Synthesis of enantiopure ABC-type triacylglycerols

Article abstract of DOI:10.1016/j.tet.2019.130813

The synthesis of twelve enantiopure structured triacylglycerols (TAGs) of the ABC type possessing three different fatty acids is described by a six-step chemoenzymatic approach starting from (S)-solketal. Eight of the TAGs possess two different saturated fatty acyl groups located in the sn-1 and sn-2 positions with an unsaturated fatty acyl group in the remaining sn-3 position of the glycerol skeleton, whereas the remaining four possess two different saturated acyl groups in the terminal sn-1 and sn-3 positions with an unsaturated acyl group in the sn-2 position. The former group was synthesised by a six-step chemoenzymatic route involving a highly regioselective immobilised Candida antarctica lipase. The second group was prepared by a similar six-step approach, that required two separate lipase steps. Such enantiopure TAGs are strongly demanded as standards for enantiospecific analysis of intact TAGs in fats and oils.

Full text of DOI:10.1016/j.tet.2019.130813

Journal Pre-proof
Synthesis of enantiopure ABC-type triacylglycerols
Haraldur G. Gudmundsson, Kaisa M. Linderborg, Heikki Kallio, Baoru Yang,
Gudmundur G. Haraldsson
PII:
S0040-4020(19)31220-7
https://doi.org/10.1016/j.tet.2019.130813
TET 130813
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 31 August 2019
Revised Date: 12 November 2019
Accepted Date: 15 November 2019
Please cite this article as: Gudmundsson HG, Linderborg KM, Kallio H, Yang B, Haraldsson GG,
Synthesis of enantiopure ABC-type triacylglycerols, Tetrahedron (2019), doi: https://doi.org/10.1016/
j.tet.2019.130813.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier Ltd.

Graphical Abstract
Synthesis of enantiopure ABC-type triacylglycerols
Haraldur G. Gudmundsson, Kaisa M. Linderborg, Heikki Kallio, Baoru Yang and
Gudmundur G. Haraldsson*

Synthesis of enantiopure ABC-type triacylglycerols
a,b
b
b
b
Haraldur G. Gudmundsson , Kaisa M. Linderborg , Heikki Kallio , Baoru Yang and Gudmundur G.
a
Haraldsson *
a
Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavik, Iceland
b
Food Chemistry and Food Development, Department of Biochemistry, University of Turku, FI-20014 Turku,
Finland
*
Corresponding author:
Gudmundur G. Haraldsson
Science Institute, University of Iceland
Dunhaga 3, 107 Reykjavik, Iceland
Tel: +354 525 4818
Fax: +354 552 8911
E-mail: gghar@hi.is
Keywords
ABC type structured triacylclycerols; Candida antarctica lipase; chemoenzymatic synthesis;
enantiopure triacylglycerols; enantiospecific TAG analysis.
1

This paper is dedicated to Professor Steve Davies in recognition of his outstanding contribution to
the area of asymmetric and stereoselective synthesis and the Founder and Editor in Chief for
Tetrahedron Asymmetry.
2

Abstract
The synthesis of twelve enantiopure structured triacylglycerols (TAGs) of the ABC type possessing
three different fatty acids is described by a six-step chemoenzymatic approach starting from (S)-
solketal. Eight of the TAGs possess two different saturated fatty acyl groups located in the sn-1 and
sn-2 positions with an unsaturated fatty acyl group in the remaining sn-3 position of the glycerol
skeleton, whereas the remaining four possess two different saturated acyl groups in the terminal sn-
1
and sn-3 positions with an unsaturated acyl group in the sn-2 position. The former group was
synthesized by a six-step chemoenzymatic route involving a highly regioselective immobilized
Candida antarctica lipase. The second group was prepared by a similar six-step approach, that
required two separate lipase steps. Such enantiopure TAGs are strongly demanded as standards for
enantiospecific analysis of intact TAGs in fats and oils.
3

1
. Introduction
Triacylglycerols (TAGs) are by far the largest class of nonpolar lipids that occur widely in fats and
1
-3
oils of plant and animal origin. They consist of a glycerol skeleton to which three fatty acyl
groups are attached as carboxylate esters. In fats and oils the variety and number of TAG molecular
species is immense as a result of the great number of different fatty acids that differ in length,
saturation and location of the carbon-carbon double bonds in unsaturated and polyunsaturated fatty
acids (PUFAs). The location of the fatty acids within the glycerol backbone adds further to the
diversity in terms of regio- and stereoisomerism.
In the human and animal diet TAGs are of importance, not only as a source of energy but also for
supplying bioactive fatty acids for various biological roles. That includes the n-6 and n-3 PUFAs
4
-6
serving as precursors to various potent mediators including eicosanoids and docosanoids.
The
TAGs in animals and plants are known to differ significantly from species to species in terms of
their fatty acid compositions as well as their non-random distribution within the glycerol backbone.
Classical examples of such positionally structured TAGs and their major role in physical properties
7
and biological function, include cocoa butter used in chocolate manufacturing and human milk
8
TAGs. In fish oil TAGs the n-3 PUFAs show a clear preference for the mid-position, whereas in
9
marine mammals this is the other way around.
The glycerol molecule is prochiral, meaning that when two different fatty acids are attached to its
terminal 1,3-positions, a chiral centre is generated at the 2-position. The consequence of that is an
increased number of isomers due to the presence of enantiomers, as pairs or with one of them
predominating. A stereospecific numbering designated by the prefix sn- is used to distinguish
3
between the two enantiotopic terminal positions of the glycerol backbone. The pro-S
4

hydroxymethyl group refers to the sn-1 position and the pro-R group to the sn-3 position with the
remaining stereogenic carbon at the mid-position referred to as the sn-2 position. There are
1
0-12
numerous reports on the enantiospecific location of fatty acids in animals and plants.
A proper chemical analysis of TAGs is quite a challenge. Their total fatty acid composition is
readily available by routine GC analysis after converting their fatty acids into volatile methyl
1
3
esters, but all information with regard to their positional distribution within the glycerol backbone
is lost. The composition of the most abundant or selected abundant regioisomers is available, for
1
4, 15
example from atmospheric pressure chemical ionization mass spectrometry (APCI MS)
or
1
6,17
ammonia negative ion tandem mass spectrometry (NICI-MS/MS).
However, these methods do
not distinguish between the enantiomeric sn-1 and sn-3 positions. That type of information is
available by enantiospecific analyses, where multi-step procedures are combined with
1
8
19
chromatography. That includes 1,3-regioselective lipase or Grignard reagents to form a mixture
of diacylglycerols (DAGs) and monoacylglycerols (MAGs) undergoing further treatments
1
0,20
21
22,23
involving enantioselective phospholipase A2,
chiral derivatization or chiral-phase HPLC.
The reliability of the stereospecific analysis is affected by the tedious multi-step procedures and
2
4,25
acyl migration
side reaction where acyl groups migrate from one position to another within the
glycerol backbone.
The limitation of the stereospecific analyses described above is that they provide an insight into the
overall fatty acid composition of the individual stereospecific sn-1/2/3 positions for all TAG
molcular species present in a mixture rather than information on individual TAG enantiomers.
Obviously, stereospecific analysis of intact TAGs in a natural oil mixture is far more of a challenge.
Nonaqueous reversed phase and silver-ion HPLC/MS are the most commonly used analytical
1
4,26
techniques for separation and characterization of natural TAG molecular species,
but these
5

methods do not allow separation of TAG enantiomers. Consequently, there is an urgent need for a
non-destructive enantiospecific analysis of such intact TAGs. That important challenge has been
addressed in a few recent reports based on chiral HPLC in attempts to separate TAG
1
1,12,27,28
enantiomers.
Despite the variation in methods of chromatography and ionization, current MS- and MS/MS-based
methods for resolving the regioisomers of intact TAGs in natural oils and fats rely on the difference
in the energy required for dissociation of fatty acids between the sn-2 and sn-1/3 positions. For a
specific TAG, the fragmentation pattern depends on both the fatty acid combination and positional
distribution. Therefore, it is of crucial importance to have access to regiopure TAGs as reference
2
9,30
compounds in order to quantify accurately the regioisomer composition of TAGs.
For
sterespecific separation of enantiomers, chiral chirochromatographic analysis has proven to be a
promising method, where enantiopure TAG compounds are needed for identification of separated
2
8
enantiomers.
Natural oils and fats are complex mixtures of a large number of molecular species of TAGs, of
which human milk fat and fish oil are excellent examples with significant nutritional
2
9,31
importance.
Separation, identification, and quantification of the enantiomers would be of
3
2
crucial importance, for example in studies on gene – lipid composition interactions, as well as in
studies on composition of human milk and subsequent development of human milk formulas and
2
9,33
dietary supplements.
A major obstacle hindering the development of the methodologies is the
lack of appropriate TAG enantiomers to serve as standards for establishing calibration curves to
quantify regioisomers and for distinguishing between the intact TAG enantiomers in such
1
1,12,28
analyses.
6

In a recent report the preparation of nine enantiopure (S)-TAG enantiomers was described by
chemical and chemoenzymatic synthesis starting from enantiomerically pure (R)-solketal by use of
3
4
an immobilized Candida antarctica lipase (CAL-B) to secure the regiocontrol. They were all of
the AAB type possessing two different fatty acids, one saturated and one unsaturated, and
representing TAG species commonly present in plant oils and fats. They were intended and used as
standards for enantiospecific separation and recognition of individual enantiomers by a recently
2
8
developed double chiral column sample recycling HPLC system. In the current report the
synthesis of twelve enantiopure ABC type TAGs possessing three different fatty acids, two
saturated and one unsaturated, representing TAG molecular species commonly found in plant and
animal fats, such as human milk, is addressed by a similar chemoenzymatic approach. The
synthesised compounds are of high value as reference compounds for further development of regio-
and stereospecific analysis of TAGs. Their structures are illustrated in Figure 1.
7

Figure 1. The structures of the twelve enantiostructured TAG products synthesised.
2
. Results and Discussion
All the enantiopure structured TAGs addressed in the current study are of the ABC type, thus
constituting three different fatty acids. Two of them are saturated, with the third unsaturated, which
enables their detection by a UV detector in combination with the intended chiral HPLC analysis.
3
5,36
As previously described, the term structured TAG
refers to selected fatty acids accommodating
predetermined positions of the TAG glycerol skeleton. Accordingly, enantiostructured TAG refers
to the fatty acids being located in the enantiospecific positions of the glycerol backbone. Four
different saturated fatty acids were used in these studies, including hexadecanoic acid (C16:0;
palmitic acid), tetradecanoic acid (C14:0; myristic acid), dodecanoic acid (C12:0; lauric acid) and
8

decanoic acid (C10:0; capric acid), and two unsaturated fatty acids, the monounsaturated cis-
octadec-9-enoic acid (C18:1; oleic acid) and the diunsaturated cis-octadec-9,12-dienoic acid
(C18:2; linoleic acid).
From a synthetic organic chemistry point of view the TAGs may be divided into two categories: (a)
Those possessing the two different saturated fatty acyl groups located in the sn-1 and sn-2
positions, with the unsaturated fatty acyl group in the remaining sn-3 position of the glycerol
backbone, thus all being of the R-configuration; (b) Those possessing the two different saturated
acyl groups in the terminal sn-1 and sn-3 positions, with the unsaturated acyl group in the sn-2
position, all of the S-configuration. The preparation of the TAGs belonging to each of the two
categories requires two different synthetic approaches. A total of twelve such enantiostructured
TAGs were prepared, eight belonging to the first category and four belonging to the second one.
The structures of these TAGs are illustrated in Figure 1 with compounds (R)-1, 2, 4, 5, 7, 8, 10 and
1
1 belonging to the first category and (S)-3, 6, 9 and 12 belonging to the second one.
2
9,33
In fact this is a set of four TAG fatty acid compositions commonly found in human milk.
The
first is comprised of palmitic, myristic and oleic acids. We were interested in both combinations of
TAGs belonging to the first category (TAGs (R)-1 and 2), and one belonging to the second (TAG
(S)-3). Since only one of the two enantiomers is needed as a standard to determine the separation
order of the two TAG enantiomers in their enantiospecific TAG analysis, it was of no relevance
which one was prepared, provided that the absolute configuration was known. Therefore, to make
the synthetic work simpler, and to reduce the number of intermediates involved, it was decided to
base the syntheses on the use of (S)-solketal as the sole chiral precursor rather than using both the
(R)- and (S)-solketals. Therefore, we ended up with R-configuration of the first category TAGs and
the S-configuration of those belonging to the second category. This is simply determined by the
9

Cahn-Ingold-Prelog priority rules rather than being of any scientific necessity. Three more TAG
fatty acid compositions were then included in the current task, namely those constituting palmitic,
lauric and linoleic acids (TAGs (R)-4, 5 and (S)-6), palmitic, capric and oleic acids (TAGs (R)-7, 8
and (S)-9), and myristic, lauric and oleic acids (TAGs (R)-10, 11 and (S)-12).
2
.1 Synthesis of the first category TAGs
The synthetic route to the TAGs (R)-1, 2, 4, 5, 7, 8, 10 and 11 belonging to the first category is
illustrated in Scheme 1 for (R)-1. It was based on a six-step chemoenzymatic process starting from
enantiomerically pure 1,2-isopropylidene-sn-glycerol, (S)-solketal, as a chiral precursor. In the first
three steps a previously reported approach towards synthesis of (R)-13 obtained from (R)-solketal
3
4
was followed. In the current work (S)-solketal was benzylated at the sn-3 position and the
resulting benzylated adduct subsequently deprotected without purification to remove the
isopropylidene moiety (see Scheme 1). The enantiomerically pure key intermediate 3-O-benzyl-sn-
glycerol was obtained in 92% overall yield after purification.
Scheme 1. Chemoenzymatic synthesis of the first category of TAGs (R)-1, 2, 4, 5, 7, 8, 10 and 11
(shown for (R)-1). Reagents and conditions: (a) NaH, THF, then BnBr; (b) 1 M HCl, H
2
O/EtOH,
reflux 30 min. (92% overall); (c) Vinyl hexadecanoate, CAL, CH Cl , r.t. (87%); (d) Tetradecanoic
2
2
acid, EDCI, DMAP, CH
2
Cl
2
, r.t. (95%); (e) H
2
, 10% Pd/C, THF, r.t. (71%); (f) cis-Hexadec-9-
enoic acid, EDCI, DMAP, CH Cl , r.t. (93%).
2
2
1
0

The benzylated glycerol adduct was subsequently acylated with palmitic, myristic, lauric and capric
acids, all activated as vinyl esters, by use of the highly regioselective immobilised Candida
antarctica lipase (CAL-B from Novozymes) in dichloromethane at room temperature. Under these
3
5-37
conditions, and in accordance with numerous previous reports,
the enzymatic reaction took
place exclusively at the terminal sn-1 position with no occurrence of the acyl migration side-
reaction. It took the reaction only 90 minutes to proceed to completion. The resulting acylated
benzyl ether adducts (S)-13, 14, 15 and 16 were all prepared in very high to excellent yields (see
1
0
Table 1) after purification by column chromatography using boric acid impregnated silica gel to
prevent acyl migration.
The second saturated fatty acyl group was introduced to the free hydroxyl groups at the sn-2
position of (S)-13-16 using 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI)
as a chemical coupling agent in the presence of dimethylaminopyridine (DMAP) following a
3
5,36
previously described procedure.
The resulting diacylated adducts (S)-17-24 were all obtained in
very high to excellent yields (82-96%) after crystallisation from hot ethanol. They were all
subjected to catalytic hydrogenolysis using Pd/C catalyst in THF under atmospheric pressure at
room temperature to remove the benzyl protective group under which no acyl migration was
observed to take place. The resulting 1,2-DAGs (S)-25-32 were all obtained in moderate to good
yields after crystallisation from hot hexane, except (S)-30 that was obtained as a brownish wax and
used as such without further purification. Table 1 shows the yields obtained for these intermediate
adducts along with their specific optical rotation values. These intermediates were all obtained in
1
excellent chemical and regiopurity as was firmly established by H NMR spectroscopy at 400
3
5,36
MHz.
1
1

Table 1. Structural information based on the stereospecific numbering (sn-1/2/3), yields and
specific rotation of the benzylated 1-MAG and 1,2-DAG, and the 1,2-DAG intermediates obtained
in accordance with the reactions described in Scheme 1 for the synthesis of the first category of
TAGs
2
5
Entry
sn-1
sn-2
sn-3
Yield (%)
[α]D
(
S)-13
S)-14
S)-15
S)-16
C H COO-
-OH
-OH
-OH
-OH
-OBn
-OBn
-OBn
-OBn
87
94
96
97
+1.11
+3.64
+1.31
+1.48
15
31
(
(
(
C
13
H27COO-
C H COO-
11
23
C
9
H19COO-
(S)-17
(S)-18
(S)-19
(S)-20
(S)-21
(S)-22
(S)-23
(S)-24
C
15
H
31COO- C13
H
27COO-
-OBn
-OBn
-OBn
-OBn
-OBn
-OBn
-OBn
-OBn
95
89
96
84
90
82
86
82
+4.28
+4.32
+8.11
+5.73
+5.34
+4.53
+4.81
+4.69
C H COO- C H COO-
13
27
15 31
15
C H31COO- C11H23COO-
C H COO- C H COO-
11
23
15 31
C H COO- C H COO-
15
31
9
19
9
C H19COO- C15H31COO-
C H COO- C H COO-
13
27
11 23
C H COO- C H COO-
11
23
13 27
(S)-25
(S)-26
(S)-27
(S)-28
(S)-29
(S)-30
(S)-31
(S)-32
C
15
H
31COO- C13
H
27COO-
-OH
-OH
-OH
-OH
-OH
-OH
-OH
-OH
71
74
74
68
83
62
82
58
-2.20
-3.19
-2.94
-3.15
-3.10
*
C H COO- C H COO-
13
27
15 31
C H COO- C H COO-
15
31
11 23
11
C H23COO- C15H31COO-
C H COO- C H COO-
15
31
9
19
9
C H19COO- C15H31COO-
C H COO- C H COO-
-3.17
-2.92
13
27
11 23
C H COO- C H COO-
11
23
13 27
*
Not purified by crystallisation
In accordance with Scheme 1 and Table 1 both the benzylated adducts (S)-17-24 and their derived
,2-DAGs (S)-25-32 are intermediates for the synthesis of the first category of TAGs (R)-1, 2, 4, 5,
1
1
2

7
, 8, 10 and 11, respectively. Information on the structures of all these intermediates is provided in
Table 1 showing the enantiospecific positioning of the acyl groups and other substituents on the
glycerol skeleton for each of these compounds.
Finally, the unsaturated oleic or linoleic acids were introduced to the free hydroxyl group at the sn-
3
position of (S)-25-32 using EDCI as a chemical coupling agent in the presence of DMAP
following the same procedure as described above. The resulting first category products (R)-1, 2, 4,
, 7, 8, 10 and 11 were all obtained in high, and in most cases, excellent yields (77-98%) after
purification by flash chromatography on silica gel. No acyl migration was observed to take place
5
3
5-37
during this reaction as had been firmly established in previous studies.
Table 2 shows the type
of products and the yields obtained along with their specific optical rotation values. As may be
observed these values are extremely low and a question emerges as to whether this may possibly be
an indication of losses in enantiopurity. That possibility, on the other hand, can by ruled out by the
absence of any noticeable acyl migration as will be described below.
a)
Table 2. Structural information based on the stereospecific numbering (sn-1/2/3), yields and
specific rotation of the first category of TAG products (R)-1, 2, 4, 5, 7, 8, 10 and 11 and the second
category of TAG products (S)-3,6,9 and 12 obtained in accordance with Schemes 1 and 2,
respectively
2
5
Entry
sn-1
sn-2
sn-3
Yield (%)
[α]D
(
R)-1
R)-2
C
15
H
31COO- C13
H
27COO- C17
H
33COO-
93
96
79
+0.10
+0.11
+0.07
(
C H COO- C H COO- C H COO-
13
27
15 31
17 33
(
S)-3
C
C
15
15
H
31COO- C17
31COO- C11
H
33COO- C13
23COO- C17
H
27COO-
31COO-
(
R)-4
R)-5
H
H
H
91
94
+0.03
+0.30
(
C H COO- C H COO- C H COO-
11
23
15 31
17 31
1
3

(S)-6
C H COO- C H COO- C H COO-
93
+0.17
15
31
17 31
11 23
(
R)-7
R)-8
C H COO- C H COO- C H COO-
84
77
99
+0.14
+0.11
+0.26
15
31
9
19
17 33
(
C H COO- C H COO- C H COO-
9
19
15 31
17 33
(
S)-9
C H COO- C H COO- C H COO-
15 31 17 33 9 19
(
R)-10
R)-11
C H COO- C H COO- C H COO-
98
95
97
+0.09
+0.32
+0.17
13
27
11 23
17 33
(
C H COO- C H COO- C H COO-
11
23
13 27
17 33
(
S)-12
13
C H27COO- C17H33COO- C11H23COO-
a
Yields are based on the final acylation in Schemes 1 and 2
2
.2 Synthesis of the second category TAGs
The second category products (S)-3, 6, 9 and 12 were synthesised by the synthetic route illustrated
in Scheme 2 for (S)-3. It is a three-step chemoenzymatic route based on the benzylated acylglycerol
adduct (S)-13 as a precursor, shared with the synthesis of the first category TAG products, that
already involved an enzymatic step. This time a second enzymatic step was required to secure the
regiocontrol in introducing the second saturated fatty acyl group exclusively to the sn-3 position of
the glycerol skeleton, after removal of the benzyl protective group, leaving the sn-2 position free
for an unsaturated fatty acid.
Scheme 2. Chemical synthesis of the second category of TAGs (S)-3,6,9 and 12 (shown for (S)-3).
2
Reagents and conditions: (a) H , 10% Pd/C, THF, r.t. (87%); (b) Vinyl tetradecanoate, CAL,
CH Cl , r.t. (83%); (c) cis-Hexadec-9-enoic acid, EDCI, DMAP, CH Cl , r.t. (79%).
2
2
2
2
The benzylated acylglycerol adducts (S)-13 and 14, obtained from the synthesis of the first category
TAGs in accordance with Scheme 1, were subjected to the catalytic hydrogenolysis deprotection of
1
4

the benzyl protective group. The resulting 1-MAGs (S)-33 and 34 were obtained in very good
yields after crystallisation from hot hexane (see Table 3). A previous synthesis of (R)-33 inolved a
similar two-step overall process taking place without purification of the corresponding (R)-13 prior
3
4
to the deprotection. The second step involved a highly regioselective acylation of (S)-33 and 34
with myristic, lauric and capric acids activated as vinyl esters by use of the immobilised Candida
3
7
antarctica lipase. As in a previous report the acylations took place exclusively at the terminal sn-3
position in only 90 minutes, and the resulting 1,3-DAGs (S)-35-38 were accomplished in good to
very high yields after crystallisation from hot hexane (see Table 3). Their excellent regio- and
hence enantiopurity was established by the fact that no acyl migration was observed to take place
during these reactions. Also, the use of the activated vinyl esters and the consequent rapid reaction
should rule out any side reactions involving the lipase to act on the acyl group already present in
the 1-MAGs (S)-33 and 34 to affect the enantiocontrol in the second step. The 1,3-DAGs (S)-35-38
served as intermediates for the synthesis of the second category of TAGs (S)-3, 6, 9 and 12,
respectively. Their structural information along with yields and specific rotation is provided in
Table 3.
Table 3. Structural information based on the stereospecific numbering (sn-1/2/3), yields and
specific rotation of the 1-MAG and 1,3-DAG intermediates obtained in accordance with the
reactions described in Scheme 2 for the synthesis of the second category of TAGs
2
5
Entry
sn-1
sn-2
sn-3
Yield (%)
[α]D
(
S)-33
C H COO-
-OH
-OH
-OH
-OH
87
73
-0.32
-0.40
15
31
(S)-34
C
C
13
15
H
27COO-
(
S)-35
S)-36
H31COO-
-OH
-OH
C
13
H27COO-
83
84
+0.24
+0.61
(
C H COO-
C H COO-
11 23
15
31
1
5

(
S)-37
S)-38
C H COO-
-OH
-OH
C H COO-
86
62
+0.79
+0.68
15
31
9
19
(
C H COO-
C H COO-
11 23
13
27
The asymmetric 1,3-DAG intermediates were subsequently acylated with the unsaturated fatty
acids using the same chemical coupling method as before, (S)-35, 37 and 38 with oleic acid and
(S)-36 with linoleic acid, to prepare the corresponding TAGs belonging to the second category, (S)-
3
, 9, 12 and 6, respectively. These products were accomplished in high and most cases excellent
yields (79-99%) after flash chromatography purification as noted in Table 2. The specific optical
rotation values for these enantiopure TAGs remain of low magnitude, comparable to those
described above (see Table 2).
3
4,37
As reported earlier,
the specific optical activity of chiral TAGs remains extremely low and that
warrants a remark. Enantiopure TAGs are generally known for their extremely low optical activity
3
8,39
as was reported by Baer and Fischer in the late 1930s.
Such compounds are known to display
4
0
41,42
cryptochirality or cryptoactivity,
terms referring to their optical rotation remaining close to
zero and hardly measurable. As can be noticed from the results in Table 2 the specific optical
rotation values for all the enantiopure TAGs of both categories are all of positive sign and range
from +0.03 to +0.32. The specific optical activity values for the asymmetric MAG and DAG
intermediates displayed in Tables 1 and 3 were significantly higher, or roughly two orders of
3
4,37
magnitude, which is consistent with previous reports.
2
.3 Regio- and enantiocontrol
All individual partial acylglycerols possibly involved in the reactions described herein, 1-MAGs, 2-
MAGs, 1,3-DAGs and 1,2-DAGs, as well as their benzylated adducts and the TAGs, display
1
6

1
34-37
characteristic H NMR spectra in their glyceryl skeleton proton region.
This makes it quite
straightforward to spot individual acylglycerol constituents present in a reaction mixture and to
3
6
quantify them with reasonable or good accuracy. Therefore, this is an excellent tool to monitor
both the progress of the reactions as they proceed as well as their regiocontrol in terms of acyl
1
3
migration. The C NMR spectroscopy is also of use in supporting the regiocontrol of the
3
5,36
reactions.
1
The limit of quantification as detected by 400 MHz H NMR spectroscopy for a possible acyl-
migration of 1,3-DAGs into 1,2-DAGs as well as 1-MAGs to 2-MAGs has been determined by
3
6
thorough intensive studies (0.25 mol% for a practical concentration level). In that context it may
be borne in mind that an equilibrium composition involving 1(3)-MAGs and 2-MAGs is roughly
4
3
1
0% 2-MAGs and 90% 1(3)-MAGs, whereas the corresponding equilibrium composition for 1,3-
2
5
DAGs and 1(3),2-DAGs is roughly 70% 1,3-DAGs and 30% 1(3),2-DAGs. It may be further
added that the activation of the fatty acids involved in the lipase promoted reactions as vinyl esters
not only allows the lipase to act very fast under very mild room temperature conditions, under
which no acyl migration takes place, but also a very fast irreversible reaction is promoted, strongly
reducing the risk of any side-reactions taking place, including acyl migration. From all this it is
beyond doubt that losses of enantiocontrol in these reactions caused by acyl migration can be ruled
out in the current work.
All intermediates including the asymmetric MAGs and DAGs and the final TAG products were
obtained in high chemical purity and were all fully characterized by traditional synthetic organic
1
13
chemistry methods including H (400 MHz) and C NMR and IR spectroscopy methods as well as
satisfactory high-resolution accurate mass spectrometry analyses. Specific optical rotation was
determined for all chiral compounds involved. A full regiocontrol, and therefore enantiocontrol, in
1
7

1
13
all the reactions involved was established as based on detailed studies by the H and C NMR
3
5-37
spectroscopy in accordance with previous reports.
That is anticipated in due time to be
unequivocally confirmed in the intended reversed phase chiral HPLC analysis, as certainly was the
2
8,34
case regarding our enantiopure AAB type TAGs previously prepared by comparable methods.
In that case we had access to commercially available TAGs as racemic mixtures, unlike in the
current work. It should be emphasised that most natural chiral TAGs are not separable into
enantiomers without the use of more sophisticated techniques based on double chiral column and
1
1,12,27,28
recycling HPLC systems.
Hence, the determination of the enantiopurity of the current
TAGs must await synthesis of the opposite enantiomers or a racemic mixture of the TAGs and
cannot take place until that has been accomplished.
The synthetic strategy that was followed is suitable for preparing enantiopure structured TAGs of
the ABC type possessing three different fatty acids with the prerequisite that not more than one of
them being unsaturated. This relates to the use of catalytic hydrogenolysis to bring about
disconnection of the protective group from the mono- and diacylated benzyl protected
intermediates that essentially need to be saturated in terms of the fatty acids. TAGs belonging to the
two ABC type categories addressed in the current report, possessing two different saturated and one
unsaturated fatty acids, are readily accessible by the current synthetic approach.
It is of interest that despite the superb regioselectivity offered by the CAL to acylate glycerol and 1-
O-alkylglycerols exclusively at the terminal primary alcohol positions, the same lipase is known to
exhibit no detectable preference for one enantiomer of such substrates over the other in these
reactions. This has been demonstrated in our previous acylation of (S)-1-O-benzylglycerol with
vinyl esters of both palmitic and capric acids taking place with the same rate and efficiency as
3
4
described above for the R-enantiomer. Similar behavior was also observed in the acylation of 1-
1
8

4
4-46
O-alkylglycerol enantiomers in structured ether lipid synthesis
as well as enantiomers of 1-
MAGs in the synthesis of asymmetric 1,3-DAGs (Kristinsson and Haraldsson, unpublished
results). From this it may be concluded that the methodology described herein can be used to
synthesise both enantiomers of the ABC type enantiostructured TAGs addressed in the current
work.
3
. Conclusion
A major obstacle in enantiospecific analysis of intact TAGs in fats and oils of plant and animal
origin is a shortage of enantiopure structured TAG enantiomers to serve as standards to determine
the separation order of two TAG enantiomers in such analysis. That impediment has been
34
addressed to somewhat extent, in a previous report and the current report, describing the synthesis
of twelve enantiopure ABC type structured TAGs, this time possessing three different fatty acids.
Eight of them possess two different saturated fatty acyl groups located in one of the terminal
positions and the mid position, with an unsaturated fatty acyl group in the remaining terminal
position of the glycerol backbone. They were synthesised by a six-step chemoenzymatic route
involving a highly regioselective immobilised Candida antarctica lipase, starting from an
enantiopure solketal. The four remaining TAGs constitute two different saturated fatty acyl groups
in the terminal 1,3-positions and an unsaturated acyl group in the 2-position and were prepared by a
similar six-step approach based on the same chiral precursor, this time requiring two enzymatic
steps with the same lipase. This highly efficient synthetic strategy should prove of high use to
synthesise a variety of similar and related enantiopure ABC type structured TAGs to serve as
standards in stereospecific analysis of intact TAGs in fats and oils.
4
. Experimental
1
9

4
.1 General
1
13
H and C nuclear magnetic resonance spectra were recorded on a Bruker Avance 400
spectrometer in deuterated chloroform as a solvent at 400.12 and 100.61 MHz, respectively.
Chemical shifts (δ) are reported in parts per million (ppm) and the coupling constants (J) in Hertz
(Hz). The following abbreviations are used to describe the multiplicity: s, singlet; d, doublet; t,
triplet; q, quartet; quin, quintet; dd, doublet of doublets; m, multiplet. Infrared spectra were
conducted on a Thermo Nicolet FT-IR iS10 Spectrophotometer on a KBr pellet (crystalline
material) or as a neat liquid (oils). Melting points were determined on a Büchi 520 melting point
apparatus and are uncorrected. The high-resolution mass spectra (HRMS) were acquired on a
Bruker micrOTOF-Q mass spectrometer. All data analysis was done on Bruker software. Optical
R
activity measurements were performed on an Autopol V Automatic Polarimeter from Rudolph
Research Analytical using a 40T-2.5-100-0.7 Temp Trol™ polarimetric cell with 2.5 mm inside
diameter, 100 mm optical path length and 0.7 mL volume with c referring to g sample/100 mL.
The immobilised Candida antarctica lipase (Novozym 435; CAL-B) was supplied as a gift from
Novozymes A/S (Bagsvaerd, Denmark). All chemicals and solvents were used without further
purification unless otherwise stated. (S)-(+)-2,2-Dimethyl-1,3-dioxolane-4-methanol, (S)-solketal,
(98% purity and 99% ee) were purchased from Sigma-Aldrich (Steinheim, Germany). EDCI (1-(3-
dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride, 98%) was obtained commercial grade
from Sigma-Aldrich (Steinheim, Germany) and 4-dimethylaminopyridine (DMAP, 99%) from
Acros Organics (Geel, Belgium). Benzyl bromide (98%), sodium hydride (60% dispersion in
mineral oil) and 10% palladium on carbon catalyst were from Sigma-Aldrich (Steinheim). Linoleic,
oleic, palmitic, myristic, lauric and capric acids were all obtained as free acids in high (>99%)
purity from Larodan Fine Chemicals (Malmö, Sweden). Vinyl palmitate (96%), vinyl myristate
(99%), vinyl laurate (99%) and vinyl caprate (99%) were all obtained from TCI Europe
2
0

(Zwinderecht, Belgium). Dichloromethane and ethyl acetate were obtained HPLC grade from
Sigma-Aldrich (Steinheim, Germany), n-hexane as p.a. from Merck (Darmstadt, Germany),
petroleum ether, boiling range 40-60 °C, and ethanol (99.8%) from Honeywell (Seelze, Germany).
2
Dichloromethane was dried over CaH under dry nitrogen atmosphere. Tetrahydrofuran for analysis
was obtained from Acros Organics (Geel, Belgium) and dried over Na wire in presence of
benzophenone under dry nitrogen atmosphere. Column chromatography was performed on Silica
gel 60 (Silicycle, Ontario). Reactions were monitered by TLC on Silica gel 60 F254 (Silicycle,
Ontario), with detection by quenching of fluorescence and/or with phosphomolybdic acid in
ethanol.
4
.2 3-O-Benzyl-sn-glycerol
To a suspension of NaH (9.08 g, 60 wt% dispersion in mineral oil, 227 mmol, 3.0 equiv.) in THF
189 mL) at 0 °C was added (S)-solketal (9.43 mL, 75.7 mmol, 1.0 equiv.) dropwise via syringe.
(
The cooling bath was removed, and the resulting suspension was stirred at room temperature for 1
h before being cooled back to 0 °C followed by dropwise addition of benzyl bromide (10.8 mL,
9
0.8 mmol, 1.2 equiv.). The resulting mixture was brought to reflux and stirred for 4 h after which
it was cooled to 0 °C, diluted carefully with EtOH (116 mL), followed by addition of a 2.0 M
aqueous solution of HCl (114 mL). After stirring at 80 °C for 1 h, the resulting mixture was cooled
3
to room temperature and quenched by addition of a saturated aqueous solution of NaHCO (250
mL). The organic layer was separated and the aqueous layer further extracted with EtOAc (3 × 150
mL), the combined organic layers were dried over anhydrous Na SO , filtered and concentrated in
2
4
vacuo to give the crude product which was purified via column chromatography (petroleum ether /
EtOAc (50%) → petroleum ether / EtOAc (80%)) to give 3-O-benzyl-sn-glycerol (12.7 g, 69.8
mmol, 92%) as a yellow oil;
2
1

2
D
4
1
[α]
3 f
+5.10 (c 19.6, CHCl ); R 0.20 (petroleum ether / EtOAc (70%)); H NMR (400 MHz,
CDCl ) δ 7.40-7.27 (5H, m, ArH), 4.56 (2H, s, OCH Ar), 3.95-3.85 (1H, m, CHOH), 3.77-3.69
3
H
2
(1H, m, CH OH), 3.67-3.62 (1H, m, CH ’OH), 3.59 (1H, dd, J = 9.5 and 4.0 Hz, CH OBn), 3.55
2 2 2
(
1H, dd, J = 9.5 and 6.0 Hz, CH ’OBn), 2.57 (1H, d, J = 4.5 Hz, OH), 2.06 (1H, app. t, J = 6.0 Hz,
2
1
3
OH); C NMR (100 MHz, CDCl ) δ 137.7, 128.5, 127.9, 127.8, 73.6, 71.8, 70.6, 64.1.
3
C
3
4
Spectroscopic data in agreement with that reported for the R-enantiomer by Kristinsson et al.
4
.3 General Procedure A: Enzymatic Acylation Procedure
To a mixture of benzyl protected glycerol (1.0 equiv.) and vinyl ester of the appropriate saturated
fatty acid (1.05-1.3 equiv.) in CH Cl (3.5 mL / mmol of substrate) was added immobilized
2
2
Candida antarctica lipase B (CAL) (4-10 wt% of total combined mass of substrates). The resulting
suspension was stirred at room temperature at a rate not to imperil the solid supported lipase. After
stirring for 1.5 h (or until full consumption of starting material as indicated by TLC) the lipase was
separated by filtration and the filtrate concentrated in vacuo to give the crude product which was
purified by either flash column chromatography or recrystallisation (where appropriate). Note: 1-
Acyl-3-O-alkylglycerols should be purified via column chromatography using 4% boric acid
impregnated silica gel as they are prone to acyl migration.
4
.4 3-O-Benzyl-1-hexadecanoyl-sn-glycerol, (S)-13
S)-13 was prepared using General Procedure A, with 3-O-benzyl-sn-glycerol (100 mg, 0.55 mmol,
.0 equiv.), vinyl palmitate (117 mg, 0.63 mmol, 1.15 equiv.) and CAL (18 mg) in CH Cl (2.0
(
1
2
2
mL), with a 1.5 h reaction time. Purification via column chromatography (4% boric acid
impregnated silica gel, petroleum ether / EtOAc (40%)) gave (S)-13 (90 mg, 0.48 mmol, 87%) as a
colourless oil;
2
2

2
D
5
-1
[α]
+1.11 (c 9.0, CH
2 2
Cl ); IR (NaCl, νmax / cm ) 3459, 3064, 3031, 2924, 2853, 1739, 1174;
1
H NMR (400 MHz, CDCl ) δ 7.39-7.27 (5H, m, ArH), 4.56 (2H, s, OCH Ar), 4.19 (1H, dd, J =
3
H
2
1
1.5 and 4.5 Hz, CH OCO), 4.14 (1H, dd, J = 11.5 and 6.0 Hz, CH ’OCO), 4.07-4.00 (1H, m,
2 2
CHOCO), 3.56 (1H, dd, J = 9.5 and 4.5 Hz, CH OBn), 3.50 (1H, dd, J = 9.5 and 6.0 Hz,
2
CH ’OBn), 2.51 (1H, bs, OH), 2.32 (2H, t, J = 7.5 Hz, CH COO), 1.67-1.56 (2H, m,
2
2
13
CH CH COO), 1.38-1.24 (24H, m, 12 × CH ), 0.88 (3H, t, J = 6.5 Hz, CH ); C NMR (100 MHz,
2
2
2
3
CDCl
3
) δ
C
174.1, 137.8, 128.6, 128.0, 127.9, 73.7, 71.0, 69.1, 65.5 34.3, 32.1, 29.83, 29.80, 29.75,
+
2
9.6, 29.5, 29.4, 29.3, 25.1, 22.8, 14.3; HRMS (ESI+) calc. for C26
H O
44 4
Na [M+Na] 443.3132,
found 443.3127.
4
.5 3-O-Benzyl-1-tetradecanoyl-sn-glycerol, (S)-14
S)-14 was prepared using General Procedure A, with 3-O-benzyl-sn-glycerol (50 mg, 0.27 mmol,
.0 equiv.), vinyl myristate (0.091 mL, 0.31 mmol, 1.15 equiv.) and CAL (6 mg) in CH Cl (1.0
(
1
2
2
mL), with a 1.5 h reaction time. Purification via column chromatography (4% boric acid
impregnated silica gel, petroleum ether / EtOAc (40%)) gave (S)-14 (100 mg, 0.25 mmol, 94%) as
a colourless oil;
2
5
-1
[
α]D +3.64 (c 5.0, CH Cl ); IR (NaCl, ν
/ cm ) 3458, 3064, 3031, 2925, 2854, 1739, 1174;
max
2
2
1
H NMR (400 MHz, CDCl ) δ 7.38-7.28 (5H, m, ArH), 4.56 (2H, s, OCH Ar), 4.19 (1H, dd, J =
3
H
2
1
1.5 and 4.5 Hz, CH
2
OCO), 4.14 (1H, dd, J = 11.5 and 6.0 Hz, CH
2
’OCO), 4.06-4.00 (1H, m,
2
CHOCO), 3.56 (1H, dd, J = 9.5 and 4.5 Hz, CH OBn), 3.50 (1H, dd, J = 9.5 and 6.0 Hz,
CH ’OBn), 2.48 (1H, bs, OH), 2.32 (2H, t, J = 7.5 Hz, CH COO), 1.66-1.58 (2H, m,
2
2
13
CH CH COO), 1.32-1.26 (20H, m, 10 × CH ), 0.89 (3H, t, J = 6.5 Hz, CH ); C NMR (100 MHz,
2
2
2
3
CDCl ) δ 174.1, 137.8, 128.6, 128.0, 127.9, 73.6, 71.0, 69.1, 65.5 34.3, 32.1, 29.8, 29.7, 29.6,
3
C
2
3

+
2
4
9.5, 29.4, 29.3, 25.1, 22.8, 14.2; HRMS (ESI+) calc. for C H O Na [M+Na] 415.2819, found
24 40 4
15.2806.
4
.6 3-O-Benzyl-1-dodecanoyl-sn-glycerol, (S)-15
(
S)-15 was prepared using General Procedure A, with 3-O-benzyl-sn-glycerol (103 mg, 0.57 mmol,
.0 equiv.), vinyl laurate (0.19 mL, 0.72 mmol, 1.26 equiv.) and CAL (20 mg) in CH Cl (2.0 mL),
1
2
2
with a 1.5 h reaction time. Purification via column chromatography (4% boric acid impregnated
silica gel, petroleum ether / EtOAc (40%)) gave (S)-15 (199 mg, 0.55 mmol, 96%) as a colourless
oil;
2
D
5
-1
[α]
+1.31 (c 4.70, CH
2 2
Cl ); IR (NaCl, νmax / cm ) 3459, 3031, 2925, 2854, 1739, 1174;
1
H NMR (400 MHz, CDCl ) δ 7.38-7.27 (5H, m, ArH), 4.56 (2H, s, OCH Ar), 4.19 (1H, dd, J =
3
H
2
1
1.5 and 4.5 Hz, CH OCO), 4.14 (1H, dd, J = 11.5 and 6.0 Hz, CH ’OCO), 4.06-4.00 (1H, m,
2 2
CHOCO), 3.56 (1H, dd, J = 9.5 and 4.5 Hz, CH OBn), 3.50 (1H, dd, J = 9.5 and 6.0 Hz,
2
CH
2
2
2
’OBn), 2.48 (1H, bs, OH), 2.32 (2H, t, J = 7.5 Hz, CH COO), 1.65-1.57 (2H, m,
13
CH
2
CH COO), 1.32-1.26 (16H, m, 8 × CH
2
), 0.88 (3H, t, J = 6.5 Hz, CH
3
); C NMR (100 MHz,
CDCl
3
) δ
C
174.1, 137.8, 128.6, 128.0, 127.9, 73.6, 71.0, 69.1, 65.5 34.3, 32.0, 29.7, 29.6, 29.5,
+
2
3
9.4, 29.3, 25.1, 22.8, 14.2; HRMS (ESI+) calc. for C H O Na [M+Na] 387.2506, found
2
2
35
4
87.2498.
4
.7 3-O-Benzyl-1-decanoyl-sn-glycerol, (S)-16
(
S)-16 was prepared using General Procedure A, with 3-O-benzyl-sn-glycerol (250 mg, 1.37 mmol,
.0 equiv.), vinyl caprate (0.28 mL, 1.76 mmol, 1.28 equiv.) and CAL (24 mg) in CH Cl (4.0 mL),
1
2
2
with a 1.5 h reaction time. Purification via column chromatography (4% boric acid impregnated
silica gel, petroleum ether / EtOAc (40%)) gave (S)-16 (447 mg, 1.33 mmol, 97%) as a colourless
oil;
2
4

2
D
5
-1
[α]
+1.48 (c 10.0, CH
2 2
Cl ); IR (NaCl, νmax / cm ) 3459, 3031, 2926, 2855, 1739, 1173;
1
H NMR (400 MHz, CDCl ) δ 7.38-7.28 (5H, m, ArH), 4.56 (2H, s, OCH Ar), 4.19 (1H, dd, J =
3
H
2
1
1.5 and 4.5 Hz, CH OCO), 4.14 (1H, dd, J = 11.5 and 6.0 Hz, CH ’OCO), 4.06-4.01 (1H, m,
2 2
CHOCO), 3.55 (1H, dd, J = 9.5 and 4.5 Hz, CH OBn), 3.49 (1H, dd, J = 9.5 and 6.0 Hz,
2
CH ’OBn), 2.57 (1H, bs, OH), 2.32 (2H, t, J = 7.5 Hz, CH COO), 1.66-1.57 (2H, m,
2
2
13
CH CH COO), 1.35-1.26 (12H, m, 6 × CH ), 0.87 (3H, t, J = 6.5 Hz, CH ); C NMR (100 MHz,
2
2
2
3
CDCl
3
) δ
C
174.1, 137.8, 128.6, 128.0, 127.9, 73.6, 71.0, 69.1, 65.5 34.3, 32.0, 29.5, 29.4, 29.3,
+
2
4.0, 22.8, 14.2; HRMS (ESI+) calc. for C20
H
32
O
4
Na [M+Na] 359.2193, found 359.2193.
4
.8 General Procedure B: Chemical Acylation Procedure
To a flask containing glycerol (1.0 equiv.), the appropriate fatty acid (1.10 equiv.), EDCI (1.20
equiv.) and DMAP (0.20 equiv.) was added CH Cl (2.5 mL / mmol substrate). After stirring at
2
2
room temperature for 16 h, the reaction mixture was concentrated in vacuo and purified via column
chromatography.
4
.9 3-O-Benzyl-1-hexadecanoyl-2-tetradecanoyl-sn-glycerol, (S)-17
S)-17 was prepared using General Procedure B, with (S)-13 (436 mg, 1.04 mmol, 1.0 equiv.),
myristic acid (260 mg, 1.14 mmol, 1.1 equiv.), EDCI (237 mg, 1.24 mol, 1.2 equiv.) and DMAP
25 mg, 0.21 mmol, 0.20 equiv.) in CH Cl (2.6 mL) with a reaction time of 16 h. Purification via
(
(
2
2
recrystallisation from hot ethanol gave (S)-17 (620 mg, 0.983 mmol, 95%) as a colourless wax;
2
D
5
-1
m.p. 37.7-39.8 °C; [α]
3
+4.28 (c 1.94, CHCl ); IR (NaCl, νmax / cm ) 2915, 1721, 1646, 1298;
1
H NMR (400 MHz, CDCl ) δ 7.38-7.26 (5H, m, ArH), 5.28-5.20 (1H, m, CHOCO), 4.56 (1H, d,
3
H
J = 12.0 Hz, OCH Ar), 4.52 (1H, d, J = 12.0 Hz, OCH ’Ar), 4.35 (1H, dd, J = 12.0 and 4.0 Hz,
2
2
CH OCO), 4.19 (1H, dd, J = 12.0 and 6.5 Hz, CH ’OCO), 3.63-3.55 (2H, m, CH OBn), 2.32 (2H,
2
2
2
2
5

t, J = 7.5 Hz, CH COO), 2.28 (2H, t, J = 7.5 Hz, CH COO), 1.68-1.55 (4H, m, 2 × CH CH COO),
2
2
2
2
1
3
1
1
2
.38-1.18 (44H, m, 22 × CH
2
3 3 C
), 0.88 (6H, t, J = 6.5 Hz, 2 × CH ); C NMR (100 MHz, CDCl ) δ
73.6, 173.3, 137.9, 128.6, 127.9, 127.8, 73.5, 70.2, 68.4, 62.8, 34.5, 34.3, 32.1, 29.84, 29.81,
9.78, 29.63, 29.50, 29.44, 29.28, 29.24, 25.1, 25.0, 22.8, 14.4; HRMS (ESI+) calc. for
+
C H O Na [M+Na] 653.5115, found 653.5094.
40
70
5
4
.10 3-O-Benzyl-2-hexadecanoyl-1-tetradecanoyl-sn-glycerol, (S)-18
S)-18 was prepared using General Procedure B, with (S)-14 (496 mg, 1.26 mmol, 1.0 equiv.),
palmitic acid (356 mg, 1.39 mmol, 1.1 equiv.), EDCI (291 mg, 1.52 mol, 1.2 equiv.) and DMAP
31 mg, 0.25 mmol, 0.20 equiv.) in CH Cl (3.0 mL) with a reaction time of 16 h. Purification via
recrystallisation from hot ethanol gave (S)-18 (710 mg, 1.26 mmol, 89%) as a colourless solid;
(
(
2
2
2
5
-1
m.p. 31.3-33.9 °C; [α]D +4.32 (c 2.0, CHCl ); IR (NaCl, ν
/ cm ) 2915, 1721, 1646, 1298;
max
3
1
H NMR (400 MHz, CDCl ) δ 7.38-7.26 (5H, m, ArH), 5.28-5.20 (1H, m, CHOCO), 4.56 (1H, d,
3
H
J = 12.0 Hz, OCH
CH OCO), 4.19 (1H, dd, J = 12.0 and 6.5 Hz, CH
t, J = 7.5 Hz, CH COO), 2.28 (2H, t, J = 7.5 Hz, CH
.38-1.18 (44H, m, 22 × CH ), 0.88 (6H, t, J = 6.5 Hz, 2 × CH ); C NMR (100 MHz, CDCl ) δ
2
Ar), 4.52 (1H, d, J = 12.0 Hz, OCH
’OCO), 3.63-3.55 (2H, m, CH
COO), 1.68-1.55 (4H, m, 2 × CH
2
’Ar), 4.35 (1H, dd, J = 12.0 and 4.0 Hz,
OBn), 2.32 (2H,
CH COO),
2
2
2
2
2
2
2
13
1
1
2
2
3
3
C
73.6, 173.3, 137.9, 128.6, 127.9, 127.8, 73.5, 70.2, 68.4, 62.8, 34.5, 34.3, 32.1, 29.84, 29.81,
9.78, 29.64, 29.51, 29.44, 29.28, 29.24, 25.1, 25.0, 22.8, 14.3; HRMS (ESI+) calc. for
+
C H O Na [M+Na] 653.5115, found 653.5111.
40
70
5
4
.11 3-O-Benzyl-2-dodecanoyl-1-hexadecanoyl-sn-glycerol, (S)-19
S)-19 was prepared using General Procedure B, with (S)-13 (460 mg, 1.09 mmol, 1.0 equiv.),
lauric acid (241 mg, 1.20 mmol, 1.1 equiv.), EDCI (250 mg, 1.31 mol, 1.2 equiv.) and DMAP (27
(
2
6

mg, 0.22 mmol, 0.20 equiv.) in CH Cl (2.7 mL) with a reaction time of 16 h. Purification via
2
2
recrystallisation from hot ethanol gave (S)-19 (633 mg, 1.05 mmol, 96%) as a colourless wax;
2
5
-1
1
[
α]D +8.11 (c 1.75, CHCl ); IR (NaCl, ν / cm ) 2915, 1721, 1646, 1298; H NMR (400 MHz,
3 max
CDCl ) δ 7.38-7.26 (5H, m, ArH), 5.28-5.20 (1H, m, CHOCO), 4.56 (1H, d, J = 12.0 Hz,
3
H
OCH Ar), 4.52 (1H, d, J = 12.0 Hz, OCH ’Ar), 4.35 (1H, dd, J = 12.0 and 4.0 Hz, CH OCO), 4.19
2
2
2
(
1H, dd, J = 12.0 and 6.5 Hz, CH ’OCO), 3.63-3.55 (2H, m, CH OBn), 2.32 (2H, t, J = 7.5 Hz,
2
2
CH
2
COO), 2.28 (2H, t, J = 7.5 Hz, CH
2
COO), 1.68-1.55 (4H, m, 2 × CH
2
2
CH COO), 1.38-1.18
13
(
40H, m, 20 × CH
2
), 0.88 (6H, t, J = 6.5 Hz, 2 × CH
3
); C NMR (100 MHz, CDCl
3
) δ 173.6,
C
1
2
73.3, 137.9, 128.6, 127.9, 127.8, 73.5, 70.2, 68.4, 62.8, 34.5, 34.3, 32.06, 32.05, 29.84, 29.80,
9.77, 29.62, 29.50, 29.48, 29.43, 29.27, 29.24, 25.1, 25.0, 22.8, 14.2; HRMS (ESI+) calc. for
+
C H O Na [M+Na] 625.4802, found 625.4801.
38
66
5
4
.12 3-O-Benzyl-1-dodecanoyl-2-hexadecanoyl-sn-glycerol, (S)-20
S)-20 was prepared using General Procedure B, with (S)-15 (501 mg, 1.37 mmol, 1.0 equiv.),
palmitic acid (388 mg, 1.51 mmol, 1.1 equiv.), EDCI (316 mg, 1.65 mol, 1.2 equiv.) and DMAP
34 mg, 0.27 mmol, 0.20 equiv.) in CH Cl (3.5 mL) with a reaction time of 16 h. Purification via
recrystallisation from hot ethanol gave (S)-20 (700 mg, 1.16 mmol, 84%) as a colourless solid;
(
(
2
2
2
5
-1
1
[
α]D +5.73 (c 2.06, CHCl ); IR (NaCl, ν / cm ) 2915, 1721, 1646, 1298; H NMR (400 MHz,
3 max
CDCl
3
) δ
H
7.38-7.26 (5H, m, ArH), 5.28-5.20 (1H, m, CHOCO), 4.56 (1H, d, J = 12.0 Hz,
OCH
2
Ar), 4.52 (1H, d, J = 12.0 Hz, OCH
2
2
’Ar), 4.35 (1H, dd, J = 12.0 and 4.0 Hz, CH OCO), 4.19
(
1H, dd, J = 12.0 and 6.5 Hz, CH ’OCO), 3.63-3.55 (2H, m, CH OBn), 2.32 (2H, t, J = 7.5 Hz,
2
2
CH COO), 2.28 (2H, t, J = 7.5 Hz, CH COO), 1.68-1.55 (4H, m, 2 × CH CH COO), 1.38-1.18
2
2
2
2
1
3
(
40H, m, 20 × CH ), 0.88 (6H, t, J = 6.5 Hz, 2 × CH ); C NMR (100 MHz, CDCl ) δ 173.6,
2 3 3 C
1
73.3, 137.9, 128.6, 127.9, 127.8, 73.5, 70.2, 68.4, 62.8, 34.5, 34.3, 32.06, 32.05, 29.84, 29.80,
2
7

2
9.77, 29.62, 29.50, 29.48, 29.43, 29.27, 29.24, 25.1, 25.0, 22.8, 14.2; HRMS (ESI+) calc. for
+
C
38
H
66
O
5
Na [M+Na] 625.4802, found 625.4799.
4
.13 3-O-Benzyl-2-decanoyl-1-hexadecanoyl-sn-glycerol, (S)-21
(S)-21 was prepared using General Procedure B, with (S)-13 (460 mg, 1.09 mmol, 1.0 equiv.),
capric acid (207 mg, 1.20 mmol, 1.1 equiv.), EDCI (250 mg, 1.31 mol, 1.2 equiv.) and DMAP (27
mg, 0.22 mmol, 0.20 equiv.) in CH Cl (2.7 mL) with a reaction time of 16 h. Purification via
2
2
recrystallisation from hot ethanol gave (S)-21 (564 mg, 0.981 mmol, 90%) as a colourless wax;
2
D
5
-1
1
[α]
3
+5.34 (c 2.81, CHCl ); IR (NaCl, νmax / cm ) 2915, 1721, 1646, 1298; H NMR (400 MHz,
CDCl ) δ 7.38-7.26 (5H, m, ArH), 5.28-5.20 (1H, m, CHOCO), 4.56 (1H, d, J = 12.0 Hz,
3
H
OCH Ar), 4.52 (1H, d, J = 12.0 Hz, OCH ’Ar), 4.35 (1H, dd, J = 12.0 and 4.0 Hz, CH OCO), 4.19
2
2
2
(
1H, dd, J = 12.0 and 6.5 Hz, CH ’OCO), 3.63-3.55 (2H, m, CH OBn), 2.32 (2H, t, J = 7.5 Hz,
2
2
CH COO), 2.28 (2H, t, J = 7.5 Hz, CH COO), 1.68-1.55 (4H, m, 2 × CH CH COO), 1.38-1.18
2
2
2
2
13
(
36H, m, 18 × CH
2
3 3 C
), 0.88 (6H, t, J = 6.5 Hz, 2 × CH ); C NMR (100 MHz, CDCl ) δ 173.5,
1
2
73.2, 137.9, 128.5, 127.9, 127.8, 73.4, 70.1, 68.4, 62.8, 34.5, 34.3, 32.1, 32.0, 29.83, 29.79, 29.76,
9.62, 29.57, 29.49, 29.42, 29.26, 29.22, 25.1, 25.0, 22.82, 22.80, 14.23, 14.23; HRMS (ESI+)
+
calc. for C H O Na [M+Na] 597.4489, found 597.4476.
3
6
62
5
4
.14 3-O-Benzyl-1-decanoyl-2-hexadecanoyl-sn-glycerol, (S)-22
S)-22 was prepared using General Procedure B, with (S)-16 (403 mg, 1.20 mmol, 1.0 equiv.),
palmitic acid (338 mg, 1.32 mmol, 1.1 equiv.), EDCI (276 mg, 1.44 mol, 1.2 equiv.) and DMAP
29 mg, 0.24 mmol, 0.20 equiv.) in CH Cl (3.0 mL) with a reaction time of 16 h. Purification via
recrystallisation from hot ethanol gave (S)-22 (567 mg, 0.986 mmol, 82%) as a colourless wax;
(
(
2
2
2
8