Appl Microbiol Biotechnol DOI 10.1007/s00253-013-5449-x
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Production of D -arabitol from raw glycerol by Candida quercitrusa Jun Yoshikawa & Hiroshi Habe & Tomotake Morita & Tokuma Fukuoka & Tomohiro Imura & Hiroyuki Iwabuchi & Shingo Uemura & Takamitsu Tamura & Dai Kitamoto
Received: 19 September 2013 / Revised: 27 November 2013 / Accepted: 28 November 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract To promote the effective use of raw glycerol (a byproduct of biodiesel production), 110 yeast strains that produce D -arabitol from glycerol were isolated from environmental samples. Among them, strain 17-2A was an effective D arabitol producer in the presence of 250 g/l glycerol and was identified as Candida quercitrusa based on morphological, physicochemical, and phylogenetic analyses. C. quercitrusa type strain NBRC1022 produced the greatest quantity of D arabitol (41.7 g/l) when the ability to produce D -arabitol from raw glycerol was compared among C. quercitrusa 17-2A and its phylogenetically related strains in flask culture. Under optimized culture conditions, strain NBRC1022 produced D arabitol at a concentration of 58.2 g/l after a 7-day cultivation in 250 g/l glycerol, 6 g/l yeast extract, and 2 g/l CaCl2. The culture conditions were further investigated with raw glycerol using a jar fermenter; the concentration of D -arabitol reached 67.1 g/l after 7 days and 85.1 g/l after 10 days, respectively, which corresponded to 0.40 g/g of glycerol. To our knowledge, the present D -arabitol yield from glycerol is higher than reported previously using microbial production. Keywords D -arabitol . Candida quercitrusa . Raw glycerol . Sugar alcohol . Biodiesel Electronic supplementary material The online version of this article (doi:10.1007/s00253-013-5449-x) contains supplementary material, which is available to authorized users. J. Yoshikawa : H. Habe : T. Morita : T. Fukuoka : T. Imura : D. Kitamoto (*) Research Institute for Innovation in Sustainable Chemistry, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5-2, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan e-mail: [email protected] H. Iwabuchi : S. Uemura : T. Tamura Chemical Research Laboratories, Lion Corporation, 7-13-12 Hirai, Edogawa-ku, Tokyo 132-0035, Japan
Introduction Recently, the industrial use of renewable resources instead of petroleum has become important from an environmental standpoint (Willke and Vorlop 2004). Among some currently available resources such as starch, lignocelluloses, sugars, and organic wastes, plant oils have been used to produce biodiesel fuel (BDF) and some oleochemicals, including surfactants. Their production is based on an ester exchange reaction between triacylglycerol and alcohol (e.g., methanol) under alkaline conditions, resulting in the formation of a crude glycerol waste (raw glycerol). Due to the considerable increase in BDF production worldwide over the last few years, surplus glycerol derived from BDF production has become an alternative feedstock for production of chemicals and fuels (Yazdani and Gonzalez 2007). Since the quality of raw glycerol is usually low, refining processes to obtain pure glycerol with some cost and energy are required. Thus, the use of raw glycerol to produce value-added chemicals is important. Sugar alcohols such as erythritol, arabitol, xylitol, and mannitol are produced by osmophilic yeasts against the osmotic pressure across the cell membrane, and some sugar alcohols are used in the food and pharmaceutical industries. Among them, D -arabitol is a C5-sugar alcohol identified as one of the top 12 building block compounds based on biomass by the US Department of Energy (Werpy and Petersen 2004). Werpy and Petersen (2004) reported that some chemicals such as arabinoic acid, xylonic acid, propylene glycol and ethylene glycol can be produced from arabitol and its enantiomer, xylitol. In addition, D -arabitol can be used as a starting material for xylitol production, which has numerous applications in the food and pharmaceutical industries. Furthermore, D arabitol itself is expected to show potential applications like xylitol as an alternative low calorie sweetener (Koganti et al. 2011). Thus, these reports stress the importance of industrial production of D -arabitol.
Appl Microbiol Biotechnol
At this time, there are many reports of microbial production of D -arabitol by various osmophilic yeast species (Hajny 1964; Bernard et al. 1981; Nobre and da Costa 1985; van ECK et al. 1989; Bisping et al. 1996; Nozaki et al. 2003; Saha et al. 2007; Zhu et al. 2010; Song et al. 2011). However, to our knowledge, only two reports of D -arabitol production from glycerol, by Onishi and Suzuki (1970) and Koganti et al. (2011), have been published. In addition, there has been no report of microbial production of D -arabitol using raw glycerol as a feedstock. Raw glycerol frequently inhibits microbial growth because it contains impurities, such as non-glycerol organic matter (e.g., tar and methanol), sodium or potassium salts, and heavy metals. Hence, novel microbial strains suitable for producing D -arabitol from raw glycerol are required. In this study, we screened microbial strains suitable for producing D -arabitol from raw glycerol at high concentration. Among the tested yeast candidates, Candida quercitrusa strain NBRC1022 was selected as the best D -arabitol-producing strain in the presence of 250 g/l glycerol, and D -arabitol production from raw glycerol was optimized using a jar fermenter.
yeast extract, 3 g/l malt extract, 5 g/l peptone, 5 g/l Na2SO4, 50 mg/l streptomycin, and 50 mg/l chloramphenicol]. Environmental samples such as flowers, fruits, and leaves were added to test tubes containing 5 ml of the medium, followed by incubation on a reciprocal shaker (180 strokes/min) at 28 °C for 3 to 7 days. Then, 100 μl of each culture was inoculated into deep-well plates (96 well) containing 1 ml each of the same medium and was incubated on a reciprocal shaker (150 rpm) at 28 °C for 3 days. After assessment of D arabitol production using paper chromatography, positive cultures were appropriately diluted and spread on YM agar plates. The resultant colonies were again inoculated into deep-well plates and were assessed for their D -arabitol production by paper chromatography. Paper chromatography To detect D -arabitol production, 10 μl of each culture was applied to paper chromatography with 1-propanol/ethyl acetate/water (7:1:2, v/v) as a solvent system. Detection was then performed using Yoda’s reagent containing 5 g/l KIO4 and 150 g/l MnSO4/p,p'-tetramethyl diaminodiphenylmethane saturated with 2 M acetic acid (1:1, v/v; Yoda 1951).
Materials and methods High performance liquid chromatography Microorganisms and culture conditions A total of 10 type strains, including C. quercitrusa NBRC1022, Candida natalensis NBRC1981, Candida fragi NBRC10759, Debaryomyces hansenii var. fabryi NBRC0015, Debaryomyces hansenii var. hansenii N B R C 0 0 8 3 , M e t s c h n i k o w i a lu n a t a N B R C 1 6 0 5 , Metschnikowia pulcherrima NBRC1678, Metschnikowia reukaufii NBRC1679, Metschnikowia zobelli NBRC1680, and Hansenula anomala NBRC10213, were obtained from the Biological Resource Center, NITE (NBRC; Chiba, Japan). Strain 17-2A isolated in this study was deposited as FERM P-21420 at the NITE Patent Microorganisms Depositary (NPMD, Chiba, Japan). Each strain was maintained on YM agar plates [10 g/l glucose, 5 g/l peptone, 3 g/l yeast extract, 3 g/l malt extract, and 15 g/l agar]. Raw glycerol [composition, 850 g/l glycerol, 20 g/l organic impurities, 20 g/l inorganic impurities (Na2SO4 >5 g/l), 110 g/l H2O] was supplied from Lion Corporation, Japan. The raw glycerol was used as a substrate for D -arabitol production after passing through a column packed with granular activated charcoal to remove impurities. Screening of microorganisms producing D -arabitol from glycerol Screening of D -arabitol-producing microorganisms was performed using the enrichment medium [250 g/l glycerol, 3 g/l
To quantify D -arabitol and glycerol, high performance liquid chromatography (HPLC) analysis was performed using a SUGAR SC1011 column (8 mm×30 cm; Showa Denko, Tokyo, Japan) at 80 °C and a refractive index detector RID10A (Shimadzu, Kyoto, Japan). The column was eluted with H2O at 1 ml/min. The concentration of each product was determined from a calibration curve with the corresponding authentic reagents. Molecular phylogenetic analysis The cells of strain 17-2A were prepared by cultivation with YM medium at 25 °C. DNA extraction was performed using the DNeasy plant mini kit (QIAGEN, Hilden, Germany) according to the manufacturer’s protocol. The D1/D2 region of 26S ribosomal RNA gene was amplified by PCR using primers NL1, NL2, NL3, and NL4 (O’Donnell 1993) and puReTaq Ready-To-Go polymerase chain reaction (PCR) beads (GE Healthcare Life Sciences, Buckinghamshire, UK). After an initial denaturation at 95 °C for 7 min, 35 cycles were performed at 95 °C for 1 min, 55 °C for 1 min, and 72 °C for 1.5 min with the above primers. The program ended with a 7-min extension at 72 °C. Sequencing reactions were performed using the ABI PRISMTM BigDyeTM Terminator, v3.1 Kit (Applied Biosystems, CA, USA), as described by the manufacturer. The PCR products were sequenced using the ABI PRISMTM 3100 genetic analyzer system (Applied
Appl Microbiol Biotechnol
Biosystems). The sequence of the D1/D2 region of 26S ribosomal RNA gene in strain 17-2A was compared with closely related species obtained from the DNA Data Bank of Japan (DDBJ; http://www.ddbj.nig.ac.jp) using the BLAST search software and alignement using the Clustal W software (Thompson et al. 1994). The phylogenetic tree was visualized using the TreeView software with Candida parapsilosis as an outgroup. The nucleotide sequence for the D1/D2 region of strain 17-2A was registered in the DDBJ, EMBL, and GenBank nucleotide sequence databases with accession number AB855788.
or casamino acid was added to the basic culture medium A. For the effects of yeast extract concentration, 1, 3, 5, and 7 g/l yeast extract was added to the basic culture medium A. To explore the effects of initial glycerol concentrations, 150, 200, 300, and 350 g/l glycerol was replaced with 250 g/l glycerol in the basic culture medium B. To examine the effects of calcium addition, 0.5, 1, and 2 g/l calcium chloride was added to the basic culture medium B. The concentrations of D -arabitol and glycerol in cultures were determined by HPLC. All data shown are the averages and standard deviations from three independent experiments.
Physiological test
D -Arabitol production from raw glycerol using a jar fermenter
For strain 17-2A, the assimilation test for several carbon and nitrogen compounds, resistance tests, vitamin auxotrophic tests, urease tests, and diazonium blue B color (DBB) tests were performed as described previously (Kurtzman and Fell 1998; Barnett et al. 2000). Morphological characteristics of the microorganism were observed using a BX51 optical microscope (Olympus, Tokyo).
A single colony of C. quercitrusa NBRC1022 was inoculated into a test tube containing 5 ml of the seed culture medium. After incubation on a reciprocal shaker (180 strokes/min) at 28 °C for 1 day, 1 ml of the pre-culture was inoculated into a 300-ml Erlenmeyer flask containing 30 ml of the same medium and further incubated on a rotary shaker (250 rpm) at 28 °C for 1 day. A total of 60 ml of the culture (two flasks) was inoculated into a 1.6 l-jar fermenter containing 800 ml of the raw glycerol medium with 2 g/l CaCl2. During the jar fermenter experiments, the aeration rate and agitation speed were set to 1 vvm and 800 rpm, respectively. Temperature was maintained at 28±1 °C. The concentrations of D -arabitol and glycerol in the culture were analyzed by HPLC. Three independent experiments were carried out.
Production of D -arabitol by newly isolated strains and related strains A single colony of each strain on YM plates was inoculated into a test tube containing 5 ml of the seed culture medium [10 g/l glycerol, 3 g/l yeast extract, 3 g/l malt extract, and 5 g/l peptone]. After incubation on a reciprocal shaker (180 strokes/ min) at 28 °C for 1 day, 1 ml of culture was inoculated into 300-ml Erlenmeyer flask containing 30 ml of the raw glycerol medium [300 g/l raw glycerol, 6 g/l yeast extract, 1 g/l KH2PO4; pH 6.0], unless otherwise noted. The culture was further incubated on a rotary shaker (250 rpm) at 28 °C for 7 days. The concentrations of D -arabitol and glycerol in cultures were determined by HPLC. Optimization of culture conditions for D -arabitol production in flask culture A single colony of C. quercitrusa NBRC1022 was inoculated into a test tube containing 5 ml of the seed culture medium. After incubation on a reciprocal shaker (180 strokes/min) at 28 °C for 1 day, 1 ml of the pre-culture was inoculated into a 300-ml Erlenmeyer flask containing 30 ml of the basic culture medium A [250 g/l glycerol, 1 g/l yeast extract, 1 g/l KH2PO4, and 5 g/l Na2SO4; pH 6.0] and the basic culture medium B [250 g/l glycerol, 6 g/l yeast extract, 1 g/l KH2PO4, and 5 g/l Na2SO4; pH 6.0] with some modifications (described below), followed by another incubation on a rotary shaker (250 rpm) at 28 °C for 7 days. To evaluate the effect of nitrogenous and organic nutrients, 5 g/l of yeast extract, corn steep liquor, peptone, malt extract
Results Screening of microorganisms producing D -arabitol from glycerol To screen D -arabitol-producing microorganisms using deepwell plates, an increased concentration of glycerol (250 g/l) was used to promote the use of surplus glycerol. Also, 5 g/l Na2SO4 was added to the medium because raw glycerol used in this study contains the corresponding concentration of Na2SO4. The 2,300 microbial strains that grew on 250 g/l glycerol have been isolated from environmental samples around Japan such as flowers, fruits, and leaves. The isolated strains were investigated for their production of sugar alcohols using paper chromatography. In this solvent system, production of C5-, C6-, or both sugar alcohols was observed, but detecting C4-sugar alcohols such as erythritol was problematic because of a huge spot of C3-glycerol (Fig. 1A). Among these sugar alcohol-producers, 110 strains were predicted to produce C5-sugar alcohols by paper chromatography (an example is shown in Fig. 1B); their D -arabitol production was then assessed by HPLC (Fig. 1C). All C5-sugar alcohols produced by the 110 strains were determined to be D -arabitol,
Appl Microbiol Biotechnol
a
b
Table 1 Comparison of D -arabitol production from the raw glycerol medium by related yeast strains
C3 C5 C6
1
2
3
4
5
6
7
8
9
Strain
D -Arabitol
C. quercitrusa 17-2A C. quercitrusa NBRC1022 C. natalensis NBRC1981 C. fragi NBRC10759 D. hansenii var. fabryi NBRC0015 D. hansenii var. hansenii NBRC0083 M. reukaufii NBRC1679 H. anomala NBRC10213
40.7 41.7 ND ND 9.75 26.5 ND 15.6
(g/l)
ND not detected
c Glycerol (11.6 min)
7.5
D-Arabitol (12.4 min)
10.0 12.5 15.0 Retention time (min)
17.5
Fig. 1 Detection and quantification of D -arabitol in microbial culture. An example of a paper chromatography profile: standard samples (A) and microbial culture (B). An example of an HPLC profile: culture broth of strain 17-2A (C). The cultivation was performed in the enrichment medium. Paper chromatography was performed using 1-propanol/ethyl acetate/water (7:1:2, v/v) as a solvent system. Detection was performed by spraying 150 g/l MnSO4/p,p′-tetramethyl diaminodiphenylmethane saturated by 2 M acetic acid (1:1, v/v) after spraying 5 g/l KIO4. Lanes 1, glycerol; 2, erythritol; 3, xylitol; 4, D -arabitol; 5, sorbitol; 6, mannitol; 7, strain 15-11E; 8, stain 17-2A; 9, strain 17-12D. C3, C5, and C6 represent spot areas of 3-, 5-, and 6-carbon sugar alcohols, respectively
and their concentrations ranged from 3.4 to 12.7 g/l when cultured using the enrichment medium in 96deep-well plates. The isolated 110 strains were tested for D -arabitol production with the raw glycerol medium in flask culture. Strain 172A showed the highest D -arabitol concentration, 40.7 g/l. Strains 15-11E and 17-12D were the second (31.4 g/l) and third (14.0 g/l) highest D -arabitol producers, respectively. Thus, we next characterized strain 17-2A.
databases. The D1/D2 region of the 26S rRNA gene sequence was similar to Candida quercitrusa CBS101427 [99.8 % (564/565 bp)] and C. natalensis [98.1 % (555/566 bp)]. Physiological characteristics of 17-2Awere similar to those observed for C. quercitrusa, excluding the assimilation of D -ribose and L -arabitol, growth at 37 °C, and vitamin auxotrophy (Table S1 in supplementary materials). Based on morphological and physiological properties (Fig. S1 and Table S1; Kurtzman and Fell 1998) and molecular phylogenetic analysis using the D1/ D2 region of 26S rRNA genes (Fig. S2 in supplementary materials), we identified strain 17-2A as C. quercitrusa. Comparison of D -arabitol production among related yeast strains C. quercitrusa 17-2A was an efficient D -arabitol producer, but this species had not been reported previously to produce D arabitol. Thus, D -arabitol production was compared between strain 17-2A and the type strain C. quercitrusa . For the related type species shown in Fig. S2 (C. natalensis , C. fragi and D. hansenii) and the known D -arabitol-producing yeast from glucose (Metschnikowia reukaufii and Hansenula anomala), D -arabitol production with the raw glycerol medium was also investigated (Table 1). C. quercitrusa NBRC1022 showed the highest production (41.7 g/l), while C. natalensis NBRC1981, C. fragi NBRC10759 and M. reukaufii NBRC1679 did not produce D -arabitol. Although D. hansenii var. hansenii NBRC0083 and
Characterization of strain 17-2A After a 4-day incubation at 25˚C on YM agar plates, we assessed the morphology of strain 17-2A (Fig. S1 in supplementary materials). The cells were elongate (2.5–17.5)×(2–3.5)μm, and grew normally with polar budding. Pseudomycelia developed, and no sexual reproduction was observed within 30 days. To identify the yeast strain 17-2A, the D1/D2 region of 26S rRNA gene was determined and aligned with related available sequences from public
Table 2 Effect of nitrogenous and organic nutrients added to the basic medium A on D -arabitol production by C. quercitrusa NBRC1022 ND not detected
Nutrient (5 g/l)
D -Arabitol
Yeast extract Corn steep liquor Peptone
42.5 21.6 3.69
Malt extract Casamino acids
11.0 ND
(g/l)
60
50
50 (g/L)
60
40
D-Arabitol
D-Arabitol
(g/L)
Appl Microbiol Biotechnol
30 20
30 20 10
10 0
40
0
2
4 6 Yeast extract (g/L)
0
8
0.5
1
2
Calcium chloride (g/L)
Fig. 2 Effect of yeast extract concentration on D -arabitol production by C. quercitrusa NBRC1022. Cultivation was performed in the basic culture medium A with 2 to 8 g/l final yeast extract concentration. The vertical bar shows the standard deviation of the mean of three independent experiments
Fig. 4 Effect of calcium chloride concentration on D -arabitol production by C. quercitrusa NBRC1022. Cultivation was performed in the basic culture medium B with 0 to 2 g/l CaCl2. The vertical bar shows the standard deviation of the mean of three independent experiments
H. anomala NBRC10213 could produce D -arabitol from raw glycerol, the concentrations were considerably lower than that of C. quercitrusa. When reproducibility of D -arabitol production was compared between strains NBRC1022 and 17-2A, we confirmed that strain NBRC1022 gave about 10 % to 20 % larger amount of D -arabitol was produced (data not shown). Hence, C. quercitrusa NBRC1022 was selected as the optimum strain and was used in the following studies.
Figure 3 shows the effect of initial glycerol concentration on D -arabitol production; 250–300 g/l initial glycerol resulted in improved production compared to lower glycerol concentrations. Surprisingly, D -arabitol production occurred in the presence of relatively high osmotic stress at 350 g/l glycerol. We preliminary examined the effect of various minerals addition (Mn2+, Ca2+, Fe2+, and Zn2+) on D -arabitol production; only Ca2+ had positive effects, while most minerals inhibited D -arabitol production (data not shown). D -Arabitol production increased upon addition of more than 0.5 g/l CaCl2; the D -arabitol concentration was 58.2 g/l in the presence of 2 g/l CaCl2 (Fig. 4). Large-scale production of D -arabitol from raw glycerol by C. quercitrusa NBRC1022 Based on the above optimization experiments, large-scale production of D -arabitol from raw glycerol by C. quercitrusa NBRC1022 was examined using a jar fermenter (Fig. 5). After a 7-day cultivation, the concentration was 67.1 g/l (the yield was 0.41 g/g of glycerol). Further cultivation up to 10 days 280
100 90 80 70 60 50 40 30 20 10 0
240 Glycerol (g/L)
60
D-Arabitol
(g/L)
50 40
200 160
120
30
80
20
40
10
0 0
0 150
200
250 300 Glycerol (g/L)
350
Fig. 3 Effect of glycerol concentration on D -arabitol production by C. quercitrusa NBRC1022. Cultivation was performed in the basic culture medium B with 150 to 350 g/l glycerol. The vertical bar shows the standard deviation of the mean of three independent experiments
1
2
3
4
5
6
7
8
9
and dry cell (g/L)
Our preliminary studies showed that C. quercitrusa NBRC1022 stably produced D -arabitol at 28 °C in the pH range 5 to 7. To optimize the culture conditions for D -arabitol production, the basic culture media A or B (pH 6) were used. C. quercitrusa did not use nitrate as a nitrogen source (Table S1). Hence, D -arabitol production from various nitrogenous and organic nutrients was examined (Table 2). As yeast extract was the most suitable compound, the effect of yeast extract concentration on D -arabitol production was further investigated. The addition of 6 g/l yeast extract led to the maximum D -arabitol production (Fig. 2).
D-Arabitol
Optimization of culture conditions for producing D -arabitol by C. quercitrusa NBRC1022
10
Time (d)
Fig. 5 Large-scale production of D -arabitol from the raw glycerol medium by C. quercitrusa NBRC1022. Using a jar fermenter, the cultivation was performed with the raw glycerol medium with 2 g/l CaCl2. Symbols: D -arabitol (closed circle), glycerol (open circle), and dry cell (open triangle). A representative result of three independent experiments is shown
Appl Microbiol Biotechnol
resulted in the accumulation of 85.1 g/l D -arabitol; this corresponded to the yield of 0.40 g/g of glycerol consumed.
Discussion In this study, we obtained a novel producer of sugar alcohols from raw glycerol to create an efficient production system using non-edible feedstock. Generally, the accumulation of sugar alcohols by yeast has been proposed as a defense mechanism against osmotic stress (van Eck et al. 1989). Onishi and Suzuki (1970) reported that C. polymorpha produces 30.4 g/l of D -arabitol from 112 g/l glycerol after 6 days; this corresponded to a yield of 0.28 g/g of glycerol (volumetric productivity: 5.1 g/l/day). Koganti et al. (2011) demonstrated that Debaryomyces hansenii produces 15 g/l of D -arabitol from 150 g/l glycerol after 5 days. However, there have been no reports of D -arabitol production using>200 g/l glycerol. Here, we used 250 g/l glycerol. We isolated strain 17-2A as a novel D -arabitol producer, which was identified as C. quercitrusa. Comparative studies using related yeasts (Fig. S2) demonstrated that C. quercitrusa NBRC1022 was the best producer of D arabitol from raw glycerol. Interestingly, the known producers showed much lower D -arabitol production compared to that of the new producer. This suggests that the present screening method is effective in terms of identifying strains with the desired characteristics. C. quercitrusa NBRC1022 produced 58.2 g/l D -arabitol from 250 g/l glycerol after 7 days (Fig. 4) in flask culture, and its amount reached 67.1 g/l after 7 days in a jar fermenter (Fig. 5). This corresponded to a yield of 0.41 g/g of glycerol (volumetric productivity, 9.2 g/l/day). To our best knowledge, this is the highest D -arabitol yield from glycerol. More interestingly, the yeast showed growth and D -arabitol production even at 350 g/l glycerol (Fig. 3). C. quercitrusa NBRC1022 showed a good D -arabitol production in the presence of CaCl2. For microbial production of sugar alcohols such as erythritol and mannitol, the yields were increased by the addition of some minerals, including Ca2+ (Lee et al. 2000, 2007). Although the mechanism underlying this effect of Ca2+ remains unclear, Ca2+ is known to have many physiological activities. For example, in C. magnoliae, the addition of Ca2+ and Cu2+ improved mannitol production, presumably because Ca2+ alters the permeability of cells to mannitol and Cu2+ increases the activities of the enzymes responsible for mannitol biosynthesis (Lee et al. 2007). Likewise, Ca 2 + may improve the permeability of the C. quercitrusa membrane to D -arabitol. Further studies of the effect of Ca2+ addition on D -arabitol production from glycerol are warranted. Recently, Zygosaccharomyces rouxii was reported to produce 83.4 g/l of D -arabitol from 175 g/l of glucose; this
corresponded to a yield of 0.48 g/g of glucose (Saha et al. 2007). Kodamaea ohmeri produces 81.2 g/l of D -arabitol from 200 g/l glucose, which corresponds to 0.41 g/g of glucose (Zhu et al. 2010). Considering these yields, D -arabitol production by the present strain may be applicable for largescale production using non-edible feedstock. D -Arabitol may be applicable as a feedstock for xylitol production, which can be used as an alternative natural sweetener and for oral health care as it has a similar sweetness to sucrose and prevents caries (Dills 1989; Ohmori et al. 2004; van Loveren 2004). Xylitol is generally produced by the chemical or biotechnological reduction of D -xylose from hydrolyzed hemi-cellulose (Granström et al. 2004). However, D xylose is relatively expensive to prepare, and alternative costeffective methods are required. Microbial conversion of D arabitol to xylitol is an attractive process that does not require D -xylose, as it involves conversion of D -arabitol to D -xylulose by D -arabitol dehydrogenase (EC 1.1.1.11), and then to xylitol by xylitol dehydrogenase (EC 1.1.1.14; Suzuki et al. 2002). Accordingly, the newly identified yeast C. quercitrusa can be used for D -arabitol production and will contribute to the efficient production of xylitol.
References Barnett JA, Payne RW, Yarrow D (2000) Yeasts: Characteristics and identification, 3rd edn. Camridge University Press, Cambridge UK Bernard EM, Christiansen KJ, Tsang S, Kiehn TE, Armstrong D (1981) Rate of arabinitol production by pathogenic yeast species. J Clin Microbiol 14:189–194 Bisping B, Baumann U, Simmering R (1996) Effect of immobilization on polyol production by Pichia farinosa. Prog Biotechnol 11:395–401 Dills WL Jr (1989) Sugar alcohols as bulk sweeteners. Annu Rev Nutr 9: 161–186 Granström TB, Izumori K, Leisola M (2004) A rare sugar xylitol. Part I: the biochemistry and biosynthesis of xylitol. Appl Microbiol Biotechnol 74:277–281 Hajny GJ (1964) D-Arabitol production by Endomycopsis chodati. Appl Environ Microbiol 12:87–92 Koganti S, Kuo TM, Kurtzman CP, Smith N, Ju LK (2011) Production of arabitol from glycerol: strain screening and study of factors affecting production yield. Appl Microbiol Biotechnol 90:257–267 Kurtzman CP, Fell JW (1998) The yeasts, a taxonomic study, 4th edn. Elsevier, Amsterdam Netherlands Lee JK, Ha SJ, Kim SY, Oh DK (2000) Increased erythritol production in Torula sp. by Mn2+ and Cu2+. Biotechnol Lett 22:983–986 Lee JK, Oh DK, Song HY, Kim IW (2007) Ca2+ and Cu2+ supplementation increases mannitol production by Candida magnoliae . Biotechnol Lett 29:291–294 Nobre MF, da Costa MS (1985) Factors favouring the accumulation of arabinitol in the yeast Debaryomyces hansenii. Can J Microbiol 31: 467–471 Nozaki H, Suzuki S, Tsuyoshi N, Yokozeki K (2003) Production of Darabitol by Metschnikowia reukaufii AJ14787. Biosci Biotechnol Biochem 67:1923–1929 O’Donnell K (1993) Fusarium and its near relatives. In: Reynolds DR, Taylor JW (eds) The fungal holomorph: mitotic, meiotic and
Appl Microbiol Biotechnol pleomorphic speciation in fungal systematics. CAB International, Wallingford UK, pp 225–233 Ohmori S, Ohno Y, Makino T, Kashihara T (2004) Characteristics of erythritol and formulation of a novel coating with erythritol termed thin-layer sugarless coating. Int J Pharm 278:447–457 Onishi H, Suzuki T (1970) Production of erythritol, D-arabitol, Dmannitol and a heptitol-like compound from glycerol by yeasts. J Ferment Technol 48:563–566 Saha BC, Sakakibara Y, Cotta MA (2007) Production of D-arabitol by a newly isolated Zygosaccharomyces rouxii . J Ind Microbiol Biotechnol 34:519–523 Song W, Lin Y, Hu H, Xie Z, Zhang J (2011) Isolation and identification of a novel Candida sp. H2 producing D-arabitol and optimization of D-arabitol production. Wei Sheng Wu Xue Bao 51:332–339 Suzuki S, Sugiyama M, Mihara Y, Hashiguchi K, Yokozeki K (2002) Novel enzymatic method for the production of xylitol from Darabitol by Gluconobacter oxydans. Biosci Biotechnol Biochem 66:2614–2620 Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting. Position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680
van Eck JH, Prior BA, Brandt EV (1989) Accumulation of polyhydroxy alcohols by Hansenula anomala in response to water stress. J Gen Microbiol 135:3505–3513 van Loveren C (2004) Sugar alcohols: what is the evidence for caries-preventive and caries-therapeutic effects? Caries Res 38:286–293 Werpy T, Petersen G (2004) Top value added chemicals from biomass: volume I – results of screening for potential candidates from sugars and synthesis gas. DOE/GO-102004-1992. National Renewable Energy Laboratory. Willke T, Vorlop KD (2004) Industrial bioconversion of renewable resources as an alternative to conventional chemistry. Appl Microbiol Biotechnol 66:131–142 Yazdani SS, Gonzalez R (2007) Anaerobic fermentation of glycerol: a path to economic viability for the biofuels industry. Curr Opin Biotechnol 18:213–219 Yoda A (1951) A new method for detection of nonreducing sugars and polyalcohols on paper chromatography. J Chem Soc Japan (in Japanese) 73:18–19 Zhu HY, Xu H, Dai XY, Zhang Y, Ying HJ, Ouyang PK (2010) Production of D-arabitol by a newly isolated Kodamaea ohmeri. Bioprocess Biosyst Eng 33:565–571
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Production of D -arabitol from raw glycerol by Candida quercitrusa Jun Yoshikawa & Hiroshi Habe & Tomotake Morita & Tokuma Fukuoka & Tomohiro Imura & Hiroyuki Iwabuchi & Shingo Uemura & Takamitsu Tamura & Dai Kitamoto
Received: 19 September 2013 / Revised: 27 November 2013 / Accepted: 28 November 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract To promote the effective use of raw glycerol (a byproduct of biodiesel production), 110 yeast strains that produce D -arabitol from glycerol were isolated from environmental samples. Among them, strain 17-2A was an effective D arabitol producer in the presence of 250 g/l glycerol and was identified as Candida quercitrusa based on morphological, physicochemical, and phylogenetic analyses. C. quercitrusa type strain NBRC1022 produced the greatest quantity of D arabitol (41.7 g/l) when the ability to produce D -arabitol from raw glycerol was compared among C. quercitrusa 17-2A and its phylogenetically related strains in flask culture. Under optimized culture conditions, strain NBRC1022 produced D arabitol at a concentration of 58.2 g/l after a 7-day cultivation in 250 g/l glycerol, 6 g/l yeast extract, and 2 g/l CaCl2. The culture conditions were further investigated with raw glycerol using a jar fermenter; the concentration of D -arabitol reached 67.1 g/l after 7 days and 85.1 g/l after 10 days, respectively, which corresponded to 0.40 g/g of glycerol. To our knowledge, the present D -arabitol yield from glycerol is higher than reported previously using microbial production. Keywords D -arabitol . Candida quercitrusa . Raw glycerol . Sugar alcohol . Biodiesel Electronic supplementary material The online version of this article (doi:10.1007/s00253-013-5449-x) contains supplementary material, which is available to authorized users. J. Yoshikawa : H. Habe : T. Morita : T. Fukuoka : T. Imura : D. Kitamoto (*) Research Institute for Innovation in Sustainable Chemistry, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5-2, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan e-mail: [email protected] H. Iwabuchi : S. Uemura : T. Tamura Chemical Research Laboratories, Lion Corporation, 7-13-12 Hirai, Edogawa-ku, Tokyo 132-0035, Japan
Introduction Recently, the industrial use of renewable resources instead of petroleum has become important from an environmental standpoint (Willke and Vorlop 2004). Among some currently available resources such as starch, lignocelluloses, sugars, and organic wastes, plant oils have been used to produce biodiesel fuel (BDF) and some oleochemicals, including surfactants. Their production is based on an ester exchange reaction between triacylglycerol and alcohol (e.g., methanol) under alkaline conditions, resulting in the formation of a crude glycerol waste (raw glycerol). Due to the considerable increase in BDF production worldwide over the last few years, surplus glycerol derived from BDF production has become an alternative feedstock for production of chemicals and fuels (Yazdani and Gonzalez 2007). Since the quality of raw glycerol is usually low, refining processes to obtain pure glycerol with some cost and energy are required. Thus, the use of raw glycerol to produce value-added chemicals is important. Sugar alcohols such as erythritol, arabitol, xylitol, and mannitol are produced by osmophilic yeasts against the osmotic pressure across the cell membrane, and some sugar alcohols are used in the food and pharmaceutical industries. Among them, D -arabitol is a C5-sugar alcohol identified as one of the top 12 building block compounds based on biomass by the US Department of Energy (Werpy and Petersen 2004). Werpy and Petersen (2004) reported that some chemicals such as arabinoic acid, xylonic acid, propylene glycol and ethylene glycol can be produced from arabitol and its enantiomer, xylitol. In addition, D -arabitol can be used as a starting material for xylitol production, which has numerous applications in the food and pharmaceutical industries. Furthermore, D arabitol itself is expected to show potential applications like xylitol as an alternative low calorie sweetener (Koganti et al. 2011). Thus, these reports stress the importance of industrial production of D -arabitol.
Appl Microbiol Biotechnol
At this time, there are many reports of microbial production of D -arabitol by various osmophilic yeast species (Hajny 1964; Bernard et al. 1981; Nobre and da Costa 1985; van ECK et al. 1989; Bisping et al. 1996; Nozaki et al. 2003; Saha et al. 2007; Zhu et al. 2010; Song et al. 2011). However, to our knowledge, only two reports of D -arabitol production from glycerol, by Onishi and Suzuki (1970) and Koganti et al. (2011), have been published. In addition, there has been no report of microbial production of D -arabitol using raw glycerol as a feedstock. Raw glycerol frequently inhibits microbial growth because it contains impurities, such as non-glycerol organic matter (e.g., tar and methanol), sodium or potassium salts, and heavy metals. Hence, novel microbial strains suitable for producing D -arabitol from raw glycerol are required. In this study, we screened microbial strains suitable for producing D -arabitol from raw glycerol at high concentration. Among the tested yeast candidates, Candida quercitrusa strain NBRC1022 was selected as the best D -arabitol-producing strain in the presence of 250 g/l glycerol, and D -arabitol production from raw glycerol was optimized using a jar fermenter.
yeast extract, 3 g/l malt extract, 5 g/l peptone, 5 g/l Na2SO4, 50 mg/l streptomycin, and 50 mg/l chloramphenicol]. Environmental samples such as flowers, fruits, and leaves were added to test tubes containing 5 ml of the medium, followed by incubation on a reciprocal shaker (180 strokes/min) at 28 °C for 3 to 7 days. Then, 100 μl of each culture was inoculated into deep-well plates (96 well) containing 1 ml each of the same medium and was incubated on a reciprocal shaker (150 rpm) at 28 °C for 3 days. After assessment of D arabitol production using paper chromatography, positive cultures were appropriately diluted and spread on YM agar plates. The resultant colonies were again inoculated into deep-well plates and were assessed for their D -arabitol production by paper chromatography. Paper chromatography To detect D -arabitol production, 10 μl of each culture was applied to paper chromatography with 1-propanol/ethyl acetate/water (7:1:2, v/v) as a solvent system. Detection was then performed using Yoda’s reagent containing 5 g/l KIO4 and 150 g/l MnSO4/p,p'-tetramethyl diaminodiphenylmethane saturated with 2 M acetic acid (1:1, v/v; Yoda 1951).
Materials and methods High performance liquid chromatography Microorganisms and culture conditions A total of 10 type strains, including C. quercitrusa NBRC1022, Candida natalensis NBRC1981, Candida fragi NBRC10759, Debaryomyces hansenii var. fabryi NBRC0015, Debaryomyces hansenii var. hansenii N B R C 0 0 8 3 , M e t s c h n i k o w i a lu n a t a N B R C 1 6 0 5 , Metschnikowia pulcherrima NBRC1678, Metschnikowia reukaufii NBRC1679, Metschnikowia zobelli NBRC1680, and Hansenula anomala NBRC10213, were obtained from the Biological Resource Center, NITE (NBRC; Chiba, Japan). Strain 17-2A isolated in this study was deposited as FERM P-21420 at the NITE Patent Microorganisms Depositary (NPMD, Chiba, Japan). Each strain was maintained on YM agar plates [10 g/l glucose, 5 g/l peptone, 3 g/l yeast extract, 3 g/l malt extract, and 15 g/l agar]. Raw glycerol [composition, 850 g/l glycerol, 20 g/l organic impurities, 20 g/l inorganic impurities (Na2SO4 >5 g/l), 110 g/l H2O] was supplied from Lion Corporation, Japan. The raw glycerol was used as a substrate for D -arabitol production after passing through a column packed with granular activated charcoal to remove impurities. Screening of microorganisms producing D -arabitol from glycerol Screening of D -arabitol-producing microorganisms was performed using the enrichment medium [250 g/l glycerol, 3 g/l
To quantify D -arabitol and glycerol, high performance liquid chromatography (HPLC) analysis was performed using a SUGAR SC1011 column (8 mm×30 cm; Showa Denko, Tokyo, Japan) at 80 °C and a refractive index detector RID10A (Shimadzu, Kyoto, Japan). The column was eluted with H2O at 1 ml/min. The concentration of each product was determined from a calibration curve with the corresponding authentic reagents. Molecular phylogenetic analysis The cells of strain 17-2A were prepared by cultivation with YM medium at 25 °C. DNA extraction was performed using the DNeasy plant mini kit (QIAGEN, Hilden, Germany) according to the manufacturer’s protocol. The D1/D2 region of 26S ribosomal RNA gene was amplified by PCR using primers NL1, NL2, NL3, and NL4 (O’Donnell 1993) and puReTaq Ready-To-Go polymerase chain reaction (PCR) beads (GE Healthcare Life Sciences, Buckinghamshire, UK). After an initial denaturation at 95 °C for 7 min, 35 cycles were performed at 95 °C for 1 min, 55 °C for 1 min, and 72 °C for 1.5 min with the above primers. The program ended with a 7-min extension at 72 °C. Sequencing reactions were performed using the ABI PRISMTM BigDyeTM Terminator, v3.1 Kit (Applied Biosystems, CA, USA), as described by the manufacturer. The PCR products were sequenced using the ABI PRISMTM 3100 genetic analyzer system (Applied
Appl Microbiol Biotechnol
Biosystems). The sequence of the D1/D2 region of 26S ribosomal RNA gene in strain 17-2A was compared with closely related species obtained from the DNA Data Bank of Japan (DDBJ; http://www.ddbj.nig.ac.jp) using the BLAST search software and alignement using the Clustal W software (Thompson et al. 1994). The phylogenetic tree was visualized using the TreeView software with Candida parapsilosis as an outgroup. The nucleotide sequence for the D1/D2 region of strain 17-2A was registered in the DDBJ, EMBL, and GenBank nucleotide sequence databases with accession number AB855788.
or casamino acid was added to the basic culture medium A. For the effects of yeast extract concentration, 1, 3, 5, and 7 g/l yeast extract was added to the basic culture medium A. To explore the effects of initial glycerol concentrations, 150, 200, 300, and 350 g/l glycerol was replaced with 250 g/l glycerol in the basic culture medium B. To examine the effects of calcium addition, 0.5, 1, and 2 g/l calcium chloride was added to the basic culture medium B. The concentrations of D -arabitol and glycerol in cultures were determined by HPLC. All data shown are the averages and standard deviations from three independent experiments.
Physiological test
D -Arabitol production from raw glycerol using a jar fermenter
For strain 17-2A, the assimilation test for several carbon and nitrogen compounds, resistance tests, vitamin auxotrophic tests, urease tests, and diazonium blue B color (DBB) tests were performed as described previously (Kurtzman and Fell 1998; Barnett et al. 2000). Morphological characteristics of the microorganism were observed using a BX51 optical microscope (Olympus, Tokyo).
A single colony of C. quercitrusa NBRC1022 was inoculated into a test tube containing 5 ml of the seed culture medium. After incubation on a reciprocal shaker (180 strokes/min) at 28 °C for 1 day, 1 ml of the pre-culture was inoculated into a 300-ml Erlenmeyer flask containing 30 ml of the same medium and further incubated on a rotary shaker (250 rpm) at 28 °C for 1 day. A total of 60 ml of the culture (two flasks) was inoculated into a 1.6 l-jar fermenter containing 800 ml of the raw glycerol medium with 2 g/l CaCl2. During the jar fermenter experiments, the aeration rate and agitation speed were set to 1 vvm and 800 rpm, respectively. Temperature was maintained at 28±1 °C. The concentrations of D -arabitol and glycerol in the culture were analyzed by HPLC. Three independent experiments were carried out.
Production of D -arabitol by newly isolated strains and related strains A single colony of each strain on YM plates was inoculated into a test tube containing 5 ml of the seed culture medium [10 g/l glycerol, 3 g/l yeast extract, 3 g/l malt extract, and 5 g/l peptone]. After incubation on a reciprocal shaker (180 strokes/ min) at 28 °C for 1 day, 1 ml of culture was inoculated into 300-ml Erlenmeyer flask containing 30 ml of the raw glycerol medium [300 g/l raw glycerol, 6 g/l yeast extract, 1 g/l KH2PO4; pH 6.0], unless otherwise noted. The culture was further incubated on a rotary shaker (250 rpm) at 28 °C for 7 days. The concentrations of D -arabitol and glycerol in cultures were determined by HPLC. Optimization of culture conditions for D -arabitol production in flask culture A single colony of C. quercitrusa NBRC1022 was inoculated into a test tube containing 5 ml of the seed culture medium. After incubation on a reciprocal shaker (180 strokes/min) at 28 °C for 1 day, 1 ml of the pre-culture was inoculated into a 300-ml Erlenmeyer flask containing 30 ml of the basic culture medium A [250 g/l glycerol, 1 g/l yeast extract, 1 g/l KH2PO4, and 5 g/l Na2SO4; pH 6.0] and the basic culture medium B [250 g/l glycerol, 6 g/l yeast extract, 1 g/l KH2PO4, and 5 g/l Na2SO4; pH 6.0] with some modifications (described below), followed by another incubation on a rotary shaker (250 rpm) at 28 °C for 7 days. To evaluate the effect of nitrogenous and organic nutrients, 5 g/l of yeast extract, corn steep liquor, peptone, malt extract
Results Screening of microorganisms producing D -arabitol from glycerol To screen D -arabitol-producing microorganisms using deepwell plates, an increased concentration of glycerol (250 g/l) was used to promote the use of surplus glycerol. Also, 5 g/l Na2SO4 was added to the medium because raw glycerol used in this study contains the corresponding concentration of Na2SO4. The 2,300 microbial strains that grew on 250 g/l glycerol have been isolated from environmental samples around Japan such as flowers, fruits, and leaves. The isolated strains were investigated for their production of sugar alcohols using paper chromatography. In this solvent system, production of C5-, C6-, or both sugar alcohols was observed, but detecting C4-sugar alcohols such as erythritol was problematic because of a huge spot of C3-glycerol (Fig. 1A). Among these sugar alcohol-producers, 110 strains were predicted to produce C5-sugar alcohols by paper chromatography (an example is shown in Fig. 1B); their D -arabitol production was then assessed by HPLC (Fig. 1C). All C5-sugar alcohols produced by the 110 strains were determined to be D -arabitol,
Appl Microbiol Biotechnol
a
b
Table 1 Comparison of D -arabitol production from the raw glycerol medium by related yeast strains
C3 C5 C6
1
2
3
4
5
6
7
8
9
Strain
D -Arabitol
C. quercitrusa 17-2A C. quercitrusa NBRC1022 C. natalensis NBRC1981 C. fragi NBRC10759 D. hansenii var. fabryi NBRC0015 D. hansenii var. hansenii NBRC0083 M. reukaufii NBRC1679 H. anomala NBRC10213
40.7 41.7 ND ND 9.75 26.5 ND 15.6
(g/l)
ND not detected
c Glycerol (11.6 min)
7.5
D-Arabitol (12.4 min)
10.0 12.5 15.0 Retention time (min)
17.5
Fig. 1 Detection and quantification of D -arabitol in microbial culture. An example of a paper chromatography profile: standard samples (A) and microbial culture (B). An example of an HPLC profile: culture broth of strain 17-2A (C). The cultivation was performed in the enrichment medium. Paper chromatography was performed using 1-propanol/ethyl acetate/water (7:1:2, v/v) as a solvent system. Detection was performed by spraying 150 g/l MnSO4/p,p′-tetramethyl diaminodiphenylmethane saturated by 2 M acetic acid (1:1, v/v) after spraying 5 g/l KIO4. Lanes 1, glycerol; 2, erythritol; 3, xylitol; 4, D -arabitol; 5, sorbitol; 6, mannitol; 7, strain 15-11E; 8, stain 17-2A; 9, strain 17-12D. C3, C5, and C6 represent spot areas of 3-, 5-, and 6-carbon sugar alcohols, respectively
and their concentrations ranged from 3.4 to 12.7 g/l when cultured using the enrichment medium in 96deep-well plates. The isolated 110 strains were tested for D -arabitol production with the raw glycerol medium in flask culture. Strain 172A showed the highest D -arabitol concentration, 40.7 g/l. Strains 15-11E and 17-12D were the second (31.4 g/l) and third (14.0 g/l) highest D -arabitol producers, respectively. Thus, we next characterized strain 17-2A.
databases. The D1/D2 region of the 26S rRNA gene sequence was similar to Candida quercitrusa CBS101427 [99.8 % (564/565 bp)] and C. natalensis [98.1 % (555/566 bp)]. Physiological characteristics of 17-2Awere similar to those observed for C. quercitrusa, excluding the assimilation of D -ribose and L -arabitol, growth at 37 °C, and vitamin auxotrophy (Table S1 in supplementary materials). Based on morphological and physiological properties (Fig. S1 and Table S1; Kurtzman and Fell 1998) and molecular phylogenetic analysis using the D1/ D2 region of 26S rRNA genes (Fig. S2 in supplementary materials), we identified strain 17-2A as C. quercitrusa. Comparison of D -arabitol production among related yeast strains C. quercitrusa 17-2A was an efficient D -arabitol producer, but this species had not been reported previously to produce D arabitol. Thus, D -arabitol production was compared between strain 17-2A and the type strain C. quercitrusa . For the related type species shown in Fig. S2 (C. natalensis , C. fragi and D. hansenii) and the known D -arabitol-producing yeast from glucose (Metschnikowia reukaufii and Hansenula anomala), D -arabitol production with the raw glycerol medium was also investigated (Table 1). C. quercitrusa NBRC1022 showed the highest production (41.7 g/l), while C. natalensis NBRC1981, C. fragi NBRC10759 and M. reukaufii NBRC1679 did not produce D -arabitol. Although D. hansenii var. hansenii NBRC0083 and
Characterization of strain 17-2A After a 4-day incubation at 25˚C on YM agar plates, we assessed the morphology of strain 17-2A (Fig. S1 in supplementary materials). The cells were elongate (2.5–17.5)×(2–3.5)μm, and grew normally with polar budding. Pseudomycelia developed, and no sexual reproduction was observed within 30 days. To identify the yeast strain 17-2A, the D1/D2 region of 26S rRNA gene was determined and aligned with related available sequences from public
Table 2 Effect of nitrogenous and organic nutrients added to the basic medium A on D -arabitol production by C. quercitrusa NBRC1022 ND not detected
Nutrient (5 g/l)
D -Arabitol
Yeast extract Corn steep liquor Peptone
42.5 21.6 3.69
Malt extract Casamino acids
11.0 ND
(g/l)
60
50
50 (g/L)
60
40
D-Arabitol
D-Arabitol
(g/L)
Appl Microbiol Biotechnol
30 20
30 20 10
10 0
40
0
2
4 6 Yeast extract (g/L)
0
8
0.5
1
2
Calcium chloride (g/L)
Fig. 2 Effect of yeast extract concentration on D -arabitol production by C. quercitrusa NBRC1022. Cultivation was performed in the basic culture medium A with 2 to 8 g/l final yeast extract concentration. The vertical bar shows the standard deviation of the mean of three independent experiments
Fig. 4 Effect of calcium chloride concentration on D -arabitol production by C. quercitrusa NBRC1022. Cultivation was performed in the basic culture medium B with 0 to 2 g/l CaCl2. The vertical bar shows the standard deviation of the mean of three independent experiments
H. anomala NBRC10213 could produce D -arabitol from raw glycerol, the concentrations were considerably lower than that of C. quercitrusa. When reproducibility of D -arabitol production was compared between strains NBRC1022 and 17-2A, we confirmed that strain NBRC1022 gave about 10 % to 20 % larger amount of D -arabitol was produced (data not shown). Hence, C. quercitrusa NBRC1022 was selected as the optimum strain and was used in the following studies.
Figure 3 shows the effect of initial glycerol concentration on D -arabitol production; 250–300 g/l initial glycerol resulted in improved production compared to lower glycerol concentrations. Surprisingly, D -arabitol production occurred in the presence of relatively high osmotic stress at 350 g/l glycerol. We preliminary examined the effect of various minerals addition (Mn2+, Ca2+, Fe2+, and Zn2+) on D -arabitol production; only Ca2+ had positive effects, while most minerals inhibited D -arabitol production (data not shown). D -Arabitol production increased upon addition of more than 0.5 g/l CaCl2; the D -arabitol concentration was 58.2 g/l in the presence of 2 g/l CaCl2 (Fig. 4). Large-scale production of D -arabitol from raw glycerol by C. quercitrusa NBRC1022 Based on the above optimization experiments, large-scale production of D -arabitol from raw glycerol by C. quercitrusa NBRC1022 was examined using a jar fermenter (Fig. 5). After a 7-day cultivation, the concentration was 67.1 g/l (the yield was 0.41 g/g of glycerol). Further cultivation up to 10 days 280
100 90 80 70 60 50 40 30 20 10 0
240 Glycerol (g/L)
60
D-Arabitol
(g/L)
50 40
200 160
120
30
80
20
40
10
0 0
0 150
200
250 300 Glycerol (g/L)
350
Fig. 3 Effect of glycerol concentration on D -arabitol production by C. quercitrusa NBRC1022. Cultivation was performed in the basic culture medium B with 150 to 350 g/l glycerol. The vertical bar shows the standard deviation of the mean of three independent experiments
1
2
3
4
5
6
7
8
9
and dry cell (g/L)
Our preliminary studies showed that C. quercitrusa NBRC1022 stably produced D -arabitol at 28 °C in the pH range 5 to 7. To optimize the culture conditions for D -arabitol production, the basic culture media A or B (pH 6) were used. C. quercitrusa did not use nitrate as a nitrogen source (Table S1). Hence, D -arabitol production from various nitrogenous and organic nutrients was examined (Table 2). As yeast extract was the most suitable compound, the effect of yeast extract concentration on D -arabitol production was further investigated. The addition of 6 g/l yeast extract led to the maximum D -arabitol production (Fig. 2).
D-Arabitol
Optimization of culture conditions for producing D -arabitol by C. quercitrusa NBRC1022
10
Time (d)
Fig. 5 Large-scale production of D -arabitol from the raw glycerol medium by C. quercitrusa NBRC1022. Using a jar fermenter, the cultivation was performed with the raw glycerol medium with 2 g/l CaCl2. Symbols: D -arabitol (closed circle), glycerol (open circle), and dry cell (open triangle). A representative result of three independent experiments is shown
Appl Microbiol Biotechnol
resulted in the accumulation of 85.1 g/l D -arabitol; this corresponded to the yield of 0.40 g/g of glycerol consumed.
Discussion In this study, we obtained a novel producer of sugar alcohols from raw glycerol to create an efficient production system using non-edible feedstock. Generally, the accumulation of sugar alcohols by yeast has been proposed as a defense mechanism against osmotic stress (van Eck et al. 1989). Onishi and Suzuki (1970) reported that C. polymorpha produces 30.4 g/l of D -arabitol from 112 g/l glycerol after 6 days; this corresponded to a yield of 0.28 g/g of glycerol (volumetric productivity: 5.1 g/l/day). Koganti et al. (2011) demonstrated that Debaryomyces hansenii produces 15 g/l of D -arabitol from 150 g/l glycerol after 5 days. However, there have been no reports of D -arabitol production using>200 g/l glycerol. Here, we used 250 g/l glycerol. We isolated strain 17-2A as a novel D -arabitol producer, which was identified as C. quercitrusa. Comparative studies using related yeasts (Fig. S2) demonstrated that C. quercitrusa NBRC1022 was the best producer of D arabitol from raw glycerol. Interestingly, the known producers showed much lower D -arabitol production compared to that of the new producer. This suggests that the present screening method is effective in terms of identifying strains with the desired characteristics. C. quercitrusa NBRC1022 produced 58.2 g/l D -arabitol from 250 g/l glycerol after 7 days (Fig. 4) in flask culture, and its amount reached 67.1 g/l after 7 days in a jar fermenter (Fig. 5). This corresponded to a yield of 0.41 g/g of glycerol (volumetric productivity, 9.2 g/l/day). To our best knowledge, this is the highest D -arabitol yield from glycerol. More interestingly, the yeast showed growth and D -arabitol production even at 350 g/l glycerol (Fig. 3). C. quercitrusa NBRC1022 showed a good D -arabitol production in the presence of CaCl2. For microbial production of sugar alcohols such as erythritol and mannitol, the yields were increased by the addition of some minerals, including Ca2+ (Lee et al. 2000, 2007). Although the mechanism underlying this effect of Ca2+ remains unclear, Ca2+ is known to have many physiological activities. For example, in C. magnoliae, the addition of Ca2+ and Cu2+ improved mannitol production, presumably because Ca2+ alters the permeability of cells to mannitol and Cu2+ increases the activities of the enzymes responsible for mannitol biosynthesis (Lee et al. 2007). Likewise, Ca 2 + may improve the permeability of the C. quercitrusa membrane to D -arabitol. Further studies of the effect of Ca2+ addition on D -arabitol production from glycerol are warranted. Recently, Zygosaccharomyces rouxii was reported to produce 83.4 g/l of D -arabitol from 175 g/l of glucose; this
corresponded to a yield of 0.48 g/g of glucose (Saha et al. 2007). Kodamaea ohmeri produces 81.2 g/l of D -arabitol from 200 g/l glucose, which corresponds to 0.41 g/g of glucose (Zhu et al. 2010). Considering these yields, D -arabitol production by the present strain may be applicable for largescale production using non-edible feedstock. D -Arabitol may be applicable as a feedstock for xylitol production, which can be used as an alternative natural sweetener and for oral health care as it has a similar sweetness to sucrose and prevents caries (Dills 1989; Ohmori et al. 2004; van Loveren 2004). Xylitol is generally produced by the chemical or biotechnological reduction of D -xylose from hydrolyzed hemi-cellulose (Granström et al. 2004). However, D xylose is relatively expensive to prepare, and alternative costeffective methods are required. Microbial conversion of D arabitol to xylitol is an attractive process that does not require D -xylose, as it involves conversion of D -arabitol to D -xylulose by D -arabitol dehydrogenase (EC 1.1.1.11), and then to xylitol by xylitol dehydrogenase (EC 1.1.1.14; Suzuki et al. 2002). Accordingly, the newly identified yeast C. quercitrusa can be used for D -arabitol production and will contribute to the efficient production of xylitol.
References Barnett JA, Payne RW, Yarrow D (2000) Yeasts: Characteristics and identification, 3rd edn. Camridge University Press, Cambridge UK Bernard EM, Christiansen KJ, Tsang S, Kiehn TE, Armstrong D (1981) Rate of arabinitol production by pathogenic yeast species. J Clin Microbiol 14:189–194 Bisping B, Baumann U, Simmering R (1996) Effect of immobilization on polyol production by Pichia farinosa. Prog Biotechnol 11:395–401 Dills WL Jr (1989) Sugar alcohols as bulk sweeteners. Annu Rev Nutr 9: 161–186 Granström TB, Izumori K, Leisola M (2004) A rare sugar xylitol. Part I: the biochemistry and biosynthesis of xylitol. Appl Microbiol Biotechnol 74:277–281 Hajny GJ (1964) D-Arabitol production by Endomycopsis chodati. Appl Environ Microbiol 12:87–92 Koganti S, Kuo TM, Kurtzman CP, Smith N, Ju LK (2011) Production of arabitol from glycerol: strain screening and study of factors affecting production yield. Appl Microbiol Biotechnol 90:257–267 Kurtzman CP, Fell JW (1998) The yeasts, a taxonomic study, 4th edn. Elsevier, Amsterdam Netherlands Lee JK, Ha SJ, Kim SY, Oh DK (2000) Increased erythritol production in Torula sp. by Mn2+ and Cu2+. Biotechnol Lett 22:983–986 Lee JK, Oh DK, Song HY, Kim IW (2007) Ca2+ and Cu2+ supplementation increases mannitol production by Candida magnoliae . Biotechnol Lett 29:291–294 Nobre MF, da Costa MS (1985) Factors favouring the accumulation of arabinitol in the yeast Debaryomyces hansenii. Can J Microbiol 31: 467–471 Nozaki H, Suzuki S, Tsuyoshi N, Yokozeki K (2003) Production of Darabitol by Metschnikowia reukaufii AJ14787. Biosci Biotechnol Biochem 67:1923–1929 O’Donnell K (1993) Fusarium and its near relatives. In: Reynolds DR, Taylor JW (eds) The fungal holomorph: mitotic, meiotic and
Appl Microbiol Biotechnol pleomorphic speciation in fungal systematics. CAB International, Wallingford UK, pp 225–233 Ohmori S, Ohno Y, Makino T, Kashihara T (2004) Characteristics of erythritol and formulation of a novel coating with erythritol termed thin-layer sugarless coating. Int J Pharm 278:447–457 Onishi H, Suzuki T (1970) Production of erythritol, D-arabitol, Dmannitol and a heptitol-like compound from glycerol by yeasts. J Ferment Technol 48:563–566 Saha BC, Sakakibara Y, Cotta MA (2007) Production of D-arabitol by a newly isolated Zygosaccharomyces rouxii . J Ind Microbiol Biotechnol 34:519–523 Song W, Lin Y, Hu H, Xie Z, Zhang J (2011) Isolation and identification of a novel Candida sp. H2 producing D-arabitol and optimization of D-arabitol production. Wei Sheng Wu Xue Bao 51:332–339 Suzuki S, Sugiyama M, Mihara Y, Hashiguchi K, Yokozeki K (2002) Novel enzymatic method for the production of xylitol from Darabitol by Gluconobacter oxydans. Biosci Biotechnol Biochem 66:2614–2620 Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting. Position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680
van Eck JH, Prior BA, Brandt EV (1989) Accumulation of polyhydroxy alcohols by Hansenula anomala in response to water stress. J Gen Microbiol 135:3505–3513 van Loveren C (2004) Sugar alcohols: what is the evidence for caries-preventive and caries-therapeutic effects? Caries Res 38:286–293 Werpy T, Petersen G (2004) Top value added chemicals from biomass: volume I – results of screening for potential candidates from sugars and synthesis gas. DOE/GO-102004-1992. National Renewable Energy Laboratory. Willke T, Vorlop KD (2004) Industrial bioconversion of renewable resources as an alternative to conventional chemistry. Appl Microbiol Biotechnol 66:131–142 Yazdani SS, Gonzalez R (2007) Anaerobic fermentation of glycerol: a path to economic viability for the biofuels industry. Curr Opin Biotechnol 18:213–219 Yoda A (1951) A new method for detection of nonreducing sugars and polyalcohols on paper chromatography. J Chem Soc Japan (in Japanese) 73:18–19 Zhu HY, Xu H, Dai XY, Zhang Y, Ying HJ, Ouyang PK (2010) Production of D-arabitol by a newly isolated Kodamaea ohmeri. Bioprocess Biosyst Eng 33:565–571
