Exploring the hydroxylation - Dehydrogenation connection: Novel catalytic activity of castor stearoyl-ACP Δ9 desaturase

Article abstract of DOI:10.1021/ja012252l

The novel product profile obtained by incubating chiral fluorinated substrate analogues with castor stearoyl-ACP Δ⁹ desaturase has been rationalized through a series of labeling studies. It was found that the introduction of the Z-double bond b

Full text of DOI:10.1021/ja012252l

Published on Web 03/06/2002
Exploring the Hydroxylation-Dehydrogenation Connection:
Novel Catalytic Activity of Castor Stearoyl-ACP ∆9 Desaturase
Behnaz Behrouzian,†,‡ Christopher K. Savile,§ Brian Dawson,| Peter H. Buist,*,§ and
John Shanklin*,†
Contribution from the Department of Biology, BrookhaVen National Laboratory,
Upton, New York 11973, OttawasCarleton Chemistry Institute, Department of Chemistry,
Carleton UniVersity, 1125 Colonel By DriVe, Ottawa, Ontario, Canada, K1S 5B6, and
Research SerVices DiVision, Health Products and Food Branch, Health Canada,
Tunney’s Pasture, Ottawa, Ontario, Canada, K1A 0L2
Received September 27, 2001
Abstract: The novel product profile obtained by incubating chiral fluorinated substrate analogues with castor
stearoyl-ACP ∆9 desaturase has been rationalized through a series of labeling studies. It was found that
the introduction of the Z-double bond between C-9 and C-10 of the parent substrate occurs with pro-R
enantioselectivitysa result that accounts for the observed stereochemistry of oxidation products derived
from (9R)- and (9S)-9-fluorostearoyl-ACP. Oxidation of (9R)-9-fluorostearoyl-ACP occurs via at least two
rapidly interchanging substrate conformations in the active site as detected by reaction pathway branching
induced by deuteration at C-10 and C-11. Hydroxylation and desaturation of this substrate share the same
site of initial oxidative attack.
Scheme 1
Introduction
Non-heme diiron-containing enzymes have emerged as
important catalysts for a wide variety of enzymatic processes.1a-d
A prominent member of this family of proteins is the soluble,
plant ∆9 desaturase,2 which carries out the chemo-, regio-, and
stereoselective dehydrogenation (desaturation) of stearoyl-acyl
carrier protein (ACP) 1-ACP (Scheme 1). The product of this
reaction, 2-ACP is a key intermediate in the biosynthesis of
cellular lipids.3 Stearoyl-ACP ∆9 desaturase is also significant
because it is the only desaturase that has been purified in
sufficient quantities to permit detailed structural characterization.
Recent X-ray crystallographic studies4 of the castor enzyme have
determined that the catalytic diiron center is located midway
along a narrow, hydrophobic binding pocket. The shape of the
active site favors a gauche conformation at C9-C10 of the
stearoyl substrate, thus facilitating formation of an olefinic
product with Z-stereochemistry.
desaturase.5a,b A unique feature of this enzyme is the substrate-
enhanced reactivity of its diiron core with O2, which leads to
the formation of a (FeIII)2-peroxo intermediate. It is proposed
that the high-valent iron-oxidant required for the desaturation
reaction is similar to compound Qsan Fe(IV) Fe(IV) species
involved in methane monooxygenase-mediated attack of un-
activated C-H bonds.5c
Recently, mechanistic probes originally developed for mem-
brane-bound fatty acyl desaturases6a-c have been applied to the
study of stearoyl-ACP ∆9 desaturase. We have shown7 that the
C-H cleavage steps at C-9 and C-10 are not subject to an
intermolecular primary deuterium kinetic isotope effect (KIE
∼ 1)san observation that is consistent with a kinetic scheme
Considerable progress has been achieved in understanding
the details of the overall catalytic cycle of stearoyl-ACP ∆9
† Brookhaven National Laboratory.
‡ Current address: Maxygen Inc., 515 Galveston Drive, Redwood City,
CA 94063.
§ Carleton University.
| Health Canada.
(1) (a) Wilkens, R. G. Chem. Soc. ReV. 1992, 21, 171-178. (b) Feig, A. L.;
Lippard, S. L. Chem. ReV. 1994, 94, 759-805. (c) Wallar, B. J.; Lipscomb,
J. D. Chem. ReV, 1996, 96, 715-726. (d) Broadwater, J. A.; Fox, B. G.;
Haas, J. A. Fett/Lipid 1998, 100, 103-113.
(5) (a) Yang, Y.-S.; Broadwater, J.; Pulver, S. C.; Fox, B. G.; Solomon, E. I.
J. Am. Chem. Soc. 1999, 121, 2770-2783. (b) Broadwater, J. A.; Ai, J.;
Loehr, T. M.; Sanders-Loehr, J.; Fox, B. G. Biochemistry 1998, 37, 14664-
14671. (c) Brazeau, B. J.; Austin, R. N.; Tarr, C.; Groves, J. T.; Lipscomb,
J. D. J. Am. Chem. Soc. 2001, 123, 11831-11837.
(2) Shanklin, J.; Cahoon, E. B. Annu. ReV. Plant. Physiol. Plant. Mol. Biol.
1998, 49, 611-641.
(6) (a) Buist, P. H.; Marecak, D. M. J. Am. Chem. Soc. 1992, 114, 5073-
5080. (b) Buist, P. H.; Behrouzian, B. J. Am. Chem. Soc. 1996, 118, 6295-
6296. (c) Buist, P. H.; Behrouzian, B.; Alexopoulos, K.; Dawson, B.; Black,
B. J. Chem. Soc., Chem. Commun. 1996, 2671-2672.
(3) Cook, H. In Biochemistry of Lipids and Membranes; Vance, D. E., Vance,
J. E., Eds.; The Benjamin Cumming Publishing Co. Ltd.: Menlo Park,
CA, 1985; pp 191-203.
(4) Lindqvist, Y.; Huang, W.; Schneider, G.; Shanklin, J. EMBO J. 1996, 15,
4081-4092.
(7) Behrouzian, B.; Buist, P. H.; Shanklin, J. J. Chem. Soc., Chem. Commun.
2001, 401-402.
9
10.1021/ja012252l CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 3277-3283
3277

A R T I C L E S
Behrouzian et al.
sodium benzophenone ketyl. All air- and moisture-sensitive reactions
were performed under N2. Organic extracts were typically shaken with
saturated NaCl and dried over Na2SO4, and solvents were evaporated
in vacuo on a Bu¨chi RE 111 Rotavapor. [1,1-2H2]-1-Bromononane was
prepared by reduction11 of nonanoic acid with LiAlD4 followed by
treatment with HBr/H2SO4.12 [2,2-2H2]-1-Bromononane was prepared
by R-exchange of methyl nonanoate with Na in MeOD13 followed by
reduction with LiAlH411 and treatment with HBr/H2SO4.12 Dimorphet-
heca sinuata seeds were obtained from Butchart Gardens, Victoria, BC,
Canada. Pseudomonas cultures were obtained from the Center for
Catalysis and Bioprocessing, University of Iowa, Iowa City, and
cultured as previously described.14
All buffers and salts, NADPH, bovine serum albumin (BSA), and
other biochemicals were purchased from Sigma-Aldrich. Protein
concentrations were measured by the method of Bradford15 with the
use of BSA as standard protein. The purification of castor stearoyl-
ACP ∆9 desaturase, required cofactors and the synthesis of substrate
ACP derivatives has been previously described.16
Synthesis of Substrates. Racemic 9-Fluorostearates. All racemic
9-fluorostearates were prepared by a Grignard addition reaction between
the appropriate alkenal and a bromoalkane of the appropriate chain
length followed by a sequence of reactions including fluorination,
oxidative cleavage, and methylation as previously described.17 The
relevant analytical data are given below:
Methyl (R,S)-9-Fluorostearate. The analytical data for this com-
pound have been reported previously17 except for 13C NMR δ 173.28
(CdO), 95.54 (C-9, JCF 165.6), 51.44 (OCH3), 35.21 (C-10, JCCF 20.7),
35.16 (C-8, JCCF 20.8), 34.09 (C-2), 31.92 (C-16), 29.57 (C-13), 29.55
(C-12), 29.33, 29.17, 29.07 (C-4 to C-6, C-14 to C-15), 25.17 (C-11,
JCCCF 4.4), 25.09 (C-7, JCCCF 4.4), 24.94 (C-3), 22.70 (C-17), 14.12
(C-18).
dominated by other enzymatic events such as substrate binding
or product release.5b We also discovered that the putative diiron
oxidant acts as an oxygenase with thia analogues. Incubation
of enzyme with 10-thiastearoyl-ACP leads to exclusive sulfoxide
formation, while a mixture of 9-sulfoxide and an as yet
unidentified product was obtained from the 9-thia analogue.7
In a more recent development, it was demonstrated for the first
time that stearoyl-ACP ∆9 desaturase can act as an hydroxylase
when presented with 9-fluorinated stearates.8 This finding is
particularly important since it points to a fundamental mecha-
nistic link6a between dehydrogenation and the hydroxylation
chemistry catalyzed by the structurally related enzyme, methane
monooxygenase (MMO) alluded to above.9a Interestingly,
the latter enzyme was recently shown to give a mixture of
hydroxylated and desaturated products from benzylic substrates.9b
Revealing the latent hydroxylase capability of a structurally
characterized desaturase through the use of a substrate analogue
is intriguing and clearly deserves closer scrutiny.10 In this paper,
we document the results of deuterium-labeling experiments
which demonstrate that the stereochemistry of the parent
dehydrogenation reaction correlates with that of stearoyl-ACP
desaturase-mediated oxidation of 9-fluorinated substrates. In
addition, we show that hydroxylation and desaturation are
initiated at the same carbon through the observation of isoto-
pically induced reaction pathway branching.
Materials and Methods
General Methods. Except where noted otherwise, 1H and 13C NMR
spectra were obtained at 400 and 100 MHz, respectively, on a Bru¨ker
AMX 400 spectrometer with the use of dilute CDCl3 solutions.
Chemical shifts are expressed in ppm (δ) and are referenced to
Methyl [10,10-2H2]-(R,S)-9-Fluorostearate: from decen-9-en-1-
al and [1,1-2H2]-1-bromononane. The spectral data of this compound
were similar to those described for the parent unlabeled compound17
except for the following: 1H NMR δ 1.63 (m, 2H, C(10)-H2, absent),
4.44 (ddd, 1 H, JHF 49.2, JHH 3.6, 7.9, CHF); 13C NMR δ 94.47 (C-9,
upfield â-isotope shift (0.07 ppm)), 35.21 (C-10, absent), 35.12 (C-8,
upfield γ-isotope shift (0.04 ppm)), 29.51 (C-12, upfield γ-isotope shift
(0.04 ppm)), 25.17 (C-11, upfield â-isotope shift (0.20 ppm)); 19F NMR
δ -180.60 (broadened, upfield â-isotope shift (0.41 ppm), unlabeled
material: δ -180.19); MS (EI, 70 eV) m/z 298 (M+ - HF), 266 (M+
- HF, CH3OH).
1
tetramethylsilane. J values are reported in hertz (Hz). H -Decoupled
19F NMR (376.5 MHz) spectra were recorded on a Bru¨ker AM 400
spectrometer with a dedicated 5-mm 19F/1H probe and 19F-specific
amplifier. Under these conditions, nanomolar quantities of enzymatic
products dissolved in CDCl3 could be analyzed. All 19F NMR spectra
were referenced to external trichlorofluoromethane at 0.00 ppm.
Deuterium isotope effects on 13C and 19F NMR shifts were estimated
by running spectra of mixtures (1:2) of labeled and unlabeled material.
Mass spectra were obtained by GC/MS using a HP5973 mass
spectrometer in EI+ mode (70 eV), coupled to a HP6890 GC equipped
with a SP2340 capillary column (60 m × 0.25 mm), temperature
programmed from 100 to 160 °C at 25 °C/min and 160 to 240 °C at
10 °C/min. Isotopic content of analytes was estimated by integrating
the intensities of selected individual ions through the use of HP-
ChemStation software and correcting for natural isotopic abundances
(scan rate, 2 s-1 ). Isotopic ratios for the stereochemical studies were
determined using the corrected intensities for the ions: m/z 298 (299
for d1 analogue), M+ (1); m/z 325 (326 for d1 analogue), (M - 15)+
(2-1-OTMS).
Methyl [11,11-2H2]-(R,S)-9-Fluorostearate: from decen-9-en-1-
al and [2,2-2H2]-2-bromononane. The spectral data of this compound
were similar to that described for the parent unlabeled compound17
except for the following: 13C NMR δ 35.02 (C-10, upfield â-isotope
shift (0.19 ppm)), 29.53 (C-13, upfield â-isotope shift (0.04 ppm)),
29.36 (C-12, upfield â-isotope shift (0.19 ppm)), 25.17 (C-11, absent);
19F NMR δ -180.20 (upfield γ-isotope shift (0.01 ppm), unlabeled
material: δ -180.19); MS (EI, 70 eV) m/z 298 (M+ - HF), 266 (M+
- HF, CH3OH).
Chiral Deuterated Substrates. All chiral deuterated substrates were
prepared from samples of methyl (R)- or (S)-9-hydroxystearate and
methyl (R)- or (S)-10-hydroxystearate.
Methyl (R)-9-Hydroxystearate. D. sinuata seeds (6 g) were ground
in a mortar filled with liquid N2. The resultant coarse powder was stirred
Flash chromatography with silica gel (230-400 mesh) was used to
purify all intermediates and substrates. Visualization of UV-inactive
materials on TLC was accomplished by a combination of water spray
and I2 vapor.
All reagents and starting materials for organic synthesis were
purchased from Sigma-Aldrich and used without purification. Tetra-
hydrofuran (THF) and diethyl ether (Et2O) were freshly distilled from
(11) Crombie, L.; Heaves, A. D. J. Chem. Soc., Perkin Trans. 1 1992, 1929-
1937.
(12) Vogel, A. In Vogel’s Textbook of Practical Organic Chemistry, 4th ed.;
revised by Furniss, B. S., Hannaford, A. J., Rogers, V., Smith, P. W. G.,
Tatchell, A. R., Eds.; Longman: New York, 1978; pp 387-388.
(13) Rackoff, H. Prog. Lipid Res. 1982, 42, 225-254.
(14) Yang, W.; Dostal, L.; Rosazza, J. P. N. Appl. EnViron. Microbiol. 1993,
59, 281-284.
(8) Behrouzian, B.; Buist, P. H.; Shanklin, J. J. Chem. Soc., Chem. Commun.
2001, 765-766. Hydroxylation of parent stearoyl-ACP is not detectable
(<0.1% of total products).
(9) (a) Gherman, B. F.; Dunietz, B. D.; Whittington, D. A.; Lippard, S. J.;
Friesner, R. A. J. Am. Chem. Soc. 2001, 123, 3836-3837. (b) Jin, Y.;
Lipscomb, J. D. JBIC 2001, 6, 717-725.
(15) Bradford, M. M. Anal. Biochem. 1976, 72, 248-254.
(16) (a) Studts, J. M.; Fox, B. G. Protein Expression Purif. 1999, 16, 109-
119. (b) Shanklin, J. Protein Expression Purif. 2000, 19, 319.
(17) Buist, P. H.; Behrouzian, B.; Alexopoulos, K. A.; Dawson, B.; Black, B.
J. Chem. Soc., Perkin. Trans. 1 1997, 2617-2624.
(10) Zhou, J.; Kelly, W. L.; Bachmann, B. O.; Gunsior, M.; Townsend, C. A.;
Solomon, E. I. J. Am. Chem. Soc. 2001 123, 7388-7398.
9
3278 J. AM. CHEM. SOC. VOL. 124, NO. 13, 2002

Castor Stearoyl-ACP ∆9 Desaturase
A R T I C L E S
under N2 in 50 mL of CHCl3/MeOH (2:1 v/v) for 24 h after which the
slurry was filtered by suction and rinsed with CHCl3/MeOH (2:1 v/v,
50 mL). The filtrate was acidified with 0.1 M HCl/0.1M NaCl and
extracted with CHCl3 (2 × 30 mL). The combined organics were dried
and evaporated to give a yellow oil. This residue was treated with 20
mL of 1 M NaOCH3 at 0 °C for 2 h. Addition of 1 N HCl (15 mL),
followed by extraction with hexane (3 × 10 mL) gave crude methyl
dimorphecolate (10(E),12(E)-(R)-9-hydroxyoctadecadienoate) as a clear
viscous oil. Hydrogenation of this material over Pt (Adam’s catalyst)
in 95% ethanol and purification by flash chromatography (15% EtOAc/
hexanes) gave the title compound as a white solid (225 mg): mp 50-
Methyl (S)-10-hydroxystearate was prepared from methyl (R)-10-
hydroxystearate as described above. The title compound was obtained
as a white solid (85 mg, 76%). The spectral data of this material were
identical to those reported for the R-enantiomer. 50% ee (1H NMR of
the corresponding (S)-(+)-O-acetylmandelate derivative). The low %
ee of this material was traced to the use of one batch of microbially
produced (R)-10-hydroxystearate of exceptionally low stereochemical
purity.
Methyl (9S)-[9-2H1]-Stearate ((9S)-[9-2H1]-1). A solution of methyl
(R)-9-hydroxystearate (106 mg, 0.34 mmol) in dry pyridine was treated
with TsCl (133 mg, 0.70 mmol) at 0 °C. The solution was stirred for
1 h at 0 °C and left at 4 °C until complete as determined by TLC
(silica gel, 20% EtOAc/hexanes) (3-4 days). The reaction was
quenched with H2O (10 mL) and extracted with Et2O (3 × 15 mL).
The combined organics were washed with 1 N HCl (3 × 20 mL) and
saturated NaCl (1 × 25 mL), dried over Na2SO4, and concentrated in
vacuo. Purification by flash chromatography (10% EtOAc/hexanes)
gave the tosylate as a viscous oil: 92 mg, 58% yield; 1H NMR δ 0.88
(t, J 6.6, 3H, CH2CH3), 1.20 (br s, 24 H, methylenes), 1.55 (m, 4H,
CH2CH(OTs)CH2), 2.29 (t, J 7.4, 2H, CH2COOMe), 3.67 (s, 3H,
OCH3), 4.53 (p, 1H, CHOTs), 7.31 (d, 2H, phenyl), 7.79 (d, 2H,
phenyl). This intermediate (92 mg, 0.20 mmol) was dissolved in dry
THF (2 mL), and a solution of 1.0 M lithium triethyl borodeuteride
(1.5 mL) in THF was added at 0 °C over 10 min and stirred for 1 h at
room temperature. The reaction was quenched with 1 N HCl (3 mL)
and H2O (3 mL) and extracted with Et2O (3 × 10 mL). The combined
ethereal layers were dried and concentrated in vacuo to give the crude
deuterated alcohol (49 mg) as a low melting solid: 1H NMR δ 0.88 (t,
J 6.8, 3H, CH2CH3), 1.26 (br s, 29 H, methylenes), 1.51 (br t, 2H,
CH2CD2OH). This intermediate was oxidized with 1.5 mL of Jones
reagent (4 g of CrO3, 15 mL of 3 M H2SO4, 2.5 mL of H2O) in acetone
(5 mL) for 4 h at room temperature. Reduction of excess oxidant with
SO2 gas and extraction with Et2O gave the crude carboxylic acid. This
material was subsequently methylated by BF3/MeOH and purified by
flash chromatography (1% EtOAc/hexanes) to give the title compound
as a white solid (31 mg, 30% overall yield). The spectral data of the
title compound were identical to that of methyl stearate except for the
following: MS (EI, 70 eV) m/z 299 (M+), 268 (M+ - CH3O).
Methyl (9R)-[9-2H1]-stearate ((9R)-[9-2H1]-1): from methyl (S)-
9-hydroxystearate. This was obtained as a white solid (24 mg, 32%
overall yield). The spectral data of the title compound were identical
to that of corresponding S-enantiomer (see above).
1
52 °C (lit.18mp 49-51 °C); H NMR δ 0.89 (t, J 6.7, 3H, CH2CH3),
1.27 (br s, 24H, methylenes), 1.58 (m, 4H, CH2-CHOHCH2), 2.30 (t,
J 7.4, 2H, CH2COOMe), 3.55 (p, 1H, CHOH), 3.67 (s, 3H, OCH3);
13C NMR δ 174.56 (CdO), 72.07 (C-9), 51.69 (OCH3), 37.74 (C-8 or
C-10), 37.65 (C-8 or C-10), 34.29 (C-2), 32.15 (C-16), 29.99, 29.91,
29.84, 29.73, 29.59, 29.31 (C-4 to C-6, C-12 to C-15), 25.93 (C-7 or
C-11), 25.82 (C-7 or C-11), 25.14 (C-3), 22.93 (C-17), 14.36 (C-18);
MS (TMS derivative, EI, 70 eV) m/z 259 (TMSOC8H15COOMe), 229
(C10H20OTMS); >98% ee (1H NMR of (S)-1-(1-naphthyl)ethylcarbam-
ate19 or (S)-(+)-O-acetylmandelate derivative14).
Methyl (R)-10-Hydroxystearate. Oleic acid (1.0 g, 3.5 mmol) was
incubated for 48 h with growing cultures (500 mL) of Pseudomonas
NRLL 3266 in the presence of tergitol (1% w/v) according to the
method of Rosazza.14 The culture medium was adjusted to pH 2 with
50% HCl and extracted with ethyl acetate/propanol (9:1 v/v, 1 L). The
combined extracts were washed with water, dried over anhydrous Na2-
SO4, and concentrated in vacuo at 40 °C. The brown residue was
methylated (BF3/MeOH) and chromatographed on flash silica gel (15%
EtOAc/hexanes) to give 105 mg of the title compound as a colorless
solid: mp 53-55 °C (lit.20a mp 56.5-57.0). The spectral data were
similar to those obtained for methyl (R)-9-hydroxystearate except for
the following: 13C NMR δ 174.33 (CdO), 71.88 (C-10), 51.46 (OCH3),
37.56 (C-9 or C-11), 37.50 (C-9 or C-11), 34.10 (C-2), 31.94 (C-16),
29.80, 29.68, 29.46, 29.35, 29.25, 29.16 (C-4 to C-7, C-13 to C-15),
25.73 (C-8 or C-12), 25.68 (C-8 or C-12), 24.97 (C-3), 22.72 (C-17),
14.36 (C-18); MS (TMS derivative, EI, 70 eV) m/z 273 (TMSOC9H17-
COOMe), 215 (C9H18-OTMS); % ee > 98% (1H NMR of the
corresponding (S)-(+)-O-acetylmandelate derivative).14
Methyl (S)-9-Hydroxystearate. Diethyl diazodicarboxylate (DEAD)
(345 mg, 1.98 mmol) in 1 mL of dry THF was added to a solution of
methyl (R)-9-hydroxystearate (205 mg, 0.65 mmol), benzoic acid (240
mg, 1.97 mmol), and triphenylphosphine (670 mg, 2.56 mmol) in 5
mL of anhydrous THF over 5 min at room temperature. The reaction
was monitored by TLC (silica gel, 20% EtOAC/hexane), and additional
DEAD was added as required (∼200 mg) until starting material was
consumed (2 h). The reaction mixture was concentrated in vacuo and
flash chromatographed (7.5% EtOAc/hexane) to give the corresponding
benzoate ester as an oil (125 mg, 46% estimated yield): 1H NMR δ
0.87 (t, J 6.6, 3H, CH2CH3), 1.29 (br s, 24 H, methylenes), 1.59 (m,
4H, CH2CH(OC(O)Ph)CH2), 2.28 (t, J 7.4, 2H, CH2COOMe), 5.11 (p,
1H, CHO(CO)Ph) 3.66 (s, 3H, OCH3), 7.50 (m, 3H, phenyl), 8.04 (m,
2H, phenyl). Hydrolysis of this intermediate followed by methylation
(BF3/MeOH) and flash chromatography (15% EtOAc/Hexanes) yielded
the title compound as a white solid (80 mg, 39% overall). The spectral
data of this material were identical to those reported for the R-
enantiomer. >98% ee (1H NMR of (S)-(+)-O-acetylmandelate deriva-
tive14).
Methyl (10S)-[10-2H1]-stearate ((10S)-[10-2H1]-1): from methyl
(R)-10-hydroxystearate. This was obtained as a white solid (14 mg,
23% overall yield). The spectral data of the title compound were
identical to those reported for methyl stearate except for the follow-
ing: (EI, 70 eV) m/z 299 (M+), 268 (M+ - CH3O).
Methyl (10R)-[10-2H1]-stearate ((10R)-[10-2H1]-1): from methyl
(S)-10-hydroxystearate. This was obtained as a white solid (32 mg,
40% overall yield). The spectral data of the title compound were
identical to those of corresponding S-enantiomer (see above).
Desaturase Assay. Reactions of acyl-ACP derivatives with ∆9
desaturase were carried out at room temperature. Each reaction mixture
consisted of ∆9 desaturase (0.6 nmol), ferredoxin (1.0 nmol), NADPH:
ferredoxin oxidoreductase (1.8 nmol), and acyl-ACP (8.1 nmol) in a
total volume of 200 µL of buffer. The reaction was initiated by the
addition of NADPH (1.0 mg, 1.2 µmol) in buffer (50 µL) and allowed
to continue for 30 min. The reaction was terminated with the addition
of THF (100 µL), and the thioester linkage of ACP derivatives was
reduced to the corresponding primary alcohol with NaBH4 at 37 °C
for 15 min. The residue was diluted with H2O (1 mL) and extracted
with hexane (2 × 2 mL). (Similar product profiles were obtained when
CH2Cl2 was used as the extracting solvent.) The phases were separated
by centrifugation, and the organic layers were collected. The combined
organics were evaporated under a steady stream of N2 and derivatized
with BSTFA/TMCS (50 µL) at 37 °C for 1 h. Excess reagent was
(18) Smith, C. R.; Wilson, T. L.; Melvin, E. H.; Wolff, I. A. J. Am. Chem.
Soc., 1960, 82, 1417.
(19) Sonnet, P. E.; Osman, S. F.; Gerard, H. C.; Dudley, R. L. Chem. Phys.
Lipids 1994, 69, 121.
(20) (a) Schroepfer, G. J.; Bloch, K. J. Biol. Chem. 1965, 240, 54-63. (b) Morris,
L. J.; Harris, R. V.; Kelly, W.; James, A. T. Biochem J. 1968, 109, 673-
678. (c) The stereochemical studies described in refs 20a and 20b were
conducted using in vivo microbial ∆9 desaturases; the structural relationship
of these systems to stearoyl-ACP ∆9 desaturase has not been defined.
9
J. AM. CHEM. SOC. VOL. 124, NO. 13, 2002 3279

A R T I C L E S
Behrouzian et al.
Scheme 2
both incubations under conditions of partial and complete
substrate conversion. The recovery of total products and
remaining products was estimated to be ∼70-90% of theoreti-
cal. The structural assignments of enzymatic products are based
on a combination of 1H-decoupled 19F NMR and GC/MS
analysis with the help of reference standards and deuterium-
labeled substrates.8
To rationalize these results, we felt that it would be instructive
to elucidate the enantioselectivity of the parent ∆9 desaturation
reaction.20a-c Our stereochemical study was accomplished by
incubating ACP derivatives of four stearates that were ste-
reospecifically deuterated at the C-9 and C-10 positions: (9R)-
and (9S)-[9-2H1]-1 and (10R)- and (10S)-[10-2H1]-1.
evaporated under a stream of N2, and the residue was diluted with 50
µL of hexane for analysis by GC/MS.
Results and Discussion
Stereochemical Studies. We have previously shown8 that
separate incubation of (9S)-3-ACP and (9R)-3-ACP with
stearoyl-ACP ∆9 desaturase generates two different sets of
oxidation products (Scheme 2).21 A threo-9,10-fluorohydrin 5
along with a fluoroolefin of “normal” stereochemistry 4 was
formed from (9S)-3-ACP. In contrast, a 1:2 mixture of threo-
and erythro-9,10 fluorohydrins 8 and 922 as well as traces of
diastereomeric 9,11 fluoro alcohols 10 and 1123 accompanied
two novel olefinic products 6 and 7 upon incubation of (9R)-
3-ACP. Products attributable to hydroxylation at C-9 were not
detected (detection limit, 0.1% of total products). Oxygenated
products accounted for ∼10% of the total product mixture in
These compounds were prepared in straightforward fashion
from the corresponding chiral hydroxystearates by a tosylation/
superdeuteride displacement sequence as described in Materials
and Methods. The source of (R)- and (S)-9-hydroxystearic acid
(>98% ee) was identical to that employed for the chiral
9-fluorostearate syntheses.8 (R)-10-Hydroxystearic acid was
obtained by the known stereo- and regioselective hydration of
oleic acid using growing cultures of Pseudomonas NRRL
3266.14 The required (S)-10-hydroxystearic acid was prepared
from the corresponding R-enantiomer by the same inversion
sequence as was utilized for the 9-hydroxy series.8 The outcome
of the four incubations is shown in Table 1 and demonstrates
clearly that ∆9 desaturation catalyzed by stearoyl ACP desatu-
rase involves syn removal of the pro-R hydrogens at C-9 and
C-10 by the putative compound Q-type oxidant: Essentially
complete loss of deuterium was observed upon desaturation of
R-deuterated substrates while olefin produced from S-labeled
stearates retained deuterium label (Scheme 3).
These findings account for the product profile obtained from
the two 9-fluorostearoyl ACP derivatives (Scheme 2). We note
that the desaturation of (9S)-3-ACP generates the fluoroolefinic
product 4 with the “normal” configuration since the enzymic
oxidant can carry out syn removal of the pro-R hydrogen at
C-10 and the corresponding hydrogen at C-9 of substrate
residing in a gauche conformation. Importantly, this picture is
also consistent with the production of threo-fluorohydrin 5 from
(9S)-3-ACP via an apparent MMO-type hydroxylation process
which would be expected to proceed by attack at the pro-R
(21) A series of control experiments8 have confirmed that the fluorohydrin
products were not derived by hydration of fluoroolefin during the workup
procedure. We have also demonstrated high (75-85%) incorporation of
18O into the fluorohydrin products when incubations were carried out under
an atmosphere of 18O2 (unpublished experiments).
(22) As noted previously,8 the erythro-9, 10-fluorohydrin 9 appeared to be
accompanied by varying amounts of the regioisomeric erythro-10,9-
fluorohydrin as evidenced by the appearance of two partially resolved
components in the erythro-fluorohydrin region of the 19F NMR spectrum
and GC/MS chromatogram (SIMS mode). The presence of erythro-10,9-
fluorohydrin was confirmed by the observation of a substantial upfield
â-isotope shift (0.5 ppm) on one of the 19F NMR signals due to the erythro-
11-d2-labeled fluorohydrins derived from 11-d2-(R,S)-9-fluorostearoyl ACP.
(This large isotope shift is not seen for the threo -fluorohydrin 11-d2-8, as
expected.) The incorporation of 18O (see above, ref 21) into the fluorine-
shifted compound was similar to that of the other fluorohydrins: 18O
incorporation, 85% ( 2 as measured at m/z 303/305. The origin of this
product of fluorine migration is currently under investigation.
(23) These previously unreported compounds appeared as two very minor peaks
(<1%) in the fluorohydrin region of the GC/MS chromatogram and were
identified by their MS (diTMS derivatives: m/z 201 (C8H16OTMS)+, 329
[M - (C7H15) - HF]+). The correctness of the assignment of hydroxyl
position was confirmed by subsequent deuterium labeling experiments (vide
infra).
9
3280 J. AM. CHEM. SOC. VOL. 124, NO. 13, 2002

Castor Stearoyl-ACP ∆9 Desaturase
A R T I C L E S
Table 1. Isotopic Contenta of Stereospecifically Monodeuterated Stearates and ∆9 Desaturated Products
substrates
products
%d0
%d1
%d0
%d1
% retention oflabelb
(9R)-[9 -2H1]-1
(9S)-[9 -2H1]-1
4.5 ( 1.3
4.2 ( 0.7
3.9 ( 0.6
6.2 ( 0.2
95.5 ( 1.3
95.8 ( 0.7
96.1 ( 0.6
93.8 ( 0.2
96.4 ( 0.2
4.4 ( 0.6
67.5 ( 1.4
8.2 ( 0.4
3.6 ( 0.2
95.6 ( 0.6
32.5 ( 1.4c
91.8 ( 0.4
3.8 ( 0.2
99.8 ( 0.9
2.2 ( 2.2c
97.9 ( 0.5
(10R)-[10 -2H1]-1c
(10S)-[10 -2H1]-1
b
a Each incubation was repeated two times, and the deuterium content is given as an average value ( standard deviation. Percent retention of label ) [%
c
d1 (product)/(% d1 (substrate) × 100]. Calculation corrected for the fact that the deuterated substrate contained ∼33% of the corresponding (10S)-enantiomer
as assessed by the % ee of the synthetic precursor-methyl (10S)-10-hydroxystearate (see Materials and Methods).
synthesis6c (see Materials and Methods). We were then able to
Scheme 3
prepare the corresponding ACP derivatives, which were greatly
enriched in the (R)-9-fluorostearoyl stereoisomer due to a
substantial diastereoselective discrimination which occurs during
the ACP synthase-mediated acylation of (R,S)-9-fluorostearic
acid as previously noted.8
hydrogen at C-10 with predominately retention of configuration.24a
An entirely different set of products is obtained upon oxidation
of (9R)-3-ACP, because the C-9 hydrogen at the pro-R position
of this substrate is replaced by fluorine: These include fluo-
roolefins with “abnormal” configuration 6 and regiochemistry
7 as well as a mixture of C-10- and C-11-hydroxylated products
8-11 (Scheme 2). Possible substrate conformations leading to
these novel products are described in the next section.
The Hydroxylation/Desaturation Connection. The observa-
tion that both olefins and fluorohydrins derived from (9R)-3-
ACP were obtained as regioisomeric pairs (∆9,∆10; 10-ol,11-
ol) prompted us to examine how these products might be related
in a mechanistic sense. According to a previously proposed
model,6a hydroxylation and desaturation pathways share a
common initial, irreversible, C-H activation step such as
H-abstraction. The location of this event in the case of (9R)-
Each deuterated substrate, as well as a nonlabeled control,
was incubated with stearoyl-ACP ∆9 desaturase under conditions
of high (>∼85%) conversion. The product profiles obtained in
these experiments are presented in Table 2 and displayed
graphically in Figure 1. Examination of these data reveals that
the amount of fluoroolefin 10-d1-6 and 9,10-fluorohydrins
10-d1-8 and 10-d1-9 derived from the [10,10-2H2]-(9R)-3 was
greatly reduced relative to the levels obtained from the corre-
sponding d0-substrate. In contrast, the intensity of peaks due to
allylic fluoride 10-d1-7 and the 9,11-fluorohydrins, 10-d2-10 and
10-d2-11 (ratio of 10/11 ∼1:3)26 was substantially enhanced.
3-ACP oxidation can potentially be probed through isotopically
Precisely the opposite trends were observed upon oxidation of
induced branching of the reaction pathway. That is, if this
the [11-2H2]-fluorostearate: The amount of 11-d1-7, 11-d1-10
and 11-d1-11 produced was attenuated while the levels of
11-d2-6, 11-d2-8 and 11-d2-9 increased.
Our interpretation of these results is presented in Scheme 4:
Oxidation of (9R)-3-ACP proceeding via an anti conformation
similar to A is initiated by the energetically difficult C-H
cleavage at C-10 to give a short-lived intermediate, which
collapses to olefinic product 6 by a relatively fast,6b second C-H
substrate presents two different methylene groups to enzymatic
attack via rapidly interchanging conformations in the active site,
and if the initial reaction leads to two different sets of products,
then the presence of deuterium at either methylene position
could, in principle, influence the proportion of product pairs
obtained from each conformation. Precedent for this type of
phenomenon has been established for other enzyme systems.25
To test our hypothesis, two samples of labeled (9R)-3-ACPs
one bearing deuterium at C-10 and the other at C-11swere
required. The synthesis of these materials was accomplished
by first synthesizing methyl [10,10-2H2]- and [11,11-2H2]-(R,S)-
bond breaking step at C-9. A closely related reaction pathway
yields mainly erythro-fluoro alcohol 9, assuming, as is reason-
able, that hydroxylation occurs predominately with retention of
configuration.24a,b In contrast, formation of allylic fluoride 7 and
9-fluorostearate via a modification of a previously published
11, the putative,26 major hydroxylated product, involves initial
reaction at C-11 on a coiled (9R)-3 (conformation similar to
B).27 In both cases, the point of attack is dictated by a strong
regiochemical imperative. Because equilibration between alter-
nate substrate conformations is faster than the initial C-H
(24) (a) Valentine, A. M.; Wilkinson, B.; Liu, K. E.; Komar-Panicucci, S.;
Priestley, N. D.; Williams, P. G.; Morimoto, H.; Floss, H. G.; Lippard, S.
J. J. Am. Chem. Soc. 1997, 119, 1818-1827. (b) The production of the
minor stereoisomers 8 (and 10) is due to the fact that either hydroxylation
proceeds with less than 100% stereospecificity for this substrate or that
substrate conformations in addition to those depicted in Scheme 4 are
involved in this reaction.
(25) Atkinson, J. K.; Hollenberg, P. F.; Ingold, K. U.; Johnson, C.; Le Tadic,
M.; Newcomb, M.; Putt, D. A. Biochemistry 1994, 33, 10630-10637 and
references therein.
(26) The assignment of 11(R)-configuration to the predominant stereoisomer
11 must be regarded as tentative at this time as it is based solely on the
mechanistic considerations depicted in Scheme 4.
9
J. AM. CHEM. SOC. VOL. 124, NO. 13, 2002 3281

A R T I C L E S
Behrouzian et al.
Table 2. Profile of Products Obtained by Oxidation of Predominately (9R)-3-ACP, [10,10 -2H2]-3-ACP, and [11,11-2H2]-3-ACP by
Stearoyl-ACP ∆9-Desaturase
product profilea
b
b
(Z)−9F-9-ene
9F-10-ene
9F-10-OH
9F-11-OH
(E)-9F-9-ene
substrates
6
7
8, 9
10, 11
4
otherc
3
28.7 ( 1.7
4.5 ( 0.2
51.4 ( 1.0
46.5 ( 1.4
71.5 ( 1.3
4.6 ( 0.8
11.6 ( 1.1
1.7 ( 0.3
24.5 ( 0.6
0.9 ( 0.4
11.0 ( 0.6
0.3 ( 0.1
6.0 ( 0.7
4.4 ( 1.0
4.6 ( 0.8
6.4 ( 1.0
6.9 ( 0.2
5.3 ( 0.3
[10 -2H2]-3
[11 -2H2]-3
a Each incubation was carried out to greater than 85% conversion and repeated at least three times; percent composition is given as an average value (
standard deviation and was determined by GC/MS analysis and corroborated by 19F NMR. The mass spectra of analytes derived from deuterated substrates
b
were in accord with the anticipated isotopic content. Contributions of both diastereoisomers were combined since the ratios of 8:9 and 10:11 appeared to
be invariant in all three experiments. Contribution of 5 from (S)-9F to amount of 9F-10-OH was neglected. c Total of unidentified peakssassumed to be
artifacts of workup procedure.
Figure 1. Isotopic discrimination in oxidation of [10,10 -2H2]-(9R)-3-ACP and [11,11-2H2]-(9R)-3-ACP by stearoyl-ACP ∆9 desaturase. The structures of
the analytes are given in Scheme 2.
Scheme 4
activation step, the oxidant can select the energetically lowest
pathway when presented with a substrate that is dideuterated at
C-10 or C-11sa phenomenon that leads to the manifestion of
a kinetic isotope effect on the initial C-H cleavage at C-10
and C-11 leading to olefin formation. The intramolecular nature
of this substantial isotope effect renders it less subject to masking
by other enzymatic steps.28,29 Large “intrinsic” primary30
deuterium KIEs are typical of C-H cleavages involving
(27) There is some precedent for the production of a mixture of (Z)- ∆9- and
(28) A detailed treatment of enzymatic kinetic isotope effects and the sensitivity
to masking effects is given in: Miwa, G. T.; Garland, W. A.; Hodshon, B.
J.; Lu, A. Y. H.; Northrop, D. B. J. Biol. Chem. 1980, 255, 6049-6054.
(29) Rataj, M. J.; Kauth, J. E.; Donnelly, M. I. J. Biol. Chem. 1991, 18684-
18690.
10
∆
-desaturated products in this system. Reaction of 7(E)-octadecenoyl-
ACP by stearoyl-ACP ∆9 desaturase gave a (4:1) mixture of (7E,9Z)-
octadecadienoyl-ACP and (7E,10Z)-octadecadienoyl-ACP: Broadwater, J.
A.; Laundre, B. J.; Fox, B. G. J. Inorg. Biochem. 2000, 78, 7-14.
9
3282 J. AM. CHEM. SOC. VOL. 124, NO. 13, 2002

Castor Stearoyl-ACP ∆9 Desaturase
A R T I C L E S
oxidation of unactivated C-H bonds and are reminiscent of
values obtained for the initial oxidation step involved in a
number of membrane-bound desaturase systems.6b,31a-e
9-Fluorinated substrates are processed by stearoyl-ACP ∆9
desaturase to give novel oxygenated and olefinic products. In
the case of (9R)-fluorostearoyl ACP oxidation, we were able to
show that desaturation and hydroxylation are initiated at the
same site and possibly via an identical transition state. How
these two pathways diverge, and the effect of fluorine substitu-
tion on the ratio of the energy barriers to each route, remains
to be identified; the use of emerging techniques for modeling
the reactivity of diiron-containing enzymes may shed further
light on this fascinating mechanistic problem.9
Conclusions
The stereochemistry of C-H cleavage mediated by stearoyl-
ACP ∆9 desaturase has been unambiguously determined for the
first time since the discovery of this class of enzymes some
thirty-five years ago.32 The observed pro-R enantioselectivity
correlates well with the markedly different product profiles
obtained by oxidation of stearoyl substrates bearing a (9R)- and
(9S)-monofluorine substituent. These data are critical to the
interpretation of ongoing biophysical studies involving substrate
analogue-enzyme complexes.
Acknowledgment. This work was supported by an NSERC
PDF (to B.B.), NSERC Grant OGP-2528 (to P.H.B.), by the
Office of Basic Energy Sciences of the United States Department
of Energy, and in part by Dow and Dow Agrosciences (to J.S.).
We are indebted to Dr. John Broadwater (BNL) for useful
discussions and Jamie Cote´ and Christine Caputo at Carleton
University for preparing some of the substrates. The technical
assistance of Dr. Ted Edwards is also gratefully acknowledged.
Cultures of Pseudomonas were a generous gift of Dr. John
Rossaza, University of Iowa, Iowa City, IA.
(30) These values may include secondary kinetic isotope effects (∼10% of
primary KIE): Jones, J. P.; Trager, W. F. J. Am. Chem. Soc. 1987, 109,
2171-2173.
(31) (a) Buist, P. H.; Behrouzian, B. J. Am. Chem. Soc. 1998, 120, 871-876.
(b) Pinilla, A.; Camps, F.; Fabrias, G. Biochemistry 1999, 38, 15272-
15277. (c) Meesapyodsuk, D.; Reed, D. W.; Savile, C. K.; Buist, P. H.;
Covello, P. S. Biochemistry 2000, 39, 11948-11954. (d) Fauconnot, L.;
Buist, P. H. J. Org. Chem. 2001, 66, 1210-1215. (e) Behrouzian, B.;
Fauconnot, L.; Daligault, F.; Nugier-Chavin, C.; Patin, H.; Buist, P. H.
Eur. J. Biochem. 2001, 268, 3545-3549.
(32) Nagai, J.; Bloch, K. J. Biol. Chem. 1966, 241, 1925-1927.
JA012252L
9
J. AM. CHEM. SOC. VOL. 124, NO. 13, 2002 3283