Lipase‐catalysed enrichment of DHA and EPA in acylglycerols resulting from squid oil ethanolysis

The fatty acid specificity of four lipases towards eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) was evaluated when performing ethanolysis of squid oil. During the first part of ethanolysis, no DHA ethyl esters were detected when using the lipases from Thermomyces lanuginosus, Pseudomonas cepacia or Pseudomonas fluorescens (in the case of the second and third lipases, no EPA ethyl esters were detected either). This indicates that these three lipases could not catalyse the conversion of DHA located in a triacylglycerol to ethyl ester, and that the Pseudomonas lipases could not catalyse the conversion of EPA either. This pattern was not found for the lipase from Rhizomucor miehei. The lipase from Thermomyces lanuginosus showed the lowest specificity towards DHA and the highest DHA recovery during DHA enrichment in the acylglycerol fraction. It was thus used to catalyse the ethanolysis of squid oil on a larger scale. The ethyl esters formed were removed using short‐path distillation, resulting in a product containing mainly mono‐ and diacylglycerols. The product contained 34 mol‐% DHA and 17 mol‐% EPA, compared with 19 mol‐% DHA and 12 mol‐% EPA in the original squid oil.     

TMAH-Catalyzed Transesterification of EPA and DHA in Encapsulated Fish Oil Products

Tetramethylammonium hydroxide (TMAH)-catalyzed transesterification was developed as a rapid and reliable method using gas chromatography (GC) to determine the total fatty acid profile and to quantify the ethyl esters of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in 20 brands of encapsulated fish oil products. The AOAC method with boron trifluoride (BF₃) as a catalyst was used as a reference. After the respective transesterifications of BF₃ and TMAH, seven brands of encapsulated fish oil showed a single peak of EPA or DHA in the chromatograms, while 13 brands showed a single peak in the chromatograms after BF₃ esterification, but doublet peaks of EPA or DHA after TMAH esterification. By comparing with the GC/MS NIST library and authentic standard fatty acids of ethyl esters, the two pairs of doublet peaks were confirmed the ethyl and methyl esters of EPA and DHA, while the sum of the peak areas of the doublet represented the content of EPA or DHA. The reaction time course concluded that optimal TMAH transesterification was obtained at 25 °C for 10 min and using GC columns of low to medium polarity including Rtx-wax and Rtx-2330 were able to differentiate and quantify the ethyl- and methyl-esterified EPA and DHA, while RT-2560 column with higher polarity than the two other columns was unable to resolve the ethyl ester from the methyl ester of EPA or DHA. An EPA/DHA ratio of ≥1.10 may serve as an indicator of fish oil fortified with the ethyl ester of EPA.     

Fractionation of urea-pretreated squid visceral oil ethyl esters

Ethyl esters of squid (Illex argentinus) visceral oil contained 11.8% eicosapentaenoic acid (EPA) and 14.9% docosahexaenoic acid (DHA). The esters were treated with urea to increase the contents of EPA and DHA. The non-urea complexing ethyl esters of squid visceral oil contained 28.2% EPA and 35.6% DHA. This mixture was fractionated by molecular distillation to further increase the EPA or DHA content. The fraction collected in the 110°C distillate had an EPA content of 39.0% with 0.26 g/100 g of cholesterol, while the 130°C distillate contained 65.6% DHA and 0.42 g/100 g of cholesterol. Ethyl esters prepared from visceral oil of squid Ommastrephes bartrami had 4.5% EPA and 12.7% DHA. After urea pretreatment, the EPA and DHA contents were raised to 10.1 and 30.0%, respectively. When this mixture was further fractionated by molecular distillation, 16.9% EPA with 0.35 g/100 g cholesterol was found in the 110°C distillate and 52.6% DHA with 0.70 g/100 g cholesterol was found in the 130°C distillate. Cholesterol in the squid visceral oil ethyl esters was concentrated in the final residue of molecular distillation when the polyunsaturated ethyl esters were enriched by the urea complexation method prior to molecular distillation. For example, the cholesterol content in the ethyl esters from O. bartrami squid visceral oil was 2.28 g/100 g originally. It was enriched to 64.15 g/100 g in the final residue from the molecular distillation.     

EPA,but not DHA,decreases mean platelet volume in normal subjects

The first indication of platelet activation is an increase in mean platelet volume (MPV). n−3 FA are known to inhibit platelet function and to reduce the risk for coronary heart disease. The purpose of this study was to determine the effects of FPA and DHA on MPV. Healthy subjects received olive oil placebo for 4 wk and then were randomly assigned to receive 4g of ethyl esters of either safflower oil (n=11), EPA (n=10), or DHA (n=12) for 4 wk. At the end of placebo run-in and treatment periods, MPV (fL; mean±SEM) and platelet count (PLT-CT; 10³/μL blood) were measured in the basal state and after ex vivo stimulation with collagen (10 μg/mL), cold (4°C), and heat (37°C). Unlike DHA, EPA lowered MPV as compared with safflower oil (7.2±0.1 vs. 7.5±0.1 fL; P<0.05) and raised PLT-CT (211±18 vs. 192±18 10³/μL; P<0.05) in the fasting state. Collagen and cold significantly increased MPV whereas heat lowered MPV regardless of treatment. All stimuli decreased PLT-CT. EPA significantly increased platelet EPA (0.2±0.1 vs. 3.3±0.4%) and docosapentaenoic acid (DPA; 2.2±0.3 vs. 2.9±0.3%) concentrations, but not DHA. DHA treatment significantly increased DHA (1.4±0.2 vs. 4.1±0.5%) and DPA (2.0±0.4 vs. 3.0±0.4%) concentrations, but not EPA. In conclusion, EPA, but not DHA, reduces platelet activation, an early step in platelet aggregation.     

Fractionation of menhaden oil ethyl esters using supercritical fluid CO2

Supercritical fluid CO₂ was used to fractionate menhaden oil fatty acid ethyl esters to obtain concentrates of the esters of allcis-5,8,11,14,17-eicosapentaenoic acid (EPA) and allcis-4,7,10,13,16,19-docosahexaenoic acid (DHA). Separation of the ethyl esters was found to occur primarily by carbon number, thus limiting the degree to which the ethyl esters of EPA and DHA could be concentrated. Urea fractionation of whole esters in order to remove saturates, monoenes and dienes prior to fractionation with supercritical fluid CO₂ resulted in concentrates of EPA and DHA in purities exceeding 90%. Several criteria are given for the selection of crude oils in order to maximize both purity and yield of concentrates.     

Synthesis of triacylglycerol from polyunsaturated fatty acid by immobilized lipase

More than 95% of polyunsaturated acid (PUFA) was converted to triacylglycerol by immobilized lipase fromCandida antarctica orRhizomucor miehei. The esterification was carried out at 50–60°C with shaking and dehydration for 24 h. The substrates consisted of glycerol and free fatty acid or ethyl esters of the fatty acid at a 1∶3 molar ratio. The docosahexaenoic acid (DHA) or eicosapentaenoic acid (EPA) in the substrate polymerized during the reaction, and they required 5–10% more than the stoichiometric amount to compensate for the PUFA loss. On the contrary, ethyl esters of DHA and EPA were not polymerized. Pure tridocosahexaenoyl, trieicosapentaenoly and triarachidonoyl glycerol were isolated after passing the product through a basic aluminum oxide column. Industrial feasibility of this process was discussed for the ethyl ester as substrate. Portions of this article were presented at the annual meeting of the Japan Society for Bioscience, Biotechnology, and Agrochemistry and at the annual meeting of The Japan Oil Chemists' Society held in Kyoto, Japan, March 31, 1991, and in Hamamatsu, Japan, October 4, 1991, respectively.     

Purification of polyunsaturated fatty acid esters from tuna oil with supercritical fluid chromatography

The technical and economic feasibility of producing docosahexaenoic acid (DHA)- and eicosapentaenoic acid (EPA)-ethyl ester concentrates from transesterified tuna oil using supercritical fluid chromatography (SFC) was studied. A systematic experimental procedure was used to find the optimal values for process parameters and the maximal production rate. DHA ester concentrates up to 95 wt% purity were obtained in one chromatographic step with SFC, using CO₂ as the mobile phase at 65°C and 145 bar and octadecyl silane-type reversed-phase silica as the stationary phase. DHA ester, 0.85 g/(kg stationary phase · h) and 0.23 g EPA ester/(kg stationary phase · h) can be simutaneously produced at the respective purities of 90 and 50 wt%. The process for producing 1,000 kg DHA concentrate and 410 kg EPA concentrate per year requires 160 kg stationary phase and 2.6 tons/h carbon dioxide eluant recycle. The SFC operating cost is U.S. $550/kg DHA and EPA ethyl ester concentrate.     

Separation of eicosapentaenoic acid and docosahexaenoic acid in fish oil by kinetic resolution using lipase

The objective of this study was to investigate the use of lipases as catalysts for separating eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in fish oil by kinetic resolution. Transesterification of various fish oil triglycerides with a stoichiometric amount of ethanol by immobilized Rhizomucor miehei lipase under anhydrous solvent-free conditions resulted in a good separation. When free fatty acids from the various fish oils were directly esterified with ethanol under similar conditions, greatly improved results were obtained. By this modification, complications related to regioselectivity of the lipase and nonhomogeneous distribution of EPA and DHA into the various positions of the triglycerides were avoided. As an example, when tuna oil comprising 6% EPA and 23% DHA was transesterified with ethanol, 65% conversion into ethyl esters was obtained after 24 h. The residual glyceride mixture contained 49% DHA and 6% EPA (8:1), with 90% DHA recovery into the glyceride mixture and 60% EPA recovery into the ethyl ester product. When the corresponding tuna oil free fatty acids were directly esterified with ethanol, 68% conversion was obtained after only 8h. The residual free fatty acids comprised 74% DHA and only 3% EPA (25:1). The recovery of both DHA into the residual free fatty acid fraction and EPA into the ethyl ester product remained very high, 83 and 87%, respectively.     

Strategies for the enzymatic enrichment of PUFA from fish oil

PUFA from oil extracted from Nile perch viscera were enriched by selective enzymatic esterification of the free fatty acids (FFA) or by hydrolysis of ethyl esters of the fatty acids from the oil (FA‐EE). Quantitative analysis was performed using RP‐HPLC coupled to an evaporative light scattering detector (RP‐HPLC‐ELSD). The lipase from Thermomyces lanuginosus discriminated against docosahexaenoic acid (DHA) most, resulting in the highest DHA/DHA‐EE enrichment while lipase from Pseudomonas cepacia discriminated against eicosapentaenoic acid (EPA) most, resulting in the highest EPA/EPA‐EE enrichment. The lipases discriminated between DHA and EPA with a higher selectivity when present as ethyl esters (EE) than when in FFA form. Thus when DHA/EPA were enriched to the same level during esterification and hydrolysis reactions, the DHA‐EE/EPA‐EE recoveries were higher than those of DHA/EPA‐FFA. In reactions catalysed by lipase from T. lanuginosus, at 26 mol% DHA/DHA‐EE, DHA recovery was 76% while that of DHA‐EE was 84%. In reactions catalysed by lipase from P. cepacia, at 11 mol% EPA/EPA‐EE, EPA recovery was 79% while that of EPA‐EE was 92%. Both esterification of FFA and hydrolysis of FA‐EE were more effective for enriching PUFA compared to hydrolysis of the natural oil and are thus attractive process alternatives for the production of products highly enriched in DHA and/or EPA. When there is only one fatty acid residue in each substrate molecule, the full fatty acid selectivity of the lipase can be expressed, which is not the case with triglycerides as substrates.     

Jojoba oil wax esters and derived fatty acids and alcohols: Gas chromatographic analyses

HCl-catalyzed ethanolysis followed by saponification readily surmounts the resistance of long chain wax esters to direct hydrolysis by alkali. Additionally, choosing ethyl instead of methyl esters allows baseline separations between long-chain alcohols and corresponding esters in gas liquid chromatographic (GLC) analysis of total alcohol and acid components before saponification. Liquid wax esters were analyzed on a temperature-programmed 3% OV-1 silicone column. Geographical and genetic effects on the variability of jojoba oil composition were investigated with five different seed samples. Major constituents in jojoba seed oil from shrubs in the Arizona deserts, as indicated by GLC analyses of oil, ethanolysis product, isolated fatty alcohols and methyl esters of isolated fatty acids, were C₄₀ wax ester 30%, C₄₂ wax ester 50% and C₄₄ wax ester 10%; octadecenoic acid 6%; eicosenoic acid 35%, docosenoic acid 7%, eicosenol 22%, docosenol 21% and tetracosenol 4%. Oil from smaller leaved prostrate plants growing along California’s oceanside showed a slight tendency toward higher molecular size than oils from the California desert and Arizona specimens. The wax esters are made up of a dispro-portionately large amount of docosenyl eicosenoate and are not a random combination of constituent acids and alcohols.Lunaria annua synthetic wax ester oil was used as a model for evaluating the analytical procedures. Presented at the AOCS Meeting, Chicago, September 1970 No. Utiliz, Res. Dev. Div., ARS, USDA.     

Lipase-catalyzed incorporation of n−3 polyunsaturated fatty acids into vegetable oils

The ability of immobilized lipases IM60 fromMucor miehei and SP435 fromCandida antarctica to modify the fatty acid composition of selected vegetable oils by incorporation of n−3 polyunsaturated fatty acids into the vegetable oils was studied. The transesterification was carried out in organic solvent with free acid and ethyl esters of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) as acyl donors. With free EPA as acyl donor, IM60 gave higher incorporation of EPA than SP435. However, when ethyl esters of EPA and DHA were the acyl donors, SP435 gave higher incorporation of EPA and DHA than IM60. When IM60 and free acid were used, the addition of 5 μL water increased EPA incorporation into soybean oil by 4.9%. With ethyl ester of EPA as acyl donor, addition of 2 μL water increased EPA incorporation by 3.9%. For SP435, addition of water up to 2μL resulted in increased EPA incorporation, but the incorporation declined when the added water exceeded this amount. The addition of water increased the EPA incorporation into Trisun 90 after 24 h reaction but not the reaction rate at early stages of the reaction.     

Lipase selectivity toward fatty acids commonly found in fish oil

The main objective of this study was to compare the fatty acid selectivity of numerous commercially available lipases toward the most ubiquitous fatty acids present in fish oils in form of their corresponding ethyl esters. Special interest was taken in their ability to separate the n‐3 long‐chain polyunsaturated fatty acids (PUFA), mainly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), from the more saturated fatty acids as well as exploiting the putative discrimination between these highly valuable n‐3 PUFA. Hydrolysis of sardine oil ethyl esters in a Tris buffer solution by 12 microbial lipases is described. The results reveal that all of the lipases strongly discriminate against the n‐3 PUFA and prefer the more saturated fatty acids as substrates. Most of the lipases discriminate between EPA and DHA in favor of EPA, however, 2 bacterial lipases from Pseudomonas were observed to prefer DHA to EPA. Digestive lipolytic enzymes isolated from salmon and rainbow trout intestines displayed reversed fatty acid selectivity when their fish oil triacylglycerol hydrolysis was studied. Thus, the n‐3 PUFA including EPA and DHA were observed to be hydrolyzed at a considerably higher rate than the more saturated fatty acids.     

Fractionation of squid visceral oil ethyl esters by short-path distillation

Squid visceral oil contains high levels of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Its ethyl esters were fractionated by short-path distillation in this study. The elimination temperatures of squid visceral oil ethyl esters (SVOEE) ranged from 50 to 140°C, increasing with the carbon number of ethyl esters. The elimination temperature of cholesterol was higher than those of SVOEE. The SVOEE of Illex argentinus (SVOEE-A) was more advantageous as the raw material (feed) than that of Ommastrephes bartrami (SVOEE-B) for the isolation of EPA and DHA, because SVOEE-A contained less 20∶1 and 22∶1. When SVOEE-A originally containing 9.0% EPA, 14.7% DHA, and 1,121 mg/100 g of cholesterol was distilled from 50 to 150°C with 20°C interval, the 130°C distillate could give 15.5% EPA and 34.7% DHA with 99 mg/100 g of cholesterol, and the yield was 21.8%. The 150°C distillate could give 43.1% DHA with 496 mg/100 g of cholesterol. Furthermore, the distillates collected from 110 to 150°C contained 24.4 to 50.2% of EPA plus DHA, and their total yield was 58.3%. The final residue after 150°C distillation contained 77% of the total cholesterol in the initial SVOEE-A, and the yield was 6.0%.     

n‐3 PUFA Induce Microvascular Protective Changes During Ischemia/Reperfusion

Ischemia/reperfusion (I/R) injury can occur in consequence of myocardial infarction, stroke and multiple organ failure, the most prevalent cause of death in critically ill patients. I/R injury encompass impairment of endothelial dependent relaxation, increase in macromolecular permeability and leukocyte‐endothelium interactions. Polyunsaturated fatty acids (n‐3 PUFA), such as eicosapentaenoic acid (EPA, 20:5n‐3) and docosahexaenoic acid (DHA, 22:6n‐3) found in fish oil have several anti‐inflammatory properties and their potential benefits against I/R injury were investigated using the hamster cheek pouch preparation before and after ischemia. Before the experiments, hamsters were treated orally with saline, olive oil, fish oil and triacylglycerol (TAG) and ethyl ester (EE) forms of EPA and DHA at different daily doses for 14 days. Fish oil restored the arteriolar diameter to pre ischemic values during reperfusion. At onset and during reperfusion, Fish oil and DHA TAG significantly reduced the number of rolling leukocytes compared to saline and olive oil treatments. Fish oil, EPA TAG and DHA TAG significantly prevented the rise on leukocyte adhesion compared to saline. Fish oil (44.83 ± 3.02 leaks/cm²), EPA TAG (31.67 ± 2.65 leaks/cm²), DHA TAG (41.14 ± 3.63 leaks/cm²), and EPA EE (30.63 ± 2.25 leaks/cm²), but not DHA EE (73.17 ± 2.82 leaks/cm²) prevented the increase in macromolecular permeability compared to saline and olive oil (134.80 ± 1.49 and 121.00 ± 4.93 leaks/cm², respectively). On the basis of our findings, we may conclude that consumption of n‐3 polyunsaturated fatty acids, especially in the triacylglycerol form, could be a promising therapy to prevent microvascular damage induced by ischemia/reperfusion and its consequent clinical sequelae.     

Production of Structured Phosphatidylcholine with High Content of DHA/EPA by Immobilized Phospholipase A1-Catalyzed Transesterification

This paper presents the synthesis of structured phosphatidylcholine (PC) enriched with docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) by transesterification of DHA/EPA-rich ethyl esters with PC using immobilized phospholipsase A₁ (PLA₁) in solvent-free medium. Firstly, liquid PLA₁ was immobilized on resin D380, and it was found that a pH of 5 and a support/PLA₁ ratio (w/v) of 1:3 were the best conditions for the adsorption. Secondly, the immobilized PLA₁ was used to catalyze transesterification of PC and DHA/EPA-rich ethyl esters. The maximal incorporation of DHA and EPA achieved was 30.7% for 24 h of reaction at 55 °C using a substrate mass ratio (PC/ethyl esters) of 1:6, an immobilized PLA₁ loading of 15% and water dosage of 1.25%. Then the reaction mixture was analyzed by ³¹P nuclear magnetic resonance (NMR). The composition of reaction product included 16.5% PC, 26.3% 2-diacyl-sn-glycero-3-lysophosphatidylcholine (1-LPC), 31.4% 1-diacyl-sn-glycero-3-lysophosphatidylcholine (2-LPC), and 25.8% sn-glycerol-3-phosphatidylcholine (GPC).     

Validation of a method for gas chromatographic analysis of eicosapentaenoic acid and docosahexaenoic acid as active ingredients in medicinal products

A gas chromatographic method for the determination of all-cis-5,8,11,14,17-eicosapentaenoic acid (EPA) and all-cis-4,7,10,13,16,19-docosahexaenoic acid (DHA) as active ingredients in medicinal products was developed and validated. In accordance with the rules governing medicinal products in the European Community, the method establishes relations between label claims of active ingredients and known reference standards. A routine for examining instrument status is proposed. The relative standard deviation was 1% (n=26) for determination of the empirical response factors of EPA ethyl ester and DHA ethyl ester relative to the internal standard, C23:0 methyl ester. This experiment included two columns and EPA and DHA standards from two different suppliers and was carried out over a five-month period. Repeatability (n=6) for low and medium concentrates of glycerides and high concentrates of ethyl esters, expressed as coefficient of variation, was 4, 0.7 and 0.7%, respectively. Accuracy (n=6) determined as percent recovery was better than 98% for all sample types. Analytical results from a twelve-month stability study of the high concentrate are shown. Presented at the AOCS Annual Meeting in Chicago, IL, May 1991.     

Purification and biochemical characterization of an intracellular lipase by Rhizopus chinensis under solid‐state fermentation and its potential application in the production of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)

BACKGROUND: Purification and characterization of an intracellular lipase produced by Rhizopus chinenesis cultured in solid‐state fermentation was investigated. The potential application in concentrating eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) from fish oil by the pure enzyme was also studied. RESULTS: Through four successive purification steps, the enzyme was purified to homogeneity with an apparent molecular mass of 36 kDa. The lipase was active for pH between 7.0 and 9.0 and temperatures 20–45 °C. Lipase activity was slightly increased in the presence of Ca²⁺ and Mg²⁺, but strongly inhibited by Hg²⁺ and SDS. The pure enzyme was most active on medium chain p‐nitrophenol esters, with the highest activity towards pNP‐caprylate (C8). The enzyme is a non‐specific lipase, because it cleaved not only the 1,3‐positioned ester bonds but also the 2‐positioned bond in triolein. High EPA (17.6%) and DHA (32.9%) contents were achieved using the pure lipase (100 U) within 10 h. CONCLUSION: The enzymatic activity of the lipase on a wide variety of substrates and its stability in the presence of some organic solvents suggest that the lipase should be investigated for a range of commercial applications. The pure lipase was proved to possess potential ability for the production and concentration of EPA and DHA from fish oil. Copyright © 2008 Society of Chemical Industry     

Effects of highly purified eicosapentaenoic acid and docosahexaenoic acid on fatty acid absorption, incorporation into serum phospholipids and postprandial triglyceridemia

Fourteen healthy volunteers were randomly allocated to receive 4 g highly purified ethyl esters of eicosapentaenoic acid (EPA) (95% pure, n=7) or docosahexaenoic acid (DHA) (90% pure, n=7) daily for 5 wk in supplement to their ordinary diet. The n−3 fatty acids were given with a standard high-fat meal at the beginning and the end of the supplementation period. EPA and DHA induced a similar incorporation into chylomicrons which peaked 6 h after the meal. The relative uptake of EPA and DHA from the meal was >90% compared with the uptake of oleic acid. During absorption, there was no significant elongation or retroconversion of EPA or DHA in total chylomicron fatty acids. The concentration of EPA decreased by 13% and DHA by 62% (P<0.001) between 6 and 8 h after the meal. During the 5-wk supplementation period, EPA showed a more rapid and comprehensive increase in serum phospholipids than did DHA. DHA was retroconverted to EPA, whereas EPA was elongated to docosapentaenoic acid (DPA). The postprandial triglyceridemia was suppressed by 19 and 49% after prolonged intake of EPA and DHA, respectively, indicating that prolonged intake of DHA is equivalent to or even more efficient than that of EPA in lowering postprandial triglyceridemia. This study indicates that there are metabolic differences between EPA and DHA which may have implications for the use of n−3 fatty acids in preventive and clinical medicine.     

超临界CO2萃取精馏制取高纯度EPA乙酯和DHA乙酯

Fish oil ethyl esters complexed with aqueous silver nitrate solution were extracted and rectified by supercritical CO₂ to obtain DHA ester and EPA ester with high purity. The effects of some independent variables,such as extraction pressure, temperature gradient of rectifying column and programmed pressure,on rectification were investigated.The results showed that programmed pressure is suitable for purification of EPA and DHA esters. Increase of column temperature gradient from bottom to top is one of the key elements in rectification. Furthermore, higher temperature gradient leads to better separation effect.     

Combined Urea Complexation and Argentated Silica Gel Column Chromatography for Concentration and Separation of PUFAs from Tuna Oil: Based on Improved DPA Level

Eicosapentaenoic acid (EPA, 20:5n‐3), docosapentaenoic acid (DPA) isomers (22:5n‐6 and 22:5n‐3) and docosahexaenoic acid (DHA, 22:6n‐3) derived from tuna oil were concentrated by three stages of urea fractionation at various crystallization temperatures and different fatty acid/urea ratios. Thereafter, polyunsaturated fatty acids concentrate containing comparatively enriched DPA levels was purified by argentated silica gel column chromatography. A product containing 22.2 ± 0.6 % EPA, 4.6 ± 0.0 % DPAn‐6, 5.9 ± 0.1 % DPAn‐3 and 42.3 ± 1.2 % DHA was obtained at 1:1.6 fatty acid/urea ratio (w/w) by crystallization at ?8 °C for 16 h, ?20 °C for 8 h, and ?8 °C for 16 h. A DPA isomer concentrate containing 26.1 ± 0.5 % DPAn‐6 and 22.3 ± 0.4 % DPAn‐3 was achieved by argentated silica gel chromatography in the 6 % acetone/n‐hexane solvent fraction (v/v), and the recovery of both fatty acids was 66.1 ± 3.2 and 70.7 ± 2.2 %, respectively. Furthermore, 91.9 ± 2.5 % EPA and 99.5 ± 2.1 % DHA with recoveries of 47.8 ± 2.0 and 56.7 ± 3.3 %, respectively, were obtained in various fractions.