / Bioresource Technology 121 (2012) 369–378

Lima, L.H.A., Felipe, M.G.A., Vitolo, M., Torres, F.A.G., 2004. Effect of acetic acid present in bagasse hydrolysate on the activities of xylose reductase and xylitol dehydrogenase in Candida guilliermondii. Appl. Microbiol. Biotechnol. 65, 734– 738. Martin, C., Jönsson, L.J., 2003. Comparison of the resistance of industrial and laboratory strains of Saccharomyces and Zygosacccharomyces to lignocellulosederived fermentation inhibitors. Enzyme Microb. Technol. 32, 386–395. Modig, T., Liden, G., Taherzadeh, M.J., 2002. Inhibition effects of furfural on alcohol dehydrogenase, aldehyde dehydrogenase and pyruvate dehydrogenase. Biochem. J. 363, 769–776. Mussatto, S.I., Roberto, I.C., 2004. Alternatives for detoxification of diluted acid lignocellulosic hydrolysates for use in fermentative processes: a review. Bioresour. Technol. 93, 1–10. Nigam, P., Singh, D., 1995. Processes for fermentative production of xylitol – a sugar substitute. Proc. Biochem. 30, 117–124. Palmqvist, E., Almeida, J., Hägerdal, B.H., 1999. Influence of furfural on anaerobic glycolytic kinetics of Saccharomyces cerevisiae in batch culture. Biotechnol. Bioeng. 62, 447–454. Palmqvist, E., Hägerdal, B.H., 2000. Fermentation of lignocellulosic hydrolysates.II: inhibitors and mechanisms of inhibition. Bioresour. Technol. 74, 25–33. Rodrigues, R.C.L.B., Felipe, M.G.A., Roberto, I.C., Vitolo, M., 2003. Batch xylitol production by Candida guilliermondii FTI 20037 from sugarcane bagasse hemicellulosic hydrolysate at controlled pH values. Bioprocess Biosyst. Eng. 26, 103–107. Rodrigues, R.C.L.B., Sene, L., Matos, G.S., Roberto, I.C., Pessoa Jr, A., Felipe, M.G.A., 2006. Enhanced xylitol production by precultivation of Candida guilliermondii cells in sugarcane bagasse hemicellulosic hydrolysate. Curr. Microbiol. 53, 53–59. Russell, J.B., 1992. Another explanation for the toxicity of fermentation acids at low pH: anion accumulation versus uncoupling. J. Appl. Bacteriol. 73, 363–370. Sampaio, F.C., Torre, P., Passps, F.M.L., de Moraes, C.A., Perego, P., Converti, A., 2007. Influence of inhibitory compounds and minor sugars on xylitol production by Debaryomyces hansenii. Appl. Biochem. Biotechnol. 136, 165–180.

Sears, K.D., Beelik, A., Casebier, R.L., Engen, R.J., Hamilton, J.K., Hergert, H.L., 1971. Southern pine prehydrolyzates: characterization of polysaccharides and lignin fragments. J. Polym. Sci. 36, 425–443. Smiley, K.L., Bolen, P.L., 1982. Demonstration of D-xylose reductase and D-xylitol dehydrogenase in Pachysolen tannophilus. Biotechnol. Lett. 4, 607–610. Stouthamer, A.H., 1979. The search for a correlation between theoretical and experimental growth yields. Int. Rev. Biochem. 21, 1–47. Tanaka, M., Hirokane, Y., 2000. Oxidation of aromatic aldehyde to aromatic carboxylic acid by Burkholderia cepacia TM1 isolated from humus. J. Biosci. Bioeng. 90, 341–343. Ulbricht, R.J., Northup, S.J., Thomas, J.A., 1984. A review of 5-hydroxymethyl furfural (HMF) in parenteral solutions. Fundam. Appl. Toxicol. 4, 843–853. Verduyn, C., Postma, E., Scheffers, W.A., Dijken, J.P., 1992. Effect of benzoic acid on the regulation of respiration and alcoholic fermentation. Yeast 8, 501–517. Vilímková, L., Páca Jr., J., Kremlácˇková, V., Páca, J., Stiborová, M., 2008. Isolation of cytoplasmic NADPH-dependent phenol hydroxylase and catechol-1, 2dioxygenase from Candida tropicalis yeast. Interdiscip. Toxicol. 1, 225–230. Villarreal, M.L.M., Prata, A.M.R., Felipe, M.G.A., Silva, J.B.A., 2006. Detoxification procedures of eucalyptus hemicelluloses hydrolysate for xylitol production by Candida guilliermondii. Enzyme Microb. Technol. 40, 17–24. Winkelhausen, E., Kuzmanova, S., 1998. Microbial conversion of D-xylose to xylitol. J. Ferment. Bioeng. 86, 1–14. Zaldivar, J., Martinez, A., Ingram, L.O., 1999. Effect of selected aldehydes on growth and fermentation of ethanologenic Escherichia coli. Biotechnol. Bioeng. 65, 24– 33. Zhang, J.M., Geng, A.L., Yao, C.Y., Lu, Y.H., Li, Q.B., 2012. Xylitol production from Dxylose and horticultural waste hemicellulosic hydrolysate by a new isolate of Candida athensensis SB18. Bioresour. Technol. 105, 134–141.