ω-Fluoro Thiafatty Acids: New Mechanistic Probes of Desaturase-Mediated Reactions

Article abstract of DOI:10.1002/hlca.200390312

A series of 18-fluoro thiastearates were prepared and incubated with a yeast Δ9-desaturating system. The relative efficiency of desaturase-mediated sulfoxidation was monitored via ¹⁹F-NMR analysis of the sulfoxide products, and a strong prefere

Full text of DOI:10.1002/hlca.200390312

3688
Helvetica Chimica Acta Vol. 86 (2003)
w-Fluoro Thiafatty Acids: New Mechanistic Probes of
Desaturase-Mediated Reactions
by Derek J. Hodgsona)b), Kim Y. Y. Laoa), Brian Dawsonb), and Peter H. Buist*a)
a) Department of Chemistry, Carleton University, Ottawa, ON Canada K1S 5B6
(phone: ‡‡ 1613-520-2600 (3643); fax: ‡‡ 1613-520-3749; e-mail: pbuist@ccs.carleton.ca)
b) Centre for Biologics Research, Biologics and Genetic Therapies Directorate, Health Canada,
Tunney×s Pasture, Ottawa, Ontario, Canada K1A0L2
Dedicated to Duilio Arigoni on the occasion of his 75th birthday
A series of 18-fluoro thiastearates were prepared and incubated with a yeast D9-desaturating system. The
relative efficiency of desaturase-mediated sulfoxidation was monitored via 19F-NMR analysis of the sulfoxide
products, and a strong preference for oxo transfer to the S-atom occupying the 9-position was confirmed. The
oxidation profile obtained in this manner matched that of analogous experiments with non-fluorinated
substrates. These results form the basis of a versatile 19F-NMR-based method for mapping the position of the
putative diiron oxidant relative to substrate, and has potential application to the study of membrane-bound
desaturases in vitro.
1. Introduction. Fatty acid desaturases are an important superfamily of enzymes
found in a wide variety of aerobic organisms [1]. The first desaturation reaction to be
studied in detail was the conversion of stearoyl CoA to oleyl CoA as it occurs in baker×s
yeast (Scheme 1,a) [2]. Interest in this particular enzyme system has risen recently with
the discovery that enhanced stearoyl CoA D9 desaturase (SCD) activity in mammals
occurs in a number of disorders, including obesity, diabetes, and other metabolic
diseases [3]. Thus, the development of mechanism-based SCD inhibitors would have
potential therapeutic significance. A major stumbling block to progress in this area is
that stearoyl CoA D9 desaturase is membrane-bound and difficult to isolate in purified
form. However, some important mechanistic information has been obtained on the
structurally related yeast enzyme [4], which can be conveniently studied in vivo by an
appropriate experimental design. A KIE study showed that the H-atom at C(9) of
substrate is removed first in an isotopically sensitive step, followed by rapid cleavage of
the CÀH bond at C(10) [5]. This result correlates well with the observation that
desaturase-mediated oxo transfer occurs most efficiently when a S-atom occupies the 9-
position of a thiastearoyl substrate (Scheme 1,b) [6]; the enantioselectivity of
sulfoxidation matches the stereochemistry (syn, pro-R) of the corresponding parent
reaction [7][8]. Taken together, these results strongly suggest that the putative diiron
oxidant involved in yeast D9 desaturase mediated oxidation is asymmetrically located
between C(9) and C(10) of substrate.
The success of the thia-analogue approach in providing critically important
information on desaturase-active-site topology prompted us to consider modifications
to the probe that would obviate the need to isolate milligram amounts of product for

Helvetica Chimica Acta Vol. 86 (2003)
3689
Scheme 1. The Oxidation Reactions Catalyzed by Stearoyl CoA Desaturase. a) Desaturation; b) sulfoxidation
(CoA ˆ Coenzyme A).
analysis [6][8]. Development of a more-convenient analytical method would greatly
enhance the versatility of the −thia-test× by permitting the study of membrane-bound
desaturases such as SCD in vitro. As a first step in this direction, we report here the
validation of a 19F-NMR-based method for monitoring D9-desaturase-mediated
sulfoxidation of F-tagged thiasubstrates. A preliminary publication of portions of this
work has appeared [9].
2. Results.
The required methyl w-fluoro-thiaoctadecanoates 2a 2d were
synthesized by fluorination of the corresponding methyl 18-hydroxy-thiaoctadeca-
noates 1a 1d via a conventional tosylation/Bu4NF(TBAF) sequence (Scheme 2). Use
of DAST ((diethylamino)sulfur trifluoride) in this step was compromised by the fact
that a significant reaction at the S-atom occurred to give the sulfoxy by-product. The
intermediate alcohols 1a 1d were readily available via alkylation of the appropriate w-
thioacid [10] with the 3,4,5,6-tetrahydro-2H-pyran-2-yl (THP)-protected w-hydroxy
alkyl bromides of the correct chain length. Oxidation of 2a 2d with 1 or 2 equiv. of m-
chloroperbenzoic acid (MCPBA) furnished the corresponding sulfoxides 3a 3d and
sulfones 4a 4d [6], respectively. The analytical and spectral data of all thiastearates
and their oxidized derivatives were consistent with the assigned structures (see Exper.
1
Part). The H-decoupled 19F-NMR spectra of these compounds were recorded in
CDCl3, and the trends in chemical shift for sulfide, sulfoxide, and sulfone as a function
of the position of the S-atom are presented in Fig. 1. 19F-NMR Resonances of each
series of w-fluorothiastearate and the corresponding oxidized derivatives were resolved
to baseline in all cases. The absolute value of the 19F chemical shifts for each analyte is
given in the Exper. Part.
To evaluate the suitability of the w-fluorothiastearates as substrates for D9
desaturation by SCD-type enzymes, we turned to our well-characterized yeast model

3690
Helvetica Chimica Acta Vol. 86 (2003)
Scheme 2. Synthesis of w-Fluoro Thiaoctadecanoate Isomers and Their Oxidized Derivatives
system. Each w-fluorothiastearate 2a d (25 mg) was incubated separately with
actively growing cultures (200 ml) of Saccharomyces cerevisiae S522C to maximize
production of the corresponding sulfoxide. Our whole-cell system allows us to use
methyl ester substrates that are presumably converted intracellularly to the corre-
sponding CoA thioesters prior to desaturase-mediated sulfoxidation. The product
sulfoxide is then hydrolyzed and excreted into the medium. Control experiments with a
yeast mutant have demonstrated previously that sulfoxidation of thiastearate is
dependent on active expression of the D9 desaturase [11].
19F-NMR Analysis of the CH2Cl2 extracts of the supernatant derived from each w-
fluorothiastearate incubation revealed the presence of sulfide and the corresponding
sulfoxide as the major analytes in each case. The sulfoxy products were identified on the

Helvetica Chimica Acta Vol. 86 (2003)
3691
Fig. 1. The effect of sulfur oxidation on the 19F chemical shifts of methyl w-fluoro thiastearates as a function of S-
position
basis of 19F-NMR chemical-shift data (agreement within 0.01 ppm) as well as TLC
analysis of each extract by using authentic standards. Only small (< 5% of total)
amounts of w-fluoro sulfone were observed; in the case of 2d oxidation, three
additional minor, unidentified products (12% of total), presumably derived from chain
cleavage of a 9,10-dehydro-11-thia derivative, were observed in the 19F-NMR spectrum
of the extract.
The relative amounts of sulfoxidized product were quantitated by the use of an
external standard (flunisolide), and the % sulfoxidation of 2a, 2b, 2c, and 2d was
calculated to be 1.2, 7, 4.5, 2.0%, respectively. These data are presented in graphical
form (normalized to % sulfoxidation of 2b) in Fig. 2 along with data obtained
previously for the parent, non-fluorinated thiastearoyl substrates.
3. Discussion. The success of a 19F-NMR-based approach to studying desaturase-
mediated sulfoxidation depends on three considerations: 1) synthetic accessibility of
the fluorinated probes; 2) sensitivity of the remote 19F-NMR reporter resonance to
changes to sulfide oxidation state; 3) nonperturbing nature of the fluorine tag. We were
pleased to see that all of these criteria were met in these experiments:
1) The synthetic scheme we employed to generate 2a 2d is straightforward and can
be easily adapted to generate a number of w-fluoro thia-analogues that vary in S-

3692
Helvetica Chimica Acta Vol. 86 (2003)
Fig. 2. Comparison of yeast-mediated % sulfoxidation of non-fluorinated and fluorinated 8-, 9- 10-, and 11-
thiastearates 2a
d
position and chain length. This will allow further study of desaturases with different
positional specificities [12][13].
2) The Dd between 1a 1d and the corresponding sulfoxy products is sufficiently
different to permit facile monitoring of desaturase-mediated sulfoxidation. As might be
expected, the substituent effects on the 19F chemical shifts decreased in magnitude as
the distance (seven to eleven bonds) between the S-atom and the F-nucleus increases
(Fig. 1). The direction of the Dd effects for increasingly electron-withdrawing
substituents (shielding) match previous observations for rigid cycloalkyl systems [14].
3) Perhaps the most important finding of this work was the good agreement
between the regioselectivity profile of sulfoxidation for F-tagged thiastearates and the
corresponding non-fluorinated substrates (Fig. 2). The latter data set was obtained by
mass spectrometric analysis of sulfoxide mixtures isolated by silica-gel chromatography
from direct competition experiments between non-fluorinated thiastearates. It is clear
that the presence of a substrate w-F-tag did not significantly affect the pronounced
preference of the desaturase system to oxygenate 9-thia substrate analogues as we have
previously documented (see Introduction). The ratio of 9- to 10-sulfoxidized products
(1.6 :1) obtained in this work compares favorably with the values obtained from both
direct competition (2.1:1 [6]) and noncompetitive incubation of thiastearates (1.7 [5]).
Good correlation between sulfoxidation ratios derived from experiments with non-
fluorinated and w-fluorinated thiastearates have also been obtained for a soluble, plant
D9-desaturase system [9][15].
The relatively low levels of 8- and 11-sulfoxidation observed in these experiments
(Fig. 2) may be due to autoxidation and/or oxidation by unselective bio-oxidation by
non-desaturase-related enzymes, since the 8- and 11-positions are not expected to be
accessible to the D9-desaturase oxidant. A more-detailed assessment of the fate of 8-

Helvetica Chimica Acta Vol. 86 (2003)
3693
and 11-thiastearoyl substrates must await development of an in vitro assay; the
predicted C(9)ˆC(10) sulfide products are not expected to be stable under our in vivo
conditions.
4. Conclusions.
The results presented in this paper open a new window of
opportunity in the mechanistic characterization of desaturases. In principle, it should be
possible to generate a 19F-NMR-based sulfoxidation profile for any given desaturase in
a single competition experiment provided the sulfoxide products derived from the w-
fluorinated thiafatty acid mixtures have unique chemical shifts. A number of possible
applications of this approach are envisaged. For example, it should be possible to take
advantage of the excellent inherent sensitivity of 1H-decoupled 19F-NMR (mm
concentration of fluorinated analyte) coupled with the lack of interference to obtain
the in vitro oxidation signature of medically important desaturases such as SCD.
Furthermore, it would be interesting to characterize the regioselectivity of sulfox-
idation for desaturases with novel cryptoregiochemistry as determined by KIE
experiments [16]. Finally, our 19F-NMR-based method may offer a unique way of
tracking site-directed-mutagenesis experiments by obtaining a snapshot of the position
of the diiron oxidant relative to substrate for a series of modified desaturases.
Experimental Part
General. Abbreviations: MCPBA: meta-chloroperbenzoic acid, FC: flash chromatography. THF was freshly
distilled from Na before use. MeCN was dried over 4-ä molecular sieves. Pyridine was dried over NaOH. Unless
otherwise noted, all reagents and starting materials for organic synthesis were purchased from Sigma-Aldrich
and used without purification. TsCl was recrystallized from hexane/CHCl3 before use. 8-Sulfanyloctanoic and
10-sulfanyldecanoic acids were prepared from the corresponding, commercially available bromo acids as
described in [10]. In a similar manner, 7-sulfanylheptanoic and 9-sulfanylnonanoic acid were obtained from 7-
bromoheptanoic acid and 9-bromononanoic acid, resp.; the latter two compounds were available from the
corresponding w-bromoalkan-1-ol via Jones oxidation [17]. Protection of the required w-bromoalkan-1-ols as
the THP ethers was carried out according to a standard procedure [18]. All air- and moisture-sensitive reactions
were performed under N2. Org. extracts were typically shaken with sat. NaCl soln., dried (Na2SO4) and solvents
were evaporated in vacuo on a B¸chi RE 111 Rotavapor. FC: Merck silica gel (230 400 mesh, at ca. 0.5 atm).
TLC: Merck silica gel 60 F 254 plates; visualization by a combination of water spray and I2 vapor. M.p.: Fisher-
Johns melting-point apparatus; uncorrected. IR Spectra: ABB Bomem MB spectrometer; in cmÀ1. NMR
Spectra: Bruker AM-400 (1H-decoupled 19F: 376.5 MHz); Bruker AMX-400 (1H: 400 MHz, 13C: 100 MHz);
Bruker Avance 700 (1H: 700 MHz, 13C: 176 MHz); 1H and 13C chemical shifts (d) in ppm downfield from internal
TMS (d ˆ 0.00 ppm); 19F chemical shifts (d) in ppm downfield from external CFCl3 (d ˆ 0.00 ppm), J values in
Hz. EI-MS: Kratos Concept 1 H mass spectrometer; direct introduction or via GC (HP 5980 Series 2 gas
chromatograph (J. & W. 30 m  0.21 mm, DB-5 capillary column)). Saccharomyces cerevisiae S288C cultures
were obtained from the Yeast Genetic Stock Center, University of California at Berkeley and cultured as
described in [6].
Methyl 18-Hydroxy-8-thiaoctadecanoate (1a). To a soln of 7-sulfanylheptanoic acid (1.63 g, 10.0 mmol) in
THF/HMPA 3 :1 (66 ml) was added a 1.6m BuLi (14.0 ml ) reagent via syringe, under N2 at 08. After stirring the
mixture at 08 for 30 min, a soln of 10-bromodecan-1-ol THP ether (3.23 g, 10.1 mmol) was added via syringe.
The mixture was stirred at r.t. for 22 h, and the reaction was quenched by the addition of H2O (60 ml), and the
mixture was acidified to a pH of 2 with 3m HCl (15 ml) and extracted with hexanes (4 Â 60 ml). The combined
org. extracts were washed with sat. aq. NaCl soln., dried (Na2SO4) and evaporated to give crude w-hydroxy-8-
thiastearic acid THP ether (3.19 g). This compound was deprotected and methylated with TsOH (115 mg) /
MeOH (30 ml; reflux) and, after evaporation of the reaction solvent, the crude product (1.99 g) was purified by
FC (20% AcOEt/hexanes) to give 1a (558 mg, 17% based on 7-sulfanylheptanoic acid). White solid. TLC
(hexane/AcOEt 40 :60): Rf 0.25. M.p. 43 458 (amorphous crystals, recrystallized from hexane). IR (KBr):

3694
Helvetica Chimica Acta Vol. 86 (2003)
3340, 2931, 2851, 1738, 1471, 1462, 1437, 1254, 1176, 1171, 1073. 1H-NMR (400 MHz, CDCl3): 3.66 (s, 3 H); 3.63
(t, J ˆ 6.6, 2 H); 2.489, 2.487 (partially resolved overlapping t, J ˆ 7.4, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.53 1.67 (m,
9 H); 1.25 1.45 (m, 16 H). 13C-NMR (100 MHz, CDCl3): 174.23; 63.10; 51.47; 34.03; 32.82; 32.23; 32.11; 29.73;
‡
‡
29.53; 29.53; 29.45; 29.41; 29.24; 28.94; 28.78; 28.55; 25.74; 24.84. EI-MS: 332 (8, M ), 301 (11, [M À 31] ), 283
‡
‡
(2), 259 (4), 231 (3), 203 (6), 189 (16, [HO(CH2)10S] ), 171 (51, [CH2ˆCH(CH2)8S] ), 159 (10), 143(89,
‡
‡
[(CH2)6CO2Me] ), 129 (8), 111 (26),101 (25), 87 (54), 69 (60), 55 (100), 41 (66). HR-EI-MS: 332.2392 (M ,
C18H36O3S; calc. 332.2385).
Methyl 18-Hydroxy-9-thiaoctadecanoate (1b). From 8-sulfanyloctanoic acid and 9-bromononan-1-ol THP
ether: white solid. TLC (hexane/AcOEt 40 :60): Rf 0.25. M.p. 42 438 (plates, recrystallized from hexane). IR
(KBr): similar to that of 1a. 1H-NMR (400 MHz, CDCl3): 3.66 (s, 3 H); 3.63 (t, J ˆ 6.6, 2 H); 2.49 (unresolved
overlapping t, J ꢀ 7, 4 H); 2.30 (t, J ˆ 7.4, 2 H); 1.53 1.66 (m, 9 H); 1.5 1.45 (m, 16 H). 13C-NMR (100 MHz,
CDCl3): 174.31; 63.08; 51.49; 34.07; 32.80; 32.19; 32.14; 29.70; 29.64; 29.47; 29.36; 29.18; 29.03; 28.91; 28.89;
‡
‡
‡
28.74; 25.72; 24.89. EI-MS: 332 (5, M ), 301 (10, [M À 31] ), 175 (8, [HO(CH2)9S] ), 157 (100,
‡
‡
‡
[CH2ˆCH(CH2)7S] , [(CH2)7CO2Me] ), 87 (47), 69 (37) , 55 (100), 41 (53). HR-EI-MS: 332.2389 (M ,
C18H36O3S; calc. 332.2385).
Methyl 18-Hydroxy-10-thiaoctadecanoate (1c). From 9-sulfanylnonanoic acid and 8-bromooctan-1-ol THP
ether: a white solid. TLC (hexane/AcOEt 40 :60): Rf 0.25. M.p. 41.5 42.58 (amorphous crystals, recrystallized
from hexane). IR (KBr): similar to that of 1a. 1H-NMR (400 MHz, CDCl3): 3.66 (s, 3 H); 3.63 (t, J ˆ 6.6, 2 H);
2.49 (unresolved overlapping t, J ꢀ 7, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.53 1.66 (m, 9 H); 1.25 1.45 (m, 16 H).
13C-NMR (100 MHz, CDCl3): 174.36; 63.05; 51.48; 34.09; 32.77; 32.17; 32.17; 29.69; 29.69; 29.30; 29.21; 29.14;
‡
‡
‡
29.08; 29.05; 28.86; 28.86; 25.68; 24.92. EI-MS: 332 (4, M ), 301 (11, [M À 31] ), 171(42, [(CH2)8CO2Me] ), 161
‡
‡
(6, [HO(CH2)8S] ), 143 (64, [CH2ˆCH(CH2)6S] ), 87 (54), 69 (59), 55 (100), 41 (57). HR-EI-MS: 332.2383
‡
(M , C18H36O3S; calc. 332.2385).
Methyl 18-hydroxy-11-thiaoctadecanoate (1d). From 10-sulfanyldecanoic acid and 7-bromoheptan-1-ol
THP ether: a white solid. TLC (hexane/AcOEt 40 :60): Rf 0.25. M.p. 40 418 (amorphous crystals, recrystallized
from hexane). IR (KBr): similar to that of 1a. 1H-NMR (400 MHz, CDCl3): 3.66 (s, 3 H); 3.63 (t, J ˆ 6.6, 2 H);
2.495, 2.501 (partially resolved, overlapping t, J ˆ 7.4, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.53 1.66 (m, 9 H); 1.25 1.45
(m, 16 H). 13C-NMR (100 MHz, CDCl3): 174.38; 63.02; 51.48; 34.11; 32.73; 32.19; 32.15; 29.70; 29.64; 29.32;
‡
‡
29.20; 29.20; 29.13; 29.05; 28.91; 28.89; 25.64; 24.95. EI-MS: 332 (4, M ), 301 (13, [M À 31] ), 185 (28,
‡
‡
‡
[(CH2)9CO2Me] ), 147 (5, [HO(CH2)7S] ), 129 (65, [CH2ˆCH(CH2)6S] ), 87 (52), 69 (47) , 55 (100), 41 (51).
‡
HR-EI-MS: 332.2386 (M , C18H36O3S; calc. 332.2385).
Methyl 18-Fluoro-8-thiaoctadecanoate (2a). To a soln of 1a (0.307 g, 0.92 mmol) in dry pyridine (2 ml) was
added TsCl (0.352 g) under N2 at r.t. After stirring at r.t. for 3 h, the reaction was quenched with the addition of
H2O (20 ml), and the mixture was extracted with Et2O ( 4Â 30 ml). The combined org. layers were washed with
2m HCl (2 Â 20 ml), dried (Na2SO4) and evaporated to give the crude tosylate (0.266 g, 59% est.): TLC (hexane/
AcOEt 30 :70): Rf 0.35. 1H-NMR (200 MHz , CDCl3): 7.79 (d, J ˆ 8.3, 2 H); 7.35 (d, J ˆ 8.3, 2 H); 4.01 (t, J ˆ 6.3,
2 H); 3.67 (s, 3 H); 2.49 (overlapping t, J ˆ 6.7, 4 H); 2.45 (s, 3 H), 2.31 (t, J ˆ 7.3, 2 H); 1.45 1.70 (m, 8 H);
1.10 1.40 (m, 14 H). This compound was stirred at r.t. with 6 equiv. of TBAF in dry THF (50 ml) over molecular
sieves (0.5 g), for 3 h. THF was removed in vacuo, and the residue was partitioned between H2O (10 ml) and
hexane (4 Â 20 ml). The hexane layer was washed with sat. aq. NaCl soln. (1 Â 50 ml), dried (Na2SO4) and
evaporated. The crude product was purified by FC with 5% AcOEt/hexanes to give 2a (111 mg, 36% based on
1a). Colorless oil. TLC (hexane/AcOEt 90 :10): Rf 0.34. M.p. 24 24.58. IR (film): 2928, 2855, 1741, 1463, 1436,
1248, 1198, 1171. 1H-NMR (400 MHz , CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.497, 2.494 (partially
resolved overlapping t, J ˆ 7.4, 4 H); 2.31 (t, J ˆ 7.5, 2 H); 1.53 1.80 (m, 8 H); 1.25 1.45 (m, 16 H). 13C-NMR
(100 MHz , CDCl3): 174.23; 84.25 (d, J ˆ 163.9); 51.49; 34.01; 32.04; 32.20; 30.41 (d, J ˆ 19.3); 29.71; 29.50; 29.45;
29.42; 29.22; 29.22; 28.94; 28.76; 28.55; 25.15 (d, J ˆ 5.5 ); 24.83. 19F NMR (376.5 MHz, CDCl3): À218.18. EI-
‡
‡
‡
‡
MS: 334 (24, M ), 303 (10, [M À 31] ), 191(49, [F(CH2)10S] ), 171 (22, [CH2ˆCH(CH2)8S] ), 143 (100,
‡
‡
[(CH2)6CO2Me] ), 111 (38), 87 (45), 69(62), 55 (95), 41 (71). HR-EI-MS: 334.2323 (M , C18H35O2FS; calc.
334.2342).
Methyl 18-Fluoro-9-thiaoctadecanoate (2b). From 1b: white plates. TLC (hexane/AcOEt 90 :10): Rf 0.34.
M.p. 22.5 23.58. IR (film): similar to that of 2a. 1H-NMR (400 MHz , CDCl3): 4.43 (dt, J ˆ 47.4, 2 H); 3.67 (s,
3 H); 2.50 (unresolved overlapping t, J ˆ 7.4, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.53 1.83 (m, 8 H); 1.25 1.45 (m,
16 H). 13C-NMR (100 MHz , CDCl3): 174.25; 84.21 (d, J ˆ 164.0); 51.45; 34.04; 32.17; 32.12; 30.39 (d, J ˆ 19.3);
29.68; 29.62; 29.37; 29.16; 29.13; 29.01; 28.89; 28.87; 28.72; 25.13 (d, J ˆ 5.5); 24.87. 19F-NMR (376.5 MHz,
‡
‡
‡
CDCl3): À218.18. EI-MS: 334 (37, M ), 303 (20, [M À 31] ), 189 (28, [S(CH2)7CO2Me] ), 177 (63,

Helvetica Chimica Acta Vol. 86 (2003)
3695
‡
‡
‡
[F(CH2)9S] ), 157 (100, [CH2ˆCH(CH2)7S] , [(CH2)7CO2Me] ), 125 (56), 87 (62), 55 (92), 41 (63). HR-EI-
‡
MS: 334.2352 (M , C18H35O2FS; calc. 334.2342).
Methyl 18-Fluoro-10-thiaoctadecanoate (2c). From 1c: colorless oil at r.t. TLC (hexane/AcOEt 90 :10): Rf
0.34. M.p. 19.5 208. IR (film): similar to that of 2a. 1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s,
3 H); 2.500, 2.498 (partially resolved, overlapping t, J ˆ 7.4, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.53 1.80 (m, 8 H);
1.25 1.45 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 174.24; 84.06 (d, J ˆ 164.1); 51.39; 34.04; 32.14; 32.14; 30.33
(d, J ˆ 19.3); 29.65; 29.64; 29.10; 29.09; 29.04; 29.01; 28.87; 28.82; 28.78; 25.07 (d, J ˆ 5.4); 24.88. 19F-NMR
‡
‡
‡
(376.5 MHz, CDCl3): À218.24. EI-MS: 334 (47, M ), 303 (32, [M À 31] ), 203 (22, [S(CH2)8CO2Me] ), 171
‡
‡
‡
(100, [(CH2)8CO2Me] ), 163 (60, [F(CH2)8S] ), 143 (65, [CH2ˆCH(CH2)7S] ), 109 (32) 87 (69), 69 (73), 55
‡
(72), 41 (51). HR-EI-MS: 334.2330 ( M , C18H35O2FS; calc. 334.2342).
Methyl 18-Fluoro-11-thiaoctadecanoate (2d). From 1d: colorless oil at r.t. TLC (hexane/AcOEt 90 :10): Rf
0.34. M.p. 24 258. IR (film): similar to that of 2a. 1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s,
3 H); 2.503, 2.496 (unresolved, overlapping t, J ˆ 7.4, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.53 1.80 (m, 8 H); 1.25 1.45
(m, 16 H). 13C-NMR (100 MHz, CDCl3): 174.34; 84.16 (d, J ˆ 164.1); 51.46; 34.22; 32.22; 32.15; 30.35 (d, J ˆ
19.6); 29.72; 29.60; 29.33; 29.19; 29.19; 29.13; 28.92; 28.87; 28.79; 25.08 (d, J ˆ 5.4); 24.95. 19F-NMR (376.5 MHz,
‡
‡
‡
‡
CDCl3): À218.30. EI-MS: 334 (40, M ), 303 (37, [M À 31] ), 185 (69, [(CH2)9CO2Me] ), 149 (45, [F(CH2)7S] ),
‡
‡
129 (18, [CH2ˆCH(CH2)7S] ), 95 (37) 87 (69), 55 (100), 41 (69). HR-EI-MS: 334.2329 (M , C18H35O2FS; calc.
334.2342).
(RS)-Methyl 18-Fluoro-8-thiaoctadecanoate S-Oxide (3a). To a soln. of 2a (16.6 mg, 0.044 mmol) in CH2Cl2
(1 ml) was added MCPBA (0.5 equiv., 8.0 mg) at 08. After standing at 08 for 30 min, the mixture was filtered to
remove a precipitated white solid (MCPBA), and the filtrate was washed with 3m NaOH (2 Â 2.5 ml), dried
(Na2SO4) and evaporated to give the crude sulfoxide, which was purified by recrystallization from AcOEt/
hexanes to give 3a (6.4 mg, 42%). White solid. TLC (AcOEt): Rf 0.40. M.p. 66 66.58 (amorphous crystals,
recrystallized from hexane/AcOEt). IR (KBr): 2921, 2850, 1737, 1464, 1437, 1254, 1207, 1174, 1011. 1H-NMR
(400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.57 2.73 (m, 4 H); 2.32 (t, J ˆ 7.5, 2 H); 1.78, 1.76
(overlapping m, 4 H); 1.60 1.68 (m, 4 H), 1.24 1.56 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 174.07; 84.23 (d,
J ˆ 164.0); 52.51; 52.32; 51.53; 33.90; 30.40 (d, J ˆ 19.3); 29.38; 29.24; 29.18; 29.18; 28.88; 28.71; 28.54; 25.14 (d,
‡
J ˆ 5.4); 24.64; 22.62; 22.48. 19F-NMR (376.5 MHz, CDCl3): À218.23. EI-MS: 333 (100, [M À OH] ), 303(3),
‡
‡
191(18), 160 (57), 143 (32, [(CH2)6CO2Me] ), 111 (32), 55 (95). HR-EI-MS: 333.2263 ([ M À OH] ,
‡
C18H34FO2S ; calc. 333.2264).
(RS)-Methyl 18-Fluoro-9-thiaoctadecanoate S-Oxide (3b). From 2b: a white solid. TLC (AcOEt): Rf 0.40.
M.p. 66 66.58 (amorphous crystals, recrystallized from hexane/AcOEt). IR (KBr): similar to that of 3a.
1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.57 2.73 (m, 4 H); 2.31 (t, J ˆ 7.5, 2 H); 1.78
(m, 4 H); 1.6 1.68 (m, 4 H), 1.24 1.56 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 174.20; 84.20 (d, J ˆ 164.0);
52.48; 52.41; 51.42; 33.99; 30.38 (d, J ˆ 19.2); 29.23; 29.11; 28.86; 28.86; 28.85; 28.85; 28.69; 25.13 (d, J ˆ 5.4);
‡
24.80; 22.62; 22.57. 19F-NMR (376.5 MHz, CDCl3): À218.25. EI-MS: 333 (26, [M À OH] ), 303 (8), 177 (49),
‡
‡
174 (16), 157 (73, [(CH2)7CO2Me] ), 125 (41), 55 (100). HR-EI-MS: 333.2256 ([M À OH] , C18H34O2FS; calc.
333.2264).
(RS)-Methyl 18-Fluoro-10-thiaoctadecanoate S-oxide (3c). From 2c: a white solid. TLC (AcOEt): Rf 0.41.
M.p. 65.5 66.58 (amorphous crystals, recrystallized from hexane/AcOEt). IR (KBr): similar to that of 3a.
1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.57 2.73 (m, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.77,
1.76 (two overlapping m, 4 H); 1.6 1.68 (m, 4 H), 1.24 1.56 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 174.27;
84.14 (d, J ˆ 164.3); 52.64; 52.44; 51.48; 34.05; 30.35 (d, J ˆ 19.2); 29.11; 29.01; 29.01; 28.99; 28.96; 28.82; 28.79;
28.69; 25.10 (d, J ˆ 5.4); 24.87; 22.61; 22.61. 19F-NMR (376.5 MHz, CDCl3): À218.37. EI-MS: 333 (26, [M À
‡
‡
OH] ), 303 (9), 188(12), 171 (49, [(CH2)8CO2Me] ), 163 (26), 139 (30), 55 (100). HR-EI-MS: 333.2247 ([M À
‡
OH] , C18H34O2FS; calc. 333.2264).
(RS)-Methyl 18-Fluoro-11-thiaoctadecanoate S-Oxide (3d). From 2d: a white solid. TLC (AcOEt): Rf 0.41.
M.p. 59 608 (amorphous crystals, recrystallized from hexane/AcOEt). IR (KBr): similar to that of 3a.
1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.3, 2 H); 3.67 (s, 3 H); 2.57 2.73 (m, 4 H); 2.30 (t, J ˆ 7.5, 2 H);
1.78,1.76 (two overlapping m, 4 H); 1.58 1.68 (m, 4 H); 1.58 1.68 (m, 4 H); 1.38 1.52 (m, 8 H); 1.24 1.38 (m,
8 H). 13C-NMR (176 MHz , CDCl3): 174.34; 84.07 (d, J ˆ 164.4); 52.50; 52.38; 51.49; 34.09; 30.29 (d, J ˆ 19.2);
29.15; 29.14; 29.08; 28.86; 28.86; 28.78; 28.69; 25.01 (d, J ˆ 5.7); 24.92; 22.62; 22.57. 19F-NMR (376.5 MHz,
‡
‡
CDCl3): À218.46. EI-MS: 333 (44, [M À OH] ), 303 (3), 202 (25), 185 (6, [(CH2)9CO2Me] ), 153 (19), 149 (7),
‡
55 (100). HR-EI-MS: 333.2271 ([M À OH] , C18H34O2FS; calc. 333.2264).
Methyl 18-Fluoro-8-thiaoctadecanoate S,S-Dioxide (4a). To a soln. of 3a (14.8 mg, 0.044 mmol) in CH2Cl2
(1 ml) was added MCPBA (2 equiv., 31.4 mg) at 08. After standing at 08 for 30 min, the mixture was filtered to

3696
Helvetica Chimica Acta Vol. 86 (2003)
remove a precipitated white solid (MCPBA), and the filtrate was washed with 3m NaOH (2 Â 2.5 ml), dried
(Na2SO4) and evaporated to give the crude sulfone, which was purified by FC with 25% AcOEt/hexanes to give
4a (14.8 mg, 92%). White solid. TLC (Hexane/AcOEt 75:35): Rf 0.19. M.p. 67 688 (amorphous crystals,
recrystallized from hexane/AcOEt). IR (KBr): 2938, 2850, 1738, 1474, 1438, 1262, 1213, 1174, 1137. 1H-NMR
(400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.940 (m, 4 H); 2.32 (t, J ˆ 7.5, 2 H); 1.85, 1.83
(overlapping m, 4 H); 1.6 1.78 (m, 4 H); 1.28 1.52 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 173.98; 84.21 (d,
J ˆ 164.1); 52.82; 52.61; 51.55; 33.82; 30.39 (d, J ˆ 19.3); 29.35; 29.15; 29.15; 29.04; 28.56; 28.51; 28.20; 25.14 (d,
‡
J ˆ 5.4); 24.52; 21.94; 21.78. 19F-NMR (376.5 MHz, CDCl3): À218.26. EI-MS: 335 (7, [M À MeO] ), 315 (3), 293
‡
‡
(9), 143 (85, [(CH2)6CO2Me] ), 111 (73), 83 (100), 55 (91). HR-EI-MS: 335.2039 ([ M À MeO] , C17H32O3FS;
calc. 335.2056).
Methyl 18-Fluoro-9-thiaoctadecanoate S,S-Dioxide (4b). From 3b: a white solid. TLC (hexane/AcOEt
75 :25): Rf 0.22. M.p. 69 708 (plates, recrystallized from hexane/AcOEt). IR (KBr): similar to that of 4a.
1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.94 (m, 4 H); 2.31 (t, J ˆ 7.5, 2 H); 1.83(m,
4 H); 1.6 1.78 (m, 4 H); 1.28 1.52 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 174.14; 84.17 (d, J ˆ 164.1); 52.77;
52.70; 51.51; 33.95; 30.37 (d, J ˆ 19.6); 29.13; 29.01; 28.73; 28.73; 28.73; 28.49; 28.32; 25.13 (d, J ˆ 5.4); 24.87;
‡
21.93; 21.86. 19F-NMR (376.5 MHz, CDCl3): À218.33. EI-MS: 335 (9, [M À MeO] ), 315 (13), 293 (19), 157 (59,
‡
‡
[(CH2)7CO2Me] ), 125 (89), 55 (100). HR-EI-MS: 335.2037 ([ M À MeO] , C17H32O3FS; calc. 335.2056).
Methyl 18-Fluoro-10-thiaoctadecanoate S,S-Dioxide (4c). From 3c: a white solid. TLC (hexane/AcOEt
75 :25): Rf 0.24. M.p. 66 678 (plates, recrystallized from hexane/AcOEt). IR (KBr): similar to that of 4a.
1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.4, 2 H); 3.67 (s, 3 H); 2.94 (m, 4 H); 2.30 (t, J ˆ 7.5, 2 H); 1.84, 1.83
(overlapping m, 4 H); 1.6 1.78 (m, 4 H); 1.28 1.52 (m, 16 H). 13C-NMR (100 MHz , CDCl3): 174.23; 84.10 (d,
J ˆ 164.1); 52.76; 52.72; 51.48; 34.02; 30.32 (d, J ˆ 19.4); 28.96; 28.96; 28.88; 28.87; 28.87; 28.45; 28.42; 25.09 (d,
‡
J ˆ 5.2); 24.83; 21.92; 21.90. 19F-NMR (376.5 MHz, CDCl3): À218.43. EI-MS: 335 (11, [M À MeO] ), 334 (14,
‡
‡
[M À HOMe] ), 315 (9), 293 (8), 171 (37, [(CH 2)8CO2Me] ), 139 (89), 55 (100). HR-EI-MS: 334.1964 ([ M À
‡
HOMe] , C17H31O3FS; calc. 335.1978).
Methyl 18-Fluoro-11-thiaoctadecanoate S,S-Dioxide (4d). From 3d: a white solid. TLC (hexane/AcOEt
75 :25): Rf 0.22. M.p. 66.5 678 (plates, recrystallized from hexane/AcOEt). IR (KBr): similar to that of 4a.
1H-NMR (400 MHz, CDCl3): 4.44 (dt, J ˆ 47.3, 2 H); 3.67 (s, 3 H); 2.944, 2.938 (overlapping m, 4 H); 2.30 (t, J ˆ
7.5, 2 H); 1.6 1.78 (m, 4 H); 1.28 1.52 (m, 16 H). 13C-NMR (100 MHz, CDCl3): 174.30; 83.99 (d, J ˆ 164.2);
52.81; 52.66; 51.47; 34.07; 30.25 (d, J ˆ 19.4); 29.09; 29.04; 29.04; 28.99; 28.72; 28.50; 28.43; 24.95 (d, J ˆ 5.4);
‡
24.89; 21.94; 21.85. 19F-NMR (376.5 MHz, CDCl3): À218.56. EI-MS: 335 (11, [M À MeO] ), 315 (1), 293 (13),
‡
‡
185 (27, [(CH2)9CO2Me] ), 153 (48), 55 (100). HR-EI-MS: 335.2058 ([M À MeO] ; C17H31O3FS; calc.
335.2056).
Incubation Experiments with S. cerevisiae. A typical feeding experiment was carried out as follows: the
strain of baker×s yeast used in this experiment was wild type Saccharomyces cerevisiae # S288C. Cultures were
grown in YEPD medium (1.0% yeast extract, 2.0% bactopeptone, 2.0% d-glucose), at 308 in a rotary incubator-
shaker set at 150 rpm. The starter culture (1 ml) was used to innoculate 200 ml of sterile medium contained in a
1-l Erlenmeyer flask. A 25-mg portion of thiafatty acid methyl ester was added as a soln. (5% (w/v)) in 95%
EtOH to each culture flask, and the innoculated cultures were then incubated for 24 h. The addition of
exogenous fatty acid did not affect the growth of the organism with respect to control cultures.
Culture media was collected by centrifugation (6000 Â g, 10 min), and the supernatant was adjusted to
pH 3.5 with AcOH (ca. 5 ml) and extracted with five 80-ml portions of CH2Cl2. The combined org. layers were
dried and the solvent was evaporated in vacuo to give a crude extract, which was methylated with an ethereal
CH2N2 soln. The residue was dissolved in AcOEt and passed through a small column of Fluorosil to give a final
extract (ca. 25 mg). TLC Analysis of the extract was carried out on silica-gel plates with 80% AcOEt/hexane.
The presence of sulfoxide product was confirmed by comparison with Rf values of authentic synthetic standards.
This work was supported by NSERC Grant OGP-2528 to P. H. B. The technical assistance of Frances
Carvalho (Carleton University) in performing the yeast incubations is gratefully acknowledged. We thank Clem
Kazakoff (University of Ottawa Mass Spectrometry Centre) and Keith Bourque (Carleton University) for
recording the mass and NMR spectra, respectively.

Helvetica Chimica Acta Vol. 86 (2003)
3697
REFERENCES
[1] B. Behrouzian, P. H. Buist, Curr. Opin. Chem. Biol. 2002, 6, 577.
[2] D. K. Bloomfield, K. Bloch, J. Biol. Chem. 1960, 235, 337.
[3] J. M. Ntambi, M. Miyazaki, J. P. Stoehr, H. Lan, C. M. Kendziorski, B. S. Yandell, Y. Song, P. Cohen, J. M.
Friedman, A. D. Attie, Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 11482.
[4] J. E. Stukey, V. M. McDonough, C. E. Martin, J. Biol. Chem. 1990, 265, 20144.
[5] P. H. Buist, B. Behrouzian, J. Am. Chem. Soc. 1996, 118, 6295.
[6] P. H. Buist, H. G. Dallmann, R. R. Rymerson, P. M. Seigel, P. Skala, Tetrahedron Lett. 1988, 29, 435.
[7] G. J. Schroepfer, K. Bloch, J. Biol. Chem. 1965, 240, 54.
[8] P. H. Buist, D. M. Marecak, J. Am. Chem. Soc. 1992, 114, 5073.
[9] B. Behrouzian, D. Hodgson, C. K. Savile, B. Dawson, P. H. Buist, J. Shanklin, Mag. Res. Chem. 2002, 40,
524.
[10] P. H. Buist, H. G. Dallmann, R. R. Rymerson, P. M. Seigel, Tetrahedron Lett. 1987, 28, 857.
[11] P. H. Buist, D. M. Marecak, Tetrahedron Lett. 1991, 32, 891.
[12] J. Shanklin, E. B. Cahoon, Annu. Rev. Plant Physiol. Plant Mol. Biol. 1998, 49, 611.
[13] P. Sperling, P. Ternes, T. K. Zank, E. Heinz, Prostaglandins Leukot. Essent. Fatty Acids 2003, 68, 73.
[14] W. Adcock, N. A.Trout, Chem. Rev. 1999, 99, 1415.
[15] B. Behrouzian, P. H. Buist, J. Shanklin Chem. Commun. 2001, 401.
[16] C. Beckmann, J. Rattke, N. J. Oldham, P. Sperling E. Heinz, W. Boland, Angew. Chem., Int. Ed. 2002, 41,
2298.
[17] K. Bowden, I. M. Heilbron, E. R. H. Jones, B. C. L. Weedon, J. Chem. Soc. 1946, 39.
[18] N. Green, M. Jacobson, T. Henneberry, A. N. Kishaba, J. Med. Chem. 1967, 10, 533.
Received August 19, 2003