Tuning the regio- and stereoselectivity of C-H activation in n-octanes by cytochrome P450 BM-3 with fluorine substituents: Evidence for interactions between a C-F bond and aromatic π systems

Article abstract of DOI:10.1002/chem.201003631

We employed the water- soluble cytochrome P450 BM-3 to study the activity and regiospecificity of oxidation of fluorinated n-octanes. Three mutations, A74G, F87V, and L188Q, were introduced into P450 BM-3 to allow the system to undergo n-octane oxidation. In addition, the alanine at residue 328 was replaced with a phenylalanine to introduce an aromatic residue into the hydrophobic pocket to examine whether or not van der Waals interactions between a C-F substituent in the substrate and the polarizable π system of the phenylalanine may be used to steer the positioning of the substrate within the active-site pocket of the enzyme and control the regioselectivity and stereoselectivity of hydroxylation. Interestingly, not only was the regioselectivity controlled when the fluorine substituent was judiciously positioned in the substrate, but the electron input into the iron-heme group became tightly coupled to the formation of product, essentially without abortive side reactions. Remarkable enhancement of the coupling efficiency between electron input and product formation was observed for a range of fluorinated octanes in the enzyme even without the A328F mutation, presumably because of interactions of the C-F substituent with the π system of the porphyrin macrocycle within the active-site pocket. Evidently, tightening the protein domain containing the heme pocket tunes the distribution of accessible enzyme conformations and the associated protein dynamics that activate the iron porphyrin for substrate hydroxylation to allow the reactions mediated by the high-valent Fe^IV=O to become kinetically more commensurate with electron transfer from the flavin adenine dinucleotide (FAD)/flavin mononucleotide (FMN) reductase. These observations lend compelling evidence to support significant van der Waals interactions between the CF₂ group and aromatic π systems within the heme pocket when the fluorinated octane substrate is bound. Activation of F-octanes: Cytochrome P450 BM-3-A74GF87VL188Q (see figure) and the A328F variant regionselectively converted fluorinated C8 alkanes to the corresponding secondary alcohols. The pattern of reactivity, especially the unprecedented regio- and stereoselectivity, observed for 4,4-difluorooctane suggested that specific interactions of the fluorinated substituent with aromatic π systems within the active site could tune the reactivity. Copyright

Full text of DOI:10.1002/chem.201003631

DOI: 10.1002/chem.201003631
À
Tuning the Regio- and Stereoselectivity of C H Activation in n-Octanes by
Cytochrome P450 BM-3 with Fluorine Substituents: Evidence for
À
Interactions Between a C F Bond and Aromatic p Systems
Li-Lan Wu,[a, b] Chung-Ling Yang,[a, c] Feng-Chun Lo,[a] Chih-Hsiang Chiang,[a, b]
Chun-Wei Chang,[a, b] Kok Yaoh Ng,[a] Ho-Hsuan Chou,[a] Huei-Ying Hung,[a, b]
Sunney I. Chan,[a] and Steve S.-F. Yu*[a, b]
Abstract: We employed the water-
soluble cytochrome P450 BM-3 to
study the activity and regiospecificity
of oxidation of fluorinated n-octanes.
Three mutations, A74G, F87V, and
L188Q, were introduced into P450
BM-3 to allow the system to undergo
n-octane oxidation. In addition, the
alanine at residue 328 was replaced
with a phenylalanine to introduce an
aromatic residue into the hydrophobic
pocket to examine whether or not van
was the regioselectivity controlled
when the fluorine substituent was judi-
ciously positioned in the substrate, but
the electron input into the iron–heme
group became tightly coupled to the
formation of product, essentially with-
out abortive side reactions. Remark-
able enhancement of the coupling effi-
ciency between electron input and
product formation was observed for a
range of fluorinated octanes in the
enzyme even without the A328F muta-
tion, presumably because of interac-
within the active-site pocket. Evidently,
tightening the protein domain contain-
ing the heme pocket tunes the distribu-
tion of accessible enzyme conforma-
tions and the associated protein dy-
namics that activate the iron porphyrin
for substrate hydroxylation to allow
the reactions mediated by the high-
valent FeIV=O to become kinetically
more commensurate with electron
transfer from the flavin adenine di-
AHCTUNGTRENNnGUN ucleotide (FAD)/flavin mononucleo-
À
der Waals interactions between a C F
tide (FMN) reductase. These observa-
tions lend compelling evidence to sup-
port significant van der Waals interac-
tions between the CF2 group and aro-
À
substituent in the substrate and the po-
larizable p system of the phenylalanine
may be used to steer the positioning of
the substrate within the active-site
pocket of the enzyme and control the
regioselectivity and stereoselectivity of
hydroxylation. Interestingly, not only
tions of the C F substituent with the p
system of the porphyrin macrocycle
matic
p systems within the heme
À
Keywords: C H activation · cyto-
pocket when the fluorinated octane
substrate is bound.
chrome P450 · enzyme catalysis ·
fluorine · mutagenesis
Introduction
flavin mononucleotide (FMN) reductase[1] that accepts re-
ducing equivalents from nicotinamide adenine dinucleotide
phosphate (NADPH) to activate O2 at the heme iron and
mediates the hydroxylation of saturated and unsaturated
fatty acids with a chain length of 12 to 20 carbon atoms at
their subterminal ends.[2] Because of its high yield of heterol-
ogous overexpression in Escherichia coli and profound turn-
over efficiency,[3] the BM-3 enzyme has been subjected to di-
rected evolution for the activation of small alkanes,[4] fine
chemical conversions,[5] and pharmaceutical lead optimiza-
tion.[6]
Cytochrome P450 BM-3 (CYP102A1) is a water-soluble
monooxygenase with a flavin adenine dinucleotide (FAD)/
[a] L.-L. Wu, C.-L. Yang, Dr. F.-C. Lo, C.-H. Chiang, C.-W. Chang,
K. Y. Ng, Dr. H.-H. Chou, H.-Y. Hung, Prof. Dr. S. I. Chan,
Dr. S. S.-F. Yu
Institute of Chemistry, Academia Sinica
Taipei 115 (Taiwan)
Fax : (+886)2-2783-1237
It is usually difficult to achieve activation of aliphatics by
E-mail: sfyu@chem.sinica.edu.tw
À
conventional synthetic methods due to the high C H bond
[b] L.-L. Wu, C.-H. Chiang, C.-W. Chang, H.-Y. Hung, Dr. S. S.-F. Yu
Department of Chemistry
strength.[7] However, it has been demonstrated that P450
BM-3 with the mutations A74G, F87V, and L188Q (3mt pro-
tein) hydroxylates n-octane into the 2-, 3-, and 4-ol with rea-
sonable activity.[8] Whereas the asymmetric environment
within the active-site pocket offers opportunities to exploit
amino acid substitutions to tune substrate specificity, it is
seldom possible to accomplish controlled regioselectivity
and stereoselectivity for a given substrate.[5,9] Other strat-
National Cheng Kung University
Tainan 701 (Taiwan)
[c] C.-L. Yang
Graduate Institute of Engineering
National Taiwan University of Science and Technology
Taipei 106 (Taiwan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003631.
4774
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

FULL PAPER
egies need to be developed to achieve highly regiospecific
and stereoselective chemical conversions.
examples. Recently, we have demonstrated that 1,1,1-tri-
fluoropropane and -propene can also serve as substrates for
the particulate methane monooxygenase from Methylococ-
cus capsulatus (Bath), with varied chiral selectivity or great-
er stereoselectivity than the corresponding parent hydrocar-
bons.[10] The regioselectivity of the hydroxylation of 5,5-di-
fluorocamphor by cytochrome P450cam is dramatically
changed from 5-exo to C-9.[16] Herein, we show that it is also
possible to tune the reactivity, regiospecificity, and stereo-
Research in our laboratory has been directed toward un-
derstanding the effects of fluorinated substituents on the
regio- and stereoselectivity of oxidation of aliphatics mediat-
ed by pMMO from MethylACTHNUTRGNEUNG
ococcus capuslatus (Bath)[10] and
cytochrome P450 BM-3 from Bacillus megaterium. Fluorine
substitution can usually improve the penetration of aliphat-
ics through the membranes, tune pKa values to render them
more biologically compatible, and enhance their metabolic
stability.[6b,11] Metabolic stability is especially susceptible in
cytochrome P450 proteins.[11a,12] However, much less is
known about the effects of a fluorinated substituent on the
stereochemical control of substrate oxidation by P450. In
AHCTUNGTREGsNNNU electivity of the hydroxylation of octane by the P450 BM-3
3mt enzyme by the introduction of fluorinated substituents.
Results
À
terms of electronegativity and dipole moment, the C F
À
bond is similar to the C OH bond. Molecules containing a
Preparation of fluorinated n-octanes: A series of designed
and synthesized fluorinated n-octanes 1–7 (Schemes 1 and
2) were deployed as the substrates for the P450 BM-3 var-
iants. 1-Fluorooctane (5) was obtained from Sigma–Aldrich.
Fluorinated octanes 1–4 and 6 were prepared by employing
the fluorination reagent Deoxo-Fluor (Matrix Scientific) in
CH2Cl2 (Supporting Information).[14a] 1-Octanal (8); 1,8-
octandiol (12); and 2-, 3-, and 4-octanone (9–11) were used
as the starting materials to generate the corresponding prod-
ucts, 1,1-, 2,2-, 3,3-, and 4,4-difluorooctanes (1–4) and 1,8-di-
fluorooctane 6, respectively. (Table S1 in the Supporting In-
formation)
À
C F bond typically prefer positively polarized environments
À
À
À
À
À
to form C F···H N, C F···C=O, and C F···H Ca associa-
tions.[11b] However, there are many examples in biological
À
systems that suggest that the C F bond is more similar to
[10,13]
À
the C H bond.
Because of the bioisosteric property, flu-
orinated compounds have also become pharmaceutically sig-
nificant for lead optimization.[6b,11] Organofluorine deriva-
tives of potential inhibitors or drugs are also extensively
used as therapeutics targeted at proteins and cellular enzy-
mes.[11,13b]
In this study, we have introduced one, two, or three fluo-
rine atoms into specific carbon positions of the n-octane
molecule to study the effects of the fluorinated substrate on
the oxidation. To achieve fluorination at specific carbon po-
sitions, we have considered the two powerful nucleophilic
fluorinating reagents bis(2-methoxyethyl)aminosulfur tri-
fluoride (Deoxo-Fluor) and diethylaminosulfur trifluoride
(DAST).[14] Deoxo-Fluor appears to be a somewhat more ef-
fective fluorinating reagent than DAST because of its signif-
icantly higher thermal stability and greater reactivity. Gener-
ally, Deoxo-Fluor readily converts alcohols into alkyl fluo-
rides, aldehydes/ketones into gem-difluorides, and carboxylic
acids into the corresponding trifluoromethyl derivatives.
If P450 activity toward the oxidation of aliphatics is main-
tained, or even becomes enhanced, with fluorinated sub-
stituents, then most likely, these bioisosteric derivatives
would behave just like the parent n-octanes when encapsu-
lated into the hydrophobic pocket of the enzyme. Any stabi-
lization of the fluorinated substrate or enhancement in the
oxidation of octane would presumably come from additional
electrostatic interactions, such as hydrogen-bonding and
dipole–dipole interactions, and/or van der Waals interactions
(dipole–induced-dipole and London dispersive forces) of the
Product formation was monitored by gas chromatography
(GC) and gas chromatography/mass selective detector (GC/
MSD) at 1008C under isothermal conditions (see the Sup-
porting Information). We observed 1–4 and 6 at tR =4.24,
3.97, 4.06, 4.00, and 5.99 min, respectively, whereas 8–12 ap-
peared at tR =6.56, 6.46, 6.33, 6.02, and 15.32 min, respec-
tively, under the same separation conditions. Mass spectral
analysis of 1–4 did not reveal a signal with molecular mass
m/z 150. Among the molecular mass fragments, we usually
observed a signal at m/z 130 corresponding to the elimina-
tion of hydrogen fluoride during ionization. Mass fragments
were also detected at m/z 135 and 121, which were derived
from the mass fragmentation of 2 and 3, respectively. These
+
two signals represented the mass fragment CH
and CH3A
3ACHTUGNTERN(NUNG CH2)5CF2
(CH2)4CF2+, respectively, which confirmed the for-
mation of the target products, 2,2-difluorooctane and 3,3-di-
fluorooctane.
The fluorinated products were further identified by 1H,
13C, and 19F NMR spectroscopy (see the Supporting Infor-
mation). A 1H NMR signal with a chemical shift at d=
5.77 ppm (tt, 57.2 and 4.4 Hz) for the primary carbon indi-
cated the formation of 1. Similarly, triplets in the ratio of
1:2:1 with splitting constants of 8.0 and 18.4 Hz, 6.8 and
7.6 Hz, and 8.0 and 8.0 Hz also appeared at d=0.86 and
1.55 ppm, d=0.87 and 0.98 ppm, and d=0.90 and 0.94 ppm,
respectively, for the methyl groups of 2–4, respectively, after
fluorination. Evidently, we had incorporated two F atoms
specifically at the target carbon position. Furthermore, com-
pound 6,with two hydrogen atoms at the primary carbon,
À
polar C F bonds with residues lining the heme pocket.
Unlike hydrogen-bonding or dipole–dipole interactions, van
der Waals interactions do not usually provide specific direc-
tionality or orientations like hydrogen bonding. Moreover,
the strengths are smaller than hydrogen bonding, typically
in the range of about 0.02–2.0 kcalmolÀ1,[15] but these inter-
actions are highly dependent on the shape and volume of
the interacting molecular partners. To illustrate, we cite two
1
gave a H NMR signal at d=4.41 ppm (dt, 47.4 and 6.0 Hz).
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4775

S. S.-F. Yu et al.
Scheme 1. The enzymatic conversions of substrates 1–4 mediated by P450 BM-3_A74G F87V L188Q. More details are given in Scheme S3 in the Sup-
porting Information. DCC=dicyclohexylcarbodiimide, DMAP=4-dimethylaminopyridine.
1,1,1-Trifluorooctane (7) was synthesized by a modified
sulfinatodehalogenation system (Scheme S2 in the Support-
ing Information). 1,1,1-Trifluoro-2-iodoethane was coupled
with 1-hexene in acetonitrile/water (1:1) in the presence of
sodium dithionite and potassium carbonate at 458C for
8.0 h.[17] We then heated 1,1,1-trifluoro-4-iodooctane (com-
pound 15 in the Supporting Information) at 808C for 8.0 h
in the presence of Zn metal[18] and acetic acid to remove the
iodide atom in the molecule to give 7 in 73% yield. The
13C NMR spectrum revealed signals at d=127.33 (q, J=
274.0 Hz) and 33.75 ppm (q, J=28.0 Hz), consistent with the
replacement of one of the methyl groups with a CF3
group.[19] The observation of a signal at m/z 168 in the EI-
MS spectrum further confirmed the successful synthesis of
compound 7.
Scheme 2. The enzymatic conversions of substrates 5–7 mediated by P450
BM-3_A74G F87V L188Q. More details are given in Scheme S4 in the
Supporting Information.
The integration ratio of 1:1:2 of the absorptions at d=4.41,
1.75–1.56, and 1.40–1.33 ppm for 6 indicated that this eight-
carbon molecule possessed C2 symmetry with one fluorine
atom substituted at each end of the octane molecule. The
corresponding 13C NMR spectrum provided auxiliary evi-
dence with only four absorptions at d=84.10, 30.33, 29.06,
and 25.04 ppm.
Whole-cell conversion of the fluorinated substrates 1–7
mediated by cytochrome P450 BM-3_A74G F87V L188Q:
The gene of cytochrome P450 BM-3_A74G F87V L188Q
(3mt protein) fused with one C-terminal His6 tag was con-
structed and heterologously overexpressed in E. coli BL21
(DE3). This protein was “self-sufficient”[1] either in whole
cells or as the purified homogeneous enzyme with the sup-
plement of NADPH. Control experiments carried out in
whole cells without the induction of protein expression re-
vealed no activation of aliphatics. There was also no evi-
dence of interference by an alcohol dehydrogenase in
whole-cell catalysis.
Indeed, the overexpressed 3mt proteins were competent
in catalyzing the oxidation of fluorinated compounds 1–7
under ambient conditions. The chemoenzymatic conversion
of the various fluorinated octanes was first carried out in
whole-cell experiments because this approach quickly gener-
The 13C NMR spectrum provided additional support that
we had successfully synthesized our target fluorinated oc-
tanes 1–4 (see the Supporting Information). There were
eight independent signals in the 1H spin-decoupled
13C NMR spectrum, which indicated that these were
straight-chain octane derivatives. The downfield triplet sig-
nals appearing at d=117.50, 124.40, 125.83, and 125.37 ppm
with a 1:2:1 splitting pattern with extremely large splitting
constants (236.0–8.0 Hz) indicated two fluorine atoms were
substituted at the C-1, C-2, C-3, and C-4 positions of the
13
octane molecule in each case. The two-bond C C 19F
À À
spin–spin couplings between the geminal carbon and the flu-
orine atoms of the fluorinated carbon were also identified
for 1–4 and the splitting constants were in the range of 21.0–
28.0 Hz.
4776
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

À
C H Activation in n-Octanes
FULL PAPER
1
ated sufficient quantities of products to facilitate separation
and structural characterization. In all cases, the products
consisted of a mixture of secondary alcohols (Schemes 1 and
2, and Schemes S3 and S4 in the Supporting Information).
The position of hydroxylation of the fluorinated octane was
established in each case by EI-MS through identification of
C-6 and C-5 carbon atoms, respectively. The H absorption
of these protons appeared as triplets at d=0.98, 0.98, and
0.94 ppm, with weak coupling constants of J=7.5, 7.5, and
7.2 Hz, for 3a, 3b and 4a, respectively.
Other features in the 1H NMR spectra showed no interfer-
ence from the fluorine atoms. Upon the formation of octan-
2-ol derivatives 1–5a, the protons of one of the terminal
CH3 groups in each secondary alcohol were coupled with
À
the molecular fragments derived from cleavage of the C C
bonds adjacent to the oxygen of the alcohol. The structures
1
1
of the products were further characterized by H, 13C, and
the CH(OH) group proton to give a doublet in the H NMR
19F NMR spectroscopy; GC; and GC/MSD.
spectra at d=1.16–1.28 ppm (1a: d=1.16 ppm; 2a: d=
1.17 ppm; 3a: d=1.17 ppm; 4a: d=1.21 ppm; 5a: d=
1.28 ppm) with coupling constants ranging from J=4.0 to
6.3 Hz (1a: J=6.0 Hz; 2a: J=4.0 Hz; 3a: J=6.3 Hz; 4a: J=
6.0 Hz; 5a: J=6.0 Hz). For each of the product alcohols 1a–
d, 2a–c, 3a, 3b, 4a, 5a, 6b, and 6c, there was one new
downfield absorption feature in the 13C NMR spectra ap-
pearing in the range from d=67.50 to 72.95 ppm (1a: d=
68.0 ppm; 1b: d=72.9 ppm; 1c: d=71.2 ppm; 1d: d=
71.1 ppm; 2a: d=67.9 ppm; 2b: d=73.0 ppm; 2c: d=
71.0 ppm; 3a: d=67.8 ppm; 3b: d=72.7 ppm; 4a: d=
67.5 ppm; 5a: d=68.1 ppm; 6b: d=68.3 ppm; 6c: d=
71.2 ppm), which was readily assigned to the carbon of the
CH(OH) group, providing strong evidence for the formation
of secondary alcohol by the enzymatic conversion.
The products 1a–d, 2a–c, 3a, 3b, 4a, 5a, 6b, and 6c all
gave eight distinct signals in the 1H spin-decoupled
13C NMR spectra, indicating that they were straight-chain
octanol derivatives. The downfield 1:2:1 triplet appearing at
d=117.3–4, (1a: d=117.4 ppm; 1b: d=117.3 ppm; 1c: d=
117.3 ppm; 1d: d=117.4 ppm), 124.3–4 (2a: d=124.3 ppm;
2b: d=124.3 ppm; 2c: d=124.4 ppm), 125.4–6 (3a: d=
125.4 ppm; 3b, d=125.6 ppm), and 125.2 ppm (4a), with ex-
tremely large splitting constants of J=235.0–239.0 Hz (1a:
J=238.0 Hz; 1b: J=237.0 Hz; 1c: J=237.0 Hz; 1d: J=
237.0 Hz; 2a: J=236.0 Hz; 2b: J=237.0 Hz; 2c: J=
235.0 Hz; 3a: J=239.0 Hz; 3b: J=239.0 Hz; 4a: J=
238.0 Hz) (Scheme 1 and Scheme S3 in the Supporting Infor-
mation) for compounds 1a–d, 2a–c, 3a, 3b, and 4a, indicat-
ing that there were two fluorine atoms substituted at the C-
1, C-2, C-3, and C-4 positions of the secondary octanol de-
rivatives, respectively. On the other hand, 5a, 6b, and 6c all
displayed a strong 1:1 splitting pattern with coupling con-
stants of J=163.0–4.0 Hz (5a: J=164.0 Hz; 6b and 6c: both
with J=163.0 Hz) (Scheme 2 and Scheme S4 in the Support-
ing Information). There was only one such pattern for 5a
appearing at d=84.13 ppm. However, both 6b and 6c
showed two similar pairs of doublets at d=83.96 and
81.66 ppm and d=84.14 and 83.94 ppm, respectively. These
observations were consistent with one terminal CH2F in 5a
and two in 6b and 6c, which corresponded to the CH2F
groups at the C-1 and C-8 positions of the octanol deriva-
tives, respectively.
When 1 was used as the substrate, we observed mass frag-
ments from the products 1a–d at m/z 151, 137, 123, and 109,
corresponding to CHF2ACTHNUTRGNEUNG
(CH2)nOH+ (n=3–6), respectively,
indicating hydroxylation of 1 at the positions w-1 to w-4.
Applying the same strategy enabled us to identify the sites
of oxidation of 2, 3, and 4 to give the products 2a–2c, 3a,
3b, and 4a (Scheme 1).
When monofluorinated octane 5 (purchased from Sigma–
Aldrich) was used as the substrate for 3mt catalysis, EI-MS
analysis of the product alcohols revealed signals at m/z 133,
119, 105, and 91, which indicated hydroxylation at w-1 to w-
À
4, respectively, through C H activation of 5a–5d
(Scheme 2). For 6, the major products were 6b and 6c, as
evidenced by signals at m/z 119 and 105, respectively, in a
ratio of about 1:1 corresponding to w-2 and w-3 activation.
There was only very minor w-1 activation observed in this
case.
In the 1H NMR spectra, the fluorine atoms associated
with the fluorinated substituent in the fluorinated octanols
should be strongly spin-coupled to their neighboring protons
two and three bonds away and give rise to large 19F/1H spin–
spin splitting. The enzymatically converted fluorinated prod-
ucts 1a–1d had a terminal CHF2 group adjacent to a CH2
1
group in each case. These H NMR signals appeared at d=
5.77–5.84 ppm (1a: d=5.77 ppm; 1b: d=5.77 ppm; 1c: d=
5.78 ppm; 1d: d=5.84 ppm) as a superposition of two sets
of triplets with coupling constants of 56.8–57.2 and 4.0–
4.4 Hz, (1a: J=57.2 and 4.0 Hz; 1b: J=57.2 and 4.0 Hz; 1c:
J=56.8 and 4.4 Hz; 1d: J=56.8 and 4.4 Hz), respectively.
The CH3 groups in products 2a–2c were coupled to two vici-
1
nal 19F nuclei in each case, and the H NMR absorption ap-
pears as a triplet at d=1.55–1.58 ppm (2a: d=1.55 ppm;
2b: d=1.58 ppm; 2c: d=1.55 ppm) (Scheme 1 and
Scheme S3 in the Supporting Information). For product 5a,
1
the H NMR spectrum showed a doublet of triplets at d=
4.42 ppm. The two coupling constants, J=45.0 and 6.0 Hz,
corresponded to the coupling of the proton of the CH2F
group to the terminal fluorine atoms and to an adjacent
1
CH2 group. Similar signals were observed in the H NMR
spectra of these products, at d=4.39 (dt, 2H, J=47.1 and
6.0 Hz) and 4.71–4.05 ppm (m, 2H) for 6b, and at d=4.42
(dt, 2H, J=47.2 and 6.0 Hz) and 4.45 ppm (dt, 2H, J=47.1
and 5.6 Hz) for 6c. These spectral features arose from b-
and g-hydroxylation relative to the terminal fluorine sub-
stituents in 6b and 6c. In products 3a, 3b, and 4a, one of
the CH3 groups exhibited coupling adjacent to a CH2 group
that is proximal to the two fluorine atoms substituted at the
We noted that the two fluorine atoms of CF2 in 1d, 2c,
3b, and 4a became diastereotopic when the chiral hydroxyl-
ation occurred at a carbon within three C C bonds. There-
À
fore, for products 1d, 2c, 3b, and 4a, we observed two sig-
nals in the 19F NMR spectra with equal intensities at d=
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4777

S. S.-F. Yu et al.
À115.88 and À116.14 ppm, d=À90.92 and À91.28 ppm, d=
À100.50 and À100.61 ppm, and d=À98.17 and À98.25 ppm,
respectively, due to the appearance of two inequivalent 19F
nuclei.
Finally, it was difficult to separate products 5b–5d by
gravity column chromatography. Only minor amounts of 6a
were formed. The product yields of 7 were also lower than
those obtained for the other substrates. Therefore, we could
only identify these products through GC and GC/MSD at
the present time.
P450 BM-3 3mt A328F mutant mediated regioselective hy-
droxylation of the fluorinated n-octanes at the C-2 position:
In the case of the 3mt A328F mutant, we found that the
major product of n-octane activation was 2-octanol as pre-
dicted (94% abundance, Table 1).[9b,20] Unsurprisingly, the
Scheme 3. Fluorinated alcohols 5a and 5b were further derivatized by
(R)-O-acetylmandelic acid.
proximal phenyl substituent
acted as a chiral resolving re-
Table 1. The throughputs of secondary alcohol formation were obtained from ratios of the hydroxylated prod-
ucts produced by purified P450 BM-3 3mt (GVQ) and 3mt F328 (GVQF) by using n-octane and 1–7 as the
substrates.
agent and exerted an upfield
shift of the corresponding ter-
minal methyl group protons in
P450 BM-3 3mt (GVQ)
3-ol [%] 4-ol [%]
(ee [%])[a]
P450 BM-3 3mt F328 (GVQF)
Substrates
2-ol [%]
(ee [%])[a]
5-ol [%]
2-ol [%]
3-ol [%]
4-ol [%] the 1H NMR spectra for the
G
N
ACHTUNGTRENNUNG
product in the
S configura-
tion.[10,21] In this manner, we
could determine the ee ratios.
Commercially available (S)-
and (R)-octan-2-ol (Sigma–Al-
drich) were employed as au-
thentic standards of the (R)-O-
acetylmandelic acid derivatives
(see the Supporting Informa-
n-octane
28 (20)
43 (39)
42 (65)
48 (70, 80[b]
78 (79, 80[b]
n.d.[c]
43 (61)
46
32
29
16
30
n.d.[c]
10
94 (43)
6.0
2.9
2.6
n.d.[c]
n.d.[c]
n.d.[c]
n.d.[c]
n.d.[c]
n.d.[c]
n.d.[c]
7.8
1
2
3
4
5
6
7
32 (60, 56[b]
22 (56, 56[b]
22 (64, 67[b]
)
)
)
97 (38, 33[b]
97 (70, 67[b]
)
)
)
)
n.d.[c]
n.d.[c]
n.d.[c]
14
n.d.[c]
n.d.[c]
20
96 (43,43[b]
)
4.0
100 (52, 56[b]
)
100 (70, 67[b]
97 (31, 33)
n.d.
)
n.d.[c]
2.6
23 (58,60[b]
)
1.9
51
52
17
n.d.[c]
n.d.[c]
n.d.[c]
4.1
88
[a] The ee (ee=
(RRÀSR)/
ACHTUNGTRNE(NUNG RR+SR)) ratios of the diastereomeric (R)-O-acetylmandelic acid derivatives were
determined by the corresponding signal areas appearing in GC/MSD chromatograms. [b] The ee ratios were
determined by the ratios of the integrated areas of the stereoselective shifted methyl resonances in the
1H NMR spectra for the (R)-O-acetylmandelic acid derivatives derived from (R)- and (S)-1–5a’, 1–3, and 5b’.
[c] n.d.=not detected.
1
tion). As shown in the H NMR
spectrum, the absorption of the
terminal methyl for the (S,R)-
octan-2-ol ester occurred at d=
1.05 ppm, which was d=
major products for the fluorinated substrates 1–7 were also
activated at subterminal carbon atoms, resulting in the for-
mation of fluorinated octan-2-ol derivatives 1a–5a and 7a.
However, there was no detectable product derived from 6.
Other than this, the reactions were actually well controlled
to give the w-1 alcohol distal to the fluorine substituent.
0.16 ppm upfield of the parent (R,R)-octan-2-ol ester (d=
1.21 ppm). Similar patterns were observed from the product
alcohols derived from substrates 1–5 mediated by both P450
BM-3 variants (Table S1 in the Supporting Information). In
the w-1 products, the methyl group of the fluorinated (S,R)-
octan-2-ol esters derived from (S)-1–5a’ were also shifted
upfield d=0.12–0.18 ppm relative to that of the (R,R)-deriv-
atives derived from (R)-1–5a’. The stereochemical results
were verified by GC by comparison with the retention times
of the product octanols with authentic standards.
The five fluorinated octanes 1–5 all showed higher stereo-
selectivity toward the R configuration with comparable
ratios for their w-1 products (56–67% ee by NMR spectros-
copy and 52–64% ee by GC) when the conversion was car-
ried out by the 3mt protein (Table 1). However, in the case
of the 3mt F328 protein, the ee ratios of the w-1 products
obtained for substrates 2 and 4 increased significantly to 67–
70% ee according to both GC and NMR spectroscopy. In
contrast, the corresponding ee ratios decreased to 31–
43% ee for the C-1 fluorinated 1 and 5 as well as the C-3
fluorinated 3. For comparison, the ee ratios of the product
Stereoselectivity of the subterminal activation of the fluori-
nated n-octanes mediated by P450 BM-3 3mt and 3mt
A328F: In principle, the surface roughness of the hydropho-
bic pocket could be modified upon the introduction of an ar-
omatic residue at amino acid 328 and changing the ee ratios
of the fluorinated w-l products formed by activation of the
fluorinated octanes by the 3mt enzyme and the A328F
mutant. To compare the stereoselectivity of hydroxylation of
3mt and 3mt-F328, we esterified the converted 1a–5a into
the diatereomeric (R)-O-acetylmandelic acid derivatives
(Schemes 1 and 3). The absolute configurations at C-2 for
each of the octan-2-ol derivatives (R)- and (S)-1–5a’ were
then determined from the chemical shift of the C-1 methyl
terminus. The anisotropic magnetic susceptibility of the
4778
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

À
C H Activation in n-Octanes
FULL PAPER
(R)-octan-2-ols and (S)-octan-2-ols derived from n-octane
mediated by 3-mt and 3-mt A328F proteins were 20 and
43% ee, respectively, according to GC by comparison with
the retention times of authentic standards. It was not possi-
ble to ascertain the stereoselective ratios by NMR spectros-
copy here due to the relatively poor yields of the corre-
sponding conversions when using n-octane. From these re-
sults, it is evident that the influence of Phe328 on the stereo-
chemical outcome of the w-1 hydroxylation of the various
fluorinated octanes mediated by the A328F 3mt protein is
manifested only every other CF2 substituent along the hy-
drocarbon chain in an alternating fashion, beginning with C-
2.
We have also compared the stereoselectivity of the hy-
droxylation of C-3 in octane and the w-2 hydroxylation of
the fluorinated octanes mediated by the BM-3 3mt enzyme.
(Table 1 and Table S1 in the Supporting Information). To ac-
complish this, we examined the diastereomeric (R)-O-acetyl-
mandelic acid derivatives of (R)- and (S)-octan-3-ols and
their corresponding fluorinated alcohols by NMR spectros-
copy. Derivatization of (S)-octan-3-ol (Sigma–Aldrich) with
(R)-O-acetylmandelic acid gave
we concluded that the octan-3-ol product ee ratio was 39%
in favor of the R configuration in the case of n-octane
(Table 1). Substrates of the fluorinated octanes 1–3 and 5
gave even better enantiomeric selectivity toward the R con-
figuration with ee ratios in the range 61–80%.
Dramatic high coupling efficiencies between electron trans-
fer and substrate activation in the P450 BM-3 variants for
the fluorinated substrates: To analyze the specific activities
of the two P450 variants quantitatively, we purified the two
proteins to homogeneity by using a HisTrap HP column
(GE Healthcare). The fluorinated octanes 1–7 were subject-
ed to the purified enzyme preparations for 1.0 min conver-
sions at room temperature. For each substrate, the through-
puts of the various product secondary alcohols are summar-
ized in Table 1 for 3mt and the A328F variant. The corre-
sponding specific activities of the purified proteins are ex-
pressed in terms of the rate of NADPH consumption and
the rate of product formation (molecules of NADPH or
molecules of product alcohol formed per enzyme per min)
and are given in Table 2. In case of the 3mt, the rates of
two 1H NMR spectroscopy sig-
nals for the terminal methyl
groups at d=0.56 and
Table 2. Rates of NADPH consumption, product formation, and the corresponding coupling efficiencies were
measured for P450 BM-3 3mt (GVQ) and 3mt F328 (GVQF) by using n-octane and 1–7 as the substrates.
[a,b]
[b]
Substrates
NADPH consumption [minÀ1
]
Product formation [minÀ1
3mt 3mt F328
]
CE ratio[c] [%]
0.85 ppm, respectively (Sup-
porting Information). From the
corresponding 1H NMR spec-
troscopy signals obtained from
the derivatives of (Æ)-octan-3-
ol (the racemic mixtures of (R)-
and (S)-octan-3-ols), the termi-
nal methyl groups of the (R)-
octan-3-ol derivative were ob-
3mt
3mt F328
3mt
9.3
98
3mt F328
n-octane
1608Æ80
1809Æ171
1809Æ57
1326Æ57
1795Æ46
2010Æ113
2010Æ80
1085Æ57
1034Æ52
744Æ51
150Æ14
215Æ14
610Æ34
491Æ32
446Æ19
1197Æ104
1647Æ150
n.d.
21
82
75
70
68
82
n.d.
51
1
2
3
4
5
6
7
1765Æ89
1617Æ150
993Æ64
653Æ24
89
75
72
88
38
41
619Æ28
1759Æ107
2012Æ128
1085Æ106
857Æ43
1289Æ68
1758Æ149
769Æ14
442Æ27
433Æ13
[a] Protein concentrations were quantified by CO-binding difference spectra of the reduced heme by using an
served at d=0.74 and 0.87 ppm. extinction coefficient of 91 mmÀ1 cmÀ1 for the 450 minus 490 nm signals.[9b] [b] The background NADPH con-
sumption rate (without substrate) was (266Æ2.4) minÀ1. [c] CE=coupling efficiency.
From the chemical shifts of the
terminal methyl groups adja-
cent to gem-difluorination at C-
7 of (R)- and (S)-octan-3-ol 2b’ appearing at d=1.38 ppm
with a coupling constant of J=21 Hz, we could identify the
corresponding 1H NMR absorption of the other methyl
groups proximal to the hydroxylation positions for (R)- and
(S)-2b’ at d=0.88 and 0.57 ppm, respectively; a 0.31 ppm
upfield shift for (S)-3-ol 2b’ relative to enantiomer (R)-2b’
(Table S1 in the Supporting Information). Comparable ste-
reoselective anisotropic shift patterns were observed for the
parent authentic standards octan-3-ol (RÀS: d=0.31 ppm,
see the Supporting Information) and 3b’ (RÀS: d=
0.30 ppm).
The (R,R)- and (S,R)-octan-3-ol derivatives could also be
separated by GC and the retention times were 89 and
90 min, respectively (see the Supporting Information). In
the GC chromatograms, the order of appearance based on
the retention times was (R)-1b’, -2b’, and -3b’ (tR =192, 164,
and 148 min, respectively) always ahead of (S)-1b’, -2b’, and
-3b’ (tR =195, 166, and 149 min, respectively; Table S1 in the
Supporting Information). On the basis of these observations,
NADPH consumption (160 nm protein, 320 mm NADPH,
13 mg bovine liver catalase, and 4.0 mm substrates)[9b]
ranged from 1085–2010 minÀ1 for substrates 1–7, similar to
the consumption rate previously reported for n-octane
(1760 minÀ1)[22] and repeated in the present study ((1608Æ
80) minÀ1) as a control. In contrast, the rates of product for-
mation from substrates 1–7 (160 nm protein, 320 mm
NADPH, 13 mg bovine liver catalase, and 4.0 mm substra-
tes),[9b] which ranged from 442 to 1765 minÀ1, were signifi-
cantly higher than the corresponding rate for n-octane
(150 minÀ1). Thus, the incorporation of a fluorine substituent
has greatly improved the coupling of electron transfer to
compound conversion in the 3mt protein. The coupling effi-
ciency (CE ratio) ranged from 38 to 98% in the case of the
fluorinated octanes compared with a mere 9.3% for n-
octane.
Further insights into the influence of the fluorine substitu-
À
ents on the C H activation of n-octane were provided by
the data obtained on the 3mt A328F mutant. Here, except
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4779

S. S.-F. Yu et al.
for substrates 4 and 5, the NADPH consumption rates de-
creased by a factor of about 1/3 to 4/5 relative to the 3mt.
However, there was a corresponding drop in the rate of
product formation, so that the CE ratios were not dramati-
cally affected (only a change of ꢀ10% relative to the 3mt
protein). In the case of substrates 4 and 5, the NADPH con-
sumption and product formation rates were, in fact, compa-
rable so that the CE ratios were essentially the same for the
two enzymes. Thus, although these substrates exhibited the
same regioselectivity as n-octane in favor of formation of
the 2-ol, the product throughputs and the corresponding
coupling efficiencies were substantially different.
Discussion
The activation of the P450 enzyme for substrate oxidation
requires the initial expulsion of water from within the hy-
drophobic pocket. This process, which is accompanied by
the binding of a substrate molecule, switches the heme iron
from six to five coordinate, and raises the redox potential of
the iron to facilitate the transfer of an electron from
NADPH to reduce the heme iron for subsequent dioxygen
activation.[7b,23] Clearly, the sequestering of the substrate
into the heme pocket requires cooperation of the iron por-
phyrin with the residues lining the substrate channel
(Figure 1).
There have been attempts to discern the location and spa-
tial orientations of the substrates in the crystal structures of
P450 BM-3 to delineate those features of the hydrophobic
pocket that might control the regio- and stereoselectivity of
À
the C H activation. The bound substrates have always been
Figure 1. Three-dimensional models (PDB code 1JPZ, modified by Dis-
covery Studio 2.0 (Accelerys)) show the encapsulation of substrate 4
within the pocket of P450 BM-3 with F328 to generate the proposed
CF2···p interaction and the heme iron mediating the regiospecific w-1 oxi-
dation.
encapsulated at a site within a pocket some 6–8 ꢁ above
(distal) the iron porphyrin. Accordingly, these structural re-
sults have not provided the desired insight into how interac-
tions between the substrate and the iron porphyrin might
contribute to the control of the oxidation.[20,24]
Evidence for the existence of a distinct pocket for the
binding of the substrate not associated with the polycyclic
porphyrin system has mainly come from X-ray crystallo-
graphic studies. In the cytochrome P450 (BM-3, 3mt, and
3mt F328) proteins, this site is formed by the amino acids
Ser72, Ala74 (Gly74 for 3mt), Leu75, Val78, Ala82, Phe87
(Val87 for 3mt), Thr88, Leu181, Ile259, Thr260, Ile263,
Ala264, Thr268, Ala328 (Phe328 for 3mt F328), Ala330,
Thr348, Met354, Leu437, and Thr438. The natural substrates
of the wild-type enzyme are C12 and C14–C16 fatty acids, and
these fatty acids are typically hydroxylated at the w-1, w-2,
and w-3 positions with stereoselectivities of 86–99, 86–96,
and 44–74% ee, respectively, toward the formation of the
secondary alcohol in the R configuration.[24–25] Two wild-type
structures, 1JPZ[23] and 1FAG,[26] are shown in Figure 2. In
(Figure 2). The remainder of the secondary substrate pocket
can be divided into three zones: large (L), medium (M), and
small (S). Most likely, the S zone is located at the small cleft
around the positions F87 (wild type) or V87 (3mt). When
octane is used as the substrate in BM-3 3mt, the principal
product is the secondary alcohol at C-3 (Figure 3). Thus, for
this small substrate, we expect the aliphatic chain to have a
principal binding site in the M and L zones, with the ethyl
group in the M zone and the pentyl group in the L zone.
However, there should be a distribution of binding sites, cor-
responding to moving the aliphatic chain into other sites
within these two zones. Given the location and size of the M
zone from the crystal structures, the maximum hydrocarbon
chain length for this cavity would be n-propyl or n-butyl. If
the hydroxylation of the hydrocarbon occurs with the sub-
strate more or less in place, then the regiospecificity will be
determined by the distribution of binding sites in the M and
L zones. This distribution is controlled by the energetics as-
sociated with the substrates occupying different binding
À
these structures, the shortest distance of the C H bond to
the heme iron occurs at the w-3 carbon (6–8 ꢁ). For the pur-
pose of this discussion, we can set the position of this
carbon in N-palmitoylglycine or palmitoleic acid as the ste-
reocenter for the recognition of the aliphatic chain
4780
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

À
C H Activation in n-Octanes
FULL PAPER
Figure 2. The hydrophobic pocket of P450 BM-3 for substrates N-palmi-
toylglycine or palmitoleic acid, respectively, reproduced from PDB codes
1JPZ (a) and 1FAG (b) by Discovery Studio 2.0 (Accelerys). L and M
refers to the large and medium zones, respectively.
Figure 3. Three-dimensional models (PDB code 1JPZ, modified by Dis-
covery Studio 2.0) show the encapsulation of substrate 4 within the
pocket of the P450 BM-3 3mt (A74G F87V L188Q) protein for a) the
electrostatic surface of 4 and b) the conformer for regiospecific oxidation
À
at w-2. The a and/or b C H bonds from the C-4 position with the fluo-
rine substituent are not susceptible to oxidation by the high-valent iron
heme center.
sites. This analysis provides a reasonable scenario for hy-
droxylation at C-2, C-3, and C-4 of octane in the case of the
BM-3 3mt enzyme, and hydroxylation at w-1, w-2, w-3, and
w-4 with 1 and 5 (Table 1 and Schemes 1 and 2). For the re-
maining fluorinated octanes, when the fluorine substituent is
moved to internal carbon atoms, as in the case of substrates
2, 3, and 4, the distribution of products is now restricted by
the location of the fluorine substitution, and in the case of 4,
hydroxylation is totally regiospecific to yield only the 2-ol
(Figure 3). When the substrate is switched to 6, oxidation at
C-2 becomes negligible. It is interesting to note that the a
and/or b positions to the fluorinated carbon atoms within
the octane derivatives 1–7 are not susceptible to oxidation
by 3mt protein (see throughput data in Table 1, Schemes 1
and 2, and Schemes S3 and S4 in the Supporting Informa-
tion).
the corresponding substrate will be preferably oxidized to
give the alcohol in the R configuration. However, if oxida-
tion of the C HR bond involves a mechanism requiring
closer interaction of the C H bond with the activated Fe =
O, then binding of the substrate to the hydrophobic pocket
must be eventually followed by a conformational change to
allow hydrogen abstraction/radical rebound or direct O-
atom insertion to form the product.[7b,27] To accomplish this,
according to the crystal structures of BM-3 (Figure 2), the
residues Phe87 (Val87), Ala264, Thr268, and Ala328 must
À
IV
À
À
move to bring the C HR bond in closer proximity to the
porphyrin with the activated oxyferryl species. If there is
direct interaction between some parts of the substrate with
the porphyrin system, we surmise that this conformational
À
In the above model, as long as the C HR bonds are pre-
sented directly to the activated high-valent iron–oxo species,
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4781

S. S.-F. Yu et al.
switching will become more likely or facile. In any case, the
stereoselectivity will reflect the orientation of the hydrocar-
bon in the binding pocket before conformational switching
for product formation, with the enantiomeric excess deter-
mined by the free energy of orientating the substrate in the
binding site. This is the kind of scenario that has been pro-
the substrate is presumably optimized for w-1 activation by
the high-valent Fe=O species of the porphyrin. Such a spe-
cific interaction would, of course, limit the activation of sub-
strates in certain positions, such as the w-2 and w-3 posi-
tions.[29]
Apparently, the mutations do not affect the protein fold-
ing dramatically or the catalytic cycle, that is, compound I is
still formed and electron and proton transfers during the
catalytic cycle are not affected. The dramatic difference in
the product throughputs and coupling efficiencies observed
between the fluorinated octanes and n-octane in the case of
3mt, as well as the A328F variant, suggests that the fluori-
nated substituent is tuning the distribution of accessible
enzyme conformations and associated protein dynamics that
activate the iron porphyrin for substrate hydroxylation to
allow reactions mediated by the high-valent Fe=O to
become kinetically more commensurate with electron trans-
fer from the FAD/FMN reductase. In the case of the 3mt
protein, given that the enhanced coupling efficiency tran-
scends the entire set of fluorinated octanes studied, irrespec-
tive of the location of the substituent, we surmise that there
À
posed for direct oxene insertion into the C H bonds of ali-
phatics by the particulate methane monooxy
ACHTUNGTRNEgNUNG enase
(pMMO).[10,13d,21,28]
In the case of the 3mt F328 enzyme, we have observed
mainly w-1 selectivity. To account for this behavior, we pro-
pose a specific CF2 interaction with the Phe p system at resi-
due 328, which must be stronger than Ala at this position in
the BM-3 3mt enzyme to limit the distribution of the sub-
strates to one major binding site. Molecular modeling re-
veals a possible conformer of substrate 4 that would account
for the high regiospecific activation at w-1 in the 3mt F328
(Figure 4). In this almost fully extended all-trans conformer,
À
must be a similar interaction involving the CF2 group or C
F bond with the porphyrin p system (26 p electrons) that is
altering the energetics and dynamics of the P450. Such an
interaction would tighten up the domain of the protein con-
taining the active site.[30] Strong evidence for a similar
CF2···p interaction has also been demonstrated in a recent
study of the binding of tumor promoters to peptide ana-
logues of the C1B domain of the protein kinase C isozyme
PKCd, in which the replacement of Pro11 by 4,4-difluoro-
proline was found to exert a strong promoting effect on the
binding of the teleocidins indolelactam V and naphtholac-
tam-V8 (aromatic rings with 10 p electrons) relative to the
benzolactam-V8 (6 p electrons) to the model peptide.[31]
It is also proposed that there is a van der Waals interac-
tion between CF2 and the Phe p system at residue 328 when
4,4-difluorooctane is used as a substrate with BM-3 3mt
A328F (Figure 1). This interaction involves polarization of
À
the Phe p cloud by the C F dipole moment. The polarizabil-
[32]
À
ity of the C F bond is quite low (molecular polarizabili-
ties of ethane, 1-fluoroethane, and 1,1,1-trifluoroethane are
4.47, 4.96, and 4.4ꢂ10À24 cm3, respectively), but the dipole
À
moment of the C F bond is large (molecular dipole mo-
ments of CH3F, CH2F2, and CHF3 are 1.86, 1.98, and 1.65 D,
[33]
À
respectively; the C F bond moment is 1.4 D). The molec-
ular polarizability of Phe is estimated to be 10–15ꢂ
10À24 cm3. For comparison, the electric dipole polarizability
of a methyl group is about 2.5ꢂ10À24 cm3.
Conclusion
In this study, we have shown that it is possible to tune the
reactivity, regioselectivity, and stereoselectivity of CACHTUNGTRENNUNGH acti-
vation of n-octanes by cytochrome P450 BM-3 with fluorine
substituents. Evidence was obtained for van der Waals inter-
actions between a CF2 group with aromatic p systems, which
Figure 4. Three-dimensional model (PDB code 1JPZ, modified by Dis-
covery Studio 2.0) show the encapsulation of substrate 4 within the
pocket of the P450 BM-3 3mt F328 (A74G F87V L188Q A328F) protein
for a) the electrostatic surface of 4 and b) the conformer for regiospecific
oxidation at w-1.
4782
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

À
C H Activation in n-Octanes
FULL PAPER
Phe are underlined. The two variants were verified by DNA sequencing
and transformed into E. coli BL21 (DE3) for induction and reactivity
measurements.
could be exploited to tune the details of the binding of the
substrate to the binding pocket, as well as the protein con-
formational states and protein dynamics that participate in
Expression of the recombinant P450 BM-3 3mt (A74G F87V L188Q)
and P450 BM-3 3mt F328 (A74G F87V L188Q A328F): Both of the vec-
tors pET21_BM-3_3mt and pET21_BM-3_3mt_F328 were transformed
into E. coli BL21 (DE3). The corresponding variants were cultured in
Luria-Bertani (LB) broth (100 mL) containing ampicillin (100 mgLÀ1) at
378C. When the optical density of the cell culture reached 0.6–0.8
(OD595 =0.6–0.8), expression of the recombinant proteins was induced
with 0.35 mm isopropyl-b-d-thiogalactoside (IPTG) at 378C for 6.0 h.
After centrifugation at 6000 rpm, the cells were resuspended in 50 mm
Tris-HCl buffer (pH 7.5) for further reactivity measurements or protein
characterization. After cell lysis, we examined the expression of P450
BM-3 protein variants by 10% sodium dodecyl sulfate polyacrylamide
gel electrophoresis (SDS-PAGE). The P450 BM-3 protein fused with the
reductase and 6-His tags appeared on the SDS-PAGE gel at a molecular
weight of about 119 kDa.
À
electron transfer and C H activation.
Experimental Section
General: All chemicals were purchased from commercial sources as re-
agent-grade quality and were used as received unless stated otherwise.
Analytical TLC was performed on precoated plates (silica gel 60 F-254)
purchased from Merck. Visualization of spots on TLC was accomplished
by use of KMnO4 (aq). Mixtures of ethyl acetate and hexane were used
as eluents. Purification by gravity column chromatography was carried
out with EM Reagents Silica Gel 60 (particle size 0.063–0.200 mm, 70–
230 mesh ASTM). NMR spectra were recorded on Bruker AVA-300
(300 MHz), AV-400 (400 MHz), or AV-800 (800 MHz) spectrophotome-
ters. Chemical shifts (in ppm) were referenced either to the internal stan-
dard CFCl3 (19F NMR) or to the residual solvent peak (1H and
13C NMR). UV/Vis spectra were recorded on a HP 8453 diode array
spectrometer or Beckman Coulter DU-800 spectrometer. HR-ESI
(Waters LCT Premier XE) mass spectra were obtained with dual ioniza-
tion source options and the limitation was m/z <1500. Gas chromatogra-
phy was performed by using an Agilent HP6890 plus instrument either
equipped with a flame ionization detector (FID) or with a 5971 EI-MS
detector and separated by a HP-1 or HP-5 capillary column (60 m ꢂ
0.25 mm ꢂ 0.25 mm film thickness). The analysis conditions were as fol-
lows unless stated otherwise: carrier gas, nitrogen, 1.0–2.1 mLminÀ1; oven
temperature, 1008C; injection port, 2808C with a splitless ratio.
Purification of P450 BM-3 3mt and 3mt F328: Growth of E. coli BL21
(DE3; 600 mL; OD595 ꢀ0.6–0.8) containing vectors for the expression of
BM-3 variants in LB medium, supplied with ampicillin (100 mgLÀ1) and
0.5% glucose, were induced with 0.35 mm IPTG for 6.0 h. After centrifu-
gation at 8000 rpm (20 min, 48C), the cell pellet was resuspended in load-
ing buffer (0.1m, Tris-HCl, pH 8.2, 1 mm phenylmethanesulfonyl fluoride
(PMSF)) and lyzed by French press on ice (20000 psi, SLM-AMINCO
Instruments). After the lysate was subjected to ultracentrifugation at
36000 rpm for 1.0 h (48C), the supernatant was filtered through
a
0.22 mm filter, then reloaded onto a pre-equilibrated 1 mL HisTrap HP
column (GE Healthcare). The column was subsequently washed with
loading buffer, washing buffer A (50 mm NaH2PO4, 300 mm NaCl,
pH 8.0), and washing buffer B (100 mm imidazole, 50 mm NaH2PO4,
300 mm NaCl, pH 8.0). The purified protein was desalting with PD-10,
and finally, supplemented with 10% glycerol and stored at À808C. Con-
centrations of P450 enzymes were measured by CO difference spectros-
copy of the reduced heme by using an extinction coefficient of
91 mmÀ1 cmÀ1 for the 450 minus 490 nm signal.[9b,34]
Construction of expression vector pET21_BM-3: The growth of Bacillus
megaterium (BCRC10608 (ATCC14581), Bioresource and Collection Re-
search Center in Hsinchu, Taiwan) was carried out by following the
American Type Culture Collection culture conditions using ATCC
medium3 (50 mL) in a 250 mL flask. The genomic DNA was prepared by
using the Gene-Spin Genomic DNA kit (Protech, Taipei, Taiwan). The
four nucleic acid primers expression of the P450 BM-3 and associated re-
ductase genes (BM-1af: 5’-AATTCACAATTAAAGAAATGCCT-
CAGCCAA-3’; BM-1ar: 5’-GCCCAGCCCACACGTCTTTT-3’; BM-1bf:
5’-CACAATTAAAGAAATGCCTCAGCCAAAAA-3’, and BM-1br: 5’-
TCGAGCCCAGCCCACACG-3’) were synthesized by Protech or Tri-I
(Taipei, Taiwan). Together with PfuTurbo DNA polymerase (Stratagene),
the P450 BM-3 gene fused with FAD/FMN reductase was amplified from
genomic DNA by the polymerase chain reaction (PCR). After 5’-phos-
phorylation, denaturation, and renaturation of PCR products, the cohe-
sive end, P450_BM-3, including the reductase gene with 5’-EcoR I and
3’-Xho I sites was generated, and cloned into the corresponding restric-
tion sites of pET21a (Novagen) to produce pET21_BM-3.
Hydroxylation of the fluorinated octane derivatives by whole-cell cataly-
sis mediated by E. coli P450 BM-3 3mt and E. coli BM-3 3mt F328.: A
culture of the recombinant E. coli containing P450 BM-3 3mt and BM-3
3mt F328, including cofactors, were cultured in LB medium (100 mL) to
OD595 =0.6–0.8 and then induced with 0.35 mm IPTG at 378C for 6.0 h.
After centrifugation at 8000 rpm, cells were resuspended to a density cor-
responding to OD595 =15 with 50 mm LB buffer containing 0.5% glucose.
Substrates 1–7 (1.0 mL) were then added into reaction mixtures (1 mL),
and incubated at 378C at 170 rpm for 22 h. The reactions were then
quenched and the mixtures were extracted immediately for product anal-
ysis with dichloromethane (1.0 mL). The organic layers were separated
by centrifugation at 12000 rpm, and monitored by GC or GC/MSD. Fi-
nally, the alcohols obtained, 1a–d, 2a–c, 3a, 3b, 4a, 5a–d, 6a–c, and 7a–
c, were separated by gravity column chromatography (EtOAc in
hexane=20%) and the pure secondary alcohols were analyzed by 13C,
Site-directed mutagenesis of two variants of P450 BM-3 3mt
(A74G F87V L188Q) and P450 BM-3 3mt F328: The mutant P450 BM-3
3mt was introduced by using a Quick-change site-directed mutagenesis
kit (Stratagene) according to the manufacturerꢃs protocol. PCR amplifi-
cations were carried out with complementary mutagenesis primers se-
1
19F, and H NMR spectroscopy.
Compound 1a: Rf =0.510 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
1
2
3
tR =8.18 min; H NMR (400 MHz, CDCl3): d=5.77 (tt, J
ACHTUNGTREN(NUNG H,F)=57.2, J-
quentially
for
A74G_f,
5’-GCTTTGATAAAAACTTAAGT-
A
ACHTUNGTRENNUNG
CAAGGCCTTAAATTTGTACGTGA-3’ and A74G_r, 5’-TCACGTA-
CAAATTTAAGGCCTTGACTTAAGTTTTTATCAAAGC-3’, F87V_f,
5’-GCAGGAGACGGGTTAGTGACAAGCTGGACGCAT-3’
3
1.81–1.68 (m, 2H; CH2), 1.50–1.28 (m, 8H; CH2), 1.16 ppm (d, JACHTUNGTRENNUNG
6.0 Hz, 3H, CH3); 13C NMR (100 MHz, CDCl3): d=117.4 (t, 1J
ACHTUNGTRENNUNG
and
238.0 Hz; CHF2), 68.0 (s; CHOH), 39.0 (s; CH2), 34.0 (t, 2J
ACHTUNGTRENNUNG
F87V_r, 5’-ATGCGTCCAGCTTGTCACTAACCCGTCTCCTGC-3’,
21.0 Hz; CCHF2), 29.0 (s; CH2), 25.5 (s; CH2), 23.5 (s; CH3), 22.0 ppm (t,
L188Q_f, 5’-GAAGCAATGAACAAGCAGCAGCGAGCAAATCCA-
GACG-3’ and L188Q_r, 5’-CTGGATTTGCTCGCTGCTGCTTGTT-
CATTGCTTCATCC-3’ by using the parent plasmid pET21_BM-3 as a
template to create the pET21_BM-3_3mt for 3mt protein expression. The
vector pET21_BM-3_3mt_F328 for cloning of the mutant P450 BM-3
3mt F328 protein was constructed by A328F_f, 5’-CTGCGCTTATGGC-
CAACTTTTCCTGCGTTTTCCCTATATG-3’ and A328F_r, 5’-CATA-
ACTHNUTRGNEUNG
3J(C,F)=5.0 Hz; CCH2CHF2); 19F NMR (282 MHz, CDCl3): d=
À115.90 ppm; EI-MS: m/z (%): 151 (2) [MÀCH3] , 148 (1) [MÀH2O]+,
131 (0.5), 113 (2), 93 (0.3), 87 (2), 73 (2), 67 (5), 45 (100); HRMS (ESI)
[M+Na]+: m/z calcd for C8H16F2ONa: 189.1067; found: 189.1065.
Compound 1b: Rf =0.520 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
ACTHNUTRGNEUNG
tR =8.02 min; 1H NMR (400 MHz, CDCl3): d=5.77 (tt, 2J(H,F) 57.2, 3J-
TAGGGAAAACGCAGGAAAAGTTGGCCATAAGCGCAG-3’
using pET21_BM-3_3mt as the template. The mutated primer sequences,
with GGC encoding for Gly, GTG for Val, CAG for Gln, and TTT for
by
ACHTUNGTREN(NGNU H,H)=4.0 Hz, 1H; CHF2), 3.51 (m, 1H; CHOH), 1.78–1.63 (m, 2H;
CH2), 1.52–1.33 (m, 8H; CH2), 0.92 ppm (t, 3J
ACHTUNGTRENNUNG
13C NMR (100 MHz, CDCl3): d=117.3 (t, 1J
ACHTUNGTRENNUNG
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4783

S. S.-F. Yu et al.
72.9 (s; CHOH), 36.5 (s; CH2), 34.0 (t, 2J
CH2), 25.1 (s; CH2), 22.1 (t, 3J
N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Compound 3b: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =7.06 min; 1H NMR (300 MHz, CDCl3): d=3.56–3.48 (m, 1H;
CHOH), 1.97–1.62 (m, 4H; CH2), 1.60–1.36 (m, 4H; CH2), 0.98 (t, 3J-
Compound 1c: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
1
2
3
tR =7.78 min; H NMR (400 MHz, CDCl3): d=5.78 (tt, J
(H,H)=4.4 Hz, 1H; CHF2), 3.61 (m, 1H; CHOH), 1.91–1.79 (m, 2H;
CH2), 1.54–1.29 (m, 8H; CH2), 0.90 ppm (t, 3J
(H,H)=7.6 Hz, 3H; CH3);
13C NMR (100 MHz, CDCl3): d=117.3 (t,1J
(C,F)=237.0 Hz; CHF2), 71.2
(s; CHOH), 39.6 (s; CH2), 36.6 (s; CH2), 34.0 (t, 2J
(C,F)=21.0 Hz;
(C,F)=5.0 Hz; CCH2CHF2), 13.9 ppm (s;
ACHTUNGTREN(NUNG H,F)=56.8, J-
(H,H)=7.5 Hz, 3H; CH3), 0.92 ppm (t, 3J
13C NMR (75 MHz, CDCl3): d=125.6 (t, 1J
(s; CHOH), 32.0 (t, 2J(C,F)=26.0 Hz; CCF2), 30.3 (s; CH2), 29.7 (t, 2J-
(C,F)=26.0 Hz; CCF2), 29.3 (t, J
6.6 ppm (t, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
3
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
CCF2), 18.7 (s; CH2), 18.3 (t, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
CH3); 19F NMR (282 MHz, CDCl3): d=À115.90 ppm; EI-MS: m/z (%):
123 (16) [MÀC3H7]+, 103 (0.9), 83 (2), 73 (46), 55 (100), 43 (28); HRMS
(ESI) [M+Na]+: m/z calcd for C8H16F2O Na: 189.1067; found: 189.1065.
d=À100.50, À100.61 ppm; EI-MS: m/z (%): 137 (4) [MÀC2H5]+, 117
(64) [MÀC2H5ÀHF]+, 99 (12), 88 (46), 79 (11), 69 (51), 59 (100), 41 (36);
HRMS (ESI) [M+Na]+: m/z calcd for C8H16F2ONa: 189.1067; found:
189.1064.
Compound 1d: Rf =0.535 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
1
2
3
tR =7.64 min; H NMR (400 MHz, CDCl3): d=5.84 (tt, J
(H,H)=4.4 Hz, 1H; CHF2), 3.60 (m, 1H; CHOH), 1.99–1.90 (m, 2H;
CH2), 1.65–1.56 (m, 2H; CH2), 1.54–1.29 (m, 6H; CH2), 0.90 ppm (t, 3J-
(H,H)=7.6 Hz, 3H; CH3); 13C NMR (75 MHz, CDCl3): d=117.4 (t, 1J-
(C,F)=237.0 Hz; CHF2), 71.1 (s; CHOH), 37.3 (s; CH2), 30.5 (t, 2J-
ACHTUNGTREN(NUNG H,F)=56.8, J-
Compound 4a: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =7.01 min; 1H NMR (400 MHz, CDCl3): d=3.81 (sextet, 3J
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
6.4 Hz, 1H; CHOH), 2.02–1.45 (m, 8H; CH2), 1.21 (d, 3J
G
3
ACHTUNGTRENNUNG
3H; CH3), 0.94 ppm (t, J
N
1
ACHTUNGTRENNUNG
CDCl3): d=125.2 (t, J
(C,F)=25.0 Hz; CCF2), 32.5 (t, 2J
23.6 (s; CH2), 15.8 (t, 3J
ACHTUNGTRENNUNG ACHUTGNTRNE(NUGN C,F)=5.3 Hz; CCH2CHF2), 27.7 (s;
(C,F)=21.0 Hz; CCF2), 29.4 (t, 3J
G
ACHTUNGTRENNUNG
CH2), 22.6 (s; CH2), 13.99 ppm (s; CH3); 19F NMR (282 MHz, CDCl3):
d=À115.88, À116.14 ppm; EI-MS: m/z (%): 109 (47) [MÀC4H9]+, 89
(47), 87 (46), 71 (3), 69 (100), 51 (13), 41 (92); HRMS (ESI) [M+Na]+:
m/z calcd for C8H16F2ONa: 189.1067; found: 189.1072.
AHCTUNGTRENNUNG
19F NMR (282 MHz, CDCl3): d=À98.17, À98.25 ppm; EI-MS: m/z (%):
151 (0.6) [MÀCH3]+, 131 (21) [MÀCH3ÀHF]+, 113 (4), 102 (83), 93 (1),
87 (18), 74 (100), 67 (11), 59 (15), 41 (16); HRMS (ESI) [M+Na]+: m/z
calcd for C8H16F2ONa: 189.1067; found: 189.1065.
Compound 2a: Rf =0.520 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =7.12 min; 1H NMR (400 MHz, CDCl3): d=3.77 (sextet, 3J
ACHTUNGTRENNUNG
Compound 5a: Rf =0.540 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
8.0 Hz, 1H; CHOH), 1.87–1.75 (m, 2H; CH2), 1.55 (t, 3J
3H; CH3CF2), 1.47–1.31 (m, 6H; CH2), 1.17 ppm (d, 3J
3H; CH3); 13C NMR (100 MHz, CDCl3): d=124.3 (t, 1J
CF2), 67.9 (s; CHOH), 38.9 (s; CH2), 37.9 (t, 2J
25.4 (s; CH2), 23.5 (s; CH3), 23.2 (t, 2J
(t, 3J
ACHTUNGTRENNUNG
tR =9.03 min; 1H NMR (300 MHz, CDCl3): d=4.42 (dt, 2J
ACHTUNGTRENNUNG
and 3J(H,H)=6.0 Hz, 2H; CH2F), 3.77 (sextet, 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHUTNGRENNUG ACHTUNGTRENNUNG(C,F)=
(C,F)=19.5 Hz; CCH2F), 29.2 (s; CH2), 25.6 (s; CH2), 25.1 (d, 3J
ACHTUNGTRENNUNG
À90.62 ppm; EI-MS: m/z (%): 151 (0.5) [MÀCH3]+, 131 (1)
[MÀCH3ÀHF]+, 113 (21), 102 (14), 93 (5), 82 (16), 73 (54), 45 (100);
HRMS (ESI) [M+Na]+: m/z calcd for C8H16F2ONa: 189.1067; found:
189.1065.
5.3 Hz; CCH2CH2F), 23.4 ppm (s; CH2); 19F NMR (282 MHz, CDCl3):
d=À218.16 ppm; EI-MS: m/z (%): 133 (0.5) [MÀCH3]+, 130 (1)
[MÀH2O]+, 95 (9), 69 (1), 55(16), 45 (100), 41 (22); HRMS (ESI)
[M+Na]+: m/z calcd for C8H17FONa: 171.1161; found: 171.1167.
Compound 2b: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =6.93 min; 1H NMR (400 MHz, CDCl3): d=3.54–3.48 (1H, m,
Compound 5b: Rf =0.550 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =8.87 min; EI-MS: m/z (%): 130 (0.6) [MÀH2O]+, 119 (9) [MÀC2H5]+,
101 (1), 99 (1), 81 (47), 59 (100), 55 (41), 41 (33), 31 (26).
CHOH), 1.89–1.77 (m, 4H; CH2), 1.58 (t, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Compound 5c: Rf =0.550 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =8.58 min; EI-MS: m/z (%): 130 (0.6) [MÀH2O]+, 105 (21)
[MÀC3H7]+, 85 (18), 73 (58), 67 (23), 55 (100), 41 (5), 29 (29).
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
Compound 5d: Rf =0.550 (EtOAc/hexane 1:4); GC (isothermal, 1008C)
tR =8.42 min; EI-MS: m/z (%): 130 (1) [MÀH2O]+, 91 (57), 87 (35), 71
(69), 69 (100), 55(15).
CCH2CF2), 9.8 ppm (s; CH3); 19F NMR (282 MHz, CDCl3): d=
À90.55 ppm; EI-MS: m/z (%): 137 (1) [MÀC2H5]+, 117 (22), 99 (48), 73
(77), 59 (100), 41 (30); HRMS (ESI) [M+Na]+: m/z calcd for
C8H16F2ONa: 189.1067; found: 189.1067.
Compound 6a: Rf =0.540 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =14.78 min; 19F NMR (282 MHz, CDCl3): d=À218.29, À228.37; EI-
MS: m/z (%): 133 (4) [MÀCH2F]+, 95 (100), 69 (42), 63 (22), 55 (49), 47
(10), 45 (14), 41 (65), 33 (10).
Compound 2c: Rf =0.540 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =6.61 min; 1H NMR (400 MHz, CDCl3): d=3.64–3.58 (m, 1H;
CHOH), 1.71–1.52 (m, 4H; CH2), 1.55 (t, 3J (H,F)=18.0 Hz, 3H;
CH3CF2), 1.48–1.41 (m, 4H, CH2), 0.92 ppm (t, 3J
ACHTUNGTRENNUNG
Compound 6b: Rf =0.535 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =16.52 min; 1H NMR (300 MHz, CDCl3): d=4.71–4.05 (m, 2H; CH2),
CH3); 13C NMR (100 MHz, CDCl3): d=124.4 (t,1J
ACHTUNGTRENNUNG
4.39 (dt, 2J(H,F)=47.1 and 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
71.0 (s; CHOH), 39.7 (s; CH2), 34.1 (t, 2J
N
2
CH2), 23.4 (t, J
N
19F NMR (282 MHz, CDCl3): d=À90.92, À91.28 ppm; EI-MS: m/z (%):
123 (10) [MÀC3H7]+, 103 (67), 83 (18), 73 (56), 65 (20), 55 (100), 43 (35),
29 (17); HRMS (ESI) [M+Na]+: m/z calcd for C8H16F2ONa: 189.1067;
found: 189.1068.
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
20.0 Hz; CCH2F), 25.1 ppm (s; CH2); 19F NMR (282 MHz, CDCl3): d=
À218.29, À220.67 ppm; EI-MS: m/z (%): 119 (16) [MÀC2H4F]+, 99 (2),
81 (70), 77 (22), 55 (65), 47 (16), 41 (43), 33 (8); HRMS (ESI) [M+Na]+:
m/z calcd for C8H16F2ONa: 189.1067; found: 189.1070.
Compound 3a: Rf =0.520 (EtOAc/hexane 1:4); GC (isothermal, 1008C)
tR 7.37 min; 1H NMR (300 MHz, CDCl3): d=3.79 (sextet, 3J
ACHTUNGTRENNUNG
A
N
Compound 6c: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =17.36 min; 1H NMR (400 MHz, CDCl3): d=4.45 (dt, 2J (H,F)=47.2,
AHCTUNGTRENNUNG
N
3J(H,H)=5.6 Hz, 2H; CH2F), 4.42 (dt, 2J(H,F)=47.2 and 3J
ACHTUNGTRENNNUG ACHTUNGTENRNUG ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
4784
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

À
C H Activation in n-Octanes
FULL PAPER
CH2); 13C NMR (100 MHz, CDCl3): d=84.1 (d, 1J
G
Compound (S)-2a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =187 min; 1H NMR (400 MHz,
CDCl3): d=7.45–7.43 (m, 2H; Ar-H), 7.36–7.33 (m, 3H; Ar-H), 5.83 (s,
1H; PhCHOCO), 4.96–4.86 (m, 1H; CHOCO), 2.17 (s, 3H; CH3CO),
CH2F), 83.9 (d, 1J
ACHTUNGTRENNUNG
CH2), 33.0 (d, 3J
A
E
1.64–1.52 (m, 2H; CH3CF2CH2), 1.49 (t, 3J
CH3CF2), 1.46–1.37 (m, 4H; CH2), 1.05 (d, 3J
ACHTUNGTRENNUNG
5.0 Hz; CCH2CH2F); 19F NMR (282 MHz, CDCl3): d=À217.92,
À218.44 ppm; EI-MS: m/z (%): 105 (25) [MÀC3H6F]+, 91 (64)
[MÀC4H8F]+, 85 (18), 81 (2), 71 (49), 67 (30), 61 (5), 55 (27), 53 (6), 47
(9); HRMS (ESI) [M+Na]+: m/z calcd for C8H16F2ONa: 189.1067;
found: 189.1071.
AHCTUNGTRENNUNG
1.10–0.93 ppm (m, 2H; CH2); 19F NMR (282 MHz, CDCl3): d=
À90.60 ppm; EI-MS: m/z (%): 177(3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1536.
Compound 7a: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =7.50 min; EI-MS: m/z (%): 183 (0.2) [MÀH]+, 169 (2) [MÀCH3]+,
166 (1) [MÀH2O]+, 151 (2), 131 (1), 105 (0.5), 91 (1), 77 (4), 45 (100).
Compound 7b: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =7.34 min; EI-MS: m/z (%): 183 (0.2) [MÀH]+, 166 (0.4) [MÀH2O]+,
155 (13) [MÀC2H5]+, 137 (5), 117 (10), 97 (5), 59 (100), 41 (21).
Compound (R)-2b’: Rf =0.90 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =164 min; 1H NMR (400 MHz,
CDCl3): d=7.47–7.43 (m, 2H; Ar-H), 7.37–7.33 (m, 3H; Ar-H), 5.85 (s,
1H; PhCHOCO), 4.87–4.78 (m, 1H; CHOCO), 2.17 (s, 3H; CH3CO),
1.67–1.38 (m, 6H; CH2), 1.38 (t, 3J
ACHTUNGTRENNUNG
1.04 (m, 2H; CH2), 0.88 ppm (t, 3J
AHCTUNGTRENNUNG
Compound 7c: Rf =0.530 (EtOAc/hexane 1:4); GC (isothermal, 1008C):
tR =7.22 min; EI-MS: m/z (%): 166 (0.2) [MÀH2O]+, 141 (19)
[MÀC3H7]+, 121 (33), 101 (0.8), 83 (2), 73 (67), 57 (9), 55 (100), 43 (31).
(282 MHz, CDCl3): d=À90.38, À90.49 ppm; EI-MS: m/z (%): 177 (3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15); HRMS (ESI)
[M+Na]+: m/z calcd for C18H24F2O4Na: 365.1540; found: 365.1548.
Preparation of 1a’–5a’, 1b’–3b’ and 5b: (R)-O-Acetylmandelic acid
(76 mg, 0.39 mmol, 1.3 equiv) and DMAP (4.0 mg) were dissolved in di-
chloromethane (5.0 mL) at À408C in a 10 mL round-bottomed flask.
Over the next 5 min, a solution of DCC (75 mg, 0.39 mmol, 1.3 equiv) in
dichloromethane (5.0 mL) was added dropwise to the flask. A white pre-
cipitate resulted after 10 min. Then, a solution of authentic standards of
the octanol stereoisomers or fluorinated alcoholic products (0.30 mmol,
1.0 equiv) in dichloromethane (1.0 mL) was added over a second 5 min
period, and the mixture was stirred overnight at room temperature. The
suspension was filtered through a pad of silica gel and washed with di-
chloromethane (5–10 mL). The filtrate was evaporated to near dryness
under vacuum and the residue was suspended in dichloromethane (ca.
2.0 mL). The corresponding products were further purified by gravity
column chromatography (EtOAc in hexane=20%) and analyzed by 13C
Compound (S)-2b’: Rf =0.90 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =166 min; 1H NMR (400 MHz,
CDCl3): d=7.47–7.43 (m, 2H; Ar-H), 7.37–7.33 (m, 3H; Ar-H), 5.84 (s,
1H; PhCHOCO), 4.87–4.78 (m, 1H; CHOCO), 2.17 (s, 3H; CH3CO),
1.67–1.38 (m, 6H; CH2), 1.38 (t, 3J
1.04 (m, 2H; CH2), 0.57 ppm (t, 3J
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
(282 MHz, CDCl3): d=À90.38, À90.49 ppm; EI-MS: m/z (%): 177 (3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15); HRMS (ESI)
[M+Na]+: m/z calcd for C18H24F2O4Na: 365.1540; found: 365.1548.
Compound (R)-2c’: Rf =0.91 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =139 min; 1H NMR (400 MHz,
CDCl3): d=7.48–7.44 (m, 2H; Ar-H), 7.38–7.34 (m, 3H; Ar-H), 5.84 (s,
1H; PhCHOCO), 4.91–4.88 (m, 1H; CHOCO), 2.18 (s, 3H; CH3CO),
1.62–1.42 (m, 4H; CH2), 1.32 (t, 3J
1.21 (m, 4H; CH2), 0.89 ppm (t, 3J
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
1
and H NMR spectroscopy and GC/MSD.
(282 MHz, CDCl3): d=À92.33, À92.42 ppm; EI-MS: m/z (%): 177 (3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15); HRMS (ESI)
[M+Na]+: m/z calcd for C18H24F2O4Na: 365.1540; found: 365.1544.
Compound (R)-1a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =214 min; 1H NMR (300 MHz,
CDCl3): d=7.46–7.43 (m, 2H; Ar-H), 7.37–7.34 (m, 3H; Ar-H), 5.84 (s,
1H; PhCHOCO); 5.70 (tt, 2J(H,F)=57.0, 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
Compound (S)-2c’: Rf =0.91 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =139 min; 1H NMR (400 MHz,
CDCl3): d=7.48–7.44 (m, 2H; Ar-H), 7.38–7.34 (m, 3H; Ar-H), 5.82 (s,
1H; PhCHOCO), 4.91–4.88 (m, 1H; CHOCO), 2.18 (s, 3H; CH3CO),
ACHTUNGTRENNUNG
À115.89 ppm; EI-MS: m/z (%): 177 (3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H25FO4Na: 365.1540; found: 365.1537.
1.62–1.42 (m, 4H; CH2), 1.32 (t, 3J
1.21 (m, 4H; CH2), 0.69 ppm (t, 3J
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
(282 MHz, CDCl3): d=À92.33, À92.42 ppm; EI-MS: m/z (%): 177 (3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15); HRMS (ESI)
[M+Na]+: m/z calcd for C18H24F2O4Na: 365.1540; found: 365.1544.
Compound (S)-1a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =216 min; 1H NMR (300 MHz,
CDCl3): d=7.46–7.43 (m, 2H; Ar-H), 7.37–7.34 (m, 3H; Ar-H), 5.83 (s,
1H; PhCHOCO), 5.76 (tt, 2J(H,F)=57.0, 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
Compound (R)-3a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1308C; flow rate: 0.5 mLminÀ1): tR =412 min; 1H NMR (300 MHz,
CDCl3): d=7.46–7.42 (m, 2H; Ar-H), 7.38–7.33 (m, 3H; Ar-H), 5.85 (s,
1H; PhCHOCO), 4.98–4.86 (m, 1H; CH3CHOCO), 2.17 (s, 3H;
ACHTUNGTRENNUNG
À115.89 ppm; EI-MS: m/z (%): 177 (3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H25FO4Na: 365.1540; found: 365.1537.
CH3CO), 1.81–1.37 (m, 8H; CH2), 1.24 (d, 3J
0.90 ppm (t, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
d=À100.22 ppm; EI-MS: m/z (%): 177 (3); 149 (59); 135 (2); 118 (4);
107 (100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1547.
Compound (R)-1b’: Rf =0.90 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =192 min; EI-MS: m/z (%): 177(3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15).
Compound (S)-3a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1308C; flow rate: 0.5 mLminÀ1): tR =414 min; 1H NMR (300 MHz,
CDCl3): d=7.46–7.42 (m, 2H; Ar-H), 7.38–7.33 (m, 3H; Ar-H), 5.83 (s,
1H; PhCHOCO), 4.98–4.86 (m, 1H; CH3CHOCO), 2.17 (s, 3H;
Compound (S)-1b’: Rf =0.90 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =195 min; EI-MS: m/z (%): 177(3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15).
CH3CO), 1.81–1.37 (m, 8H; CH2), 1.07 (d, 3J
0.97 ppm (t, 3J
ACHTUNGTRENNUNG
Compound (R)-2a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =185 min; 1H NMR (400 MHz,
CDCl3): d=7.45–7.43 (m, 2H; Ar-H), 7.36–7.33 (m, 3H; Ar-H), 5.85 (s,
1H; PhCHOCO), 4.96–4.86 (m, 1H; CHOCO), 2.17 (s, 3H; CH3CO),
AHCTUNGTRENNUNG
d=À100.22 ppm; EI-MS: m/z (%): 177 (3); 149 (59); 135 (2); 118 (4);
107 (100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1547.
1.64–1.52 (m, 2H; CF2CH2), 1.49 (t, 3J
1.46–1.37 (m, 6H; CH2), 1.22 ppm (d, 3J
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
Compound (R)-3b’: Rf =0.90 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =148 min; 1H NMR (300 MHz,
CDCl3): d=7.47–7.43 (m, 2H; Ar-H), 7.36–7.31 (m, 3H; Ar-H), 5.83 (s,
1H; PhCHOCO), 4.85–4.77 (m, 1H; CHOCO), 2.15 (s, 3H; CH3CO),
19F NMR (282 MHz, CDCl3): d=À90.60 ppm; EI-MS: m/z (%): 177 (3);
149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15); HRMS (ESI)
[M+Na]+: m/z calcd for C18H24F2O4Na: 365.1540; found: 365.1536.
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4785

S. S.-F. Yu et al.
3
1.63–1.19 (m, 8H; CH2), 0.87 (t, J
3J
(H,H)=7.2 Hz, 3H; CH3), 0.78 ppm (t,
and 4.0 mm substrate in DMSO (1% v/v). After the enzymatic reaction
proceeded for 1.0 min at 308C (measurement of the initial oxidation
rate), the reaction mixtures (1 mL) were quenched with CH2Cl2 (300 mL,
with p-xylene as internal standard) and HCl (6.0m, 100 mL) in a 2 mL mi-
crocentrifuge tube. The tube was vortexed and then centrifuged at
13000 rpm for 5.0 min. The organic layer was removed with a pipette and
analyzed by GC to determine the product concentrations and throughput
ratios. All reactions were carried out in triplicate.
ACHTUNGTRENNUNG
À101.61 ppm; EI-MS: m/z (%): 177(3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1532.
Compound (S)-3b’: Rf =0.90 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =149 min; 1H NMR (300 MHz,
CDCl3): d=7.47–7.43 (m, 2H; Ar-H), 7.36–7.31 (m, 3H; Ar-H), 5.81 (s,
1H; PhCHOCO), 4.85–4.77 (m, 1H; CHOCO), 2.15 (s, 3H; CH3CO),
Quantification of enzymatic alcoholic products was calibrated by the pu-
rified 2-alcohols 1--7a or commercially available octan-2-ol (Sigma–Al-
drich). The GC intensities of 1a–d, 2a–c, 3a, and 3b were determined by
the calibration curve based on 1a. Meanwhile, the GC intensities of 4a,
5a–d, 6a–c, and 7a–d were determined by the calibration curve based on
4a, 5a, 6a, and 7a, respectively. Aliquots of 2-octanol, 1a, 4a, 5a, 6a,
and 7a (0.03, 0.05, 0.10, 0.21, and 0.41 mg, respectively) were dissolved in
CH2Cl2 (10 mL) containing xylene (0.43 mg). The corresponding GC in-
tensity of standard alcohols was normalized by the intensity derived from
xylene as the internal standard. The measured ratios of Ialcohol versus Ixylene
then were fitted to a linear function of the weight of the alcohols that
were added. The regressions for these linear plots from 0.01 to 0.50 mg
were 0.997 to 0.999 and indicated that all the measurements within this
range were valid.
3
1.63–1.19 (m, 8H; CH2), 0.95 (t, J
G
3J
ACHTUNGTRENNUNG
À101.61 ppm; EI-MS: m/z (%): 177 (3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1532.
Compound (R)-4a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =174 min; 1H NMR (300 MHz,
CDCl3): d=7.47–7.43 (m, 2H; Ar-H), 7.38–7.33 (m, 3H; Ar-H), 5.85 (s,
1H; PhCHOCO), 4.98–4.86 (m, 1H; CHOCO), 2.17 (s, 3H; CH3CO),
1.78–1.28 (m, 8H; CH2), 1.25 (d, 2J
G
3
(t, J
U
À99.17 ppm; EI-MS: m/z (%): 177 (3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1542.
Compound (S)-4a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =175 min; 1H NMR (300 MHz,
CDCl3): d=7.47–7.43 (m, 2H; Ar-H), 7.38–7.33 (m, 3H; Ar-H), 5.82 (s,
1H; PhCHOCO), 4.98–4.86 (m, 1H; CHOCO), 2.17 (s, 3H; CH3CO),
Acknowledgements
1.78–1.28 (m, 8H; CH2), 1.09 (d, 3J
A
3
This work was supported by Academia Sinica and grants from the Na-
tional Science Council, R.O.C. (NSC 97-2113M-001-006-MY3 and 98-
3114-E-006-014). We thank Dr. Mei-Chun Tseng for her assistance with
ESI-MS.
(t, J
A
À99.17 ppm; EI-MS: m/z(%): 177 (3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H24F2O4Na: 365.1540; found: 365.1542.
Compound (R)-5a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =255 min; 1H NMR (300 MHz,
CDCl3): d=7.45–7.43 (m, 2H; Ar-H), 7.37–7.34 (m, 3H; Ar-H), 5.85 (s,
1H; PhCHOCO), 4.95–4.85 (m, 1H; CH3CHOCO), 4.35 (dt, 2J (H,F)=
48, 3J (H,H)=4.0 Hz, 2H; CH2F), 2.17 (s, 3H; CH3CO), 1.68–1.29 (m,
[1] L. O. Narhi, A. J. Fulco, J. Biol. Chem. 1987, 262, 6683–6690.
[2] V. B. Urlacher, R. D. Schmid, Methods Enzymol. 2004, 388, 208–
224.
[3] S. N. Daff, S. K. Chapman, K. L. Turner, R. A. Holt, S. Govindaraj,
T. L. Poulos, A. W. Munro, Biochemistry 1997, 36, 13816–13823.
[4] R. Fasan, Y. T. Meharenna, C. D. Snow, T. L. Poulos, F. H. Arnold, J.
Mol. Biol. 2008, 383, 1069–1080.
[5] V. B. Urlacher, R. D. Schmid, Curr. Opin. Chem. Biol. 2006, 10,
156–161.
[6] a) A. M. Sawayama, M. M. Y. Chen, P. Kulanthaivel, M. S. Kuo, H.
Hemmerle, F. H. Arnold, Chem. Eur. J. 2009, 15, 11723–11729;
b) A. Rentmeister, F. H. Arnold, R. Fasan, Nat. Chem. Biol. 2008, 5,
26–28.
6H; CH2), 1.22 (d, 3J
ACHTUNGTRENNUNG(H,H)=4.0 Hz, 3H; CH3), 1.22–1.11 ppm (m, 4H;
CH2); 19F NMR (282 MHz, CDCl3): d=À218.34 ppm; EI-MS: m/z (%):
177 (3); 149 (59); 135 (2); 118 (4); 107 (100); 43 (44); 28 (15); HRMS
(ESI) [M+Na]+: m/z calcd for C18H25FO4Na: 347.1635; found: 347.1637.
Compound (S)-5a’: Rf =0.89 (EtOAc/hexane 1:4); GC (isothermal,
1408C; flow rate: 0.8 mLminÀ1): tR =258 min; 1H NMR (300 MHz,
CDCl3): d=7.45–7.43 (m, 2H; Ar-H), 7.37–7.34 (m, 3H; Ar-H), 5.84 (s,
1H; PhCHOCO), 4.95–4.85 (m, 1H; CH3CHOCO), 4.42 (dt, 2J
ACHTUNGTRENNUNG
48, 3J
6H; CH2), 1.22–1.11 (m, 2H; CH2), 1.10 (d, 3J
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
[7] a) A. Cherkasov, M. Jonsson, J. Chem. Inf. Comput. Sci. 2000, 40,
1222–1226; b) S. Shaik, S. Cohen, Y. Wang, H. Chen, D. Kumar, W.
Thiel, Chem. Rev. 2010, 110, 949–1017.
[8] Q. S. Li, U. Schwaneberg, P. Fischer, R. D. Schmid, Chem. Eur. J.
2000, 6, 1531–1536.
0.98–0.88 ppm (m, 2H; CH2); 19F NMR (282 MHz, CDCl3): d=
À218.34 ppm; EI-MS: m/z (%): 177(3); 149 (59); 135 (2); 118 (4); 107
(100); 43 (44); 28 (15); HRMS (ESI) [M+Na]+: m/z calcd for
C18H25FO4Na: 347.1635; found: 347.1637.
Determination of NADPH consumption rate:[9b] Rate of NADPH con-
sumption was determined by using a Beckman Coulter DU-800 UV/Vis
spectrophotometer and a 1 cm path length cuvette. NADPH turnovers
were carried out at 308C in a reaction solution containing the purified
enzyme (800 mL, 200 nm) in Tris-HCl buffer (100 mm, pH 8.2), bovine
liver catalase (13 mg), and 4.0 mm substrate (added as a 400 mm stock so-
lution in DMSO). P450 enzyme concentrations were determined by re-
duced CO difference spectra of the reduced heme by using an extinction
coefficient of 91 mmÀ1 cmÀ1 for the 450 minus 490 nm peak.[9b,34] Assays
were held for 1–2 min prior to NADPH addition at a final concentration
of 320 mm, and the decrease in absorption at 340 nm was monitored.
Rates were calculated based on the extinction coefficient of NADPH
(6.22 mmÀ1 cmÀ1). Background NADPH consumption rates were mea-
sured without substrates. All reactions were carried out in triplicate.
[9] a) M. W. Peters, P. Meinhold, A. Glieder, F. H. Arnold, J. Am.
Chem. Soc. 2003, 125, 13442–13450; b) P. Meinhold, M. W. Peters,
A. Hartwick, A. R. Hernandez, F. H. Arnold, Adv. Synth. Catal.
2006, 348, 763–772.
[10] K. Y. Ng, L.-C. Tu, Y.-S. Wang, S. I. Chan, S. S.-F. Yu, ChemBio-
Chem 2008, 9, 1116–1123.
[11] a) H.-J. Bçhm, D. Banner, S. Bendels, M. Kansy, B. Kuhn, K.
Mꢄller, U. Obst-Sander, M. Stahl, ChemBioChem 2004, 5, 637–643;
b) K. Muller, C. Faeh, F. Diederich, Science 2007, 317, 1881–1886;
c) S. Purser, P. R. Moore, S. Swallow, V. Gouverneur, Chem. Soc.
Rev. 2008, 37, 320–330.
[12] B. K. Park, N. R. Kitteringham, P. M. OꢃNeill, Annu. Rev. Pharma-
col. Toxicol. 2001, 41, 443–470.
[13] a) E. N. G. Marsh, Chem. Biol. 2000, 7, R153-R157; b) K. Ahn, D. S.
Johnson, M. Mileni, D. Beidler, J. Z. Long, M. K. McKinney, E.
Weerapana, N. Sadagopan, M. Liimatta, S. E. Smith, S. Lazerwith,
C. Stiff, S. Kamtekar, K. Bhattacharya, Y. Zhang, S. Swaney, K. V.
Determination of product throughputs or product formation rates:[9b]
NADPH (200 mL) to a final concentration of 320 mm was added to Tris-
HCl buffer (100 mm, pH 8.2, 800 mL) containing purified protein (200 nm)
4786
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4774 – 4787

À
C H Activation in n-Octanes
FULL PAPER
Becelaere, R. C. Stevens, B. F. Cravatt, Chem. Biol. 2009, 16, 411–
420; c) R. Pongdee, H.-W. Liu, Bioorg. Chem. 2004, 32, 393–437;
d) S. I. Chan, S. S.-F. Yu, Acc. Chem. Res. 2008, 41, 969–979.
[14] a) R. P. Singh, J. M. Shreeve, Synthesis 2002, 2561–2578; b) P.
Kirsch, Modern Fluoroorganic Chemistry-Synthesis Reactivity, Appli-
cations, Wiley-VCH, Weinheim, 2004.
[15] W. H. Brown, T. Poon, Introduction to Organic Chemistry, 4th ed.,
Wiley, New York, 2010.
[16] K. S. Eble, J. H. Dawson, J. Biol. Chem. 1984, 259, 14389–14393.
[17] Z.-Y. Long, Q.-Y. Chen, Tetrahedron Lett. 1998, 39, 8487–8490.
[18] N. O. Brace, J. Fluorine Chem. 2001, 108, 147–175.
[19] R. Cloux, E. Kovꢅts, Synthesis 1992, 409–412.
[25] a) M. J. Cryle, J. J. De Voss, Tetrahedron: Asymmetry 2007, 18, 547–
551; b) M. J. Cryle, N. J. Matovic, J. J. De Voss, Tetrahedron Lett.
2007, 48, 133–136.
[26] H. Li, T. L. Poulos, Nat. Struct. Mol. Biol. 1997, 4, 140–146.
[27] P. R. Ortiz de Montellano, Chem. Rev. 2010, 110, 932–948.
[28] a) S. I. Chan, K. H.-C. Chen, S. S.-F. Yu, C.-L. Chen, S. S.-J. Kuo,
Biochemistry 2004, 43, 4421–4430; b) S. J. Elliott, M. Zhu, L. Tso,
H. H. T. Nguyen, J. H. K. Yip, S. I. Chan, J. Am. Chem. Soc. 1997,
119, 9949–9955.
[29] T. C. Bruice, S. J. Benkovic, Biochemistry 2000, 39, 6267–6274.
[30] G. B. Crull, J. W. Kennington, A. R. Garber, P. D. Ellis, J. H.
Dawson, J. Biol. Chem. 1989, 264, 2649–2655.
[20] E. Weber, A. Seifert, M. Antonovici, C. Geinitz, J. Pleiss, V. B. Ur-
lacher, Chem. Commun. 2011, 47, 944–946.
[31] Y. Nakagawa, K. Irie, R. C. Yanagite, H. Ohigashi, K. Tsuda, J. Am.
Chem. Soc. 2005, 127, 5746–5747.
[21] S. S.-F. Yu, L.-Y. Wu, K. H.-C. Chen, W.-I. Luo, D.-S. Huang, S. I.
Chan, J. Biol. Chem. 2003, 278, 40658–40669.
[22] D. Appel, S. Lutz-Wahl, P. Fischer, U. Schwaneberg, R. D. Schmid,
J. Biotechnol. 2001, 88, 167–171.
[23] D. C. Haines, D. R. Tomchick, M. Machius, J. A. Peterson, Biochem-
istry 2001, 40, 13456–13465.
[32] B. E. Smart, J. Fluorine Chem. 2001, 109, 3–11.
[33] David R. Lide ed. , CRC Handbook of Chemistry and Physics, 90th
ed., CRC Press/Taylor and Francis, Boca Raton, 2010.
[34] a) S. C. Maurer, H. Schulze, R. D. Schmid, V. B. Urlacher, Adv.
Synth. Catal. 2003, 345, 802–810; b) T. Omura, R. Sato, J. Biol.
Chem. 1964, 239, 2379–2385.
[24] G. Truan, M. R. Komandla, J. R. Falck, J. A. Peterson, Arch. Bio-
chem. Biophys. 1999, 366, 192–198.
Received: December 16, 2010
Published online: March 11, 2011
Chem. Eur. J. 2011, 17, 4774 – 4787
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4787