O-Silylated C3-halohydrins as a novel class of protected building blocks for total, regio- and stereocontrolled synthesis of glycerolipid frameworks

Article abstract of DOI:10.1039/b915533c

We propose O-silylated C3-halohydrins [1(3)-O-silyl-2-O-acyl-, 1,2(2,3)-O-bis(silyl)-, and 1(3)-O-acyl-2-O-silyl-3(1)-halo-sn-glycerides] as new chirons in the total synthesis of glycerolipid constructs. These are efficiently producible via opening of the

Full text of DOI:10.1039/b915533c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
O-Silylated C3-halohydrins as a novel class of protected building blocks
for total, regio- and stereocontrolled synthesis of glycerolipid frameworks†
Stephan D. Stamatov*a and Jacek Stawinski*b,c
Received 29th July 2009, Accepted 22nd October 2009
First published as an Advance Article on the web 26th November 2009
DOI: 10.1039/b915533c
We propose O-silylated C3-halohydrins [1(3)-O-silyl-2-O-acyl-, 1,2(2,3)-O-bis(silyl)-, and
1(3)-O-acyl-2-O-silyl-3(1)-halo-sn-glycerides] as new chirons in the total synthesis of glycerolipid
constructs. These are efficiently producible via opening of the oxirane ring of the corresponding glycidyl
derivatives and permit (i) displacement of the iodine by a requisite carboxylate in the presence of
O-triisopropylsilyl (O-TIPS), O-tert-butyldimethylsilyl (O-TBDMS), and O-acyl substituents; (ii)
selective acylation across an appropriate silyloxy system [e.g., O-TBDMS or O-triethylsilyl (O-TES)] of
monoesterified haloglycerides; (iii) direct exchange of an O-silyl protection (e.g., O-TBDMS or O-TIPS)
for a trichloroacetyl group; (iv) conversion of a terminal TBDMS group into the corresponding
trifluoroacetate without affecting O-TIPS-, O-acyl- and iodo functions. The above transformations
secure flexible routes to a variety of otherwise difficult-to-access key-intermediates [e.g., 1,2(2,3)-O-
bis(acyl)-3(1)-trichloroacetyl-, 1,3-O-bis(acyl)-2-trichloroacetyl-, 1,2(2,3)-O-bis(acyl)-3(1)-O-
TBDMS/TIPS-, 1,3-O-bis(acyl)-2-O-TIPS/TBDMS-, 1(3)-O-acyl-2-O-TIPS-, 1,2(2,3)-O-
bis(acyl)-3(1)-iodo-sn-glycerols, etc.] and lend themselves to a powerful methodology for the
preparation of di- and triacylglycerols as well as glycerol-based cationic lipids. The reactions involved
are entirely regio- and stereospecific, avoid acyl migration, and can provide target compounds with a
chosen absolute configuration from a single synthetic precursor.
vital physiological phenomena via signal transduction pathways.12
Growing interest in lipid mediators concerns site-terminal iso-
1. Introduction
The ever-increasing experimental evidence that glycerolipids might
mers of 1,2-DAGs as well, namely 1,3-O-diacyl-sn-glycerols (1,3-
be a pivotal regulatory element implicated in normal cell function
DAGs), in view of their distinctive role as endogenous vehicles
and disease,1 made this class of biomolecules a focal point of scien-
tific inquiry in once divergent fields of nucleic acid,2 carbohydrate,3
and lipid research,4-7 relevant to clinical diagnostics8 and rational
drug design.9,10
Naturally occurring 1,2-O-diacyl-sn-glycerols (1,2-DAGs) oc-
cupy an important position in this regard as they have recently
been shown to govern the activity of particular protein families
[e.g., protein kinase C (PKC), protein kinase D (PKD), RasGRP,
the chimaerins, Unc-13, etc.]11 involved in a wide range of
for transporting essential fatty acids through intestinal mucosa13
or biomodulators per se.14 Such structurally defined DGs are
valuable precursors to triacylglycerols (TAGs)15,16 or related
isosteres (e.g., phospholipids,5,17,18 cationic lipids,19,20 and others9)
and have attracted much attention as micromolecular vectors for
either gene transfection21 or organ-addressed delivery of contrast
agents,8,22 therapeutics,23 antioxidants,24,25 and others.26 In this
context, availability of appropriate chemistry allowing flexible
derivatization of the glycerol skeleton or parent frameworks is
crucial when aiming at new glycerolipid analogues of biochemical
and pharmacological relevance.9,27,28
aDepartment of Chemical Technology, University of Plovdiv, 24 Tsar Assen
St., Plovdiv 4000, Bulgaria
While our recent research in the aforementioned field resulted
in an efficient triester approach to a stereospecific prepara-
tion of DGs and TAGs,29 the method developed has some
limitations. Firstly, the synthetic protocol relied on configura-
tionally pure 1(3)-monoacyl-sn-glycerols [1(3)-MAGs], which are
often difficult to obtain by means of conventional methods30,31
and show high propensity towards racemization.31-33 Secondly,
since the two primary carbinol groups of the glycerol back-
bone are diastereotopic,34 it precludes admission to both enan-
tiomers of 1,3-DAGs and TAGs, including regioisomeric 1,2(2,1)-
DAGs, from one generic C3-unit. Thirdly, the methodology
involved derivatization of intermediary 1(3)-O-acyl-3(1)-O-silyl-
sn-glycerols and thus access to other glycerolipid analogues, as for
example, cationic amphiphiles that are of high demand in gene
therapy,35 is restricted.
bDepartment of Organic Chemistry, Arrhenius Laboratory, Stockholm
University, S-106 91 Stockholm, Sweden
cInstitute of Bioorganic Chemistry, Polish Academy of Sciences,
Noskowskiego, 12/14, 61-704 Poznan, Poland
† Electronic supplementary information (ESI) available: Mechanistic
insights into a trifluoroacetate anion/4-N,N-dimethylaminopyridine-
assisted synthesis of C2-O-acylated- (7–11, 61) and C2-O-silylated (12–19)
C3-vicinal halohydrins from glycidyls (1–6). Selective trifluoroacetylation
across tert-butyldimethylsilyloxy systems of 1-O-tert-butyldimethylsilyl-
2-O-triisopropylsilyl-3-iodo- (13) or -3-O-acyl-sn-glycerols (26, 27) as
mediated by trifluoroacetic anhydride in the presence of methanol. The
incidence of long-range acyloxy migration during silver trifluoroacetate-
promoted replacement of halogen in 1-oleoyl-2-O-tert-butyldimethylsilyl-
3-iodo-sn-glycerol (15). Preparative details of transformations and charac-
teristics of compounds not shown in the Experimental section. Analytical
criteria for assessing regiochemistry of the preparations by high-resolution
1H and 13C NMR spectroscopy. H- and 13C NMR data for compounds
7–63. See DOI: 10.1039/b915533c
1
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 463
©

View Article Online
To address these problems, we report here on three types
of O-silylated C3-halohydrin derivatives I–III (Scheme 1) that,
to our knowledge, find no literature precedents as building
blocks for regio- and stereospecific construction of glycerolipids.
Unlike other glyceride precursors,5 I–III combine stability un-
der a wide range of reaction conditions with susceptibility to
selected reagents (e.g., carboxylate, tetra-n-butylammonium halide
(Bu4NX)–trimethylsilyl halide (TMSX)–carboxylic acid anhy-
dride (CAA),36 Et3N·3HF–trichloroacetic anhydride (TCAA),37
etc.). These permit introduction of various acyloxy moieties at
sn-C1-, sn-C2-, and sn-C3-centres on the glycerol skeleton in a
strictly chemo- and regioselective manner without exposure of a
free hydroxyl function.
these instances could be done by 1H and 13C NMR spectroscopy,29
without recourse to the enzymatic or chemo-enzymatic analytical
techniques often needed for structure elucidation of long-chain
triacylglycerols43,44 lacking as a rule detectable optical activity16,45
and characteristic spectral features.46
Synthesis of halohydrins of types I, II, and III (route A in
Scheme 1 and Scheme 2)
At the crux of the stereospecific synthesis of mixed-acid DGs,
TAGs or related conjugates is the need to generate and maintain
chirality of a glycerol unit during its sequential derivatization
to allow the acyl residues to be introduced at sn-1-, sn-2-, and
sn-3-positions.7,15 This has always been problematic when using
acetal-7,15 or oxirane-derived30,47 C3-synthons during glycerolipid
construction.5,16 It is noteworthy that access to both enantiomers
from these types of classic building blocks6,48 usually requires using
two different starting materials.5,7,15,18
The above situation turned our attention to enantiopure mono-
and bis(silylated) C3-vicinal halohydrins as alternative glyceride
precursors. Unlike their acetal- and oxirane counterparts, that are
highly susceptible to racemization/solvolysis also upon storage,49
the combination of silyl and halogen functionalities should (i)
confer stability to a molecule and prevent migration of a fatty
acid radical if present; (ii) contribute to differentiation of the
two primary carbinol groups since conversion of silyloxy-,36,37
hydroxyl-,50 and halide47,51 systems into a variety of carboxylates
proceeds via chemically independent routes. As to reactivity of
halohydrins, iodo derivatives appeared to be superior to their less
reactive chloro-47 and bromo-homologues44,51 that typically require
more forcing conditions (e.g., elevated temperature or the use of
highly polar solvents) to effect a reaction with nucleophiles.19,52
Since 1(3)-O-silyl- and 1(3)-O-acyl-3(1)-halo-sn-glycerols are
directly obtainable in virtually quantitative yields from glycidyl
precursors,53 we became interested in the development of an
efficient protocol for their in situ transformation into the respective
1(3)-O-silyl-2-O-acyl-, 1,2(2,3)-O-bis(silyl)-, and 1(3)-O-acyl-2-O-
silyl-3(1)-halo-sn-derivatives (preferably iodides) of significance
to the present studies. Our previous findings that (i) tetra-n-
butylammonium trifluoroacetate (TBATFA) represented the only
by-product in the reaction mixture after opening of the oxirane
ring of glycidyls with trifluoroacetic acid (TFA) in the presence of
a halide anion,53 and (ii) that treatment of TBATFA and acyl-/or
silyl chlorides in chloroform with 4-DMAP resulted in powerful
acylating-/or silylating system (see ESI, Section 1), provided
additional strong rationale for adopting such an approach.
To this end, glycidyls having silyl ether (1–4) or acyl (5, 6)
residues were chosen as representative substrates with different
electronic- and steric requirements.
Scheme 1
Since the synthons I–III are accessible in high yields and under
mild reaction conditions from readily available, homochiral gly-
cidyl substrates in a way that also eradicates constraints of existing
procedures for converting either secondary alcohols into sterically
hindered silyl ethers38 or silyl-protected glycidols to C2-O-acylated
C3-vicinal halohydrins,39 the chemistry as a whole delineates a
novel strategy for total, regio- and stereocontrolled syntheses of
1,2(2,3)-DAGs (route B), 1,3-DAGs (route C), structured TAGs
(routes D or E), and cationic lipids (route F) (Scheme 1).
2. Results and discussion
Pertinent to the scope and generality of the synthetic objectives
outlined in Scheme 1, we required distinct molecular targets that
have already been synthesized in an independent way and are
known to be highly predisposed to acyl migration, transesterifica-
tion, hydrolysis, etc., to evaluate the efficiency and mildness of the
reaction conditions employed.
In light of the above, a chemically labile 1-oleoyl-2-acetyl-
(OAG) and 1-acetyl-2-oleoyl-sn-glycerol (AOG), their 1,3-sn-
isomers or triester isosters (e.g., bearing acetyl, palmitoyl and
oleoyl fragments), which are also known as exogenous effectors of
PKC40,41 and Ca2+ mobilizing agents,42 appeared as excellent can-
didates for this purpose. Another argument justifying this choice
of substrates was that identification of regioisomeric species in
Acid catalyzed opening of the oxirane system of glycidol
derivatives 1–6 with TFA (1.5 equiv.) in the presence of Bu4NX
(3.0 equiv.)53 in chloroform at room temperature left cleanly within
5–60 min the expected C3-vicinal haloalkanol (Scheme 2, step i).
To produce halohydrins of type I (compounds 7–11), the reaction
mixture was treated with the requisite acyl chloride (1.50 equiv.)
and 4-DMAP (6.0 equiv.) [or pyridine (20.0 equiv.) alternatively
applicable to this particular route only] (Scheme 2, step ii a),
or for halohydrins of type II and III (compounds 12–14 and
15–19, respectively) with a silyl chloride (3.0 equiv.) needed and
464 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 2
4-DMAP (6.0 equiv.) (Scheme 2, step ii b). This gave quantitatively
and in a strictly chemo- and regioselective way (>99%, H and
species [e.g., D-mannitol,7 D- or L-serine,55 L-glyceric acid,56
L-erythrulose,57 or sn-glycero-3-phosphocholine41] or synthetic
building blocks (e.g., glycerol acetals,15,45,58 various glycidol
derivatives,47,59 and others5,18) that can be elaborated by chemical5,16
or chemoenzymatic18,60 techniques to an enantiomerically pure
1(3)-monoacyl-sn-glycerol [1(3)-MAG] bearing a transient pro-
tection at the other primary hydroxyl function.
1
13C NMR spectroscopy) C2-acylates 7–11 (10–20 min; if pyridine
used instead, ca. 2 h) and C2-O-silyl ethers 12-19 (5 min–
1.5 h; for the reaction with TIPSCl, 24 h), which were isolated
in 90–98% yields after flash column silica gel chromatography.
The chemistry seemed to be rather general as other glycidyl
conjugates (e.g., acetyl, arachidonoyl, 4-nitrobenzoyl, hexadecyl
or isopropyl; data not shown) also underwent smooth conversion
to the analogous 2-O-functionalized vicinal chloro-, bromo-,
and iodohydrins of types I–III. Some mechanistic aspects of
these reactions were investigated and shortly discussed in the
Supplementary Information (see ESI, Table S1 and Scheme S1).
The formation of haloglycerides 7–10, 12, 13, and 15–17
with defined stereochemistry and the lack of acyl migration (or
accumulation of by-products through exchange of a halogen in the
C3-unit) are in conformity with a putative mechanism (pathway
2) described in ESI (Scheme S1).
Regioselective and stereospecific preparation of such key inter-
mediates from the above congeners is a rather tedious task due
to separation problems,45,59,61 low to moderate isolated yields,45,61,62
and extensive side reactions (e.g., acyl migration, formation of
cyclic systems, hydrolysis of an ester function, etc.)31,33,63 that
erode stereochemistry. For quaternary ammonium salt-promoted
cleavage of the epoxide ring of glycidyl substrates by carboxylic
acids,47,59 harsh reaction conditions and poor regioselectivity of
the transformations are the main setbacks of these strategies.
Contrary to these, we show in Scheme 3 that halohydrins of type
I (compounds 8 and 9, obtained from one common precursor as in
Scheme 2), bearing C2-acetate (8) or C2-oleate (9) can be efficiently
converted into isomeric diglycerides 24 and 25, respectively.
The reaction sequence commenced with substitution of iodide in
8 and 9 by the corresponding carboxylic acid moiety. This was ef-
fected by treatment of halohydrin 8 with tetra-n-butylammonium
oleate (3.0 equiv.), and 9 with tetra-n-butylammonium acetate
(3.0 equiv.), in toluene at 80 ◦C for ca. 1.5 h. 1H and 13C NMR spec-
tra of the isolated products (see Experimental) revealed that under
these conditions the conversion to the target compounds 20 vs. 21
was nearly quantitative and entirely chemo- and regioselective
(>99%;) as assessed by the comparative analytical NMR criteria
presented in ESI (Section 5). No by-products due to possible acyl
migration or dehydroiodination could be detected by means of
TLC or 1H and 13C NMR spectroscopy. In all cases, the rates of the
reactions were not appreciably affected by the structural features
the iodohydrin derivatives (i.e. 8 vs. 9) or the attacking carboxylate
(i.e. oleate vs. acetate). Also other alkyl iodides [e.g., methyl io-
dide, oleyl iodide, 1-O-hexadecyl-2-acetyl-3-iodo-rac-glycerol, or
1-iodo-2-oleoyl-3-O-tert-butyldiphenylsilyl-sn-glycerol (7); data
As the increase of steric bulk on silicon renders silyl chlo-
rides progressively less reactive towards secondary alcohols,54
the strategy depicted in Scheme 2 permits efficient silylation of
C3-vicinal haloalkanols at C2, without recourse to forceful
reaction conditions.38
Having developed a methodology for a one-pot synthesis of
halohydrins 7–19 from the corresponding oxirane educts 1–6,
the important question to be addressed next was whether satu-
rated, unsaturated and hydrolytically labile (e.g., trifluoroacetyl
or trichloroacetyl) acid-chains can be incorporated in a chemo-,
regioselective, and stereospecific manner at a given position within
the glycerol framework by acylating directly an incipient hydroxyl
group, in the presence of O-silyl-, O-acyl- and halo-functionalities.
Total synthesis of 1,2(2,1)-O-diacyl-sn-glycerols from iodohydrins
of type I (route B in Scheme 1 and Scheme 3)
Total synthesis of 1,2(2,3)-O-diacyl-sn-glycerols and deriva-
tives thereof usually makes use of naturally occurring chiral
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 465
©

View Article Online
Scheme 3
not shown] could be quantitatively converted to the corresponding
esters.
able chiral glycidols,66 both enantiomers of a 1,2- or 2,3-
O-diacylglycerol could be accessed by the protocol described
above.
Since nucleophilic substitution of the iodine atom within
these substrates could be carried out using readily available
quaternary ammonium carboxylates in a non-polar environment,
this eliminated the necessity of using costly44,51 and poorly soluble
in common organic media metal salts of carboxylic acids44,47,51
deemed as the reagents of limited synthetic utility.64
Total synthesis of 1,3-O-diacyl-sn-glycerols from iodohydrins of
type II (route C in Scheme 1 and Scheme 4)
Interchange of two acyl substituents (e.g., acetyl vs. oleoyl) at C1-
and C3-sn-sites of a glycerol unit should produce enantiomeric
pairs of the respective 1,3-O-diacyl-sn-glycerol (e.g., 32 vs. 33),
provided that no scission of a C–O bond at C2-stereocenter takes
place during the reaction sequence.
Due to high susceptibility to acyltropy of unprotected 1,2(2,3)-
diglycerides5,32 or during removal of a silyl group from their
synthetic precursors,65 we converted the silyl derivatives 20 and
21 into the corresponding trichloroacetyl derivatives (22 and 23,
respectively), that can be considered as a convenient storage form
for diglycerides.29,37 To this end the silyl ethers 20 and 21 were
treated with trichloroacetic anhydride (TCAA, 9.0 equiv.) and
Et3N·3HF (2.0 equiv.) at 80 ◦C for 2 h to effect quantitatively and
Despite its apparent triviality, the concept has not been
exploited to any significant extent as a useful approach to
stereochemically defined 1,3-DAGs, due to acute problems in
regioselective incorporation of individual acyl moieties at primary
positions of a glycerol backbone.5,16 For example, treatment of
glycidyl esters with another fatty acid in the presence of a
quaternary ammonium salt at 100–110 ◦C (2–4 h), was shown
to produce mixtures of 1,2- and 1,3-O-diacylglycerols in erratic
proportions.55 Although advocated as superior to protection–
deprotection protocols, this method and its latter modifications67
suffer from lack of generality and harsh reaction conditions that
contribute to formation of transesterification products, acyl mi-
gration, epimerization, oxidation, etc.59 One-step methodologies
based on esterification of 1-monoglycerides with various acyl
donors, seem to be equally inefficient16 and afford 1,3-diglycerides
in low yields (44–46%).25,42
1
in a highly chemo- and regiospecific fashion (>99%, H and 13C
NMR spectroscopy) a direct transformation of O-TIPS group
into trichloroacetates in 22 and 23 (Scheme 3, step ii). These
compounds were isolated in 90–92% yields and no decomposition
(1H and 13C NMR spectra; optical activity) could be detected upon
storage for several months (-20 ◦C, under argon).
In addition to a stabilizing effect to labile constructs like 22,
23, the importance of having a trichloroacetyl functionality was
that this group could be selectively removed even in the presence
of acetyl esters. Thus, treatment of trichloroacetates 22 and 23 in
THF with pyridine (50 equiv.) and methanol (500 equiv.) at room
temperature for 2 h (Scheme 3, step iii), followed by evaporation of
volatile products under reduced pressure, afforded OAG (24) and
the regioisomeric AOG (25) of purity >99% (1H and 13C NMR
spectroscopy).
These difficulties prompted us to consider 1-O-tert-
butyldimethylsilyl-2-O-triisopropylsilyl-3-iodo-sn-glycerol (13) as
a new building block (halohydrin of type II) in the synthesis of
1,3-DAGs (Scheme 4), to permit a direct incorporation of an acyl
group into any position of the glycerol skeleton.
As (R) and (S)-2-(silyloxymethyl)oxiranes with diverse steric
features at silicon are easily prepared from commercially avail-
466 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 4
◦
The first step of the reaction sequence, i.e. replacement of the
iodide group in halohydrin 13 either by tetra-n-butylammonium
acetate or oleate (3.0 equiv.) in toluene at 80 C was uneventful,
and afforded in highly chemo- and regiospecific manner (>99%,
1H and 13C NMR spectroscopy) 3-sn-monoesters 26 and 27 in
93–95% isolated yields.
Having at the C2 position in 26 and 27 an unreactive O-TIPS
functionality we attempted to introduce a second acyl group via
a selective acylation across the terminal tert-butyldimethylsilyloxy
system using Bu4NBr–TMSBr–CAA36 reagent system.
in the presence of Et3N·3HF (3.0 equiv.) at 80 C for 6 h.29,37
This furnished 2,2,2-trichloroacetates 30 and 31 (isolated yields:
91-93%) in a strictly chemo- and regioselective manner (>99%, 1H-
and 13C NMR spectroscopy). Comparison of the optical rotations
of 30 vs. 31, ([a]2D0 = -0.69 vs. [a]2D0 = +0.66), with that of the
◦
reference 1-oleoyl-2-trichloroacetyl-3-acetyl-sn-glycerol ([a]D20
=
-0.68),36 indicated retention of configuration in the C3-unit. These
findings were in agreement with a putative mechanism proposed
previously for the reactions involving primary silyloxy systems of
glycerol.37
For this purpose, a solution of monoester 26 or 27 and Bu4NBr
(2.0 equiv.) in chloroform was treated at 80 ◦C with a mixture of
oleic or acetic anhydride (3.0 equiv.) and TMSBr (1.5 equiv.) for
2 h. 1H- and 13C NMR spectra revealed that transformation of 26
and 27 into the corresponding 1,3-O-diacyl-sn-derivative 28 and
29 was practically quantitative and entirely regioselective (>99%).
No by-products due to possible intramolecular rearrangements or
side reactions involving either the olefinic part of the oleic acid or
the O-TIPS moiety at C2, were observed.
Critical to a successful synthesis of diglycerides 28 and 29 was
the fact that a more reactive silyl group (TBDMS) was at the
primary position of the glycerol system, while a less reactive one
(TIPS), at the C2 centre. Exchanging the positions of TBDMS- and
TIPS-groups within 26 or 27 (e.g., to form 1-O-triisopropylsilyl-2-
O-tert-butyldimethylsilyl-3-acetyl-/or -3-oleoyl-sn-glycerol) led to
indiscriminate acylolysis of both O-silyl protecting groups under
the reaction conditions. Similar results were obtained when 1,2-O-
bis(tert-butyldimethylsilyl)-3-stearoyl-rac-glycerol (prepared from
halohydrin 14) was used for the reaction.
As for 1,2-sn-digycerides, trichloroacetates 30 and 31 could
be efficiently converted into unprotected 1,3-sn-digycerides 32
([a]2D0 = -0.27) and 33 ([a]2D0 = +0.28), respectively, via treatment
with methanol–pyridine in THF (Scheme 3, step iv). Enantiomeric
purity of the isolated compounds was additionally confirmed
by conversion to the corresponding Mosher esters,29 and can
serve as a proof of stereochemical integrity within the entire
protocol.
In Scheme 5, a useful variant of this method for the
preparation of asymmetric of 1,3-DAGs is shown, that in-
volves derivatization of halohydrin II precursor 13 (or derived
from it 1-O-tert-butyldimethylsilyl-2-O-triisopropylsilyl-3-O-acyl-
sn-glycerols 26 and 27) to produce C3-synthons bearing a free
hydroxy-function at sn-C1 (e.g., 36-38) or halohydrins of type III
(compounds 39 and 40).
Attempted direct transformation of tert-butyldimethylsilyl into
trifluoroacetyl groups by means of a reagent system we have
recently developed [e.g., Bu4NI (2.0 equiv.)–TFAA (2.0 equiv.)]
for trifluoroacetylation of TMS68 ethers was, unfortunately,
unsuccessful due to sluggish reaction (ca. 48 h at 80 ◦C for
the completion) and formation 1,2-bis(trifluoroacetyl)-derivatives
(7–12%) from 13, 26 and 27 (1H- and 13C NMR spectroscopy).
In the next step of this synthetic protocol, we replaced O-TIPS
ethers with trichloroacetate groups (Scheme 3, step iii), by reacting
silyl ethers 28 or 29 with trichloroacetic anhydride (12.0 equiv.)
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 467
©

View Article Online
traditional chiral pool of 3,4-isopropylidene-D-mannitol7 and 1,2-
isopropylidene-sn-glycerol15 require at the least ten protection–
deprotection steps to produce a mixed-chain glyceryl triester.
While a one-pot acylolytic cleavage of esterified glycerol acetals,69
as well as a consecutive esterification of monoglycerides24,25,42,70
have been recently proposed as less arduous approaches to STGs,
these methodologies often suffer from extended reaction time (1–3
days), mediocre regioselectivity, and afford the desired compounds
usually in poor to moderate yields (7–68%). Although opening of
the epoxide ring of glycidyl esters with reagent systems consisting
of LiBr–carboxylic acid anhydrides51 or LiBr–oleic anhydride–
benzyltributylammonium bromide,44 followed by a caesium car-
boxylate effected acidolysis of bromine in the resulted 3-bromo-
1,2-propanediol diacylates, has been suggested to provide a short
entry to STGs, the chemistry involved does not prevent completely
formation of the isomeric products.44
Within the framework of the presented here synthetic method-
ology, triglycerides can be prepared from suitable diglycerides
precursors (Scheme 6), or directly from halohydrins I or III
(Scheme 7).
For the first approach, the most suitable precursor seemed to
be 1,2-O-diacyl-3-trichloroacetyl- (e.g., 22, 23) or 1,3-O-diacyl-
3-trichloroacetyl-sn-glycerols (e.g., 30, 31), easily accessible from
O-silylated glycidyl 2 or iodohydrin 13, as described above
(Scheme 3 and Scheme 4).
Introduction of the third fatty acid moiety to diglycerides
22, 23 or 30, 31, involved two reaction steps: (i) removal of a
trichloroacetate moiety with methanol–pyridine and (ii) acylation
of an unprotected diglyceride with an acyl chloride (Scheme 6).
These, produced quantitatively and in a regioselective way (>99%,
1H- and 13C NMR spectroscopy) the target triacylglycerols 41–46
in isolated 92–96%.
Scheme 5
After exploring various reaction conditions, the best results
were achieved when 13, 26 or 27 was reacted with trifluoroacetic
anhydride (12.0 equiv.) in chloroform at 70 ◦C for 2 h in the
presence of methanol (3.0 equiv.). Removal of the volatile prod-
ucts under reduced pressure provided positionally homogeneous
(purity >99%,1H- and 13C NMR spectroscopy) the corresponding
trifluoroacetyl esters without supplementary chromatography
(Scheme 5, step i). This method of trifluoroacetylation across
a silyloxy system worked well also for other TBDMS-protected
compounds [e.g., iodohydrins of type 10, 4-chloro-1-(O-TBDMS)-
butanol, etc.] or C2-O-silyl ethers (16 and 17, data no shown). The
produced trifluoroacetyl derivatives could be used as a storage
form for these precursors, or be converted into compounds 36–
38 with a free hydroxyl function (Scheme 5, step ii) by treatment
with pyridine–methanol in THF at room temperature for 30 min
(purity >99%; 1H- and 13C NMR spectroscopy).
Due to the strictly quantitative character of both incorporation
(step i) and cleavage of a trifluoroacetyl group (step ii), the
conversion of the starting materials such as 13, 26 and 27 into
the suitably modified precursors of 1,3-DAGs (e.g., 36–38 or 39,
40) could be executed as a one-pot procedure, and then subjected
to additional functionalization as shown in Scheme 5 (step iii).
It is worth noting that replacement of TFAA by trichloroacetic
anhydride (TCAA) resulted in very slow reaction (after 8 days,
conversion ~80%). For some mechanistic aspects of the tri-
fluoroacetylation reaction across the silyloxy system, see ESI
(Scheme S2).
Synthesis of mixed-acid triacylglycerols from diglyceride
precursors (route D in Scheme 1 and Scheme 6) or from
iodohydrins I and III (route E in Scheme 1 and Scheme 7)
The already discussed problems associated with a regioselective
and stereospecific incorporation of two different acyl substituents
into a glycerol unit apply also to the synthesis of structured
triacylglycerols (STGs). For example, methods relying on a
Scheme 6
468 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 7
One should note that the above approach can provide also a
convenient access to the various triglyceride derivatives bearing
a reporter or a new functional group (e.g., fluorescent, photoac-
tivable, isotopic or spin-labelled probes, phosphoester headgroups,
etc.)5,9,18,28,30 at the predetermined position of the glycerol moiety.
Synthesis of triglycerides from halohydrins type I (e.g., 10) and
III (e.g., 15) is shown in Scheme 7.
was of interest. As for the reactions involving halohydrins of
type I (Scheme 7), the formations of STGs 52 and 53 (89–92%
isolated yields) was entirely chemo- and regioselective (>99%,
1H- and 13C NMR spectroscopy). In addition, by comparing
optical rotations of 1,2-dioleoyl-3-acetyl- 52 ([a]2D0 = -0.66) and
1-oleoyl-2,3-diacetyl-sn-glycerol 53 ([a]2D0 = -1.28) with those of
the reference compounds29 and 1-acetyl-2,3-dioleoyl-sn-glycerol
44 ([a]2D0 = +0.70) (this work) we could conclude that replacement
of the silyloxy group in 49 by an acyl function (Scheme 7, step ii)
occurred with retention of configuration at C2.
Finally, we were interested in conversion of halohydrins of
type III (1-oleoyl-2-O-tert-butyldimethylsilyl-3-iodo-sn-glycerol,
15) into a useful intermediate 55 (Scheme 8), which upon
acylation with common acyl chlorides (step iii) and the subsequent
derivatization (steps iv and v in Scheme 8) would delineate a com-
plementary approach to 1,3-diacyl-sn-glycerols (e.g., compound
58) and ultimately, STGs (compound 59, Scheme 8). Relevant
to this goal, the presence of the TBDMS group at the glycerol
C2-position was believed to secure the stereochemical integrity
of 15 during displacement of the iodine (step i) and prevent acyl
migration after exposure of a free hydroxyl function in 55.
Unfortunately, somewhat unexpectedly we experienced prob-
lems in the first reaction step, namely, the replacement of
iodide in 15 by a trifluoroacetyl function. In contrast to the
tetra-n-butylammonium carboxylates investigated (Schemes 3, 4
and 7), tetra-n-butylammonium trifluoroacetate was essentially
unreactive in nucleophilic substitution in 15, and we had to use
silver trifluoroacetate as an alternative reagent.
The protocol seems to be more versatile than that making using
of diglyceride intermediates (Scheme 6) and can provide access to
enantiomerically pure triglycerides from a single chiral halohydrin
precursor. This was demonstrated in a reaction sequence in
Scheme 7 for the synthesis of enantiomeric triglycerides 50 and
51, starting from the same halohydrin I precursor (compound
10). The strategy involved (i) displacement of the iodide in 10
by means of either tetra-n-butylammonium acetate or tetra-n-
butylammonium palmitate to produce diglycerides 47 and 48 in
92–94% isolated yields, and (ii) treatment of silyl ethers 47 and
48 and Bu4NBr (2.0 equiv.) in chloroform at 80 ◦C (pressure
tube) for 2 h with, respectively, palmitic or acetic acid anhydride
(3.0 equiv.) and TMSBr (1.5 equiv.) to afford triglycerides 50
1
and 51 (Scheme 7). H- and 13C NMR analysis indicated, that
under the reaction conditions conversion of 47 and 48 to the
final products was nearly quantitative (90–93% yields after flash
column silica gel chromatography) and provided enantiomerically
pure 1-palmitoyl-2-oleoyl-3-acetyl-sn-glycerol 50 ([a]2D0 = -0.74)
and 1-acetyl-2-oleoyl-3-palmitoyl-sn-glycerol 51 ([a]2D0 = +0.76).
In the synthesis of triglycerides from halohydrins of type III
(compound 15 in Scheme 7), acylation across the silyloxy system
of glycerol effected by Bu4NBr–CAA–TMSBr36 occurred at the
C2 chiral centre, and thus also stereochemistry of this step
The reaction sequence in Scheme 8 was very efficient but
all synthesized compounds 54–58 consistently lacked any ap-
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 469
©

View Article Online
Scheme 8
preciable optical activity. Conversion of the final product 58
into the Mosher ester 59 (step vi, Scheme 8) revealed (1H and
13C NMR spectroscopy) that 58 represented virtually an equimolar
mixture of 1-oleoyl-3-acetyl-sn-glycerol and its 1-acetyl-3-oleoyl-
sn-enantiomer. The observed epimerization occurred most likely
during the replacement of iodine in 15 by silver trifluoroacetate,
probably due to a transient generation of symmetrical 1,3-
dioxonium cation and 1,3-acyloxy group migration (see, ESI
and Scheme S3 for some mechanistic details). This observation
pointed also to the advantage of using tetra-n-butylammonium
carboxylates vs. metal carboxylates44,71 in the reactions involving
nucleophilic substitution in halohydrins.
dride, (3.0 equiv.) and resulted in regioisomerically homogeneous
(>99%, 1H and 13C NMR spectroscopy) iodohydrins 60 and
61 (isolated in 92–94% yields after flash column silica gel
chromatography). Trifluoroacetic acid-catalysed opening of the
oxirane system of glycidyl oleate (5) in the presence of iodide,53
followed by treatment with palmitoyl chloride [Scheme 9, steps
i(a) and i(b)] represents a complementary strategy to a homochiral
1-oleoyl-2-palmitoyl-3-iodo-sn-glycerol 61.
As a final stage of this synthetic protocol, the obtained
O-bis(acylated) C3-iodohydrins 60 and 61 were reacted under
argon with pyridine at 80 ◦C for ca. 18 h to give the target
pyridinium cation lipids 62 and 63 in 87–90% isolated yields.
In conclusion, we have developed an efficient chemistry for
producing halohydrins of type I–III (Scheme 1) as a novel class of
protected chirons of broad interest to regio- and stereocontrolled
synthesis of diglycerides (e.g., 24, 25, 32 or 33), triglycerides
(e.g., 22, 23, 30, 31, 41–46, 50–53) and glycerol backbone-derived
cationic lipids (e.g., 62, 63).
The main advantages of halohydrins of type I–III as key
intermediates in glycerolipid synthesis are: (i) facile accessibility
from suitable glycidol derivatives, (ii) stability towards decompo-
sition or intramolecular rearrangements, (iii) possibility for highly
regioselective and stereospecific introduction of diverse acid-
chains at any position of a glycerol framework without exposure
of a free hydroxyl group, (iv) possibility to obtain enantiomeric
pairs of 1,3-DAGs and TAGs from one homochiral precursor.
This new strategy is high yielding, minimizes the number of
synthetic/purification operations, and can easily be scaled up. It
seems to be rather general and can be used for the synthesis of
glyceride derivatives of relevance to biochemical or biophysical
studies or to other applications where regioisomeric-/isosteric
forms of chiral lipid mediators are required.
A short synthesis of cationic lipids from halohydrins of type I and
III (route F in Scheme 1 and Scheme 9)
To demonstrate the synthetic utility of halohydrins as key
intermediate in the synthesis of a glycerolipid framework, we
prepared two cationic lipids from halohydrins of type I and
III. Cationic lipids based on glyceride frameworks are gain-
ing increasing importance as alternative synthetic gene deliv-
ery vectors to recombinant viruses,20,72 or as agents for bio-
chemical and medicinal intervention.73 Quaternary ammonium
salts of diacylglycerols, such as N-[1-(2,3-dioleoyloxy)propyl]-
N,N,N-trimethylammonium methyl sulfate (DOTAP)74 and its
analogues,19 have attracted particular attention in this respect. Due
to the presence of enzymatically labile acyloxy functionalities, this
class of cationic amphiphiles exhibits a better transfection profile
and reduced cytotoxicity compared to classical systems.75,76 It was
demonstrated that the presence of the iodide counterion increased
significantly transfection activity,75,77 while the replacement of an
alkyl ammonium head group by a pyridinium moiety, resulted in
lower cytotoxicity and increased tissue penetration ability.78
In Scheme 9, the synthesis of two cationic lipids, pyridinium
derivative 62 [( )-N-(1-oleoyl-2-acetyl-3-propyl)pyridinium io-
dide] and 63 [(-)-N-(1-oleoyl-2-palmitoyl-3-propyl)pyridinium io-
dide] as analogues of DOTAP, is depicted.
3. Experimental
3.1. General
The first step of the reaction sequence involved Bu4NI–
TMSI-mediated acylation of halohydrins I (compound 11) or
III (compound 16) with an appropriate carboxylic acid anhy-
All reagents were commercial grade (Fluka, Lancaster, Merck,
Sigma) with purity >98% and were used as provided without
470 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 9
further purification. Solvents were dried and distilled prior to use
according to standard protocols.79 Reaction conditions were kept
strictly anhydrous unless stated otherwise.
starting substrates 1–6 with spectral and physicochemical
parameters comparable to those reported in the literature.66,80,82
The detailed account of analytical criteria proposed for as-
1
Progress of the reactions was monitored by analytical thin-
layer chromatography on pre-coated glass plates of silica gel 60
F254 (Merck). The spots were visualized using the commercially
available 3.5% molybdatophosphoric acid spray reagent (Merck)
or 50% sulfuric acid followed by heating at 140 ◦C. Column
chromatography was carried out on silica gel 60 (70–230 mesh
ASTM, Merck). For preparative parameters of transformations
not shown in the main body text, see ESI (Section 4).
sessing regiochemistry of the prepared compounds by H and
13C NMR spectroscopy is given in ESI (Section 5: Table S2). For
1
full H and 13C NMR spectral characteristics of the compounds
synthesized, see ESI (Section 6).
3.2. General procedure for one-pot, two-step conversion of
glycidyl derivatives (1–6) into halohydrins I (7–11, 61), halohydrins
II (12–14), and halohydrins III (15–19)
1H and 13C NMR spectra were recorded on a Varian 400 MHz
machine and chemical shifts are reported in ppm relative to TMS.
The assignment of proton and carbon resonances of 1–63 was done
on the basis of known or expected chemical shifts in conjunction
with 1H–1H, 1H–13C, and DEPT correlated NMR spectroscopy. In
A solution of glycidyl substrate 1–6 (1.00 mmol) and tetra-n-
butylammonium halide (3.00 mmol) in alcohol-free chloroform
(10.0 mL) at room temperature was treated under argon with
trifluoroacetic acid (0.115 mL, 1.50 mmol) for 5 min–1 h. Then,
the corresponding either acyl chloride (1.50 mmol) or silyl chloride
(3.00 mmol) and, after ca. 5 min, 4-N,N-dimethylaminopyridine
(0.733 g, 6.00 mmol) [or pyridine (1.61 mL, 20.0 mmol) used
in the case of acylation only] were added successively and the
reaction mixture was left at room temperature for 5 min–24 h.
The solution was passed through a chloroform-filled silica gel
pad (~5 g), which was washed with the same solvent (~100 mL).
Chloroform was removed under reduced pressure and the fully-
protected C3-synthons 7–19, 61 were isolated in pure state (>99%,
1H NMR spectroscopy) by flash column silica gel chromatography
(eluent for 7, 9, 10: toluene–pentane = 80 : 20, v/v; eluent for 8, 11:
toluene–EtOAc = 98 : 2, v/v; eluent for 12–14: pentane–toluene =
80 : 20, v/v; eluent for 15: toluene–pentane = 50 : 50, v/v; eluent
for 16-19, 61: toluene).
1
certain instances, only the most informative parts of H and 13C
NMR spectra of known compounds,29,68,80 were provided for the
sake of convenience.
Optical rotations were measured on a Perkin-Elmer 241 digital
polarimeter. The melting points were determined on a Kofler
melting point apparatus and are uncorrected.
Tetra-n-butylammonium acetate was purchased from Aldrich.
Tetra-n-butylammonium palmitate, tetra-n-butylammonium
oleate, and tetra-n-butylammonium trifluoroacetate were
synthesized from tetra-n-butylammonium hydroxide and the
corresponding carboxylic acids (all from Fluka) following routine
approaches.71,81
(R)-(+)-2-(tert-Butyldiphenylsilyloxymethyl)oxirane 1, (R)-(+)-
2-(triisopropylsilyloxymethyl)oxirane 2, (S)-(+)-2-(tert-butyl-
dimethylsilyloxymethyl)oxirane 3, (rac)-( )-2-(tert-butyldime-
thylsilyloxymethyl)oxirane 4, (S)-(+)-2-(oleoyloxymethyl)oxirane
5, and (rac)-( )-2-(oleoyloxymethyl)oxirane 6 were prepared
from chiral or racemic glycidols (all from Fluka) analogously to
conventional methods66 or as described elsewhere.80 No attempts
were made to optimize these particular procedures that afforded
1-Iodo-2-acetyl-3-O-triisopropylsilyl-sn-glycerol 8. Obtained
from (R)-(+)-2-(triisopropylsilyloxymethyl)oxirane (2; 0.230 g,
1.00 mmol) using Bu4NI (1.108 g, 3.00 mmol), acetyl chloride
(0.107 mL, 1.50 mmol) and pyridine (reaction times, stage I: 5 min;
stage II: 1.5 h). Yield: 0.380 g (95%, colorless oil); Rf (pentane–
toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.73; [a]2D0 = +9.82 (c 11.19,
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 471
©

View Article Online
CHCl3); Found: C, 40.96; H, 7.33; I, 31.75%. C14H29IO3Si (400.37)
requires C, 42.00; H, 7.30; I, 31.70%.
3.3. Typical procedure for the transformation of halohydrins I
(8–10), halohydrin II (13) and halohydrin III (15) to terminal
O-acylates (20, 21, 26, 27, 47–49, 54)
1-Iodo-2-oleoyl-3-O-triisopropylsilyl-sn-glycerol 9. Obtained
from (R)-(+)-2-(triisopropylsilyloxymethyl)oxirane (2; 0.230 g,
1.00 mmol) using Bu4NI (1.108 g, 3.00 mmol), oleoyl chloride
(0.496 mL, 1.50 mmol) and pyridine (reaction times, stage I: 5 min;
stage II: 2 h). Yield: 0.579 g (93%, colorless oil); Rf (toluene–
pentane = 80 : 20, v/v) = 0.88; [a]2D0 = +5.77 (c 10.15, CHCl3);
Found: C, 57.93; H, 9.51; I, 20.41%. C30H59IO3Si (622.78) requires
C, 57.86; H, 9.55; I, 20.38%.
To a solution of the requisite iodohydrin 8–10, 13 or 15 (1.00 mmol)
in anhydrous toluene (7.0 mL) was added the corresponding
tetra-n-butylammonium carboxylate (3.00 mmol) and the reaction
system was stirred under argon in a pressure-proof glass ampoule
at 80 ◦C (bath) for 1.5 h. The solution was passed through a short
silica gel pad (~5 g), the support was washed with chloroform
(~150 mL) and the solvents were evaporated in vacuum. The target
esters 20, 21, 26, 27, 47–49 were isolated in pure state (>99%,
1H NMR spectroscopy) by flash column silica gel chromatography
(mobile phase for 20, 21, 47–49: toluene–EtOAc, 98 : 2, v/v; mobile
phase for 26: toluene; mobile phase for 27: pentane–toluene:
50 : 50, v/v).
1-O-tert-Butyldimethylsilyl-2-oleoyl-3-iodo-sn-glycerol
10.
Obtained from (S)-(+)-2-(tert-butyldimethylsilyloxymethyl)-
oxirane (3; 0.188 g, 1.00 mmol) using Bu4NI (1.108 g,
3.00 mmol), oleoyl chloride (0.496 mL, 1.50 mmol) and 4-
DMAP (reaction times, stage I: 5 min; stage II: 10 min). Yield:
0.545 g (94%, colorless oil); Rf (toluene–pentane = 80 : 20, v/v) =
0.73; [a]2D0 = -3.21 (c 7.88, CHCl3); Found: C, 55.93; H, 9.17;
I, 21.90%. C27H53IO3Si (580.70) requires C, 55.84; H, 9.20; I,
21.85%.
Using silver trifluoroacetate (3.00 mmol) instead, the trifluo-
roacetyl ester 54 of 1-oleoyl-2-O-tert-butyldimethylsilyl-3-iodo-sn-
glycerol 15 was obtained in a similar way as described below.
1-Oleoyl-2-acetyl-3-O-triisopropylsilyl-sn-glycerol
20. Pre-
pared from 1-iodo-2-acetyl-3-O-triisopropylsilyl-sn-glycerol (8;
0.400 g, 1.00 mmol) and tetra-n-butylammonium oleate (1.57 g,
3.00 mmol). Yield: 0.511 g (92%, colorless oil); Rf (pentane–
toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.60; [a]2D0 = +13.45 (c
3.67, CHCl3); lit.29 [a]D20 = +11.28 (c 9.87, CHCl3); Found: C,
69.31; H, 11.24%. C32H62O5Si (554.92) requires C, 69.26; H,
11.26%.
1-O-tert-Butyldimethylsilyl-2-O-triisopropylsilyl-3-iodo-sn-gly-
cerol 13. Obtained from (S)-(+)-2-(tert-butyldimethylsily-
loxymethyl)oxirane (3; 0.188 g, 1.00 mmol) using Bu4NI (1.108 g,
3.00 mmol), TIPSCl (0.636 mL, 3.00 mmol) and 4-DMAP
(reaction times, stage I: 5 min; stage II: 24 h). Yield: 0.434 g (92%,
colorless oil); Rf (pentane) = 0.29; [a]2D0 = +1.44 (c 10.34, CHCl3);
Found: C, 45.83; H, 8.70; I, 26.90%. C18H41IO2Si2 (472.59) requires
C, 45.75; H, 8.74; I, 26.85%.
1-Acetyl-2-oleoyl-3-O-triisopropylsilyl-sn-glycerol 21. Acqu-
ired by treatment of 1-iodo-2-oleoyl-3-O-triisopropylsilyl-sn-
glycerol (9; 0.623 g, 1.00 mmol) with tetra-n-butylammonium
acetate (0.90 g, 3.00 mmol). Yield: 0.508 g (92%, colorless oil);
1-Oleoyl-2-O-tert-butyldimethylsilyl-3-iodo-sn-glycerol
15.
Rf (pentane–toluene–EtOAc, 40 : 50 : 10, v/v/v) = 0.58; [a]D20
=
Obtained from (S)-(+)-2-(oleoyloxymethyl)oxirane (5; 0.338 g,
1.00 mmol) using Bu4NI (1.108 g, 3.00 mmol), TBDMSCl
(0.452 g, 3.00 mmol) and 4-DMAP (reaction times, stage I:
5 min; stage II: 1.5 h). Yield: 0.552 g (95%, colorless oil); Rf
(pentane–toluene = 80 : 20, v/v) = 0.71; [a]2D0 = +6.44 (c 12.79,
CHCl3); Found: C, 55.77; H, 9.22; I, 21.81%. C27H53IO3Si (580.70)
requires C, 55.84; H, 9.20; I, 21.85%.
+15.72 (c 11.38, CHCl3); Found: C, 69.22; H, 11.30%. C32H62O5Si
(554.92) requires C, 69.26; H, 11.26%.
1-O-tert-Butyldimethylsilyl-2-O-triisopropylsilyl-3-acetyl-sn-
glycerol 26. Obtained from 1-O-tert-butyldimethylsilyl-2-O-
triisopropylsilyl-3-iodo-sn-glycerol (13; 0.473 g, 1.00 mmol) and
tetra-n-butylammonium acetate (0.90 g, 3.00 mmol). Yield: 0.384 g
(95%, colorless oil); Rf (toluene) = 0.40; [a]2D0 = -25.51 (c
10.11, CHCl3); Found: C, 59.59; H, 10.83%. C20H44O4Si2 (404.73)
requires C, 59.35; H, 10.96%.
1-Oleoyl-2-O-triethylsilyl-3-iodo-sn-glycerol
16. Obtained
from (S)-(+)-2-(oleoyloxymethyl)oxirane (5; 0.338 g, 1.00 mmol)
using Bu4NI (1.108 g, 3.00 mmol), triethylsilyl chloride (0.316 mL,
3.00 mmol) and 4-DMAP (reaction times, stage I: 5 min; stage II:
5 min). Yield: 0.540 g (93%, colorless oil); Rf (pentane–toluene–
EtOAc, 40 : 50 : 10, v/v/v) = 0.89; [a]2D0 = +4.94 (c 10.58, CHCl3);
Found: C, 55.90; H, 9.17; I, 21.89%. C27H53IO3Si (580.70) requires
C, 55.84; H, 9.20; I, 21.85%.
1-O-tert-Butyldimethylsilyl-2-O-triisopropylsilyl-3-oleoyl-sn-
glycerol 27. Obtained from 1-O-tert-butyldimethylsilyl-2-O-
triisopropylsilyl-3-iodo-sn-glycerol (13; 0.473 g, 1.00 mmol) and
tetra-n-butylammonium oleate (1.57 g, 3.00 mmol). Yield: 0.583 g
(93%, colorless oil); Rf (pentane–toluene: 50 : 50, v/v) = 0.61;
[a]2D0 = -15.65 (c 9.06, CHCl3); Found: C, 69.15; H, 11.79%.
C36H74O4Si2 (627.14) requires C, 68.94; H, 11.89%.
1-Oleoyl-2-palmitoyl-3-iodo-sn-glycerol 61. Obtained from
(S)-(+)-2-(oleoyloxymethyl)oxirane (5; 0.338 g, 1.00 mmol) using
Bu4NI (1.108 g, 3.00 mmol), palmitoyl chloride (0.453 mL,
1.50 mmol) and 4-DMAP (reaction times, stage I: 5 min; stage II:
10 min). Yield: 0.683 g (97%, colorless oil); Rf (pentane–toluene–
EtOAc, 40 : 50 : 10, v/v/v) = 0.80; [a]2D0 = +3.56 (c 12.63, CHCl3);
lit.68 (for 1-iodo-2-palmitoyl-3-oleoyl-sn-glycerol) [a]2D0 = -3.62 (c
10.28, CHCl3); Found: C, 62.95; H, 9.90; I, 18.10%. C37H69IO4
(704.85) requires C, 63.05; H, 9.87; I, 18.00%.
1-O-tert-Butyldimethylsilyl-2-oleoyl-3-acetyl-sn-glycerol
47.
Obtained from 1-O-tert-butyldimethylsilyl-2-oleoyl-3-iodo-sn-
glycerol (10; 0.581 g, 1.00 mmol) and tetra-n-butylammonium
acetate (0.90 g, 3.00 mmol). Yield: 0.482 g (94%, colorless oil);
Rf (pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.58; [a]D20
=
-12.60 (c 8.21, CHCl3); Found: C, 67.69; H, 11.06%. C29H56O5Si
(512.85) requires C, 67.92; H, 11.01%.
472 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
1-O-tert-Butyldimethylsilyl-2-oleoyl-3-palmitoyl-sn-glycerol 48.
Obtained from 1-O-tert-butyldimethylsilyl-2-oleoyl-3-iodo-sn-
glycerol (10; 0.581 g, 1.00 mmol) and tetra-n-butylammonium
palmitate (1.49 g, 3.00 mmol). Yield: 0.652 g (92%, colorless oil);
C, 55.27; H, 7.60; Cl, 19.57%. C25H41Cl3O6 (543.95) requires C,
55.20; H, 7.60; Cl, 19.55%.
1-Acetyl-2-trichloroacetyl-3-oleoyl-sn-glycerol 31. Obtained
from
1-acetyl-2-O-triisopropylsilyl-3-oleoyl-sn-glycerol
(29;
Rf (pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.68; [a]D20
=
0.555 g, 1.00 mmol), trichloroacetic anhydride (2.192 mL,
12.00 mmol) and triethylamine tris(hydrofluoride) (0.488 mL,
-7.03 (c 3.08, CHCl3); Found: C, 72.95; H, 11.87%. C43H84O5Si
(709.22) requires C, 72.82; H, 11.94%.
3.00 mmol) for 6 h. Yield: 0.506 g (93%, colorless oil); [a]D20
+0.66 (c 8.33, CHCl3); all other physicochemical and spectral
=
1-Oleoyl-2-O-tert-butyldimethylsilyl-3-trifluoroacetyl-rac-gly-
cerol 54. To a solution of 1-oleoyl-2-O-tert-butyldimethylsilyl-
3-iodo-sn-glycerol (15; 0.581 g, 1.00 mmol) in anhydrous toluene
(6.0 mL) was added silver trifluoroacetate (0.663 g, 3.00 mmol) and
the reaction mixture was stirred under argon at 60 ◦C for 2 h in a
tightly stoppered glass ampoule. The solution was passed through
a silica gel pad (~5 g) prepared in the same solvent. The support
was washed with toluene (~150 mL) and fractions containing the
target product were combined. Evaporation of the solvent under
reduced pressure gave the title compound 54 (0.538 g, 95%) as a
colorless oil. Rf (pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) =
0.80; [a]2D0 = 0.00 (c 8.39, CHCl3); Found: C, 61.02; H, 9.49%.
C29H53F3O5Si (566.81) requires C, 61.45; H, 9.43%.
characteristics were identical with those of the previous product.
1-Oleoyl-2-trichloroacetyl-3-acetyl-rac-glycerol 57. Obtained
from
1-oleoyl-2-O-tert-butyldimethylsilyl-3-acetyl-rac-glycerol
(56; 0.513 g, 1.00 mmol), trichloroacetic anhydride (2.192 mL,
12.00 mmol) and triethylamine tris(hydrofluoride) (0.488 mL,
3.00 mmol) for 2.5 h. Yield: 0.504 g (93%, colorless oil); excluding
the lack of optical activity, all other physicochemical and spectral
characteristics were identical with those of compounds 30 and 31.
3.5. Typical procedure for the direct functionalization of silyl
ethers (26, 27, 47–49, 11, and 16) to fatty acid esters (28, 29,
50–53, 60, and 61)
To a solution of the silyl ether 26, 27, 47–49, 11 or 16
(1.00 mmol) and tetra-n-butylammonium bromide (Bu4NBr;
0.645 g, 2.00 mmol) [or tetra-n-butylammonium iodide (Bu4NI;
0.739 g, 2.00 mmol)] in alcohol-free chloroform (3.0 mL), a
mixture of the appropriate carboxylic acid anhydride (3.00 mmol)
and trimethylbromosilane (TMSBr; 0.195 mL, 1.50 mmol) [or
trimethyliodosilane (TMSI; 0.204 mL, 1.50 mmol)] in the same
solvent (3.0 mL) was added, and the reaction system was
kept under argon, in a pressure-proof glass ampoule at 80 ◦C
(bath) for 30 min–7 h. Chloroform was removed under reduced
pressure and the residue was subjected to flash column silica gel
chromatography (mobile phase: toluene–EtOAc = 98 : 2, v/v) to
give the expected acylates 28, 29, 50–53, 60, and 61 (purity >99%,
1H NMR spectroscopy).
3.4. Typical procedure for the direct conversion of silyl ethers (20,
21, 28, 29, and 56) into trichloroacetates (22, 23, 30, 31, and 57)
A mixture of the starting silyl ether 20, 21, 28, 29 or 56
(1.00 mmol), neat trichloroacetic anhydride (9.00–12.00 mmol)
and triethylamine tris(hydrofluoride) (2.00–3.00 mmol) was kept
under argon, in a pressure-proof glass ampoule at 80 ◦C (bath)
for 2–6 h. The system was taken in toluene–EtOAc (98 : 2, v/v;
5 mL) and the trichloroacetyl derivatives 22, 23, 30, 31, and 57
1
were isolated in pure state (>99%, H NMR spectroscopy) by
flash column silica gel chromatography (mobile phase: toluene–
EtOAc = 98 : 2, v/v).
1-Oleoyl-2-acetyl-3-trichloroacetyl-sn-glycerol 22. Obtained
from
1-oleoyl-2-acetyl-3-O-triisopropylsilyl-sn-glycerol
(20;
0.555 g, 1.00 mmol), trichloroacetic anhydride (1.644 mL,
9.00 mmol) and triethylamine tris(hydrofluoride) (0.326 mL,
2.00 mmol) for 2 h. Yield: 0.500 g (92%, colorless oil); Rf
1-Oleoyl-2-O-triisopropylsilyl-3-acetyl-sn-glycerol
28. Ob-
tained from 1-O-tert-butyldimethylsilyl-2-O-triisopropylsilyl-3-
acetyl-sn-glycerol (26; 0.405 g, 1.00 mmol), oleic anhydride
(1.641 g, 3.00 mmol), Bu4NBr, and TMSBr for 2 h. Yield:
(pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.58; [a]D20
=
-0.41 (c 5.17, CHCl3); lit.29 [a]D20 = -0.40 (c 7.18, CHCl3); Found:
C, 55.29; H, 7.52; Cl, 19.70%. C25H41Cl3O6 (543.95) requires C,
55.20; H, 7.60; Cl, 19.55%.
0.510
g (92%, colorless oil); Rf (pentane–toluene–EtOAc:
40 : 50 : 10, v/v/v) = 0.56; [a]2D0 = -1.20 (c 18.05, CHCl3); Found:
C, 69.05; H, 11.30%. C32H62O5Si (554.92) requires C, 69.26; H,
11.26%.
1-Acetyl-2-oleoyl-3-trichloroacetyl-sn-glycerol 23. Obtained
from
1-acetyl-2-oleoyl-3-O-triisopropylsilyl-sn-glycerol
(21;
1-Acetyl-2-O-triisopropylsilyl-3-oleoyl-sn-glycerol
29. Ob-
0.555 g, 1.00 mmol), trichloroacetic anhydride (1.644 mL,
9.00 mmol) and triethylamine tris(hydrofluoride) (0.326 mL,
2.00 mmol) for 2 h. Yield: 0.490 g (90%, colorless oil); Rf
tained from 1-O-tert-butyldimethylsilyl-2-O-triisopropylsilyl-3-
oleoyl-sn-glycerol (27; 0.627 g, 1.00 mmol), acetic anhydride
(0.284 mL, 3.00 mmol), Bu4NBr, and TMSBr for 2 h. Yield:
0.527 g (95%, colorless oil); [a]2D0 = +1.14 (c 17.03, CHCl3); all
other physicochemical and spectral characteristics were identical
with those of the previous product.
(pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.51; [a]D20
=
+1.78 (c 10.31, CHCl3); Found: C, 55.25; H, 7.56; Cl, 19.60%.
C25H41Cl3O6 (543.95) requires C, 55.20; H, 7.60; Cl, 19.55%.
1-Oleoyl-2-trichloroacetyl-3-acetyl-sn-glycerol 30. Obtained
1-Palmitoyl-2-oleoyl-3-acetyl-sn-glycerol 50. Obtained from
from
1-oleoyl-2-O-triisopropylsilyl-3-acetyl-sn-glycerol
(28;
1-O-tert-butyldimethylsilyl-2-oleoyl-3-acetyl-sn-glycerol
(47;
0.555 g, 1.00 mmol), trichloroacetic anhydride (2.192 mL,
12.00 mmol) and triethylamine tris(hydrofluoride) (0.488 mL,
3.00 mmol) for 6 h. Yield: 0.495 g (91%, colorless oil); Rf
0.513 g, 1.00 mmol), palmitic anhydride (1.484 g, 3.00 mmol),
Bu4NBr, and TMSBr for 2 h. Yield: 0.592 g (93%, colorless oil);
Rf (pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.54; [a]D20
=
(pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.50; [a]D20
=
-0.74 (c 4.13, CHCl3); Found: C, 73.50; H, 11.43%. C39H72O6
-0.69 (c 9.15, CHCl3); lit.29 [a]D20 = -0.68 (c 9.77, CHCl3); Found:
(636.98) requires C, 73.54; H, 11.39%.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 473
©

View Article Online
[a]D20 = -5.47 (c, 4.82, CHCl3); lit.29 [a]2D0 = -5.42 (c, 5.07, CHCl3);
Found: C, 69.38; H, 10.58%. C23H42O5 (398.58) requires C, 69.31;
H, 10.62%.
1-Acetyl-2-oleoyl-3-palmitoyl-sn-glycerol 51. Prepared from
1-O-tert-butyldimethylsilyl-2-oleoyl-3-palmitoyl-sn-glycerol (48;
0.709 g, 1.00 mmol), acetic anhydride (0.284 mL, 3.00 mmol),
Bu4NBr, and TMSBr for 2 h. Yield: 0.574 g (90%, colorless
oil); [a]2D0 = +0.76 (c 3.12, CHCl3); lit.29 [a]2D0 = +0.72 (c 11.73,
CHCl3); all other physicochemical and spectral characteristics
were identical with those of 50.
1-Acetyl-2-oleoyl-sn-glycerol 25. Obtained from 1-acetyl-2-
oleoyl-3-trichloroacetyl-sn-glycerol (23) for 2 h. Yield: 0.399 g
(100%, colorless oil); Rf (toluene–EtOAc = 80 : 20, v/v) = 0.27;
[a]2D0 = -1.62 (c 6.53, CHCl3); Found: C, 69.51; H, 10.62%.
C23H42O5 (398.58) requires C, 69.31; H, 10.62%.
1-Oleoyl-2-acetyl-3-iodo-rac-glycerol 60. Obtained from 1-O-
tert-butyldimethylsilyl-2-acetyl-3-iodo-rac-glycerol (11; 0.358 g,
1.00 mmol), oleic anhydride (1.641 g, 3.00 mmol), Bu4NI, and
TMSI for 2 h. Yield: 0.467 g (92%, colorless oil); Rf (pentane–
toluene–EtOAc, 40 : 50 : 10, v/v/v) = 0.67; Found: C, 54.43; H,
8.11; I, 24.80%. C23H41IO4 (508.47) requires C, 54.33; H, 8.13; I,
24.96%.
1-Oleoyl-3-acetyl-sn-glycerol 32. Obtained from 1-oleoyl-2-
trichloroacetyl-3-acetyl-sn-glycerol (30) for 2 h. Yield: 0.397 g
(100%, colorless oil); Rf (toluene–EtOAc = 80 : 20, v/v) = 0.32;
[a]D20 = -0.27 (c 10.85, CHCl3); lit.29 [a]D20 = -0.28 (c 9.15, CHCl3);
Found: C, 69.25; H, 10.70%. C23H42O5 (398.58) requires C, 69.31;
H, 10.62%.
3.6. Typical procedure for the direct transformation of silyl
ethers (26, 27) into trifluoroacetates (34, 35) [Scheme 5: step (i)]
1-Acetyl-3-oleoyl-sn-glycerol 33. Obtained from 1-acetyl-2-
trichloroacetyl-3-oleoyl-sn-glycerol (31) for 2 h. Yield: 0.399 g
(100%, colorless oil); [a]2D0 = +0.28 (c 9.05, CHCl3); all other
physicochemical and spectral characteristics were identical with
those of 32.
To a solution of silyl ether 26 or 27 (1.00 mmol) and trifluoroacetic
anhydride (1.67 mL, 12.00 mmol) in chloroform (3.0 mL),
methanol (0.122 mL, 3.0 mmol) was added and the reaction system
was kept under argon in a pressure-proof glass ampoule at 70 ◦C
(bath) for 2 h. Solvents were evaporated under reduced pressure
to give the target compounds 34 and 35 (purity: >99%, 1H NMR
spectroscopy) in a straightforward manner.
2-O-Triisopropylsilyl-3-oleoyl-sn-glycerol 37. Prepared from
1-trifluoroacetyl-2-O-triisopropylsilyl-3-oleoyl-sn-glycerol
(35;
0.609 g, 1.00 mmol) for 30 min. Yield: 0.511 g (100%, colorless
oil); Rf (pentane–toluene–EtOAc, 40 : 50 : 10, v/v/v) = 0.44;
[a]2D0 = +9.01 (c 10.05, CHCl3); Found: C, 70.48; H, 11.60%.
C30H60O4Si (512.88) requires C, 70.25; H, 11.79%.
1-Trifluoroacetyl-2-O-triisopropylsilyl-3-acetyl-sn-glycerol 34
Obtained from 1-O-tert-butyldimethylsilyl-2-O-triisopropylsilyl-
3-acetyl-sn-glycerol (26; 0.405 g, 1.00 mmol). Yield: 0.386 g (100%,
2-O-Triisopropylsilyl-3-iodo-sn-glycerol 38. To
a solution
of
1-O-tert-butyldimethylsilyl-2-O-triisopropylsilyl-3-iodo-sn-
pale yellowish oil); Rf (toluene–EtOAc, 98 : 2, v/v) = 0.39; [a]D20
=
glycerol (13; 0.473 g, 1.00 mmol) in chloroform (3.0 mL)
were consecutively added trifluoroacetic anhydride (1.67 mL,
12.00 mmol) and methanol (0.122 mL, 3.0 mmol), and the
+11.77 (c 14.82, CHCl3); Found: C, 49.83; H, 7.53%. C16H29F3O5Si
(386.48) requires C, 49.72; H, 7.56%.
◦
reaction system was heated at 70 C for 2 h (pressure tube). The
1-Trifluoroacetyl-2-O-triisopropylsilyl-3-oleoyl-sn-glycerol 35.
Obtained from 1-O-tert-butyldimethylsilyl-2-O-triisopropylsilyl-
3-oleoyl-sn-glycerol (27; 0.627 g, 1.00 mmol). Yield: 0.608 g (100%,
pale yellowish oil); Rf (pentane–toluene: 50 : 50, v/v) = 0.55;
[a]2D0 = +9.12 (c 15.73, CHCl3); Found: C, 63.24; H, 9.70%.
C32H59F3O5Si (608.89) requires C, 63.12; H, 9.77%.
solvents were distilled off, the residue was taken in tetrahydrofuran
(5.0 mL) and the solution was treated at room temperature for
30 min with a mixture of pyridine (4.0 mL, 50 mmol) and
methanol (20.3 mL, 500 mmol). Evaporation of the solvents
under reduced pressure afforded the title compound 38. Yield:
0.357 g (100%, pale yellowish oil); Rf (pentane–toluene–EtOAc =
40 : 50 : 10, v/v/v) = 0.55; [a]2D0 = +11.25 (c 8.81, CHCl3); Found:
C, 40.30; H, 7.51; I, 35.50%. C12H27IO2Si (358.33) requires C,
40.22; H, 7.59; I, 35.41%.
3.7. Typical procedure for removal of trichloroacetyl- (22, 23, 30,
31, and 57) and trifluoroacetyl groups (34, 35 and 54) to produce
alcohols (24, 25, 32, 33, 36–38, 55, and 58)
To a solution of either trichloroacetyl- 22, 23, 30, 31, or 57 (0.544 g,
1.00 mmol) or trifluoroacetyl derivative 34, 35 or 54 (1.00 mmol) in
tetrahydrofuran (5.0 mL), a mixture of pyridine (4.0 mL, 50 mmol)
and methanol (20.3 mL, 500 mmol) was added and the reaction
system was left at room temperature for 30 min–2 h. Solvents were
evaporated under reduced pressure (bath temp. 50 ◦C) and the
residue was kept under high vacuum at room temperature for 2–
3 h to give the deprotected compounds 24, 25, 32, 33, 36-38, 55,
and 58 directly (purity >99%, 1H NMR spectroscopy).
2-O-Triisopropylsilyl-3-iodo-sn-glycerol 38 was obtained in a
similar way by following the one-pot, two-step procedure de-
scribed bellow.
3.8. Typical procedure for one-pot conversion of trichloroacetyl-
(22, 23, 30, 31, and 57) and trifluoroacetyl derivatives (54) into the
corresponding fatty acid esters (39–46, 56 and 59)
First stage. A solution of 22, 23, 30, 31, 54 or 57 (1.00 mmol)
in tetrahydrofuran (5.0 mL) was treated at room temperature with
a mixture of pyridine (4.0 mL, 50 mmol) and methanol (20.3 mL,
500 mmol) for 30 min–2 h, and the volatile materials were removed
under reduced pressure.
Second stage. The residue containing the deprotected alcohol
species was dissolved in alcohol-free chloroform (10.0 mL)
containing pyridine (1.61 mL, 20.0 mmol), and the mixture was
reacted at -20 ◦C with a solution of a requisite acyl chloride
(2.00 mmol) in alcohol-free chloroform (10.0 mL). After keeping
the reaction system at room temperature for 2–3.5 h, the solution
1-Oleoyl-2-acetyl-sn-glycerol 24. Produced from 1-oleoyl-2-
acetyl-3-trichloroacetyl-sn-glycerol (22) for 2 h. Yield: 0.397 g
(100%, colorless oil); Rf (toluene–EtOAc = 80 : 20, v/v) = 0.26;
474 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
was passed through a chloroform-filled silica gel pad (~5 g),
which was washed with the same solvent (~100 mL). Chloroform
was removed under reduced pressure and the thus acylated
(-)-N-(1-Oleoyl-2-palmitoyl-3-propyl)pyridinium iodide 63.
Acquired from 1-oleoyl-2-palmitoyl-3-iodo-sn-glycerol (61;
0.705 g, 1.0 mmol). Yield: 0.682 g (87%, yellowish solid); mp:
83.6–85.0 ◦C (from diethyl ether); Rf (CH2Cl2–MeOH–H2O,
65 : 25 : 4, v/v/v) = 0.74; [a]2D0 = -16.26 (c 3.26, CHCl3); Found:
C, 64.44; H, 9.48; N, 1.81%. C42H74INO4 (783.95) requires C,
64.35; H, 9.51; N, 1.79%.
1
products 41–46, 56 were isolated in pure state (>99%, H NMR
spectroscopy) by flash column silica gel chromatography (mobile
phase: toluene–EtOAc = 98 : 2, v/v).
1-Acetyl- (39) and 1-oleoyl-2-O-triisopropylsilyl-3-iodo-sn-
glycerol (40) were synthesized analogously by
a three-
step, one-pot derivatization of 1-O-tert-butyldimethylsilyl-2-O-
triisopropylsilyl-3-iodo-sn-glycerol (13) (see below).
The Mosher ester (59) of 1,3-diglyceride (58) was acquired from
C2-trichloroacetate (57) in a slightly modified manner concerning
the purification process only.
Acknowledgements
Financial support from the Swedish Natural Science Research
Council and the Swedish Foundation for Strategic Research is
gratefully acknowledged.
1-Oleoyl-2-O-triisopropylsilyl-3-iodo-sn-glycerol 40. Synthe-
References
sized in
a
one-pot, three-step procedure from 1-O-tert-
(13;
butyldimethylsilyl-2-O-triisopropylsilyl-3-iodo-sn-glycerol
1 G. S. Denyer, Med. J. Aust., 2002, 176(Suppl.), 109–110; A. H. Merrill,
Nutrition Rev., 1989, 47, 161–169.
2 K. Parang, L. I. Wiebe and E. E. Knaus, Curr. Med. Chem., 2000,
7, 995–1039; D. W. Will and T. Brown, Tetrahedron Lett., 1992, 33,
2729–2732.
3 H. M. VonMinden, M. Morr, G. Milkereit, E. Heinz and V. Vill,
Chem. Phys. Lipids, 2002, 114, 55–80; P. Ferraboschi, D. Colombo,
F. Compostella and S. Reza-Elahi, Synlett, 2001, 1379–1382.
4 N. M. Carballeira, Prog. Lipid Res., 2002, 41, 437–456.
5 F. Paltauf and A. Hermetter, Prog. Lipid Res., 1994, 33, 239–328.
6 B. T. Golding, A. L. Griffin and D. H. Robinson, Tetrahedron Lett.,
1993, 34, 6459–6462.
0.473 g, 1.00 mmol) and oleoyl chloride (0.66 mL, 2.00 mmol)
(stage III: r.t./2 h), as described for 39. Subsequent flash column
silica gel chromatography (mobile phase: pentane–toluene =
50 : 50, v/v) gave the title compound 40 (purity >99%, 1H
NMR spectroscopy). Yield: 0.585 g (94%, colorless oil); Rf
(pentane–toluene = 50 : 50, v/v) = 0.69; [a]2D0 = +8.12 (c 6.03,
CHCl3); Found: C, 58.01; H, 9.42; I, 20.45%. C30H59IO3Si (622.78)
requires C, 57.86; H, 9.55; I, 20.38%.
1-Oleoyl-2-[R-(-)-a-methoxy-a-trifluoromethylphenylacetyl]-3-
acetyl-rac-glycerol 59. Acquired according to the typical
procedure from 1-oleoyl-2-trichloroacetyl-3-acetyl-rac-glycerol
(57; 0.544 g, 1.00 mmol) via 58 and R-(-)-a-methoxy-a-
trifluoromethylphenylacetyl chloride (0.374 mL, 2.00 mmol) (stage
I; 2 h; stage II: 20 h) with the exception that after removing the
solvents, the residue was taken in toluene–EtOAc (98 : 2, v/v,
5 mL) and the solution was passed through a silica gel pad
(~5 g) prepared in the same solvent system. The support was
washed with this eluent (50 mL), fractions containing the product
were combined and the solution was concentrated under reduced
pressure to afford the crude Mosher ester 59, which was examined
7 H. Eibl and P. Woolley, Chem. Phys. Lipids, 1988, 47, 47–53.
8 J. P. Weichert, M. A. Longino, D. A. Bakan, M. G. Spigarelli, T. Chou,
S. W. Schwendner and R. E. Counsell, J. Med. Chem., 1995, 38, 636–
646; B. Gallez, R. Demeure, R. Debuyst, D. Leonard, F. Dejehet and
P. Dumont, Magn. Reson. Imaging, 1992, 10, 445–455.
9 S. Bhattacharya and A. Bajaj, Curr. Opin. Chem. Biol., 2005, 9, 647–
655.
10 V. E. Marquez, K. Nacro, S. Benzaria, J. Lee, R. Sharma, K. Teng,
G. W. A. Milne, B. Bienfait, S. Wang, N. E. Lewin and P. M.
Blumberg, Pharmacol. Ther., 1999, 82, 251–261; J. A. Boomer and D. H.
Thompson, Chem. Phys. Lipids, 1999, 99, 145–153; E. L. Arrese and
M. A. Wells, J. Lipid Res., 1997, 38, 68–76; J. N. Jacob, G. W. Hesse and
V. E. Shashoua, J. Med. Chem., 1990, 33, 733–736; H. K. Mangold,
F. Paltauf, (Ed.) Ether Lipids: Biochemical and Biomedical Aspects,
Academic Press, New York, 1983; A. Brohult, J. Brohult and S. Brohult,
Experientia, 1973, 29, 81–82.
1
next by H and 13C NMR without supplementary purification.
11 M. J. Caloca, M. L. Garcia-Bermejo, P. M. Blumberg, N. E. Lewin, E.
Kremmer, H. Mischak, S. Wang, K. Nacro, B. Bienfait, V. E. Marquez
and M. G. Kazanietz, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 11854–
11859; E. Oancea and T. Meyer, Cell Death and Differentiation, 1998,
95, 307–318; J. O. Ebinu, D. A. Bottorff, E. Y. W. Chan, S. L. Stang, R. J.
Dunn and J. C. Stone, Science, 1998, 280, 1082–1086; J. Van Lint, J.
Sinnett-Smith and E. Rozengurt, J. Biol. Chem., 1995, 270, 1455–1461;
M. G. Kazanietz, N. E. Lewin, J. D. Bruns and P. M. Blumberg, J. Biol.
Chem., 1995, 270, 10777–10783; R. M. Bell and D. J. Burns, J. Biol.
Chem., 1991, 266, 4661–4664.
12 C.-H. Liu, F. S. Machado, R. Guo, K. E. Nichols, A. W. Burks, J. C.
Aliberti and X.-P. Zhong, J. Exp. Med., 2007, 204, 781–792; S. M.
Okamura, C. E. Oki-Idouchi and P. S. Lorenzo, J. Biol. Chem., 2006,
281, 36132–36139; S. Madani, A. Hichami, M. Charkaoui-Malki and
N. A. Khan, J. Biol. Chem., 2004, 279, 1176–1183; F. C. Gomez-Merino,
C. A. Brearley, M. Ornatowska, M. E. F. Abdel-Haliem, M.-I. Zanor
and B. Mueller-Roeber, J. Biol. Chem., 2004, 279, 8230–8241.
13 Y. Iwasaki and T. Yamane, J. Mol. Catal. B: Enzym., 2000, 10, 129–140.
14 D. P. Siegel, J. Banschbach, D. Alford, H. Ellens, L. J. Lis, P. J. Quinn,
P. L. Yeagle and J. Bentz, Biochemistry, 1989, 28, 3703–3709; H. Kido,
N. Fukussen, K. Ishidoh and N. Katunuma, Biochem. Biophys. Res.
Commun., 1986, 138, 275–282.
Yield calcd for C33H49F3O7 (614.73): 0.565 g (92%, colorless oil);
Rf (pentane–toluene–EtOAc = 40 : 50 : 10, v/v/v) = 0.47.
3.9. Typical procedure for transformation of terminal
C3-iodohydrins (60, 61) into quaternary pyridinium salts (62, 63)
A mixture of iodoglyceride 60 or 61 (1.00 mmol) and pyridine
(2.4 mL, 30 mmol) was heated in a pressure glass ampoule
◦
under argon at 80 C for 18 h. The excess pyridine was removed
under reduced pressure and the residue was subjected to flash
column silica gel chromatography using CH2Cl2–MeOH–H2O
(65 : 25 : 4, v/v/v) as the mobile phase. The thus isolated products
were additionally recrystallized with dry diethyl ether (-20 ◦C) to
1
give the target compounds 62 and 63 (purity >99%, H NMR
spectroscopy).
( )-N-(1-Oleoyl-2-acetyl-3-propyl)pyridinium iodide 62. Ob-
tained from 1-oleoyl-2-acetyl-3-iodo-rac-glycerol (60; 0.508 g,
1.00 mmol). Yield: 0.528 g (90%, amorphous yellowish solid); Rf
(CH2Cl2–MeOH–H2O, 65 : 25 : 4, v/v/v) = 0.69; Found: C, 57.31;
H, 7.81; N, 2.40%. C28H46INO4 (587.58) requires C, 57.24; H, 7.89;
N, 2.38%.
15 H. Eibl and P. Woolley, Chem. Phys. Lipids, 1986, 41, 53–63.
16 L. Hartman, Chem. Rev., 1958, 58, 845–867.
17 R. Haftendorn, G. Schwarze and R. Ulbrich-Hofmann, Chem. Phys.
Lipids, 2000, 104, 57–66; M. L. M. Mannesse, J.-W. P. Boots, R.
Dijkman, A. J. Slotboom, H. T. W. M. van der Hijden, M. R. Egmond,
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 475
©

View Article Online
H. M. Verheij and G. H. de Haas, Biochim. Biophys. Acta, 1995, 1259,
56–64.
R. L. Dudley, Chem. Phys. Lipids, 1994, 72, 185–191; J. D. Baty and
N. W. Rawle, J. Chromatogr., A, 1987, 395, 395–402.
44 P. Villeneuve, M. Pina, D. Montet, G. Renard and J. Graille, Chem.
Phys. Lipids, 1994, 72, 135–141.
18 F. Marguet, J.-F. Cavalier, R. Verger and G. Buono, Eur. J. Org. Chem.,
1999, 1671–1678.
19 T. Ren and D. Liu, Tetrahedron Lett., 1999, 40, 209–212.
20 I. Solodin, C. S. Brown and T. D. Heath, Synlett, 1996, 457–458.
21 P. P. Karmali, V. V. Kumar and A. Chaudhuri, J. Med. Chem., 2004,
47, 2123–2132; X. Guo and F. C. J. Szoka, Acc. Chem. Res., 2003, 36,
335–341.
22 S. M. Weber, K. A. Peterson, B. Durkee, C. Qi, M. A. Longino, T.
Warner, F. T. Lee and J. P. Weichert, J. Surg. Res., 2004, 119, 41–45; J. K.
Doerr-Stevens, D. A. Bakan, F. T. Lee, S. G. Chosy, B. K. Markhardt,
A. K. Bonneville, C. Burrascano, C. Delaney, M. A. Longino, Y.
Greener and J. P. Weichert, Acad. Radiol., 2002, 9, S200–S204; D. A.
Bakan, J. K. Doerr-Stevens, J. P. Weichert, M. A. Longino, F. T. Lee Jr.
and R. E. Counsell, Am. J. Ther., 2001, 8, 359–365.
23 O. V. Gerasimov, Y. Rui and D. H. Thompson, Surfactant Sci. Ser.,
1996, 62, 679–746; A. Garzon-Aburbeh, J. H. Poupaert, M. Claesen and
P. Dumont, J. Med. Chem., 1986, 29, 687–691; A. Garzon-Aburbeh,
J. H. Poupaert, M. Claesen, P. Dumont and G. Atassi, J. Med. Chem.,
1983, 26, 1200–1203.
24 T. Naalsund, K. E. Malterud, V. Partali and H.-R. Sliwka, Chem. Phys.
Lipids, 2001, 112, 59–65.
45 P. R. J. Gaffney and C. B. Reese, Tetrahedron Lett., 1997, 38, 2539–2542.
46 L. Mannina, C. Luchinat, M. Patumi, M.-C. Emanuele, E. Rossi and
A. Segre, Magn. Reson. Chem., 2000, 38, 886–890; G. Vlahov, Magn.
Reson. Chem., 1998, 36, 359–362; M. M. Bergana and T. W. Lee, J. Am.
Oil Chem. Soc., 1996, 73, 551–556.
47 C. M. Lok, Chem. Phys. Lipids, 1978, 22, 323–337.
48 J. M. C. Golec, C. J. R. Hedgecock, R. Murdoch and W. R. Tully,
Tetrahedron Lett., 1992, 33, 551–554; J. C. Briggs, A. H. Haines, R. J. K.
Taylor, A. P. Dawson, I. Gibson, J. Hook, A. Lloyd and S. Meiners,
Carbohydr. Res., 1992, 234, 23–35; M. K. Jain, W. J. Tao, J. Rogers, C.
Arenson, H. Eibl and B.-Z. Yu, Biochemistry, 1991, 30, 10256–10268;
G. K. Chacko and D. J. Hanahan, Biochim. Biophys. Acta, 1968, 164,
252–271.
49 J. Gorzynski Smith, Synthesis, 1984, 629–656; J. J. Baldwin, A. W. Raab,
K. Mensler, B. H. Arison and D. E. McClure, J. Org. Chem., 1978, 43,
4876–4878; J. K. Addy and R. E. Parker, J. Chem. Soc., 1963, 915–921.
50 K. Ishihara, H. Kurihara and H. Yamamoto, J. Org. Chem., 1993, 58,
3791–3793; G. Hofle, W. Steglich and H. Vorbruggen, Angew. Chem.,
Int. Ed. Engl., 1978, 17, 569–583.
25 H. Houte, V. Partali, H.-R. Sliwka and E. G. K. Quartey, Chem. Phys.
Lipids, 2000, 105, 105–113.
26 K. W. Ferrara, M. A. Borden and H. Zhang, Acc. Chem. Res., 2009,
42, 881–892.
51 P. E. Sonnet, Chem. Phys. Lipids, 1991, 58, 35–39.
52 M. C. Moschidis, Chem. Phys. Lipids, 1985, 36, 297–302.
53 S. D. Stamatov and J. Stawinski, Tetrahedron Lett., 2007, 48, 1887–
1889.
27 M. Halter, Y. Nogata, O. Dannenberger, T. Sasaki and V. Vogel,
Langmuir, 2004, 20, 2416–2423; M. M. Wright and C. R. McMaster,
Biol. Res., 2002, 35, 223–229.
28 J. E. Browne, R. T. Freeman, J. C. Russell and P. G. Sammes, J. Chem.
Soc., Perkin Trans. 1, 2000, 645–652.
54 C. Yu, B. Liu and L. Hu, Tetrahedron Lett., 2000, 41, 4281–4285; M.
Lalonde and T. H. Chan, Synthesis, 1985, 817–845; S. K. Chaudhary
and O. Hernandez, Tetrahedron Lett., 1979, 20, 99–102; K. K. Ogilvie,
K. L. Sadana, E. A. Thompson, M. A. Quilliam and J. B. Westmore,
Tetrahedron Lett., 1974, 15, 2861–2863; K. K. Ogilvie, E. A. Thompson,
M. A. Quilliam and J. B. Westmore, Tetrahedron Lett., 1974, 15, 2865–
2868; E. J. Corey and A. Venkateswarlu, J. Am. Chem. Soc., 1972, 94,
6190–6191; A. D. Allen, J. C. Charlton, C. Eaborn and G. Modena,
J. Chem. Soc., 1957, 3668–3670.
55 C. M. Lok, J. P. Ward and D. A. vanDorp, Chem. Phys. Lipids, 1976,
16, 115–122.
56 F. S. Roodsari, D. Wu, G. S. Pum and J. Hajdu, J. Org. Chem., 1999,
64, 7727–7737.
29 S. D. Stamatov and J. Stawinski, Org. Biomol. Chem., 2007, 5, 3787–
3800.
30 C. E. Burgos, D. E. Ayer and R. A. Johnson, J. Org. Chem., 1987, 52,
4973–4977.
31 D. R. Kodali, J. Lipid Res., 1987, 28, 464–469.
32 B. Serdarevich, J. Am. Oil Chem. Soc., 1967, 44, 381–393.
33 E. Baer and H. O. L. Fisher, J. Am. Chem. Soc., 1945, 67, 2031–
2037.
34 IUPAC-IUB Commission on biochemical nomenclature. The nomen-
clature of lipids, J. Biol. Chem., 1967, 242, 4845–4849.
35 C. A. H. Prata, Y. Li, D. Luo, T. J. McIntosh, P. Barthelemy and
M. W. Grinstaff, Chem. Commun., 2008, 1566–1568; A. El-Aneed,
J. Controlled Release, 2004, 94, 1–14.
36 S. D. Stamatov and J. Stawinski, Synlett, 2005, 2587–2590.
37 S. D. Stamatov, M. Kullberg and J. Stawinski, Tetrahedron Lett., 2005,
46, 6855–6859.
38 S. A. Hardinger and N. Wijaya, Tetrahedron Lett., 1993, 34, 3821–3824;
P. G. McDougal, J. G. Rico, Y.-I. Oh and B. D. Condon, J. Org. Chem.,
1986, 51, 3388–3390; J. M. Renga and P.-C. Wang, Tetrahedron Lett.,
1985, 26, 1175–1178; G. H. Hakimelahi, Z. A. Proba and K. K. Ogilvie,
Tetrahedron Lett., 1981, 22, 4775–4778; T. J. Barton and C. R. Tully,
J. Org. Chem., 1978, 43, 3649–3653.
57 H. De Wilde, P. De Clerq, M. Vandewalle and H. Ro¨per, Tetrahedron
Lett., 1987, 28, 4757–4758.
58 K. Nacro, B. Bienfait, N. E. Lewin, P. M. Blumberg and V. E. Marquez,
Bioorg. Med. Chem. Lett., 2000, 10, 653–655; J.-L. Gras and J.-F.
Bonfanti, Synlett, 2000, 248–250.
59 A. Froling, H. J. J. Pabon and J. P. Ward, Chem. Phys. Lipids, 1984, 36,
29–38.
60 C. Teixeira, S. L. Stang, Y. Zheng, N. S. Beswick and J. C. Stone, Blood,
2003, 102, 1414–1420.
61 L. M. Strawn, R. E. Martell, R. U. Simpson, K. L. Leach and R. E.
Counsell, J. Med. Chem., 1989, 32, 2104–2110.
62 M. Kurz and G. K. E. Scriba, Chem. Phys. Lipids, 2000, 107, 143–157;
A. Hermetter, H. Stutz, R. Franzmair and F. Paltauf, Chem. Phys.
Lipids, 1989, 50, 57–62; P. Bette-Bobillo, A. Bienvenue, C. Broquet and
L. Maurin, Chem. Phys. Lipids, 1985, 37, 215–226.
63 D. R. Kodali, T. G. Redgrave, D. M. Small and D. Atkinson,
Biochemistry, 1985, 24, 519–525; L. Hartman, J. Chem. Soc., 1959,
4134–4135.
64 C. G. Rao, Org. Prep. Proced. Int., 1980, 12, 225–228.
65 R. J. Batten, A. J. Dixon, R. J. K. Taylor and R. F. Newton, Synthesis,
1980, 234–236; G. H. Dodd, B. T. Golding and P. V. Ioannou, J. Chem.
Soc., Chem. Commun., 1975, 249–250.
66 J.-C. Wang and G. Just, J. Org. Chem., 1999, 64, 8090–8097; D. Laine,
M. Fujita and S. V. Ley, J. Chem. Soc., Perkin Trans. 1, 1999, 1639–1645.
67 I. Ikeda, X. P. Gu, I. Miyamoto and M. Okahara, J. Am. Oil Chem.
Soc., 1989, 66, 822–824; W. T. M. deGroot, C. M. Lok and J. P. Ward,
Chem. Phys. Lipids, 1988, 47, 75–80; C. M. Lok, A. P. J. Mank and J. P.
Ward, Chem. Phys. Lipids, 1985, 36, 329–334; A. P. J. Mank, J. P. Ward
and D. A. VanDorp, Chem. Phys. Lipids, 1976, 16, 107–114.
68 S. D. Stamatov and J. Stawinski, Eur. J. Org. Chem., 2008, 2635–2643.
69 J.-L. Gras and J.-F. Bonfanti, Synlett, 1999, 1835–1837.
70 L. Fauconnot, F. Robert, R. Villard and F. Dionisi, Chem. Phys. Lipids,
2006, 139, 125–136.
39 S. D. Stamatov and J. Stawinski, Bioorg. Med. Chem. Lett., 2006, 16,
3388–3391.
40 J. C. Go´mez-Ferna´ndez and S. Corbala´n-Garc´ıa, Chem. Phys. Lipids,
2007, 148, 1–25; Y. Kambayashi, S. Takekoshi, Y. Tanino, K. Watanabe,
M. Nakano, Y. Hitomi, T. Takigawa, K. Ogino and Y. Yamamoto,
J. Clin. Biochem. Nutr., 2007, 41, 68–75; C. Hisatsune, K. Nakamura,
Y. Kuroda, T. Nakamura and K. Mikoshiba, J. Biol. Chem., 2005, 280,
11723–11730; S. J. Slater, J. L. Seiz, B. A. Stagliano, A. C. Cook, S. K.
Milano, C. Ho and C. D. Stubbs, Biochemistry, 2001, 40, 6085–6092;
E. G. Lapetina, B. Reep, B. R. Ganong and R. M. Bell, J. Biol. Chem.,
1985, 260, 1358–1361; T. J. Rink, A. Sanchez and T. J. Hallam, Nature,
1983, 305, 317–319; T. Mori, Y. Takai, B. Yu, J. Takahashi, Y. Nishizuka
and T. Fujikura, J. Biochem. (Tokyo), 1982, 91, 427–432.
41 J. G. Ebeling, G. R. Vandenbark, L. J. Kuhn, B. R. Ganong, R. M. Bell
and J. E. Niedel, Proc. Natl. Acad. Sci. U. S. A., 1985, 82, 815–819.
42 S.-Y. Han, S.-H. Cho, S.-Y. Kim, J.-T. Seo, S.-J. Moon and G.-J. Jhon,
Bioorg. Med. Chem. Lett., 1999, 9, 59–64.
43 R. O. Adlof and G. R. List, J. Agric. Food Chem., 2003, 51, 2096–2099;
F. Turon, F. Bonnot, Y. Caro, M. Pina and J. Graille, Chem. Phys.
Lipids, 2003, 125, 41–48; J. J. Myher, A. Kuksis, K. Geher, P. W. Park
and D. A. Diersen-Schade, Lipids, 1996, 31, 207–215; P. E. Sonnet and
71 A. Bhati, R. J. Hamilton, D. A. Steven, R. Aneja and F. B. Padley,
J. Chem. Soc., Perkin Trans. 2, 1983, 1553–1558.
476 | Org. Biomol. Chem., 2010, 8, 463–477
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
72 A. D. Miller, Angew. Chem., Int. Ed., 1998, 37, 1768–1785; S. Walker,
M. J. Sofia, R. Kakarla, N. A. Kogan, L. Wierichs, C. B. Longley, K.
Bruker, H. R. Axelrod, S. Midha, S. Babu and D. Kahne, Proc. Natl.
Acad. Sci. U. S. A., 1996, 93, 1585–1590; P. L. Felgner, T. R. Gadek,
M. Holm, R. Roman, H. W. Chan, M. Wenz, J. P. Northrop, G. M.
Ringold and M. Danielsen, Proc. Natl. Acad. Sci. U. S. A., 1987, 84,
7413–7417.
73 M. C. Filion and N. C. Phillips, Br. J. Pharmacol., 1997, 122, 551–557;
I. D. Konstantinova and G. A. Serebrennikova, Russ. Chem. Rev., 1996,
65, 537–553; L. Susan, S. Morris-Natschke, F. Gumus, C. J. Marasco,
K. L. Meyer, M. Marx, C. Piantadosi, M. D. Layne and E. J. Modest,
J. Med. Chem., 1993, 36, 2018–2025; C. J. Marasco, C. Piantadosi, K. L.
Meyer, S. Morris-Natschke, K. S. Ishaq, G. W. Small and L. W. Daniel,
J. Med. Chem., 1990, 33, 985–992; L. Susan, S. Morris-Natschke, K. L.
Meyer, C. J. Marasco, C. Piantadosi, F. Rossi, P. L. Godwin and E. J.
Modest, J. Med. Chem., 1990, 33, 1812–1818.
75 B. Martin, M. Sainlos, A. Aissaoui, N. Oudrhiri, M. Hauchecorne,
J.-P. Vigneron, J.-M. Lehn and P. Lehn, Curr. Pharm. Des., 2005, 11,
375–394.
76 I. Solodin, C. S. Brown, M. S. Bruno, C.-Y. Chow, E.-H. Jang, R. J.
Debs and T. D. Heath, Biochemistry, 1995, 34, 13537–13544.
77 D. Liu, W. Qiao, Z. Li, Y. Chen, X. Cui, K. Li, L. Yu, K. Yan, L. Zhu,
Y. Guo and L. Cheng, Chem. Biol. Drug Des., 2008, 71, 336–344.
78 M. A. Ilies, B. H. Johnsonb, F. Makorib, A. Millerb, W. A. Seitza, E. B.
Thompsonb and A. T. Balaban, Arch. Biochem. Biophys., 2005, 435,
217–226; M. A. Ilies, W. A. Seitz, T. Miron, M. T. Caproiu, M. Wentz,
R. E. Garfield and A. T. Balaban, Eur. J. Org. Chem., 2003, 2645–2655.
79 D. D. Perrin and W. L. F. Armarego, (Ed.) Purification of Laboratory
Chemicals, Pergamon Press, New York, 1988.
80 S. D. Stamatov and J. Stawinski, Tetrahedron, 2005, 61, 3659–
3669.
81 H. M. Weiss and K. M. Touchette, J. Chem. Soc., Perkin Trans. 2, 1998,
74 R. Leventis and J. R. Silvius, Biochim. Biophys. Acta, 1990, 1023, 124–
132.
1517–1522.
82 P. N. Guivisdalsky and R. Bittman, J. Org. Chem., 1989, 54, 4637–4642.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 463–477 | 477
©