Epoxidation of polyunsaturated fatty acid double bonds by dioxirane reagent: Regioselectivity and lipid supramolecular organization

Article abstract of DOI:10.1002/hlca.200690209

The use of dimethyldioxirane (DMD) as the epoxidizing agent for polyunsaturated fatty acids was investigated. With fatty acid methyl esters, this is a convenient method for avoiding acidic conditions, using different solvents, and simplifying the isolation procedures, with less contamination due to by-products. The reagent was also tested with free fatty acids in water. In this case, the supramolecular organization of fatty acids influenced the reaction outcome, and the epoxidation showed interesting regioselective features. The C=C bonds closest to the aqueous-micelle interface is the most favored for the interaction with dimethyldioxirane. The preferential epoxidation of linoleic acid (=(9Z,12Z)-octadeca-9,12-dienoic acid) to the 9,10-monoepoxy derivative was achieved, with a high yield and 65% regioselectivity. In case of arachidonic acid (=(5Z,8Z,11Z,14Z)-eicosa-5,8,11,14-tetraenoic acid) micelles, the regioselective outcome with formation of the four possible monoepoxy isomers was studied under different conditions. It resulted to be a convenient synthesis of 'cis-5,6-epoxyeicosatrienoic acid' (=3-[(2Z,5Z,8Z)-tetradeca-2,5,8- trienyl]oxiran-2-butanoic acid), whereas in reverse micelles, epoxidation mostly gave 'cis-14,15-epoxyeicosatrienoic acid (= (5Z,8Z,11Z)-13-(3-pentyloxiran-2- yl)trideca-5,8,11-trienoic acid).

Full text of DOI:10.1002/hlca.200690209

Helvetica Chimica Acta – Vol. 89 (2006)
2243
Epoxidation of Polyunsaturated Fatty Acid Double Bonds by Dioxirane
Reagent: Regioselectivity and Lipid Supramolecular Organization
by Stanislav A. Grabovskiya)b), Natalia N. Kabalꢀnovab), Chryssostomos Chatgilialoglua), and Carla
Ferreri*a)
a) I.S.O.F., Consiglio Nazionale delle Ricerche, Via Piero Gobetti 101, I-40129 Bologna, Italy
b) Institute of Organic Chemistry, Ufa Research Center of the Russian Academy of Science, 71 prosp.
Oktyabrya, 450054 Ufa, Russian Federation
Dedicated to Professor Hanns Fischer, an unforgettable teacher and friend
The use of dimethyldioxirane (DMD) as the epoxidizing agent for polyunsaturated fatty acids was
investigated. With fatty acid methyl esters, this is a convenient method for avoiding acidic conditions,
using different solvents, and simplifying the isolation procedures, with less contamination due to by-prod-
ucts. The reagent was also tested with free fatty acids in water. In this case, the supramolecular organi-
zation of fatty acids influenced the reaction outcome, and the epoxidation showed interesting regioselec-
tive features. The C=C bonds closest to the aqueous-micelle interface is the most favored for the inter-
action with dimethyldioxirane. The preferential epoxidation of linoleic acid (=(9Z,12Z)-octadeca-9,12-
dienoic acid) to the 9,10-monoepoxy derivative was achieved, with a high yield and 65% regioselectivity.
In case of arachidonic acid (=(5Z,8Z,11Z,14Z)-eicosa-5,8,11,14-tetraenoic acid) micelles, the regioselec-
tive outcome with formation of the four possible monoepoxy isomers was studied under different condi-
tions. It resulted to be a convenient synthesis of ꢀcis-5,6-epoxyeicosatrienoic acidꢁ (=3-[(2Z,5Z,8Z)-tet-
radeca-2,5,8-trienyl]oxiran-2-butanoic acid), whereas in reverse micelles, epoxidation mostly gave ꢀcis-
14,15-epoxyeicosatrienoic acid (=(5Z,8Z,11Z)-13-(3-pentyloxiran-2-yl)trideca-5,8,11-trienoic acid).
Introduction. – After the early discovery as constituents of vegetable oils [1] and
metabolites by the epoxygenase pathway [2], epoxy derivatives of polyunsaturated
fatty acid (PUFA) have attracted a multidisciplinary research interest in chemistry,
biology and pharmacology. In particular, monoepoxy derivatives of linoleic acid
((9Z,12Z)-octadeca-9,12-dienoic acid) and arachidonic acid (=(5Z,8Z,11Z,14Z)-
eicosa-5,8,11,14-tetraenoic acid) have been identified as products of cytochrome
P450 enzymes, and a series of biological effects have been linked to their in vivo forma-
tion. ꢀcis-9,10-Epoxy-12cis-octadec-12-enoic acidꢁ (=rel-(2R,3S)-3-[(2Z)-oct-2-enyl]ox-
iraneoctanoic acid; 9,10-EOA) and ꢀcis-12,13-epoxy-9cis-octadecenoic acidꢁ (=rel-
(9Z)-11-[(2R,3S)-3-pentyloxiran-2-yl]undec-9-enoic acid; 12,13-EOA) are known as
leukotoxin and isoleukotoxin, respectively, and are formed during the functioning of
leukocytes, as well as the lipid peroxidation processes [3]. In the field of natural-prod-
uct chemistry, 9,10-EOA and 12,13-EOA are known as coronaric and vernolic acids,
respectively [1]. The ꢀcis-epoxyeicosatrienoic acidsꢁ (cis-EETs, Scheme 1) are synthe-
sized in many tissues, including kidney, liver, adrenal cells, ovarian cells, endothelial
cells and brain. Potent effects have been observed in modulating various ion channels,
membrane-bound transport proteins, mitogenesis, tyrosine kinase cascades [4]. Specific
ꢂ 2006 Verlag Helvetica Chimica Acta AG, Zürich

2244
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 1. Preparation of Four Regioisomers by Monoepoxidation of Arachidonic Acid Methyl Ester
activities have been linked to each regioisomer of PUFA monoepoxides, as for exam-
ple, the previously recalled leukotoxin which is specifically produced and hydrolyzed to
the corresponding diol in the acute respiratory distress syndrome [5], or the 8,9-and
11,12-EET, which are activators of KATP channels with a resulting modulation of cardiac
electrophysiology [6]. In case of the 14,15-EET, this compound is known as an epider-
mal growth-factor signaling [7].
Therefore, fatty acid epoxidation addressing the regioselectivity issue is of relevant
importance. Chemical methods are limited to the classical procedure with peroxy acid
reagents, such as 3-chloroperbenzoic acid (m-CPBA), which are essentially nonselec-
tive toward the different C=C bonds of polyunsaturated structures. Unselective bis-
epoxidation of linoleic acid and with m-CPBA, has been used in the route to tetrahy-
drofurans [8]. In the case of arachidonic acid, the four monoepoxy derivatives are
obtained (Scheme 1). The only selective epoxidation protocol to date is the ꢀinternalꢁ
strategy reported by Corey and co-workers [9] applied to peroxyarachidonic acid, in
which an intramolecular O-transfer operates to afford preferentially the 14,15-EET.
However, the peroxy acid method does not avoid acidic conditions and the need of
chromatographic removal of by-products. For conjugated linoleic acid, a recent report
comparing different epoxidizing conditions has been published, without evidencing any
selectivity feature [10].
In connection with our work on the regioselectivity of radical reactions involving
the supramolecular organization of a lipid assembly [11][12], the study of the epoxida-
tion with dimethyldioxirane (DMD) [13][14] can be interesting, based on two features:
i) the advantage of nonacidic reaction conditions, coupled with the production of vol-

Helvetica Chimica Acta – Vol. 89 (2006)
2245
atile by-products and the easy isolation procedure; ii) the tolerance of DMD to aque-
ous conditions. This latter aspect can be crucial for a regioselective strategy. In fact,
fatty acids are immiscible in water and assume a supramolecular organization, forming
micelles at a critical micelle concentration (CMC). The supramolecular disposition of
the hydrophobic fatty acid chains in this assembly can drive specific interactions with
the added reagents, so that molecular properties can be developed at the interface
between H2O and hydrophobic phases. This concept has been recently applied in sev-
eral aqueous synthetic procedures ꢀon waterꢁ, the immiscibility of the reagents becom-
ing an advantage instead of being a drawback [15][16]. The behavior in micelles can
also be compared with reverse micelles that fatty acids form in organic solvents, with
an organization opposite to the aqueous conditions.
We now report the results of epoxidation of linoleic and arachidonic acid and their
methyl esters with DMD, together with the study of regioselectivity obtained with the
free acids in micelles.
Results. – 1. PUFAMethyl Ester Epoxidation. The 0.08–1.4M solution of DMD in
acetone was synthesized following the standard procedure from Oxoneꢃ (2
KHSO5 ·KHSO4 ·K2SO4) [13][14]. To perform the epoxidation in other organic sol-
vents, an extraction from acetone was preliminarily carried out to get a final DMD con-
centration of 0.15M, as determined by spectrophotometric measurement [14]. First, we
tested the reactivity of linoleic acid methyl ester with an equimolar amount of DMD
under different conditions of solvent and temperature. The results are reported in
Table 1. As depicted in Scheme 2, the monoepoxy derivatives, 9,10-EOA (1) and
12,13-EOA (2) are the intermediates for the half reaction.
Table 1. DMD Epoxidation of Linoleic Acid Methyl Ester (see Scheme 2)
Entry Temperature [8], solvent, equiv.
Yield
[%]
Monoepoxy derivatives Diepoxy derivatives
DMD
1/2
3/4
1
2
3
4
5
0, CH2Cl2, 1
À20, CH2Cl2, 1
0, CH2Cl2, 2
0, acetone, 2
45, acetone, 2
51a)
50a)
99
99
99
1:1 (47%)
1:3 (4%)
1:2.3
1:2.3
1:1.7
1:1.7
–
–
–
–
a) 49% recovery of starting material.
The best solvent and temperature for monoepoxidation were found to be CH2Cl2 at
08 and 1 equiv. of DMD (Table 1, Entry 1). Under these conditions, only a 4% yield of
diepoxy derivatives was obtained, whereas the two positional isomers were equally
formed in 47% yield, and a full recovery of the starting methyl ester was possible.
The same reaction at À208 (Entry 2) afforded only ꢀsynꢁ and ꢀantiꢁ diepoxy derivatives
3 and 4 in 50% yield. With 2 equiv. of DMD and raising the temperature to 08 to
improve the solubility, 3 and 4 were obtained in quantitative yield (Entry 3). By replac-
ing CH2Cl2 with acetone as the solvent, a similar result was obtained (Entry 4), which
did not change at higher temperature (Entry 5). The two solvents slightly influenced the
ꢀsynꢁ and ꢀantiꢁ diepoxy derivative ratio, which is always in favor of ꢀantiꢁ 4 and similar to

2246
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 2. Synthesis of Mono- and Diepoxy Derivatives by DMD Epoxidation of Methyl Linoleate
the reaction with peracids [17]. The separation between the ꢀsynꢁ and ꢀantiꢁ products
was achieved by crystallization at À208 in hexane, and their spectroscopic characteris-
tics were in agreement with the literature data [18].
The methodology was then applied to methyl arachidonate. Fig. 1 shows the forma-
tion of the epoxy products, and each point corresponds to the experiment run with the
corresponding amount of DMD. For example, a ca. 40% yield of monoepoxy deriva-
tives was obtained with 0.5 equiv. DMD, which also afforded 4% of diepoxy derivatives.
Based the recovery of the unreacted starting material, the yield was quantitative. As
shown in Fig. 1, formation of di-and triepoxy derivatives increased with higher
amounts of DMD.
Fig. 1. Product distribution in the epoxidation of methyl arachidonate with DMD. The equiv. of DMD
refer to 1 equiv. of substrate, in CCl4 at 08.

Helvetica Chimica Acta – Vol. 89 (2006)
2247
Silica gel chromatography is a well known method for the separation of EET as
methyl esters [9]. The isolated monoepoxy derivatives were identified by comparison
with commercially available references, resulting in a mixture of 5,6-EET/8,9-EET/
11,12-EET/14,15-EET in the ratio of 1.6 :1.6 :2.6 :1.
The exhaustive epoxidation of methyl arachidonate was carried out with 4 equiv. of
DMD in CCl4 at 08 with a quantitative formation of a mixture of the diastereoisomeric
tetraepoxy derivatives. The presence of diastereoisomeric tetraepoxy derivatives was
also confirmed by GC and GC/MS analyses, but the mixture was not further resolved.
2. PUFAEpoxidation in Micelle. The epoxidation was then performed under aque-
ous conditions with free fatty acids, which is very convenient for a direct access to the
biologically active compounds. In H2O the CMCs for linoleic and arachidonic acids are
ca. 10À4M [19], and the supramolecular arrangement of the fatty acid chains in the
micelles could be relevant for the regioselectivity of the transformation. An optimiza-
tion of the fatty acid concentrations and pH conditions was preliminarily carried out.
DMD in H2O was formed by an in situ procedure, i.e., by adding the potassium peroxy-
monosulfate salt to a vigorously stirred basic suspension of the fatty acid, containing
<2% of acetone [20]. In case of linoleic acid, a 0.1M fatty acid suspension was prepared
in aqueous 0.9M KOH containing 1.12M NaHCO3 (pH 9.5), and Oxoneꢃ was added por-
tion-wise up to a final concentration of 0.05M. The mixture was then stirred overnight at
room temperature, followed by extraction and transformation of the fatty acids to the
corresponding methyl esters by diazomethane for GC analyses. Under these conditions,
a 15% conversion was found with a complete recovery of the starting linoleic acid
(85%). Interestingly, the resulting mixture of monoepoxy derivatives was in favor of
the 9,10-EOA (1), the 9,10-EOA/12,13-EOA (1/2) ratio being 1.8 :1. Remarkably,
this is the first report of regioselective epoxidation of linoleic acid.
The more complex case of arachidonic acid epoxidation in a micelle environment
was then examined. A vigorously stirred 0.05M suspension of arachidonic acid was
used in the above described alkaline aqueous medium. DMD was progressively formed
via the in situ addition (in portions) of Oxoneꢃ during a few hours. The progress of the
reaction was monitored by TLC (silica gel), which allows arachidonic acid to be sepa-
rated from the monoepoxy, and also from the disepoxy derivatives. On TLC, the mono-
epoxy regioisomer 14,15-EET (Rf 0.62) was separated from 11,12-EET/8,9-EET (Rf
0.58) and 5,6-EET (Rf 0.50). The results of the product distribution at two different
Oxoneꢃ concentrations are shown in Table 2 (Entries 1 and 2).
In the first case, a total of 1.9 equiv. of Oxoneꢃ was added (Table 2, Entry 1). Due to
the mild conditions, easy workup, and good analytical procedures, a quantitative mate-
rial balance was obtained, composed by 46% of monoepoxy derivatives, 27% of diep-
oxy derivatives, and 27% of recovered starting material. The identification of the EETs
was performed by isolation of the regioisomers by means of silica gel chromatography
(see above), characterization by NMR spectroscopy, and comparison with authentic
references which are all commercially available. The final mixture of monoepoxy deriv-
atives consisted of equal amounts of 5,6-and 8,9-EETs, and ca. 3–4 times less amount
of 11,12-EET and 14,15-EET. This indicated that in micelles the C=C bonds have a
nonequivalent reactivity toward the epoxidizing agent, with the C(5)=C(6) and
C(8)=C(9) bonds more reactive compared to the other two positions. It is worth noting
that in this reaction the 5,6-EET was isolated as a pure product in 15% yield.

2248
Helvetica Chimica Acta – Vol. 89 (2006)
Table 2. DMD Epoxidation of Arachidonic Acid (AA) in Organized Systems
Entry AA Solvent, micelle type, Monoepoxy 5,6-EET/8,9-EET/
Diepoxy deriva- Recovered
[M]
reagent
derivatives 11,12-EET/14,15-EET tives (yield [%]) AA [%]
(yield [%])
1
2
3
0.19 H2O, micelles, 1.9
equiv. Oxoneꢃ
46a)
30
3 :3 :1:0.7
27
15
19
27
55
23
0.19 H2O, micelles, 0.8
3 :1:0.3 :0
equiv. Oxoneꢃ
0.035 CCl4, reverse
micelles, 0.8 equiv.
DMD
0.030 CCl4, reverse
micelles, 1.9 equiv.
DMD
47
1:1.3 :3.3 :4.5
4
23
0 :1:3 :3
44b)
2
a) 5,6-EET in a 15% isolated yield. b) Formation of triepoxy derivatives (24% yield).
With 0.8 equiv. of Oxoneꢃ, a lower conversion was observed, i.e., 30% of monoep-
oxy derivatives, 15% of diepoxy derivatives, and complete recovery of the unreacted
arachidonic acid (55%) (Table 2, Entry 2). Interestingly, in this low-conversion proto-
col, a better regioselectivity was observed, the mixture of monoepoxy derivatives
being highly in favor of the 5,6-EET regioisomer, present in a 3 :1:0.3 ratio with 8,9-
EET and 11,12-EET, whereas the 14,15-EET was completely absent.
Finally, arachidonic acid epoxidation was also run in CCl4 as the solvent where a
reverse micelle organization can be expected [21]. In this assembly, a different dispo-
sition of the hydrophobic chains occurs. Table 2 shows the results for low-and high-con-
version protocols (Entries 3 and 4), which were carried out by the direct DMD addition
to the reaction mixture. It is worth noting that in the direct addition, a better control of
the DMD is achieved, and the efficiency of the reaction in the organic solvent was much
higher than in the aqueous system. The arachidonic acid concentration was optimized
at ca. 0.030–0.035M. DMD was first added in a 0.8 equiv. amount with respect to the
starting material, and the reaction was instantaneous (Table 2, Entry 3). After a few
minutes, the workup and transformation to methyl esters were performed, resulting
in 47% of monoepoxy and 19% of diepoxy derivatives. After silica gel chromatography,
the monoepoxy fraction had the composition 14,15-EET/11,12-EET/8,9-EET/5,6-EET
4.5 :3.3 :1.3 :1. This is in agreement with the supramolecular organization in the micelle
system that differentiates the reactivity of the C=C bonds. Interestingly, the
regioisomer composition was different from that obtained in the previously reported
micelle experiments.
The experiment was repeated with 1.9 equiv. of DMD, but this time, the mixture of
the epoxy derivatives was much more complicated since 24% of triepoxy derivatives
were also present (Table 2, Entry 4), besides 23% of monoepoxy and 44% of diepoxy
derivatives. Only a 2% of starting material was recovered. The EET mixture was com-
posed of 14,15-EET/11,12-EET/8,9-EET in a 3 :3 :1 ratio.

Helvetica Chimica Acta – Vol. 89 (2006)
2249
Discussion. – PUFA Epoxidation with DMD proved to be convenient for avoiding
acidic conditions and facilitating workup and isolation of the products. The epoxidation
of linoleic acid methyl ester with DMD furnished the same results as the classical per-
acid method both for mono-and diepoxy derivatives [17]. The epoxidation of methyl
arachidonate with DMD in CCl4 is a step-by-step process, as depicted in Fig. 1. With
0.5 equiv. of DMD, the four monoepoxy derivatives 5,6-EET/8,9-EET/11,12-EET/
14,15-EET were obtained in the ratio 1.6 :1.6 :2.6 :1 as the main products in a 40%
of conversion. For comparison, the reported ratios obtained after mCPBA [22] and
enzymatic [23] epoxidations are 1:2 :2 :2 and 1.5 :1:1.6 :5, respectively.
When a supramolecular organization of the fatty acid is involved, the regioselectiv-
ity of the reaction can improve. DMD Epoxidation of linoleic acid in micelles repre-
sents the first case of a regioselective process, with the preferential formation of the
9,10-monoepoxy derivative. It indicates how effective the micellar supramolecular
arrangement can be in orienting the reactivity of the epoxidation toward the C=C
bond closest to the aqueous phase. In fact, DMD is generated in situ by the addition
of the potassium peroxymonosulfate salt to the suspension. Once formed, it diffuses
from the aqueous phase to the micelle compartment and reacts with the C=C bonds.
The C=C bonds in the micelle arrangement of polyunsaturated fatty acids result to
be nonequivalent, C(9)=C(10) being closer to the aqueous phase than C(12)=C(13),
as shown in Fig. 2. Therefore, the regioselective outcome is due to a combination of
DMD diffusion, local concentration and C=C bond interaction allowed by the supra-
molecular assembly [24].
Fig. 2. Sketch of linoleic acid micelles, and result of DMD epoxidation
The regioselectivity is again found in the case of arachidonic acid epoxidation in
micelles, where the preferential formation of 5,6-EET and 8,9-EET was observed. At
lower conversion, that is, with less DMD equivalents, 5,6-EET is the prevalent isomer.
In the absence of data on the most favored conformation of arachidonic acid in
micelles, we can only say that the interaction of the epoxidizing agent with the first
C=C bond of the chain occurs by a combination of factors, among which the diffusion
of DMD and its local concentration play a role. At a low DMD concentration, chemo-
selectivity is achieved as well, with reduced formation of the diepoxy derivatives (cf.,

2250
Helvetica Chimica Acta – Vol. 89 (2006)
Table 2, Entries 1 and 2). Due to the satisfactory chromatographic separation of 5,6-
EET, the DMD epoxidation in micelles can represent a convenient synthetic method
for this biologically relevant monoepoxy derivative.
In CCl4 the organization of free fatty acids takes the form of reverse micelles, and
our data establish that in this case, 14,15-EET is the major compound. However, by
using a higher DMD concentration, the selectivity drops, due to the local concentration
that the reagent reaches in the organic-micellar interface, with a reduced yield of mono-
epoxy derivatives and increased formation of diepoxy derivatives. Under these condi-
tions, 5,6-EET is not formed, which again can be explained by an effect of the local con-
centration of the reagent (Table 2, Entry 4).
Conclusions. – In this paper, a valuable strategy for the DMD epoxidation of fatty
acids is reported, which expands the available selective synthetic methodologies for lip-
idomic libraries. Further work on the synthesis of specific mono-trans isomers of arach-
idonic acid [25], by using the micelle reactivity and formation of EETregioisomers, is in
progress.
The authors thank the Consiglio Nazionale delle Ricerche (CNR) and the Russian Academy of Sci-
ences (RAS) for the support given for the exchange visits. S. A. G. thanks the Russian Science Support
Foundation for partial financial support. S. A. G. and N. K. thank the Program of Basic Research, Depart-
ment of Chemistry and Material Science (DCMS), of RAS for partial financial support (ꢀTheoretical and
Experimental Studies of the Nature of Chemical Bond and Mechanisms of the Important Chemical
Reactions and Processesꢁ; contracts No. 10002-251/DCMS – 01/118 – 141/160603 – 687 and No.
171414/9, April 1, 2003).
Experimental Part
1. General. Arachidonic and linoleic acids and their corresponding methyl esters were commercially
available from Sigma-Aldrich. EETs of arachidonic acid methyl ester were commercially available from
Biomol International (Exeter, UK). The solns. of DMD (0.08–0.15M in CCl4, emax 14 MÀ1 cmÀ1) were pre-
pared as already reported in detail [13][14]. HPLC-Grade CCl4 and acetone were used as received. GC:
HP-5890-II (gas chromatography; 30 m”0.25 mm”0.25 mm cross-linked 5% phenylsilicone capillary col-
umn (HP-5) with hexadecanoic acid methyl ester as the internal standard; initial temp. 1008, held for 2
min, then 208/min up to 2108, held for 25 min. GC (linoleic acid reaction): Varian-CP-3800 gas chromato-
graph; flame ionization detector, Rtx-2330 column (90% biscyanopropyl/10% phenylcyanopropyl poly-
siloxane capillary column; 60 m, 0.25 mm i. d., 0.20 mm film thickness); initial temp. 1608, held for 55 min,
then 58/min up to 1958, held for 10 min, followed by a second increase of 108/min up to 2508 (FC=flash
chromatography). UV-VIS Spectra: Perkin-Elmer lamda-20 UV/VIS spectrometer with a thermostating
block Perkin-Elmer PTP-6 (Peltier system), in CCl4. IR Spectra: Perkin-Elmer Spectrum-BX-FT-IR
instrument; film or CHCl3; in cmÀ1. NMR Spectra: Varian-Mercury instrument at 400 (1H) or 100.6
(13C) MHz; d in ppm; CDCl3 as the solvent and the reference peak at d 7.26. GC/MS: HP-5890-II instru-
ment with a mass-selective detector HP5972. ESI/MS: Bruker Esquire-3000-Plus instrument; in m/z.
2. Epoxidation of Linoleic Acid Methyl Ester. ꢀcis-9,10-Epoxy-12cis-octadecenoic Acid Methyl Esterꢁ
(=rel-(2R,3S)-3-[(2Z)-Oct-2-enyl]oxiraneoctanoic Acid Methyl Ester; 1) and ꢀcis-12,13-Epoxy-9cis-octa-
decenoic Acid Methyl Ester (=rel-(9Z)-11-[(2R,3S)-3-Pentyloxiranyl]undec-9-enoic Acid Methyl Ester;
2). At 08, 0.15M DMD in CCl4 (2.2 ml, 0.34 mmol) was added to linoleic acid methyl ester (100 mg,
0.34 mmol) in CCl4 (200 mL). The reaction of DMD was followed spectrophotometrically at 335 nm
and monitored by TLC (Et
derivative) 0.15). After a few minutes, the mixture was poured into brine and extracted with Et CTHREUNG
O/hexane 1:4; Rf (starting material) 0.75, Rf (1) 0.49, Rf (2) 0.44, Rf (diepoxy
2AO. The

Helvetica Chimica Acta – Vol. 89 (2006)
2251
combined org. layer was dried (MgSO4) and concentrated. The crude residue was purified by FC (25%
Et2AO/hexane): recovered linoleic acid methyl ester (51 mg, 50%), 1 (24 mg, 22%), 2 (24 mg, 22%), having
CTHREUNG
characteristics similar to those reported in [21].
ꢀsyn-ꢁ and ꢀanti-9,10:12,13-Diepoxyoctadecanoic Acid Methyl Estersꢁ (=rel-(2R,3S)-3-{[(2R,3S)-3-
Pentyloxiran-2-yl]methyl}oxiraneoctanoic Acid Methyl Ester and rel-(2R,3S)-3-{[(2S,3R)-3-Pentylox-
iran-2-yl]methyl}oxiraneoctanoic Acid Methyl Ester, resp.; 3 and 4, resp.). At 08, 0.15M DMD in CCl4
(0.07 mmol) was added to linoleic acid methyl ester (10 mg, 0.034 mmol) in CCl4 (2 ml). The reaction
of DMD proceeded as described above. After completion, the solvent was evaporated: 3/4 (11 mg,
99%). Separation of the diastereoisomers 3 and 4 was carried out according to the literature procedure.
Spectroscopic data: identical to those reported in [8][17].
3. Epoxidation of Arachidonic Acid Methyl Ester. Monoepoxy Derivatives. At 08, 0.15M DMD in CCl4
(1 ml, 0.16 mmol) was added to arachidonic acid methyl ester (100 mg, 0.31 mmol) in CCl4 (20 ml). The
reaction of DMD was followed spectrophotometrically at 335 nm and was monitored by TLC (hexane/
Et2ACHTREOUGN 4 :1) as described in [20]. The formation of monoepoxy derivatives was almost instantaneous, and
the solvent was evaporated under vacuum. Chromatographic separation gave fractions of monoepoxy
derivatives (42 mg, 40%), diepoxy derivatives (4 mg, 4%), and recovered starting material (54 mg,
55%). The monoepoxy derivatives were examined by 13C-NMR: characteristic signals of epoxy and ole-
finic C-atoms of 5,6-EET/8,9-EET/11,12-EET/14,15-EET, the ratio being 1.6 :1.6 :2.6 :1. Spectroscopic
data (including MS): identical to references and in agreement with those reported in [9][22][23].
5,6:8,9:11,12:14,15-Tetraepoxyeicosanoic Acid Methyl Ester (=3-{{3-{{3-[(3-Pentyloxiran-2-yl)me-
thyl]oxiran-2-yl}methyl}oxiran-2-yl}methyl}oxiranebutanoic Acid Methyl Ester). At 08, 0.15M DMD in
CCl4 (0.9 ml, 0.13 mmol) was added to arachidonic acid methyl ester (10 mg, 0.033 mmol) in CCl4 (2
ml). The reaction was followed spectrophotometrically at 335 nm and monitored by TLC (Et2ACHTREUONG ; Rf (tet-
raepoxide) 0.43). After completion, the solvent was evaporated: methyl tetraepoxyarachidonate (12 mg,
95%). GC: eight diastereoisomers, three of which are major isomers summing up to 60% of the products.
1
IR (film): 2957, 2931, 2860, 1738, 1456, 1438, 1390, 1250, 1199, 1172, 1011, 847, 827. H-NMR: 0.888 (t,
J=7.0, Me); 1.28–1.91 (m, 18 H, CH2); 2.31–2.47 (m, MeOC(O)CH2); 2.93–3.02 (m, HÀC(5), HÀ
C(15)); 3.06–3.23 (m, 6 H, CH); 3.661 (s, MeO). 13C-NMR: 13.94 (Me); 21.83, 21.93 (C(3)); 22.53
(C(19)); 22.12, 22.21 (C(17)); 26.97–27.46 (clustered peaks); 27.78, 27.83, 31.60 (C(18)); 31.63
(C(18)); 33.47 (C(2)); 51.55 (MeO); 53.80–54.25 (clustered peaks); 56.09, 56.39, 56.69, 56.97, 173.55
(C(1)). ESI-MS: 405.2 ([M+Na]+). Anal. calc. for C21H34O6 (382.24): C 65.49, H 8.96, O 25.10; found:
C 65.47, H 8.95, O 25.08.
4. Epoxidation of Linoleic Acid in Micelles. In 100 ml of an aq. medium formed by 0.9M KOH con-
taining 1.12M NaHCO3 (pH 9.5), linoleic acid (29 mg, 0.01 mmol) was suspended. Then 2.7M acetone in
H2O (0.005 mmol, 1.8 ml) was added. The mixture was vortexed until a suspension was obtained. To the
vigorously stirred suspension kept at 08, Oxoneꢃ (31 mg, 0.005 mmol) was added in portions within 3 h.
The mixture was left overnight at r.t. Workup was performed by addition of 1M HCl (pH 4) followed by
extraction with CHCl3/EtOH 3 :1 (3”). The org. phase was dried (MgSO4) and the solvent evaporated to
afford a crude, which was converted to the corresponding methyl ester by standard reaction with diazo-
methane in Et2ACHTREOUNG for the GC and chromatographic analyses. Composition of the crude: starting fatty acid/
monoepoxy derivatives 85 :15. GC: 1.8 :1 ratio of 9,10-and 12,13-monoepoxy derivatives.
5. Epoxidation of Arachidonic Acid in Micelles. 4-{3-[(2Z,5Z,8Z)-Tetradeca-2,5,8-trienyl]oxiran-2-
yl}butanoic Acid (5,6-EET). Epoxidation of arachidonic acid was carried out by the in situ procedure
as described in Exper. 4 for linoleic acid. Two different experiments were carried out varying the amounts
of Oxoneꢃ and acetone for the DMD in situ formation, corresponding to 0.8 and 1.9 equiv. compared to
the arachidonic acid equiv. In the latter experiment, arachidonic acid (190 mg, 0.63 mmol) was suspended
in the standard medium composed by 0.9M KOH containing 1.12M NaHCO3 (pH 9.5). Then 2.7M acetone
in H2O (0.8 mmol, 0.3 ml) was added. The mixture was vortexed until a suspension was obtained. To the
vigorously stirred suspension kept at 08, Oxoneꢃ was added in portions within 5 h (500 mg, 0.8 mmol).
TLC Monitoring (hexane/Et2ACHTREOUGN /AcOH 1:3 :0.01) was done on aliquots of the mixture treated with 1M
HCl (pH 4), extracted with CHCl3/EtOH 3 :1 (3”). After 5 h, the mixture was worked up, as described
for the aliquot. The solvent was evaporated to give a colorless oil. An aliquot of this crude was converted
to the corresponding methyl esters by treatment with diazomethane in Et2ACHTREOUNG , and examined by GC. Prod-

2252
Helvetica Chimica Acta – Vol. 89 (2006)
uct composition: starting fatty acid (27%), monoepoxy derivatives (46%), and diepoxy derivatives
(27%). The composition of the monoepoxy derivatives was deduced from 13C-NMR spectra of the
crude, through the relative distribution of the signals of the olefinic C-atoms: 5,6-EET/8,9-EET/11,12-
EET/14,15-EET 3 :3 :1:0.7. Confirmation of this ratio was obtained by isolation after FC (silica gel,
TLC eluent): arachidonic acid (Rf 0.70; 51 mg, 0.17 mmol), 14,15-EET (Rf 0.62; 7 mg, 0.021 mmol),
11,12-EET/8,9-EETs (Rf 0.58; 38 mg, 0.12 mmol), and 5,6-EET (Rf 0.50; 29 mg, 0.094 mmol; isolated
yield 15%).
The same experiment was repeated with 0.8 equiv. of Oxoneꢃ after workup and separation as descri-
bed above, the results reported in Table 2 were obtained.
6. Epoxidation of Arachidonic Acid in CCl4. DMD (0.8 equiv. of a 0.15M CCl4 soln., 2.75 ml) was
added to arachidonic acid (100 mg, 0.33 mmol; final conc. 0.035M) in CCl4 (6.6 ml) (TLC monitoring
as described in Exper. 5). After 10 min, the solvent was evaporated to give a colorless oil. An aliquot
of this crude was converted to the corresponding methyl esters by treatment with diazomethane and
examined by GC. Final composition: starting material (23%), monoepoxy derivatives (47%), diepoxy
derivatives (19%). The composition of the monoepoxy derivatives was deduced from 13C-NMR spectra
of the fraction, through the relative distribution of the signals of the olefinic C-atoms: 14,15-EET/11,12-
EET/8,9-EET/5,6-EET 4.5 :3.3 :1.3 :1.
The experiment was repeated with 1.9 equiv. of DMD (0.627 mmol, 4.2 ml of a 0.15M CCl4 soln.) and
arachidonic acid (100 mg, 0.33 mmol; final conc. 0.030M) in CCl4 (6.6 ml). The analysis was performed as
described above: monoepoxy derivatives (23%), diepoxy derivatives (44%), and triepoxy derivatives
(24). Only 2% of starting material was recovered. 13C-NMR analysis: 14,15-EET/11,12-EET/8,9-EET
3 :3 :1 (Table 2).
REFERENCES
[1] C. R. Smith, M. O. Bagby, R. L. Lohmar, C. A. Glass, I. A. Wolff, J. Org. Chem. 1960, 25, 218.
[2] W. L. Smith, R. C. Murphy, in ꢀBiochemistry of Lipids, Lipoproteins and membranesꢁ, Eds. D. E.
Vance, and J. E. Vance, Elsevier, Amsterdam, 2002, p. 341.
[3] A. J. Draper, B. D. Hammock, Arch. Biochem Biophys. 2000, 376, 199.
[4] A. A. Spector, X. Fang, G. D. Snyder, N. L. Weintraub, Prog. Lipid Res. 2004, 43, 55.
[5] K. Kosaka, K. Suzuki, M. Hayakawa, S. Sugiyama, T. Ozawa, Mol. Cell. Biochem. 1994, 139, 141.
[6] T. Lu, T. Hoshi, N. L. Weintraub, A. A. Spector, H.-C. Lee, J. Physiol. 2001, 537, 811
[7] J.-C. Chen, D.-W. Wang, J. R. Falck, J. H. Capdevila, R. C. Harris, J. Biol. Chem. 1999, 274, 4764.
[8] R. J. Capon, R. A. Barrow, J. Org. Chem. 1998, 63, 75.
[9] E, J. Corey, H. Niwa, J. R. Falck, J. Am. Chem. Soc. 1979, 101, 1586.
[10] M. S. F. Lie Ken Lie, C. N. W. Lam, J. C. M. Ho, M. M. L. Lau, Eur. J. Lipid Sci. Technol. 2003, 105,
391.
[11] C. Ferreri, C. Costantino, L. Perrotta, L. Landi, Q. G. Mulazzani, C. Chatgilialoglu, J. Am. Chem.
Soc. 2001, 123, 4459.
[12] C. Ferreri, A. Samadi, F. Sassatelli, L. Landi, C. Chatgilialoglu, J. Am. Chem. Soc. 2004, 126, 1063.
[13] W. Adam, J. Bialas, L. Hadjiarapoglou, Chem. Ber. 1991, 124, 2377.
[14] M. Gibert, M. Ferrer, F. Sꢄnchez-Baeza, A. Messeguer, Tetrahedron 1997, 53, 8643.
[15] S. Narayan, J. Muldoon, M. G. Finn, V. V. Fokin, H. C. Kolb, K. B. Sharpless, Angew. Chem., Int. Ed.
2005, 44, 3275.
[16] A. Postigo, C. Ferreri, M. L. Navacchia, C. Chatgilialoglu, Synlett 2005, 2854.
[17] M. Ferrari, E. L. Ghisalberti, U. M. Pagnoni, F. Pelizzoni, J. Am. Oil Chem. Soc. 1968, 45, 649.
[18] R. J. Capon, R. A. Barrow, S. Rochfort, M. Jobling, C. Skene, Tetrahedron 1998, 54, 2227.
[19] S. D. Nguyen, D.-E. Sok, Biochem. J. 2003, 375, 275.
[20] A. R. Gallopo, J. O. Edwards, J. Org. Chem. 1981, 46, 1684.
[21] P. L. Luisi, M. Giomini, M. P. Pileni, B. H. Robinson, Biochim. Biophys. Acta 1988, 947, 209.
[22] J. R. Falck, P. Yadagiri, J. Capdevila, Methods Enzymol. 1990, 187, 357.
[23] J. R. Falck, S. Manna, H. R. Jacobson, R. W. Estabrook, N. Chacos, J. Capdevila, J. Am. Chem. Soc.
1984, 106, 3334.

Helvetica Chimica Acta – Vol. 89 (2006)
2253
[24] M. Ollmann, G. Schwarzmann, K. Sandhoff, H. J. Galla, Biochemistry 1987, 26, 5943
[25] H. Jiang, J. Quilley, L. Manmohan Reddy, J. R. Falck, P. Y.-K Wong, J. C. McGiff, Prostag. Other
Lipid Mediators 2005, 75, 65.
Received April 19, 2006