Inhibitory Effect of Dihydrosphingosine with ... - JST

713

Journal of Oleo ScienceCopyright ©2016 by Japan Oil Chemists’ Societydoi : 10.5650/jos.ess16071J. Oleo Sci. 65, (9) 713-722 (2016)

Inhibitory Effect of Dihydrosphingosine with α-Tocopherol on Volatile Formation during the Autoxidation of Polyunsaturated TriacylglycerolsMariko Uemura1, Ako Shibata1, Masashi Hosokawa1, Ai Iwashima-Suzuki2, Makoto Shiota2 and Kazuo Miyashita1,＊

1 Faculty of Fisheries Sciences, Hokkaido University (3-1-1 Minato, Hakodate, Hokkaido 041-8611, JAPAN)2 Milk Science Research Institute, Megmilk Snow Brand Co. Ltd. (1-1-2 Minamidai, Kawagoe, Saitama 350-1165, JAPAN)

1 IntroductionEicosapentaenoic acid（20:5n-3, EPA）and docosahexae-

noic acid（22:6n-3, DHA）are active forms of n-3 polyunsat-urated fatty acids（PUFA）and are found in fish oils1）. Con-sidering the strong link between the consumption of both PUFA and the reduction of cardiovascular disease risk, the American and European Heart Associations recommend a high intake of fish oil-containing foods and supplements for the prevention of sudden cardiac death and other cardio-vascular dysfunctions2, 3）. However, fish oil is highly suscep-tible to oxidation, which causes undesirable flavors and loss of the functionality of n-3 PUFA. Natural antioxidants are often added to protect EPA and DHA from oxidative deterioration4, 5）. Because of their low cost, synthetic anti-oxidants have been widely used; however, consumer pref-erences for “natural” foods and the potential toxicological effects of synthetic antioxidants have prompted the food industry to search for more effective natural antioxidants6, 7）.

The first and rate-limiting step of lipid oxidation is ab-straction of a hydrogen radical by free radical initiators from substrate lipids. The reaction proceeds thereafter through a free radical chain reaction to form lipid hydro-peroxides as the primary oxidation products8）. Therefore, the chain-breaking antioxidants are of considerable practi-cal importance in protecting lipids against oxidative deteri-

＊Correspondence to: Kazuo Miyashita, Faculty of Fisheries Sciences, Hokkaido University, 3-1-1 Minato, Hakodate, Hokkaido 041-8611, JAPANE-mail: [email protected] April 5, 2016 (received for review March 22, 2016)Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 onlinehttp://www.jstage.jst.go.jp/browse/jos/　　http://mc.manusriptcentral.com/jjocs

oration. These antioxidants inhibit or retard the oxidation by interfering with either chain propagation or initiation by readily donating hydrogen atoms to free radicals. Among natural chain-breaking antioxidants, tocopherols are the most well-known and are widely used for foods and cos-metics. However, because of the high unsaturation, it is ex-tremely difficult to completely inhibit the oxidation of EPA- and DHA-rich fish oils by tocopherols alone.

Tocopherols can be more active when used in combina-tion with other antioxidants, such as citric acid and ascor-bic acid. Amine-containing phospholipids and sphingolipids have also been reported to effectively inhibit the oxidation of PUFA in the presence of α-tocopherol, whereas little an-tioxidant activity was found in these polar lipids without α-tocopherol9, 10）. Moreover, we have found stronger anti-oxidant activity of sphingoid base, dihydrosphingosine（d18:0）, with α-tocopherol on fish oil triacylglycerol（TAG）oxidation10）（Fig. 1）. The addition of 1.0 wt％ d18:0 with 0.05 wt％ α-tocopherol effectively inhibited the oxygen ab-sorption and peroxide formation during the fish oil TAG oxidation for more than 500 h, whereas a rapid increase in both oxidation indicators was found with the addition of α-tocopherol alone. This result suggests the possible appli-cation of d18:0 and tocopherol combination to prevent fish oil oxidation; however, EPA- and DHA-hydroperoxides are

Abstract: The effect of dihydrosphingosine (d18:0) on triacylglycerol (TAG) oxidation was examined with and without α-tocopherol. Three types of TAG from fish, linseed, and soybean oil were oxidized at 50°C to determine the effect of dihydrosphingosine (d18:0) with or without α-tocopherol on triacylglycerol (TAG) oxidation. The analysis of oxygen consumption and total volatile formation demonstrated a small effect of d18:0 on TAG oxidation in the absence of α-tocopherol. On the other hand, the combination of d18:0 with α-tocopherol showed strong antioxidant activity and completely inhibited volatile formation within 1400 h for soybean oil TAG, 650 h for linseed oil TAG, and 380 h for fish oil TAG.

Key words: dihydrosphingosine, α-tocopherol, fish oil oxidation, flavor deterioration, volatiles, acrolein, oxidative stability

M. Uemura, A. Shibata and M. Hosokawa et al.

J. Oleo Sci. 65, (9) 713-722 (2016)

714

easily decomposed into volatile secondary oxidation prod-ucts; then, it is possible for fish oils to have an undesirable flavors and tastes, even with low peroxide levels, as found in the former study10）.

Thus, in the present study, we examined whether the combination of d18:0 and α-tocopherol can prevent the formation of volatile compounds from fish oil TAG oxida-tion. In addition, the effect of d18:0 and α-tocopherol was analyzed using other TAG with different fatty acid compo-sitions.

2 EXPERIMENTAL PROCEDURES2.1 Standards and substrate lipids

Dihydrosphingosine（d18:0）was purchased from Avanti Polar Lipids Inc., Alabaster, AL, USA. α-Tocopherol, soybean oil, and triolein（purity＞99％）were the products of Wako Pure Chemical Ind. Ltd., Osaka, Japan. Linseed oil was obtained from Summit Oil Mill Co. Ltd., Chiba, Japan. Two types of fish oil, DHA concentrated oil（DHA-55）and EPA concentrated oil（EPA-28MN）, were gifts from Maruha Nichiro Co., Tokyo, Japan. Both oils were mixed in equal parts and were used as fish oil. Silica gel（BW-60F）for column chromatography was obtained from Fuji Sylysia Chem. Ltd., Kasugai, Japan. Activated carbon and Celite（545 RVS）were obtained from Nacalai Tesque Inc., Kyoto, Japan. All other chemicals and solvents were of analytical grade.

2.2 Lipid substrate puri�cationEach oil（ca. 25 g）was passed through a column（50 cm

×4 cm i.d.）packed with an n-hexane slurry mixture of ac-tivated carbon（100 g）and Celite（100 g）to remove the to-copherols and pigments by eluting with n-hexane（1200 mL）. The obtained oil（ca. 10 g）was refined using a silicic acid column（50 cm×4 cm i.d.）packed with an n-hexane slurry of Silica gel BW-60F（200 g）by eluting with n-hexane（200 mL）and a mixture of n-hexane-diethyl ether（98:2（200 mL）and 90:10（1200 mL）, v/v）. The fraction eluted with the n-hexane-diethyl ether（90:10）was used as TAG.

To confirm the absence of peroxides, tocopherols, and minor lipid classes, such as free fatty acids, monoacylglyc-erols, and diacylglycerols in the purified TAG, high-perfor-mance liquid chromatography（HPLC）and thin-layer chro-matography（TLC）analyses were performed. The hydroperoxide content in TAG was determined using quantitative conversion of non-fluorescent diphenyl-1-py-renylphosphine（DPPP）to fluorescent DPPP oxide by the reaction with hydroperoxides11）. Briefly, 1 mg of TAG was weighed and dissolved in 5 mL chloroform（containing 10 mg/mL butylhydroxytoluene（BHT））/methanol（2:1, v/v）. To a test tube with a screw cap, 100 μL of the sample solu-tion and 50 μL of DPPP solution（1 mg/10 mL chloroform）were added and left for 60 min at 60℃ in a water bath. Then, the solution was cooled in an ice bath, and 3 mL of 2-propanol was added. The reaction mixture was diluted 1:10（v/v）before measurement by reversed phase HPLC（Hitachi Seisakusho, Co., Tokyo, Japan）. The HPLC analy-sis was performed at 40℃ using a reversed-phase column（Develosil-ODS-UG-5, Nomura Chem. Co., Seto, Japan）protected with a guard column（10×4.0 mm i.d.）with the same stationary phase. The mobile phase was 1-butanol-methanol（10:90, v/v）, and the flow rate was 1.0 mL/min. The fluorescence detector（Hitachi L-2485）was set at Ex. 352 nm and Em. 380 nm. The hydroperoxide concentration in the sample solution was calculated from the DPPP oxide detected using a DPPP oxide standard curve. The hydro-peroxides in the TAG were expressed as meq/kg TAG. Each purified TAG had less than 0.5 meq/kg peroxide.

HPLC for tocopherol analysis was also performed with a Hitachi HPLC system equipped with a pump（Hitachi L-2130）and a fluorescence detector（Hitachi L-2485）. The analysis was conducted on a silica column（Si 60, 250×4.6 mm i.d.; Kanto Chem. Co., Tokyo, Japan）protected with a guard column（15×3.2 mm）with the same stationary phase. The mobile phase was n-hexane-2-propanol（99.2:0.8, v/v）with a flow rate of 1.0 mL/min. The fluores-cence detector was set at Ex. 298 nm and Em. 325 nm. To check the lipid class purification of TAG, analytical TLC was performed on a 0.25 mm silica gel plate（Silica gel 60G; Merck, Darmstadt, Germany）developed with n-hexane-di-ethyl ether-acetic acid（70:30:1, v/v/v）. Lipid spots were detected with iodine vapor or 60％ aqueous sulfuric acid charring. Identification of the spot was performed using standard TAG（triolein）. HPLC and TLC analysis showed the complete removal of tocopherols and minor lipid classes from the crude lipids.

The fatty acid composition of the TAG was determined using gas chromatography（GC）after conversion of the fatty acyl groups in the lipid to their methyl esters by the method of Prevot and Mordret12）. Briefly, to an aliquot of total lipid（ca. 10 mg）, 1 mL n-hexane and 0.2 mL of 2 N NaOH in methanol were added, vortexed and incubated at 50℃ for 30 min. After the incubation, 0.2 mL of 2 N HCl in

Fig. 1　�Chemical structure of α-tocopherol and dihydrosphingosine（d18:0）.

Inhibitory effect of dihydrosphingosine with α-tocopherol on volatile formation from oxidized PUFA

J. Oleo Sci. 65, (9) 713-722 (2016)

715

methanol solution was added to the solution and vortexed. The mixture was separated by centrifugation at 1000 g for 5 min. The upper hexane layer containing fatty acid methyl esters was recovered and subjected to GC. The GC analysis was performed on a Shimadzu GC-14B（Shimadzu Corpora-tion, Kyoto, Japan）equipped with a flame-ionization detec-tor and a capillary column（Omegawax-320; 30 m×0.32 mm i.d.; Sigma-Aldrich, St. Louis, MO, USA）. The detector, injector, and column temperatures were 260, 250, and 200℃, respectively. The carrier gas was helium, and it had a flow rate of 50 kPa. The fatty acid content was expressed as a weight percentage of the total fatty acids.

2.3 Oxidation and analysis α-Tocopherol and d18:0 were dissolved into n-hexane

and chloroform-methanol（7:3, v/v）, respectively. A certain volume of tocopherol and/or d18:0 solutions were added to TAG dissolved in n-hexane. After homogeneously mixing the solution, the solvent was removed in vacuum so that the tocopherol and d18:0 concentrations were 0.05 wt％ and 1.0 wt％, respectively. Each 300 mg sample was placed in a 20 mL aluminum sealed vial with a butyl-gum septum（GL Science, Tokyo, Japan）and then incubated at 50℃ in the dark. Before the incubation, the level of oxygen in the headspace gas of the vial was estimated using a GC（Shi-madzu GC-14B）13）. The GC was equipped with a thermal conductivity detector and a stainless steel column（3 m×2.1 mm i.d.）packed with a molecular Sieves 5A（60/80 mesh, GL Science）. The temperatures at the injection port, detector port and column oven were 120℃, 120℃ and 70℃, respectively. The helium flow was 50 kPa. More than three separate vials containing similar samples were pre-pared and incubated. A small portion（20 μL）of the head-space gas was taken from each vial using a microsyringe through the butyl gum septum at selected times during the oxidation. The decrease（％）in oxygen was calculated from the changes in the oxygen to nitrogen ratio compared with the ratio before incubation. Each data value at different oxidation times of the different samples was expressed as the mean±SD（n＝3）.

The oxidation of the sample was also monitored via GC analysis of the volatile compounds. For the static head-space GC analysis, after a definite time of incubation, the sample vial was transferred into the HS-20 headspace auto-sampler（Shimadzu Corporation）of the GC apparatus. The headspace gas in the vial was automatically pressurized at 60℃ for 2 min and then immediately injected through a loop into a GC（Shimadzu GC-2014AFSC）equipped with a HP-1 capillary column（50-m length, 0.32 mm i.d. and 1.05 μm film thickness; Agilent Technologies, CA, USA）and a flame ionization detector. An initial oven temperature of 40℃ for 5 min was used, followed by heating at 3℃/min to 70℃, then 20℃/min to 200 ℃, and finally, the temperature was held at 200℃ for 4 min. Both the injection port and

the flame ionization detector were set at 250℃. Three rep-licate measurements of each stored sample were per-formed, and the data were expressed as the mean±SD（n＝3）.

2.4 GC-MS analysisThe identities of the volatile compounds were obtained

using solid phase micro-extraction（SPME）and gas chroma-tography-mass spectrometry（GC-MS）. The volatiles were collected from the sealed vials containing different TAG samples after incubation using a 50/30 μm DVB/CAR/PDMS SPME fiber（Sigma-Aldrich）. The SPME fiber was exposed to the headspace for 5 min at room temperature under the same conditions. The absorbed volatiles were then de-sorbed in the injection port of a GC-2010 gas chromato-graph equipped with a Model GCMS-QP2010 Ultra mass spectrometer（Shimadzu Corporation）. The GC conditions were the same as described above. The mass spectrometer was operated in the electron impact ionization mode（70 eV）. The identification of the volatile compounds was per-formed by comparison with the mass spectra from the NIST Standard Reference Database and by injection of au-thentic standards.

3 Results and Discussion3.1 Oxidative stability of TAG

The rate-limiting step in the lipid oxidation is abstraction of the hydrogen atom that occurs at the bis-allylic position（CH＝CH-CH2-CH＝CH）of PUFA; therefore, the oxidative stability of polyunsaturated lipids increase with a decreas-ing number of bis-allylic positions8）. The number of bisallyl-ic positions per molecule of soybean, linseed, and fish oil TAG can be calculated from mol concentration of PUFA in each TAG to be 0.613, 1.103, and 2.187, respectively. The mol concentration of the PUFA was computed using the weight ％ of the PUFA（Table 1）and the molecular weight of each PUFA. When the oxidative stability of substrate TAG was analyzed in the bulk phase by measuring the de-crease in oxygen in the headspace of the sample vial and the total volatile formation, the oxidative stability was found to be the highest for the soybean oil TAG, followed by the linseed and fish oil TAG（Fig. 2A and B）. This order was in agreement with the expectations based on the number of bisallylic positions per molecule of three kinds of TAG. Although the oxygen consumption（Fig. 2A）and the total volatile formation（Fig. 2B）increased with the in-cubation time for all TAG samples, the soybean oil TAG re-quired several incubations to provide a rapid increase in the oxygen consumption and in the total volatiles.

3.2 Effect of α-tocopherol and d18:0 on TAG oxidationBased on the analysis of the oxygen consumption rate


J. Oleo Sci. 65, (9) 713-722 (2016)

716

（Fig. 3）, the addition of α-tocopherol increased the oxida-tive stability of linseed oil and soybean oil TAG; however, only a small effect was found on fish oil TAG（Fig. 3）. More significant antioxidant activity was obtained by the combi-nation of α-tocopherol with d18:0, whereas d18:0 alone had no effect（Fig. 3）. Amine containing polar lipids, such as glycerophospholipids（PL）and sphingolipids（SL）, have been reported to effectively inhibit the oxidation of PUFAs in the presence of α-tocopherol, whereas little antioxidant ac-tivity is found in these polar lipids without α-tocopherol9, 10）. Our previous study reported the strong antioxidant activity of sphingoid bases in the presence of α-tocopgherol10）.

When the antioxidant activity of amine-containing PL, SL, and their sphingosyl backbone, sphingoid bases, was com-pared in the presence of α-tocopherol, the highest antioxi-dant activity was shown by sphingoid bases, followed by sphingomyelin and other amine-containing PL and SL. This result showed that the activity increased with increasing concentration of amine group of PL, SL, or sphingoid bases in the reaction mixture. The previous study also showed the higher activity of the primary amine structure com-pared to other types of amine group. The molecular weight of d18:0, having a primary amine group, is relatively lower than those of PL and SL; therefore, d18:0 could have much stronger ant ioxidant act iv ity in the presence of α-tocopherol.

Although the detailed mechanism responsible for the combination effect of d18:0 is not fully understood, the amine group of d18:0 has been postulated to have an im-portant role, such as a hydrogen or electron donor to re-generate and recycle tocopheroxyl radical intermediate to the parent phenol, tocopherol14）. Another possible role of the amine group would be a source of antioxidants formed by the non-enzymatic browning reaction with fatty acid ox-idation products containing a keto group, mainly alde-hydes15, 16）. Hidalgo and others15, 16） found the browning an-tioxidants, kinds of pyrrole compounds, which are formed by the non-enzymatic reaction with amine group of amino acids and fatty acid oxidation products, mainly low molecu-lar products. Therefore, the formation of antioxidants by the interaction between oxidized lipids and d18:0 might be a reason for the increase in the oxidative stability of TAG added by d18:0. On the other hand, no antioxidant activity was found in the addition of d18:0 to TAG without α-tocopherol in the present study（Fig. 3）and our former study10）. Therefore, the formation of antioxidant com-

Table 1　 Composition（weight ％）of major fatty acids of TAG.

Fatty acid Soybean Linseed Fish14:0 －－ 5.4116:0 11.64 5.89 12.9618:0 4.09 4.08 3.65

16:1n-7 －－ 4.0118:1n-9 24.82 25.32 6.9218:1n-7 1.47 － 2.0020:1n-11 －－ 1.7318:2n-6 50.89 16.32 －18:3n-6 －－－18:3n-3 4.71 45.31 －18:4n-3 －－ 1.5620:4n-6 －－ 2.7020:5n-3 －－ 13.9422:6n-3 －－ 25.44

Fig. 2　 Oxidative stability of soybean oil TAG（open circle）, linseed oil TAG（open square）, and fish oil TAG（open triangle）. The stability was analyzed by measuring the decrease in oxygen（A）and the increase in total volatiles（B）in the headspace gas of the vial. Oxidation was performed at 50℃ in the dark. The data were expressed as the mean±SD of three separate experiments.


J. Oleo Sci. 65, (9) 713-722 (2016)

717

pounds from d18:0 and oxidation products might require the presence of α-tocopherol. Mild oxidation conditions controlled by α-tocopherol may be important for the for-mation of the antioxidant compounds（Fig. 4）. Although the structures of the antioxidant compounds have not been made clear, they could effectively inhibit the fish oil TAG oxidation as following mechanisms: a）regeneration of α-tocopherol to inhibit the lipid oxidation; b）direct inhibi-tion of lipid oxidation; c）a）and b）.

When d18:0 was added to fish oil TAG in the presence of α-tocopherol, a little decrease in the tocopherol level and a little oxidation of TAG were observed at the first stage of the incubation, but thereafter, there were no change in the tocopherol and no TAG oxidation10）. This result suggests the formation of antioxidant compounds during the first stage of the fish oil TAG incubation. On the other hand, α-tocopherol linearly decreased in non-oxidized tricaprylin mixture in the presence of α-tocopherol, though the trica-

prylin substrate mixture contained the same amount of d18:0 and α-tocopherol as that found in fish oil TAG10）. Thus, only d18:0 could not produce antioxidants to inhibit the α-tocopherol oxidation. For the formation of antioxi-dant compounds, oxidation products as seen in fish oil TAG oxidation would be needed.

3.3 Effect of α-tocopherol and d18:0 on volatile formationThe typical GC analysis of the volatile compounds from

oxidized lipids is the dynamic headspace method with a SPME fiber17－20）. The dynamic headspace method with SPME extraction permits enhancement of trace com-pounds in complex mixtures of a wide range of volatile compounds using lower temperatures. However, there is a limitation in the analysis of several characteristic com-pounds when using the dynamic headspace method21）. Lower-boiling compounds may be lost during the purging cycle in the SPME method, whereas relatively higher mo-lecular weight compounds, such as heptadienal and deca-dienal, can be concentrated in the trap. On the other hand, we suggested that a static headspace GC analysis using a lower operating temperature of 60℃ could measure the actual level of volatile compounds, including low molecular weight or low boiling volatile compounds, such as acrolein（2-propenal）, propanal and pentane22）.

Figures 5, 6, and 7 show the representative chromato-grams of volatiles from oxidized soybean, linseed, and fish oil TAG, respectively. Each chromatogram was obtained by static GC analysis of the oxidized sample at relatively lower oxidation level. The oxidized soybean oil produced only small amounts of volatiles at the early stage of the oxida-tion, but after 507.5 h of incubation, pentane was detected as a major volatile, followed by hexanal, propanal, and acrolein（Fig. 5A）. In the oxidation of linseed oil TAG after 183 h incubation, acrolein, propanal, and pentane were de-tected as major volatiles（Fig. 6A）. Several major peaks were found at earlier retention times than that of acrolein;

Fig. 3　 Effect of d18:0 and α-tocopherol（Toc）on the oxidative stability of soybean（A）, linseed（B）, and fish（C）oil TAG during the incubation at 50℃ in the dark. The stability was analyzed by measuring the decrease in oxygen in the headspace gas of the vial. Control: open circle, ＋α-tocopherol: open square, ＋d18:0: solid diamond, ＋α-tocopherol and d18:0: solid triangle. The data were expressed as the mean±SD of three separate experiments.

Fig. 4　 Poss ib le mechanism for the format ion of antioxidants from d18:0 and aldehydes formed during the first stage of TAG oxidation.


J. Oleo Sci. 65, (9) 713-722 (2016)

718

Fig. 5　 Representative GC of volatile compounds from oxidized soybean oil TAG after 507.5 h incubation at 50℃. Volatile compounds from the oxidized TAG were analyzed by static headspace GC and were identified by GC-MS. Control: （A）, ＋α-tocopherol（Toc）: （B）, ＋d18:0: （C）, ＋α-tocopherol and d18:0（Toc＋d18:0）: （D）.

Fig. 6　 Representative GC of volatile compounds from oxidized linseed oil TAG after 183 h incubation at 50℃. Volatile compounds from the oxidized TAG were analyzed by static headspace GC and were identified by GC-MS. Control: （A）, ＋α-tocopherol（Toc）: （B）, ＋d18:0: （C）, ＋α-tocopherol and d18:0（Toc+d18:0）: （D）.


J. Oleo Sci. 65, (9) 713-722 (2016)

719

however, they could not be identified. The other main vola-tile detected in the linseed oil oxidation was 1-pentene-3-ol. In the fish oil TAG oxidation, acrolein was the most abundant volatile after 26.5 h incubation, followed by propanal and 1-pentane-3-ol（Fig. 7A）.

The combination of d18:0 with α-tocopherol almost completely inhibited the volatile formation（Fig. 5D, 6D, and 7D）, whereas α-tocopherol alone was not enough to prevent volatile formation（Fig. 5B, 6B, and 7B）. The addi-tion of d18:0 without α-tocopherol had no effect on the volatile formation（Fig. 5C, 6C, and 7C）. The most proba-ble mechanism for the antioxidant activity of d18:0 with α-tocopherol is the formation of antioxidants by the amino-carbonyl reaction between the amine group of d18:0 and the keto group of aldehydes formed in the very early stages of TAG oxidation（Fig. 4）. In the presence of α-tocopherol, d18:0 might effectively react with aldehydes, resulting in few aldehyde peaks detected on the chromatograms（Fig. 5D, 6D, and 7D）. However, abundant aldehydes were found with the addition of d18:0 without α-tocopherol（Fig. 5C, 6C, and 7C）. The presence of α-tocopherol is essential for the reaction of d18:0 and aldehydes. Mild oxidation conditions controlled by α-tocopherol may be important for the formation of antioxidants by the amino-carbonyl reaction（Fig. 4）.

3.4 Inhibition of total volatile formation by d18:0 and α-tocopherol combination

Although the strong antioxidant activity of d18:0 with α-tocopherol has been already reported by analyzing the oxygen consumption and peroxide levels of fish oil TAG in our previous study10）, no attempt has been made on the combined effect of d18:0 with α-tocopherol on the forma-tion of volatile compounds. Figure 8 shows the effective-ness of the combination of d18:0 with α-tocopherol on the formation of total volatiles. The combination could almost completely inhibit volatile formation within 1400 h for soybean oil TAG, 650 h for linseed oil TAG, and 380 h for fish oil TAG. The rapid formation of volatile aldehydes is the most serious challenge that limits the addition of fish oil to general food products. Various natural antioxidants, including tocopherols, ascorbic acid, and rosemary ex-tracts, have been used to prevent volatile formation in fish oil oxidation; however, a satisfactory effect has not yet been achieved5）. Thus, it is important to evaluate the inhib-itory effect of antioxidants on volatile compounds forma-tion in practical use. From this viewpoint, the combination of d18:0 with α-tocopherol could successfully inhibit vola-tile formation, even in fish oil oxidation.

Widely used methods for assessing lipid oxidation include oxygen consumption, peroxide value, anisidine value, 2-thiobarbituric acid value, and conjugated dienes. Although the data obtained by these methods indicate the

Fig. 7　 Representative GC of volatile compounds from oxidized fish oil TAG after 26.5 h incubation at 50℃. Volatile compounds from the oxidized TAG were analyzed by static headspace GC and were identified by GC-MS. Control:（A）, ＋α-tocopherol（Toc）: （B）, ＋d18:0:（C）, ＋α-tocopherol and d18:0（Toc＋d18:0）: （D）.


J. Oleo Sci. 65, (9) 713-722 (2016)

720

state of lipid oxidation, none of these methods correlate well with the sensory data of unsaturated lipids, especially fish oil23, 24）. On the other hand, the data on the decom-posed volatile compounds level have been demonstrated to correlate well with sensory data24）. The development of oxi-dative fishy and metallic off-flavors dissuades people from consuming fish oils. The effective inhibition of volatile compounds by the addition of α-tocopherol and d18:0 to fish oil found in Fig. 8 may be useful to develop a new ef-fective prevention system for fish oil oxidation, especially flavor deterioration.

3.5 Effect of α-tocopherol and d18:0 on acrolein forma-tion in �sh oil TAG oxidation

Many types of volatiles have been reported from fish oil oxidation using different analytical methods4, 17, 18, 20, 24－26）. Although each volatile may have some role in the oxidative deterioration, the most notable volatiles are those formed during the early stage of the oxidation. Recently, we re-ported that acrolein was formed preferentially as a major volatile in fish oil oxidation22）. Acrolein produces undesir-able and irritating odors, with an odor threshold of 3.6 ppb27）. In addition, acrolein is approximately 100 times more reactive than 4-hydroxy-2-nonenal, a well-known toxic lipid oxidation compound28－30）. Thus, acrolein is a key volatile with a strong impact on the flavor and nutritional deterioration of fish oil, especially at the early stage of fish oil oxidation. The present study showed the complete inhi-bition of acrolein formation during more than 400 h oxida-tion of fish oil TAG（Fig. 9A）. The formation of other major volatiles from fish oil TAG oxidation, propanal and 1-pen-tane-3-ol, was also inhibited by the combination of d18:0 and α-tocopherol（Fig. 9B and C）.

Our recent study demonstrated the preferential forma-tion of acrolein and propanal in an early stage of the oxida-tion of soybean oil, linseed oil, echium oil, and fish oil TAG, whereas the major volatile composition changed thereaf-

ter22）. Both acrolein and propanal are mainly formed from omega-3 PUFA, such as DHA（22:6n-3）, EPA（20:5n-3）, stearidonic acid（18:4n-3）, and α-linolenic acid（ALA, 18:3n-3）18, 27）. These n-3 PUFA are relatively more easily oxidized and decomposed than another major PUFA, lin-oleic acid（18:2n-6）. Propanal is the main aldehyde formed from outer hydroperoxides of the terminal methyl from omega-3 PUFA, such as 16-OOH-ALA, 18-OOH-EPA, and 20-OOH-DHA. The hydroperoxides are decomposed by the homolytic cleavage of the oxygen-oxygen bond to yield alkoxy radicals and a hydroxyl radical. The alkoxy radicals are further decomposed by carbon-carbon cleavage, result-ing in propanal and the corresponding 1-olefin radical31, 32）. Acrolein could be formed from the reaction of the 1-olefin radical and hydroxyl radical22）. These free radical reactions proceed very quickly and form propanal and acrolein in the very early stage of omega-3 PUFA oxidation. The present study suggests that d18:0 might react with these aldehydes to form strong ant iox idants in the presence o f α-tocopherol, which could effectively inhibit TAG oxidation by the regeneration of α-tocopherol and/or direct inhibition of lipid oxidation（Fig. 4）.

4 CONCLUSIONThe combination of d18:0 with α-tocopherol showed

strong antioxidant activity on polyunsaturated TAG. Acro-lein was found to be the major volatile in fish oil TAG oxi-dation and the most probable contributor to fish oil deteri-oration; however, d18:0 was able to completely inhibit acrolein formation during 380 h oxidation of fish oil at 50℃ in the dark. Thus, the d18:0 and α-tocopherol combination can prevent the oxidative deterioration of fish oil, although further study will be required to clarify the mechanism of antioxidant formation from the reaction of d18:0 and alde-hydes produced in the early stage of fish oil oxidation.

Fig. 8　 Effect of d18:0 and α-tocopherol on the oxidative stability of soybean（A）, linseed（B）, and fish（C）oil TAG during the incubation at 50℃ in the dark. The stability was analyzed by measuring the increase in total volatiles in the headspace gas of the vial. Control: open circle, ＋α-tocopherol: open square, ＋d18:0: solid diamond, ＋α-tocopherol and d18:0: solid triangle. The data were expressed as the mean±SD of three separate experiments.


J. Oleo Sci. 65, (9) 713-722 (2016)

721

ACKNOWLEDGEMENTSThis work was supported by “Scientific technique re-

search promotion program from agriculture, forestry, fish-eries and food industry” from the Ministry of Agriculture, Forestry and Fisheries in Japan.

References1） Miyashita, K.; Mikamia, N.; Hosokawa, M. Chemical

and nutritional characteristics of brown seaweed lip-ids: A review. J. Funct. Foods 5, 1507-1517（2013）.

2） De Backer, G.; Ambrosioni, E.; Borch-Johnsen, K.; Brotons, C.; Cifkova, R.; Dallongeville, J.; Ebrahim, S.; Faergeman, O.; Graham, I.; Mancia, G.; Cats, V. M.; Orth-Gomér, K.; Perk, J.; Pyörälä, K.; Rodicio, J. L.; Sans, S.; Sansoy, V.; Sechtem, U.; Silber, S.; Thomsen, T.; Wood, D. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and Other Societies on Car-diovascular Disease Prevention in Clinical Practice. Eur. Heart J. 24, 1601-1610（2003）.

3） Smith Jr, S. C.; Allen, J.; Blair, S. N.; Bonow, R. O.;

Brass, L. M.; Fonarow G. C.; Grundy, S. M.; Hiratzka, L.; Jones, D.; Krumholz, H. M.; Mosca, L.; Pasternak, R. C.; Pearson, T.; Pfeffer, M. A.; Taubert, K. A. AHA/ACC guidelines for secondary prevention for patients with coronary and other atherosclerotic vascular disease: 2006 update: endorsed by the National Heart, Lung, and Blood Institute. Circulation 113, 2363-2372（2006）.

4） Benjamin, B. A.; Cameron-Smith, D.; Hofman, P. L.; Cutfield, W. S. Oxidation of marine omega-3 supple-ments and human health. Biomed. Res. Int. 2013, 464921（2013）.

5） Taneja, A.; Singh, H. Challenges for the delivery of long-chain n-3 fatty acids in functional foods. Annu. Rev. Food Sci. Technol. 3, 105-123（2012）.

6） Mcclements, D. J.; Decker, E. A. Lipid oxidation in oil-in-water emulsions: impact of molecular environment on chemical reactions in heterogeneous food systems. J. Food Sci. 65, 1270-1282（2000）.

7） Shah, M. A.; Bosco, S. J. D.; Mir, S. A. Plant extracts as natural antioxidants in meat and meat products. Meat Sci. 98, 21-33（2014）.

8） Miyashita, K. Paradox of omega-3 PUFA oxidation.

Fig. 9　 Effect of d18:0 and α-tocopherol on the formation of major volatiles during the fish oil oxidation, acrolein（A）, propanal（B）, 1-peneten-3-ol（C）, pentane（D）. Control: open circle, ＋α-tocopherol: open square, ＋d18:0: solid diamond, ＋α-tocopherol and d18:0: solid triangle. The data were expressed as the mean±SD of three separate experiments.


J. Oleo Sci. 65, (9) 713-722 (2016)

722

Eur. J. Lipid Sci. Tech. 116, 1268-1279（2014）.9） Lu, F. S. H.; Nielsen, N. S.; Timm-Heinrich, M.; Jacob-

sen, C. Oxidative stability of marine phospholipids in the liposomal form and their applications. Lipids 46, 3-23（2011）.

10） Shimajiri, J., Shiota, M., Hosokawa, M., Miyashita, K. Synergistic antioxidant activity of milk sphingomyeline and its sphingoid base with α-tocopherol on fish oil triacylglycerol. J. Agric. Food Chem. 61, 7969-7975（2013）.

11） Akasaka, K.; Sasaki, I.; Ohrui, H.; Meguro, H. Simple fluorometry of hydroperoxides in oils and foods. Bios-ci. Biotechnol. Biochem. 56, 605-607（1992）.

12） Prevot, A. F.; Mordret, F. X. Utilisation des colonnes capillaries de verre pour l’analyse des corps gras par chromotographie en phase gazeuse. Rev. Fr. Corps. Gras. 23, 409-423（1976）.

13） Cho, S.-Y.; Miyashita, K.; Miyazawa, T.; Fujimoto, K.; Kaneda, T. Autoxidation of ethyl eicosapentaenoate and docosahexaenoate. J. Am. Oil Chem. Soc. 64, 876-879（1987）.

14） Takenaka, A.; Hosokawa, M.; Miyashita, K. Unsaturat-ed phosphatidylethanolamine as effective synergist in combination with α-tocopherol. J. Oleo Sci. 56, 511-516（2007）.

15） Ahmad, I.; Alaiz, M.; Zamora, R.; Hidalgo, F. J. Effect of oxidized lipid/amino acid reaction products on the an-tioxidative activity of common antioxidants. J. Agric. Food Chem. 46, 3768-3771（1998）.

16） Zamora, R.; Nogales, F.; Hidalgo, F. J. Phospholipid ox-idation and nonenzymatic browning development in phosphatidylethanolamine/ribose/lysine model sys-tems. Eur. Food Res. Technol. 220, 459-465（2005）.

17） Dehaut, A.; Himber, C.; Mulak, V.; Grard, T.; Krzewins-ki, F.; Le Fur, B.; Duflos, G. Evolution of volatile com-pounds and biogenic amines throughout the shelf life of marinated and salted anchovies（Engraulis encra-sicolus）. J. Agric. Food Chem. 62, 8014-8022（2014）.

18） Gómez-Cortés, P.; Sacks, G. L.; Brenna, J. T. Quantita-tive analysis of volatiles in edible oils following accel-erated oxidation using broad spectrum isotope stan-dards. Food Chem. 174, 310-318（2015）.

19） Sae-leaw, T.; Benjakul, S. Oxidative stability of lipids from hepatopancreas of Pacific white shrimp（Litope-naeus vannamei）as affected by essential oils incorpo-ration. Eur. J. Lipid Sci. Tech. 116, 987-995（2014）.

20） Serra, A.; Buccioni, A.; Rodriguez-Estrada, M. T.; Con-te, G.; Cappucci, A.; Mele, M. Fatty acid composition,

oxidation status and volatile organic compounds in “Colonnata” lard from Large White or Cinta Senese pigs as affected by curing time. Meat Sci. 97, 504-512（2014）.

21） Snyder, J. M.; Frankel, E. N.; Selke, E.; Warner, K. Comparison of gas chromatographic methods for vola-tile lipid oxidation compounds in soybean oil. J. Am. Oil Chem. Soc. 65, 1617-1620（1988）.

22） Shibata, A.; Uemura, M.; Hosokawa, M.; Miyashita, K. Formation of acrolein in the autoxidation of triacylg-lycerols with different fatty acid compositions. J. Am. Oil Chem. Soc. 92, 1661-1670（2016）.

23） Jacobsen, C. Sensory impact of lipid oxidation in com-plex food systems. Fett/Lipid 101, 484-492（1999）.

24） Venkateshwarlu, G; Let, M. B.; Meyer, A. S.; Jacobsen, C. Modeling the sensory impact of defined combina-tions of volatile lipid oxidation products on fishy and metallic off-flavors. J. Agric. Food Chem. 52, 1635-1641（2004）.

25） Karahadian, C.; Lindsay, R. C. Evaluation of com-pounds contributing characterizing fishy flavors in fish oils. J. Am. Oil Chem. Soc. 66, 953-960（1989）.

26） Sae-leaw, T.; Benjakul, S. Fatty acid composition, lipid oxidation, and fishy odour development in seabass（Lates calcarifer）skin during iced storage. Eur. J. Lipid Sci. Tech. 116, 885-894（2014）.

27） Endo, Y.; Hayashi, C.; Yamanaka, T.; Takayose, K.; Ya-maoka, M.; Tsuno, T.; Nakajima, S. Linolenic acid as the main source of acrolein formed during heating of vegetable oils. J. Am. Oil Chem. Soc. 90, 959-964（2013）.

28） Shibamoto, T. Analytical methods for trace levels of reactive carbonyl compounds formed in lipid peroxi-dation systems. J. Pharm. Biomed. Anal. 41, 12-25（2006）.

29） Singh, M.; Nam, D. T.; Arseneault, M.; Ramassamy, C. Role of by-products of lipid oxidation in Alzheimer’s disease brain: a focus on acrolein. J. Alzheimer’s Dis. 21, 741-756（2010）.

30） Moghe, A.; Ghare, S.; Lamoreau, B.; Mohammad, M.; Barve, S.; McClain, C.; Joshi-Barve, S. Molecular mech-anisms of acrolein toxicity: relevance to human dis-ease. Toxicol. Sci. 143, 242-255（2015）.

31） Frankel, E. N. Chemistry of free radical and singlet ox-idation of lipids, Prog. Lipid Res. 23, 197-221（1985）.

32） Porter, N. A.; Caldwell, S. E.; Mills, K. A. Mechanisms of free radical oxidation of unsaturated lipids. Lipids 30, 277-290（1995）.

Documents

Inhibitory Effect of Dihydrosphingosine with ... - JST