Chemoenzymatic synthesis of enantiopure structured triacylglycerols

Article abstract of DOI:10.1055/s-2008-1077981

A highly efficient chemoenzymatic method for the synthesis of enantiopure ABC-type asymmetrically structured triacylglycerols has been developed starting from enantiopure (S)-solketal and involving two lipase steps. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2008-1077981

2178
LETTER
Chemoenzymatic Synthesis of Enantiopure Structured Triacylglycerols
Synthesis of
E
nan
j
tiopur
öStructured
r
nlycerols Kristinsson, Gudmundur G. Haraldsson*
Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavik, Iceland
E-mail: gghar@raunvis.hi.is
Received 2 June 2008
(CAL)7 was observed to display excellent regioselectivity
toward the end-positions of glycerol using vinyl esters as
acylating agents. This is based on a rapid, irreversible
transesterification of glycerol using 1.25-fold stoichio-
metric amount of the vinyl esters of MCFA in dichlo-
romethane at 0–4 °C. The lipase acted exclusively on the
glycerol end-positions and excellent yields (≥ 90%) were
obtained as based on chemically and regioisomerically
pure material after recrystallization. It took the reaction
only 3–5 hours to proceed to completion resulting in
quantitative conversion into the desired 1,3-DAG with
only traces of 1-monoacylglycerol (1-MAG) intermediate
present throughout the reaction. The n-3 PUFA were sub-
sequently introduced into the remaining mid-position of
the symmetric 1,3-diacylglycerol key adducts, highly ef-
ficiently using 1-(3-dimethylaminopropyl)-3-ethylcarbo-
diimide hydrochloride (EDAC) as a chemical coupling
agent.2
Abstract: A highly efficient chemoenzymatic method for the syn-
thesis of enantiopure ABC-type asymmetrically structured triacyl-
glycerols has been developed starting from enantiopure (S)-solketal
and involving two lipase steps.
Key words: asymmetrically structured triacylglycerols, regioselec-
tivity, lipase, eicosapentaenoic acid, acyl migration
Structured triacylglycerols (TAG) constituting saturated,
medium-chain fatty acids (MCFA) at the terminal (sn-1/
sn-3) positions and long-chain, biologically active poly-
unsaturated fatty acids (PUFA) at the mid-position (sn-2)
of the glycerol backbone have gained increased attention
of scientists as dietary and health supplements.1 Recently,
a highly efficient synthesis of such structured MLM (me-
dium-long-medium) type TAG possessing pure eicosa-
pentenoic acid (EPA) or docosahexenoic acid (DHA) at
the mid-position has been described by a two-step
chemoenzymatic process.2 Numerous beneficial effects
on human health have been firmly established for the
long-chain n-3 PUFA that are characteristic of marine
fat.3 These effects are almost exclusively attributed to
EPA and DHA, the two most prevalent n-3 PUFA in ma-
rine fat.4 Compound 1 (Figure 1) represents a structured
TAG of the type described with capric acid at the end-po-
sitions and DHA at the mid-position.
The current work describes the design of synthesis of a
novel type of structured TAG by a chemoenzymatic ap-
proach starting with optically pure solketal and involving
two lipase steps. This is an enantiopure ABC-type asym-
metrically structured TAG possessing two different types
of pure saturated fatty acids at the sn-1 and sn-3 positions
and a bioactive PUFA such as EPA or DHA at the sn-2 po-
sition of the glycerol backbone. In compound 2 (Figure 2)
the structure of such asymmetrically structured TAG is re-
vealed for EPA located at the sn-2 position, stearic acid at
the sn-1 position, and capric acid at the sn-3 position of
the glycerol moiety. This is the synthetic objective of the
work described in the present report.
Synthesis of such positionally labeled structured TAG by
traditional synthetic organic chemistry methods requires a
full regioselectivity control and can hardly be undertaken
without multistep protection–deprotection processes. Ow-
ing to their regioselectivity, lipases are ideally suited as
biocatalysts for preparing structured TAG.5 By acting
preferably or exclusively at the primary alcoholic posi-
tions of the glycerol backbone they may be employed to
introduce fatty acids of certain type or composition at
these positions by esterification or transesterification pro-
cesses.6 An immobilized Candida antarctica lipase
Such asymmetrically labeled ABC-type structured TAG
may find value in various applications including screen-
ing for biological effects of individual fatty acids, possi-
bly related to their location in stereospecific positions of
the glycerol backbone. They may also become useful as
chiral substrates to investigate lipase enantioselectivity
O
O
C9H19
O
O
O
O
1
C9H19
Figure 1
SYNLETT 2008, No. 14, pp 2178–2182
2
2
.0
8
.2
0
0
8
Advanced online publication: 31.07.2008
DOI: 10.1055/s-2008-1077981; Art ID: D19508ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
Synthesis of Enantiopure Structured Triacylglycerols
2179
O
O
C17H35
O
O
O
O
2
C9H19
Figure 2
towards TAG, as standards, fine chemicals, and drug sup- tempts to prepare asymmetric AAB- and ABC-type struc-
plements. The methodology may also be utilized to intro- tured TAG.11 Although moderate enantioselectivity was
duce isotopically labeled fatty acids into predetermined obtained in these attempts enantiopurities were nowhere
positions of TAG and in design of prodrugs based on the close to what the current aims require. However, excellent
glycerol framework.
enantioselectivity (>99% ee) was obtained in a recently
reported enantioselective ethanolysis of homogeneous
TAG using an immobilized Rhizomucor miehei lipase.12
That lipase displayed strong enantiopreference for the
sn-1 position of the TAG, but the resulting optically pure
sn-2,3-DAG was obtained in only 60% chemical purity in
a mixture with 2-MAG, unreacted TAG and glycerol.
It is pretty obvious that a synthetic route towards enan-
tiopure ABC-type TAG would have to involve an enan-
tiopure 1,3-DAG possessing all required regio- and
stereochemistry as a key intermediate adduct. An ideal
way to bring about synthesis of such enantiopure 1,3-
DAG is by an all-enzymatic two-step process starting
from nonsubstituted prochiral glycerol (Scheme 1).
An acyl group can be easily introduced to the sn-1 posi-
tion of enantiopure (R)-solketal to afford the isopro-
pylidene-protected MAG adduct using CAL and vinyl
esters. However, the sn-1-MAG did not survive mild acid-
ic conditions required for the deprotection of the isopro-
O
O
C17H35
O
C17H35
O
OH
OH
O
OH
OH
OH
OH
O
pylidene moiety. This resulted in
a mixture of
'lipase 1'
'lipase 2'
approximately 90% 1-MAG and 10% 2-MAG, which is
C9H19
close to a reported equilibrium.13
Scheme 1 Hypothetical enzymatic two-step process to prepare
asymmetric 1,3-DAG
A more extensive multistep protection–deprotection ap-
proach was clearly needed for the synthesis of (S)-2. The
proposed synthetic route is based on a six-step chemoen-
zymatic process starting from optically pure (S)-solketal
as is illustrated in Scheme 2.
The first step in such an approach requires a highly regio-
and enantioselective lipase (‘lipase 1’ in Scheme 1) acting
exclusively at the sn-1 position of glycerol to afford an
enantiopure (S)-1-MAG adduct (or its R-enantiomer, de-
pending on the enantiopreference of the lipase). A second
lipase (‘lipase 2’ in Scheme 1) subsequently acts highly
regioselectively or exclusively at the nonaccommodated
sn-3 position to introduce a different fatty acyl moiety at
that position to afford the asymmetric 1,3-DAG. For that
lipase no enantioselectivity is needed.
The first four steps were needed for sorting out the ste-
reochemistry and introducing stearic acid into the glycerol
sn-1 position to afford a regioisomerically and optically
pure 1-MAG (S)-4. This involves a benzyl ether protec-
tion of the free hydroxyl group, deprotection of the isopro-
pylidene moiety, a highly regioselective introduction of
pure stearic acid to the primary hydroxyl group of the re-
sulting 1-O-benzyl-sn-glycerol (R)-3 by CAL, and a cata-
lytic hydrogenolysis of the benzyl protective group. The
remaining part of the synthesis is rather straightforward
by introduction of capric acid exclusively to the vacant
primary sn-3 hydroxyl group by CAL providing the asym-
metric regioisomerically pure 1,3-DAG key adduct (S)-5,
and a subsequent introduction of EPA to its mid-position
by chemical coupling.
Despite its simplicity this route turned out not to be very
realistic after all. Lipases of the second type are certainly
known, one of the most efficient being CAL as has been
demonstrated in our previously described synthesis of
structured MLM-type TAG.2 But, there are apparently no
reports on enantioselective acylations of nonsubstituted
prochiral glycerol involving lipase and fatty acids or their
esters to discriminate between the sn-1 and sn-3 positions,
although there is a report of moderate enantioselectivity
with vinyl benzoate.8 There are, however, numerous re-
ports on asymmetric biotransformations involving lipase
and variously substituted glycerol derivatives by kinetic
resolution.9
Full enantiocontrol was secured by the use of commercial-
ly available enantiopure (S)-solketal as a chiral precur-
sor.14 Regioselective lipase was used to control the
regiochemistry. The regiopurity, and therefore the enan-
tiopurity, was maintained by mild conditions offered by
the lipase acting at mild temperature. The good success is
believed to predominantly relate to the enzyme displaying
a superb regioselectivity and elimination of any acyl-mi-
gration side reaction.2,5 The acyl migration is interrelated
to various important factors including temperature, appar-
Lipases are known to act enantioselectively on TAG mol-
ecules, but usually their enantiopreference for the sn-1 or
sn-3 positions is only moderate or relatively low.10 Based
on such lipase enantioselectivity there are reports on at-
Synlett 2008, No. 14, 2178–2182 © Thieme Stuttgart · New York

2180
B. Kristinsson, G. G. Haraldsson
LETTER
O
O
OH
O
C17H35
O
a
b
c
O
O
O
OH
O
OH
O
OH
Ph
Ph
Ph
(R)-3
O
O
O
O
EPA
C17H35
O
C17H35
O
C17H35
O
e
d
f
OH
OH
OH
O
O
O
O
O
C9H19
C9H19
(S)-4
(S)-5
(S)-2
Scheme 2 Reagents and conditions: (a) NaH, THF, then BnBr; (b) 1 M HCl, H2O–EtOH, reflux 30 min, 87% (2 steps); (c) vinyl stearate,
CAL, CH2Cl2, r.t.; (d) H2, 10% Pd/C, THF–hexane, 85% (2 steps); (e) vinyl capriate, CAL, THF, r.t., 85%; (f) EPA, EDAC, DMAP, CH2Cl2,
r.t., 91%.
ently the most crucial single parameter, reaction rate, sup- The second enzymatic step involved the highly regiose-
port material of enzyme, type of reaction, acyl donor, and lective CAL a second time, this time to acylate exclusive-
reaction conditions.2,13,15,16 The temperature was main- ly at the sn-3 position of the 1-MAG (S)-4 with vinyl
tained sufficiently low to keep the acyl migration and li- capriate in dry THF at room temperature. Only 2–3 hours
pase regioselectivity completely under control with the were needed to complete the reaction. This is in good
lipase still acting fast enough, since the acyl-migration agreement with our previous observation that only traces
process is clearly time-dependent. This was brought about of 1-MAG were present during the two-step process when
by use of activated vinyl esters offering a fast and irrevers- preparing the symmetric 1,3-DAG from glycerol suggest-
ible reaction.2
ing that 1(3)-MAG, once formed, reacts significantly fast-
er with the MCFA vinyl ester than glycerol under the
reaction conditions used.2 It should be kept in mind that
the reaction conditions were different from the previous
case, room temperature instead of 0–4 °C and, for solubil-
ity reasons, THF replacing dichloromethane. The enan-
tiopure asymmetric 1,3-DAG adduct (S)-5 was obtained
in 85% yield after crystallization from methanol.22 The
specific optical-rotation value is extremely low for this
type of compounds, [a]D20 +0.02 (c 10, CH2Cl2), but fur-
ther investigations by varying the solvent are under way.
In the first step sodium hydride was used as a base in THF
to introduce a benzyl protective group to the sn-3 position
of the glycerol moiety by Williamson ether synthesis us-
ing benzyl bromide. The benzylated solketal adduct was
not isolated but directly submitted to deprotection of the
isopropylidene protective moiety using acidic aqueous
ethanol. The resulting 3-O-benzyl-sn-glycerol adduct (R)-
3 was isolated by Kugelrohr distillation in vacuo in 87%
overall yield.17 Both (R)-3 and its optical antipode (S)-3
are commercially available.18 The latter adduct has been
used previously as a chiral precursor in synthesis of asym- The final step involved chemical coupling with EDAC in
metric enantiopure 2,3-DAG.19
the presence of 4-dimethylaminopyridine (DMAP) to in-
troduce pure EPA into the sn-2 position of adduct (S)-5 to
afford the asymmetrically structured ABC-type TAG fi-
nal product (S)-2 in 91% yield after purification by silica
gel chromatography.23 The reaction was conducted in
dichloromethane at room temperature for twelve hours.
Stoichiometric amount of EPA was used, 20% molar ex-
cess of EDAC and 0.4 equivalents of DMAP. As noticed
before2 no acyl migration was observed to take place dur-
ing this reaction. This was further confirmed by detailed
The benzylated glycerol adduct (R)-3 was subsequently
acylated with vinyl stearate using the immobilized CAL at
room temperature in dichloromethane. It took the reaction
only 70 minutes to proceed to completion. The acylated
benzyl ether adduct was not purified but directly intro-
duced to Pd/C-promoted catalytic hydrogenolysis depro-
tection of the benzyl protective group in THF–hexane. As
had been anticipated no acyl migration took place during
this reaction as was firmly established by thorough inves-
1
stability studies of 1,3-DAG using 400 MHz H NMR
1
tigations based on H NMR spectroscopy at 400 MHz.
spectroscopy under the reaction conditions of the cou-
pling reaction in absence of EPA. The specific optical-ro-
tation value of (S)-2 was somewhat an order of magnitude
higher than for its precursor, [a]D20 +0.16 (c 10, CH2Cl2),
but again, further investigations by varying the solvent are
under way.
The absence of acyl migration rules out possibilities that
any losses in enantiopurity were taking place during these
reactions. The overall two-step yield of the enantiopure 1-
MAG (S)-4 was 81% after crystallization from hexane.20
This adduct along with its optical antipode has been pre-
viously prepared from D-mannitol by use of xanthen-9-
ylidene protecting groups.21
The enantiopurity of the compounds involved in this study
has not been determined explicitly, but the fact that no
acyl migration was detected at any stage of the reactions
Synlett 2008, No. 14, 2178–2182 © Thieme Stuttgart · New York

LETTER
Synthesis of Enantiopure Structured Triacylglycerols
2181
(9) (a) Wang, Y.-F.; Wong, C.-H. J. Org. Chem. 1988, 53,
3127. (b) Guanti, G.; Banfi, L.; Basso, A.; Bevilacqua, E.;
Bondanza, L.; Riva, R. Tetrahedron: Asymmetry 2004, 15,
2889. (c) Halldorsson, A.; Thordarson, P.; Kristinsson, B.;
Magnusson, C. D.; Haraldsson, G. G. Tetrahedron:
Asymmetry 2004, 15, 2893. (d) Ransac, S.; Rogalska, E.;
Gargouri, Y.; Deveer, A. M. T. J.; Paltauf, F.; de Haas, G. H.;
Verger, R. J. Biol. Chem. 1990, 265, 20263. (e) Theil, F.;
Lemke, K.; Ballschuh, S.; Kunath, A.; Schick, H.
implies that no depletions in enantiopurity were taking
place during the reactions involved. Loss in enantiopurity
would most certainly have to take place through acyl mi-
gration. The detection limits for such migration products
by 1H NMR spectroscopy (400 MHz) are estimated below
0.25% as based on careful intensive studies, whereas an
equilibrium composition between 1(3)-MAG and 2-MAG
is roughly 10% 2-MAG and 90% 1(3)-MAG.13 The corre-
sponding equilibrium composition for 1,3-DAG and
1(3),2-DAG is roughly 70% 1,3-DAG and 30% 1(3),2-
DAG.16 Therefore, losses in enantiopurity through lack of
regiocontrol are rather unlikely, but this will be confirmed
later by accurate measurements by chiral HPLC.
Tetrahedron: Asymmetry 1995, 6, 1323.
(10) (a) Rogalska, E.; Ransac, S.; Verger, R. J. Biol. Chem. 1990,
265, 20271. (b) Villeneuve, P.; Pina, M.; Montet, D.;
Graille, J. Chem. Phys. Lipids 1995, 76, 109.
(c) Villeneuve, P.; Pina, M.; Graille, J. Chem. Phys. Lipids
1996, 83, 161. (d) Lang, D. A.; Dijkstra, B. W. Chem. Phys.
Lipids 1998, 93, 115. (e) Rogalska, E.; Cudrey, C.; Ferrato,
F.; Verger, R. Chirality 1993, 5, 24. (f) Lang, D. A.;
Mannesse, M. L. M.; de Haas, G. H.; Verheij, H. M.;
Dijkstra, B. W. Eur. J. Biochem. 1993, 254, 333.
(g) Uzawa, H.; Nishida, Y.; Ohrui, H.; Meguro, H. Biochem.
Biophys. Res. Commun. 1990, 168, 506. (h) Uzawa, H.;
Noguchi, T.; Nishida, Y.; Ohrui, H.; Meguro, H. Biochim.
Biophys. Acta 1993, 1168, 253.
Acknowledgment
The University of Iceland Research Fund is acknowledged for fi-
nancial support, Novozyme AS in Denmark for the lipase, and Olav
Thorstad of Pronova Bicocare ASA in Norway for pure EPA. Dr.
Sigridur Jonsdottir and Dr. Sigurdur V. Smarason at University of
Iceland are acknowledged for high-resolution NMR and accurate
MS measurements.
(11) (a) Chandler, I. C.; Quinlan, P. T.; McNeill, G. P. J. Am. Oil
Chem. Soc. 1998, 75, 1513. (b) Iwasaki, Y.; Yamane, T.
J. Mol. Catal. B: Enzym. 2000, 10, 129.
(12) Piyatheerawong, W.; Yamane, T.; Nakano, H.; Iwasaki, Y.
J. Am. Oil Chem. Soc. 2006, 83, 603.
(13) Compton, D. L.; Vermillion, K. E.; Laszlo, J. A. J. Am. Oil
Chem. Soc. 2007, 84, 343.
(14) Sigma-Aldrich (Steinheim, Germany): (S)-(+)-1,2-O-
isopropylidene-sn-glycerol of 98% purity and 99% ee.
(15) (a) Kodali, D. R.; Tercyak, A.; Fahey, D. A.; Small, D. M.
Chem. Phys. Lipids 1990, 52, 163. (b) Bloomer, S.;
Adlercreutz, P.; Mattiasson, B. Biocatalysis 1991, 5, 145.
(c) Fureby, A. M.; Virto, C.; Adlercreutz, P.; Mattiasson, B.
Biocat. Biotrans. 1996, 14, 89.
References and Notes
(1) (a) Miura, S.; Ogawa, A.; Konishi, H. J. Am. Oil Chem. Soc.
1999, 76, 927. (b) Fomuso, L. B.; Akoh, C. C. J. Am. Oil
Chem. Soc. 1998, 75, 405. (c) Christensen, M. S.; Höy,
C.-E.; Becker, C. C.; Redgrave, T. G. Am. J. Clin. Nutr.
1995, 61, 56.
(2) Halldorsson, A.; Magnusson, C. D.; Haraldsson, G. G.
Tetrahedron 2003, 59, 9101.
(3) (a) Haraldsson, G. G.; Hjaltason, B. In Structured and
Modified Lipids; Gunstone, F. D., Ed.; Marcel Decker: New
York, 2001, 313–350. (b) Hjaltason, B.; Haraldsson, G. G.
In Modifying Lipids for Use in Food; Gunstone, F. D., Ed.;
Woodhead Publishing: Cambridge UK, 2006, Chap. 4, 56–
79.
(4) For reviews, see: (a) Stansby, M. E. In Fish Oils in
Nutrition; Stansby, M. E., Ed.; van Nostrand Reinhold: New
York, 1990, Chap. 10, 268–288. (b) Nettleton, J. A.
Omega-3 Fatty Acids and Health; Chapman and Hall: New
York, 1995. (c) Lands, W. E. M. Fish, Omega-3 and Human
Health, 2nd ed.; AOCS Press: Champaign IL, 2005.
(5) Haraldsson, G. G. In Handbook of Industrial Biocatalysis;
Hou, C. T., Ed.; CRC Press, Taylor and Francis Group: Boca
Raton, 2005, Chap. 18, 18-1–18-21.
(6) (a) Berger, M.; Laumen, K.; Schneider, M. P. J. Am. Oil
Chem. Soc. 1992, 69, 955. (b) Aha, B.; Berger, M.; Jakob,
B.; Machmüller, G.; Waldinger, C.; Schneider, M. P. In
Enzymes in Lipid Modification; Bornscheuer, U. T., Ed.;
Wiley-VCH: Weinheim, 2000, 100–115. (c) Irimescu, R.;
Furihata, K.; Hata, K.; Iwasaki, Y.; Yamane, T. J. Am. Oil
Chem. Soc. 2001, 78, 285. (d) Irimescu, R.; Iwasaki, Y.;
Hou, C. T. J. Am. Oil Chem. Soc. 2002, 79, 879.
(e) Kawashima, A.; Shimada, Y.; Yamamoto, M.; Sugihara,
A.; Nagao, T.; Komemushi, S.; Tominaga, Y. J. Am. Oil
Chem. Soc. 2001, 78, 611.
(16) Laszlo, J. A.; Compton, D. L.; Vermillion, K. E. J. Am. Oil
Chem. Soc. 2008, 85, 307.
(17) [a]D20 +5.5 (c 20, CHCl3), identical to that in ref. 18.
(18) Aldrich Chemical Catalog, (R)-(+)-benzyloxy-1,2-
propanediol, 99%: [a]D20 +5.5 (c 20, CHCl3).
(19) Vilcheze, C.; Bittman, R. J. Lipid Res. 1994, 35, 734.
(20) Procedure for the Preparation of 1-O-Octadecanoyl-sn-
glycerol [(S)-4]
To a solution of 3-O-benzyl-sn-glycerol [(R)-3, 1.497 g, 6.73
mmol] and vinyl stearate (3.827 mg, 12.3 mmol) in CH2Cl2
(8.2 mL) was added immobilized CAL (99 mg). The
resulting suspension was stirred at r.t. for approx. 70 min
when TLC monitoring (EtOAc–PE, 1:1) indicated a
complete reaction. The lipase preparation was separated by
filtration and the solvent removed in vacuo on a rotary
evaporator. The resulting residue was dissolved in THF (30
mL) without further purification followed by addition of n-
hexane (70 mL) and Pd/C catalyst (370 mg). The reaction
was performed in a PARR reactor under hydrogen pressure
(5 bar) during which the product precipitated. When the
reaction had proceeded to completion (about 2 h), THF was
added until all the product had been dissolved. The catalyst
was separated off by filtration by the aid of Celite and the
solvent removed in vacuo on a rotary evaporator. The
residue was redissolved in minimum amount of THF and a
fourfold volume of n-hexane added. The resulting mixture
was allowed to stand at r.t. overnight to afford the product as
white crystals (2.057 g, 5.73 mmol) in 85% overall yield; mp
59.5–60.7 °C. [a]D20 +2.43 (c 6, THF). 1H NMR (400 MHz,
CDCl3): d = 4.21 (dd, J = 11.6, 4.8 Hz, 1 H, OCOCH2), 4.15
(7) Novozym 435, CAL-B; a gift from Novozyme A/S
(Bagsvaerd, Denmark).
(8) Kato, Y.; Fujiwara, I.; Asano, Y. J. Mol. Catal. B: Enzym.
2000, 9, 193.
Synlett 2008, No. 14, 2178–2182 © Thieme Stuttgart · New York

2182
B. Kristinsson, G. G. Haraldsson
LETTER
C), 29.1 (2 C), 24.9 (2 C), 22.7 (2 C), 14.1 (2 C). FT-IR:
3300–3600 (br, OH), 2914 (vs, CH), 2849 (vs, CH), 1732
(vs, C=O), 1708 (vs, C=O) cm–1. HRMS (APCI): m/z calcd
for C31H60O5 – OH 495.4408; found: 495.4412 amu.
(23) Procedure for 3-O-Decanoyl-2-eicosapentaenoyl-1-
octadecanoyl-sn-glycerol [(S)-2]
(dd, J = 11.6, 6.0 Hz, 1 H, OCOCH2), 3.96–3.90 (m, 1 H,
CH2CHCH2), 3.72–3.67 (m, 1 H, CH2OH), 3.63–3.57 (m, 1
H, CH2OH), 2.52 (d, J = 5.2 Hz, 1 H, CHOH), 2.35 (t,
J = 7.6 Hz, 2 H, CH2COO), 2.08 (t, J = 6.0 Hz, 1 H,
CH2OH), 1.65–1.59 (m, 2 H, CH2CH2COO), 1.36–1.18 (m,
28 H, CH2), 0.88 (t, J = 6.5 Hz, 3 H, CH2CH3). 13C NMR
(100 MHz, CDCl3): d = 174.3, 70.3, 65.2, 63.3, 34.1, 31.9,
29.7 (5 C), 29.6 (2 C), 29.4 (2 C), 29.3, 29.2, 29.1, 24.9, 22.7,
14.1. FT-IR: 3150–3600 (br, OH), 2918 (vs, CH), 2849 (vs,
CH), 1735 (vs, C=O) cm–1. HRMS (APCI): m/z calcd for
C21H42O4 – OH: 341.3050; found: 341.3042 amu.
To a solution of 3-O-decanoyl-1-octadecanoyl-sn-glycerol
[(S)-5, 100 mg, 0.195 mmol] and EPA as free acid (66.0 mg,
0.218 mmol) in CH2Cl2 (2 mL) were added DMAP (20 mg,
0.16 mmol) and EDAC (50 mg, 0.26 mmol). The resulting
solution was stirred on a magnetic stirrer hotplate at r.t. for
12 h. The reaction was disconnected by passing the reaction
mixture through a short column packed with silica gel by use
of Et2O–CH2Cl2 (10:90). Solvent removal in vacuo afforded
the pure product as a yellowish oil (141 mg, 0.177 mmol,
91% yield). [a]D20 +0.16 (c 10, CH2Cl2). 1H NMR (400
MHz, CDCl3): d = 5.42–5.33 (m, 10 H, =CH), 5.32–5.23 (m,
1 H, CH2CHCH2), 4.29 (dd, J = 12.0, 4.4 Hz, 2 H,
CH2CHCH2), 4.14 (dd, J = 12.0, 6.0 Hz, 2 H, CH2CHCH2),
2.86–2.78 (m, 8 H, =CCH2C=), 2.34 (t, J = 7.6 Hz, 2 H,
CH2COO in EPA), 2.31 (t, J = 7.6 Hz, 4 H, CH2COO), 2.14–
2.04 (m, 4 H, =CCH2CH2 and =CCH2CH3), 1.70 (quint,
J = 7.6 Hz, 2 H, CH2CH2COO in EPA), 1.64–1.55 (m, 4 H,
CH2CH2COO), 1.34–1.20 (m, 40 H, CH2), 0.97 (t, J = 7.6
Hz, 3 H, CH3 in EPA), 0.88 (t, J = 6.8 Hz, 6 H, CH3 in C18).
13C NMR (100 MHz, CDCl3): d = 173.3 (2 C), 172.6, 132.0,
128.9, 128.8, 128.6, 128.3, 128.2, 128.1, 128.0, 127.8,
127.0, 69.0, 62.1 (2 C), 34.0 (2 C), 33.6, 31.9, 31.8, 29.7 (5
C), 29.6 (2 C), 29.5, 29.4 (2 C), 29.3 (2 C), 29.2 (3 C), 29.1,
26.5, 25.6 (3 C), 25.5, 24.8 (2 C), 24.7, 22.7, 22.6, 20.5, 14.3,
14.1 (2 C). FT-IR: 3013 (s, CH), 2925 (vs, CH), 2854 (vs,
CH), 1745 (vs, C=O) cm–1. HRMS (APCI): m/z calcd for
C51H88O6 + NH4: 814.6919; found: 814.6913 amu.
(21) Reese, C. B.; Yan, H. J. Chem. Soc., Perkin Trans. 1 2001,
1807.
(22) Procedure for 3-O-Decanoyl-1-octadecanoyl-sn-glycerol
[(S)-5]
Immobilized Candida antarctica lipase (45 mg) was added
to a 10 mL round-bottom flask containing a mixture of 1-
octadecanoyl-sn-glycerol [(S)-4, 203 mg, 0.566 mmol] and
vinyl decanoate (168 mg, 0.849 mmol) dissolved in THF (2
mL). The resulting suspension was stirred at r.t. for 2–3 h or
until TLC monitoring (EtOAc–PE, 1:1) indicated a complete
reaction. The lipase preparation was removed by filtration
and the solvent removed in vacuo on a rotary evaporator to
afford pure 3-O-decanoyl-1-octadecanoyl-sn-glycerol [(S)-
5, 246 mg, 0.480 mmol] as a white crystalline material in
85% yield after recrystallization from MeOH; mp 53.4–
54.0 °C; [a]D20 +0.02 (c 10, CH2Cl2). 1H NMR (400 MHz,
CDCl3): d = 4.20–4.06 (m, 5 H, CH2CHCH2), 2.47 (d,
J = 3.6 Hz, 1 H, OH), 2.34 (t, J = 7.6 Hz, 4 H, CH2COO),
1.68–1.59 (m, 4 H, CH2CH2COO), 1.36–1.18 (m, 40 H,
CH2), 0.87 (t, J = 6.8 Hz, 6 H, CH3). 13C NMR (100 MHz,
CDCl3): d = 173.9 (2 C), 68.4, 65.1 (2 C), 34.1 (2 C), 32.0,
31.9, 29.7 (5 C), 29.6 (2 C), 29.5 (2 C), 29.4 (3 C), 29.3 (2
Synlett 2008, No. 14, 2178–2182 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.