Optimization of the Lipase Catalyzed Production of Structured Acylglycerols With Polyunsaturated Fatty Acids Isolated From Sardine Oil

In the present work, direct enzyme-catalyzed esterification of n-3 polyunsaturated fatty acids (n-3 PUFA) isolated from sardine oil was optimized to obtain structured acyglycerols. A n-3 PUFA concentrate was prepared by urea crystallization of refined sardine oil and esterification was carried out mixing free fatty acids and glycerol at different molar ratios (M = 0.48, 1.5, 3.0, 4.5 and 5.52 mol/mol), using an immobilized lipase preparation from Candida antarctica (NV-435) at different temperatures (T = 38, 45, 55, 65 and 72 °C) and reaction times (t = 0.7, 2.75, 5.75, 8.75 and 10.8 h) in a rotatable central composition design. The degree of esterification was determined by analysis of the acylglycerides produced, using liquid chromatography (HPLC-ELSD). Optimization by response surface methodology (RSM) showed that in order to obtain higher esterification levels of n-3 PUFA to glycerol (99.5%), a molar ratio of 1.3 mol n-3 PUFA/mol glycerol, time 8.3 h and temperature 38 °C, are required. However, results of this work show that it is possible to drive the reaction to any determined product (MAG, DAG or TAG) by modifying the reaction conditions.


Introduction
N-3 polyunsaturated fatty acids (n-3 PUFA) are essential fatty acids because they cannot be synthesized by humans and animals; their tissues lack the enzymatic mechanism to insert double bonds before carbon nine from the end methyl group (Makrides, Neumann, & Gibson, 1996).The well documented health beneficial effects of n-3 PUFA for cardiovascular disease, rheumatoid arthritis, immune function and cancer (Mantzioris, Cleland, & Gibson, 2000;Li, Bode, Drummond, & Sinclair, 2003;Cleland, Caughey, James, & Proudman, 2006) have promoted the rapid development of the nutraceutical and pharmaceutical markets (Young, 2003).One of the most important effects of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids is the prevention of arrhythmias (Sellmayer & Koletzco, 1999); furthermore, researchers have concluded that n-3 PUFA can reduce the amount of triacylglycerides (TAG) in 25 and 30%, with a dose of 2 and 3 g/day (Harris & Isley, 2001).
Fish oil is well known for its high content of n-3 fatty acids (FA), with documented benefits on human health (Nettleton, 1994).DHA has proven capable to reduce the risk of heart disease and inflammatory cytokines (Simopoulos, 2002;Von Schacky, 2007).Additionally, DHA inhibits tumour cells growth (Zhang, Long, Zhang, & Wang, 2007) and it can be consumed as acylglycerol for its moderate absorption and less oxidation compared to the free fatty acid form (Valenzuela, Valenzuela, Sanhueza, & Nieto, 2005).Different methods have been developed to obtain n-3 PUFA extracts (as acylglicerides or free fatty acids) from marine oils; some of those methods have been combined to increase both efficiency and yield of the extraction (Gamez, Noriega, Medina, Ortega, García, & Angulo, 2003).n-3 PUFA are easily oxidized, for that reason enzymatic reactions have been studied for the production of oils with elevated content of n-3 PUFA, because these reactions are conducted in mild conditions (Haraldsson, Kristinsson, Sigurdardottir, Gudmundsson, & Breivik, 1997).The enzyme-catalyzed enrichment of fish oil with n-3 PUFA have been carried out via transesterification, direct esterification of glycerol with EPA and DHA, interesterification of tributyrin with ethyl esters of EPA and DHA (Haraldsson, Gudmundson, & Almarsson, 1995) and the direct esterification of the free fatty acid (FFA) from tuna oil with alcohols (Shimada, Sugihara, Nakano, Kuramoto, Nagao, Gemba, & Tominaga, 1997).The aim of the present work was the optimization, employing response surface methodology (RSM), of the enzyme-catalyzed esterification of the n-3 PUFA isolated from sardine oil.

Concentrate Extracts of n-3 PUFA
Fresh crude oil from whole sardine (Sardinops sagax caeruleus) was obtained from a fishmeal plant located at Guaymas, Mexico.Refining (R), bleaching (B), and deodorizing (D) of the sardine oil was carried out according to recommended procedures for fish oil (Noriega, Ortega, Angulo, García, Medina, & Gámez, 2009).To obtain the n-3 PUFA from RBD sardine oil, we followed the method described by Gamez, Noriega, Medina, Ortega, García and Angulo (2003).ca. 100 g of RBD sardine oil (containing 0.02 g of TBHQ) were saponified with 200 mL 7M KOH in ethanol (70%) under reflux at 90 °C for 1h.The saponifiable fraction was extracted with distilled water (240 mL) while the unsaponifiable material was extracted with hexane (200 mL) and discarded.The aqueous layer containing the saponified matter was acidified to pH=1.0 with 3 N HCl, for the extraction of FFA with hexane (200 mL, twice), which was further evaporated in a rotary evaporator at 40 °C and 25 mm Hg.Anhydrous Na 2 SO 4 was added to dry the concentrate extracts.ca. 15 g of the obtained FFA were placed in Erlenmeyer flasks with urea (25 g) and ethanol (95%, 100 mL) to be heated and stirred until the whole mixture turned into a clear homogeneous solution.The mixture was transferred to centrifugue tubes and rapidly cooled by immersion in cold water, then kept refrigerated (4 °C, 8 h).Crystals were removed by centrifugation (6000xg) for 20 min at 5 °C.The supernatant was kept at -30 °C for 12 h then centrifuged again (6000xg) at -30 °C for 20 min.Non-complexing supernatant (containing the PUFA) was acidified at pH 4.0 and equal volumes of warm (65 °C) water and hexane were added and stirred thoroughly for 30 min.The n-3 PUFA concentrate was obtained after separation of the phases and evaporation of the solvent (40 rpm, 40 °C, 25 mmHg).

Analysis of Fatty Acids
FFA were transformed into the corresponding methyl esters with 12% borontrifluoride in methanol (Ce 2-66 AOCS, 2009).A Varian 3400 gas chromatograph, equipped with a flame-ionization detector analyzed the composition of FFA.The column used was Omegawax 250 (30 m x 0.25 mm i.d., 0.25 mm film thickness; Supelco, Inc., Bellefonte, PA).The oven temperature was held at 205 °C for 5 min, then increased to 240 °C at 4 °C/min and held at 240 °C for 8 min.The injector and flame ionization detector were held at 250 and 260 °C, respectively.Nitrogen was used as carrier gas at 20 cm/s flow rate.Identification of the fatty acids was based on a menhaden oil fish standard obtained from Supelco (4-7116).Heptadecanoic acid (C17:0) was used as internal standard.

Enzyme-Catalyzed Esterification
For all of the esterification reaction trials, different temperatures (T = 38, 45, 55, 65 and 72 °C), substrates molar ratios (M = 0.48, 1.5, 3.0, 4.5 and 5.52 FFA:glycerol) and reaction times (t = 0.7, 2.75, 5.75, 8.75 and 10.8 h) were employed, according to a rotatable central composition design (Montgomery, 2009).1.0 g of the substrates mixture (containing 0.02% w/w of TBHQ and molecular sieves 20% w/w of substrates) was mixed in a 10 mL glass vials and placed into an incubator with continuous shaking (220 rpm) at constant temperature.To start the reaction, 50 mg of lipase were added.The vials were flushed with nitrogen, sealed with rubber caps and parafilm, and continuously shaken for the trial time.

Analysis of Acylglicerols
Samples were withdrawn (10 L) and dissolved with 5 mL of solvent extractor CHCl 3 :CH 3 OH (2:1).Alicuots (1 mL) of this mixture were transferred to 13x100 tubes for solvent evaporation by flushing nitrogen.The remaining matter was redissolved in 1 mL hexane:2-propanol (90:10) to be analyzed by HPLC as reported previously (Liu, Lee, Bobik, Guzman, & Hastilow, 1993).The analysis was carried out using a Varian 9012 HPLC system, fitted with an ELSD 500 (Evaporative Light Scatering Detector; Alltech) using Nitrogen as nebulizer gas at 2.1 bar, and the drift tube temperature was set at 90 °C.The analytical column (250 mm x 4.6 mm ID x 5m) was a Lichrosorb Si60 (Supelco).A loop of 10 L was employed to inject of samples into the column.The chromatographic separation was carried out at 40 o C using a jacket (Alltech) connected to a recirculation water bath (Thermomix 1420; B. Braun).

Experimental Design and Statistical Analysis
Experiments were conducted using a central composite design to investigate the linear, quadratic, and cross-product effects of three factors, each varied at five levels and also included four central points for replicates.The three factors chosen were substrates molar ratio (x 1 = 0.48, 1.5, 3.0, 4.5 and 5.52 FFA/glycerol), reaction time (x 2 = 0.7, 2.75, 5.75, 8.75 and 10.8 h) and reaction temperature (x 3 = 38, 45, 55, 65 and 72 °C) for the global esterification (y 1 ) and the production of monoacylglycerols (y 2 ), diacylglycerols (y 3 ) and triacylglycerols (y 4 ).
The design of the experiments employed is depicted in Table 2. To predict the dependent variables (y i ) a polynomial regression of a second-order model was assumed as follow: where: o, i, ii and ij, represent the coefficients for combined, lineal, quadratic and interaction effects of the regression, respectively.In order to screen the effect of independent variables and to further determine the optimal conditions for the enzyme-catalyzed esterification, analysis of variance and RSM were performed respectively, with the statistical program JMPin 4.0.4(SAS Institute, Inc.).x 1 =substrate molar ratio (FFA/glycerol), x 2 =reaction time (h), x 3 =temperature (°C), y 1 = global esterification (%), y 2 =monoacylglycerols (%), y 3 =diacylglycerols (%), y 4 =triacylglycerols (%).

Results & Discussion
The concentrate of FFA contained 81.43% (w/w) of n-3 PUFA, and consisted of 52.8% DHA, 18.2% EPA and 10.4% octadecatetraenoic acid (Table 1).The remaining matter was composed by monounsaturated fatty acids, mainly oleic (7%).Table 2 lists the experimental parameter settings and the results based on the experimental design.

Global Esterification (GE)
The extent of esterification reached 93% for a 1.5 FFA/glycerol substrates molar ratio, 8.75 h and 45 °C (Table 2).Esteban, Robles, Jiménez, Ibáñez & Molina (1998), used a cod liver oil concentrate and obtained 68.5% of esterification after 48 h and for the following 48 h, it only increased by 5.3%.In other study, 92.5% of esterification was reported by Robles, Esteban, Giménez, Camacho, Ibáñez & Molina (1999) after 24 h.The second order polynomial regression to experimental data generated the following model for esterification: where y 1 is the extent of global esterification, x 1 the substrates molar ratio, x 2 the reaction time and x 3 the temperature.
Figure 1 shows the response surface generated for the experimental data of the global esterification.It can be observed that the greatest extent of global esterification could be obtained with the lower substrates molar ratio (0.48 FFA/glycerol) and temperature (38 °C), after 5 h of reaction.Low temperatures have been suggested for esterification of n-3 PUFA in order to prevent polymerization during the reaction (Kosugy & Azuma, 1994). www.ccsen

Produc
Even

Table 2 .
Experimental setup for enzyme-catalyzed esterification of n-3 PUFA of sardine oil