Synthesis of a fluorinated ether lipid analogous to a platelet activating factor

Article abstract of DOI:10.1002/1099-0690(200112)2001:23<4501::AID-EJOC4501>3.0.CO;2-J

The synthesis of racemic 2-fluoro-3-hexadecyloxy-2-methyl-prop-1-yl 2′-(trimethylammonio)ethyl phosphate (10), a fluorinated analogue of the anticancer active and blood platelet activating ether lipids 8 and 9, has been achieved in a six-step sequence from methallyl alcohol (11). Etherification of 11 with hexadecyl bromide gave allylic ether 12, bromofluorination of which afforded the bromo-substituted fluoride 13, which was subsequently transformed into the acetate 14. Hydrolysis of 14 gave the key intermediate 2-fluoro-2-(hexadecyloxymethyl)propanol (15), which was attached to the phosphocholine residue in the final step. Enzyme-catalyzed deracemization of the fluorohydrin 15 by acetylation with several lipases gave optically active compounds 14 and 15, with a maximum ee of 61% for 15 with Candida antarctica lipase catalysis after 74% conversion. An enantioselective synthesis of 10, based on a planned eight-step synthesis starting from α-methylstyrene, failed. The anticancer activity of racemic 10 has been observed in an in vivo model of methylcholanthrene-induced mouse fibrosarcoma.

Full text of DOI:10.1002/1099-0690(200112)2001:23<4501::AID-EJOC4501>3.0.CO;2-J

FULL PAPER
Synthesis of a Fluorinated Ether Lipid Analogous to a Platelet Activating
Factor
Günter Haufe*[a] and Annegret Burchardt[a]
Dedicated to Professor Dieter Hoppe on the occasion of his 60th birthday
Keywords: Allylic ether / Antitumor agents / Bromofluorination / Fluorine / Lipase-catalyzed deracemization / Lipids
The synthesis of racemic 2-fluoro-3-hexadecyloxy-2-methyl-
phocholine residue in the final step. Enzyme-catalyzed dera-
prop-1-yl 2Ј-(trimethylammonio)ethyl phosphate (10), a flu-
cemization of the fluorohydrin 15 by acetylation with several
orinated analogue of the anticancer active and blood platelet lipases gave optically active compounds 14 and 15, with a
activating ether lipids 8 and 9, has been achieved in a six- maximum ee of 61% for 15 with Candida antarctica lipase
step sequence from methallyl alcohol (11). Etherification of
11 with hexadecyl bromide gave allylic ether 12, bromofluo- of 10, based on a planned eight-step synthesis starting from
rination of which afforded the bromo-substituted fluoride 13,
α-methylstyrene, failed. The anticancer activity of racemic 10
catalysis after 74% conversion. An enantioselective synthesis
which was subsequently transformed into the acetate 14. Hy- has been observed in an in vivo model of methylchol-
drolysis of 14 gave the key intermediate 2-fluoro-2-(hexade- anthrene-induced mouse fibrosarcoma.
cyloxymethyl)propanol (15), which was attached to the phos-
Introduction
gation of the chemistry and biochemistry of biologically
active ether lipids. The cytotoxic effects of the PAF 1, and
in particular its lyso compounds 2, stimulated systematic
investigations into the cytotoxic properties of several syn-
thetic analogues of PAF.[6] The metabolic instability of 1
and 2[7] encouraged the development of more stable ether
lipids. One of the most intensely investigated antitumoral
ether lipids of this type is 2-O-methyl-1-O-octadecyl-rac-
glycero-3-phosphocholine (edelfosine, 3).[8] This compound
exhibits strong cytotoxicity against numerous human and
murine tumor cell strains both in vitro and in vivo. Clinical
tests have been performed.[9] Variation of the substituent
pattern around the stereogenic center of compounds 1 re-
sulted in substances of high potency against different tumor
cell strains. Tests of ilmofosine (4) and its oxygen analogue
5, as well as the C18 homologues, were very promising.[9,10]
A fluorinated analogue 6 also exhibited anticancer activ-
ity.[11]
Since the discovery of the physiological activity of com-
pounds 1 as platelet activating factor (PAF) in the early
1970s[1] and their first isolation from basophilic leuko-
cytes,[2] there has been growing interest in both natural and
synthetic ether lipids.[3] In 1979 the structure of natural
PAF was assigned as a mixture of two (R)-2-acetyl-1-O-
alkyl-sn-glycero-3-phosphocholines
1
(Scheme 1), with
C16H33 and C18H37 long-chain alkyl groups.[4]
The etiology of the cytotoxic activity of ether lipids is as
yet largely unknown. The anticancer activity of these com-
pounds is probably different from that of most other anti-
cancer drugs, which interfere with DNA synthesis or func-
tion.[12] Because of the structural relationship of these ether
lipids to PAF, the question arose of whether or not the PAF
receptor is decisive for their cytotoxicity. The PAF activity
of edelfosine (3) might be an indication of that cytotoxic
origin.[13] This cytotoxicity against tumor cells is not influ-
enced by PAF receptor antagonists;[14] the cyclic ether ana-
logue SRI 62-834 (7) binds to the receptor independently
of its cytotoxicity.[15] The ether lipids 8 and 9 exhibit pro-
nounced cytotoxicity against a number of human tumor cell
strains; this cytotoxicity is different for each enantio-
mer.[9,10] These ether lipids also influence the aggregation
of blood platelets (Scheme 2).[9,16]
Scheme 1
Structural elucidation and several syntheses[5] of the nat-
urally occurring enantiomers of 1 were followed by investi-
[a]
Organisch-Chemisches Institut, Westfälische Wilhelms-
Universität Münster,
Corrensstraße 40, 48149 Münster, Germany
Fax: (internat.) ϩ 49-(0)251/83-39772
E-mail: haufe@uni-muenster.de
Supporting information for this article is available on the
WWW under http://www.eurjoc.com or from the author.
Eur. J. Org. Chem. 2001, 4501Ϫ4507
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/1223Ϫ4501 $ 17.50ϩ.50/0
4501

G. Haufe, A. Burchardt
FULL PAPER
of 11 with hexadecyl bromide and potassium hydroxide in
DMSO, in analogy with a procedure used by Bittman et
al.,[10a] gave the ether 12 in 50% yield (Scheme 3). Sub-
sequent bromofluorination of 12 with NBS/Et3N·3HF[26]
(or even better with NBS/Me3N·3HF[27]) gave the Markov-
nikov-oriented bromo-substituted fluoride 13 with good re-
gioselectivity and in 77% yield. Neither the regioisomer of
13 nor other fluorinated compounds were detected in the
crude reaction product (19F NMR).[28]
Scheme 2
Many molecules are known in which substitution of a
fluorine for a hydrogen atom or a hydroxy function pro-
duces different chemical and physical properties, and biolo-
gical activity in the fluoro analog distinctly different from
that in the unmodified species. This change of activity is
presumably due to modification of some electronic, steric,
or general metabolic behavior.[17] The extreme electronega-
tivity of fluorine effects changes of electron density in mole-
cules and hence changes the acidities and reactivities of
neighboring groups.[18] The high electron density of the
fluorine substituent causes its low polarizability. However,
this high electron density also gives rise to the ability of
fluorine substituents to act as acceptors in intra- and inter-
molecular hydrogen bonds,[19] and these interactions can re-
sult in modified binding to a receptor.[20] The generally
higher metabolic stability of fluorinated compounds may be
responsible for inhibition of essential biochemical reac-
tions.[17b,20] Presumably, because of the similarity in the
Scheme 3. Reagents and conditions: (i) KOH, DMSO, 1-bromo-
hexadecane, room temp., 4 h; (ii) NBS, Me3N·3HF, CH2Cl2,
Ϫ10 °C, 96 h; (iii) KOAc, DMF, reflux, 24 h; (iv) KOH, MeOH,
room temp., 5 h; (v) Et3N, 2-chloro-2-oxo-1,3,2-dioxaphospholane,
THF, 0 °C 30 min, then room temp., 6 h, separation of Et3N·HCl;
(vi) Ϫ78 °C, Me3N, MeCN, then 60 °C, sealed tube, 68 h
The bromofluoride 13 was transformed into 14 in 73%
yield without significant elimination of hydrogen bromide,
by use of a previously developed recipe (KOAc, DMF, re-
flux[11]).[29] Fluorohydrin 15 was obtained in 77% yield by
saponification of 14.
sizes of hydrogen and fluorine substituents (van der Waals
[21]
˚
˚
radii of H ϭ 1.20 A and of F ϭ 1.47 A),
fluorinated
analogues of biologically active compounds often mimic the
natural metabolism of the parent compounds.[22] In many
cases partial fluorine substitution increases both the lipo-
philicity and the hydrophilicity of a particular molecule,
which improves its resorption and its transport in the or-
ganism.[17b,23]
Among the known synthetic ether lipids with cytotoxic
and PAF-antagonistic properties there are only a few
monofluorinated compounds.[7,24] Brachwitz et al. have
shown that the racemic 2-deoxy-2-fluoro-1-O-hexadecyl-
rac-glycero-3-phosphocholine (3, X ϭ F, R ϭ C16H33) and
edelfosine (3) have almost identical cytotoxic activities
against EhrlichϪAscites tumor. Compound 3 also has PAF-
antagonistic properties.[25] In view of these facts we thought
it desirable to synthesize a fluorinated analogue 10 of both
the PAF active ether lipids 8 and 9. We believed that the
analogue 10 might exhibit some interesting physiological
behavior.
In order to synthesize the enantiomers of 15, we screened
several lipases previously applied with related sub-
strates[11,30] for deracemization by acetylation in organic
solvents, using vinyl acetate as the acylating reagent
(Scheme 4, Table 1). Porcine pancreas lipase (PPL) was the
slowest enzyme, giving 48% conversion after 168 h reaction
time. The observed enantiomeric excess was as low as 17%
ee for the residual fluorohydrin 15 and 23% ee for the
formed acetate 14. The results with Candida rugosa lipase
(CRL) or immobilized Pseudomonas cepacia lipase (PCL,
Amano PS) were only slightly better. With Candida ant-
arctica lipase, the fastest enzyme under the conditions used,
a 61% ee for the fluorohydrin 15 and a 25% ee for the
acetate 14 were obtained after 74% conversion. Thus, the
enantioselectivities of all the tested lipases were quite low
(Table 1).
For the final coupling of the racemic fluorohydrin 15 to
the choline moiety, a procedure described by Guivisdalsky
and Bittman[33] for a similar (but non-fluorinated) long-
chain hydroxy ether was used. Compound 15 was treated
with a threefold excess of 2-chloro-2-oxo-1,3-dioxa-2-phos-
Results and Discussion
Retrosynthetic analysis of racemic 10 suggested a linear pholane and triethylamine in THF at room temperature.
synthesis starting from methallyl alcohol (11). O-Alkylation The crude cyclic phosphoric acid triester intermediate 16
4502
Eur. J. Org. Chem. 2001, 4501Ϫ4507

Synthesis of a Fluorinated Ether Lipid Analogous to a Platelet Activating Factor
FULL PAPER
periodate in a two-phase system of tetrachloromethane and
water.[34] This procedure was improved by Sharpless et
al.[35] and was used by Murato and Kitazume for the oxida-
tion of (1R)-1-acetoxy-2,2-difluoro-1-phenylethane to (2R)-
2-acetoxy-3,3-difluoropropanoic acid in 35% yield.[36] We
adapted the Sharpless conditions for the oxidation of 19 to
20, which was obtained in a 67% yield in a preparative scale
transformation by use of 6.6 mol % of RuCl3·H2O and 23
equiv. of NaIO4 in a 2:2:3 mixture of water/tetrachloro-
methane/acetonitrile at room temperature (Scheme 6).[37]
Table 1. Results of the lipase-catalyzed deracemization of the
fluorohydrin 15 by acetylation with vinyl acetate in toluene
Entry Enzyme
[mg]
Time Conversion Product Yield ee[a] E[b]
[h]
[%]
[%]
[%]
1
2
3
4
PPL[150]
CRL[35]
168 48
15 48
15
14
15
14
15
14
15
14
33
29
45
33
47
48
22
62
17 1.7
23 1.1
28 2.4
29 2.3
27 2.2
23 2.0
61 2.6
25 3.2
immobilized
PCL[70]
4
1
50
74
CAL[5]
[a] Determined by 19F NMR spectroscopy after esterification of the
fluorohydrins 15 (after hydrolysis of 14 if necessary) by Mosher’s
method,[31] by integration of the signals at δ ϭ Ϫ158.22 and δ ϭ
Ϫ158.43. Ϫ [b] The E value is a measure for the selectivity of an en-
zyme.[32]
Scheme 6. Reagents and conditions: (i) NaIO4 (23 equiv.), RuCl3
(6.6 mol %), H2O/CCl4/CH3CN, 48 h, room temp.; (ii) DMP,
MeOH
This procedure was applied to oxidize the optically active
19 in order to examine whether or not the oxidation pro-
ceeds without racemization. Oxidation of (S)-(ϩ)-19 (53%
ee), obtained by a preparative-scale enzyme-catalyzed
acetylation of the corresponding fluorohydrin according to
the procedure described in ref.[30b], thus gave the optically
active carboxylic acid 20. The determination of the enantio-
meric excess of (R)-20 was difficult.[38] After several failures,
the acetate (R)-20 was finally transformed into the hydroxy-
carboxylic ester (R)-21 in 74% yield by treatment with 2,2-
dimethoxypropane (DMP) in methanol. The enantiomeric
excess (50% ee) was determined by 19F NMR spectroscopy
after derivatization with Mosher’s acid[31] and DCC/
DMAP. The optically active building block 21 has previ-
ously been synthesized by several different procedures and
was used in the various syntheses.[39]
Table 1. Reagents and conditions: lipase, vinyl acetate, toluene,
room temp. (cf. Table 1)
was used without isolation. After 68 h at 60 °C in a sealed
tube with trimethylamine in acetonitrile, the resulting phos-
phocholine derivative 10 was isolated in 40% yield; 9% over-
all, based on methallyl alcohol (11) (Scheme 3).
Since deracemization of 15 had given the desired prod-
ucts with quite a low enantiomeric excess, an alternative
sequence was designed. Retrosynthetic analysis of 10 sug-
gested a highly functionalized, crucial C4 moiety 21, which
should be available from α-methylstyrene (17) via 18, 19,
and 20 (Scheme 5).
We tested the next step of the planned synthetic sequence
to form 10 with racemic 21. However, treatment of 21 with
NaH and either hexadecyl bromide or hexadecyl mesyl-
ate[42] gave only 4% of the desired long-chain ether 22
(Scheme 7). The main reaction was most probably an inter-
molecular transesterification resulting in higher molecular
weight material, which was not fully characterized. Most of
the starting mesylate was recovered.
Scheme 5
The synthesis of 19 from 17 by bromofluorination using
the NBS/Me3N·3HF combination[27] followed by removal
of the bromo substituent in the produced 1-bromo-2-
fluoro-2-phenylpropane (18) by acetate exchange with
KOAc in DMF has already been described.[30b] Oxidation
Scheme 7. Reagents and conditions: NaH, hexadecylmesylate,
Et2O, 6 h reflux
We therefore decided to modify the components of the
of the phenyl ring is a necessary next step. This type of Williamson ether synthesis. The bromo-substituted fluoride
oxidation of aromatic compounds was first achieved by 18 was oxidized in 70% yield to the C4 building block 23,
Caputo and Fuchs, with ruthenium tetraoxide and sodium which was esterified with 2,2-dimethoxypropane in meth-
Eur. J. Org. Chem. 2001, 4501Ϫ4507
4503

G. Haufe, A. Burchardt
FULL PAPER
saturated aqueous solution of NaCl and extracted with 3 ϫ 200 mL
of diethyl ether. The combined ethereal layers were washed with 2
ϫ 200 mL of water and dried with magnesium sulfate. The solvent
was removed and the residue chromatographed (silica gel; pentane/
diethyl ether, 20:1) to give 12 as a colorless oil (10% 1-bromohexa-
anol to give 24 in 50% yield (Scheme 8). Unfortunately, the
conditions used by Bittman et al.[10a] for etherification of a
similar, but non-fluorinated, compound did not work in our
case. Treatment of 24 with copper hexadecanolate, formed
in situ from hexadecanol, butyllithium, and copper(I) iod-
ide, did not give the desired ether 22. In addition to un-
changed 24, 14% of the transesterification product 25 was
isolated as the only reaction product. Thus, the attempt to
synthesize the optically active key intermediate 21 starting
from α-methylstyrene 17 was not successful.
1
decane impurity). Yield: 8.3 g (50%). The H NMR spectroscopic
data agree with those published.[16a] 13C NMR: δ ϭ 14.1 (q, 16Ј-
C), 19.4 (q, 4-C), 22.7 (t, 15Ј-C), 26.3 (t, 3Ј-C), 29.4Ϫ29.8 (11 t, 4Ј-
C to 14Ј-C), 31.9 (t, 2Ј-C), 70.3 (t, 1Ј-C), 74.8 (t, 3-C), 111.6 (t, 1-
C), 142.6 (q, 2-C). GC/MS: m/z (%) ϭ 296 (26) [Mϩ·], 72 (100)
[C4H8Oϩ·].
1-Bromo-2-fluoro-3-hexadecyloxy-2-methylpropane (13): A solution
of 12 (4.0 g, 13.5 mmol) and Me3N·3HF (2.4 mL, 26.1 mmol) was
treated in portions with NBS (2.5 g, 14.9 mmol) in CH2Cl2 (50 mL)
and stirred at Ϫ10 °C for 96 h. The mixture was poured into
250 mL of ice/water and neutralized with 26% aqueous ammonia.
The organic layer was separated and the aqueous layer was ex-
tracted with 3 ϫ 50 mL of ethyl acetate. The combined organic
layers were washed with 25 mL of 0.1  HCl followed by 25 mL of
aqueous NaHCO3 solution and dried with magnesium sulfate. The
solvent was removed in vacuo, and the residue was purified by
column chromatography (silica gel; pentane/diethyl ether, 100:1) to
give 13 as a colorless oil (10% impurity of 1-bromohexadecane).
Yield: 4.1 g (77%). 1H NMR: δ ϭ 0.88 (t, 3JH,H ϭ 6.7 Hz, 3 H, 16Ј-
H), 1.21Ϫ1.38 (m, 26 H, 3Ј-H to 15Ј-H), 1.47 (d, 3JH,F ϭ 21.5 Hz, 3
H, 4-H), 1.55Ϫ1.62 (m, 2 H, 2Ј-H), 3.43Ϫ3.65 (m, 6 H, 1Ј-H, 1-H,
Scheme 8. Reagents and conditions: (i) NaIO4 (23 equiv.), RuCl3
(6.6 mol %), H2O/CCl4/CH3CN, 48 h, room temp.; (ii) DMP,
MeOH; (iii) hexadecanol, nBuLi, THF, CuI (10 mol %), Ϫ78 °C,
6 h
The anticancer activity of racemic 10 was tested in an in
vivo model of methylcholanthrene-induced mouse fibrosar-
coma, comparing the activity of 10 to that of cisplatin and
ilmofosine (4). While significant anticancer activity (IC50 ϭ
1.82 mg/mL, 48 h incubation time) was found for 10, this
activity was lower than that of cisplatin (IC50 ϭ 0.17 mg/
mL) or ilmofosine (4) (IC50 ϭ 0.23 mg/mL), all measured
under identical experimental regimes.
2
3-H). 13C NMR: δ ϭ 14.1 (q, 16Ј-C), 21.3 (dq, JC,F ϭ 22.9 Hz, 4-
C), 22.7 (t, 15Ј-C), 26.0 (t, 3Ј-C), 29.3 to 29.7 (11t, 4Ј-C to 14Ј-C),
2
31.9 (t, 2Ј-C), 35.6 (dt, JC,F ϭ 28.0 Hz, 1-C), 72.1 (t, 1Ј-C), 73.4
2
1
(dt, JC,F ϭ 25.4 Hz, 3-C), 94.3 (ds, JC,F ϭ 175.5 Hz, 2-C). 19F
NMR: δ ϭ Ϫ152.6 (m). GC/MS: m/z (%) ϭ 374/376 (0.7/0.7) [Mϩ·
Ϫ HF], 315 (31) [Mϩ· Ϫ Br], 295 (1) [315 Ϫ HF], 255 (100)
[C16H33OCH2ϩ], 225 (31) [C16H33ϩ], 224 (24) [C16H32ϩ·], 196 (6)
[C14H28ϩ·], 183 (6) [C13H27ϩ], 168 (8) [C12H24ϩ·], 167 (4) [C12H23ϩ],
155 (16) [C11H23ϩ], 153 (11) [C11H21ϩ], 141 (19) [C10H21ϩ], 139 (12)
[C10H19ϩ], 127 (19) [C9H19ϩ], 125 (19) [C9H17ϩ], 113 (16) [C8H17ϩ],
111 (30) [C8H15ϩ], 99 (41) [C7H15ϩ], 98 (10) [C7H14ϩ·], 97 (57)
[C7H13ϩ], 96 (10) [C7H12ϩ·], 85 (44) [C6H13ϩ], 84 (38) [C6H12ϩ·], 83
(68) [C6H11ϩ], 82 (24) [C6H10ϩ·], 71 (93) [C5H11ϩ], 70 (26)
[C5H10ϩ·], 69 (71) [C5H9ϩ], 68 (20) [C5H8ϩ·], 57 (95) [C4H9ϩ]. GC/
MS (CI): m/z ϭ 412/414 [M ϩ NH4ϩ].
Experimental Section
General Remarks: 1H (300.1 MHz), 13C (75.5 MHz), and 19F NMR
(282.3 MHz) were recorded in ca. 20% solutions in CDCl3 (if not
stated otherwise). Chemical shifts are reported as δ values [ppm]
relative to TMS (1H), CDCl3 (13C), or CFCl3 (19F), respectively, as
internal standards. For 31P NMR (121.5 MHz), 80% phosphoric
acid was used as an external standard. The multiplicity of 13C sig-
nals was determined by DEPT. Mass spectra (electron impact ion-
ization, 70 eV) were recorded by GC/MS coupling. The composi-
tions of crude reaction products and the conversions of substrates
during enzymatic transformations were monitored by GC with
quartz capillary columns: 25 m ϫ 0.33 mm, 0.52 µm HP-1
(HewlettϪPackard) and 30 m ϫ 0.32 mm, 0.25 µm SPB-1
(Supelco), temperature program, 40 °C Ǟ 280 °C with a 10 °C/min
heating rate, N2 as the carrier gas. Compound ratios were deter-
mined by integration of the peak areas. The products, including
those of enzymatic transformations, were separated by column
chromatography (silica gel, Merck 60, 70Ϫ230 mesh). Microana-
lyses were carried out by the ‘‘Mikroanalytisches Laboratorium,
Organische Chemie’’, University of Münster, with a CHN-O ana-
lyzer.
1-Acetoxy-2-fluoro-3-hexadecyloxy-2-methylpropane (14): A mix-
ture of 13 (3.2 g, 8.4 mmol) and potassium acetate (1.9 g,
33.5 mmol) in dry DMF (50 mL) was refluxed for 24 h. After the
mixture had cooled to room temperature, a 1:1 mixture of cyclo-
hexane and ethyl acetate (50 mL) was added, and the precipitate
was filtered off. The organic layer was washed with 20 mL of water
and dried with magnesium sulfate. The solvent was vacuum-
evaporated and the residue was purified by column chromato-
graphy (silica gel; pentane/diethyl ether, 20:1) to give pure 14 as a
colorless oil, which became a waxy solid in the refrigerator at 4 °C.
Yield: 2.3 g (73%). 1H NMR: δ ϭ 0.88 (t, 3JH,H ϭ 6.7 Hz, 3 H, 16Ј-
H), 1.23Ϫ1.36 (m, 26 H, 3Ј-H to 15Ј-H), 1.37 (d, 3JH,F ϭ 21.7 Hz, 3
H, 4-H), 1.51Ϫ1.58 (m, 2 H, 2Ј-H), 2.09 (s, 3 H, 6-H), 3.47 (t,
3JH,H ϭ 6.5 Hz, 2 H, 1ЈH,), 3.46 and 3.53 (2 dd, 3JHa,F ϭ 15.3 Hz,
2
3JHb,F ϭ 16.5 Hz, JHa,Hb ϭ 10.3 Hz, 2 H, 3-H), 4.17 and 4.23 (2
3
3
2
dd, JHc,F ϭ 19.3 Hz, JHd,F ϭ 19.8 Hz, JHc,HdЈ ϭ 11.9 Hz, 2 H,
2
3-Hexadecyloxy-2-methylpropene (12): In analogy with the proced-
1-H). 13C NMR: δ ϭ 14.1 (q, 16Ј-C), 19.4 (dq, JC,F ϭ 22.9 Hz, 4-
ure given in ref.[40], powdered KOH (4.3 g, 15.0 mmol) and meth-
C), 20.7 (q, 6-C), 22.6 (t, 15Ј-C), 26.0 (t, 3Ј-C), 29.3 to 29.7 (11t,
2
allyl alcohol (11, 4.3 g, 55.6 mmol) in DMSO (40 mL) were treated 4Ј-C to 14Ј-C), 31.9 (t, 2Ј-C), 66.2 (dt, JC,F ϭ 25.4 Hz, 1-C), 72.1
with 1-bromohexadecane (58.1 g, 0.2 mol) and stirred at room tem-
(t, 1Ј-C), 73.0 (dt, 2JC,F ϭ 28.0 Hz, 3-C), 94.4 (ds, 1JC,F ϭ 172.9 Hz,
perature for 3.5 h. The mixture was then poured into 500 mL of a 2-C), 170.4 (s, 5-C). 19F NMR: δ ϭ Ϫ160.0 (m). GC/MS: m/z (%) ϭ
4504
Eur. J. Org. Chem. 2001, 4501Ϫ4507

Synthesis of a Fluorinated Ether Lipid Analogous to a Platelet Activating Factor
FULL PAPER
374 (1) [Mϩ·], 354 (18) [Mϩ· Ϫ HF], 311 (5) [354 Ϫ CH3CO], 294 6.8 Hz, 3 H, 16Ј-H), 1.24Ϫ1.37 (m, 26 H, 3Ј-H to 15Ј-H), 1.38 (d,
(30) [354 Ϫ CH3COOH], 255 (20) [C16H33OCH2ϩ], 225 (15) 3JH,F ϭ 21.9 Hz, 3 H, 4-H), 1.53Ϫ1.62 (m, 2 H, 2Ј-H), 3.24 (s, 9
[C16H33ϩ], 196 (2) [C14H28ϩ·], 183 (4) [C13H27ϩ], 169 (8) [C12H25ϩ],
H, 7-H to 9-H), 3.47Ϫ3.58 (m, 4 H, 1Ј-H, 3-H), 3.65Ϫ3.68 (m, 2
167 (3) [C12H23ϩ], 155 (7) [C11H23ϩ], 153 (3) [C11H21ϩ], 151 (27) H, 6-H), 3.88Ϫ4.08 (m, 2 H, 1-H), 4.26Ϫ4.35 (m, 2 H, 5-H). 13C
[Mϩ· Ϫ C16H31], 141 (9) [C10H21ϩ], 139 (5) [C10H19ϩ], 133 (18) [151 NMR: δ ϭ 13.2 (q, 16Ј-C), 18.1 (dq, JC,F ϭ 23.6 Hz, 4-C), 22.0
2
Ϫ H2O], 131 (28) [Mϩ· Ϫ C16H33 Ϫ H2O], 127 (10) [C9H19ϩ], 125 (t, 15Ј-C), 25.4 (t, 3Ј-C), 28.7 to 29.1 (11t, 4Ј-C to 14Ј-C), 31.3 (t,
1
2
(6) [C9H17ϩ], 113 (16) [C8H17ϩ], 111 (13) [C8H15ϩ], 99 (18) 2Ј-C), 53.4 (3tq, JC,N ϭ 3.5 Hz, 7-C to 9-C), 58.7 (dt, JC,P
ϭ
2
2
[C7H15ϩ], 98 (6) [C7H14ϩ·], 97 (31) [C7H13ϩ], 96 (9) [C7H12ϩ·], 85
(47) [C6H13ϩ], 84 (8) [C6H12ϩ·], 83 (33) [C6H11ϩ], 82 (13) [C6H10ϩ·],
71 (100) [C5H11ϩ], 70 (38) [C5H10ϩ·], 69 (37) [C5H9ϩ], 68 (9)
4.2 Hz, 5-C), 65.8 (mt, 6-C), 67.7 (ddt, JC,F ϭ 25.0 Hz, JC,P
ϭ
2
5.0 Hz, 1-C), 71.6 (t, 1Ј-C), 72.3 (dt, JC,F ϭ 26.4 Hz, 3-C), 94.9
(dds, JC,F ϭ 172.7 Hz, JC,P ϭ 7.5 Hz, 2-C). 19F NMR (CDCl3):
1
3
[C5H8ϩ·], 57 (88) [C4H9ϩ]. C22H43FO3 (374.6): calcd. C 70.54, H δ ϭ Ϫ159.7 (ttq, JH,F ϭ 19.1 Hz, JH,F ϭ 21.0 Hz). 31P NMR
3
3
11.57; found C 70.36, H 11.48.
(CDCl3/CD3OD, 1:1): δ ϭ 5.59 (m). MS (Maldi-TOF): m/z ϭ
498Ϯ0.05% [Mϩ ϩ H].
2-Fluoro-3-hexadecyloxy-2-methylpropan-1-ol (15): A solution of 14
(2.0 g, 5.3 mmol) and KOH (1.68 g, 30.0 mmol) in methanol
(10 mL) was stirred at room temperature for 3 h. The mixture was
poured into 25 mL of water and extracted with 3 ϫ 10 mL of di-
ethyl ether. The combined ethereal extracts were washed with
10 mL of water, dried with magnesium sulfate, and filtered through
silica gel (5 cm). After evaporation of the solvent, the fluorohydrin
15 was obtained as a white solid. Yield: 1.37 g (77%). M.p. 41 °C
Lipase-Catalyzed Acetylation of 2-Fluoro-3-hexadecyloxy-2-methyl-
propan-1-ol (15): The enzyme (occasionally immobilized, cf.
Table 1) was added to a solution of the fluorohydrin 15 (134 mg,
0.4 mmol) and vinyl acetate (35 mg, 0.4 mmol) in toluene (3 mL).
The mixture was stirred at room temperature and the progress of
the acetylation was monitored by gas chromatography after
workup of 100-µL aliquots of the reaction mixture. After the length
of time given in Table 1, the reaction was stopped by filtration of
the mixture through a short column of silica gel (toluene as the
eluent). The solvent was evaporated, and the residue was separated
by column chromatography (silica gel; pentane/diethyl ether, 10:1).
The enantiomeric excesses of the fluorohydrin 15 and of the acetate
14, after hydrolysis according to the procedure given above, were
determined by 19F NMR spectroscopy (Table 1) after esterification
with (R)-(ϩ)-Mosher’s acid[31] (cf. below).
3
(diethyl ether). 1H NMR: δ ϭ 0.88 (t, JH,H ϭ 6.7 Hz, 3 H, 16Ј-
3
H), 1.23Ϫ1.35 (m, 26 H, 3Ј-H to 15Ј-H), 1.34 (d, JH,F ϭ 22.2 Hz,
3 H, 4-H), 1.52Ϫ1.62 (m, 2 H, 2Ј-H), 2.02 (s, 1 H, 1-OH),
3
3.42Ϫ3.64 (m, 4 H, 1Ј-H, 3-H), 3.67 (dd, JHa,F ϭ 16.7 Hz,
3
2JHa,Hb ϭ 12.2 Hz, 1 H, 1-Ha), 3.73 (dd, JHb,F ϭ 19.8 Hz,
2JHa,Hb ϭ 12.2 Hz, 1 H, 1-Hb). 13C NMR: δ ϭ 14.1 (q, 16Ј-C), 19.5
2
(dq, JC,F ϭ 22.9 Hz, 4-C), 22.7 (t, 15Ј-C), 26.0 (t, 3Ј-C), 29.3 to
2
29.7 (11t, 4Ј-C to 14Ј-C), 31.9 (t, 2Ј-C), 66.6 (dt, JC,F ϭ 25.4 Hz,
1-C), 72.2 (t, 1Ј-C), 73.8 (dt, 2JC,F ϭ 28.0 Hz, 3-C), 95.9 (ds, 1JC,F ϭ
170.4 Hz, 2-C). 19F NMR: δ ϭ Ϫ163.1 (m). GC/MS: m/z (%) ϭ
332 (0.8) [Mϩ·], 312 (25) [Mϩ· Ϫ HF], 255 (6) [C16H33OCH2ϩ], 253
(7) [C16H31OCH2ϩ], 225 (17) [C16H33ϩ], 224 (5) [C16H32ϩ·], 196 (5)
[C14H28ϩ·], 183 (3) [C13H27ϩ], 169 (4) [C12H25ϩ], 167 (1) [C12H23ϩ],
155 (5) [C11H23ϩ], 153 (4) [C11H21ϩ], 141 (6) [C10H21ϩ], 139 (5)
[C10H19ϩ], 127 (8) [C9H19ϩ], 125 (11) [C9H17ϩ], 113 (11) [C8H17ϩ],
111 (19) [C8H15ϩ], 99 (18) [C7H15ϩ], 98 (6) [C7H14ϩ·], 97 (31)
[C7H13ϩ], 96 (10) [C7H12ϩ·], 89 (17) [Mϩ· Ϫ C16H33 Ϫ H2O], 88
(18), 85 (49) [C6H13ϩ], 84 (9) [C6H12ϩ·], 83 (38) [C6H11ϩ], 82 (17)
[C6H10ϩ·], 72 (40) [Mϩ· Ϫ C16H33OH Ϫ H2O], 71 (76) [C5H11ϩ], 70
(28) [C5H10ϩ·], 69 (38) [C5H9ϩ], 68 (11) [C5H8ϩ·], 57 (100) [C4H9ϩ].
C20H41FO2 (332.6): calcd. C 72.24, H 12.43; found C 72.29, H
12.45.
3-Acetoxy-2-fluoro-2-methylpropanoic Acid (20): A suspension of
NaIO4 (60.02 g, 280.6 mmol) and RuCl3·H2O (374 mg, 6.6 mol %)
in a mixture of water (96 mL), CCl4 (96 mL), and CH3CN
(192 mL) was stirred at room temperature for 1 h. The fluoro acet-
ate 19 (2.39 g, 12.2 mmol), prepared in 69% overall yield from α-
methylstyrene (17) according to ref.[30b], was then added and the
mixture was stirred at room temperature for 72 h. The solid was
filtered off and washed with 2 ϫ 25 mL of CHCl3. The aqueous
layer was separated and extracted with 3 ϫ 25 mL of CHCl3. The
combined organic layers were dried with magnesium sulfate and
the solvent was removed in vacuo to give a hygroscopic, white solid.
1
3
Yield: 1.21 g (61%). H NMR: δ ϭ 1.57 (d, JH,F ϭ 21.0 Hz, 3 H,
4-H) 2.04 (s, 3 H, 6-H) 4.24Ϫ4.35 (m, 2 H, 3-H), 9.55 (br, 1 H,
2
OH). 13C NMR: δ ϭ 20.3 (dq, JC,F ϭ 25.4 Hz, 4-C), 20.5 (q, 6-
2
C), 66.6 (dt, JC,F ϭ 22.9 Hz, 3-C), 92.7 (ds,1JC,F ϭ 188.2 Hz, 2-
2-Fluoro-3-hexadecyloxy-2-methylprop-1-yl 2Ј-(Trimethylammonio)-
2
C), 170.5 (s, 5-C), 174.2 (ds, JC,F ϭ 28.0 Hz, 1-C). 19F NMR: δ ϭ
ethyl Phosphate (10): In analogy with the procedure given in ref.[41]
,
3
Ϫ162.4 (tq, JH,F ϭ 21.0 Hz). GC/MS of the Si(CH3)3 ester: m/z
a mixture of the fluorohydrin 15 (183 mg, 0.55 mmol) and triethyl-
amine (250 µL, 1.73 mmol) in dry THF (20 mL) was prepared in
an argon-flushed, dried Schlenk vessel. A solution of 2-chloro-2-
oxo-1,3,2-dioxaphospholane (160 µL, 1.70 mmol) in dry THF
(8 mL) was added to this flask at 0 °C, and the resulting mixture
was stirred at this temperature for 30 min. The mixture was then
allowed to warm to room temperature and stirred for an additional
6 h. The precipitated triethylammonium chloride was filtered
through silica gel under argon (1 cm, THF as eluent), and the sol-
vent was evaporated in vacuo. The residue was dissolved in 20 mL
of dry acetonitrile under argon, placed in a pressure vessel, and
(%) ϭ 221 (8) [Mϩ Ϫ CH3], 179 (98) [221 Ϫ C2H2O], 164 (16), 161
(11) [221
Ϫ C2H4O2], 133 (34) [161 Ϫ CO], 117 (72)
[CO2Si(CH3)3ϩ], 85 (42), 77 (41), 75 (54) [C2H7OSiϩ], 73 (97)
[Si(CH3)3ϩ], 43 (100) [C2H3Oϩ]. GC/MS [CI, of the Si(CH3)3 ester]:
m/z (%) ϭ 254 [M ϩ NH4ϩ]. (S)-(Ϫ)-1-Acetoxy-2-fluoro-2-phenyl-
propane, (S)-(Ϫ)-19 (53% ee), was also oxidized to (R)-20 (50% ee)
by this procedure. The enantiomeric excess of (R)-20 was deter-
mined by 19F NMR after transformation into (R)-21 and ester-
ification with (R)-(ϩ)-Mosher’s acid.[31]
Methyl 2-Fluoro-3-hydroxy-2-methylpropanoate (21): A solution of
cooled to Ϫ78 °C. Trimethylamine (2.5 mL) was condensed into 20 (1.0 g, 6.1 mmol) in dry methanol (21 mL) was treated with 2,2-
the vessel, which was sealed and heated at 60 °C for 68 h. The
resulting liquid was separated and maintained at Ϫ20 °C for 4 h.
The precipitated white solid was filtered off and purified by column
dimethoxypropane (200 µL) and one drop of conc. HCl. The solu-
tion was stirred at room temperature for 32 h, methanol was re-
moved in vacuo, and the residue was dissolved in CHCl3 (10 mL).
chromatography (silica gel; chloroform/methanol/water, 65:25:4) to The solution was washed with 5 mL of 5% aqueous NaHCO3 solu-
obtain, in addition to starting material, the ether lipid 10 as a white,
tion and dried with magnesium sulfate, and the solvent was re-
moved in vacuo to yield 20 as a yellowish oil. Yield: 610 mg (74%).
1
3
waxy solid. Yield: 110 mg (40%). H NMR: δ ϭ 0.89 (t, JH,H
ϭ
Eur. J. Org. Chem. 2001, 4501Ϫ4507
4505

G. Haufe, A. Burchardt
FULL PAPER
3
1H NMR: δ ϭ 1.53 (d, JH,F ϭ 21.5 Hz, 3 H, 4-H) 2.08 (br. s, 1 zymes, respectively, and Dr. D. Herrmann, Heidelberg Pharma
2
3
H, OH) 3.75Ϫ3.96 (2dd, JHa,Hb ϭ 12.4 Hz, JHb,F ϭ 16.0 Hz,
Holding, GmbH & Co KG, Germany, for biological tests.
3JHa,F ϭ 25.3 Hz, 2 H, 3-H), 3.83 (s, 3 H, 5-H). 13C NMR: δ ϭ
2
2
19.7 (dq, JC,F ϭ 22.9 Hz, 4-C), 52.8 (q, 5-C), 66.8 (dt, JC,F
ϭ
ϭ
22.9 Hz, 3-C), 95.5 (ds,1JC,F ϭ 185.60 Hz, 2-C), 171.1 (ds, JC,F
2
[1]
J. Benveniste, P. M. Henson, C. G. Cochrane, J. Exp. Med.
25.4 Hz, 1-C). 19F NMR: δ ϭ Ϫ165.0 (m). GC/MS: m/z (%) ϭ 136
(0.2) [Mϩ·], 135 (0.1) [Mϩ· Ϫ 1], 116 (4) [Mϩ· Ϫ HF], 106 (100)
[Mϩ· Ϫ CH2O], 105 (5) [135 Ϫ CH2O], 77 (36) [Mϩ· Ϫ C2H3O2],
74 (57) [C3H6O2ϩ], 59 (23) [C3H4Fϩ; C2H3O2ϩ]. GC/MS (CI):
m/z ϭ 154 [M ϩ NH4ϩ]. C5H9FO3 (136.1): calcd. C 44.12, H 6.66;
found C 43.82, H 6.87. By the same procedure, (R)-20 (40 mg,
0.24 mmol) was transformed into (R)-21. Yield: 20 mg (61%). The
enantiomeric excess of this compound was determined by Mosher’s
method:[31] DCC (23 mg, 110 µmol) and DMAP (a few crystals)
were added to a solution of (R)-21 (5 mg, 37 µmol) and (R)-(ϩ)-
Mosher’s acid (26 mg, 110 µmol) in CH2Cl2 (1 mL). The resulting
mixture was stirred at room temperature for 14 h, and the solid
material was filtered off and washed with 1 mL of CH2Cl2. The
organic layers were combined and the solvent was removed. The
proton-decoupled 19F NMR spectrum of the crude residue (re-
duced measuring range δ ϭ Ϫ150 to Ϫ180) showed two baseline-
separated signals for the CF groups at δ ϭ Ϫ162.1 and δ ϭ Ϫ162.2.
The integration showed a ratio of 75:25, hence 50% ee [the small
difference in the ee values of (S)-19 and (R)-21 is probably due to
the different methods, GC or 19F NMR, respectively, used for the
determination]. The signals of the CF3 groups at δ ϭ Ϫ72.44 and
δ ϭ Ϫ72.49 were not baseline-separated and could not be used for
the determination of the ee.
1972, 136, 1356Ϫ1375.
[2]
D. J. Hanahan, C. A. Demopoulos, L. Liehr, R. N. Pinckard,
J. Biol. Chem. 1980, 255, 5514Ϫ5516.
[3] [3a]
H. K. Mangold, Angew. Chem. 1979, 91, 550Ϫ560; Angew.
[3b]
Chem. Int. Ed. Eng. 1979, 18, 493Ϫ502.
N. Weber, Lipids 1987, 22, 789Ϫ799.
H. K. Mangold,
[4]
[5]
C. A. Demopoulos, R. N. Pinckard, D. J. Hanahan, J. Biol.
Chem. 1979, 254, 9355Ϫ9358.
[5a] H. K. Mangold, Prog. Biochem. Pharmacol. 1988, 22, 1Ϫ16.
[5b]
R. K. Erukulla, H.-S. Byun, R. Bittman, J. Org. Chem.
1995, 60, 7706Ϫ7708.
[6] [6a]
M. H. Runge, R. Andreesen, A. Pfleiderer, P. G. Munder,
[6b]
J. Natl. Cancer Inst. 1980, 64, 1301Ϫ1306.
E. A. M: Fleer,
D.-J. Kim, G. A. Nagel, H. Eibl, C. Unger, Onkologie 1990,
13, 295Ϫ300.
[7]
[8]
F. Snyder, Med. Res. Rev. 1985, 5, 107Ϫ140.
P. G. Munder, H. Fischer, H. U. Weltzein, Proc. Am. Assoc.
Cancer Res. 1976, 17, 174Ϫ178.
[9]
W. J. Houlihan, M. Lohmeyer, P. Workman, S. H. Cheon, Med.
Res. Rev. 1995, 15, 157Ϫ223.
[10] [10a]
K. C. Reddy, H.-S. Byun, R. Bittman, Tetrahedron Lett.
[10b]
1994, 35, 2679Ϫ2682.
R. Bittman, H.-S. Byun, K. C.
Reddy, P. Samadder, G. Arthur, J. Med. Chem. 1997, 40,
1391Ϫ1395.
[11]
[12]
[13]
[14]
[15]
A. Burchardt, T. Takahashi, Y. Takeuchi, G. Haufe, J. Org.
Chem. 2001, 66, 2078Ϫ2084.
W. E. Berdel, W. R. E. Bausert, U. Fink, J. Raststetter, P. G.
Munder, Anticancer Res. 1981, 1, 345Ϫ352.
I. Kudo, S. Nojima, H. W. Chang, R. Yanoshita, H. Hayashi,
E. Kondo, H. Nomura, K. Inoue, Lipids 1987, 22, 862Ϫ867.
P. Workman, J. Donaldson, M. Lohmeyer, Biochem. Pharma-
col. 1991, 41, 319Ϫ322.
3-Bromo-2-fluoro-2-methylpropanoic Acid (23): According to the
procedure given above for the synthesis of 20, the bromo-substi-
tuted fluoride 18 (2.60 g, 12 mmol) was oxidized (reaction time
6 d) to 23, and isolated as a hygroscopic white solid. Yield: 1.56 g
(70%). 1H NMR: δ ϭ 1.75 (d, 3JH,F ϭ 20.5 Hz, 3 H, 4-H) 3.62 (dd,
3JHb,F ϭ 14.5 Hz, 2JHa,Hb ϭ 11.5 Hz, 1 H, 3-Hb), 3.76 (dd, 3JHa,F ϭ
W. J. Houlihan, M. Lee, P. G. Munder, G. M. Nemecek, D. A.
Handley, Ch. M. Winslow, J. Happy, Ch. Jaeggi, Lipids 1987,
22, 884Ϫ890.
2
23.8 Hz, JHa,Hb ϭ 11.5 Hz, 1 H, 3-Ha), 7.74 (br, 1 H, OH). 13C
2
NMR (360 MHz): δ ϭ 22.5 (dq, JC,F ϭ 25.0 Hz, 4-C), 34.6 (dt,
1
2JC,F ϭ 25.0 Hz, 3-C), 92.8 (ds, JC,F ϭ 191.4 Hz, 2-C), 173.6 (ds,
[16] [16a]
R. Bittman, N. M. Witzke, T.-C. Lee, M. L. Blank, F.
[16b]
2JC,F ϭ 26.4 Hz, 1-C). 19F NMR: δ ϭ Ϫ155.6 (m). GC/MS [of the
Si(CH3)3 ester]: m/z (%) ϭ 212/214 (2/2) [Mϩ· Ϫ CO2], 136/138 (4/
4), 116/118 (1/1) [136/138 Ϫ HF], 93/95 (3/3) [CH2Brϩ], 85 (5), 77
(78), 73 (100) [Si(CH3)3ϩ], 59 (15) [C3H4Fϩ]. GC/MS [CI, of the
Si(CH3)3 ester]: m/z ϭ 274/276 [M ϩ NH4ϩ].
Snyder, J. Lipid Res. 1987, 28, 733Ϫ738.
H.-P. Kertscher,
G. Ostermann, A. Lang, W. Weissflog, K. Gawrisch, Pharmazie
1985, 40, 702Ϫ704.
[17] [17a]
Organofluorine Compounds in Medicinal Chemistry and
Biomedical Applications (Eds.: R. Filler, Y. Kobayashi, L. M.
Yagupolski), Elsevier, Amsterdam, 1993.
Chemistry: Principles and Commercial Applications (Eds.: R. E.
[17b]
Organofluorine
Methyl 3-Bromo-2-fluoro-2-methylpropanoate (24): A solution of 23
(1.52 g, 8.2 mmol) in dry methanol (30 mL) was treated with 2,2-
dimethoxypropane (300 µL) and one drop of conc. HCl and stirred
at room temperature for 32 h. After workup by the procedure given
above for compound 21, the ester 24 was isolated as a yellow oil.
Yield: 850 mg (48%). The 1H NMR spectroscopic data of this com-
pound agree with those published.[41] 13C NMR: δ ϭ 22.4 (dq,
Banks, B. E. Smart, J. C. Tatlow), Plenum Press, New York,
[17c]
1994.
Biomedical Frontiers of Fluorine Chemistry (Eds.: I.
Ojima, J. R. McCarthy, J. T. Welch), ACS Symposium Series
[17d]
639, American Chemical Society, Washington, 1996.
D.
[17e]
O’Hagan, H. S. Rzepa, Chem. Commun. 1997, 645Ϫ652.
R. E. Banks, J. Fluorine Chem. 1998, 87, 1Ϫ17. [17f] Asymmetric
Fluoroorganic Chemistry. Synthesis, Applications, and Future
Directions (Ed.: P. V. Ramachandran), ACS Symposium Series
746, American Chemical Society, Washington, 2000.
2
2JC,F ϭ 22.9 Hz, 4-C), 35.1 (dt, JC,F ϭ 22.9 Hz, 3-C), 52.9 (q, 5-
1
2
C), 93.0 (ds, JC,F ϭ 190.7 Hz, 2-C), 169.7 (ds, JC,F ϭ 22.9 Hz, 1-
[18] [18a] A. Guidi, F. Canfarini, A. Giolitti, F. Pasqui, F. Arcamone,
C). 19F NMR: δ ϭ Ϫ155.4 (m). GC/MS: m/z (%) ϭ 198/200 (6/6)
[18b]
Pure Appl. Chem. 1994, 66, 2319Ϫ2322.
M. Schlosser, An-
[Mϩ·], 178/180 (19/19) [Mϩ· Ϫ HF], 139/141 (11/11) [Mϩ·
Ϫ
gew. Chem. 1998, 110, 1538Ϫ1556; Angew. Chem. Int. Ed. Engl.
C2H3O2], 119 (11) [Mϩ· Ϫ Br], 99 (21) [178/180 Ϫ Br], 77 (38), 71
1998, 37, 1496Ϫ1513.
(11), 60 (22), 59 (100) [C3H4Fϩ; C2H3O2ϩ].
[19] [19a]
A. W. Baker, A. T. Shullgin, Nature 1965, 206, 712Ϫ713.
[19b]
M. C. Chapeau, P. A. Frey, J. Org. Chem. 1994, 59,
[19c]
6994Ϫ6998.
J. A. K. Howard, V. J. Hoy, D. O’Hagan, G.
[19d]
T. Smith, Tetrahedron 1996, 52, 12613Ϫ12622.
T. Yama-
Acknowledgments
zaki, T. Kitazume, Coordination Ability of Fluorine to Proton
or Metals Based on Experimental and Theoretical Evidence, in:
Enantiocontrolled Synthesis of Fluoro-Organic Compounds (Ed.:
V. A. Soloshonok), John Wiley & Sons, Chichester, 1999, pp.
575Ϫ600.
This work was generously supported by the Deutsche Forschungs-
gemeinschaft and the Fonds der Chemischen Industrie. We wish to
thank the Bayer AG, Leverkusen, and the Amano Pharmaceutical
Co., Ltd., Nagoya, Japan, for kind donations of chemicals and en-
4506
Eur. J. Org. Chem. 2001, 4501Ϫ4507

Synthesis of a Fluorinated Ether Lipid Analogous to a Platelet Activating Factor
FULL PAPER
G. R. Sullivan, J. A. Dale, H. S. Mosher,
[20]
[31b]
G. A. Pantani, E. J. LaVoie, Chem. Rev. 1996, 96, 3147Ϫ3176.
A. Bondi, J. Phys. Chem. 1964, 68, 441Ϫ451.
34, 2543Ϫ2549.
J. Org. Chem. 1973, 38, 2143Ϫ2147.
C.-S. Chen, Y. Fujimoto, G. Girdaukas, C. S. Sih, J. Am. Chem.
Soc. 1982, 104, 2794Ϫ2799.
P. N. Guivisdalsky, R. Bittman, J. Org. Chem. 1989, 54,
4637Ϫ4642.
J. A. Caputo, R. Fuchs, Tetrahedron Lett. 1967, 4729Ϫ4731.
H. J. Carlsen, T. Katsuki, V. S. Martin, K. B. Sharpless, J. Org.
Chem. 1981, 46, 3936Ϫ3938.
K. Murata, T. Kitazume, Tetrahedron: Asymmetry 1993, 4,
889Ϫ892.
[21]
[22] [22a]
[32]
[33]
[22b]
J. T. Welch, Tetrahedron 1987, 43, 3123Ϫ3197.
J. T.
Welch, S. Eswarakrishnan, Fluorine in Organic Chemistry,
Wiley Interscience Publications, Wiley & Sons, New York,
1991, chapter 8.
[34]
[35]
[23]
[24]
R. Filler, Y. Kobayashi, Biomedicinal Aspects of Fluorine Chem-
istry, Kodansha, Ltd., Tokyo, Elsevier Biomedical Press, Am-
sterdam, New York, Oxford, 1982.
N. Weber, Pharm. Unserer Zeit 1986, 15, 107Ϫ112, and refs.
cited therein.
[36]
[37]
[38]
A. Burchardt, PhD Thesis, University of Münster, 1999.
The often used one-pot reaction between a carboxylic acid and
thionyl chloride, followed by treatment of the crude mixture
with (Ϫ)-menthol and pyridine, gave up to five fluorinated
products (19F NMR). Esterification of the acid with (Ϫ)-men-
thol and DCC/DMAP according to the method described by
A. Hassner, V. Alexanian, Tetrahedron Lett. 1978, 4475Ϫ4476,
and simultaneously by B. Neises, W. Steglich, Angew. Chem.
1978, 90, 556Ϫ557; Angew. Chem. Int. Ed. Engl. 1978, 17,
522Ϫ523, gave a 31% yield of a side product, in addition to
the two diastereomeric esters (69%, 19F NMR). The side prod-
uct was identified as 3-acetoxy-2-fluoro-2-methylpropanoic
acid N-cyclohexylamide (ref.[37]). Partial kinetic resolution dur-
ing the esterification therefore cannot be ruled out. In fact, a
55:45 ratio of diastereomeric esters was obtained (19F NMR).
[25] [25a]
H. Brachwitz, P. Langen, R. Hintsche, J. Schildt, Chem.
[25b]
Phys. Lipids 1982, 31, 33Ϫ52.
G. Ostermann, H. Brach-
witz, U. Till, Biomed. Biochim. Acta 1984, 43, 349Ϫ355.
[26] [26a]
G. Alvernhe, A. Laurent, G. Haufe, Synthesis 1987,
562Ϫ664. [26b] G. Haufe, G. Alvernhe, A. Laurent, T. Ernet, O.
Goj, S. Kröger, A. Sattler, Org. Synth. 1999, 76, 159Ϫ168.
O. G. J. Meyer, R. Fröhlich, G. Haufe, Synthesis 2000,
1479Ϫ1490.
[27]
[28]
Small amounts of the corresponding anti-Markovnikov prod-
ucts have been detected in bromofluorinations of methallyl
chloride or methallyl phenyl ether, or of other allylic alcohol
[28a]
derivatives, with NBS/Et3N·3HF.
G. Haufe, U. Weßel, K.
Schulze, G. Alvernhe, J. Fluorine Chem. 1995, 74, 283Ϫ291.
[28b]
A. Hedhli, A. Baklouti, J. Org. Chem. 1994, 59,
[28c]
[39] [39a]
5277Ϫ5279.
M. Lübke, R. Skupin, G. Haufe, J. Fluorine
T. Kitazume, T. Sato, N. Ishikawa, Chem. Lett. 1984,
[39b]
Chem. 2000, 102, 125Ϫ133, and refs. cited therein.
1811Ϫ1814.
T. Kitazume, K. Murata, T. Ikey, J. Fluorine
[29]
[39c]
Nucleophilic substitution of a bromine atom with an acetate
function (step iii in Scheme 4) was significantly slower with
vicinal bromofluoroalkanes than with the non-fluorinated par-
ent compounds. As well as the modification of the electron
density at the reaction center, electrostatic repulsion between
Chem. 1986, 32, 233Ϫ238.
T. Kitazume, T. S. Kobayashi,
J. T. Lin, J. Org. Chem. 1986, 51, 1003Ϫ1006. [39d] M. Ihara, T.
Kai, N. Taniguchi, K. Fukumoto, J. Chem. Soc., Perkin Trans.
1 1990, 2357Ϫ2358.
[39e]
M. Ihara, N. Taniguchi, T. Kai, K.
Satoh, K. Fukumoto, J. Chem. Soc., Perkin Trans. 1 1992,
221Ϫ227. [39f] M. Ihara, T. Kawabuchi, Y. Tokunaga, K. Fuko-
moto, Tetrahedron: Asymmetry 1994, 5, 1041Ϫ1050.
R. A. W. Johnstone, M. E. Rose, Tetrahedron 1979, 35,
2169Ϫ2173.
L. S. Boguslavskaya, N. N. Chuvatkin, I. Y. Panteleeva, J. Org.
Chem. USSR (Engl. Trans.) 1982, 18, 814Ϫ820.
L. Bushong, J. Am. Chem. Soc. 1903, 30, 212Ϫ224.
Received May 15, 2001
the fluorine substituent and the approaching nucleophile seem
[29a]
to be important.
N. Mongelli, F. Animati, R. D’Alessio,
L. Zuliani, Synthesis 1988, 310Ϫ313. [29b] A. Sattler, G. Haufe,
Tetrahedron 1996, 52, 5469Ϫ5474.
[40]
[41]
[42]
[30] [30a]
A. Sattler, G. Haufe, Tetrahedron: Asymmetry 1995, 6,
2841Ϫ2848. [30b] O. Goj, A. Burchardt, G. Haufe, Tetrahedron:
Asymmetry 1997, 8, 399Ϫ408. [30c] M. Runge, G. Haufe, J. Org.
Chem. 2000, 65, 8737Ϫ8742.
[31] [31a]
J. A. Dale, D. L. Hull, H. S. Mosher, J. Org. Chem. 1969,
[O01240]
Eur. J. Org. Chem. 2001, 4501Ϫ4507
4507