Preparation of High Purity Δ5-Olefinic Acids from Pine Nut Oil Via

Preparation of High Purity Δ5-Olefinic Acids from Pine Nut Oil via Repeated Lipase-Catalyzed Esterification Heejin Kim1, Nakyung Choi2, Hak-Ryul Kim3, Junsoo Lee4, and In-Hwan Kim1,2＊ 1 Department of Public Health Sciences, Graduate School, Korea University, 145, Anam-Ro, Sungbuk-Gu, Seoul, 02841, Republic of KOREA 2 Department of Integrated Biomedical and Life Sciences, Graduate School, Korea University, 145, Anam-Ro, Sungbuk-Gu, Seoul, 02841, Republic of KOREA 3 School of Food Science and Biotechnology, Kyungpook National University, Daegu, 702-701, Republic of KOREA 4 Division of Food and Animal Sciences, Chungbuk National University, Cheongju, Chungbuk, 28644, Republic of KOREA

Abstract: Δ5-Olefinic acids have been characterized in gymnosperm plants and have been reported to have several biological health benefits. Δ5-Olefinic acids from pine nut oil were effectively concentrated by repeated lipase-catalyzed esterification. The pine nut oil contained three major Δ5-olefinic acids, namely taxoleic acid (C18:2 Δ5,9), pinolenic acid (C18:3 Δ5,9,12), and sciadonic acid (C20:3 Δ5,11,14). The fatty acids present in pine nut oil were selectively esterified with ethanol using Lipozyme RM IM from Rhizomucor miehei as a biocatalyst. The Δ5-olefinic acids were concentrated in the unesterified fatty acid fraction. The optimum molar ratio of the substrates (fatty acid:ethanol), temperature, the enzyme loading, and the reaction time were 1:7, 25℃, 5% of total substrate weight, and 6 h, respectively. There was no significant effect in the concentration of Δ5-olefinic acids when water was added in the reaction mixture. The same protocol and optimum conditions were employed for two times repeated lipase-catalyzed esterifications. In first lipase-catalyzed esterification, the Δ5-olefinic acids content in the pine nut oil increased from 17 mol% to 51 mol% with a yield of 40 mol%. In a second lipase-catalyzed esterification, with the Δ5-olefinic acids-concentrated fatty acids obtained from the first reaction as the substrate, the Δ5- olefinic acids content increased to 86 mol% with a yield of 15 mol%. Finally, a maximum Δ5-olefinic acids content of ca. 96 mol% with a yield of 6 mol% was obtained via a third lipase-catalyzed esterification.

Key words: concentration, lipase-catalyzed esterification, Lipozyme RM IM, Δ5-olefinic acids, pine nut oil

1 INTRODUCTION varies depending on the genus of the Pinaceae2）. Among Pine nut oil, one of the edible seed oils obtained from various pine species, Pinus koraiensis nuts（Korean pine genus Pinus, is a good source of unusual Δ5-olefinic acids nuts）, which are consumed as a condiment oil, have Δ5-OA （Δ5-OA）such as taxoleic acid（C18:2 Δ5,9）, pinolenic acid content of ca. 17–18％, including ca. 13％ PLA. （PLA, C18:3 Δ5,9,12）, and sciadonic acid（C20:3 Δ5,11,14）. Several researchers have reported the effects of Δ5-OA Δ5-OA, also known as Δ5-unsaturated polymethylene-in- on rats and mice. Asset et al. reported that PLA has a lipid- terrupted fatty acids, is commonly present in gymnosperm lowering effect by transforming the expression of various seed oils but they are rarely present in angiosperm seed apo genes3, 4）. Lee et al.5）reported a relationship between oils1）. The presence of Δ5-OA is characteristic of conifer dietary PLA and a cholesterol-lowering effect. According to seeds and is the basis for differentiating several genera2）. these reports, the ingestion of PLA lowers the serum low- Pinaceae family seed oils generally contain various types of density-lipoprotein（LDL）level by enhancing hepatic LDL Δ5-OA. These include 5,9-18:2（taxoleic）, 5,9,12-18:3（pino- uptake and reduces triglycerol and very-low-density-lipo- lenic）, 5,9,12,15-18:4（coniferonic）, 5,11-20:2, 5,11,14-20:3 protein levels. Ferramosca et al. and Pasman et al.6, 7）re- （sciadonic）, and 5,11,14,17-20:4（juniperonic）acids. The ported that PLA has an appetite-suppressing effect. Some fatty acid composition, particularly the Δ5-OA content, reports on the effects of the sciadonic acid from conifer

＊Correspondence to: In-Hwan Kim, Department of Public Health Sciences, Graduate School, Korea University, 145, Anam-Ro, Sungbuk-Gu, Seoul, 02841, Republic of KOREA E-mail: [email protected] Accepted August 6, 2018 (received for review July 9, 2018) Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 online http://www.jstage.jst.go.jp/browse/jos/ http://mc.manusriptcentral.com/jjocs

1435 H. Kim, N. Choi, H.-R. Kim et al.

seed oils have suggested that sciadonic acid can replace ar- 2.2 Preparation of Fatty Acids from Pine Nut Oil achidonic acid in tissues and cells. Endo et al.8）observed Pine nut oil was initially converted to fatty acids, which that torreya（Torreya nucifera）seed oil containing sciad- were used as the substrate. Pine nut oil（150 g）was added onic acid decreased triacylglycerol（TAG）levels in the to a solution of sodium hydroxide（60 g）in a mixture of dis- plasma and liver of rats. Ikeda et al.9）reported that Biota tilled water（150 mL）and ethanol（99％, 450 mL）. The orientalis seed oil containing juniperonic and sciadonic mixture was refluxed with stirring at 500 rpm for 1 h and acids lowered the contents of rat liver TAG and plasma then transferred to a 2 L separatory funnel. Distilled water cholesterol. （300 mL）was added to the saponified mixture. The Two studies on the concentration of PLA from pine nut aqueous layer containing the saponifiable material was oil by enzymatic methods have been performed in our labo- acidified by addition of aqueous 6 N HCl to pH 1.0 to ratory. It was reported that PLA was predominantly esteri- release the free fatty acids. The lower layer was discarded, fied at the sn-3 position of pine nut oil TAG10）. In the first and the upper layer containing the fatty acids was extract- study11）, Novozym 435 from Candida antarctica in the ed into n-hexane（300 mL）and washed twice with distilled presence of ethanol showed sn-3 regiospecificity with water（150 mL）. The n-hexane layer containing the fatty respect to pine nut oil TAG and PLA was concentrated in acids was dried over anhydrous sodium sulfate. n-Hexane the fatty acid ethyl ester fraction from 13 mol％ of starting was removed from the fatty acids by evaporation at 40℃ materials to 39 mol％. In the second study12）, PLA was with a rotary evaporator. The residual solvent in the fatty concentrated to 43％ in the unesterified fatty acid fraction acids was removed completely by nitrogen flushing in a via selective esterification of lauryl alcohol and fatty acids water bath at 65℃. from pine nut oil using C. rugosa lipase. There have been several reports on the lipase-catalyzed 2.3 Lipase-Catalyzed Esteri cation enrichment of n-3 PUFAs using Lipozyme RM IM from Lipase-catalyzed esterifications were performed in a sol- Rhizomucor meihei as a biocatalyst13－15）. For example, vent-free system using a 50 mL water-jacketed glass vessel. Halldorsson et al.13）reported that DHA was concentrated to The vessel was preheated to the desired temperature with ca. 71％ via the esterification of tuna oil fatty acids with a water circulator. Lipozyme RM IM was used as a biocata- glycerol using Rhizomucor miehei lipase. Baik et al.14）re- lyst. A portion of the fatty acids from the pine nut oil and ported that stearidonic acid（SDA）was concentrated by Li- 99.9％ ethanol were placed in the vessel with the desired pozyme RM IM-catalyzed esterification of echium oil fatty amount of water or without water. The reaction was initiat- acids with ethanol. Lee et al.15）reported that high-purity ed by addition of the enzyme to the substrate mixture with DHA（purity 100％）from microalgae was obtained via a li- stirring at 300 rpm. Samples（100 μL）were withdrawn peri- pase-catalyzed esterification with Lipozyme RM IM in a odically from the reaction mixture and dissolved in chloro- packed-bed reactor. form（300 μL）. Individual samples were then filtered In this study, Δ5-OA was effectively concentrated by li- through a 0.45 μm nylon microfilter（Pall Corporation, Port pase-catalyzed esterification with fatty acids from pine nut Washington, NY, USA）to completely remove the enzyme oil and ethanol using Lipozyme RM IM as a biocatalyst. To and then analyzed by thin-layer chromatography and gas optimize the lipase-catalyzed esterification for concentra- chromatography. The second and third reactions were per- tion of Δ5-OA, the molar ratio of fatty acid to ethanol, tem- formed repeatedly with the unesterified fatty acid fractions perature, and amount of added water were investigated as from the first and second reaction mixtures, respectively. parameter studies by monitoring the time course of these The repeated reactions were performed under the reactions. Lipase-catalyzed esterification was repeated optimum conditions determined for the first reaction. All three times to obtain highly purified Δ5-OA. experiments were conducted in duplicate.

2.4 Separation of Reaction Product after Lipase-Cata- lyzed Esteri cation 2 MATERIALS AND METHODS Before the second and third reactions, the unesterified 2.1 Materials fatty acid fractions were separated from the reaction mix- Pine nut oil was purchased from the K-Omega Co., Ltd. tures of the first and second reactions, respectively, by sa- （Chungcheongbukdo, Republic of Korea）. Lipozyme RM ponification and subsequent acidification. A scaled-up re- IM from R. miehei was purchased from Novozymes（Seoul, action was conducted under the optimum conditions for Republic of Korea）. Silica gel 60 for column chromatogra- the first reaction. A portion of the fatty acids（325.6 g, 1.2 phy was purchased from Merck KGaA（Darmstadt, mol）from the pine nut oil and 99.9％ ethanol（374.4 g, 8.1 Germany）. All other chemicals used in this study were ana- mol）were placed in a 1 L water-jacketed glass vessel lytical grade, unless otherwise stated. without addition of water and the vessel was preheated to 25℃. The reaction was initiated by addition of 5％ enzyme

1436 J. Oleo Sci. 67, (11) 1435-1442 (2018) Preparation of High Purity Δ5-OA

with respect to the total substrate weight, with stirring at Δ5－OA yield in unesterified fatty acid fraction（mol％） 500 rpm. The Δ5-OA -concentrated fatty acid fraction was b ＝ ×100 ［2］ separated by neutralization of the first reaction mixture, b＋c which contained free fatty acids, fatty acid ethyl esters, where a is the number of moles of fatty acids in the reac- and ethanol. The reaction mixture（100 g）was dissolved in tion mixture, b is the number of moles of Δ5-OA in the un- n-hexane（1 L）and the resulting solution was filtered esterified fatty acids in the reaction mixture, and c is the through anhydrous sodium sulfate to remove the enzyme number of moles of Δ5-OA in the fatty acid ethyl esters in and any residual water. The filtrate was collected and the reaction mixture. In thus study, Δ5-OA is defined as the transferred to a 2 L separatory funnel together with a 1.9％ sum of taxoleic acid（C18:2 Δ5,9）, PLA（C18:3 Δ5,9,12）, and （w/v）sodium hydroxide solution（200 mL）and 95％ ethanol sciadonic acid（C20:3 Δ5,11,14）. （200 mL）. The lower layer was collected, placed in a second separatory funnel, and was washed with n-hexane （1 L）. A 6 N HCl solution（40 mL）was then added to acidify the mixture, and the fatty acid fraction was recovered by 3 RESULTS AND DISCUSSION extraction with n-hexane（200 mL）. The n-hexane solution 3.1 Effects of Molar Ratio containing the fatty acid fraction was washed several times The effects of the molar ratio of the substrates on the with distilled water and then filtered through anhydrous Δ5-OA content and yield in the unesterified fatty acid frac- sodium sulfate. The solvent was removed in vacuo to give tion obtained via lipase-catalyzed esterification as a func- the fatty acid product and a stream of nitrogen gas was tion of reaction time are shown in Fig. 1. The fatty passed over the product mixture to remove any residual acid:ethanol molar ratio was tested in the range 1:1 to 1:9. solvent. For these experiments, the temperature, amount of added water, and enzyme loading were set at 25℃, 0％（ no added 2.5 Product Analysis water）, and 5％ of the total substrate weight, respectively. The sample（35 μL）was dissolved in chloroform（105 μL） At a molar ratio of 1:1, during the first 1 h of the reac- and loaded on to thin-layer chromatography plates. The tion, the Δ5-OA content in the unesterified fatty acid frac- plates were developed with a mixture of petroleum ether, tion increased from 17 mol％ of starting material to 28 diethyl ether, and acetic acid（100:20:1, v/v/v）. The plates mol％. However, no further increase in Δ5-OA content was were subsequently developed with a 2,7-dichlorofluorosce- observed when the reaction time was increased from 1 to 8 in solution（0.2％ in 95％ methanol）to enable visualization h. of the fatty acid and fatty acid ethyl ester fractions. The The maximum Δ5-OA content at each molar ratio in-

corresponding methyl esters were prepared using 14％ BF3 creased significantly with increasing molar ratio from 1:1 to in methanol. Samples（1 μL）of the extract were injected 1:5, although longer reaction times were required to into a gas chromatographic system（model 3800; Varian achieve the maximum values. In addition, there was a Inc., Palo Alto, CA, USA）equipped with a Supelcowax 10 steady increase in the maximum Δ5-OA content when the fused-silica capillary column（30 m×0.32 mm i.d.; Supelco, molar ratio was increased from 1:5 to 1:7 and the maximum Bellefonte, PA, USA）. A flame ionization detector（FID）was Δ5-OA content of ca. 51 mol％ was obtained at 6 h. used for the analysis. The injector and FID temperatures However, no significant difference was observed between were set at 240 and 250℃, respectively. The column tem- the maximum Δ5-OA contents at 1:7 and 1:9, and similar perature was held at 180℃ for 1 min and then increased to yields of ca. 40 mol％ were obtained. In addition, the time 210℃ at a rate of 1.5℃/min. Helium at a flow rate of 1.5 needed to reach the maximum Δ5-OA content at a 1:9 mL/min was used as the carrier gas and the split ratio was molar ratio was longer than that at 1:7. Gudmynder et al.16） 1:50. The fatty acid methyl esters were identified by com- reported that excess alcohol can decrease the activities paring their retention times with those of known standards. and performances of lipases because many lipases do not Heptadecanoic acid（0.2 mg）was used as an internal refer- tolerate polar conditions. A similar result15）was observed in ence standard. a previous study, in which DHA was concentrated from mi- The Δ5-OA content（mol％）in the unesterified fatty acid croalgae oil by lipase-catalyzed ethanolysis in a packed bed fraction and yield（mol％）of Δ5-OA in the unesterified fatty reactor. On the basis of these results, a molar ratio of 1:7 acid fraction were calculated as follows: was selected as the optimum condition for Δ5-OA concentration via lipase-catalyzed esterification. Δ5－OA content in unesterified fatty acid fraction（mol％） b 3.2 Effects of Temperature ＝ ×100 ［1］ a The reaction temperature is a key factor in lipase-catalyzed esterification because it effectively controls the activity and selectivity of the enzyme. In general, increasing the

1437 J. Oleo Sci. 67, (11) 1435-1442 (2018) H. Kim, N. Choi, H.-R. Kim et al.

Fig. 1 Effects of fatty acid:ethanol molar ratio on Δ5-OA Fig. 2 Effects of temperature on Δ5-OA content（a）and content（a）and Δ5-OA yield（b）in unesterified fatty Δ5-OA yield（b）in unesterified fatty acid fraction acid fraction after lipase-catalyzed esterification as after Lipozyme RM IM-catalyzed esterification as a function of reaction time. For these experiments, function of reaction time. For these experiments, the temperature, amount of added water, and the fatty acid:ethanol molar ratio, amount of added enzyme loading were kept at 25℃, 0％, and 5％ of water, and enzyme loading were kept at 1:7, 0％, the total substrate weight, respectively. All and 5％ of the total substrate weight, respectively. experiments were performed in duplicate. All experiments were performed in duplicate. temperature induces an increase in the reaction rate. effect of temperature on the concentration of DHA from However, extremely high temperatures can cause a reduc- tuna oil showed that an increase in temperature led to a tion in the enzymatic activity and irreversible denaturation decrease in the selectivity of the enzyme in DHA concen- of the enzyme protein17）. Furthermore, the optimum tem- tration with Lipozyme RM IM19）. In the temperature range perature for an enzymatic reaction varies depending on the 15–35℃, the maximum Δ5-OA content was achieved in 6 h reaction conditions, the type of enzyme, and melting points and then the Δ5-OA content in the unesterified fatty acid of the reactants18）. fraction decreased. The highest Δ5-OA contents were ob- The effects of temperature on the Δ5-OA content and tained at temperatures of 15 and 25℃, and there was no yield in the unesterified fatty acid fraction obtained via li- significant difference between the maximum Δ5-OA con- pase-catalyzed esterification as a function of reaction time tents at these temperatures. are shown in Fig. 2. The temperature range tested was The optimum temperature for an enzymatic reaction 15–45℃. For these experiments, the fatty acid:ethanol differ depending on the type of substrate, the type of reac- molar ratio, amount of added water, and enzyme loading tion, and the type of reactor, even if the same enzyme is were set at 1:7, 0％（ no added water）, and 5％ of the total used. Baik et al.14）reported that in experiments on the con- substrate weight, respectively. centration of SDA by Lipozyme RM IM-catalyzed esterifica- The Δ5-OA content decreased sharply as the tempera- tion of echium oil fatty acids with ethanol, the maximum ture decreased from 35 to 45℃. A previous study of the SDA content was achieved at 30℃. However, Rahmatullah

1438 J. Oleo Sci. 67, (11) 1435-1442 (2018) Preparation of High Purity Δ5-OA

et al.20）found that 50℃ was the optimum temperature for GLA concentration via Lipozyme RM IM-catalyzed esterification of borage oil fatty acids with 1-butanol in n-hexane as the solvent. The highest Δ5-OA content, ca. 51 mol％, was achieved at temperatures of 15 and 25℃; a temperature of 25℃, i.e., room temperature, is more efficient because of the energy costs associated with cooling. Moreover, the Δ5-OA yield at 25℃ was higher than that at 15℃ for a reaction time of 6 h. We therefore selected 25℃ as the optimum temperature for Δ5-OA concentration, because of cost considerations.

3.3 Effects of Additional Water The water content of the reaction mixture is generally a significant factor when a lipase is used as a catalyst for esterification. Lipase-catalyzed reactions require appropriate amounts of water to maintain the catalytic activity and enzyme structure21）. However, the use of too much water can cause a decrease in lipase activity and an increase in by-product formation22, 23）. The effects of additional water on the Δ5-OA content and yield in the unesterified fatty acid fraction obtained via lipase-catalyzed esterification as a function of reaction time are shown in Fig. 3. The added water content range tested was 0％ to 1％ of the total substrate weight. A charge of water was added to each reaction at the beginning of the process. For these experiments, the fatty acid:ethanol molar ratio, temperature, and enzyme loading were set at Fig. 3 Effects of added water on 5-OA content a and 1:7, 25℃, and 5％ of the total substrate weight, respective- Δ （） ly. Δ5-OA yield（b）in unesterified fatty acid fraction after Lipozyme RM IM-catalyzed esterification as As the amount of water added increased from 0％ to 1％, the maximum Δ5-OA content in the unesterified fatty acid function of reaction time. For these experiments, fraction decreased, but the time to achieve the maximum the fatty acid:ethanol molar ratio, temperature, Δ5-OA content was shorter. Furthermore, the Δ5-OA yield and enzyme loading were kept at 1:7, 25℃, and 5％ from the reaction with no added water was higher than of the total substrate weight, respectively. All experiments were performed in duplicate. those obtained with water contents of 0.25％ to 1％, when the Δ5-OA content in the unesterified fatty acid fraction reached the maximum value. The highest Δ5-OA content quently, on the basis of the Δ5-OA content and yield, no （51 mol％）was achieved in 6 h for the reaction with no added water（0％）was identified as the most effective con- added water（0％）. These results indicate that the addi- dition for Δ5-OA concentration. tional water has no effect on Δ5-OA concentration. Addi- tional water in the reaction mixture led to an increase in 3.4 Repeated Lipase-Catalyzed Esteri cations the reaction rate, but could have an adverse impact on the In the first reaction, a maximum Δ5-OA content of ca. 51 selectivity of the enzyme. mol％ with a Δ5-OA yield of 40 mol％ was achieved under The enzymatic activity depended on the amount of water the optimum conditions. in the reaction medium and the optimum amount of added For further Δ5-OA concentration, repeated lipase-cata- water varied depending on the type of reaction. In esterifi- lyzed esterifications were performed. The Δ5-OA- cations using lipase from R. miehei, additional water im- concentrated fatty acids from the first reaction were used proves the selectivity of the lipase and facilitates the reac- as the substrate in the second reaction. The Δ5-OA- tion19, 24, 25）. concentrated fatty acids from the second reaction were However, Baik et al.14）reported that the lipase from R. then used as the substrate for the third reaction. The re- miehei exhibits better activity for selective concentration peated reactions were carried out under the optimum con- of SDA from echium oil in the absence of added water. This ditions for the first reaction. result is consistent with those of our present study. Conse- The effects of repeated reactions on Δ5-OA concentra-

1439 J. Oleo Sci. 67, (11) 1435-1442 (2018) H. Kim, N. Choi, H.-R. Kim et al.

Fig. 4 Effects of second reaction on Δ5-OA content and Fig. 5 Effects of third reaction on Δ5-OA content and Δ5- Δ5-OA yield in unesterified fatty acid fraction after OA yield in unesterified fatty acid fraction after Lipozyme RM IM-catalyzed esterification as Lipozyme RM IM-catalyzed esterification as function of reaction time. For these experiments, function of reaction time. For these experiments, the fatty acid:ethanol molar ratio, temperature, the fatty acid:ethanol molar ratio, temperature, amount of added water, and enzyme loading were amount of added water, and enzyme loading were kept at 1:7, 25℃, 0％, and 5％ of the total kept at 1:7, 25℃, 0％, and 5％ of total substrate substrate weight, respectively. All experiments weight, respectively. All experiments were were performed in duplicate. performed in duplicate.

tion are shown in Fig. 4 and Fig. 5. The fatty acid:ethanol a 6 mol％ yield. molar ratio, temperature, and enzyme loading were set at 1:7, 25℃, and 5％ of the total substrate weight, respectively. In the second reaction（Fig. 4）, the Δ5-OA content in the 4 CONCLUSIONS unesterified fatty acid fraction increased significantly from High purity Δ5-OA from pine nut oil was successfully ob- 51 to 86 mol％ in 10 h, whereas the Δ5-OA yield decreased tained via repeated lipase-catalyzed esterification using Li- sharply from 40 to 15 mol％. In the third reaction（Fig. 5）, pozyme RM IM. In all reactions, Δ5-OA including taxoleic the maximum Δ5-OA content was achieved in 16 h, with a acid, PLA, and sciadonic acid were selectively concentrated Δ5-OA content of 96 mol％ and a total yield of 6 mol％. in the unesterified fatty acid fraction of reaction mixture. Through the first reaction, Δ5-OA was enriched from initial 3.5 Fatty Acid Compositions of Initial Pine Nut Oil and 16.7 to 50.8 mol％. The Δ5-OA was concentrated from 50.8 Δ5-OA Concentrates to 85.8 mol％ in the second reaction and it was raised up The fatty acid compositions of the initial pine nut oil and to 95.9 mol％ in the third reaction. As a result, we could three Δ5-OA concentrates obtained by Lipozyme RM IM- obtain maximum Δ5-OA content of ca. 95.9 mol％. catalyzed esterifications are shown in Table 1. The dominant fatty acid in the pine nut oil was linolenic acid（C18:2 Δ9,12）with a content of 47 mol％, followed by oleic acid（C18:1 Δ9）, PLA（C18:3 Δ5,9,12）, palmitic acid ACKNOWLEDGMENTS （C16:0）, taxoleic acid（C18:2 Δ5,9）, stearic acid（C16:0）, This research was supported by Basic Science Research and others. The initial pine nut oil contained three types of Program through the National Research Foundation of Δ5-OA, namely taxoleic acid（C18:2 Δ5,9）of 2 mol％, PLA Korea（NRF）, funded by the Ministry of Education（NRF- （C18:3 Δ5,9,12）of 14 mol％, and sciadonic acid（C20:3 2018R1D1A1B07049256） Δ5,11,14）of 1 mol％. Because Lipozyme RM IM discrimi- nated strongly against Δ5-OA, the Δ5-OA was concentrated in the unesterified fatty acids fraction, and the other fatty acids were selectively converted to ethyl esters. In the first References reaction, the Δ5-OA content increased to 51mol％ from 17 1） Wolff, R.L. Discussion of the term unusual when dis- mol％ in the starting material. The second and third reac- cussing delta 5-olefinic acids inplant lipids. J. Am Oil tions enabled maximum Δ5-OA content, i.e., 96 mol％ with Chem Soc. 74, 619-619（1997）.

1440 J. Oleo Sci. 67, (11) 1435-1442 (2018) Preparation of High Purity Δ5-OA

Table 1 Fatty acid composition（mol％）of pine nut oil and unesterified fatty acid fractions after lipase-catalyzed esterificationa.

Fatty acid Initialb First reactionc Second reactiond Third reactione C16:0 5.3±0.0 3.6±0.2 1.2±0.0 0.5±0.0 C18:0 2.1±0.0 1.6±0.2 0.3±0.4 0.3±0.0 C18:1 (Δ9) 27.1±0.0 14.2±0.9 4.0±0.2 1.1±0.0 C18:1 (Δ11) 0.3±0.0 0.0±0.0 0.0±0.0 0.0±0.0 C18:2 (Δ5,9) 2.2±0.0 5.9±0.0 9.1±0.1 9.1±0.1 C18:2 (Δ9,12) 46.6±0.0 28.6±0.8 8.7±0.9 2.2±0.0 C18:3 (Δ5,9,12) 13.5±0.0 41.6±0.9 69.8±1.3 77.7±0.0 C20:0 0.3±0.0 0.3±0.5 0.0±0.0 0.0±0.0 C20:1 (Δ11) 1.1±0.0 0.5±0.1 0.0±0.0 0.0±0.0 C20:2 (Δ11,14) 0.6±0.0 0.4±0.0 0.0±0.0 0.0±0.0 C20:3 (Δ5,11,14) 1.0±0.0 3.4±0.1 7.0±0.2 9.1±0.1 Total Δ5 fatty acidsf 16.7±0.1 50.8±0.8 85.8±1.5 95.9±0.0 Total Δ5 yield 100±0.0 39.2±7.1 14.8±1.1 5.8±0.1 a Values represent the average of duplicate determinations from different experiments. b Pine nut oil fatty acid was used as the substrate. c Unesterified fatty acid fraction obtained after 6 h in first lipase-catalyzed esterification under the optimum conditions. d Unesterified fatty acid fraction obtained after 10 h in second lipase-catalyzed esterification under the optimum conditions for the first reaction. e Unesterified fatty acid fraction obtained after 16 h in third lipase-catalyzed esterification under the optimum conditions for the first reaction. f Taxoleic acid (18:2 Δ5,9), PLA (Δ5,9,12), and sciadonic acid (20:3 Δ5,11,14).

2） Wolff, R.L.; Deluc, L.G.; Marpeau, A.M.; Comps, B. （2008）. Chemotaxonomic differentiation of conifer families 7） Pasman, W.J.; Heimerikx, J.; Rubingh, C.M.; van den and genera based on the seed oil fatty acid composi- Berg, R.; O’Shea, M.; Gambelli, L.; Hendriks, H.F.; Ein- tions: multivariate analyses. Trees 12, 57-65（1997）. erhand, A.W.; Scott, C.; Keizer, H.G. The effect of Ko- 3） Asset, G.; Leroy, A.; Bauge, E.; Wolff, R.L.; Fruchart, rean pine nut oil on in vitro CCK release, on appetite J.-C.; Dallongeville, J. Effects of dietary maritime pine sensations and on gut hormones in post-menopausal （Pinus pinaster）-seed oil on high-density lipoprotein overweight women. Lipids Health Dis. 7, 10（2008）. levels and in vitro cholesterol efflux in mice express- 8） Endo, Y.; Osada, Y.; Kimura, F.; Fujimoto, K. Effects of ing human apolipoprotein A-I. Br. J. Nutr. 84, 353-360 Japanese torreya（Torreya nucifera）seed oil on lipid （2000）. metabolism in rats. Nutrition 22, 553-558（2006）. 4） Asset, G.; Baugé, E.; Wolff, R.; Fruchart, J.; Dallon- 9） Ikeda, I.; Oka, T.; Koba, K.; Sugano, M.; Jie, M.S.L.K. geville, J. Effects of dietary maritime pine seed oil on 5c, 11c, 14c-Eicosatrienoic acid and 5c, 11c, 14c, 17c- lipoprotein metabolism and atherosclerosis develop- eicosatetraenoic acid of Biota orientalis seed oil af- ment in mice expressing human apolipoprotein B. fect lipid metabolism in the rat. Lipids 27, 500-504 Eur. J. Nutr. 40, 268-274（2001）. （1992）. 5） Lee, J.-W.; Lee, K.-W.; Lee, S.-W.; Kim, I.-H.; Rhee, C. 10） Wolff, R.L.; Dareville, E.; Martin, J.-C. Positional distri- Selective increase in pinolenic acid（all-cis-5, 9, 12– bution of Δ5-olefinic acids in triacylglycerols from co- 18:3）in Korean pine nut oil by crystallization and its nifer seed oils: General and specific enrichment in the effect on LDL-receptor activity. Lipids 39, 383-387 sn-3 position. J. Am. Oil Chem. Soc. 74, 515-523 （2004）. （1997）. 6） Ferramosca, A.; Savy, V.; Einerhand, A.; Zara, V. Pinus 11） Lee, B.-M.; Choi, J.-H.; Hong, S.I.; Yoon, S.W.; Kim, B.H.; koraiensis seed oil（PinnoThinTM）supplementation re- Kim, C.-T.; Kim, C.-J.; Kim, Y.; Kim, I.-H. Enrichment duces body weight gain and lipid concentration in liver of pinolenic acid from pine nut oil via lipase-catalyzed and plasma of mice. J. Anim. Feed Sci. 17, 621-630 ethanolysis with an immobilized Candida antarctica

1441 J. Oleo Sci. 67, (11) 1435-1442 (2018) H. Kim, N. Choi, H.-R. Kim et al.

lipase. Biocatal. Biotransform. 29, 155-160（2011）. etable oil to biodiesel using immobilized Candida 12） No, D.S.; Zhao, T.T.; Kim, Y.; Yoon, M.-R.; Lee, J.-S.; antarctica lipase. J. Am. Oil Chem. Soc. 76, 789-793 Kim, I.-H. Preparation of highly purified pinolenic acid （1999）. from pine nut oil using a combination of enzymatic es- 19） Hong, S.I.; Ma, N.; No, D.S.; Choi, N.; Baik, J.Y.; Kim, terification and urea complexation. Food Chem. 170, C.-T.; Kim, Y.; Chang, E.; Kim, I.-H. Enrichment of 386-393（2015）. DHA from tuna oil in a packed bed reactor via lipase- 13） Halldorsson, A.; Kristinsson, B.; Glynn, C.; Haraldsson, catalyzed esterification. J. Am. Oil Chem. Soc. 91, G.G. Separation of EPA and DHA in fish oil by lipase- 1877-1884（2014）. catalyzed esterification with glycerol. J. Am. Oil 20） Rahmatullah, M.S.; Shukla, V.; Mukherjee, K. Chem. Soc. 80, 915-921（2003）. γ-Linolenic acid concentrates from borage and evening 14） Baik, J.Y.; No, D.S.; Oh, S.W.; Kim, I.H. Enrichment of primrose oil fatty acidsvia lipase-catalyzed esterifica- stearidonic acid from echium oil via a two-step lipase- tion. J. Am. Oil Chem. Soc. 71, 563-567（1994）. catalyzed esterification. Eur. J. Lipid Sci. Technol. 21） Zaks, A.; Klibanov, A.M. The effect of water on enzyme 116, 618-626（2014）. action in organic media. J. Biol. Chem. 263, 8017- 15） Lee, E.-J.; Lee, M.W.; No, D.S.; Kim, H.; Oh, S.-W.; Kim, 8021（1988）. Y.; Kim, I.-H. Preparation of high purity docosahexae- 22） Yang, T.; Xu, X.; He, C.; Li, L. Lipase-catalyzed modifi- noic acid from microalgae oil in a packed bed reactor cation of lard to produce human milk fat substitutes. via two-step lipase-catalysed esterification. J. Funct. Food Chem. 80, 473-481（2003）. Foods 21, 330-337（2016）. 23） He, Y.; Shahidi, F. Enzymatic esterification of ω -3 fat- 16） Haraldsson, G.G.; Kristinsson, B. Separation of eicosa- ty acid concentrates from seal blubber oil with glycer- pentaenoic acid and docosahexaenoic acid in fish oil ol. J. Am. Oil Chem. Soc. 74, 1133-1136（1997）. by kinetic resolution using lipase. J. Am. Oil Chem. 24） Lee, C.H.; Parkin, K.L. Effect of water activity and im- Soc. 75, 1551-1556（1998）. mobilization on fatty acid selectivity for esterification 17） No, D.S.; Zhao, T.T.; Kim, B.H.; Choi, H.-D.; Kim, I.-H. reactions mediated by lipases. Biotechnol. Bioeng. Enrichment of erucic acid from crambe oil in a recir- 75, 219-227（2001）. culated packed bed reactor via lipase-catalyzed etha- 25） Chulalaksananukul, W.; Condoret, J.; Delorme, P.; Wil- nolysis. J. Mol. Catal. B: Enzym. 87, 6-10（2013）. lemot, R. Kinetic study of esterification by immobilized 18） Shimada, Y.; Watanabe, Y.; Samukawa, T.; Sugihara, A.; lipase in n-hexane. FEBS Lett. 276, 181-184（1990）. Noda, H.; Fukuda, H.; Tominaga, Y. Conversion of veg-

1442 J. Oleo Sci. 67, (11) 1435-1442 (2018)