Appl Biochem Biotechnol (2015) 175:194–208 DOI 10.1007/s12010-014-1253-z
Effect of Pretreatments and Endo-1,4-β-Xylanase Hydrolysis of Canola Meal and Mustard Bran for Production of Oligosaccharides Lin Yuan & Martin G. Scanlon & N. A. Michael Eskin & Usha Thiyam-Hollander & Ayyappan A. Aachary
Received: 10 April 2014 / Accepted: 10 September 2014 / Published online: 24 September 2014 # Springer Science+Business Media New York 2014
Abstract Alkali/acid-pretreated canola meal and mustard bran were subjected to endo-1,4-βxylanase (T. longibrachiatum) hydrolysis for oligosaccharide production. Pretreatments significantly (α=0.05) increased the relative content of pentose sugars, especially in alkali-pretreated canola meal (∼44 %) and mustard bran (∼72 %). The amounts of pentosan (g/100 g) in acid- and alkali-pretreated canola meal were 7.50 and 8.21 and in corresponding mustard bran were 8.67 and 10.39, respectively. These pretreated substrates produced a pentose content (g/100 g) of 2.10±0.14 (18 h) and 2.95±0.10 (24 h), respectively, during hydrolysis. As per UPLC-MS data, the main oligosaccharides in the hydrolyzates of alkali-pretreated substrates are xylo-glucuronic acid and xylobiose. The release of total phenolics of the hydrolyzates increased until 18 h irrespective of the type of substrate or pretreatment. Hydrolyzates of acid-pretreated substrates indicated more total antioxidant activity than alkali-pretreated substrates, attributed to its high phenolic content. The study suggests the potential of canola meal and mustard bran for the production of oligosaccharides, wherein the use of various combinations of cell-wall-degrading enzymes and its optimization may result in a better yield, with simultaneous production of endogenous phenolics. Keywords Oligosaccharides . Pentoses . Pentosan . Endoxylanase . Canola meal . Mustard bran . Phenolics . Sinapine
Introduction Hemicelluloses, especially xylans, represent an immense resource of biopolymers for industrial applications. In this context, utilization of such carbohydrate component in canola meal U. Thiyam-Hollander Richardson Centre for Functional Foods and Nutraceuticals, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada L. Yuan : N. A. M. Eskin : U. Thiyam-Hollander (*) : A. A. Aachary Department of Human Nutritional Sciences, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada e-mail: [email protected] M. G. Scanlon Department of Food Science, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada
Appl Biochem Biotechnol (2015) 175:194–208
195
and mustard bran could be of great commercial value. Canada produced around 13.3 million metric tons (MT) of canola seed, which was the top revenue crop once again in 2012 generating $8.1 billion in farm cash receipts [1]. Canola seeds are mainly used for oil production generating large quantity of canola meal in the oil industry. It was reported that the yield of canola-rapeseed meal reached 3.6 million MT during the 2010/2011, which was increased by 33 %. Canada is also the largest producer and exporter of mustard seed generating a large amount of bran for the food condiment and additive industries. Canola meal and mustard bran are primarily used as animal feed. Canola meal contains a relatively high amount of fiber due to the high content (30 %) of hull in the meal while the hulls alone contain up to 48 % dietary fiber. With respect to mustard bran, it is also relatively rich in fiber with the content of crude fiber about 15–25 % based on dry matter [2]. Up to now, little attention has been given to better utilizing the carbohydrate constituents in canola meal and mustard bran. The β-1,4-linked D-xylosyl backbone of xylan is decorated through an α-1,2-linkage with glucuronic acid and 4-O-methyl-α-D-glucuronic acid, whereas α-L-arabinofuranosyl residues are linked through α-1,3 to the polymer [3, 4]. 1,4-β-D-Xylan xylanohydrolase (EC 3.2.1.8), generally known as endoxylanase, hydrolyzes β-1,4-xylosidic bonds in xylan producing βanomeric XOS consisting of 4-O-methyl-α-D-glucuronic acid residues. XOS and arabinoxylooligosaccharides (AXOS) are a class of functional food ingredients generally regarded as prebiotics [5]. They exhibit beneficial health effects including revitalizing the growth of intestinal bifidobacteria, enhancing immunity activation, dietary-fiber-like effects, water retention capacity, and are non-cariogenic [6, 7]. XOS and AXOS have acceptable organoleptic properties and do not exhibit toxicity or negative effects on human health [6]. The preferred degree of polymerization (DP) of these oligosaccharides is 2–4 for food-related applications [8]. The sweetness of xylobiose is equivalent to 30 % that of sucrose and the sweetness of higher XOS is moderate and possesses no off-taste. XOS, being low calorific, finds use in the preparation of anti-obesity diets [9]. Exploiting such underutilized agri-by-products for the production of XOS and AXOS has caught global attention as they can now be produced from xylan using chemical methods, autohydrolysis, direct enzymatic hydrolysis of a susceptible substrate, or a combination of chemical and enzymatic treatments [10]. Enzymatic production of pentose oligosaccharides, however, is preferred by the food industry because of the environmental problems associated with the other methods for producing XOS. Endo-1,4-β-xylanase or 1,4-β-D-xylan xylanohydrolase from (T. longibrachiatum) hydrolyzes β-1,4-xylosidic bonds within the β-(1,4)-linked D-xylosyl backbone of xylan producing β-anomeric XOS. For the efficient enzymatic production of XOS from agro-residues, the xylan must be exposed to the endoxylanase action. Since the complete extraction of xylan is time-consuming and is often related with environmental issues, a mild pretreatment method was proposed for making the xylan more available for enzymatic reaction [11]. These researchers reported that a mild alkali pretreatment followed by enzymatic hydrolysis enhances the yield of XOS. Canola meal is also a richer source of phenolic compounds compared to other oilseeds [12]. The purpose of our study was to evaluate two different pretreatment methods, dilute alkali treatment and dilute acid treatment, of canola meal and mustard bran for the efficient production of pentose sugars (monosaccharides and oligosaccharides) by the endoxylanase enzyme obtained from T. longibrachiatum. The study also estimated the total phenolic content of the enzyme hydrolyzates as endoxylanases are cell-wall-breaking enzymes, which could release bound phenolics. The study also encompassed the identification of oligosaccharides by UPLC-MS to understand the effect of type of substrates and pretreatment on the variety of oligosaccharides released. Additionally, the enzyme hydrolyzates rich in oligosaccharides and
196
Appl Biochem Biotechnol (2015) 175:194–208
phenolics were assessed for 1,1-diphenyl-2-picrylhydrazyl radical (DPPH) radical scavenging activity to ascertain its antioxidant potential.
Materials and Methods Materials All chemicals were of analytical grade. Standard xylose, birch wood xylan, endoxylanase, 3,5dinitrosalicylic acid (DNS), orcinol, and 1,1-diphenyl-2-picrylhydrazyl radical (DPPH) were purchased from Sigma-Aldrich (USA) while D-xylose assay kit and standards of 1,4-β-D-xylooligosaccharides (xylobiose, xylotriose, xylotetraose, and xylopentose) were purchased from Megazyme International Ireland (Ireland). The endo-1,4-β-xylanase (1,4-β-D -xylan xylanohydrolase, EC 3.2.1.8) from T. longibrachiatum (CAS 9025-57-4) was obtained from Sigma-Aldrich (USA). Canola meal and mustard bran were provided by Bunge Canada and GS Dunn Ltd., Canada, respectively, and were dried at 40 °C for 24 h, milled (60 mesh), and stored in polycarbonate containers. Standard sinapic acid was procured from Sigma-Aldrich (St. Louis, MO, USA). Sinapine was obtained from EPL Bioanalytical Services, IL, USA. Pretreatment of Canola Meal and Mustard Bran Alkali Treatment Canola meal and mustard bran powder were treated with 2 % NaOH solution at a solid to liquid ratio of 1:6 (w/w) for 12–15 h at room temperature with occasional stirring [11, 13]. Excess alkali was washed with deionized water till the pH of the washings was 6.0. The pretreated substrates were dried at 40 °C and stored at 4 °C until further use. Acid Treatment Canola meal and mustard bran powder were soaked in 0.1 % H2SO4 solution at a solid to liquid ratio of 1:10 (w/w) for 12 h at 60 °C [14]. The excess acid was washed with deionized water till the pH of the washings was 6.0. The pretreated substrates were dried at 40 °C and stored at 4 °C until further use. Endoxylanase Activity Assay The activity of endo-1,4-β-xylanase or endoxylanase from T. longibrachiatum was estimated using standard birch wood xylan as the substrate [15]. One unit of endoxylanase activity was defined as the amount of enzyme required to produce 1 μmol of xylose per minute under the specified conditions. Enzymatic Hydrolysis of Untreated and Pretreated Canola Meal and Mustard Bran The conditions for the production of XOS from pretreated canola meal and mustard bran followed the procedure previously reported for pretreating corncobs [11] with some modifications. The reaction was carried out in 250-ml conical flasks containing 50-ml reaction mixture, consisting of 3.0 g untreated/pretreated powder of canola meal/mustard bran and
Appl Biochem Biotechnol (2015) 175:194–208
197
700 U of endoxylanase. The volume was made up to 50-ml using Mc Ilvain buffer (0.05 M, pH 5.4). The enzymatic reaction was carried out in a shaking water bath maintained at 50 °C for specific reaction periods (1, 6, 18, and 24 h). At the end of incubation, the reaction was stopped by keeping the reaction mixture in a boiling water bath for 10–15 min, filtered using 0.45-μm membrane filters and stored at −20 °C until further use. Chemical Analysis of Untreated and Pretreated Canola Meal and Mustard Bran The pentose content was determined following a modified method of ferric-orcinol assay by Pramod and Venkatesh [16]. Monosaccharides were determined by treating pretreated and untreated canola meal and mustard bran (20 mg) with 1 ml 12 M H2SO4 at 30 °C for 30 min, then diluting to 6 ml (2 M H2SO4) followed by hydrolysis at 100 °C for 2 h. Analysis was carried out using a high-performance anion-exchange chromatograph (HPAEC) as described by Wood et al. [17]. All measurements were replicated three times. Total Phenolic Content in Enzymatic Hydrolyzates The Folin-Ciocalteu reagent method was used to measure total phenolic content with a few modifications [18]. The absorption was measured at 750 nm. A standard sinapic acid calibration graph was used to express the total phenolics in the hydrolyzates as milligrams of sinapic acid equivalents (SAE) per kilogram of the sample. This assay is also sensitive to sugars that are reactive to the Folin-Ciocalteu reagent. Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry of Oligosaccharides from CM and MB Confirmation of pentose derivatives in the enzyme hydrolyzates was carried out using a Waters Acquity ultra-performance liquid chromatography (UPLC) system coupled to a Quattro micro API tandem mass spectrometer (Milford, MA, USA). A Phenomenex Synergi 4u Fusion-RP column (150×4 mm, Torrance, CA) was utilized for the separation of pentose. The mobile phase A consisted of 5 mM of ammonium acetate (pH was adjusted to 3.2 with acetic acid), and the mobile phase B was 100 % methanol. The gradient was programmed at a flow rate at 0.5 ml/min as follows: initially, phase B was set at 25 % and kept for 1 min and then increased to 95 % over 10 min in a linear manner and held for 2 min. Column was re-equilibrated for 3 min after each injection with temperature kept constant at 35 °C. Samples were maintained at 4 °C throughout the analysis. The injection volume was 10 ml. The tandem mass spectrometer was equipped with an atmospheric pressure ionization (API) probe and was operated in positive ion mode (ES+) with the conditions tuned based on each authentic standard for identification purpose. The general MS/MS parameters were as follows: capillary voltage, 3.00 kV; source temperature, 100 °C; desolvation temperature, 400 °C; cone gas (N2) flow, 50 l/h; and desolvation gas (N2) flow, 400 l/h. The cone voltage was set at 25 V, and collision energy was set at 15 eV for all analytes. Multiple reaction monitoring (MRM) mode was used to monitor the precursor to product ion transition. Daughter ion mode was used to obtain MS/ MS spectrum of their precursor ions. Mass resolution was set at maximum. The hydrolyzates were filtered through 0.45-μm cellulose nitrate membranes to remove the unreacted polysaccharide fragments before the analysis. MS analysis of the hydrolysates has been performed using underivatized samples [19]. For comparison of results, authentic standards of D-xylose and 1,4-β-D-xylo-oligosaccharides (xylobiose, xylotriose, xylotetraose, and xylopentose) were used.
198
Appl Biochem Biotechnol (2015) 175:194–208
HPLC Analysis of Phenolic Compounds The phenolic profiling of extracts was established following a reversed-phase HPLC-DAD analysis as described previously [20]. Synergi 4 μm Fusion-RP 80 Å; 150×4.0 mm–4 μm (Phenomenex, Canada) column was used for sinapic acid derivative separation (330 nm). Standards of sinapine and sinapic acid were also analyzed for comparison purpose based on retention time and UV absorption spectra. DPPH Scavenging Activity of Enzyme Hydrolyzates Antioxidant activity of the enzyme hydrolyzates was assessed using the DPPH free radical scavenging activity assay [21]. Briefly, 50 μl of enzyme hydrolysate was added to 950 μl of DPPH (0.030 mg/ml) in methanol. Then, the mixture was shaken vigorously and left in darkness for 5 min. The absorbance of the mixture was measured against DPPH blank at 515 nm by a spectrophotometer. The DPPH scavenging activity was expressed as the inhibition rate (IR). Inhibition rate ð%Þ ¼ ½1−ðAs =Ab Þ 100 where Ab is the absorbance of reagent blank, and As is the absorbance of sample. Data Expression and Analysis Data analysis was carried out using Excel and SPSS (version 18.0). One factor analysis of variance (ANOVA) and Student-Newman-Keuls (S-N-K) mean separation for multiple comparisons were used to determine significant differences (p≤0.05) between treatments.
Results and Discussion The criteria used to evaluate the efficiency of different pretreatment methods were the relative increase in xylose and pentose content and the corresponding decrease in glucose content in the pretreated substrates. The results for xylose content of untreated and pretreated canola meal and mustard bran are compared in Table 1. The xylose content of canola meal ranged from 2.07 to 3.04 g/100 g while for mustard bran from 1.73 to 2.80 g/100 g. These results indicated that alkali pretreated samples had a significantly higher relative content of xylose than untreated canola meal and mustard bran. Even though pretreated canola meal showed a higher xylose concentration than the untreated substrates, there was no significant difference observed between xylose content of the two pretreatment groups. Similar to canola meal, the xylose concentration of mustard bran also increased during pretreatment. However, its amount in alkali-pretreated mustard bran is significantly higher than its content in acid-pretreated mustard bran. Additionally, a significant relative increase of arabinose content in pretreated samples was observed in comparison to untreated substrates. The amounts of arabinose (g/100 g of substrate) in acid and alkali-pretreated canola meal were 5.79 and 6.29, respectively, while the values were 7.34 and 9.01, respectively, for mustard bran. The content of total monosaccharides (neutral sugars) in untreated canola meal and mustard bran were 14.99 and 14.86 %. The acid-pretreated canola meal and mustard bran showed a total monosaccharide content of 13.18 and 15.62 %, respectively, and the alkali-pretreated ones showed a total monosaccharide
Appl Biochem Biotechnol (2015) 175:194–208
199
Table 1 Carbohydrate composition of untreated and pretreated CM and MB Parameter
Canola meal Untreated
Mustard bran Acid pretreatment
Alkali pretreatment
Untreated
Acid pretreatment
Alkali pretreatment
Xylose
2.07±0.09 a
2.73±0.04 b
3.04±0.34 b
1.73±0.03 a
2.51±0.03 b
Arabinose
4.43±0.15 a
5.79±0.21 b
6.29±0.08 c
5.14±0.03 a
7.34±0.09 b
9.01±0.02 c
TP Glucose
6.50±0.24 a 5.30±0.16 b
8.52±0.25 b 1.99±0.09 a
9.33±0.42 c 2.00±0.17 a
6.87±0.06 a 4.16±0.19 c
9.85±0.12 b 1.09±0.03 b
11.81±0.04 c 0.78±0.00 a
Galactose
2.74±0.02 c
2.11±0.09 a
2.38±0.06 b
3.28±0.03 a
3.89±0.01 b
4.93±0.03 c
Rhamnose
0.21±0.03 a
0.24±0.00 a
0.28±0.00 b
0.35±0.02 a
0.47±0.00 b
0.53±0.00 c
Mannose TNS Pentosan
2.80±0.02 c
0.24±0.00 a
0.32±0.02 b
0.36±0.01 c
0.20±0.01 a
0.32±0.01 c
0.26±0.00 b
14.99±0.45 b
13.18±0.45 a
14.35±0.66 b
14.86±0.31 a
15.62±0.17 a
18.31±0.07 b
5.72±0.21 a
7.50±0.22 b
8.21±0.37 c
6.05±0.05 a
8.67±0.12 b
10.39±0.04 c
Mean±standard; within each row, the same lowercase letters indicated no significant difference between different treatments at the level α=0.05, namely p>0.05; n=3 TP total pentose, TNS total neutral sugars
content of 14.35 and 18.31 %. This indicated a relative reduction and a significant increase in the total neutral sugars in acid-pretreated canola meal and alkali-pretreated mustard bran, respectively, in comparison to corresponding untreated substrates. The effect of these pretreatments on the monosaccharide composition is shown in Table 1 with the pentosan content expressed as xylose equivalents (g/100 g of substrates). The weight of the pentosan in the sample was calculated by multiplying the total pentose content with a known factor (0.88), assuming that the hydrolysis during pentose assay is complete, where 0.88 accounts for the loss of water during xylosidic bond formation [14]. The relative content of pentose sugars, therefore pentosan, increased significantly during pretreatment. The initial levels of pentosan in untreated canola meal and mustard bran were 5.72 and 6.05 g xylose equivalent, respectively. These increased to 7.50–8.21 g xylose equivalent and 8.67–10.39 g xylose equivalent in canola meal and mustard bran, respectively, during pretreatment. Statistical comparison indicated a significant relative increase of pentosan in pretreated samples in comparison to untreated substrates. The amount of pentosan in acid- and alkali-pretreated canola meal were 7.50 g xylose equivalent and 8.21 g xylose equivalent, respectively, while the values were 8.67 g xylose equivalent and 10.39 g xylose equivalent, respectively, for mustard bran. Of the two types of pretreatments, the alkali pretreatment increased the relative concentration of pentose sugars in both canola meal and mustard bran more efficiently. The apparent increase in the pentose content during pretreatment was probably due to the effectiveness of pretreatments to remove some cellulose and lignin components [22, 23]. The relative increase in the xylan content of corncob during a similar pretreatment process was reported [11]. In the present study, we observed a 31.1 and 43.3 % increase in the pentose content during acid pretreatment of canola meal and mustard bran, respectively. This clearly indicated the better efficacy of acid pretreatment in canola meal matrix in comparison to mustard bran. The apparent increase in the pentose concentration, however, was more pronounced during alkali pretreatment with canola meal and mustard bran showing a 43.5 and 71.7 % increase in pentose content, respectively. This relative increase in pentose content when compared to native substrates has a positive effect on the enzymatic hydrolysis. The
200
Appl Biochem Biotechnol (2015) 175:194–208
pretreatment methods expose non-starch polysaccharide components, especially pentosans, rendering them more susceptible to endoxylanase action. The xylan of raw corncob cell wall was located within the lignin network through ester and ether lignin-carbohydrate linkages [11]. The availability of xylan directly determines the pentose content liberated from certain plant tissues. The glucose content of untreated canola meal and mustard bran were 5.30 and 4.16 %, which were significantly reduced during pretreatments (Table 1). The glucose concentration of acid- and alkali-pretreated canola meal were 1.99 and 2.00 %, respectively, while that of corresponding mustard bran samples were and 1.09 and 0.78 %, respectively. This reduction of substantial quantity of glucose can be extrapolated to the removal of cellulose, as it is a biopolymer of glucose. The pretreated substrates with such low concentration of glucose are suitable to produce oligosaccharides from pentosan. The protein contents of untreated and pretreated canola meal and mustard bran were compared (Table 2). The results indicated that alkali-pretreated samples had a significantly lower relative content of protein than untreated canola meal and mustard bran, which also reflected the efficiency of alkali pretreatment. The results also showed that the proteins had been removed from the substrates during pretreatment along with cellulose, which resulted in an apparent increase in the quantity of pentosan in alkali-pretreated canola meal and mustard bran. Even though acid-pretreated canola meal samples showed a higher concentration of protein than the corresponding untreated canola meal, there was no significant difference between the untreated and acid-pretreated mustard bran. This indicated that acid pretreatment had relatively no effect on the removal of protein. With this observation, it can be concluded that treating the substrates with dilute alkali is more efficient than acid pretreatment. Generally, similar to protein content, the concentration of oil in the substrates was also reduced considerably making the substrates suitable for enzyme hydrolysis.
Enzymatic Production of Pentose Sugars The pentose content of enzymatic hydrolyzates obtained from untreated and pretreated canola meal and mustard bran were monitored over reaction time (1, 6, 18, and 24 h) (Table 3). The total pentose content of enzyme hydrolyzates was used as an indicator for the amount of oligosaccharides produced from untreated and pretreated canola meal and mustard bran. The total pentose contents gave only a rough estimate of oligosaccharides because the hydrolyzates might contain xylose and arabinose in addition to the oligomers, and it would also contribute toward total pentose content. An increased release of pentose sugar was observed for both
Table 2 Moisture, protein, and oil contents of untreated and pretreated canola meal and mustard bran Parameter (g/100 g)
Canola meal Untreated
Moisture Protein Oil
Mustard bran Acid pretreated
Alkali pretreated
Untreated
Acid pretreated
Alkali pretreated
9.00±0.69 b
9.50±0.55 b
7.48±0.30 a
7.58±0.62 a
8.78±0.81 b
36.88±0.08 b
42.78±0.05 c
29.19±0.32 a
19.56±0.34 a
19.25±0.07 a
12.14±0.05 b
7.87±0.18 ab
4.64±0.27 b
4.19±0.51 b
2.92±0.28 a
24.66±1.38 c
15.59±2.56 b
11.21±0.64 a
Mean±standard; within each row, the same lowercase letters indicated no significant difference between different pretreatments at the level α=0.05, namely p>0.05; n=3
Appl Biochem Biotechnol (2015) 175:194–208
201
Table 3 Enzymatic production of pentoses from canola meal and mustard bran Reaction Pentose content (g/100 g) time (h) Canola meal Untreated
Acid pretreated
Mustard bran Alkali pretreated
Untreated
Acid pretreated
Alkali pretreated
1
1.00±0.261 aAB 0.74±0.08 aA 1.39±0.18 aB
1.05±0.14 aA 1.20±0.02 aAB 1.35±0.04 aB
6
1.30±0.349 aA
1.22±0.03 bA 1.83±0.09 bB
1.40±0.15 bA 1.78±0.11 bB
2.31±0.19 bC
18
1.30±0.056 aA
1.71±0.08 cB 2.10±0.14 bcC 1.84±0.12 cA 2.56±0.10 cB
2.38±0.17 bB
24
1.26±0.103 aA
2.02±0.15 cB 2.35±0.13 cC
2.95±0.10 cB
2.23±0.13 dA 2.81±0.03 dB
Mean±standard; within each column and row, the same lowercase or uppercase letters indicated no significant difference between different treatments at the level α=0.05, namely p>0.05; n=3
canola meal and mustard bran for both treatments. The pentose content (g/100 g) of hydrolyzates obtained from alkali-pretreated, acid-pretreated, and untreated canola meal were 1.39– 2.35, 0.74–2.02, and 1.00–1.30, respectively, while hydrolyzates of alkali-pretreated, acidpretreated, and untreated mustard bran (g/100 g) were 1.35–2.95, 1.20–2.81, and 1.05–2.23, respectively. Within 1 h of reaction, the maximum pentose was produced from alkali-pretreated substrates wherein 1.39 and 1.34 g of pentose were liberated from 100 g of alkali-pretreated canola meal and mustard bran, respectively. This also indicated the efficacy of alkali pretreatment over acid pretreatment. With respect to acid- and alkali-pretreated canola meal, there was no remarkable increase in the pentose after 18 h of reaction, though it increased through 1 to 18 h. However, in the case of mustard bran, irrespective of the type of the pretreatments, there observed a significant difference in the yields between 18 and 24 h (α=0.05). While the change in pentose content for enzymatic hydrolyzates for canola meal control was small, the pentose content changed significantly for the mustard bran control samples. Differences in the nature of the carbohydrates could account for these differences suggesting that mustard bran polysaccharides were far more accessible to enzymatic hydrolysis compared to canola meal. It is evident that while pretreatment appeared necessary for canola meal, the yield increases for mustard bran were not that pronounced. Overall, the enzymatic hydrolyzates of alkali-pretreated canola meal and mustard bran contained more pentose with the yields further increasing over time compared to acid pretreatment. Since there were no significant increase in the amount of between 18- and 24h samples for canola meal, 18 h was selected as the optimum time. In the case of mustard bran, the pentose significantly increased within 6 h and then reached the highest at 24 h. Optimization of reaction time with respect to yield is very important from an economic perspective. Ai et al. [13] reported that the native and immobilized xylanase of Streptomyces olivaceoviridis produced 3.9 and 15.5 mg/ml of XOS, respectively, from pretreated corncob within 24-h reaction time. A reaction time of 24 h can be a bottleneck in the development of an economically viable process. Previous research by Aachary and Prapulla [11] optimized a shorter reaction (14 h) and obtained 81±3.9 % of XOS, which corresponded to 10.2± 0.14 mg/ml in the hydrolyzate. Jeong et al. [24] showed 60 °C as the optimum temperature for xylobiose production. An optimum temperature of 50 °C was also reported [11, 14]. Above 50 °C, XOS production decreased probably as a result heat inactivation of xylanases. In the present study, therefore, we carried out oligosaccharide production at 50 °C as most fungal endoxylanases are highly active
202
Appl Biochem Biotechnol (2015) 175:194–208
at this temperature. Further studies may be required to optimize conditions, such as temperature, for enzyme hydrolysis.
Total Phenolic Content of the Enzyme Hydrolyzates Enzyme hydrolyzates from the alkali-pretreated, acid-pretreated, and untreated canola meal and mustard bran samples were monitored during the production of pentose sugars, to assess the impact of enzyme hydrolysis on the release of phenolics. The results indicated the total phenolics, although some interference from sugars reacting with Folin-Ciocalteu occurred. A large number of compounds, such as other phenols, carbohydrates, aromatic amines, and ascorbic acid, were reported to interfere with Folin-Ciocalteu assay and causing an increase in the measurements. The results showed that the total phenolics (expressed as SAE) of the hydrolyzate increased with reaction time irrespective of the type of substrate or pretreatment (Table 4). At 0 h of reaction, the total phenolics of control samples of canola meal and mustard bran were significantly higher than the pretreated substrates (∼6.6 g SAE and ∼2.9 g SAE for MB per 100 g of substrates). In order to assess the effects of enzyme treatment on the release of phenolics from the substrates, the phenolic content of 0 h reaction (prior to the addition of enzyme) was subtracted from the total phenolics estimated at 1, 6, 18, and 24 h. The initial lower content of water-extractable phenolics of pretreated canola meal and mustard bran in comparison to untreated samples is attributed to the intensive removal of such phenolics during pretreatment. As endoxylanase hydrolysis increased over time, the amount of total phenolics released also increased. During the initial hour of reaction, the majority of the phenolics released appeared to be free phenolics. Generally, phenolics are heat-sensitive and would degrade at 50 °C if incubated for a long period (6–24 h). This degradation is reflected in the case of untreated substrates where a lower content of phenolics after 6 h was observed in comparison to 1 h. The content of phenolics increased significantly after 18 h. This increase might be attributed to the release of bound phenolics following cell wall degradation by endoxylanase. Since these phenolics are unstable at high temperatures for a relatively long duration, some degradation might have occurred as indicated by the reduced phenolic contents at 24 h. During the initial hour, the content of phenolics liberated in the hydrolyzates of alkali-pretreated substrates were remarkably low as most of the free phenolics may have been removed during pretreatment, and after 1 h of enzyme reaction, there was only a small release of bound phenolics. The phenolic Table 4 Total phenolic content of hydrolyzates obtained from canola meal and mustard bran Reaction time (h)
Total phenolic content (mg SAE/kg of hydrolyzate) Canola meal Untreated
Mustard bran Acid pretreated
Alkali pretreated
Untreated
Acid pretreated
Alkali pretreated
1
294.2±5.6 dC
92.6±0.1 aB
15.1±0.7 aA
460.9±7.6 bC
60.7±0.0 aB
3.2±0.7 aA
6
119.8±2.9 aB
206.2±0.5 bC
89.2±0.2 cA
425.7±36.4 aC
201.2±0.3 bB
50.6±0.4 bA
18
250.6±2.9 cB
288.3±8.2 dC
75.6±1.8 bA
540±8.3 cC
213.8±1.3 bB
51.3±0.2 bA
24
127.3±2.6 bB
220.9±2.4 cC
77.2±0.6 bA
523±7.7 cC
204.8±0.9 bB
70.2±0.1 bA
Mean±standard; within each column and row, the same lowercase or uppercase letters indicated no significant difference between different treatments at the level α=0.05, namely p>0.05; n=3
Appl Biochem Biotechnol (2015) 175:194–208
203
content then increased significantly up to 24 h, indicating the release of bound phenolics. This trend was also observed with acid pretreated samples. The present study mainly focused on the production of pentose sugars from the hemicelluloses of mustard bran and canola meal using an endoxylanase enzyme. The optimization of this process would certainly open up new areas of research for the simultaneous production of oligosaccharides and phenolics. Recently, the simultaneous production of XOS and phenolic acids (hydroxycinnamic and hydroxybenzoic acids) from hemicelluloses using crude enzyme mixture synthesized by a novel Bacillus subtilis KCX006 was reported [25]. The enzyme mixture was tested for simultaneous production of XOS and phenolic acids from xylan-rich plant biomass, such as wheat bran, sugarcane bagasse, bamboo, and rise husk. In this context, the present study provides preliminary data on the simultaneous production of oligosaccharides and phenolics from canola meal and mustard bran. The effect of enzyme treatment on the sinapine content in the hydrolyzates was also studied, but not reported here.
UPLC-MS Analysis of Oligosaccharides Produced from Canola Meal and Mustard Bran Glucuronoxylan (GX) is one of the major polysaccharides present in plant cell walls [26]. It is composed of β-(1→4)-linked xylopyranose (Xyl) residues forming linear chains. These xylan chains are frequently substituted by terminally linked glucuronic acid (GlcA) residues or by 4O-methyl derivative (MeGlcA), arabinofuranosyl (Ara) residues, and/or acetyl groups. Several studies identified neutral oligosaccharides as unbranched xylosyl residues with varied chain length (DP0.05; n=3
Appl Biochem Biotechnol (2015) 175:194–208
207
identify the phenolics attached to the oligosaccharides produced from canola meal and mustard bran. Similarly, production of XOS from autohydrolysis liquors of wheat straw and sunflower stalk, as well as the antioxidant activity of these autohydrolysis liquors were reported [30]. The scavenging activities of enzymatic hydrolyzates obtained from alkali-pretreated, acidpretreated, and untreated canola meal (IR, %) were 9.8–28.2, 22.1–57.2, and 60.8–73.9, respectively, while for mustard bran (IR, %), they were 2.1–5.9, 5.8–19.3, and 47.3–68.7, respectively. It can be summarized that the pretreatment methods, especially the alkali treatment, significantly reduced the sinapine in comparison with control groups. Xyloglucuronic acid, the major pentose derivative in the hydrolyzates, could also contribute toward the antioxidant activity.
Conclusions Both canola meal and mustard bran were successfully value added by producing oligosaccharides from the non-starch polysaccharide components of the same, following an enzymatic approach. Pretreating substrates with mild alkali/acid improved the efficiency of enzyme hydrolysis and thereby, oligosaccharide production. Alkali pretreatment was superior to acid pretreatment for making its pentosan content available for hydrolytic breaking. The use of mild pretreatment conditions together with use of food grade microbial endoxylanases makes the process very much environmentally friendly. Xylo-glucuronic acid was the major pentose derivative, and sinapine was the principal phenolic identified in the enzymatic hydrolyzates. The DPPH antioxidant activities of the enzymatic hydrolyzates were attributed to their reactive sugars and total phenolic content. Their yields can be further improved by (a) using a combination of cell-wall-degrading enzymes and (b) by optimizing the parameters of enzyme hydrolysis. The enzyme hydrolyzates can be either purified to obtain oligosaccharides and phenolics separately or freeze-dried to obtain a crude powder, which can be applied to various food systems to impart antioxidant and other bioactive benefits upon consumption. This is the first ever study conducted with respect to the production of oligosaccharides from canola meal and mustard bran. The study would encourage large-scale utilization of these by-products and thereby, increase their market potential. Acknowledgments This study was funded by the Agri-Food Research and Development Initiative (ARDI)Growing Forward 1 Program and supported by the Bunge Canada, Canola Council of Canada and GS Dunn Ltd. The authors would like to thank Dr. Steve Cui and his technician Ms. Cathy Wang from the Guelph Food Research Centre, Agriculture and Agri-Food Canada for carrying out the neutral sugar analysis. The authors also would like to thank Hai Feng, Department of Human Nutritional Sciences, University of Manitoba (Winnipeg, MA, Canada) for technical assistance with UPLC-MS. Conflict of Interest There is no conflict of interest between the authors to declare.
References 1. Canada’s Farm Income Forecast for 2011–2012. www.agr.gc.ca. Accessed 4 Feb 2014. 2. Animal Feed Resources Information System (2002). www.fao.org. Accessed 21 Dec 2013. 3. Zui, F., Kaneko, S., Kuno, A., Kobayashi, H., Kusakabe, I., & Mizuno, H. (2004). Journal of Biological Chemistry, 279(10), 9606–9614. 4. Ishihara, M., Nojiri, M., Hayashi, N., Nishimura, T., & Shimizu, K. (1997). Enzyme and Microbial Technology, 21, 170–175.
208
Appl Biochem Biotechnol (2015) 175:194–208
5. Okazaki, M., Fujikawa, S., & Mastumoto, N. (1990). Bifidobacteria and Microflora, 9, 77–86. 6. Nabarlatz, D., Farriol, X., & Montane, D. (2005). Industrial and Engineering Chemistry Research, 44, 7746– 7755. 7. Moure, A., Gullon, P., Dominiguez, H., & Parajo, J. C. (2006). Process Biochemistry, 41(9), 1913–1923. 8. Loo, J. V., Cummings, J., Delzenne, N., Englyst, H., Franck, A., Hopkins, M., Kok, N., Macfarlane, G., Newton, D., Quigley, M., Roberfroid, M., van Vliet, T., & van den Heuvel, E. (1999). British Journal of Nutrition, 81, 121–132. 9. Toshio, I., Noriyoshi, I., Toshiaki, K., Toshiyuki, N. and Kunimasa, K. (1990). Japan Patent JP, 2119790. 10. Vazquez, M. J., Alonso, J. L., Domınguez, H., & Parajo, J. C. (2000). Trends in Food Science & Technology, 11, 387–393. 11. Aachary, A. A., & Prapulla, S. G. (2009). Bioresource Technology, 100(2), 991–995. 12. Naczk, M., Amarowicz, R., Sullivan, A., & Shahidi, F. (1998). Food Chemistry, 62(4), 489–502. 13. Ai, Z., Jiang, Z., Li, L., Deng, W., Kusakabe, I., & Li, H. (2005). Process Biochemistry, 40, 2707–2714. 14. Yang, R., Xu, S., Wang, Z., & Yang, W. (2005). LWT-Food Science and Technology, 38, 677–682. 15. Jiang, Z., Yang, S., Yang, Q., Li, L., & Tan, S. (2005). World Journal of Microbiology and Biotechnology, 21(6–7), 863–867. 16. Pramod, S. N., & Venkatesh, Y. P. (2006). Glycoconjugate Journal, 23(7–8), 481–488. 17. Wood, P. J., Weisz, J., & Blackwell, B. A. (1994). Cereal Chemistry, 71(3), 301–307. 18. Folin, O., & Ciocalteu, V. (1927). Journal of Biological Chemistry, 73, 627–650. 19. Reis, A., Pinto, P., Evtuguin, D. V., Neto, C. P., Domingues, P., Ferrer-Correia, A. J., & Domingues, M. R. M. (2005). Rapid Communications in Mass Spectrometry, 19, 3589–3599. 20. Khattab, R., Eskin, M., Aliani, M., & Thiyam, U. (2010). Journal of the American Oil Chemists Society, 87(2), 147–155. 21. Karakaya, S., & Simsek, S. (2011). Journal of the American Oil Chemists Society, 88, 1361–1366. 22. Parajo, J. C., Alonso, J. L., & Santos, V. (1995). Bioresource Technology, 51, 153–162. 23. Saska, M., & Ozer, E. (1995). Biotechnology and Bioengineering, 45, 517–523. 24. Jeong, K. J., Park, I. Y., Kim, M. S., & Kim, S. C. (1998). Applied Microbiology and Biotechnology, 50, 113–118. 25. Reddy, S. S., & Krishnan, C. (2013). Food Biotechnology, 27(4), 357–370. 26. Selvendran, R. R. (1985). Journal of Cell Science - Supplement, 2, 51–88. 27. Havlicek, J. E., & Samuelson, O. (1972). Carbohydrate Research, 22, 307–316. 28. Jacobs, A., Larsson, P. T., & Dahlman, O. (2001). Biomacromolecules, 2(3), 979–990. 29. Veenashri, B. R., & Muralikrishna, G. (2011). Food Chemistry, 126(3), 1475–1481. 30. Akpinar, O., Gunny, K., Yilmaz, Y., Levent, O., & Bostanci, S. (2010). BioResource, 5, 699–711.
Effect of Pretreatments and Endo-1,4-β-Xylanase Hydrolysis of Canola Meal and Mustard Bran for Production of Oligosaccharides Lin Yuan & Martin G. Scanlon & N. A. Michael Eskin & Usha Thiyam-Hollander & Ayyappan A. Aachary
Received: 10 April 2014 / Accepted: 10 September 2014 / Published online: 24 September 2014 # Springer Science+Business Media New York 2014
Abstract Alkali/acid-pretreated canola meal and mustard bran were subjected to endo-1,4-βxylanase (T. longibrachiatum) hydrolysis for oligosaccharide production. Pretreatments significantly (α=0.05) increased the relative content of pentose sugars, especially in alkali-pretreated canola meal (∼44 %) and mustard bran (∼72 %). The amounts of pentosan (g/100 g) in acid- and alkali-pretreated canola meal were 7.50 and 8.21 and in corresponding mustard bran were 8.67 and 10.39, respectively. These pretreated substrates produced a pentose content (g/100 g) of 2.10±0.14 (18 h) and 2.95±0.10 (24 h), respectively, during hydrolysis. As per UPLC-MS data, the main oligosaccharides in the hydrolyzates of alkali-pretreated substrates are xylo-glucuronic acid and xylobiose. The release of total phenolics of the hydrolyzates increased until 18 h irrespective of the type of substrate or pretreatment. Hydrolyzates of acid-pretreated substrates indicated more total antioxidant activity than alkali-pretreated substrates, attributed to its high phenolic content. The study suggests the potential of canola meal and mustard bran for the production of oligosaccharides, wherein the use of various combinations of cell-wall-degrading enzymes and its optimization may result in a better yield, with simultaneous production of endogenous phenolics. Keywords Oligosaccharides . Pentoses . Pentosan . Endoxylanase . Canola meal . Mustard bran . Phenolics . Sinapine
Introduction Hemicelluloses, especially xylans, represent an immense resource of biopolymers for industrial applications. In this context, utilization of such carbohydrate component in canola meal U. Thiyam-Hollander Richardson Centre for Functional Foods and Nutraceuticals, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada L. Yuan : N. A. M. Eskin : U. Thiyam-Hollander (*) : A. A. Aachary Department of Human Nutritional Sciences, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada e-mail: [email protected] M. G. Scanlon Department of Food Science, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada
Appl Biochem Biotechnol (2015) 175:194–208
195
and mustard bran could be of great commercial value. Canada produced around 13.3 million metric tons (MT) of canola seed, which was the top revenue crop once again in 2012 generating $8.1 billion in farm cash receipts [1]. Canola seeds are mainly used for oil production generating large quantity of canola meal in the oil industry. It was reported that the yield of canola-rapeseed meal reached 3.6 million MT during the 2010/2011, which was increased by 33 %. Canada is also the largest producer and exporter of mustard seed generating a large amount of bran for the food condiment and additive industries. Canola meal and mustard bran are primarily used as animal feed. Canola meal contains a relatively high amount of fiber due to the high content (30 %) of hull in the meal while the hulls alone contain up to 48 % dietary fiber. With respect to mustard bran, it is also relatively rich in fiber with the content of crude fiber about 15–25 % based on dry matter [2]. Up to now, little attention has been given to better utilizing the carbohydrate constituents in canola meal and mustard bran. The β-1,4-linked D-xylosyl backbone of xylan is decorated through an α-1,2-linkage with glucuronic acid and 4-O-methyl-α-D-glucuronic acid, whereas α-L-arabinofuranosyl residues are linked through α-1,3 to the polymer [3, 4]. 1,4-β-D-Xylan xylanohydrolase (EC 3.2.1.8), generally known as endoxylanase, hydrolyzes β-1,4-xylosidic bonds in xylan producing βanomeric XOS consisting of 4-O-methyl-α-D-glucuronic acid residues. XOS and arabinoxylooligosaccharides (AXOS) are a class of functional food ingredients generally regarded as prebiotics [5]. They exhibit beneficial health effects including revitalizing the growth of intestinal bifidobacteria, enhancing immunity activation, dietary-fiber-like effects, water retention capacity, and are non-cariogenic [6, 7]. XOS and AXOS have acceptable organoleptic properties and do not exhibit toxicity or negative effects on human health [6]. The preferred degree of polymerization (DP) of these oligosaccharides is 2–4 for food-related applications [8]. The sweetness of xylobiose is equivalent to 30 % that of sucrose and the sweetness of higher XOS is moderate and possesses no off-taste. XOS, being low calorific, finds use in the preparation of anti-obesity diets [9]. Exploiting such underutilized agri-by-products for the production of XOS and AXOS has caught global attention as they can now be produced from xylan using chemical methods, autohydrolysis, direct enzymatic hydrolysis of a susceptible substrate, or a combination of chemical and enzymatic treatments [10]. Enzymatic production of pentose oligosaccharides, however, is preferred by the food industry because of the environmental problems associated with the other methods for producing XOS. Endo-1,4-β-xylanase or 1,4-β-D-xylan xylanohydrolase from (T. longibrachiatum) hydrolyzes β-1,4-xylosidic bonds within the β-(1,4)-linked D-xylosyl backbone of xylan producing β-anomeric XOS. For the efficient enzymatic production of XOS from agro-residues, the xylan must be exposed to the endoxylanase action. Since the complete extraction of xylan is time-consuming and is often related with environmental issues, a mild pretreatment method was proposed for making the xylan more available for enzymatic reaction [11]. These researchers reported that a mild alkali pretreatment followed by enzymatic hydrolysis enhances the yield of XOS. Canola meal is also a richer source of phenolic compounds compared to other oilseeds [12]. The purpose of our study was to evaluate two different pretreatment methods, dilute alkali treatment and dilute acid treatment, of canola meal and mustard bran for the efficient production of pentose sugars (monosaccharides and oligosaccharides) by the endoxylanase enzyme obtained from T. longibrachiatum. The study also estimated the total phenolic content of the enzyme hydrolyzates as endoxylanases are cell-wall-breaking enzymes, which could release bound phenolics. The study also encompassed the identification of oligosaccharides by UPLC-MS to understand the effect of type of substrates and pretreatment on the variety of oligosaccharides released. Additionally, the enzyme hydrolyzates rich in oligosaccharides and
196
Appl Biochem Biotechnol (2015) 175:194–208
phenolics were assessed for 1,1-diphenyl-2-picrylhydrazyl radical (DPPH) radical scavenging activity to ascertain its antioxidant potential.
Materials and Methods Materials All chemicals were of analytical grade. Standard xylose, birch wood xylan, endoxylanase, 3,5dinitrosalicylic acid (DNS), orcinol, and 1,1-diphenyl-2-picrylhydrazyl radical (DPPH) were purchased from Sigma-Aldrich (USA) while D-xylose assay kit and standards of 1,4-β-D-xylooligosaccharides (xylobiose, xylotriose, xylotetraose, and xylopentose) were purchased from Megazyme International Ireland (Ireland). The endo-1,4-β-xylanase (1,4-β-D -xylan xylanohydrolase, EC 3.2.1.8) from T. longibrachiatum (CAS 9025-57-4) was obtained from Sigma-Aldrich (USA). Canola meal and mustard bran were provided by Bunge Canada and GS Dunn Ltd., Canada, respectively, and were dried at 40 °C for 24 h, milled (60 mesh), and stored in polycarbonate containers. Standard sinapic acid was procured from Sigma-Aldrich (St. Louis, MO, USA). Sinapine was obtained from EPL Bioanalytical Services, IL, USA. Pretreatment of Canola Meal and Mustard Bran Alkali Treatment Canola meal and mustard bran powder were treated with 2 % NaOH solution at a solid to liquid ratio of 1:6 (w/w) for 12–15 h at room temperature with occasional stirring [11, 13]. Excess alkali was washed with deionized water till the pH of the washings was 6.0. The pretreated substrates were dried at 40 °C and stored at 4 °C until further use. Acid Treatment Canola meal and mustard bran powder were soaked in 0.1 % H2SO4 solution at a solid to liquid ratio of 1:10 (w/w) for 12 h at 60 °C [14]. The excess acid was washed with deionized water till the pH of the washings was 6.0. The pretreated substrates were dried at 40 °C and stored at 4 °C until further use. Endoxylanase Activity Assay The activity of endo-1,4-β-xylanase or endoxylanase from T. longibrachiatum was estimated using standard birch wood xylan as the substrate [15]. One unit of endoxylanase activity was defined as the amount of enzyme required to produce 1 μmol of xylose per minute under the specified conditions. Enzymatic Hydrolysis of Untreated and Pretreated Canola Meal and Mustard Bran The conditions for the production of XOS from pretreated canola meal and mustard bran followed the procedure previously reported for pretreating corncobs [11] with some modifications. The reaction was carried out in 250-ml conical flasks containing 50-ml reaction mixture, consisting of 3.0 g untreated/pretreated powder of canola meal/mustard bran and
Appl Biochem Biotechnol (2015) 175:194–208
197
700 U of endoxylanase. The volume was made up to 50-ml using Mc Ilvain buffer (0.05 M, pH 5.4). The enzymatic reaction was carried out in a shaking water bath maintained at 50 °C for specific reaction periods (1, 6, 18, and 24 h). At the end of incubation, the reaction was stopped by keeping the reaction mixture in a boiling water bath for 10–15 min, filtered using 0.45-μm membrane filters and stored at −20 °C until further use. Chemical Analysis of Untreated and Pretreated Canola Meal and Mustard Bran The pentose content was determined following a modified method of ferric-orcinol assay by Pramod and Venkatesh [16]. Monosaccharides were determined by treating pretreated and untreated canola meal and mustard bran (20 mg) with 1 ml 12 M H2SO4 at 30 °C for 30 min, then diluting to 6 ml (2 M H2SO4) followed by hydrolysis at 100 °C for 2 h. Analysis was carried out using a high-performance anion-exchange chromatograph (HPAEC) as described by Wood et al. [17]. All measurements were replicated three times. Total Phenolic Content in Enzymatic Hydrolyzates The Folin-Ciocalteu reagent method was used to measure total phenolic content with a few modifications [18]. The absorption was measured at 750 nm. A standard sinapic acid calibration graph was used to express the total phenolics in the hydrolyzates as milligrams of sinapic acid equivalents (SAE) per kilogram of the sample. This assay is also sensitive to sugars that are reactive to the Folin-Ciocalteu reagent. Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry of Oligosaccharides from CM and MB Confirmation of pentose derivatives in the enzyme hydrolyzates was carried out using a Waters Acquity ultra-performance liquid chromatography (UPLC) system coupled to a Quattro micro API tandem mass spectrometer (Milford, MA, USA). A Phenomenex Synergi 4u Fusion-RP column (150×4 mm, Torrance, CA) was utilized for the separation of pentose. The mobile phase A consisted of 5 mM of ammonium acetate (pH was adjusted to 3.2 with acetic acid), and the mobile phase B was 100 % methanol. The gradient was programmed at a flow rate at 0.5 ml/min as follows: initially, phase B was set at 25 % and kept for 1 min and then increased to 95 % over 10 min in a linear manner and held for 2 min. Column was re-equilibrated for 3 min after each injection with temperature kept constant at 35 °C. Samples were maintained at 4 °C throughout the analysis. The injection volume was 10 ml. The tandem mass spectrometer was equipped with an atmospheric pressure ionization (API) probe and was operated in positive ion mode (ES+) with the conditions tuned based on each authentic standard for identification purpose. The general MS/MS parameters were as follows: capillary voltage, 3.00 kV; source temperature, 100 °C; desolvation temperature, 400 °C; cone gas (N2) flow, 50 l/h; and desolvation gas (N2) flow, 400 l/h. The cone voltage was set at 25 V, and collision energy was set at 15 eV for all analytes. Multiple reaction monitoring (MRM) mode was used to monitor the precursor to product ion transition. Daughter ion mode was used to obtain MS/ MS spectrum of their precursor ions. Mass resolution was set at maximum. The hydrolyzates were filtered through 0.45-μm cellulose nitrate membranes to remove the unreacted polysaccharide fragments before the analysis. MS analysis of the hydrolysates has been performed using underivatized samples [19]. For comparison of results, authentic standards of D-xylose and 1,4-β-D-xylo-oligosaccharides (xylobiose, xylotriose, xylotetraose, and xylopentose) were used.
198
Appl Biochem Biotechnol (2015) 175:194–208
HPLC Analysis of Phenolic Compounds The phenolic profiling of extracts was established following a reversed-phase HPLC-DAD analysis as described previously [20]. Synergi 4 μm Fusion-RP 80 Å; 150×4.0 mm–4 μm (Phenomenex, Canada) column was used for sinapic acid derivative separation (330 nm). Standards of sinapine and sinapic acid were also analyzed for comparison purpose based on retention time and UV absorption spectra. DPPH Scavenging Activity of Enzyme Hydrolyzates Antioxidant activity of the enzyme hydrolyzates was assessed using the DPPH free radical scavenging activity assay [21]. Briefly, 50 μl of enzyme hydrolysate was added to 950 μl of DPPH (0.030 mg/ml) in methanol. Then, the mixture was shaken vigorously and left in darkness for 5 min. The absorbance of the mixture was measured against DPPH blank at 515 nm by a spectrophotometer. The DPPH scavenging activity was expressed as the inhibition rate (IR). Inhibition rate ð%Þ ¼ ½1−ðAs =Ab Þ 100 where Ab is the absorbance of reagent blank, and As is the absorbance of sample. Data Expression and Analysis Data analysis was carried out using Excel and SPSS (version 18.0). One factor analysis of variance (ANOVA) and Student-Newman-Keuls (S-N-K) mean separation for multiple comparisons were used to determine significant differences (p≤0.05) between treatments.
Results and Discussion The criteria used to evaluate the efficiency of different pretreatment methods were the relative increase in xylose and pentose content and the corresponding decrease in glucose content in the pretreated substrates. The results for xylose content of untreated and pretreated canola meal and mustard bran are compared in Table 1. The xylose content of canola meal ranged from 2.07 to 3.04 g/100 g while for mustard bran from 1.73 to 2.80 g/100 g. These results indicated that alkali pretreated samples had a significantly higher relative content of xylose than untreated canola meal and mustard bran. Even though pretreated canola meal showed a higher xylose concentration than the untreated substrates, there was no significant difference observed between xylose content of the two pretreatment groups. Similar to canola meal, the xylose concentration of mustard bran also increased during pretreatment. However, its amount in alkali-pretreated mustard bran is significantly higher than its content in acid-pretreated mustard bran. Additionally, a significant relative increase of arabinose content in pretreated samples was observed in comparison to untreated substrates. The amounts of arabinose (g/100 g of substrate) in acid and alkali-pretreated canola meal were 5.79 and 6.29, respectively, while the values were 7.34 and 9.01, respectively, for mustard bran. The content of total monosaccharides (neutral sugars) in untreated canola meal and mustard bran were 14.99 and 14.86 %. The acid-pretreated canola meal and mustard bran showed a total monosaccharide content of 13.18 and 15.62 %, respectively, and the alkali-pretreated ones showed a total monosaccharide
Appl Biochem Biotechnol (2015) 175:194–208
199
Table 1 Carbohydrate composition of untreated and pretreated CM and MB Parameter
Canola meal Untreated
Mustard bran Acid pretreatment
Alkali pretreatment
Untreated
Acid pretreatment
Alkali pretreatment
Xylose
2.07±0.09 a
2.73±0.04 b
3.04±0.34 b
1.73±0.03 a
2.51±0.03 b
Arabinose
4.43±0.15 a
5.79±0.21 b
6.29±0.08 c
5.14±0.03 a
7.34±0.09 b
9.01±0.02 c
TP Glucose
6.50±0.24 a 5.30±0.16 b
8.52±0.25 b 1.99±0.09 a
9.33±0.42 c 2.00±0.17 a
6.87±0.06 a 4.16±0.19 c
9.85±0.12 b 1.09±0.03 b
11.81±0.04 c 0.78±0.00 a
Galactose
2.74±0.02 c
2.11±0.09 a
2.38±0.06 b
3.28±0.03 a
3.89±0.01 b
4.93±0.03 c
Rhamnose
0.21±0.03 a
0.24±0.00 a
0.28±0.00 b
0.35±0.02 a
0.47±0.00 b
0.53±0.00 c
Mannose TNS Pentosan
2.80±0.02 c
0.24±0.00 a
0.32±0.02 b
0.36±0.01 c
0.20±0.01 a
0.32±0.01 c
0.26±0.00 b
14.99±0.45 b
13.18±0.45 a
14.35±0.66 b
14.86±0.31 a
15.62±0.17 a
18.31±0.07 b
5.72±0.21 a
7.50±0.22 b
8.21±0.37 c
6.05±0.05 a
8.67±0.12 b
10.39±0.04 c
Mean±standard; within each row, the same lowercase letters indicated no significant difference between different treatments at the level α=0.05, namely p>0.05; n=3 TP total pentose, TNS total neutral sugars
content of 14.35 and 18.31 %. This indicated a relative reduction and a significant increase in the total neutral sugars in acid-pretreated canola meal and alkali-pretreated mustard bran, respectively, in comparison to corresponding untreated substrates. The effect of these pretreatments on the monosaccharide composition is shown in Table 1 with the pentosan content expressed as xylose equivalents (g/100 g of substrates). The weight of the pentosan in the sample was calculated by multiplying the total pentose content with a known factor (0.88), assuming that the hydrolysis during pentose assay is complete, where 0.88 accounts for the loss of water during xylosidic bond formation [14]. The relative content of pentose sugars, therefore pentosan, increased significantly during pretreatment. The initial levels of pentosan in untreated canola meal and mustard bran were 5.72 and 6.05 g xylose equivalent, respectively. These increased to 7.50–8.21 g xylose equivalent and 8.67–10.39 g xylose equivalent in canola meal and mustard bran, respectively, during pretreatment. Statistical comparison indicated a significant relative increase of pentosan in pretreated samples in comparison to untreated substrates. The amount of pentosan in acid- and alkali-pretreated canola meal were 7.50 g xylose equivalent and 8.21 g xylose equivalent, respectively, while the values were 8.67 g xylose equivalent and 10.39 g xylose equivalent, respectively, for mustard bran. Of the two types of pretreatments, the alkali pretreatment increased the relative concentration of pentose sugars in both canola meal and mustard bran more efficiently. The apparent increase in the pentose content during pretreatment was probably due to the effectiveness of pretreatments to remove some cellulose and lignin components [22, 23]. The relative increase in the xylan content of corncob during a similar pretreatment process was reported [11]. In the present study, we observed a 31.1 and 43.3 % increase in the pentose content during acid pretreatment of canola meal and mustard bran, respectively. This clearly indicated the better efficacy of acid pretreatment in canola meal matrix in comparison to mustard bran. The apparent increase in the pentose concentration, however, was more pronounced during alkali pretreatment with canola meal and mustard bran showing a 43.5 and 71.7 % increase in pentose content, respectively. This relative increase in pentose content when compared to native substrates has a positive effect on the enzymatic hydrolysis. The
200
Appl Biochem Biotechnol (2015) 175:194–208
pretreatment methods expose non-starch polysaccharide components, especially pentosans, rendering them more susceptible to endoxylanase action. The xylan of raw corncob cell wall was located within the lignin network through ester and ether lignin-carbohydrate linkages [11]. The availability of xylan directly determines the pentose content liberated from certain plant tissues. The glucose content of untreated canola meal and mustard bran were 5.30 and 4.16 %, which were significantly reduced during pretreatments (Table 1). The glucose concentration of acid- and alkali-pretreated canola meal were 1.99 and 2.00 %, respectively, while that of corresponding mustard bran samples were and 1.09 and 0.78 %, respectively. This reduction of substantial quantity of glucose can be extrapolated to the removal of cellulose, as it is a biopolymer of glucose. The pretreated substrates with such low concentration of glucose are suitable to produce oligosaccharides from pentosan. The protein contents of untreated and pretreated canola meal and mustard bran were compared (Table 2). The results indicated that alkali-pretreated samples had a significantly lower relative content of protein than untreated canola meal and mustard bran, which also reflected the efficiency of alkali pretreatment. The results also showed that the proteins had been removed from the substrates during pretreatment along with cellulose, which resulted in an apparent increase in the quantity of pentosan in alkali-pretreated canola meal and mustard bran. Even though acid-pretreated canola meal samples showed a higher concentration of protein than the corresponding untreated canola meal, there was no significant difference between the untreated and acid-pretreated mustard bran. This indicated that acid pretreatment had relatively no effect on the removal of protein. With this observation, it can be concluded that treating the substrates with dilute alkali is more efficient than acid pretreatment. Generally, similar to protein content, the concentration of oil in the substrates was also reduced considerably making the substrates suitable for enzyme hydrolysis.
Enzymatic Production of Pentose Sugars The pentose content of enzymatic hydrolyzates obtained from untreated and pretreated canola meal and mustard bran were monitored over reaction time (1, 6, 18, and 24 h) (Table 3). The total pentose content of enzyme hydrolyzates was used as an indicator for the amount of oligosaccharides produced from untreated and pretreated canola meal and mustard bran. The total pentose contents gave only a rough estimate of oligosaccharides because the hydrolyzates might contain xylose and arabinose in addition to the oligomers, and it would also contribute toward total pentose content. An increased release of pentose sugar was observed for both
Table 2 Moisture, protein, and oil contents of untreated and pretreated canola meal and mustard bran Parameter (g/100 g)
Canola meal Untreated
Moisture Protein Oil
Mustard bran Acid pretreated
Alkali pretreated
Untreated
Acid pretreated
Alkali pretreated
9.00±0.69 b
9.50±0.55 b
7.48±0.30 a
7.58±0.62 a
8.78±0.81 b
36.88±0.08 b
42.78±0.05 c
29.19±0.32 a
19.56±0.34 a
19.25±0.07 a
12.14±0.05 b
7.87±0.18 ab
4.64±0.27 b
4.19±0.51 b
2.92±0.28 a
24.66±1.38 c
15.59±2.56 b
11.21±0.64 a
Mean±standard; within each row, the same lowercase letters indicated no significant difference between different pretreatments at the level α=0.05, namely p>0.05; n=3
Appl Biochem Biotechnol (2015) 175:194–208
201
Table 3 Enzymatic production of pentoses from canola meal and mustard bran Reaction Pentose content (g/100 g) time (h) Canola meal Untreated
Acid pretreated
Mustard bran Alkali pretreated
Untreated
Acid pretreated
Alkali pretreated
1
1.00±0.261 aAB 0.74±0.08 aA 1.39±0.18 aB
1.05±0.14 aA 1.20±0.02 aAB 1.35±0.04 aB
6
1.30±0.349 aA
1.22±0.03 bA 1.83±0.09 bB
1.40±0.15 bA 1.78±0.11 bB
2.31±0.19 bC
18
1.30±0.056 aA
1.71±0.08 cB 2.10±0.14 bcC 1.84±0.12 cA 2.56±0.10 cB
2.38±0.17 bB
24
1.26±0.103 aA
2.02±0.15 cB 2.35±0.13 cC
2.95±0.10 cB
2.23±0.13 dA 2.81±0.03 dB
Mean±standard; within each column and row, the same lowercase or uppercase letters indicated no significant difference between different treatments at the level α=0.05, namely p>0.05; n=3
canola meal and mustard bran for both treatments. The pentose content (g/100 g) of hydrolyzates obtained from alkali-pretreated, acid-pretreated, and untreated canola meal were 1.39– 2.35, 0.74–2.02, and 1.00–1.30, respectively, while hydrolyzates of alkali-pretreated, acidpretreated, and untreated mustard bran (g/100 g) were 1.35–2.95, 1.20–2.81, and 1.05–2.23, respectively. Within 1 h of reaction, the maximum pentose was produced from alkali-pretreated substrates wherein 1.39 and 1.34 g of pentose were liberated from 100 g of alkali-pretreated canola meal and mustard bran, respectively. This also indicated the efficacy of alkali pretreatment over acid pretreatment. With respect to acid- and alkali-pretreated canola meal, there was no remarkable increase in the pentose after 18 h of reaction, though it increased through 1 to 18 h. However, in the case of mustard bran, irrespective of the type of the pretreatments, there observed a significant difference in the yields between 18 and 24 h (α=0.05). While the change in pentose content for enzymatic hydrolyzates for canola meal control was small, the pentose content changed significantly for the mustard bran control samples. Differences in the nature of the carbohydrates could account for these differences suggesting that mustard bran polysaccharides were far more accessible to enzymatic hydrolysis compared to canola meal. It is evident that while pretreatment appeared necessary for canola meal, the yield increases for mustard bran were not that pronounced. Overall, the enzymatic hydrolyzates of alkali-pretreated canola meal and mustard bran contained more pentose with the yields further increasing over time compared to acid pretreatment. Since there were no significant increase in the amount of between 18- and 24h samples for canola meal, 18 h was selected as the optimum time. In the case of mustard bran, the pentose significantly increased within 6 h and then reached the highest at 24 h. Optimization of reaction time with respect to yield is very important from an economic perspective. Ai et al. [13] reported that the native and immobilized xylanase of Streptomyces olivaceoviridis produced 3.9 and 15.5 mg/ml of XOS, respectively, from pretreated corncob within 24-h reaction time. A reaction time of 24 h can be a bottleneck in the development of an economically viable process. Previous research by Aachary and Prapulla [11] optimized a shorter reaction (14 h) and obtained 81±3.9 % of XOS, which corresponded to 10.2± 0.14 mg/ml in the hydrolyzate. Jeong et al. [24] showed 60 °C as the optimum temperature for xylobiose production. An optimum temperature of 50 °C was also reported [11, 14]. Above 50 °C, XOS production decreased probably as a result heat inactivation of xylanases. In the present study, therefore, we carried out oligosaccharide production at 50 °C as most fungal endoxylanases are highly active
202
Appl Biochem Biotechnol (2015) 175:194–208
at this temperature. Further studies may be required to optimize conditions, such as temperature, for enzyme hydrolysis.
Total Phenolic Content of the Enzyme Hydrolyzates Enzyme hydrolyzates from the alkali-pretreated, acid-pretreated, and untreated canola meal and mustard bran samples were monitored during the production of pentose sugars, to assess the impact of enzyme hydrolysis on the release of phenolics. The results indicated the total phenolics, although some interference from sugars reacting with Folin-Ciocalteu occurred. A large number of compounds, such as other phenols, carbohydrates, aromatic amines, and ascorbic acid, were reported to interfere with Folin-Ciocalteu assay and causing an increase in the measurements. The results showed that the total phenolics (expressed as SAE) of the hydrolyzate increased with reaction time irrespective of the type of substrate or pretreatment (Table 4). At 0 h of reaction, the total phenolics of control samples of canola meal and mustard bran were significantly higher than the pretreated substrates (∼6.6 g SAE and ∼2.9 g SAE for MB per 100 g of substrates). In order to assess the effects of enzyme treatment on the release of phenolics from the substrates, the phenolic content of 0 h reaction (prior to the addition of enzyme) was subtracted from the total phenolics estimated at 1, 6, 18, and 24 h. The initial lower content of water-extractable phenolics of pretreated canola meal and mustard bran in comparison to untreated samples is attributed to the intensive removal of such phenolics during pretreatment. As endoxylanase hydrolysis increased over time, the amount of total phenolics released also increased. During the initial hour of reaction, the majority of the phenolics released appeared to be free phenolics. Generally, phenolics are heat-sensitive and would degrade at 50 °C if incubated for a long period (6–24 h). This degradation is reflected in the case of untreated substrates where a lower content of phenolics after 6 h was observed in comparison to 1 h. The content of phenolics increased significantly after 18 h. This increase might be attributed to the release of bound phenolics following cell wall degradation by endoxylanase. Since these phenolics are unstable at high temperatures for a relatively long duration, some degradation might have occurred as indicated by the reduced phenolic contents at 24 h. During the initial hour, the content of phenolics liberated in the hydrolyzates of alkali-pretreated substrates were remarkably low as most of the free phenolics may have been removed during pretreatment, and after 1 h of enzyme reaction, there was only a small release of bound phenolics. The phenolic Table 4 Total phenolic content of hydrolyzates obtained from canola meal and mustard bran Reaction time (h)
Total phenolic content (mg SAE/kg of hydrolyzate) Canola meal Untreated
Mustard bran Acid pretreated
Alkali pretreated
Untreated
Acid pretreated
Alkali pretreated
1
294.2±5.6 dC
92.6±0.1 aB
15.1±0.7 aA
460.9±7.6 bC
60.7±0.0 aB
3.2±0.7 aA
6
119.8±2.9 aB
206.2±0.5 bC
89.2±0.2 cA
425.7±36.4 aC
201.2±0.3 bB
50.6±0.4 bA
18
250.6±2.9 cB
288.3±8.2 dC
75.6±1.8 bA
540±8.3 cC
213.8±1.3 bB
51.3±0.2 bA
24
127.3±2.6 bB
220.9±2.4 cC
77.2±0.6 bA
523±7.7 cC
204.8±0.9 bB
70.2±0.1 bA
Mean±standard; within each column and row, the same lowercase or uppercase letters indicated no significant difference between different treatments at the level α=0.05, namely p>0.05; n=3
Appl Biochem Biotechnol (2015) 175:194–208
203
content then increased significantly up to 24 h, indicating the release of bound phenolics. This trend was also observed with acid pretreated samples. The present study mainly focused on the production of pentose sugars from the hemicelluloses of mustard bran and canola meal using an endoxylanase enzyme. The optimization of this process would certainly open up new areas of research for the simultaneous production of oligosaccharides and phenolics. Recently, the simultaneous production of XOS and phenolic acids (hydroxycinnamic and hydroxybenzoic acids) from hemicelluloses using crude enzyme mixture synthesized by a novel Bacillus subtilis KCX006 was reported [25]. The enzyme mixture was tested for simultaneous production of XOS and phenolic acids from xylan-rich plant biomass, such as wheat bran, sugarcane bagasse, bamboo, and rise husk. In this context, the present study provides preliminary data on the simultaneous production of oligosaccharides and phenolics from canola meal and mustard bran. The effect of enzyme treatment on the sinapine content in the hydrolyzates was also studied, but not reported here.
UPLC-MS Analysis of Oligosaccharides Produced from Canola Meal and Mustard Bran Glucuronoxylan (GX) is one of the major polysaccharides present in plant cell walls [26]. It is composed of β-(1→4)-linked xylopyranose (Xyl) residues forming linear chains. These xylan chains are frequently substituted by terminally linked glucuronic acid (GlcA) residues or by 4O-methyl derivative (MeGlcA), arabinofuranosyl (Ara) residues, and/or acetyl groups. Several studies identified neutral oligosaccharides as unbranched xylosyl residues with varied chain length (DP0.05; n=3
Appl Biochem Biotechnol (2015) 175:194–208
207
identify the phenolics attached to the oligosaccharides produced from canola meal and mustard bran. Similarly, production of XOS from autohydrolysis liquors of wheat straw and sunflower stalk, as well as the antioxidant activity of these autohydrolysis liquors were reported [30]. The scavenging activities of enzymatic hydrolyzates obtained from alkali-pretreated, acidpretreated, and untreated canola meal (IR, %) were 9.8–28.2, 22.1–57.2, and 60.8–73.9, respectively, while for mustard bran (IR, %), they were 2.1–5.9, 5.8–19.3, and 47.3–68.7, respectively. It can be summarized that the pretreatment methods, especially the alkali treatment, significantly reduced the sinapine in comparison with control groups. Xyloglucuronic acid, the major pentose derivative in the hydrolyzates, could also contribute toward the antioxidant activity.
Conclusions Both canola meal and mustard bran were successfully value added by producing oligosaccharides from the non-starch polysaccharide components of the same, following an enzymatic approach. Pretreating substrates with mild alkali/acid improved the efficiency of enzyme hydrolysis and thereby, oligosaccharide production. Alkali pretreatment was superior to acid pretreatment for making its pentosan content available for hydrolytic breaking. The use of mild pretreatment conditions together with use of food grade microbial endoxylanases makes the process very much environmentally friendly. Xylo-glucuronic acid was the major pentose derivative, and sinapine was the principal phenolic identified in the enzymatic hydrolyzates. The DPPH antioxidant activities of the enzymatic hydrolyzates were attributed to their reactive sugars and total phenolic content. Their yields can be further improved by (a) using a combination of cell-wall-degrading enzymes and (b) by optimizing the parameters of enzyme hydrolysis. The enzyme hydrolyzates can be either purified to obtain oligosaccharides and phenolics separately or freeze-dried to obtain a crude powder, which can be applied to various food systems to impart antioxidant and other bioactive benefits upon consumption. This is the first ever study conducted with respect to the production of oligosaccharides from canola meal and mustard bran. The study would encourage large-scale utilization of these by-products and thereby, increase their market potential. Acknowledgments This study was funded by the Agri-Food Research and Development Initiative (ARDI)Growing Forward 1 Program and supported by the Bunge Canada, Canola Council of Canada and GS Dunn Ltd. The authors would like to thank Dr. Steve Cui and his technician Ms. Cathy Wang from the Guelph Food Research Centre, Agriculture and Agri-Food Canada for carrying out the neutral sugar analysis. The authors also would like to thank Hai Feng, Department of Human Nutritional Sciences, University of Manitoba (Winnipeg, MA, Canada) for technical assistance with UPLC-MS. Conflict of Interest There is no conflict of interest between the authors to declare.
References 1. Canada’s Farm Income Forecast for 2011–2012. www.agr.gc.ca. Accessed 4 Feb 2014. 2. Animal Feed Resources Information System (2002). www.fao.org. Accessed 21 Dec 2013. 3. Zui, F., Kaneko, S., Kuno, A., Kobayashi, H., Kusakabe, I., & Mizuno, H. (2004). Journal of Biological Chemistry, 279(10), 9606–9614. 4. Ishihara, M., Nojiri, M., Hayashi, N., Nishimura, T., & Shimizu, K. (1997). Enzyme and Microbial Technology, 21, 170–175.
208
Appl Biochem Biotechnol (2015) 175:194–208
5. Okazaki, M., Fujikawa, S., & Mastumoto, N. (1990). Bifidobacteria and Microflora, 9, 77–86. 6. Nabarlatz, D., Farriol, X., & Montane, D. (2005). Industrial and Engineering Chemistry Research, 44, 7746– 7755. 7. Moure, A., Gullon, P., Dominiguez, H., & Parajo, J. C. (2006). Process Biochemistry, 41(9), 1913–1923. 8. Loo, J. V., Cummings, J., Delzenne, N., Englyst, H., Franck, A., Hopkins, M., Kok, N., Macfarlane, G., Newton, D., Quigley, M., Roberfroid, M., van Vliet, T., & van den Heuvel, E. (1999). British Journal of Nutrition, 81, 121–132. 9. Toshio, I., Noriyoshi, I., Toshiaki, K., Toshiyuki, N. and Kunimasa, K. (1990). Japan Patent JP, 2119790. 10. Vazquez, M. J., Alonso, J. L., Domınguez, H., & Parajo, J. C. (2000). Trends in Food Science & Technology, 11, 387–393. 11. Aachary, A. A., & Prapulla, S. G. (2009). Bioresource Technology, 100(2), 991–995. 12. Naczk, M., Amarowicz, R., Sullivan, A., & Shahidi, F. (1998). Food Chemistry, 62(4), 489–502. 13. Ai, Z., Jiang, Z., Li, L., Deng, W., Kusakabe, I., & Li, H. (2005). Process Biochemistry, 40, 2707–2714. 14. Yang, R., Xu, S., Wang, Z., & Yang, W. (2005). LWT-Food Science and Technology, 38, 677–682. 15. Jiang, Z., Yang, S., Yang, Q., Li, L., & Tan, S. (2005). World Journal of Microbiology and Biotechnology, 21(6–7), 863–867. 16. Pramod, S. N., & Venkatesh, Y. P. (2006). Glycoconjugate Journal, 23(7–8), 481–488. 17. Wood, P. J., Weisz, J., & Blackwell, B. A. (1994). Cereal Chemistry, 71(3), 301–307. 18. Folin, O., & Ciocalteu, V. (1927). Journal of Biological Chemistry, 73, 627–650. 19. Reis, A., Pinto, P., Evtuguin, D. V., Neto, C. P., Domingues, P., Ferrer-Correia, A. J., & Domingues, M. R. M. (2005). Rapid Communications in Mass Spectrometry, 19, 3589–3599. 20. Khattab, R., Eskin, M., Aliani, M., & Thiyam, U. (2010). Journal of the American Oil Chemists Society, 87(2), 147–155. 21. Karakaya, S., & Simsek, S. (2011). Journal of the American Oil Chemists Society, 88, 1361–1366. 22. Parajo, J. C., Alonso, J. L., & Santos, V. (1995). Bioresource Technology, 51, 153–162. 23. Saska, M., & Ozer, E. (1995). Biotechnology and Bioengineering, 45, 517–523. 24. Jeong, K. J., Park, I. Y., Kim, M. S., & Kim, S. C. (1998). Applied Microbiology and Biotechnology, 50, 113–118. 25. Reddy, S. S., & Krishnan, C. (2013). Food Biotechnology, 27(4), 357–370. 26. Selvendran, R. R. (1985). Journal of Cell Science - Supplement, 2, 51–88. 27. Havlicek, J. E., & Samuelson, O. (1972). Carbohydrate Research, 22, 307–316. 28. Jacobs, A., Larsson, P. T., & Dahlman, O. (2001). Biomacromolecules, 2(3), 979–990. 29. Veenashri, B. R., & Muralikrishna, G. (2011). Food Chemistry, 126(3), 1475–1481. 30. Akpinar, O., Gunny, K., Yilmaz, Y., Levent, O., & Bostanci, S. (2010). BioResource, 5, 699–711.
