Enzymatic synthesis of symmetrical 1,3-diacylglycerols by direct esterification of glycerol in solvent-free system

Article abstract of DOI:10.1007/s11746-999-0074-7

1,3-Diacylglycerols were synthesized by direct esterification of glycerol with free fatty acids in a solvent-free system. Free fatty acids with relatively low melting points (<45°C) such as unsaturated and medium-chain saturated fatty acids were used. With stoichiometric ratios of the reactants and water removal by evaporation at 3 mm Hg vacuum applied at 1 h and thereafter, the maximal 1,3-diacylglycerol content in the reaction mixture was: 84.6% for 1,3-dicaprylin, 84.4% for 1,3-dicaprin, 74.3% for 1,3-dilinolein, 71.7% for 1,3-dieicosapentaenoin, 67.4% for 1,3-dilaurin, and 61.1% for 1,3-diolein. Some of the system's parameters (temperature, water removal, and molar ratio of the reactants) were optimized for the production of 1,3-dicaprylin, and the maximal yield reached 98%. The product was used for the chemical synthesis of 1,3-dicapryloyl-2-eicosapentaenoylglycerol. The yield after purification was 42%, and the purity of the triacylglycerol was 98% (both 1,3-dicapryloyl-2-eicosapentaenoylglycerol and 1,2-dicapryloyl-3-eicosapentaenoylglycerol included) by gas chromatographic analysis, of which 90% was the desired structured triacylglycerol (1,3-dicapryloyl-2-eicosapentaenoylglycerol) as determined by silver ion high-performance liquid chromatographic analysis.

Full text of DOI:10.1007/s11746-999-0074-7

Enzymatic Synthesis of Symmetrical 1,3-Diacylglycerols
by Direct Esterification of Glycerol in Solvent-Free System
Roxana Rosu, Mamoru Yasui, Yugo Iwasaki, and Tsuneo Yamane*
Laboratory of Molecular Biotechnology, Graduate School of Bio- and Agro-Sciences, Nagoya University, Nagoya 464-8601, Japan
ABSTRACT: 1,3-Diacylglycerols were synthesized by direct es-
terification of glycerol with free fatty acids in a solvent-free sys-
tem. Free fatty acids with relatively low melting points (<45°C)
such as unsaturated and medium-chain saturated fatty acids
were used. With stoichiometric ratios of the reactants and water
removal by evaporation at 3 mm Hg vacuum applied at 1 h and
thereafter, the maximal 1,3-diacylglycerol content in the reac-
tion mixture was: 84.6% for 1,3-dicaprylin, 84.4% for 1,3-di-
caprin, 74.3% for 1,3-dilinolein, 71.7% for 1,3-dieicosapen-
taenoin, 67.4% for 1,3-dilaurin, and 61.1% for 1,3-diolein.
Some of the system’s parameters (temperature, water removal,
and molar ratio of the reactants) were optimized for the produc-
High yields of 1,3-DAG cannot be obtained by direct chem-
ical methods because they lack positional selectivity. Multistep
reaction sequences are necessary, and large quantities of sol-
vents are used as reaction media and for purification (7,8). The
classical synthesis requires starting from the 1-monoacylglyc-
erol (1-MAG), which is acylated with the acid chloride in the
presence of pyridine and chloroform. Yields of 50–70% are ob-
tained after purification (9).
Better 1,3-DAG yields (85%) can be obtained by solid-
phase enzymatic glycerolysis of triglycerides (TAG), but this
method, based on the selective crystallization of 1,3-DAG, is
tion of 1,3-dicaprylin, and the maximal yield reached 98%. The limited to the glycerides of saturated fatty acids (FA) with rela-
product was used for the chemical synthesis of 1,3-dicapryloyl-
2-eicosapentaenoylglycerol. The yield after purification was
42%, and the purity of the triacylglycerol was 98% (both 1,3-di-
capryloyl-2-eicosapentaenoylglycerol and 1,2-dicapryloyl-3-
eicosapentaenoylglycerol included) by gas chromatographic
analysis, of which 90% was the desired structured triacylglyc-
erol (1,3-dicapryloyl-2-eicosapentaenoylglycerol) as determined
by silver ion high-performance liquid chromatographic analysis.
Paper no. J9081 in JAOCS 76, 839–843 (July 1999).
tively high melting points (10). The process is economically
advantageous for natural fats, but use of the products is limited
to applications where acyl species homogeneity is not impor-
tant. For obtaining 1,3-DAG of the purity necessary for drug
synthesis, pure homogeneous TAG should be synthesized first.
In another study, esterification of glycerol adsorbed be-
forehand onto a solid support was performed in organic sol-
vents with 1,3-specific lipases (11,12). The synthesis of liq-
uid 1,3-DAG (medium-chain and unsaturated 1,3-DAG) was
done via irreversible acyl transfer using vinyl esters as acyl
donors, adding an extra step to the reaction scheme plus the
requisite purification of the vinyl ester. The yields obtained
with stoichiometric ratios of reactants ranged between 52 and
KEY WORDS: 1,3-Diacylglycerol, esterification, Rhizomucor
miehei lipase, structured lipid synthesis.
Besides their use as nonionic surfactants in food, cosmetics, 76% for unsaturated fatty acid vinyl esters and between 80
and pharmaceutical industries, 1,3-diacylglycerols (1,3- and 85% for saturated acid vinyl esters.
DAG) are intermediates for synthesis of various chemical
In this work, a simple one-step procedure for synthesis of
compounds with pharmaceutical applications (1,2). Diglyc- 1,3-DAG in solvent-free system using free fatty acids (FA) in
eride prodrugs can be obtained when the second position is liquid phase is described. Glycerol was esterified directly
esterified with a drug bearing a carboxylic group such as the with a variety of FA by immobilized 1,3-specific Rhizomucor
antiinflammatory niflumic acid (3), by covalently binding an miehei lipase (Lipozyme). The equilibrium was shifted to the
alcohol, for example the β-blocker bupranolol (4), or pheny- synthetic side by removing the water formed during the reac-
toin (5) to 2-chlorosuccinyl-1,3-diacylglycerol, etc. These tion by evaporating at reduced pressure or by blowing dry ni-
forms are better adsorbed and have fewer side effects than the trogen gas or by adding molecular sieves. 1,3-Dicapryloyl-2-
original drugs (1,4). In a similar manner, structured glycerides eicosapentaenoylglycerol, a structured glyceride with high
with a polyunsaturated fatty acid such as docosahexaenoic pharmaceutical potential, was chemically synthesized from
acid (DHA), eicosapentaenoic acid (EPA), and the like in the 1,3-dicapryloylglycerol obtained by this method.
second position and medium-chain acyl groups in 1- and 3-
positions can be used as carriers of these therapeutically and
nutritionally valuable acids (6).
EXPERIMENTAL PROCEDURES
Materials. Commercially available Lipozyme (R. miehei li-
pase immobilized on an anion exchange resin) was a gift from
*To whom correspondence should be addressed.
E-mail: yamanetu@agr.nagoya-u.ac.jp
Novo Nordisk A/S, Bagsvaerd, Denmark. The chemicals used
Copyright © 1999 by AOCS Press
839
JAOCS, Vol. 76, no. 7 (1999)

840
R. ROSU ET AL.
were purchased as follows: caprylic acid (min. 99%), butyric temperature GC according to their molecular weight with an
acid (min. 98%), caproic acid (min. 99%), capric acid (min. SGE-HT5 aluminum-clad fused-silica capillary column (6 m
99%), linoleic acid (min. 88%), glycerol (min. 99%), dicyclo- length, 0.53 mm internal diameter, and 0.1 µm film thickness)
hexylcarbodiimide (DCC) (min. 95%), and 4-dimethyl- from Supelco Inc., (Bellefonte, PA). The triacylglycerol posi-
aminopyridine (4-DMAP) (min. 99%) from Wako Pure Chem- tional isomers were separated by HPLC with a ChromSpher 5
ical Industries, Ltd. (Osaka, Japan); oleic acid (min. 90%) from Lipids silver ion chromatography column (250 × 4.6 mm ×
Nippon Oil & Fats Co., Ltd. (Amagasaki, Japan); lauric acid 1/4′′) from Chrompack (Middleburg, The Netherlands).
(min. 98%) from Tokyo Kasei Industries, Ltd. (Tokyo, Japan).
Eicosapentaenoic acid ethyl ester (min. 97%) was a gift from
Nippon Suisan Co., Ltd. (Tokyo, Japan). EPA was obtained by
RESULTS AND DISCUSSION
hydrolysis of eicosapentaenoic acid ethyl ester (13).
Synthesis of 1,3-DAG. The synthesis of 1,3-DAG by direct
Typical esterification reaction. Esterification was per- esterification is based on the 1,3-positional specificity of im-
formed in a solvent-free system which contained 25 mmol mobilized R. miehei lipase (Lipozyme) and on the critical fea-
FA, 12.5 mmol glycerol (at the stoichiometric ratio of 2:1), ture of the support to retain the water essential for the cat-
and 4% Lipozyme based on the total reaction mixture. The alytic function of the lipase (18). Esterification is a reversible
reaction temperature depended on the melting point of the FA reaction, and the effect of water mass action is used to mini-
used and ranged between 25–45°C. The reaction was carried mize the hydrolytic side reactions and to allow the lipase to
out at normal pressure for 1 h and then at 3 mm Hg vacuum catalyze the synthetic reactions. Lipase speeds up the reac-
or under a stream of dry nitrogen gas (0.7 L/min). When bu- tion and equilibrium is reached faster, but the yields of the es-
tyric and caproic acid were used, instead of vacuum, 1.5 g terification products are not in the practical range at stoichio-
molecular sieves powder 4 A (Nacalai Tesque, Ltd., Kyoto, metric molar ratios of the reactants. The equilibrium compo-
Japan) was added to the reaction mixture for water removal.
sition at constant temperature can be modified in two ways:
Analysis of the reaction mixture. During the enzymatic reac- one is the use of excess acid and the other is the removal of
tion, 2-µL samples were withdrawn from the reaction vessel and water formed as by-product.
dissolved in 0.1 mL chloroform/methanol (2:1). The immobi-
The synthesis of 1,3-dicapryloylglycerol, which can be an
lized enzyme was separated by centrifugation at 20,000 × g for intermediate for synthesis of structured triacylglycerols with
2 min. The removal of glycerol from the supernatant solution a polyunsaturated fatty acid in the 2-position, was used as a
was achieved by extraction with water (0.05 mL). For effective model for studying the effect of water removal and some
extraction of partial glycerides (soluble in water to some extent), other parameters of the system.
the water layer was extracted twice with 0.1 mL chloroform
When the reaction was performed at normal pressure in a
each time. The total chloroform extract (approx. 0.3 mL) was closed vessel, equilibrium was reached after 3 h at only 60%
analyzed by thin-layer chromatography–flame-ionization detec- conversion of the acid (Fig. 1). Adding extra water to the ini-
tion (TLC–FID; Iatroscan MK-5, Iatron Laboratories, Tokyo,
Japan) using Chromarod S III quartz rods. The rods were devel-
oped in benzene/chloroform/acetic acid (70:30:2) solvent for 10
cm at room temperature. The rods were scanned under the fol-
lowing conditions: hydrogen flow rate, 0.16 L min−1; air flow
rate, 2.0 L min−1; scanning speed, 30 s per scan. The results
were expressed as percentage of peak areas of the reaction mix-
ture’s components on a glycerol-free basis and may vary slightly
from the actual weight percentages (14).
The water content of the reaction mixture was determined
with a Karl Fischer moisture meter (MKS-1; Kyoto Electron-
ics, Ltd., Kyoto, Japan).
Synthesis of 1,3-dicapryloyl-2-eicosapentaenoylglycerol. A
solution of 1,3-dicaprylin (95%, 0.172 g, 0.5 mmol), EPA
(97%, 0.227 g, 0.75 mmol), DCC (0.244 g, 2 mmol), and 4-
DMAP (0.244 g, 2 mmol) in 10 mL ethanol-free dry chloro-
form was stirred at room temperature overnight. The liberated
urea was filtered and the filtrate was concentrated. The crude
product was purified by silica gel column chromatography with
hexane/ethyl acetate (9:1), to give 0.15 g 1,3-dicapryloyl-2-
eicosapentaenoylglycerol in 42% yield (15,16). The analyses
of the final product were performed by gas chromatography
(GC) and silver ion high-performance liquid chromatography
(HPLC) (17). Triacylglycerol species were separated by high-
FIG. 1. Effect of water removal on conversion. The reaction was per-
formed with 1.150 g (12.5 mmol) glycerol, 3.605 g (25 mmol) caprylic
acid, and 0.200 g Lipozyme (Novo Nordisk A/S, Bagsvaerd, Denmark)
at 25°C and 300 rpm stirring speed. Reaction performed without water
removal at normal pressure in a closed vessel (▲), reaction with water
removal by evaporation under a stream of nitrogen gas (0.7 L/min) ap-
plied at 1 h (■), reaction with water removal by evaporation under 3
mm Hg vacuum applied at 1 h (●).
JAOCS, Vol. 76, no. 7 (1999)

ENZYMATIC SYNTHESIS OF SYMMETRICAL 1,3-DIGLYCERIDE
841
FIG. 3. Effect of temperature on reaction rate for the esterification of
glycerol with caprylic acid. Reaction performed at 15 (▲), 20 (▲), 25
(■), 35 (●), and 50°C (■). For the rest of the reaction conditions see Fig-
ure 2.
FIG. 2. Time course of esterification of glycerol with caprylic acid. The
reaction was performed with 1.150 g (12.5 mmol) glycerol, 3.605 g (25
mmol) caprylic acid, and 0.200 g Lipozyme at 25°C and 300 rpm stir-
ring speed. Water removal was started at 1 h by evaporation under 3
mm Hg vacuum. Caprylic acid (●), 1,3-dicapryloylglycerol (▲), 1,2-di-
capryloylglycerol (●),1-monocapryloylglycerol (■), and tricapryloyl-
glycerol (▲). For manufacturer see Figure 1.
glycerol content remained low throughout the reaction (less
than 1%). 2-Monocapryloylglycerol was not detected in the
reaction mixture.
tial reaction mixture to enhance the enzyme activity was un-
The reaction temperature influenced the reaction rate (Fig.
necessary because the commercially available enzyme prepa- 3) and the yield of 1,3-dicapryloylglycerol (Fig. 4). From 15
ration (Lipozyme) already contained 4.20% water. Water to 25°C, the reaction accelerated, but still higher temperatures
formed in the reaction accumulated in the reaction mixture, did not improve the reaction rate significantly (Fig. 3). The
and increased from 0.24% initially to 4.50% at equilibrium. content of 1,3-dicapryloylglycerol in the reaction mixture was
The removal of water formed during the reaction is very im- maximal at 8 h for the reaction performed at 25, 35, and 50°C
portant for shifting the reaction equilibrium toward high con- (Fig. 4). At 15 and 20°C, the reaction was slow, and maximal
versions of the acid and thereby high yields of 1,3-dicapryl- 1,3-dicapryloylglycerol content could not be attained in 8 h.
oylglycerol. Water, the most volatile component in this reac- Maximal 1,3-dicapryloylglycerol content (approximately
tion system, was removed efficiently by a stream of nitrogen
gas (0.7 L/min) or by a 3-mm Hg vacuum, bringing the con-
version at 12 h to 98% for both cases (Fig. 1). The water con-
tent of the reaction mixture at 12 h was reduced from 4.5%
for the reaction at normal pressure in a closed vessel to 0.45%
when evaporation under nitrogen was used, and to 0.33% for
the reaction at 3 mm Hg. Water removal was started at 1 h to
maintain the high initial reaction rates. A study of water re-
moval under a stream of nitrogen gas showed that applying
nitrogen from the beginning lowered the initial rate of the
acid conversion to 8.1%/h from 25.1%/h for the reaction
without water removal. Perhaps the water formed in the ini-
tial stage of the reaction helps the enzyme reach maximal ac-
tivity. The extension of the initial stage to 2 h did not bring
any further improvement of the reaction rates for the two
methods. For the rest of the experiments, 3 mm Hg vacuum
applied after 1 h was used for the water removal.
Figure 2 is an example of the time course of the reaction.
The high reaction rate ensured a consumption of more than
95% of caprylic acid in 8 h. The content of 1,3-dicapryloyl-
glycerol was maximal at 8 h (84.6%), but decreased later
owing to tricapryloylglycerol formation. 1,2-Dicapryloyl- Figure 2.
FIG. 4. Effect of temperature on glyceride content for the esterification
of glycerol with caprylic acid at 8 h. The content of 1,3-dicapryloyl-
glycerol (■), 1,2-dicapryloylglycerol (▲), tricapryloylglycerol (■), and
1-monocapryloylglycerol (●). For the rest of the reaction conditions see
JAOCS, Vol. 76, no. 7 (1999)

842
R. ROSU ET AL.
oylglycerol during the removal of acid from the reaction mix-
ture. The caprylic acid recovered can be recycled, and the pu-
rified product can be used directly as an intermediate for an-
other synthesis.
Various FA were used in this system, and the results are
shown in Table 1. The reaction did not work at all for butyric
and caproic acids. The best results were obtained for synthe-
ses of 1,3-dicaprylin and 1,3-dicaprin (their maximal contents
were 84.6 and 84.4%, respectively). The content of 1,3-di-
olein was only 61.1%, but the reaction was slow and incom-
plete in 12 h. The content of 1,2-DAG was very low in all
syntheses, making the product purification by silica gel chro-
matography easier.
The lipase specificity for the FA species can be investi-
gated using the initial reaction rates (Table 1). The initial re-
action rate was defined as mmol fatty acid consumed per
FIG. 5. Effect of molar ratio of reactants on glyceride composition for minute per milligram Lipozyme. The reaction rate was high
the esterification of glycerol with caprylic acid at 8 h: 1,3-dicapryloyl-
glycerol (▲), 1-monocapryloylglycerol (■), and tricapryloylglycerol (▲).
For the rest of the reaction conditions see Figure 2.
for caprylic and eicosapentaenoic acids, low for linoleic and
oleic acids, and zero for butyric and caproic acids. These re-
actions were performed at the same temperature (25°C) and
therefore the initial reaction rates can be compared directly.
85%) was achieved at temperatures between 25 and 35°C. At The reaction temperature was 35°C for capric acid and 45°C
higher temperatures, the maximal content of 1,3-dicapryloyl- for lauric acid because they have higher melting points. The
glycerol decreased because of the faster tricapryloylglycerol higher temperatures accelerate the acid conversion, as well as
formation. The low content of 1,2-DAG at higher tempera- the triacylglycerol formation owing to the faster spontaneous
tures suggests that the 1,2-DAG formed by the spontaneous acyl migration. For capric acid at 35°C, the content of triacyl-
acyl migration of 1,3-DAG were esterified at a higher rate to glycerol was 6.62% in the reaction mixture at 12 h at approx-
TAG and thus could not accumulate in the reaction mixture.
imately 98% consumption of the acid. The initial reaction rate
The effect of molar ratio of caprylic acid to glycerol on was one-third of that for caprylic acid at 25°C, and the same
glyceride composition is shown in Figure 5. The optimum maximal content of 1,3-DAG was obtained, but 4 h later. Ap-
was found at 4:1 molar ratio, for which 1,3-DAG content was proximately the same content of triacylglycerol (6.68%) was
maximal (98%) and 1-MAG and TAG contents were minimal formed in the case of lauric acid at 45°C, but only after 3 h at
(1% each). 1,2-DAG content was under the detection limit of 90% consumption of the acid.
the TLC–FID method. The relatively low boiling point of
From the data shown in Figure 6 for caprylic acid, it is evi-
caprylic acid (240°C at normal pressure) makes possible its dent that the reaction rate for a 4% Lipozyme content in the re-
removal from the product by evaporation under reduced pres- action mixture was much lower than the rates obtained for
sure at lower temperatures (molecular distillation). The boil- higher contents. It can be expected that for the other acid
ing point of caprylic acid at 10 mm Hg is 124°C, but at a pres- species also a higher content of enzyme in the reaction mixture
sure under 1 mm Hg, its boiling temperature should be much would make the reaction faster with higher yields in shorter
lower, thus avoiding the acyl migration of 1,3- to 1,2-dicapryl- time because the acyl migration would be avoided partially.
TABLE 1
Syntheses of Some 1,3-Diacylglycerols
Composition of reaction
Temperature
(°C)
Time
(h)
Initial reaction ratea
mixtureb (%)
1,3-DAG
Fatty acid species
(mmol/min mg × 10−4
)
TAG
FA
1,2-DAG
1-MAG
Butyric acid (4:0)
25
25
25
35
45
25
25
25
12
12
8
12
3
12
12
12
0
0
0
0
0
0
0
0
0
0
0
0
Caproic acid (6:0)
Caprylic acid (8:0)
Capric acid (10:0)
Lauric acid (12:0)
Oleic acid (18:1)
9.53
3.45
8.66
0.41
1.03
8.00
3.60
6.62
6.68
1.34
2.29
7.83
2.88
2.03
10.15
20.63
12.19
7.86
84.63
84.37
67.35
61.10
74.25
71.74
0.45
0.83
1.75
0.85
0.91
2.52
8.44
6.15
14.07
16.08
9.65
10.05
Linoleic acid (18:2)
Eicosapentaenoic acid (20:5)
aInitial reaction rate was defined as mmol fatty acid consumed per minute per mg Lipozyme (Novo Nordisk A/S, Bagsvaerd, Denmark).
bComposition of the reaction mixture at the time when the contents of 1,3-diacylglycerol were maximum. For the reaction conditions see the Experimental
Procedures section. TAG, triacylglycerols; FA, fatty acids; 1,3- and 1,2-DAG, 1,3- and 1,2-diacylglycerols; 1-MAG, 1-monoacylglycerol.
JAOCS, Vol. 76, no. 7 (1999)

ENZYMATIC SYNTHESIS OF SYMMETRICAL 1,3-DIGLYCERIDE
843
2. Haftendorn, R., and R. Ulbrich-Hofmann, Synthesis of 2-Modi-
fied 1,3-Diacylglycerols, Tetrahedron 51:1177–1186 (1995).
3. El Kihel, L., J. Bourass, P. Richomme, J.Y. Petit, and Y. Le-
tourneux, Synthesis and Evaluation of the Anti-inflammatory
Effects of Niflumic Acid Lipophilic Prodrugs in Brain Edema,
Drug Res. 46:1040–1044 (1996).
4. Mantelli, S., P. Speiser, and H. Hauser, Phase Behavior of
Diglyceride Prodrugs: Spontaneous Formation of Unilamellar
Vesicles, Chem. Phys. Lipids 37:329–343 (1985).
5. Scriba, G.K.E., Phenytoin-Lipid Conjugates as Potential Pro-
drugs of Phenytoin, Arch. Pharm. 326:477–481 (1993).
6. Soberg Christensen, M., C.E. Høy, C.C. Becker, and T.G. Red-
grave, Intestinal Absorption and Lymphatic Transport of Eicosa-
pentaenoic (EPA), Docosahexaenoic (DHA), and Decanoic
Acids: Dependence on Intramolecular Triacylglycerol Structure,
Am. J. Clin. Nutr. 61:56–61 (1995).
7. Ikeda, I., X.-P. Gu, I. Miyamoto, and M. Okahara, Preparation
of 1,3-Diacylglycerols and 1-Alkyl-3-acylglycerols in the Pres-
ence of Quaternary Ammonium Salt, J. Am. Oil Chem. Soc. 66:
822–824 (1989).
8. Benbouzid, M., A. Bhati, and J.R. Hamilton, Synthesis of Sym-
metrical Diacid Triacylglycerols from Glycerol-1,3-ditosylate,
Fat Sci. Technol. 90:292–295 (1988).
9. Jensen, R.G., and R.E. Pitas, Synthesis of Some Acylglycerols
and Phosphoglycerides, in Advances in Lipid Research, Vol. 14,
edited by R. Paoletty and D. Kritchevsky, Academic Press, New
York, 1976, p. 228.
10. Yamane, T., S.T. Kang, K. Kawahara, and Y. Koizumi, High-
Yield Diacylglycerol Formation by Solid-Phase Enzymatic
Glycerolysis of Hydrogenated Beef Tallow, J. Am. Oil Chem.
Soc. 71:339–342 (1994).
11. Berger, M., K. Laumen, and M.P. Schneider, Enzymatic Esteri-
fication of Glycerol I. Lipase-Catalyzed Synthesis of Regioiso-
merically Pure 1,3-sn-Diacylglycerols, Ibid. 69:955–960 (1992).
12. Waldinger, C., and M.P. Schneider, Enzymatic Esterification of
Glycerol III. Lipase-Catalyzed Synthesis of Regioisomerically
Pure 1,3-sn-Diacylglycerols and 1(3)-rac-Monoacylglycerols
Derived from Unsaturated Fatty Acids, Ibid. 73:1513–1519
(1996).
13. Haraldsson, G.G., B.O. Gudmundsson, and O. Almarsson, The
Synthesis of Homogeneous Triglycerides of Eicosapentaenoic
Acid and Docosahexaenoic Acid by Lipase, Tetrahedron 51:
941–952 (1995).
14. Tatara, T., T. Fujii, T. Kawase, and M. Minagawa, Determina-
tion of Tri-, Di-, Monooleins, and Free Oleic Acid by the Thin-
Layer Chromatography–Flame Ionization Detector System
Using Internal Standards and Boric Acid Impregnated Chromar-
ods, Lipids 18:732–736 (1983).
15. Baba, N., N. Hirota, H. Umino, K. Matsuo, S. Nakajima, S.
Tahara, T. Kaneko, and M. Matsuo, Synthesis of Triacylglyc-
eride Hydroperoxides Derived from Linoleic Acid, Biosci.
Biotechnol. Biochem. 58:1547–1548 (1994).
FIG. 6. Effect of Lipozyme content on the initial rate of esterification of
glycerol with caprylic acid. The initial reaction rate was defined by the
conversion of caprylic acid at 15 min. For reaction conditions see Fig-
ure 2. For manufacturer see Figure 1.
The yields of 1,3-DAG synthesis were comparable to those
of the other enzymatic methods (10–12). Higher yields can
be expected if the method is optimized for each FA employed.
This direct method has the advantage of starting from the
cheapest reactants without employing organic solvents as re-
action media.
Chemical synthesis of 1,3-dicapryloyl-2-eicosapen-
taenoylglycerol. Purified 1,3-dicapryloylglycerol (95%) was
used for the chemical synthesis of the structured glyceride
with an eicosapentaenoyl group in position 2. The yield of
1,3-dicapryloyl-2-eicosapentaenoylglycerol after purification
was 42%. The purity of the product was checked by GC and
HPLC analysis. GC analysis could differentiate triacylglyc-
erols according to their molecular weights only, giving 98%
purity for the product, but two positional isomers, 1,3-di-
capryloyl-2-eicosapentaenoylglycerol and 1,2-dicapryloyl-3-
eicosapentaenoylglycerol, were included. Silver ion HPLC
analysis was used to determine the ratio of the two isomers.
The ratio of 1,3-dicapryloyl-2-eicosapentaenoylglycerol to
1,2-dicapryloyl-3-eicosapentaenoylglycerol (and/or 1-eicosa-
pentaenoyl-2,3-dicapryloylglycerol) was 9:1.
16. Srisiri, W., H.G. Lamparski, and D.F. O’Brien, Syntheses of
Polymerizable Monoacylglycerols and 1,2-Diacyl-sn-glycerols,
J. Org. Chem. 61:5911–5915 (1996).
17. Han, J.J., Y. Iwasaki, and T. Yamane, Monitoring of Lipase-
Catalyzed Transesterification Between Eicosapentaenoic Acid
Ethyl Ester and Tricaprylin by Silver Ion High-Performance-
Liquid Chromatography and High-Temperature Gas Chroma-
tography, J. Am. Oil Chem. Soc. 76:31–39 (1999).
ACKNOWLEDGMENTS
Novo Nordisk A/S is acknowledged for the generous gift of the en-
zyme for this study. Roxana Rosu is a recipient of a scholarship for
foreign students from the Ministry of Culture and Education of Japan.
REFERENCES
18. Miller, C., H. Austin, L. Posorske, and J. Gonzlez, Characteris-
tics of an Immobilized Lipase for the Commercial Synthesis of
Esters, J. Am. Oil Chem. Soc. 65:927–931 (1988).
1. Berger, M., and M.P. Schneider, Regioisomerically Pure Mono
and Diacylglycerols as Synthetic Building Blocks, Fat Sci.
Technol. 95:169–175 (1993).
[Received November 23, 1998; accepted March 2, 1999]
JAOCS, Vol. 76, no. 7 (1999)