Production, medium optimization, and structural characterization of an extracellular polysaccharide produced by Rhodotorula minuta ATCC 10658

Abstract Several strains of microorganism are capable of converting carbohydrates into extracellular polysaccharide. The preset research is a first effort made to optimize extracellular polysaccharide (CRMEP) by Rhodotorula minuta ATCC 10658 using one factor at time and response surface methods. One factor at time was applied in the initial screening of substrates prior to optimization study. Of all the substrates examined, starch as carbon source and defatted soy bean powder as protein source were discovered to be best for CRMEP production. Response surface analysis revealed that 15 g/L starch and 30g/L defatted soy bean powder were the optimal chemical conditions. The model predicted 13.22 g/L for CRMEP, which went along with the experimentally observed result. Purification of CRMEP by anion‐exchange column of DEAE‐cellulose yielded RMEP. Structural investigation indicated that the main chain of RMEP was composed of (1 → 3) and (1 → 4)‐linked mannopyranosyl residues, with branches attached to O‐6 of some (1 → 3)‐linked mannopyranosyl residues. The branches were composed of Glcp‐(1 → residues.

namely R. mucilaginosa, R. glutinis, and R. minuta, have been rarely recognized as human pathogens (Arendrup et al., 2014). R. minuta has been recognized as a well-known source of EPS served in food, cosmetic, and pharmaceutical fields (Seveiri et al., 2019). The influence of substrate properties and their concentrations on microbial productions has been well-characterized. The quality and quantity of carbon and nitrogen in medium highly impact on the microbial proliferation and EPS synthesis (Kim et al., 2003;López et al., 2003;Nicolaus, Kambourova, & Oner, 2010). Organic carbon and nitrogen sources support microbial growth rate and EPS production (Czaczyk & Wojciechowska, 2003;Görke & Stülke, 2008). The synthesis of heteroglycan made of D-glucose, D-mannose, and D-glucuroic acid was stimulated by organic nitrogen (Elinov et al., 1988). Given this in view, enhanced productivity can be achieved by using proper substrates and optimization methods (Ma et al., 2013). Among various statistical methods, response surface methodology (RSM) has worldwide served as a pioneer in mathematical analysis of the variable factors on responses. Such a method was served to optimize the valuable products produced by fermentation method (Ma et al., 2013;Malinowska, Krzyczkowski, Łapienis, & Herold, 2009).
Prior to the optimization, screening method like one a factor at a time method was served to identify the key substrates among the various selected substrates (Singh, Singh, Tripathi, Khare, & Garg, 2011).
However, there are no available data for statistical optimization of EPS by R. minuta in submerge conditions. More importantly, the structural features of R. minuta´ EPS have not been well-characterized yet. Therefore, the objective of this research was the optimization of R. minuta' EPS production, purification of the produced EPS by anion-exchange column chromatography and finally, characterization of the main purified EPS by gas chromatography-mass spectrometry (GC-MS).

| Materials and chemicals
Rhdotorula minuta ATCC 10658 was purchased from Persian Type Culture Collection (PTCC). DEAE-Cellulose A52 and bovine serum albumin (BSA) were purchased from Pharmacia Co. and Merck, respectively. All materials were also provided from Merck. Aqueous solutions were prepared with ultra-pure water from a Milli-Q water purification system (Millipore). All other reagents used in this study were of analytical grades.

| General methods
Concentrations were performed under reduced pressure in a rotary evaporator (Heidolph Laborota 4,000 efficient rotary evaporator, Germany). The products were dried by vacuum freeze-drying (Christ Alpha 1-2 freeze-dryer). Protein in the exopolysaccharide was quantified according to the Bradford method (Bradford, 1976), using BSA as the standard. Ultraviolet-visible absorption spectra were recorded with a VarianCary100-Bio UV/visible spectrophotometer. Gas chromatography-mass spectrometry (GC-MS) was done on a HP5890 (II) instrument (Hewlett-Packard Component, USA) with an HPS quartz capillary column (25 m × 0.22 mm × 0.20 μm), and at temperatures programmed from 120 ºC (maintained for 2 min) to 260ºC (kept for 40 min) at a rate of 15ºC/min.

| Microorganisms, inoculums, and cultivation conditions
The strain was grown on potato dextrose agar (PDA), and for longterm storage, it was incubated at 4°C. 250 ml Erlenmeyer flasks with 50 ml medium was used for yeast growth. Seed cultures contained 20 g/L glucose and 10 g/L yeast extract with essential mineral elements. After sterilization, R. minuta ATCC 10658 was cultivated in seed medium and centrifuged cells were used in fermentation media.

| Fermentation media
Various substrates such as carbon and protein resources as variable factors were utilized to examine EPS productions in the fermentation media. Mineral elements (magnesium sulfate (7 mg/L), calcium chloride (2 mg/L), di-potassium hydrogen phosphate (1 g/L), ammonium sulfate (5/2 g/L), sodium chloride (0.1 g/L)), a rotary shaker at 180 RPM and 28°C were selected as constant factors. 3 percent of fresh seed culture was used in the fermentation media.

| Isolation and purification of exopolysaccharide
Yeast cells were removed from the submerge medium by dilution with distilled water (two times) and centrifugation (12,000 g, 5 min).
The cell-bound exopolysaccharides were removed with 0.25 N NaOH for 2 hr and mixed with supernatant. The supernatant was concentrated 3-fold with the rotary evaporator at 45°C. The concentrated solution was deproteinated by sevag method (1-butanol: chloroform at a ratio of 1:4, v/v) (Staub, 1965). After the removal of sevag reagent, to remove small molecules, the solution was dialyzed against deionized water for 48 hr. The non-dialyzate was then precipitated with 96% ethanol (1:4, v/v, stored for 24 hr at 4°C). Finally, the precipitate, collected by centrifugation, was lyophilized to give crude R. minuta ATCC 10658' exopolysaccharide (CRMEP). The CRMEP was dissolved in deionized water and filtered (0.45 μm). The solution was passed through an anion-exchange column of DEAE-Cellulose A52 (2.6 × 30 cm) (Jahanbin, 2018). The elution was a gradient of 0-1 M aqueous solution of NaCl. The collected fractions were monitored by the phenol-sulfuric acid colorimetric method at 490 nm (Dubois, Gilles, Hamilton, Rebers, & Smith, F., 1956).
Fractions, which corresponded to the major peak, were pooled, dialyzed, and lyophilized to result white pure polysaccharide (RMEP) and used for further study (Beigi & Jahanbin, 2019).

| Structure determination of RMEP
Methylation analysis was employed to determine the positions of glycosidic linkages and their proportions in RMEP. RMEP was methylated according to the Needs and Selvendran method (Needs & Selvendran, 1993). Briefly, 1 ml DMSO was added to dry RMEP (5 mg) in a 25 ml flask. The mixture was sonicated at room temperature for 20 min and 4 ml methyl sulfinyl methyl sodium (MSMS) was added to the solution to form a gel, and the mixture was again treated by sonication for 20 min. 0.3 ml methyl iodide was then added, and the mixture was sonicated for 15 min at 25°C once more. After incubation for 6 hr at room temperature, excess MSMS was removed by the addition of water and subsequent centrifugation (Chaplin & Kennedy, 1994).
The methylated polysaccharide was extracted with chloroform (4 ml) and was examined by IR spectrometry. Complete methylation was confirmed by the absence of an absorption peak related to the hydroxyl group in the region of 3200-3700 cm -1 . The methylated RMEP was hydrolyzed with formic acid and TFA (2 M), reduced with NaBD 4 for 24 hr, and finally acetylated with acetic anhydride-pyridine (1:1).

| Statistical method
One factor at a time method was used to identify the key substrates among the various carbons and nitrogen sources which exerted high influence on EPS production, then the selected carbon and nitrogen sources were optimized using response surface method.
On the other hand, carbon sources (starch, glucose, fructose, sorbitol, and lactose) were variable factors. As can be seen in Figure 1, starch source was ranked as the best substrate for CRMEP production, followed closely by sorbitol. By contrast, glucose had the least impact on CRMEP production.
Many research put forward hypotheses that carbon sources as a main precursor for EPS production had the highest effect on the polysaccharide's properties and supported the high quantity of polysaccharide. Various research studies were carried out to investigate the effect of the carbon and nitrogen sources on polysaccharides production (Gientka, Bzducha-Wróbel, Stasiak-Różańska, Bednarska, & Błażejak, 2016;Khani et al., 2016). Khani et al., 2016(Khani et al., 2016 stated that EPS production was stimulated by the high content of glucose. A study done by Gientka et al., 2016(Gientka et al., 2016 revealed that, of all substrates used, carbon source significantly enhanced the yeast polysaccharide. Maalej et al., 2014 stated starch was discovered to be preferred over other carbon sources for EPS by Pseudomonas stutzeri AS22. Polysaccharide was better off than other carbon sources for ease of polymerization (Fan, Soccol, Pandey, & Soccol, 2007). However, microbial EPS depends upon the types of carbon source and the yeast species. Lactose, for instances, was ranked as the best substrate for EPS production by Zunongwangia profunda SM-A87 (Liu et al., 2011), while Pavlova & Grigorova, 1999(Pavlova & Grigorova, 1999 stated that sucrose was chosen to be the best source of carbon source for EPS by Rhodotorula acheniorum MC.

| Protein sources
Temperature (21°C), starch (7 g/L), pH (5.5), stirring round 180 rpm, minerals, and the duration of 4 days fermentation were considered constant factors. The effects of nitrogen sources on CRMEP production were investigated. Inorganic nitrogen sources had a negligible impact on CRMEP yields (urea and NH 4 NO 3 ). To be more precise, the least quantity of CRMEP was obtained in medium containing urea as nitrogen sources. However, organic nitrogen sources were rated as good substrates to stimulate R. minuta ATCC 10658̕ CRMEP.

| Optimization of CRMEP production using RSM
Central Composite Rotatable Design (CCRD) was used as promising method for optimization. One factor at a time conducted disclosed that starch as carbon source and soybean as nitrogen sources were typically key factors enhancing CRMEP production by R. minuta. The data were statically analyzed using Design-Expert software (6.0.10 version) and the method of the least squares was applied to formulate second equation. R 2 index was used as an adequacy of model (Giovanni, 1983). Significance at 5 percent level of confidence was selected to identify the effect of variables at linear, quadratic, and interactive level on response. ANOVAs results indicated that the model term and linear effect of starch and soybean powder having values "Prob > F" less than 0.05 were significant (Table 1). Table 1, F value of soybean powder was higher than that of the carbon source; as a result, protein source was quantitatively more important than that of carbon source on CRMEP production. Starch and soybean powders were the variable factors as given in Table 1. Once, soybean concentration rose to 30 g/L along with the high and constant level of glucose (15 g/L), CRMEP production considerably climbed from 9.9 up to 13.2 (run 4 and 8). Similar patterns repeated at the lower rate for glucose substrate (run 4 and 10) ( Table 2).

As shown in
That is to say, soybean was more effective than starch on CRMEP production. Analysis of variance (ANOVA) was applied to fit a sec-

| Validation of the optimized culture conditions
Under the optimum conditions, the predicted CRMEP content reached the highest point (13.35 g/L). To validate the suitability of the model equation, predicted optimal condition was run. Under the actual experimental condition, the CRMEP level was 13.82 ± 0.2 g/L, which was slightly higher than the predicted maximum value (13.22 g/L).

| Purification of CRMEP and production of RMEP
absorption at 280 and 260 nm in the UV spectrum, indicating the absence of protein and nucleic acid.

| Methylation analysis of RMEP
Methylation analysis by GC-MS was employed to determine the types and proportions of glycosidic linkages of monosaccharide residues in RMEP. As summarized in Table 3, RMEP showed the presence of four components, namely 2,4,6-Me 3 -Man, 2,3,6-Me 3 -Man, 2,4-Me 2 -Man, and 2,3,4,6-Me 4 -Glc in molar ratios of 3.35:3.07:1.30:1.36 (about 3:3:1:1). This result revealed that Man and Glc accounted for about 85% and 15% of the total methylated residues, respectively. It was also revealed that some mannopyranosyl residues (around 17%) were branched and the molar number of 1,3,6-linked Manp was approximately equal to the number of the 1-linked Glcp, which implied all sugar residues appeared to have been completely methylated and the RMEP branches terminated with glucose residues. Moreover, the methoxyl groups were not observed at the C-5 position, indicated that all the sugar residues existed in the pyranose ring forms. On the basis of the aforementioned results, it can be concluded that RMEP had a backbone chain of 1,3-linked and 1,4-linked Manp residues with side chains of terminal Glcp residues substituted in O-6 position of some 1,3-linked Manp.
Our results were in accordance with that of Ramirez (2016), who reported that EPS, produced by Rhodotorula minuta BIOTECH 2,178, composed of mannose and glucose (Ramirez, 2016). The author gave no more information about molar ratios of the monosaccharides.

| CON CLUS ION
In this study, one factor at a time was used to identify the key substrates with a great impact on exopolysaccharide production, named CRMEP. The highest content of CRMEP in starch media indicated that this substrate could be assimilated by R. minuta ATCC 10658.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.