Enzymatic synthesis of steryl esters of polyunsaturated fatty acids

Article abstract of DOI:10.1007/s11746-999-0164-6

Steryl esters of long-chain fatty acids have water-holding properties, and polyunsaturated fatty acids (PUFA) have various physiological functions. Because steryl ester of PUFA can be expected to have both features, we attempted to synthesize steryl esters of PUFA by enzymatic methods. Among lipases used, Pseudomonas lipase was the most effective for the synthesis of cholesteryl docosahexaenoate. When a mixture of cholesterol/docosahexaenoic acid (3:1, mol/mol), 30% water, and 3000 units/g of lipase was stirred at 40 °C for 24 h, the esterification extent attained 89.5%. Under the same reaction conditions, cholesterol, cholestanol, and sitosterol were also esterified efficiently with docosahexaenoic, eicosapentaenoic, arachidonic, and γ-linolenic acids.

Full text of DOI:10.1007/s11746-999-0164-6

Enzymatic Synthesis of Steryl Esters
of Polyunsaturated Fatty Acids
Yuji Shimadaa,*, Yoshinori Hirotab, Takashi Babac, Akio Sugiharaa,
Shigeru Moriyamac, Yoshio Tominagaa, and Tadamasa Teraib
aOsaka Municipal Technical Research Institute, Osaka 536-8553, Japan, bOsaka Institute of Technology,
Department of Applied Chemistry, Osaka 535-8585, and cMaruha Corporation, Ibaraki 300-4295, Japan
ABSTRACT: Steryl esters of long-chain fatty acids have water- chain fatty acids were synthesized with Candida rugosa li-
holding properties, and polyunsaturated fatty acids (PUFA) have
various physiological functions. Because steryl ester of PUFA
can be expected to have both features, we attempted to synthe-
size steryl esters of PUFA by enzymatic methods. Among lipases
used, Pseudomonas lipase was the most effective for the syn-
pase (1), while steryl esters of PUFA have not been synthe-
sized. This paper deals with the synthesis of steryl esters of
PUFA with a lipase from Pseudomonas sp.
thesis of cholesteryl docosahexaenoate. When a mixture of cho- MATERIALS AND METHODS
lesterol/docosahexaenoic acid (3:1, mol/mol), 30% water, and
Chemicals. Sterols and linoleic acid were purchased from
3000 units/g of lipase was stirred at 40°C for 24 h, the esterifi-
cation extent attained 89.5%. Under the same reaction condi-
tions, cholesterol, cholestanol, and sitosterol were also esteri-
fied efficiently with docosahexaenoic, eicosapentaenoic,
arachidonic, and γ-linolenic acids.
Tokyo Kasei Kogyo Co. (Tokyo, Japan). DHA and EPA were
products of Maruha Corp. (Tokyo, Japan), and their purities
were 95%. GLA (purity, 97%) and AA (92%) were gifts from
Nippon Synthetic Chemical Industry Co. Ltd. (Osaka, Japan)
and Suntory Co. (Osaka, Japan), respectively. The molar
amount of fatty acid was calculated on the basis of its acid value.
Lipases. Lipases were gifts from the following companies:
Pseudomonas aeruginosa lipase (LPL; Toyobo Co. Ltd.,
Osaka, Japan); Pseudomonas sp. lipase (LIPOSAM; Showa
Denko, K.K., Tokyo, Japan); Pseudomonas sp. lipases (Li-
pase-AK, and Lipase-PS; Amano Pharmaceutical Co. Ltd.,
Aichi, Japan), C. rugosa lipase (Lipase-OF; Meito Sangyo
Co. Ltd., Aichi, Japan) and Rhizopus delemar lipase (Ta-li-
pase; Tanabe Seiyaku Co. Ltd., Osaka, Japan). Lipase activ-
ity was measured by titrating fatty acids liberated from olive
oil (Wako Pure Chemical Ind. Co., Osaka, Japan) with 50 mM
KOH as described previously (8). The reaction was carried
out at 30°C for 30 min with stirring at 500 rpm. One unit (U)
of lipase activity was defined as the amount of enzyme that
liberated 1 µmol of fatty acid per minute.
Reaction. A reaction mixture of sterol, PUFA, water, and li-
pase was stirred at 500 rpm in a 50-mL screw capped vial sealed
with nitrogen gas. The esterification extent was calculated from
the amount of fatty acid consumed during the reaction.
Silica gel column chromatography. Steryl ester and sterol
were extracted with 150 mL n-hexane after adding 70 mL of
0.5 N KOH (30% ethanol solution) to 10–15 g of reaction
mixture. The extracts (ca. 10 g) were applied to a silica gel
60 column (30 × 250 mm; Merck, Darmstadt, Germany), and
steryl ester was eluted with a mixture of n-hexane/ethyl ac-
etate (98:2, vol/vol).
Paper no. J9028 in JAOCS 76, 713–716 (June 1999).
KEY WORDS: Esterification, lipase, polyunsaturated fatty acid,
Pseudomonas, steryl ester.
Steryl esters of fatty acids have water-holding properties and
are widely used as ingredients of cosmetics and bath additives
(1). Furthermore, cholesteryl esters were reported to be very im-
portant as intercellular lipids (2). On the other hand, polyunsat-
urated fatty acids (PUFA) are used as pharmaceutical sub-
stances, ingredients of cosmetics, and in health foods and food
materials. Actually, the ethyl ester of eicosapentaenoic acid
(EPA; 20:5n-3) has been used in the treatment of arteriosclero-
sis obliterans and hyperlipemia (3). Tuna oil containing docosa-
hexaenoic acid (DHA; 22:6n-3) and borage oil containing γ-
linolenic acid (GLA; 18:3n-6) have been used as components in
infant formulas, health foods, and food materials (4,5). In addi-
tion, a single-cell oil containing arachidonic acid (AA; 20:4n-6)
has been desired as a component in infant formulas, because it
shows an effect on the growth acceleration of infants as well as
does DHA (6,7). Thus, we attempted to synthesize steryl esters
of PUFA which are expected to have both features.
Because PUFA are very sensitive to heat and oxidation,
enzymatic catalysis, which proceeds efficiently under mild
conditions, is attractive for the synthesis of steryl esters of
PUFA. It has already been reported that steryl esters of long-
Analysis. Fatty acids in cholesteryl ester were methylated
in methanol using sodium methylate as a methylating reagent,
and were analyzed with a Hewlett-Packard 5890 plus gas
*To whom correspondence should be addressed at Osaka Municipal Techni-
cal Research Institute, 1-6-50 Morinomiya, Joto-ku, Osaka 536-8553, Japan.
E-mail: shimaday@omtri.city.osaka.jp
Copyright © 1999 by AOCS Press
713
JAOCS, Vol. 76, no. 6 (1999)

714
Y. SHIMADA ET AL.
chromatograph (Avondale, PA) connected to a DB-5 capillary of cholesterol increased with increasing amounts of the li-
column (0.25 mm × 10 m; J&W Scientific, Folsom, CA). The pase, but did not increase significantly once more than 2000
column temperature was raised from 190 to 290°C at a rate U/g-reaction mixture of the lipase was used.
of 5°C/min, and then raised to 320°C at a rate of 10°C/min.
Effect of water content on the synthesis of ChDHA was
The temperatures of the injector and detector (FID, flame-ion- examined (Fig. 1). The highest esterification extent was ob-
ization detector) were set at 245 and 340°C, respectively. The tained in mixtures containing 30 to 50% water. In general, es-
carrier gas was helium at a flow rate of 25 cm/s. Thin-layer terification proceeded efficiently in the mixture with smaller
chromatography was performed by developing substances on amount of water. In this reaction, the enzyme exists in the
a silica gel plate 60 (Merck) with a mixture of n-hexane/ethyl water phase in the reaction mixture. Therefore, a smaller
acetate/acetic acid (80:20:1, vol/vol/vol). Infrared (IR) spec- amount of water reduces the contact of the enzyme with the
trum was measured with a Shimadzu FTIR-8100M spec- substrates and results in low esterification extent.
trophotometer (Kyoto, Japan). Nuclear magnetic resonance
Figure 2 shows the effect of the amount of DHA on esteri-
(NMR) spectrum was measured in CDCl3 using tetramethyl- fication of cholesterol. An increase in the extent of esterifica-
silane as the internal standard with a UNITY 300 spectrome- tion was not observed even though more than three molar
ter (Varian, CA). The molecular mass was analyzed by field
TABLE 2
desorption mass spectrometry (FD-MS) performed on a JMS-
Effect of Lipase Amount on Synthesis of Cholesteryl Docosa-
DX303HF (JEOL Ltd., Tokyo, Japan).
hexaenoatea
Lipase amount (U/g)b
Esterification (%)c
RESULTS AND DISCUSSION
400
700
1000
2000
4000
6.4
24.2
39.7
75.3
80.0
Screening for suitable lipase for synthesis of cholesteryl es-
ters of PUFA. Because lipases act generally on DHA most
weakly among PUFA, we screened for a suitable lipase for
the synthesis of cholesteryl docosahexaenoate (ChDHA)
(Table 1). Rhizopus delemar lipase did not synthesize
ChDHA, because the lipase is a 1,3-positional specific lipase
and does not act on secondary alcohol (9). Candida rugosa
lipase is nonspecific toward ester bonds of triglyceride and
acts on secondary alcohol (10). The enzyme synthesized cho-
lesteryl linoleate (esterification extent, 73.7%), but not
ChDHA. This result was shown to be due to the very weak
activity on DHA. Several Pseudomonas lipases act on sec-
ondary alcohol, and also on DHA somewhat (11,12).
Pseudomonas sp. lipase (LIPOSAM) was the most effective
among the enzymes tested. Thus the lipase was selected for
the following experiments.
aA mixture of 2.4 g of DHA/cholesterol (2:1, mol/mol) and 0.6 g water was
stirred at 30°C for 20 h with various amounts of Pseudomonas lipase (LI-
POSAM).
bLipase amount per 1 g of reaction mixture.
cEsterification extent of cholesterol. See Table 1 for abbreviations and com-
pany source.
Several factors affecting esterification of cholesterol with
DHA. Cholesterol was esterified with two molar equivalents
of DHA in a mixture containing 20% water, by weight, using
various amounts of lipase (Table 2). The esterification extent
TABLE 1
Suitable Lipases for Synthesis of Cholesteryl Docosahexaenoatea
Lipase
Esterification (%)b
Pseudomonas aeruginosa
Pseudomonas sp.c
Pseudomonas sp.d
Pseudomonas sp.e
Candida rugosa
38.3
62.5
14.0
10.0
3.0
Rhizopus delemar
N.D.f
aA mixture of 2.4 g docosahexaenoic acid (DHA)/cholesterol (=2:1,
mol/mol), 0.6 g water and 4500 units (U) lipase was incubated at 30°C for
20 h with stirring at 500 rpm.
FIG. 1. Effect of water content on esterification of cholesterol with do-
cosahexaenoic acid (DHA). A mixture (3.0 g) of DHA/cholesterol (=2:1,
mol/mol) and various amounts of water was stirred at 30°C for 20 h with
3000 units (U)/g-reaction mixture of Pseudomonas lipase (LIPOSAM;
Showa Denko, K.K., Tokyo, Japan). Percentages of water are expressed
on a weight basis.
bEsterification extent of cholesterol.
cLIPOSAM (Showa Denko, K.K., Tokyo, Japan).
dLipase-AK (Amano Pharmaceutical Co. Ltd., Aichi, Japan).
eLipase-PS (Amano Pharmaceutical Co. Ltd.)
fN.D., Not detected.
JAOCS, Vol. 76, no. 6 (1999)

SYNTHESIS OF STERYL ESTER OF PUFA
715
equivalents of DHA was added. The effect of temperature is CH3), 1.01 (3H, s, CH3), 1.07–1.70 (22H, m), 1.78–1.91 (2H,
shown in Figure 3. The esterification extent at the early stage m), 1.92–2.13 (4H, m), 2.26–2.42 (6H, m), 2.78–2.90 (10H,
of reaction (after 4 h) increased with raising temperature, but m), 4.53–4.67 (1H, m, >CH–O–CO–), 5.18–5.47 (13H, m,
the extent after 20 h was constant above 30°C.
–HC=CH– and >C=CH–); 13C-NMR δ = 11.8, 14.2, 18.7,
On the basis of the above results, the reaction conditions 19.3, 20.5, 21.0, 22.5, 22.8, 22.9, 23.8, 24.3, 25.5, 25.6 × 4,
were set as follows: A reaction mixture containing DHA/cho- 27.8, 28.0, 28.2, 31.8, 31.9, 34.5, 35.8, 36.2, 36.6, 37.0, 38.1,
lesterol (3:1, mol/mol), 30% water, and 3000 U/g-reaction 39.5, 39.7, 42.3, 50.0, 56.1, 56.7, 73.9, 122.6, 127.0, 127.8,
mixture of Pseudomonas lipase was incubated at 40°C with 128.0, 128.05, 128.1 × 2, 128.16 × 2, 128.2, 128.5, 129.1,
stirring at 500 rpm. The reaction period was extended to over 132.0, 139.6, 172.4. Signals from the carbonyl carbon of
24 h under these conditions, but the esterification extent did DHA and the secondary alcohol carbon at C-3 of cholesterol
not increase. Thus the reaction period was fixed at 24 h.
were observed at 179.6 and 71.7 ppm, respectively. These sig-
Structure of synthesized product. Cholesterol (2.5 g) was es- nals of the synthesized product shifted to 172.4 and 73.9 ppm,
terified with DHA under the above conditions (esterification ex- respectively. It was therefore shown that DHA was bound at
tent, 93.5%). The reaction product was extracted with n-hexane the hydroxy group at C-3 of cholesterol. The molecular
under alkaline conditions, and purified by silica gel column weight ion was observed at m/z 696 (M+) on FD-MS. These
chromatography (purified product, 2.7 g). The product was results show that the synthesized product is ChDHA.
methylated and then analyzed by gas chromatography. As a re-
Synthesis of steryl esters of PUFA. To show that the enzy-
sult, cholesterol and methyl docosahexaenoate were detected at matic method is effective for the synthesis of steryl esters of
the molar ratio of 50:48, which showed that the synthesized PUFA, several sterols were esterified with PUFA under the
product consisted of equal amounts of cholesterol and DHA.
same conditions described before (Table 3). All the sterols
The absorption maxima of the IR spectrum of the synthe- were efficiently esterified with fatty acids used as substrates.
sized product were observed at the following position: 3013, The reaction mixture contains 30% water, but the esterifica-
1736 (C=O), 1466, and 1171 (C–O–) cm−1. The absorbance tion extents of all reactions were more than 85%. When cho-
of hydroxy group was observed in cholesterol and DHA, but lesteryl, cholestanyl, and sitosteryl docosahexaenoates were
not in the synthesized product. The absorbance of the car- stirred at 40°C for 24 h in a reaction mixture containing 30%
bonyl group in the product shifted to a shorter wavelength by water, their hydrolysis extents were only 15.9, 18.5, and
23 cm-1, indicating the formation of ester bond.
11.8%, respectively. This result showed that the equilibrium
1
The synthesized product was identified by H and 13C of the reaction was shifted to esterification because steryl es-
1
NMR, and mass spectroscopies. H-NMR δ = 0.68 (3H, s, ters were poor substrates of the lipase.
CH3), 0.86 (3H, d, J = 6.6 Hz, CH3), 0.87 (3H, d, J = 6.6 Hz,
Superiority of reaction presented here. In general, esterifi-
CH3), 0.92 (3H, d, J = 6.6 Hz, CH3), 0.97 (3H, t, J = 7.5 Hz, cation reactions using lipases are disturbed by water gener-
FIG. 2. Effect of amount of DHA on esterification of cholesterol. Cho-
lesterol was esterified at 30°C for 20 h with various amounts of DHA in FIG. 3. Effect of temperature on esterification of cholesterol with DHA.
a mixture containing 30% water and 3000 U/g-reaction mixture of
A mixture containing 2.1 g of DHA/cholesterol (3:1, mol/mol), 0.9 g
Pseudomonas lipase (LIPOSAM). The amount of DHA was expressed as water and 9000 U Pseudomonas lipase (LIPOSAM) was stirred for 4 h
a molar ratio to that of cholesterol. See Figure 1 for abbreviations and
company source.
(hatched box) or 20 h (solid box) at a range of 20 to 50°C. See Figure 1
for abbreviations and company source.
JAOCS, Vol. 76, no. 6 (1999)

716
Y. SHIMADA ET AL.
TABLE 3
2. Imokawa G., and M. Hattori, A Possible Function of Structural
Esterification of Cholesterol, Cholestanol, and Sitosterol
Lipids in the Water-Holding Properties of the Stratum Corneum,
J. Invest. Dermatol. 84:282–284 (1985).
with Polyunsaturated Fatty Acidsa
3. Hara, K., Pharmaceutical Application of Icosapentaenoic Acids,
Yushi (in Japanese) 46:91–99 (1993).
4. Maruyama, K., and M. Nishikawa, Function of Fish Oil and Its
Application to Foods, Food Chemicals (in Japanese)
1995(4):31–37 (1995).
5. Crozier, G.L., and M.-C. Seretin, γ-Linolenic Acid in Infant For-
mula, in γ-Linolenic Acid: Metabolism and Its Roles in Nutri-
tion and Medicine, edited by Y.-S. Huang and D.E. Mills,
AOCS Press, Champaign, 1996, pp. 246–251.
6. Carlson, S.E., S.H. Werkman, J.M. Peeples, R.J. Cooke, and
E.A. Tolley, Arachidonic Acid Status Correlates with First Year
Growth in Preterm Infant, Proc. Natl. Acad. Sci. USA
90:1073–1077 (1993).
7. Lanting, C.I., V. Fidler, M. Huisman, B.C.L. Touwen, and E.R.
Boersma, Neurological Differences Between 9-Year-Old Chil-
dren Fed Breast-Milk as Babies, Lancet 344:1319–1322 (1994).
8. Sugihara, A., Y. Shimada, and Y. Tominaga, Separation and
Characterization of Two Molecular Forms of Geotrichum can-
didum Lipase, J. Biochem. 107:426–430 (1990).
Sterol
Fatty acid
Esterification (%)b
Cholesterol
Docosahexaenoic acid
Eicosapentaenoic acid
Arachidonic acid
γ-Linolenic acid
89.5
88.0
92.1
89.5
91.9
87.3
86.9
90.6
91.7
92.7
85.3
91.5
91.2
Linoleic acid
Cholestanol
Sitosterol
Docosahexaenoic acid
Eicosapentaenoic acid
γ-Linolenic acid
Linoleic acid
Docosahexaenoic acid
Eicosapentaenoic acid
γ-Linolenic acid
Linoleic acid
aA mixture of fatty acid/sterol (3:1, mol/mol), 30% water, and 3000 U/g-reac-
tion mixture of Pseudomonas lipase (LIPOSAM) was stirred at 40°C for 24 h.
bEsterification extent of sterol. See Table 1 for abbreviations and company
source.
9. Okumura, S., M. Iwai, and Y. Tsujisaka, Positional Specificities
of Four Kinds of Microbial Lipases, Agric. Biol. Chem.
40:655–660 (1976).
10. Tanaka, Y., J. Hirano, and T. Funada, Concentration of Docosa-
hexaenoic Acid in Glyceride by Hydrolysis of Fish Oil with
Candida cylindracea Lipase, J. Am. Oil Chem. Soc.
69:1210–1214 (1992).
11. Shimada, Y., K. Maruyama, A. Sugihara, S. Moriyama, and Y.
Tominaga, Purification of Docosahexaenoic Acid from Tuna Oil
by a Two-Step Enzymatic Method: Hydrolysis and Selective Es-
terification, Ibid. 74:1441–1446 (1997).
12. Shimada Y., A. Sugihara, M. Shibahiraki, H. Fujita, H. Nakano,
T. Nagao, T. Terai, and Y. Tominaga, Purification of γ-Linolenic
Acid from Borage Oil by a Two-Step Enzymatic Method, Ibid.
74:1465–1470 (1997).
ated through the reaction. To remove the water, methods of
using a large amount of n-hexane (13,14), of adding molecu-
lar sieves (15), and of removing water with vacuum pump
(16) have been proposed. The use of organic solvent requires
a large-scale reactor and involves a risk of explosion. The ad-
dition of molecular sieves or the use of vacuum pump is not
suitable for the industrial production because of high cost.
The method described here may be effective for a large-scale
production of steryl esters of PUFA, because the high esterifi-
cation extent can be obtained even though the reaction mix-
ture contains water.
13. Kirchner, G., M.P. Scollar, and Alexander M. Kribanov, Reso-
lution of Racemic Mixture via Lipase Catalysis in Organic Sol-
vents, J. Am. Chem. Soc. 107:7072–7076 (1985).
ACKNOWLEDGMENTS
14. Foglia, T.A., and P.E. Sonet, Fatty Acid Selectivity of Lipases:
γ-Linolenic Acid from Borage Oil, J. Am. Oil Chem. Soc.
72:417–420 (1995).
15. Yamaguchi, S., and T. Mase, High-Yield Synthesis of Mono-
glyceride by Mono- and Diacylglycerol Lipase from Penicillium
camembertii U-150, J. Ferment. Bioeng. 72:162–167 (1991).
16. Tanaka, Y., T. Funada, and J. Hirano, Esterification by Lipases:
Preparation of Triicosapentaenoyl and Tridocosahexaenoyl
Glycerols, Yukagaku (in Japanese) 41:563–567 (1992).
We thank Kazuaki Maruyama, Central Research Institute, Maruha
Corp., for his technical assistance, and Takumi Mizuno, Osaka
Municipal Technical Research Institute, for mass spectroscopic
measurement.
REFERENCES
1. Myojo K., and Y. Matsufune, Process for Preparing Sterol Fatty
Acid Esters with Enzymes, Yukagaku (in Japanese) 44:883–896
(1995).
[Received September 24, 1998; accepted February 16, 1999]
JAOCS, Vol. 76, no. 6 (1999)