Food Chemistry 155 (2014) 140–145

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Simultaneous production of fatty acid methyl esters and diglycerides by four recombinant Candida rugosa lipase’s isozymes Shu-Wei Chang b,⇑, Myron Huang c,1, Yu-Hsun Hsieh d, Ying-Ting Luo e, Tsung-Ta Wu f, Chia-Wen Tsai g,1, Chin-Shuh Chen c, Jei-Fu Shaw a,c,d,h,i,⇑ a

Department of Biological Science & Technology, I-Shou University, Kaohsiung City 84001, Taiwan Department of Medicinal Botanicals and Health Applications, Dayeh University, Changhua 515, Taiwan Department of Food Science and Biotechnology, National Chung Hsing University, Taichung 402, Taiwan d Department of Life Science, National Chung Hsing University, Taichung 402, Taiwan e Department of Bioindustry Technology, Dayeh University, Changhua 515, Taiwan f Department of Chemical Engineering, National Chung Hsing University, Taichung 402, Taiwan g Department of Nutrition, China Medical University, No. 91, Hsueh-Shih Rd., Taichung 404, Taiwan h Agricultural Biotechnology Research Center, Academia Sinica, Nankang, Taipei 11529, Taiwan i Agricultural Biotechnology Center, National Chung Hsing University, Taichung 402, Taiwan b c

a r t i c l e

i n f o

Article history: Received 16 August 2013 Received in revised form 2 December 2013 Accepted 14 January 2014 Available online 23 January 2014 Keywords: Candida rugosa lipase Diacylglycerol Emulsifier Esterification Fatty acid methyl ester Isozymes

a b s t r a c t In this study, the catalytic efficiency of four recombinant CRL (Candida rugosa lipase) isozymes (LIP1–LIP4) towards the production of fatty acid methyl ester (FAME) was compared and evaluated as an alternative green method for industrial applications. The results indicated that the recombinant C. rugosa LIP1 enzyme exhibited the highest catalytic efficiency for FAME production compared to the recombinant C. rugosa LIP2–LIP4 enzymes. The optimal conditions were as follows: pH 7.0, methanol/ soybean oil molar ratio: 3/1, enzyme amount: 2 U (1.6 lL), reaction temperature: 20 °C, 22 h of reaction time, and 3 times of methanol addition (1 mol/6 h), and resulted in 61.5 ± 1.5 wt.% of FAME conversion. The reaction product contained also 10 wt.% of DAG with a ratio of 1,3-DAG to 1,2-DAG of approximately 4:6, and can be potentially used in industrial applications as a food emulsifier. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Candida rugosa (formerly Candida cylindracea) lipase (CRL), an mixture of many individual lipase isoforms, is a versatile lipolytic enzyme which has been widely used to catalyze hydrolysis, esterification, alcoholysis, acidolysis and transesterification reactions yielding free fatty acids, diacylglycerol (DAG), monoacylglycerol (MAG), glycerol and/or specific ester compounds which have many biotechnological applications (Akoh, Lee, & Shaw, 2004; Benjamin & Pandey, 1998). Among these, methyl esters of long chain fatty acids are useful industrial products for food, pharmaceutical, cosmetic and biodiesel applications (Farris, 1979). DAG, a nontoxic ⇑ Corresponding authors. Addresses: Department of Medicinal Botanicals and Health Applications, Dayeh University, Changhua, 515, Taiwan. Tel.: +886 4 8511888x2294; fax: +886 4 8511326 (S.-W. Chang), Department of Biological Science & Technology, I-Shou University, Kaohsiung City 84001, Taiwan. Tel.: +886 7 6577 001; fax: +886 7 6577 051 (J.-F. Shaw). E-mail addresses: [email protected] (S.-W. Chang), [email protected] (J.-F. Shaw). 1 Myron Huang and Chia-Wen Tsai contributed equally to this work. http://dx.doi.org/10.1016/j.foodchem.2014.01.035 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.

product of enzymatic hydrolysis and/or reverse reactions, is useful as food emulsifier, for treating postprandial hyperlipidemia and for preventing excess adiposity (Yanai et al., 2007). These are produced by the methanolysis and are currently of great interest due to the fact that are renewable and biodegradable, and have non-toxic properties. The biochemical properties (e.g. high levels of chemo-specificity, enantio- and regio-selectivity, etc.) of lipases often result in the production of different percentage of products during chemical reactions with a wide range of natural and synthetic substrates. For example, 1,3-specific lipase can be used to hydrolyse triacylglycerols (TAG) of plant oils to produce DAG (composed of 1,2-DAG and 1,3-DAG), in particular 1,3-DAG (1,3diacyl-sn-glycerol), synthetically by acyl migration. This 1,3-DAG oil can be easily incorporated into food products and has the ability to increase b-oxidation, to enhance body weight loss, to suppress body fat accumulation, and to lower serum triacylglycerol levels postprandially (Rudkowska et al., 2005). Also, different individual CRL isoform (recombinant LIP1–LIP4) has been shown to display quite different catalytic efficiency, substrate specificity and thermostability (López et al., 2004). The utilization of such unique

S.-W. Chang et al. / Food Chemistry 155 (2014) 140–145

biocatalyst (lipases) to synthesize specific lipid/ester products using abundant natural substrates, such as plant oils, under mild conditions has attracted considerable attention in recent years (Chang, Shieh, Lee, & Shaw, 2005; Iso, Chen, Eguchi, Kudo, & Shrestha, 2001). In previous studies, Linko et al. (1998) demonstrated that a mixture of 2-ethyl-1-hexyl esters, which can be used as a solvent, can be obtained from rapeseed oil fatty acids with a good conversion (97%) using a C. rugosa lipase-catalysed transesterification reaction. De, Bhattacharyya, and Bandhu (1999) investigated the conversion of various fatty alcohol esters (C4–C18:1) using immobilized Mucor miehei lipase (Lipozyme IM-20) in a solvent-free system. The molar conversion of all corresponding alcohol esters ranged from 86.8% to 99.2%. The transesterification of different triacylglycerols, which are present in sunflower oil, fish oil and grease, with ethanol by ethanolysis reaction, has also been studied. In each case, high yield beyond 80% was achieved using lipases from M. miehei (Selmi & Thomas, 1998), C. antarctica (Breivik, Haraldsson, & Kristinsson, 1997) and Pseudomonas cepacia (Wu, Foglia, Marmer, & Phillips, 1999), respectively. Abigor et al. (2000) also found that in the conversion of palm kernel oil to alkyl esters using P. cepacia lipase, ethanol gave the highest conversion of 72%, while only 15% methyl esters was obtained with methanol. Recently, effective methanolysis reactions using extracellular lipase were developed by several researchers (Kaieda, Samukawa, Kondo, & Fukuda, 2001). The non-regiospecific lipases from C. rugosa, P. cepacia and P. fluorescens and the Rhizopus oryzae lipase with 1,3-regiospecificity showed significantly high catalytic ability towards methanolysis of soybean oil (Fukuda, Kondo, & Noda, 2001). In general, lipases are known to have a propensity to act on long-chain alcohols better than on short-chain ones, whereas the efficiency of the transesterification of triacylglycerols with methanol (methanolysis) is likely to be very low compared to that with ethanol in a solvent or solvent-free system (Fukuda et al., 2001). Previous studies have shown that commercial C. rugosa lipase mixtures displayed particularly high catalytic activity towards methanolysis of soybean oil (Kaieda et al., 2001). However, for CRL isozymes, crude enzyme preparations, which are essentially mixtures of many isozymes obtained from various commercial suppliers, exhibit remarkable variation in their catalytic efficiency, regioselectivity, and stereospecificity for the production of various valuable compounds. Therefore, it would be more important to screen separately all CRL isoforms in terms of reaction time and high yields of specific lipid/ester products, thus avoiding the drawbacks of soap formation, recovery of glycerol, requirement to remove salt residues, large amount of wastewater, and high energy cost, etc. (Chang, Li, Lee, Yeh, & Shaw, 2011). Recently, we have successfully expressed the recombinant LIP1, LIP2, LIP3 and LIP4 lipases in P. pastoris using multiple site-directed mutagenesis method without any methanol induction procedure (Chang et al., 2005; Lee, Lee, Sava, & Shaw, 2002; Tang et al., 2001). The pGAPZaC vector driven by the GAP (glyceraldehyde3-phosphate dehydrogenase) constitutive promoter (Lee et al., 2002) can avoid the accumulation of formaldehyde and hydrogen peroxide (oxidised products of methanol by alcohol oxidase) in P. pastoris (Van der Klei, Bystrykh, & Harder, 1990) and secretes functional recombinant LIP protein into a medium using a N-terminal peptide encoding the Saccharomyces cerevisiae a-factor secretion signal (Lee et al., 2002). LIP3 has been reported to be very useful for pitch treatment in the pulp industry due to its ability to catalyse the hydrolysis of both triacylglycerols and plant steryl esters (Tenkanen, Kontkanen, Isoniemi, & Spetz, 2002). Since we have demonstrated the different biochemical properties of individual recombinant CRL isoforms, this opens a new opportunity for their applications with unique properties for biotransformation and biocatalytic (acylation and deacylation) reactions to produce valuable

141

materials for food, flavour, fragrance, cosmetic, pharamaceutical, and other industrial applications. In the present work, we tested and compared the catalytic ability of four recombinant C. rugosa lipase isozymes (LIP1–LIP4) on FAME, MAG and DAG production. Our aims were to better understand the relationship between the reaction variables (reaction temperature, reaction time, enzyme amount, substrate molar ratio, and methanol addition times) and response (weight conversion; %) in order to establish the optimal condition for enzymatic FAME production. All products produced by the lipase reaction are discussed in terms of their possible applications for the food industry and their potential beneficial health effects. 2. Materials and methods 2.1. Chemicals All chemicals were of analytical reagent grade. Soybean oil (Taiwan Sugar Corp., Taipei) and methanol (Merck, 99.8% pure) were used for FAME biosynthesis reaction. All quantitative standards including methyl oleate (99% pure), Glyceryl trioleate (P99%), methyl heptadecanoate (95% pure), 1,2-diolein (P99%), 1,3-diolein (P99%), monoolein (P99%), and oleic acid (99% pure) were purchased from Sigma Chemical Co. (St. Louis, MO, USA). For HPLC analysis, acetonitrile (P99.9%), isopropanol (P99.9%) and n-hexane (99.8% pure) were purchased from Merck Chemical Co. (Darmstadt, Germany). Molecular sieve 4 Å was purchased from Davison Chemical (Baltimore, MD, USA). 2.2. Strains and plasmids The P. pastoris expression vector pGAPZaC (Invitrogen, Carlsbad, CA, USA) was manipulated in an Escherichia coli strain, DH5a [F/ þ 80dlacDM15ZD(lacZYA-argF)U169 recA1 endA1 hsdR17ðr k ; mk Þ phoA supE44k-thi-1 gyrA96 relA1; Invitrogen], which was used as a host for cloning. The P. pastoris strain SMD168H (pep4; Invitrogen), harboring the recombinant plasmids, was used for expressing recombinant LIP1, LIP2, LIP3 and LIP4, respectively. All P. pastoris transformants were cultured in a YPD (1% yeast extract, 2% peptone, and 2% dextrose; pH 6.3) broth containing 100 lg/ml of Zeocin™ (Invitrogen) at 30 °C. 2.3. Transformation and expression The plasmids (10 lg) harboring the engineered lip1–lip4 were linearised with AvrII and then transformed into P. pastoris SMD168H by electroporation. High-voltage pulses (1.5 kV) were delivered to 100 lL samples in 0.2 cm electrode gap cuvettes using a Gene PulserÒ apparatus supplied with the Pulse Controller (Bio-Rad). Transformants were plated on YPDS (1% yeast extract, 2% peptone, 2% dextrose, 1 M sorbitol, and 2% agar, pH 7.2) plates containing 100 lg/ml Zeocin™ (Invitrogen) to isolate Zeocin™resistant clones. Individual colonies containing lipase-secreting transformants were picked and patched on 1% tributyrin-emulsion YPD plates. The clear zone on the opaque tributyrin emulsion identified the lipase-secreting transformants. P. pastoris, transformed with pGAPZaC and free of any target gene sequence, was used as a negative control. 2.4. Protein concentration determination All recombinant enzymes including LIP1, LIP2, LIP3, and LIP4 were expressed by P. pastoris system and partially purified by using 30 kDa cut-off column (Minipore). Their protein amounts were calculated as 0.61 mg/mL (LIP1), 0.32 mg/mL (LIP2), 0.19 mg/mL

S.-W. Chang et al. / Food Chemistry 155 (2014) 140–145

(LIP3), and 0.39 mg/mL (LIP4), respectively. The total protein in the samples was quantified with a Bio-Rad assay kit (Bio-Rad Laboratories, Hercules, CA) using bovine serum albumin as standard. 2.5. Enzymatic synthesis of FAME Soybean oil was purchased from Taiwan Sugar Corporation. Methanol for alkylation and n-hexane for extraction were purchased from Merck KGaA (Darmstadt Germany). The standards for gas chromatography of methyl palmitate, methyl oleate, methyl linoleate were from Sigma Chemical (St. Louis, MO, USA). The experiments were performed in 50-mL screw-cap bottles and placed in a water bath at a mixing speed of 250 rpm, which provided a precision in temperature of 0.1 °C. Soybean oil, typically 0.5 g, was weighed on precision scale balance (Analytical standard with 0.0001-g accuracy; PA214C, OHAUS, TAIWAN) and made up with various amounts of water (10–80 wt.% of soybean oil). After the appropriate temperature was reached, lipase was first mixed with soybean oil followed by stepwise methanol addition at 6/ 8 h intervals for 24 h. At the pre-established time, the reaction mixture was immediately centrifuged at 8000 rpm for 1 min. The upper layer containing FAME was then transferred to other clean bottles for further analysis by capillary gas chromatography as described below. For methanolysis reaction mixture evaluation, all experiments were similar to those previously described except that triolein was used to replace the soybean oil as substrate at 20 °C and 150 rpm for 24 h. This way, the composition of LIP1-catalyzed methanolysis products could be easily identified and quantified by HPLC. 2.6. Sampling and analysis Samples (10 mg) were taken from the final reaction mixture and 200 lL of methyl heptadecanoate (10 mg/mL in n-hexane) were added as an internal standard for further quantitative analysis. The analysis of the FAME contents in the reaction mixture was carried out with a GC-18A gas chromatograph (Shimadzu Corp., Kyoto) connected to a MXT-65TG capillary column (30 m  0.25 mm I.D.  0.1 lm; Scientific Hightek Co., Ltd., Taiwan). 1.0 lL of the treated sample was injected into the GC. The oven temperature was kept at 200–220 °C for 10 min at the rate of 2 °C/min, and held for 5 min at the same temperature. The injector and detector temperatures were set at 250 and 270 °C, respectively. The weight conversion (%) of FAMEs was defined as (mg of FAMEs/mg of initial soybean oil)  100%. All ingredients were calculated through the on-line peak area integral software of Hewlett–Packard 3365 Series II ChemStation (Avondale, PA, USA). The methanolysis by-products (i.e. MAG and DAG, etc.) in the reaction mixture were identified and quantified by high performance liquid chromatography (HPLC; JASCO, Japan) equipped with an UV detector (JASCO UV-975, Japan) and a Luna C8 analysis column (3 m) (150  4.6 mm, Phenomenex, USA). The column was equilibrated with mobile phase A (80% acetonitrile and 20% water) and mobile phase B (isopropanol) at 40 °C. At 1 mL/min of flow rate, all samples were eluted with 100% of mobile phase A for 10 min and shifted to 40% of mobile phase A and 60% of mobile phase B at 34 min followed by 6 min of holding time. The % weight of conversion was defined as the weight of methanolysis product (milligram) per total weight of triolein (milligram)  100 by using an external standard calibration method. 3. Results and discussion Four recombinant enzymes (LIP1LIP4) were expressed and screened for their biocatalytic activity and specificity on enzymatic

FAME production. The most active one was then used as the major biocatalyst to obtain the optimum condition for methanolysis reaction by using a soybean oil substrate. The effect of various factors including the water content, enzyme amount, pH, substrate molar ratio (alcohol/oil), reaction temperature, reaction time, and alcohol addition time, on the bioconversion yield of FAME was evaluated and the results are discussed below. 3.1. Effect of water content on FAME production For the alkaline-catalysed production of FAME, the glycerides and alcohol must be substantially anhydrous since the existence of water might cause a partial saponification. The soap reduces the catalytic efficiency and causes an increase in viscosity, the formation of gels, and difficulty in achieving separation of glycerol (Wright et al., 1944). In contrast, for lipase-catalysed production of FAME, trace amount of water could maintain the protein configuration responsible for the catalytic function of enzyme, however, the excess water might stimulate the competing hydrolysis reaction to produce unwanted byproducts. Therefore, the optimization of water content is essential to improve the enzyme activity and production yield. As shown in Fig. 1, all recombinant CRL isozymes (LIP1LIP4) need more water (at least 30%) to maintain their configuration and catalytic activity which is effective in alleviating lipase inactivation from methanol for higher conversion of TG to FAME (Kaieda et al., 2001; Pizarro & Park, 2003). The recombinant LIP2 showed the highest conversion yield of FAME (43.2 ± 2.1%) which was 1.26-, 3.69- and 1.92-fold higher than that of recombinant LIP1 (34.4 ± 3.3%), LIP3 (11.7 ± 1.9%) and LIP4 (22.5 ± 2.5%) with 40% of H2O addition. However, when the water addition content increased (>40%), there was no significant improvement on the FAME conversion yield. This indicated that the excess water content may promote an enzyme catalysed reverse hydrolysis reaction, thereby reducing the rate of lipase-catalysed esterification and increasing the difficulty and cost of product recovery. For these reasons, 30% of water content was used in all subsequent experiments. 3.2. Effect of temperature on FAME production For the effect of reaction temperature, several trials were conducted to elucidate the most suitable reaction temperature for 50

Weight Conversion of FAME (%)

142

40

30

20

10 LIP1 LIP2 LIP3 LIP4

0

10

20

30

40

50

60

70

80

Water content (%) Fig. 1. Effect of water addition content on FAME conversion catalysed by recombinant LIP1 (d), LIP2 (s), LIP3 (.), and LIP4 (5), respectively. Each enzyme was added at 2 U of total activity and the reaction conditions were: 0.5 g soybean oil, pH 7.0, 3 mol of methanol added stepwise in three times, 37 °C for 24 h. Each amount of recombinant protein was calculated as 4.7 mg/mL (LIP1), 1.23 mg/mL (LIP2), 1.74 mg/mL (LIP3), and 1.27 mg/mL (LIP4), respectively.

143

S.-W. Chang et al. / Food Chemistry 155 (2014) 140–145

3.3. Effect of pH on FAME production To determine the pH preference of our recombinant CRL isozymes (LIP1 and LIP2), a broad pH range from 2.0 to 11.0 was therefore utilised to compare their catalytic efficiency on methanolysis reaction at 20 and 37 °C for 24 h (Fig. 3). The result shows that LIP2 has a broad pH preference from 4.0 to 11.0 with about 30% of FAME yield. The most suitable pH for recombinant LIP1 was between pH 3.0 and 7.0, with about 50% of FAME yield, whereas there was a considerable drop in FAME formation rate with a pH higher than 7.0. The highest wt.% of FAME (53.7 ± 0.2%) was therefore obtained by using LIP1 as biocatalyst at pH 6.0. Unlike chemical catalysts, all the enzyme-catalysed reactions were performed under mild conditions to maintain the active forms of the enzymes. This means that the degree of conversion esters was significantly affected by the enzyme conformational states. For example, a stronger nucleophile attack of the active-site serine on the carbonyl carbon of the substrate would benefit the conversion rate by the proper position of histidine and glutamic acid residues. Regarding to the effect of pH, it was found that the recombinant LIP1 exhibited an average 22% of FAME conversion yield higher than that of LIP2 at pH 5.0 and 6.0, whereas LIP2

60

Weight conversion of FAME (%)

recombinant LIP1 and LIP2. A broad temperature range from 10 to 60 °C was employed in these trials and the highest weight conversion yields of 50.0 ± 2.6% (LIP1) and 40.2 ± 2.0% (LIP2) was obtained at 20 and 40 °C, respectively (Fig. 2). It is obvious that the recombinant LIP1 showed 1.3-fold higher of weight conversion yield than that of LIP2 at 20 °C. Generally, enzymes are more stable at room temperature. As shown in Fig. 2, the recombinant LIP2 showed the highest production yield between 37 and 40 °C, which is similar to several previous reports (Kaieda et al., 2001; Kojima, Du, Sato, & Park, 2004; Lee et al., 2002). In contrast, LIP1 showed a lower optimum temperature range, from 10 to 20 °C, for FAME production which indicates that the native structure of LIP1 was not damaged at lower temperature. The lower hydrophobicity of the LIP1’s lid domain containing fewer hydrophobic amino acids (i.e. valine, leucine and isoleucine, etc.) than that of LIP2 might result in decreasing the lipase interfacial binding in the active site. Therefore, the solvent and substrate can easily get into the catalytic triad to increase the conversion rate of FAME at a low temperature (Jeong et al., 2002).

40

30

20

10

0

-10 0

2

4

6

8

10

12

pH Fig. 3. Effect of pH on FAME conversion catalysed by LIP1 (d) and LIP2 (s), respectively. Each enzyme was added at 2 U of total activity and the reaction condition were: 0.5 g soybean oil, 30% H2O, pH 7.0, 3 mol of methanol added stepwise in three times at (A) 20 °C or (B) 37 °C for 24 h.

was stable in a broad range from pH 4.0 to 11.0 as shown in Fig. 3. This might be due to the fact that the His-449 imidazole ring becomes protonated and positively charged as it is stabilized by the negative charge of the active-site Glu-341 residues at a conserved catalytic triad of the CRL isozymes. Although the LIP2 showed a lower conversion rate of FAME, its high tolerance ability under both extreme acidic and alkaline environment would be potentially useful for various industrial applications (Fig. 3).

3.4. Effect of substrate molar ratio on FAME production The optimum substrate concentration was measured for onestep methanolysis reaction with a varying range of methanol/oil molar ratio, from 1 to 6 (Fig. 4). As shown in Fig. 4, enzymes including lipases are generally unstable in short-chain alcohols, especially in the presence of excess methanol amount (>3 mol), since the complete conversion of triglyceride (oil) to FAME requires at least three molar equivalents of methanol (Akoh, Chang, Lee, & Shaw, 2007; Takaya et al., 2011). LIP1 showed at least 20% higher conversion yield (50.2 ± 1.2%) than LIP2 using 3:1 (alcohol: oil) of

60

70 LIP1 LIP2

50

Weight conversion of FAME (%)

Weight conversion of FAME(%)

LIP1 LIP2

50

40

30

20

10

0

LIP1 LIP2

60

50

40

30

20

10

0 0

10

20

30

40

50

60

70

Reaction temperature (oC) Fig. 2. Effect of reaction temperature on FAME conversion catalysed by recombinant LIP1 (d) and LIP2 (s), respectively. Each enzyme was added at 2 U of total activity and the reaction condition were: 0.5 g soybean oil, 30% H2O, pH 7.0, and 3 mol of methanol added stepwise in three times at 37 °C for 24 h.

0

1

2

3

4

5

6

7

Substrate molar ratio (methanol:oil) Fig. 4. Effect of substrate molar ratio on FAME conversion catalysed by LIP1 (d) and LIP2 (s), respectively. All methanol amounts were added in one step. Each enzyme was added at 2 U of total activity and the reaction conditions were: 0.5 g soybean oil, 30% H2O, pH 7.0 at 20 °C (LIP1) or 37 °C (LIP2) for 24 h.

144

S.-W. Chang et al. / Food Chemistry 155 (2014) 140–145

A

50

Weight Conversion of FAME (%)

Weight Conversion of FAME (%)

50

40

30

20

10

Add 1mole Methanol Add 2mole Methanol Add 3mole Methanol

0

B

40

30

20

10

Add 1 mole of second methanol Add 2 mole of second methanol Add 3 mole of second methanol

0 0

2

4

6

8

10

12

0

2

4

Reaction Time (h)

Weight Conversion of FAME (% )

C

6

8

10

12

14

16

18

Reaction Time (h)

70

60

50

40

30

20

Add 1 mole of fourth methanol / 6h Add 1 mole of third methanol / 6h

10 0

5

10

15

20

25

30

Reaction Time (h) Fig. 5. Effect of different methanol addition time and amount on FAME conversion catalyzed by LIP1. (A) The reaction conditions were:0.5 g soybean oil, 2 U of LIP1 (261 U/ mg), 30% H2O, pH 7.0 and 1–3 mole of methanol added in one step at 20 °C for 10 h. (B) The reaction conditions were: 0.5 g soybean oil, 2 U of LIP1 (261 U/mg), 30% H2O, pH 7.0 and 1–3 mole of the second methanol added stepwise at 20 °C for 18 h. (C) A comparison chart depicting the results from: 1 mole of the third methanol added stepwise (1 mol/6 h) at 20 °C for 20 h (s) and 1 mole of the fourth methanol added stepwise (1 mol/6 h) at 20 °C for 24 h (d). The reaction conditions were: 0.5 g soybean oil, 2 U of LIP1 (261 U/mg), 30% H2O, pH 7.0. Table 1 Composition of methanolysis reaction mixture catalyzed by recombinant C. rugosa LIP1 using triolein as substrate.

a

Reaction time (h)

Oleic acid (wt.%)a

Methyl oleate (wt.%)

Monooleate (wt.%)

1,3-Dioleate (wt.%)

1,2-Dioleate (wt.%)

6 12 18 24

8.34 ± 0.99 9.78 ± 2.57 10.06 ± 1.97 10.54 ± 1.44

4.08 ± 0.98 6.44 ± 0.79 9.59 ± 0.47 9.84 ± 1.33

0.17 ± 0.01 0.38 ± 0.09 0.63 ± 0.18 1.08 ± 0.45

2.32 ± 0.58 2.96 ± 0.90 4.08 ± 1.03 3.27 ± 0.27

2.14 ± 0.50 4.46 ± 1.13 5.94 ± 1.19 5.52 ± 0.58

% Weight conversion = weight (mg) of methanolysis product/total weight (mg) of triolein  100.

stoichiometric molar ratio at 20 °C for 24 h. The lower production yield of FAME and the bigger SD values of some data in Fig. 4 might be due to the higher instability of LIP2 in methanol, which is consistent with our previous reports (Chang, Lee, & Shaw, 2006; Chang et al., 2011). Therefore, the recombinant LIP1 was used in the following experiments. 3.5. Effect of stepwise methanol addition time and reaction time on FAME production The time course of the methanolysis reaction catalyzed by LIP1 was studied in a series of trials for 24 h. Since the enzymes are easily inactivated in 100% of methanol, a stepwise methanol addition

reaction system was therefore developed to avoid enzyme inactivation by short-chain alcohols (Shimada, Watanabe, Sugihara, & Tominaga, 2002). All reactions were started by adding methanol with a lower soluble methanol amount (8 h) (Fig. 5(B)). This might be due to a conformation change induced by the microenvironment of the reaction mixture or the methanol toxic effect leading to enzyme inactivation (Watanabe & Shimada, 2010). The

S.-W. Chang et al. / Food Chemistry 155 (2014) 140–145

step-wise methanol addition process appears to be a feasible way to reduce the activity loss of LIP1 and obtain a higher yield of FAME (>60%) at 20 °C for 22 h (Fig. 5(C) and (D)). Based on the results, we concluded that the recombinant LIP1 was the most efficient CRL isoform for FAME production in comparison with the other three CRL isozymes (LIP2–LIP4). To solve the methanol toxicity problem, the stepwise addition strategy was therefore developed and applied to obtain the optimal methanol addition frequency for LIP1. Fig. 5(C) shows that 2 times of methanol addition at an interval of 6 h (2 mol of methanol divided in 2 parts) was the optimal frequency and resulted in nearly 50.7 ± 0.8 wt.% FAME conversion at 20 h. Furthermore, the highest FAME conversion yield, 61.5 ± 1.5 wt.%, was obtained by adding 3 times the methanol addition at an interval of 6 h (3 mol of methanol divided in 3 parts) at 20 °C for 22 h (Fig. 5(D)). On the other hand, other valuable products were simultaneously produced. More specifically, the reaction mixture contained 10 wt.% of DAG (1,3-DAG and 1,2-DAG ranging from 2.3% to 4.1% and 2.1% to 5.9%, respectively) at 20 °C for 18 h (Table 1). 1,3 DAG oil, has been suggested to have a health benefit in the treatment of obesity and/ or diseases and can be easily incorporated into food products (Rudkowska et al., 2005). Such a one-step process can be devised economically in the future in order to minimise the production cost and waste and result in the simultaneous production of multiple valuable products. 4. Conclusions The current interest in using lipases or whole cells as ‘‘green’’ alternatives to chemical catalysts to synthesise ‘‘green products’’ offers an environmentally attractive option to the conventional process. In this study, we have screened and selected the recombinant LIP1 for the production of FAME. The optimal FAME conversion yield (61.5 ± 1.5%) at 22 h, 20 °C with methanol/oil molar ratio 4, and 3 methanol addition times (1 mol/6 h). Consequently, we found that LIP1, a major isoenzyme of various commercial C. rugosa lipase suppliers (Sigma or Amano Co.) (López et al., 2004), possesses the highest catalytic efficiency for FAME production. The use of immobilized enzyme and engineered enzyme with high resistance to methanol inactivation is currently underway to reduce the lipase cost and the lipase instability in short-chain alcohol, targeting industrial applications. For the simultaneous production of FAME and other valuable products such as 1,3-diglycerides, the optimum production condition can be designed according to these results to meet the industry’s needs. Acknowledgement We acknowledge financial support by Grant NSC-94-2313-B001-002-, NSC 98-2313-B-212-003-MY3, NSC 98-3111-Y-005001-, NSC 100-2313-B-212-001-MY3, 101-2324-B-214-001-CC1, 102-2324-B-214-001-CC1 from the National Science Council and the Ministry of Education, Taiwan, ROC under ATU plan. References Abigor, R. D., Uaudia, P. O., Foglia, T. A., Haas, M. J., Jones, K. C., Okpefa, E., et al. (2000). Lipase-catalyzed production of biodiesel fuel from some Nigerian lauric oils. Biochemical Society Transactions, 28, 979–981. Akoh, C. C., Chang, S. W., Lee, G. C., & Shaw, J. F. (2007). Enzymatic approach to biodiesel production. Journal of Agricultural and Food Chemistry, 55, 8995–9005. Akoh, C. C., Lee, G. C., & Shaw, J. F. (2004). Protein engineering and application of Candida rugosa lipase isoforms. Lipids, 39, 513–526. Benjamin, S., & Pandey, A. (1998). Candida rugosa lipases: Molecular biology and versatility in biotechnology. Yeast, 14, 1069–1087.

145

Breivik, H., Haraldsson, G. G., & Kristinsson, B. (1997). Preparation of highly purified concentrates of eicosapentaenoic acid and docosahexaenoic acid. Journal of the American Oil Chemists Society, 74, 1425–1429. Chang, S. W., Lee, G. C., & Shaw, J. F. (2006). Codon optimization of Candida rugosa lip1 gene for improving expression in Pichia pastoris and biochemical characterization of the purified recombinant LIP1 lipase. Journal of Agricultural and Food Chemistry, 54, 815–822. Chang, S. W., Li, C. F., Lee, G. C., Yeh, T., & Shaw, J. F. (2011). Engineering the expression and biochemical characteristics of recombinant Candida rugosa LIP2 lipase by removing the additional N-terminal peptide and regional codon optimization. Journal of Agricultural and Food Chemistry, 59, 6710–6719. Chang, S. W., Shieh, C. J., Lee, G. C., & Shaw, J. F. (2005). Multiple mutagenesis of the Candida rugosa LIP1 gene and optimum production of recombinant LIP1 expressed in Pichia pastoris. Applied Microbiology and Biotechnology, 67, 215–224. De, B. K., Bhattacharyya, D. K., & Bandhu, C. (1999). Enzymatic synthesis of fatty alcohol esters by alcoholysis. Journal of the American Oil Chemists Society, 76, 451–453. Farris, R. D. (1979). Methyl esters in the fatty acid industry. Journal of the American Oil Chemists Society, 56, 770A–773A. Fukuda, H., Kondo, A., & Noda, H. (2001). Biodiesel fuel production by transesterification of oils. Journal of Bioscience and Bioengineering, 92, 405–416. Iso, M., Chen, B., Eguchi, M., Kudo, T., & Shrestha, S. (2001). Production of biodiesel fuel from triglycerides and alcohol using immobilized lipase. Journal of Molecular Catalysis B, Enzymatic, 16, 53–58. Jeong, S. T., Kim, H. K., Kim, S. J., Chi, S. W., Pan, J. G., Oh, T. K., et al. (2002). Novel zinc-binding center and a temperature switch in the Bacillus stearothermophilus L1 lipase. Journal of Biological Chemistry, 277, 17041–17047. Kaieda, M., Samukawa, T., Kondo, A., & Fukuda, H. (2001). Effect of methanol and water contents on production of biodiesel fuel from plant oil catalyzed by various lipases in a solvent-free system. Journal of Bioscience and Bioengineering, 91, 12–15. Kojima, S., Du, D., Sato, M., & Park, E. Y. (2004). Efficient production of fatty acid methyl ester from waste activated bleaching earth using diesel oil as organic solvent. Journal of Bioscience and Bioengineering, 98, 420–424. Lee, G. C., Lee, L. C., Sava, V., & Shaw, J. F. (2002). Multiple mutagenesis of nonuniversal serine codons of the Candida rugosa LIP2 gene and biochemical characterization of purified recombinant LIP2 lipase overexpressed in Pichia pastoris. Biochemical Journal, 366, 603–611. Linko, Y. Y., Lamsa, M., Wu, X., Uosukainen, E., Seppala, J., & Linko, P. (1998). Biodegradable products by lipase biocatalysis. Journal of Biotechnology, 66, 41–50. López, N., Pernas, M. A., Pastrana, L. M., Sánchez, A., Valero, F., & Rúa, M. L. (2004). Reactivity of pure Candida rugosa lipase isoenzymes (Lip1, Lip2, and Lip3) in aqueous and organic media. Influence of the isoenzymatic profile on the lipase performance in organic media. Biotechnology Progress, 20, 65–73. Pizarro, A. V. L., & Park, E. Y. (2003). Lipase-catalyzed production of biodiesel fuel from vegetable oils contained in waste activated bleaching earth. Process Biochemistry, 38, 1077–1082. Rudkowska, I., Roynette, C. E., Demonty, I., Vanstone, C. A., Jew, S., & Jones, P. J. H. (2005). Diacylglycerol: Efficacy and mechanism of action of an anti-obesity agent. Obesity, 13, 1864–1876. Selmi, B., & Thomas, D. (1998). Immobilized lipase-catalyzed ethanolysis of sunflower oil in solvent-free medium. Journal of the American Oil Chemists Society, 75, 691–695. Shimada, Y., Watanabe, Y., Sugihara, A., & Tominaga, Y. (2002). Enzymatic alcoholysis for biodiesel fuel production and application of the reaction to oil processing. Journal of Molecular Catalysis B, Enzymatic, 17, 133–142. Takaya, T., Koda, R., Adachi, D., Nakashima, K., Wada, J., Bogaki, T., et al. (2011). Highly efficient biodiesel production by a whole-cell biocatalyst employing a system with high lipase expression in Aspergillus oryzae. Applied Microbiology and Biotechnology, 90, 1171–1177. Tang, S. J., Shaw, J. F., Sun, K. H., Sun, G. H., Chang, T. Y., Lin, C. K., et al. (2001). Recombinant expression and characterization of the Candida rugosa Lip4 lipase in Pichia pastoris: Comparison of glycosylation, activity, and stability. Archives of Biochemistry and Biophysics, 387, 93–98. Tenkanen, M., Kontkanen, H., Isoniemi, R., & Spetz, P. (2002). Hydrolysis of steryl esters by a lipase (Lip 3) from Candida rugosa. Applied Microbiology and Biotechnology, 60, 120–127. Van der Klei, I. J., Bystrykh, L. V., & Harder, W. (1990). Alcohol oxidase from Hansenula polymorpha CBS 4732. Methods in Enzymology, 188, 420–427. Watanabe, Y., & Shimada, Y. (2010). Processes for production of biodiesel fuel. In C. T. Hou & J. F. Shaw (Eds.), Biocatalysis and biomolecular engineering (pp. 225–241). Hoboken, New Jersey: John Wiley & Sons Inc.. Wright, H. J., Segur, J. B., Clark, H. V., Coburn, S. K., Langdon, E. E., & DuPuis, R. N. (1944). A report on ester interchange. Oil and Soap, 21, 145–148. Wu, W. H., Foglia, T. A., Marmer, W. N., & Phillips, J. G. (1999). Optimizing production of ethyl esters of grease using 95% ethanol by response surface methodology. Journal of the American Oil Chemists Society, 76, 517–521. Yanai, H., Tomono, Y., Ito, K., Furutani, N., Yoshida, H., & Tada, N. (2007). Diacylglycerol oil for the metabolic syndrome. Nutrition Journal, 6, 43–48.