Polyalcohol arabitol can be used in the food and pharmaceutical industries as a natural sweetener, a dental caries reducer, and texturing agent. Environmental samples were screened to isolate effective yeast producers of arabitol. The most promising isolate 27RL-4, obtained from raspberry leaves, was identified genetically and biochemically as Candida parapsilosis. It secreted 10.42– 10.72 g l-1 of product from 20 g l-1 of L-arabinose with a yield of 0.51 - 0.53 g g-1 at 28°C and a rotational speed of 150 rpm. Batch cultures showed that optimal pH value for arabitol production was 5.5. High yields and productivities of arabitol were obtained during incubation of the yeast at 200 rpm, or at 32°C, but the concentrations of the polyol did not exceed 10 g l-1. In modified medium, with reduced amounts of nitrogen compounds and pH 5.5-6.5, lower yeast biomass produced a similar concentration of arabitol, suggesting higher efficiency of yeast cells. This strain also produced arabitol from glucose, with much lower yields. The search for new strains able to successfully produce arabitol is important for allowing the utilization of sugars abundant in plant biomass.
L-arabitol is a pentitol which, together with its enantiomer xylitol, has been included on the list of top 12 biomass-derivable chemicals designated for further research within biotechnology . Arabitol can be used as a natural sweetener, a dental caries reducer, and a sugar substitute for diabetic patients due to its low caloric (only 0.2 kcal g-1), low-glycemic, low-insulinemic, anticariogenic, and prebiotic effects . It also reduces adipose tissue in the body and prevents fat deposition in the digestive tract . This polyol can be used in the food and pharmaceutical industries as a texturing agent, a humectant, a softener, and a color stabilizer. Currently, arabitol is produced chemically and requires an expensive catalyst, a high temperature, and chromatographic purification steps . Biotechnological production of arabitol from L-arabinose or other sugars obtained from hemicellulosic hydrolyzates of the plant biomass, by yeasts such as Debaryomyces, Candida, Pichia, Wickerhamomyces, and Saccharomycopsis represents an efficient and cost-effective alternative to chemical production [2, 5].
Screening of yeast culture collections has revealed yeast strains able to produce arabitol from L-arabinose as reported by several research groups [5-8]. McMillan and Boynton  showed that L-arabinose was converted to arabitol by the yeasts Candida shehatae, C. tropicalis, Pichia stipitis, Pachysolen tannophilus, and Torulopsis sonorensis. The best producer of arabitol was C. tropicalis, which gave a yield of 1.02 g g-1 during cultivation in a medium containing a yeast nitrogen base and arabinose . Saha and Bothast  observed that C. entomaea NRRL Y-7785 and P. guilliermondii NRRL Y-2075 were superior producers of L-arabitol (a yield of about 0.7 g g-1). In experiments conducted by Fonseca et al. , P. guilliermondii PYCC 3012 secreted considerable amounts of arabitol during cultivation in an L-arabinose containing medium under oxygen-limited conditions giving yields in the range of 0.21–0.58 g g-1. Kordowska-Wiater et al.  reported that C. parapsilosis DSM 70125 and C. shehatae ATCC 22984 efficiently produced arabitol from L-arabinose with yields of 0.28 - 0.78 g g-1 and 0.25 - 0.5 g g-1, respectively.
There are also some reports on the screening of yeasts from environmental samples. Watanabe et al.  selected strain NY7122, related to Candida subhashii, which was able to produce L-arabitol and ethanol from L-arabinose. Kumdam et al.  determined that Debaryomyces nepalensis, isolated from rotten apple, produced arabitol in batch cultures with a yield of 0.26 g g-1. In a study by Bura et al. , the endophytic strain Rhodotorula mucilaginosa PTD3, was found to be capable of producing arabitol from arabinose in batch cultures with a yield of 0.2 g g-1.
The additional advantage of some yeasts, enhancing productive efficiency, is their ability to metabolise glucose to arabitol. An overview of these species was published by Kordowska-Wiater . Among presented yeasts there were effective polyol producers such as Metchnikowia reukaufii (yield of 0.52 g g-1) and Zygosaccharomyces rouxii (yield of 0.48 g g-1), as well as weak producers e.g. Debaryomyces nepalensis (0.02 g g-1). The screening of strains from various habitats, able to successfully produce arabitol, allows for the effective utilization of sugars abundant in plant biomass.
In the present study, plant and soil samples were screened to isolate strains of yeasts able to efficiently produce arabitol from L-arabinose and additionally from D-glucose. This work describes the isolation of the yeasts, the characterization and molecular identification of the most efficient strain and arabitol production efficiency in batch cultures under different conditions.
2.1 Yeast strains
One hundred yeast strains were isolated in Poland from environmental samples such as rotten wood, soil, plant leaves, flowers and grain stalks. C. parapsilosis DSM 70125 was obtained from the German Collection of Microorganisms and Cell Cultures. All strains were maintained on YPG agar slants (yeast extract 10 g l-1, peptone 10 g l-1, glucose 20 g l-1) (BTL, Łódź, Poland) at 4°C. They were deposited in the Culture Collection of the Department of Biotechnology, Microbiology and Human Nutrition, University of Life Sciences in Lublin, Poland.
2.2 Yeast isolation
Five grams of each environmental sample were added to Erlenmeyer flasks with 45 ml of physiological saline, and shaken on a reciprocal shaker for 10 min. The suspensions obtained were decimally diluted and plated onto Petri dishes with YGC agar (yeast extract 5 g l-1, glucose 20 g l-1, chloramphenicol 0.1 g l-1, agar 15 g l-1) (BTL, Łódź, Poland). The dishes were incubated at 28°C for 5 days. The single colonies of yeasts were transferred separately to a new medium and incubated. This purification procedure was repeated twice. Each colony was confirmed microscopically, then inoculated into an agar slant and incubated at 28°C for 3 days.
2.3 Screening of arabinose assimilating yeasts
Screening was performed using a modified method by Subtil and Boles . Briefly, one milliliter of sterile saline was inoculated with a yeast strain and decimally diluted twice. Ten microliters of each dilution was dropped onto Petri dishes with selective agar (yeast nitrogen base 6.7, L-arabinose 20.0 and agar 20.0 g l-1). After incubation at 28°C for 5 days the strains which grew on this medium were selected for further stages.
2.4 Biotransformation of L-arabinose to L-arabitol
The inoculation medium was composed of (at g l-1): L-arabinose 20.0, yeast extract 3.0, malt extract 3.0, (NH4)2SO4 5.0, and KH2PO4 3.0, pH 5.5. The cultivation medium was composed of (at g l-1): L-arabinose 20.0 (or 32.5, 50.0, and 80.0 in experiments investigating the effect of sugar concentration), yeast extract 3.0, malt extract 3.0, (NH4)2SO4 5.0 and KH2PO4 3.0. The pH was adjusted to 5.5, except for the study of pH effect, when it was adjusted to 4.5, 5.0, or 6.0 respectively. The modified cultivation medium with lower amounts of nitrogen sources was also composed (L-arabinose 20.0, yeast extract 2.0, malt extract 3.0, (NH4)2SO4 1.0, and KH2PO4 3.0 g l-1) and initial pH 5.5 was maintained by the addition of sterile CaCO3 (2.5 g l-1) after one day of incubation. The medium for inoculum preparation was dispensed into tubes (5 ml per tube), and the medium for yeast cultivation was dispensed into 100-ml Erlenmeyer flasks (50 ml per flask) and sterilized.
Inocula were prepared by transferring a loopful of cells from a slant into the tubes with the medium and incubated at 28°C for 24 h. Then, the cultivation medium was inoculated with 2% (v v-1) inoculum and incubated in a rotary shaker (Infors HT Minitron, Infors AG, Bottmingen, Switzerland) at 150 rpm and 28°C for 4 days, for most of the experiments. During the study of the effect of rotational speed on arabitol production, speeds of 100 or 200 rpm were used, and during temperature optimization experiments, the temperature was set to 24 or 32°C. Every 24 h, samples were collected to analyze pH, biomass concentration, L-arabinose utilization, and production of arabitol and, possibly ethanol. The biotransformation experiment for the positive strains was performed in triplicate.
2.5 Production of arabitol from glucose
The inoculation medium was composed of (at g l-1): D-glucose 50.0, yeast extract 3.0, malt extract 3.0, (NH4)2SO4 2.0, and KH2PO4 3.0. The cultivation medium had the same composition with the concentrations of glucose 50 or 100 g l-1. The pH was adjusted to 5.5. Inoculation and cultivation media were dispensed into tubes (5 ml) or 100-ml Erlenmeyer flasks (50 ml per flask), respectively, and sterilized. Inoculum and productive cultures were prepared as above and incubated for 5 days in a rotary shaker at 28°C, 150 rpm. Every 24 h, an analysis of concentrations of biomass, D-glucose, arabitol, glycerol, ethanol, ribitol and pH was carried out. The cultivation experiment was performed in triplicate.
2.6 Analysis of cell and colony morphology
For macroscopic morphology observations, the strains were grown on YPG agar (BTL, Łódź, Poland) at 28°C for 2–4 days. Intravital microscopic imaging was performed using the optical microscope Delta Optical Evolution 300 (Delta Optical, Poland) equipped with an HDCE-50B camera and ScopeImage Dynamic Pro (Delta Optical, Poland) software under 1000× magnification. Then the colonies and cells were characterized according to Kurtzman et al. .
2.7 Biochemical characterization and identification by the Yeast Identification Test Panel
The yeasts were identified biochemically using the Biolog System ™ (Biolog YT MicroPlate™, Biolog Inc., Hayward, CA, USA) by studying their ability to assimilate 59 and oxidate 35 substrates as carbon sources. The system was equipped with a multichannel pipette, a computer-linked absorbance and turbidity growth reader and Biolog Microlog System 3 (5.2) software. Before use, yeast strains were cultured onto plates of Biolog Universal Yeast Agar (BUY™) (Biolog Inc., Hayward, CA, USA) and incubated at 25°C for 2 days. Cells were suspended in sterile water at a specified density (47% T). After inoculation, the Biolog YT MicroPlates were incubated at 27°C for 72 h with optical density (OD) measuring at 590 nm every day. For identification, the MicroPlates were read with the MicroStation ™ and compared to the biochemical pattern of the YT database. The intensity of assimilation of each substrate was recorded as positive (+), negative (-) or partial response (+/-).
Percentage of total carbon source utilization following oxidation tests (%) was calculated using OD data for each well, corrected by subtracting the blank well values. The carbon substrates were divided into groups: amino acids, carbohydrates, carboxylic acids, polymers, polyalcohols, and miscellaneous. For each series, the corrected absorbance values of a particular group of substrates were summarized and expressed as a percentage of the average absorbance value of the corresponding substrate group for a selected yeast strain.
2.8 Genetic identification
DNA from 20 mg of yeast cells grinded by pestle in liquid nitrogen to a fine powder was obtained using the Plant and Fungi DNA Purification Kit (EURx, Gdańsk, Poland). ITS1 (5-TCCGTAGGTGAACCTGCGG-3’) and ITS4 (5’-TCCTCCGCTTATTGATATGC-3’) primers were used, as described by White et al. . PCR reactions were run in 25 μl volume using 2 × PCR Master Mix (Thermo Scientific Fermentas, Vilnius, Lithuania) with 20 pmol of each primer and 20 ng of DNA on a SensoQuestLabcycler (SensoQuest GmbH, Göttingen, Germany). Thermal cycling conditions were as follows: an initial step at 95°C for 3 min, 40 cycles at 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s followed by 72°C for 8 min. Amplification products were visualized using the GelDoc 2000 gel documentation system (Bio-Rad Laboratories Inc., Hercules, CA, USA).
The purified PCR products obtained using ITS1 and ITS4 primers were directly sequenced with an ABI 3730xl Genetic Analyzer at Genomed, Poland, using the Big dye Terminator method. The DNA sequence obtained was compared with the sequences reported in the NCBI GenBank database and then submitted to the NCBI GenBank® under accession number KP783504.
2.9 Analytical methods
2.9.1 Sugars and polyols analysis
Samples of cultures were centrifuged at 9000 rpm for 15 min. In each supernatant, after deproteinization by acetonitrile, the concentrations of sugars and polyols were determined by HPLC (Gilson Inc., Middleton, WI, USA) equipped with an autosampler, a refractive index detector (Knauer GmbH, Berlin, Germany) and a Bio-Rad Aminex Carbohydrate HPX 87H column (Bio-Rad Laboratories Inc., Hercules, CA, USA). 0.05 M sulfuric acid was used as the mobile phase at a flow rate of 0.5 ml min-1. The temperature of separation was 42°C. Chromatograms were integrated and analyzed using Chromax 2007 software version 1.0a (Pol-lab, Poland). Arabitol (glycerol, ribitol) yield was calculated as grams of product per grams of sugar consumed. The productivity was calculated as yield per hour.
2.9.2 Biomass and pH analysis
The biomass was determined by measuring OD at 590 nm using a BioRad Smart Spec Plus spectrophotometer (Bio-Rad Laboratories Inc., Hercules, CA, USA). The samples containing calcium salts were first treated by 10% (v v-1) HCl. Then, the relationship between OD and dry cell weight was calculated using a calibration curve. The biomass yield was calculated as grams of dry cell weight per grams of sugar consumed. The pH of cultures was monitored using an electronic pH-meter (Hanna Instruments, Poland).
2.9.3 Ethanol analysis
GC/MS analysis of the supernatants after deproteinization was carried out using a gas chromatograph GC2010 (Shimadzu, Tokyo, Japan) coupled with an MS-EI apparatus QP 2010Plus and an auto injector AOC-20i. The gas chromatograph was equipped with a 25 m × 0.32 mm (0.3 μm film thickness) CP-WAX 57 CB (Agilent J&W) column operated in the splitless mode with the valve closed for 0.3 min. The carrier gas was helium, used at a flow rate of 1.8 ml min-1. The injector and detector temperature was 200°C. The oven temperature was set to 50°C for 6 min, then increased at a ramp rate of 3°C min-1 to 120°C, maintained for 3 min, and finally increased at a ramp rate of 15°C min-1 to 190°C, and maintained for 2 min. Data were acquired in the SCAN-mode (20–400 m/z). GC peaks were identified by comparing the MS fragmentation pattern and the relative retention time with those of the reference compounds.
2.10 Statistical analysis
Data obtained from arabitol production by yeast were expressed as mean ± standard deviation. Differences among mean values of arabitol concentrations, yields, and productivities for optimization experiments were tested for statistical significance at the p < 0.05 level using analysis of variance and Fisher’s test within univariate groups (STATISTICA 7.0, StatSoft Inc., Tulsa, USA).
The conducted research is not related to either human or animal use.
3.1 Screening of arabitol-producing yeasts from L-arabinose
The two-step method of screening proved effective in the isolation of the desirable strains. One hundred isolates from various habitats were obtained and microscopically confirmed. The results of screening are presented in Table 1. The first step of screening showed that 55 isolates could utilize L-arabinose. In the second step of screening in a liquid medium 16 isolates produced L-arabitol with a yield of over 0.2 g g-1. The richest sources of interesting yeasts were from the leaves. The most effective producer of arabitol was the strain designated as 27RL-4, isolated from raspberry leaves, and represents 1% of all isolates. It produced 10.42 ± 0.40 g l-1 from 20 g l-1 of L-arabinose with a yield 0.51 ± 0.01 g g-1 in batch cultures (Table 2). None of the investigated isolates were able to produce ethanol from L-arabinose during the cultivation time.
|Source||Number of isolates||Number of L-arabinose assimilating isolates||Number of L-arabitol producing isolates||Number of isolates producing L-arabitol with a yield ≥0.2 g l-1|
|Conditions (variables)||Production parameters|
|Cult. time (h)||Residual arabinose (g l-1)[*]||Maximum CA-ol (g l-1)[*]||YA-ol (g g-1) [*]||pA-ol (g (g×h)-1) [*]||CB (g l-1) [*]||YB (g l-1)[*]||Final pH|
|Initial CA-a (g l-1)|
|Rotation speed (rpm)|
CA-ol – arabitol concentration; YA-ol - arabitol yield ; pA-ol - arabitol productivity; CB - biomass concentration; YB – biomass yield; CA-a - arabinose concentration; Values with the same superscript letters within a column and one variable are not significantly different (p<0.05).
3.2 Yeast identification
The colonies of strain 27RL-4 after three-day incubation on YPG agar were white-cream, convex, smooth and butyrous with a smooth edge. The cells were elongate, ellipsoidal or oval with multilateral budding and started to form a pseudomycelium (Fig. 1).
In the genetic identification procedure, the region containing the 3’ end of 18S rDNA, ITS1, 5.8S rDNA, ITS2 and the 5’ end of 26S rDNA was amplified and the product was found to have 558 base pairs. Comparison of DNA sequences in NCBI GenBank® identified isolate 27RL-4 as Candida parapsilosis.
Identification using the Biolog system, called metabolic fingerprint, confirmed that isolate 27RL-4 belonged to the species Candida parapsilosis. Detailed biochemical characteristics of the strain obtained using assimilation tests are presented in Table 3. The most intensively assimilated carbon substrates were carbohydrates. By contrast, strain 27RL-4 did not assimilate glycosides and only weakly assimilated carboxylic acids and polyalcohols. Figure 2 shows the ability of the selected strain to oxidize different groups of substrates during incubation. The oxidation of carbohydrates, amino acids, carboxylic acids, polyalcohols and miscellaneous substrates was noticeable after 24 h and slightly increased up to 72 h.
|Group of substrates||Carbon substrate||Result||Group of substrates||Carbon substrate||Result|
|Carbohydrates||D-Cellobiose||(-)||Miscellaneous||Succinic Acid Methyl Ester + D-Xylose||(+)|
|Gentiobiose||(-)||N-Acetyl-L-Glutamic Acid + D-Xylose||(+ /-)|
|Maltose||(+/-)||Quinic Acid + D-Xylose||(+ /-)|
|Maltotriose||(-)||D-Glucuronic Acid + D-Xylose||(+ /-)|
|D-Melezitose||(+)||Dextrin + D-Xylose||(-)|
|D-Melibiose||(-)||α-D-Lactose + D-Xylose||(+ /-)|
|Palatinose||(+)||D-Melibiose + D-Xylose||(+ /-)|
|D-Raffinose||(-)||D-Galactose + D-Xylose||(+ /-)|
|Stachyose||(-)||m-Inositol + D-Xylose||(+ /-)|
|Sucrose||(+)||1,2 Propanediol + D-Xylose||(+ /-)|
|D-Trehalose||(+ /-)||Acetoin + D-Xylose||(-)|
|Turanose||(+ /-)||Others||Succinic Acid Mono-Methyl Ester||(-)|
|L-Sorbose||(+ /-)||D-Mannitol||(+ /-)|
|Carboxylic acids||Fumaric Acid||(-)||i-Erythritol||(-)|
|L-Malic Acid||(+ /-)||Glycerol||(-)|
|2 Keto-D-Gluconic Acid||(+ /-)||Amygdalin||(-)|
3.3 Arabitol production from arabinose
The isolate C. parapsilosis 27RL-4 was cultivated under various conditions (initial concentrations of L-arabinose, initial pH values, temperatures and rotational speeds) using the “one variable at a time” optimization method. The results obtained are presented in table 2 and figure 3.
For one combination, substrate concentration (32.5 g l-1), temperature (28°C), and rotational speed (150 rpm) were chosen on the basis of earlier optimization experiments for C. parapsilosis DSM 70125 carried out by Kordowska-Wiater et al.  using response surface methodology. The higher concentrations of sugar (50.0 and 80.0 g l-1) were chosen to measure the ability of the yeast strain to produce arabitol in media of higher osmolarity. The novel strain was unable to assimilate more than 20.0–25.0 g l-1 of arabinose (Fig. 3(A)), and consumed the sugar only in the medium containing 20.0 g l-1 of arabinose. Production of arabitol (Fig. 3(C)) was also the most effective in this medium. It is worth noting that biomass production was similar in all cultures within the whole incubation period (Fig. 3(E)) and depended on the amount of L-arabinose consumed (similar biomass yields – Tab. 1). The pH of all cultures decreased to 2.8–3.0 during the incubation period. For comparison, C. parapsilosis DSM 70125 was cultivated in a medium containing 20 g l-1 of L-arabinose under the same temperature and rotational speed conditions. Similar assimilation of L-arabinose and biomass production was observed (Fig. 3(A), 3(E)), but the strain produced only 5.07 g l-1 of arabitol (Fig. 3(C)); the yield was 0.3 g g-1. When the effect of the initial pH of the medium on arabitol secretion was investigated for the novel strain, the results were similar across the pH range of 4.5–6.0 (Table 2), with an optimum at pH 5.5. This shows that this variable had a weak effect on the production of arabitol, especially since pH decreased to about 3 in each culture. The effect of rotational speed and temperature of cultivation on arabinose consumption and arabitol and biomass production by C. parapsilosis strain 27RL-4 was much more salient (Fig. 3(B), (D), (F)). Result show that the rotational speed of 100 rpm was unfavorable: it slowed down substrate assimilation and the production of arabitol and biomass and required extended incubation periods, exceeding 4 days (Table 2, Fig. 3(B), (D), (F)). Analysis of all the combinations of culture conditions for novel C. parapsilosis 27RL-4 confirmed that the optimal conditions for this strain were as follows: concentration of arabinose in the medium 20 g l-1, pH 5.5, incubation temperature 28°C, and rotational speed 150 rpm. These conditions were starting points for arabitol production in modified medium, where the amount of nitrogen was reduced and pH was maintained at 5.5-6.5 by the addition of CaCO3. In those cultures, the concentrations of arabitol and production yields were similar to that obtained in the first medium, but biomass concentration was lower at 12% suggesting higher efficiency of yeast cells in more stable pH (Table 2). Results suggest that the described modifications of culture conditions are the most promising solution for arabitol production.
3.4 Arabitol production from glucose
C. parapsilosis 27RL-4 produced arabitol from glucose (50 and 100 g l-1) as the main metabolite in batch cultures. Lower amounts of glycerol and ribitol were also detected in broth samples. Table 4 presents the highest concentrations of proper polyols, which were detected during cultivation time. Ethanol was present at trace amounts. To compare, C. parapsilosis DSM 70125 produced 0.79 ± 0.17 g l-1 and about 1.48 ± 0.006 g l-1 of arabitol from 50 and 100 g l-1 of glucose, respectively, within the same time and under identical conditions. The yields and productivities obtained by strain 27RL-4 are rather low, but better than the reference strain.
|Parameters of arabitol production|
|Initial CG (g l-1)||Cult. time (h)||Residual glucose (g l-1)[*]||Maximum C (g l-1)[*]||Y (g g-1)[*]||p (g (g×h)-1)[*]|
|Parameters of glycerol production|
|Maximum C (g l-1)[*]||Y (g g-1) [*]||p (g (g×h)-1) [*]|
|Parameters of ribitol production|
|Maximum C (g l-1)[*]||Y (g g-1) [*]||p (g (g×h)-1) [*]|
CG – glucose concentration; C –concentration; Y - yield ; p - productivity;
A two-step screening procedure involved the sugar assimilation and polyol production stages to isolate polyol-producing yeast strains. The use of this method with positive results has been reported in screening of pentose-fermenting yeasts , and xylitol and D-arabitol producing strains [16, 17]. It is worth noting that in the three reports mentioned, the percentages of effective yeast strains were 1.5, 2.12, and 2.6%, respectively, while in the present work it was 1%. The novel yeast strain was identified by sequencing of the region containing the 3’ end of 18S rDNA, ITS1, 5.8S rDNA, ITS2 and the 5’ end of 26S rDNA, which is a routine and reliable method used for yeast identification. Additionally, we decided to use the Biolog system, which is much less known, but can be used not only for identification, but also for biochemical characterization. Praphailong et al. , Foshino et al. , and Wang et al.  used this system for the identification of food, beverage, and marine yeasts. Praphailong et al.  and McGinnis et al. , in evaluating studies, reported that it can sometimes provide an incomplete or incorrect identification, but C. parapsilosis was also recognized correctly. It should be noted that there are several differences between the metabolic profile obtained using this system and the biochemical characteristics according to Kurtzman et al. . These differences, which include the assimilation of pentoses, especially L-arabinose and D-xylose (Table 3), might be due to insufficient aeration of Biolog plates. The novel strain reported here has been positively screened to be able to assimilate L-arabinose, but requires oxygen and growth factors supplied in the cultivation medium. The key role of aeration in arabinose assimilation has been reported by Fonseca et al. .
The novel isolate C. parapsilosis 27RL-4 showed the ability to effectively produce arabitol from arabinose, giving the concentration 10.42 g l-1 and yield 0.51 g g-1 under optimal conditions, which was much higher than results obtained for the reference strain C. parapsilosis DSM 70125 under the same conditions. In an optimization experiment reported by Kordowska-Wiater et al. , this reference strain produced 15.45 g l-1 of polyol from 32.5 g l-1 of L-arabinose giving a yield of 0.475 g g-1 after 2 days of incubation, which was similar to what we observed for the novel strain. Other authors confirm the ability of yeasts from the genus Candida to assimilate L-arabinose and secrete arabitol into the medium. Saha and Bothast  selected C. entomaea NRRL Y-7785 from among 49 strains as an efficient producer of arabitol with a yield of 0.66 g g-1 from 50 g l-1 of L-arabinose at pH 5, 34°C, and 200 rpm. Dien et al. , who screened fifty Candida strains, found that C. auringiensis NRRL Y-11848 produced 11–37 g l-1 of arabitol from 32–75 g l-1 of pentose, giving a yield of 0.73 g g-1 under optimal conditions. Among environmental yeast strains, Candida NY7122, closely related to C. subhashii, was found to produce L-arabitol and ethanol from L-arabinose. The isolate secreted 5.63 and 10.69 g l-1 of arabitol from 20 g l-1 of substrate at 30 and 37°C, respectively, with yields in the range 0.3–0.53 g g-1 . The results obtained for the isolate C. parapsilosis 27RL-4 are generally similar to observations made on the capabilities of other Candida strains.
There are currently no reports regarding arabitol production from D-glucose by C. parapsilosis. Zakaria  reported that Candida famata R28 grew very well in the medium containing D-glucose as the sole carbon source and was able to produce D-arabitol as the single product of metabolism, but he did not present the concentration of this compound in fermentation broth. Saha et al.  reported that Candida albicans, Candida pelliculosa and C. famata could produce D-arabitol from glucose, but usually incubation time was long and yields too small to be industrially acceptable. It may be concluded based on existing reports that Candida spp. have a generally low capability for the production of arabitol from glucose.
From among one hundred polish yeast strains isolated from soil and plant habitats, the strain designated as 27RL-4 and identified as C. parapsilosis could assimilate L-arabinose and produce arabitol most efficiently. Analysis of all the combinations of culture conditions used to grow the novel strain C. parapsilosis 27RL-4 confirmed that the optimal conditions for this strain were as follows: concentration of arabinose in the medium 20 g l-1, initial pH 5.5, incubation temperature 28°C, and rotational speed 150 rpm, when the product concentration was closed to 11 g l-1. An interesting observation regarding this yeast strain was that it can be used for arabitol production on media containing limited amounts of substrate. Another promising solution was the addition of CaCO3 to yeast culture to maintain pH at a stable level and the reduction of nitrogen source amount. Further experiments showed that this novel strain could produce low concentrations of arabitol also from glucose. The results obtained in this work encourage continuation of studies on L-arabitol production by C. parapsilosis 27RL-4. The search for new strains able to successfully assimilate arabinose and produce the polyalcohol arabitol is an important branch of research, especially in the context of the utilization of pentose sugars abundant in plant biomass and new applications for the biotechnological products obtained from them.
The Biolog analyses were performed using equipment bought with European Union funds, The Eastern Poland Development Program 2007–2013, and Regional Laboratory of Renewable Energy, Institute of Agrophysics of Polish Academy of Sciences.
Conflict of interest: Authors state no conflict of interest.
 Erickson B., Nelson J.E., Winters P., Perspective on opportunities in industrial biotechnology in renewable chemicals, Biotechnol. J., 2012, 7, 176–18510.1002/biot.201100069Search in Google Scholar PubMed PubMed Central
 Koganti S., Kuo T.M., Kurtzman C.P., Smith N., Ju L.-K., Production of arabitol from glycerol: strain screening and study of factors affecting production yield, Appl. Microbiol. Biotechnol., 2011, 90, 257–26710.1007/s00253-010-3015-3Search in Google Scholar PubMed
 Mingguo J., Wang B., Yang L., Lin S., Cheng H., Microbiological purification of L-arabitol from xylitol mother liquor, J. Microbiol. Biotechnol., 2011, 21, 43–4910.4014/jmb.1006.06012Search in Google Scholar PubMed
 Kumdam H., Murthy S.N., Gummadi S.N., Production of ethanol and arabitol by Debaryomyces nepalensis: influence of process parameters, AMB Expresss, 2013, 3, 2310.1186/2191-0855-3-23Search in Google Scholar PubMed PubMed Central
 Saha B.C., Bothast R.J., Production of L-arabitol from L-arabinose by Candida entomaea and Pichia guilliermondii, Appl. Microbiol. Biotechnol., 1996, 45, 299–30610.1007/s002530050687Search in Google Scholar
 Fonseca C., Spencer-Martins I., Hahn-Hägerdal B., L-arabinose metabolism in Candida arabinofermentans PYCC 5603T and Pichia guilliermondii PYCC 3012: influence of sugar and oxygen on product formation, Appl. Microbiol. Biotechnol., 2007, 75, 303–31010.1007/s00253-006-0830-7Search in Google Scholar PubMed
 Kordowska-Wiater M., Targoński Z., Jarosz A., Biotransformation of L-arabinose to arabitol by yeasts from genera Pichia and Candida, Biotechnologia, 2008, 1, 177-188 [in Polish]Search in Google Scholar
 Watanabe I., Ando A., Nakamura T., Characterization of Candida sp. NY7122, a novel pentose-fermenting soil yeast, J. Ind. Microbiol. Biotechnol., 2012, 39, 307–31510.1007/s10295-011-1033-5Search in Google Scholar PubMed
 Bura R., Vajzovic A., Doty S.L., Novel endophytic yeast Rhodotorula mucilaginosa strain PTD3 I: production of xylitol and ethanol, J. Ind. Microbiol. Biotechnol., 2012, 39, 1003-101110.1007/s10295-012-1109-xSearch in Google Scholar PubMed
 Subtil T., Boles E., Improving L-arabinose utilization of pentose fermenting Saccharomyces cerevisiae cells by heterologous expression of L-arabinose transporting sugar transporters, Biotech. Biofuels., 2011, 4, 3810.1186/1754-6834-4-38Search in Google Scholar PubMed PubMed Central
 Kurtzman C.P., Fell J.W., Boekhout T., The Yeasts, a Taxonomic Study vol 1., Elsevier, Amsterdam, 2011Search in Google Scholar
 White T.J., Bruns T., Lee S., Taylor J., PCR protocols, a guide to methods and applications, Academic Press, New York, 1990Search in Google Scholar
 Kordowska-Wiater M., Kubik-Komar A., Targoński Z., Application of response surface methodology for the optimization of arabinose biotransformation to arabitol by Candida parapsilosis, Central Europ. J. Biol., 2013, 8, 835–84210.2478/s11535-013-0196-zSearch in Google Scholar
 Guo C., Zhao C., He P., Lu D., Shen A., Jiang N., Screening and characterization of yeasts for xylitol production, J. Appl. Microbiol., 2006, 101, 1096-110410.1111/j.1365-2672.2006.02994.xSearch in Google Scholar PubMed
 Zhu H.-Y., Xu H., Dai X.-Y., Zhang Y., Ying H.-J., Ouyang P.-K., Production of D-arabitol by a newly isolated Kodamaea ohmerii, Bioprocess Biosyst. Eng., 2010, 33, 565-57110.1007/s00449-009-0378-xSearch in Google Scholar PubMed
 Praphailong W., Van Gestel M., Fleet G.H., Heard G.M., Evaluation of the Biolog system for the identification of food and beverage yeasts, Let. Appl. Microbiol., 1997, 24, 455-45910.1046/j.1472-765X.1997.00057.xSearch in Google Scholar
 Foshino R., Gallina S., Andrighetto C., Rossetti L., Galli A., Comparison of cultural methods for the identification and molecular investigation of yeasts from sourdoughs for Italian sweet baked products, FEMS Yeast Res., 2004, 4, 609-61810.1016/j.femsyr.2003.12.006Search in Google Scholar PubMed
 Wang L., Chi Z., Wang X., Ju L., Chi Z., Guo N., Isolation and characterization of Candida membranifaciens subsp. flavinogenie W14-3, a novel riboflavin-producing marine yeast, Microbiol. Res., 2008, 163, 255-26610.1016/j.micres.2007.12.001Search in Google Scholar PubMed
 McGinnis M.R., Molina T.C., Pierson D.L., Mishra S.K., Evaluation of the Biolog MiroStation System for yeast identification, J. Med. Vet. Mycology, 1996, 34, 349-35210.1080/02681219680000591Search in Google Scholar PubMed
 Fonseca C., Romão R., Rodrigues de Sousa H., Hahn-Hägerdal B., Spencer-Martins I., L-Arabinose transport and catabolism in yeast, FEBS J., 2007, 274, 3589–360010.1111/j.1742-4658.2007.05892.xSearch in Google Scholar PubMed
 Dien B.S., Kurtzman C.P., Saha B.C., Bothast R.J., Screening for L-arabinose fermenting yeasts, Appl. Biochem. Biotechnol., 1996, 57/58, 233-24210.1007/978-1-4612-0223-3_21Search in Google Scholar
 Saha, B.C., Sakakibara, Y., Cotta, M.A., Production of D-arabitol by a newly isolated Zygosaccharomyces rouxii, J. Ind. Microbiol. Biotechnol., 2007, 34, 519-52310.1007/s10295-007-0211-ySearch in Google Scholar PubMed
© 2017 Monika Kordowska-Wiater et al.
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.