Regio- and stereoselective biohydroxylations with a recombinant escherichia coli expressing P450pyr monooxygenase of sphingomonas Sp. HXN-200

Article abstract of DOI:10.1002/adsc.201000266

A recombinant Escherichia coli expressing P450_pyr monooxygenase of Sphingomonas sp. HXN-200 was developed as a useful biocatalyst for regio- and stereoselective hydroxylations, with no side reaction and easy cell growth. The resting E. coli cells showed an activity of 4.1 U/g cdw and 9.9 U/g cdw for the hydroxylation of N-benzylpyrrolidin-2-one 1 and N-benzyloxycarbonylpyrrolidine 3, respectively, being as active as the wide-type strain. Biohydroxylation of N-benzylpyrrolidin-2-one 1 with the resting cells gave (S)-N-benzyl-4- hydroxypyrrolidin-2-one 2 in >99% ee and 10.8 mM, a 2.6 times increase of product concentration in comparison with the wild-type strain. Biohydroxylation of N-tert-butoxycarbonylpiperidin-2-one 5, N-benzylpiperidine 7 and N-tert-butoxycarbonylazetidine 9 with the E. coli cells afforded the corresponding 4-hydroxypiperidin-2-one 6, 4-hydroxypiperidine 8, and 3-hydroxyazetidine 10 in 14 mM, 17 mM, and 21 mM, respectively. Moreover, hydroxylation of (-)-β-pinene 11 with the recombinant E. coli cells showed excellent regio- and stereoselectivity and gave (1R)-trans-pinocarveol 12 in 82% yield and 4.1 mM, which is over 200 times higher than that obtained with the best biocatalytic system known thus far. The recombinant strain was also able to hydroxylate other types of substrates with unique selectivity: biohydroxylation of norbornane 13 gave exo-norbornaeol 14, with exo/endo selectivity of 95%; tetralin 15 and 6-methoxytetralin 17 were hydroxylated at the non-activated 2-position, for the first time, with regioselectivities of 83-84%. Copyright

Full text of DOI:10.1002/adsc.201000266

FULL PAPERS
DOI: 10.1002/adsc.201000266
Regio- and Stereoselective Biohydroxylations with a
Recombinant Escherichia coli Expressing P450pyr
Monooxygenase of Sphingomonas Sp. HXN-200
a,b
a
a
a,b,
Wei Zhang, Weng Lin Tang, Zunsheng Wang, and Zhi Li *
a
Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4,
Singapore 117576
Fax : (+65)-6779-1936; phone: (+65)-6516-8416; e-mail: chelz@nus.edu.sg
Singapore-MIT Alliance, National University of Singapore, 4 Engineering Drive 3, Singapore 117576
b
Received: April 5, 2010; Revised: November 5, 2010; Published online: December 7, 2010
Abstract: A recombinant Escherichia coli expressing in 14 mM, 17 mM, and 21 mM, respectively. More-
P450pyr monooxygenase of Sphingomonas sp. HXN- over, hydroxylation of (À)-b-pinene 11 with the re-
200 was developed as a useful biocatalyst for regio- combinant E. coli cells showed excellent regio- and
and stereoselective hydroxylations, with no side reac- stereoselectivity and gave (1R)-trans-pinocarveol 12
tion and easy cell growth. The resting E. coli cells in 82% yield and 4.1 mM, which is over 200 times
showed an activity of 4.1 U/g cdw and 9.9 U/g cdw higher than that obtained with the best biocatalytic
for the hydroxylation of N-benzylpyrrolidin-2-one 1 system known thus far. The recombinant strain was
and N-benzyloxycarbonylpyrrolidine 3, respectively, also able to hydroxylate other types of substrates
being as active as the wide-type strain. Biohydroxyla- with unique selectivity: biohydroxylation of norbor-
tion of N-benzylpyrrolidin-2-one 1 with the resting nane 13 gave exo-norbornaeol 14, with exo/endo se-
cells gave (S)-N-benzyl-4-hydroxypyrrolidin-2-one 2 lectivity of 95%; tetralin 15 and 6-methoxytetralin 17
in >99% ee and 10.8 mM, a 2.6 times increase of were hydroxylated at the non-activated 2-position,
product concentration in comparison with the wild- for the first time, with regioselectivities of 83–84%.
type strain. Biohydroxylation of N-tert-butoxycarbo-
nylpiperidin-2-one 5, N-benzylpiperidine 7 and N-
tert-butoxycarbonylazetidine 9 with the E. coli cells Keywords: enzyme catalysis; hydroxylations; (S)-4-
afforded the corresponding 4-hydroxypiperidin-2-one hydroxypyrrolidin-2-one; P450 monooxygenase;
6, 4-hydroxypiperidine 8, and 3-hydroxyazetidine 10 (1R)-trans-pinocarveol; recombinant Escherichia coli
[
3]
Introduction
monooxygenase (MMO), membrane-bound alkane
[
4]
hydroxylase (AlkB),
and cytochrome P450
Among them, cytochrome P450
AHCTUNGTREGUNgNN enases are very attractive, since they widely
[
5–7]
Regio- and stereoselective hydroxylations are useful monooxy
reactions for the preparation of valuable chiral alco- monooxy
ACHTUNGTRENUNNGg enases.
hols in fine chemical, pharmaceutical, as well as exist in nature and have high diversity. Many P450
flavor and fragrance industries. However, this type of monooxygenases identified from wild-type strains are
[
5a,b]
transformation remains as a great challenge in classi- able to catalyze selective hydroxylation,
and some
[1]
cal chemistry. On the other hand, selective hydroxyl- of them, such as P450
ation can be achieved by using an enzyme. For exam- characterized,
and P450BM-3, have been
cam
[
5c–e]
[5f–j]
cloned and expressed,
and engi-
[
6]
ple, monooxygenase can catalyze the insertion of an neered via directed evolution or protein engineer-
[
7]
O atom of molecular oxygen into a specific CÀH ing. However, it is still difficult to obtain a satisfac-
[2]
bond. In addition to the high regio- and stereoselec- tory P450 monooxygenase with desired substrate spe-
tivity, enzymatic hydroxylation utilizes molecular cificity and high selectivity for a given hydroxylation.
oxygen as oxidant, thus being an ideal tool for green It is also a significant challenge to construct highly
oxidation and sustainable chemical synthesis. Some active recombinant biocatalysts via genetic engineer-
progress has been achieved in the discovery and ap- ing of P450 monooxygenase, possibly due to the par-
plication of appropriate monooxygenases for enzy- ticular complexity of the P450 enzyme and its catalyt-
[
2]
matic hydroxylations, and several enzymatic systems ic system.
have been extensively investigated, such as methane
3380
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3380 – 3390

Regio- and Stereoselective Biohydroxylations with a Recombinant Escherichia coli
Previously, we discovered Sphingomonas sp. HXN- cells. Finally, the purified pRSFDuet P450pyr and pET-
[8a]
2
00 containing a P450pyr monooxygenase as a pow- Duet FdxFdR plasmids were transformed into elec-
erful hydroxylation catalyst with unique substrate spe- trocompetent E. coli BL21 (DE3) cells.
cificity as well as high selectivity. The wild-type strain
The constructed recombinant strain, E. coli BL21-
was shown to be the best catalyst known thus far for pRSFDuet P450 -pETDuet FdxFdR [E. coli
pyr
the hydroxylation of a range of alicyclic substrates (P450 )], was grown on LB medium containing am-
pyr
[
8b]
such [ as N-substituted pyrrolidines,
pyrrolidin- picillin (100 mg/mL) and kanamycin (50 mg/mL) at
and azetidi- 378C, and its P450 monooxygenase expression was in-
8c]
[8d]
[8e]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
ones,
piperidines,
piperidinones,
[8d]
nes, with high activity and good to excellent regio- duced by the addition of IPTG after the lag phase.
and stereoselectivity. For example, hydroxylation of The optimum induction condition was the addition of
N-benzylpyrrolidin-2-one 1 with this wide-type strain IPTG (0.5 mM) at 2 h followed by incubation at 308C
gave N-benzyl-4-hydroxypyrrolidin-2-one 2 in >99% for 3 h. The typical growth curve under the optimal
[8c]
ee, and product 2 is a useful intermediate for the culture condition is shown in Figure 1. Cell density
preparation of an oral carbapenem antibiotic CS- reached its maximum (1.8 g cdw/L) at 6 h and de-
[9a]
[9b]
8
34 and the nootropic drug (S)-Oxiracetame.
A
creased slowly afterwards.
Pseudomonas putida strain expressing P450 mono-
Sodium dodecyl sulfate-polyacrylamide gel electro-
pyr
[
10]
oxygenase was constructed,
but its hydroxylation phoresis (SDS-PAGE) of the cell free extracts (CFEs)
activity was rather low. Both the wild-type Sphingo- of E. coli (P450 ), that was non-induced or induced
pyr
monas strain and the recombinant P. putida strain for 2–5 h, is shown in Figure 2. Although P450pyr and
need to grow on n-octane which is a flammable and FdR cannot be distinguished in SDS-PAGE due to
relatively expensive substrate, thus being a technical the similar molecular weights, it was clear that
challenge in large-scale application.
P450 /FdR and Fdx were expressed in induced E.
pyr
Here we report our success in engineering a re-
combinant E. coli strain expressing P450pyr monooxy-
genase with high hydroxylation activity, no side reac-
tion, and easy growth on a non-flammable substrate,
applying the recombinant strain in the regio- and ste-
reoselective hydroxylation of 1 to prepare (S)-N-
benzyl-4-hydroxypyrrolidin-2-one 2 with increased
concentration, using the strain as the best catalyst
known thus far for regio- and stereoselective hydrox-
ylation of (À)-b-pinene 11 to prepare trans-pinocar-
veol 12, and exploring the unique hydroxylation of
other type of substrates such as norbornane 13, tetra-
lin 15, and 6-methoxy-tetralin 17.
Figure 1. Growth (&) and hydroxylation activity for 1 (
and 3 (~) of E. coli (P450pyr).
&
)
Results and Discussion
Genetic Engineering, Cell Growth, and Protein
Expression of E. coli (P450pyr)
To engineer a recombinant strain with active P450 ,
pyr
[
10]
plasmid pCom8-P F R
was used as a poly-
A7 200 1500
merase chain reaction (PCR) template to amplify the
coding region of cytochrome P450pyr (P ), ferredoxin
A7
(
F , Fdx), and ferredoxin reductase (R1500, FdR).
200
PCR amplified genes were digested with their respec-
tive enzymes. The digested P450pyr fragment was ligat-
ed into the pRSFDuet, while the digested Fdx frag-
ment was ligated into the pETDuet vector, and the
resulting plasmids from the ligations were separately
transformed into chemical competent E. coli DH5a
cells. Subsequently, the digested FdR fragment was li-
gated into the same pETDuet vector containing the
Figure 2. SDS-PAGE of CFE of E. coli (P450pyr). Non-in-
Fdx gene. The resulting plasmid from the ligation was duced (lane 1), induced with IPTG for 2 h (lane 2), 3 h (lane
transformed into chemical competent E. coli DH5a 3), 4 h (lane 4), and 5 h (lane 5).
Adv. Synth. Catal. 2010, 352, 3380 – 3390
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3381

Wei Zhang et al.
FULL PAPERS
Scheme 1.
Figure 3. CO difference spectra of CFEs of E. coli (P450pyr):
(
À) non-induced; (—) induced with IPTG for 3 h.
onstrated that the cloned and expressed P450pyr mono-
oxygenase was catalytically functional. These activi-
ties are nearly as high as those achieved with Sphin-
[8c]
coli cells when compared with non-induced cells in gomonas sp. HXN-200 grown on n-octane vapor,
Figure 2. and the recombinant E. coli cells could be more
To examine the percentage of the expressed active easily and economically prepared than the wild-type
P450pyr [enzyme in all soluble proteins, CO difference strain. Moreover, the specific hydroxylation activities
11]
spectra were measured with CFEs (1.7 g protein/L) are also outstanding in comparison with other well-
of induced E. coli (P450 ) cells in KP buffer known recombinant P450 monooxygeanses: P. putida
pyr
(
50 mM; pH 7.5). As shown in Figure 3, absorbance GPo12 (pGEc47DB)(pCom8-PFR1500) showed 3 U/g
[5h]
difference at 450 nm between the CO-bound reduced cdw for the hydroxylation of l-limonene, while E.
P450pyr and reduced P450pyr was 0.041. The active coli JM103 (pUC13 BM3-1) and E. coli BL21 (DE3)
[
5i]
[5j]
P450pyr enzyme in the CFE was thus established as (pETDuet-P450cam/Pdx) (pACYCDuet-PdR/GLD)
.033 g protein/L based on the molar extinction incre- gave 2.4 U/g cdw and 0.05 U/g cdw for the hydroxyl-
0
À1
[11]
ment at 450 nm of 91 cm for P450 enzyme. This ation of myristate and camphor, respectively, calculat-
corresponds to 1.92% of all soluble proteins in CFEs ed with the reported data.
of induced E. coli (P450 ). For CFE of non-induced
The hydroxylation of 1 with the recombinant E.
E. coli cells, there was no absorbance difference at coli strain was examined at different pH and tempera-
50 nm between CO-bound reduced P450pyr and re- ture. A pH of 8.0 and 258C were found to be the opti-
pyr
4
duced P450pyr as shown in Figure 3.
mum, giving a specific activity of 4.1 U/g cdw.
The kinetics of hydroxylation of pyrrolidin-2-one 1
and pyrrolidine 3 with resting cells and CFE of E. coli
Biohydroxylation of N-Substituted Pyrrolidinone 1
(P450 ) were investigated, respectively. A set of ac-
pyr
and Pyrrolidine 3
tivity assays was performed with substrates 1 and 3 at
different concentrations in the CFE (1.5 g protein/L)
The hydroxylation activity of E. coli (P450 ) cells or cell suspension (1.5 g cdw/L) in KP buffer (50 mM;
pyr
harvested at different time points (Figure 1) was ex- pH 8.0) containing glucose (2%, w/v) at 258C for
amined: cells were resuspended to 1.5 g cdw/L in po- 15 min. The initial velocities at different substrate
tassium phosphate (KP) buffer containing 2% (w/v) concentrations were used to give the plot of 1/[v] vs.
[
12]
glucose and 2 mM N-benzylpyrrolidin-2-one 1 or N- 1/[S],
and the apparent Km and apparent Vmax
benzyloxycarbonylpyrrolidine 3, the mixture was values were obtained from the plot. The kinetic data
shaken at 1200 rpm and 308C for 30 min, the forma- were summarized in Table 1. For either substrate, the
tion of product 2 or 4 (Scheme 1) was analyzed by apparent K for CFE was nearly the same as that for
m
HPLC, and the specific hydroxylation activity was cal- resting cells, suggesting that there is nearly no barrier
culated. As shown in Figure 1, the highest activity was for substrate 1 or 3 to cross the cell membrane. Since
3
9
.8 U/g cdw for the hydroxylation of substrate 1 and 1 g cdw contains ca. 0.5 g protein, the apparent Vmax
.9 U/g cdw for the hydroxylation of substrate 3, both based on the protein amount for resting cells is nearly
with cells harvested at 5 h. At this time point, cell the same as that for CFE for either hydroxylation.
growth had reached the late exponential phase. The Thus, resting cells showed nearly the same catalytic
high hydroxylation activities of E. coli (P450 ) dem- efficiency as CFE for these hydroxylations. The cata-
pyr
3382
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3380 – 3390

Regio- and Stereoselective Biohydroxylations with a Recombinant Escherichia coli
Table 1. Kinetic data of biohydroxylation of 1 and 3 with CFE and resting cells of E. coli (P450pyr), respectively.
[a]
À1
Biohydroxylation
Biocatalyst (g prot L )
or (g cdw L )
K , app
(mM)
Vmax, app
(mmol min g )
Vmax/Km, app
(mmol min g mM )
m
À1 [b]
À1 À1 [c]
À1 À1
À1 c)
From 1 to 2
From 1 to 2
From 3 to 4
From 3 to 4
CFE 1.5
Cells 1.5
CFE 1.5
Cells 1.5
2.5
2.7
0.99
1.1
11
5.1
19
11
4.4
1.9
19
10
[
a]
Hydroxylation was performed with 0.2 to 5.0 mM substrate (1 or 3), 2% (w/v) glucose, 5 mM NADH, and 1.5 g protein/L
CFE or 1.5 g cdw/L cell suspension in 1 mL KP buffer (50 mM; pH 8.0) at 258C for 15 min, and product formation was
determined by HPLC.
g protein/L for CFE; g cdw/L for resting cell suspension.
g protein for CFE; g cdw for resting cell suspension.
[
[
b]
c]
lytic efficiency (Vmax/K ) of hydroxylation of pyrroli-
m
dine 3 is about 5 times higher than that for the hy-
droxylation of pyrrolidin-2-one 1, with either resting
cells or CFE as catalysts. This was the result of the
lower K and higher V values of substrate 3.
m
max
Preparation of (S)-N-Benzyl-4-hydroxypyrrolidin-2-
one 2 by Biohydroxylation
To explore the potential of using E. coli (P450 ) for
pyr
the production of (S)-N-benzyl-4-hydroxypyrrolidin-2-
one 2, a set of reactions was performed at 258C for
25 h with N-benzylpyrrolidin-2-one 1 as substrate at
different concentrations and resting cells as biocata-
lyst (5 g cdw/L) in KP buffer (50 mM; pH 8.0) con-
Figure 4. Time course of the formation of (S)-N-benzyl-4-hy-
droxy-pyrrolidin-2-one 2 in the biohydroxylation of N-ben-
taining glucose (2%, w/v). Since the apparent K of
m
N-benzylpyrrolidin-2-one 1 for resting cells was
zylpyrrolidin-2-one 1 with resting cells of E. coli (P450pyr
)
2.7 mM, 5.0 mM and higher substrate concentrations (5 g cdw/L) in KP buffer (50 mM; pH 8.0) containing glu-
were tested for the hydroxylation. As shown in cose (2%, w/v) at 258C and at different substrate concentra-
Figure 4, when 5.0 mM substrate was initially added, tions. 5 mM (^); 10 mM (&); 15 mM (À); 20 mM (ꢁ);
2
5 mM (~).
the reaction was fast within the first 3 h and became
slow afterwards. This is possibly due to the relatively
low substrate concentration in the biotransformation
mixture after 3 h. Increase of the initial substrate con- different time points instead of adding much more
centration led to longer initial period of fast reaction substrate at the beginning of the reaction. Currently,
and higher final product concentration. In cases of we are also investigating the use of a liquid-liquid bi-
using 15, 20, and 25 mM substrate, the reaction rates phasic system as well as growing cells as catalyst to
at the first 6 h were nearly the same, since these con- further improve the productivity for hydroxylation of
centrations are much higher than the apparent Km. N-benzyl pyrrolidin-2-one 1.
Afterwards, the hydroxylation of 25 mM substrate
still showed a good reaction rate from 6 h to 21 h, and
gave 10.8 mM (S)-2 at 25 h. This is a 2.6 times in- Biohydroxylation of N-Substituted Piperidin-2-one 5,
crease of product concentration compared with the Piperidine 7, and Azetidine 9
[
8c]
use of wild-type strain (4.08 mM).
These results
also demonstrated that the P450pyr monooxygenase in To further compare the catalytic performance of the
recombinant E. coli cells is relatively stable, keeps ac- recombinant strain with wide-type Sphingomonas sp.
tivity for a relatively long period, and can survive in HXN-200,
high concentrations of substrate and product. The oxycarbonylpiperidin-2-one 5, N-benzylpiperidine 7,
conversions in these examples amounted to 76%– and N-tert-butoxycarbonylazetidine (Scheme 2)
3%. Nevertheless, higher conversions could be were examined with 1.5 g cdw/L cell suspension of E.
achieved by adding smaller portions of substrate at coli (P450 ) in KP buffer (50 mM; pH 8.0) contain-
hydroxylations
of
N-tert-but-
AHCTUNGTRENNUNG
9
4
pyr
Adv. Synth. Catal. 2010, 352, 3380 – 3390
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3383

Wei Zhang et al.
FULL PAPERS
for 20 h afforded also good conversions to the prod-
ucts, giving 14 mM 6, 17 mM 8, and 21 mM 9, respec-
tively. These results are even better than that
achieved for the hydroxylation of N-benzylpyrrolidin-
2-one 1.
Regio- and Stereoselective Allylic Biohydroxylation
of (À)-b-Pinene 11
Pinocarveol is a metabolite of b-pinene and is widely
[
13]
used in fragrance and flavour industry.
It is pro-
duced by the isolation from essence oils of natural
sources, but its availability is rather limited. There-
fore, the development of chemical and enzymatic
methods to produce pinocarveol from easily available
and low-priced b-pinene has drawn increasing atten-
tion. As chemical methods often produce mixtures of
different products, the development of a green and se-
lective enzymatic method is desirable since it may
produce pinocarveol in high purity and natural identi-
cal form. Thus far, only two biocatalytic systems have
Scheme 2.
ing 2% (w/v) glucose and 2 mM substrate, respective- been reported to directly hydroxylate b-pinene into
[
14]
ly. The formation of the corresponding 4-hydroxypi- pinocarveol. The better result was achieved by bio-
peridin-2-one 6, 4-hydroxypiperidine 8, or 3-hydroxy- transformation of (À)-b-pinene 11 with a Picea abies
azetidine 10 at 30 min was analyzed by HPLC, and suspension culture to give (1R)-trans-pinocarveol 12.
the specific activity was calculated as 8.0, 12, and However, the product concentration was very low and
1
6 U/g cdw for the hydroxylation of 5, 7 and 9, re- the reaction is slow and not clean: only 0.32 mg 12
spectively. While the activity for 7 was lower than was formed, together with other products, in 100 mL
[14b]
[
8d]
that with the wide-type strain (20 U/g cdw), the ac- plant culture suspension after 8 days of incubation.
tivities for 5 and 9 were nearly the same as those with The recombinant E. coli (P450 ) turned out to be
the wide-type strain (7.7 U/g cdw for 5 and 17 U/g a good biocatalyst for the hydroxylation of b-pinene
pyr
[8d,e]
cdw for 9).
11 to (1R)-trans-pinocarveol 12 with much higher pro-
A series of biotransformations of 5, 7 or 9 (5 mM ductivity (Scheme 3). Since b-pinene 11 is a volatile
or 25 mM) were performed in 5 g cdw/L cell suspen- compound, the biotransformation of 11 with resting
sion of E. coli (P450 ) in KP buffer (50 mM; pH 8.0) cells of E. coli (P450 ) was carried out in a shaking
pyr
pyr
containing glucose (2%, w/v). As shown in Table 2, flask with glass stopper and sealed by parafilm with-
hydroxylation of 5 mM substrates for 20 h gave the out taking any samples during the reaction. Parallel
corresponding products 6, 8, and 9 in 88%, 98%, and experiments were performed with several flasks and
99%, respectively. Hydroxylation of 25 mM substrates stopped at different time points. As shown in Table 3,
hydroxylation of 5 mM b-pinene 11 with 5 g cdw/L of
resting cells gave 3.4 mM (1R)-trans-pinocarveol 12 at
5
h, and the specific hydroxylation activity was 5.3 U/
Table 2. Selective biohydroxylation of N-substituted piperi-
dinone 5, piperidine 7, and azetidine 9 with E. coli (P450pyr).
g cdw within the first 30 min. Further increase of sub-
strate concentration to 10 mM did not improve the
final product concentration. Instead, better result was
achieved by incubating 5 mM substrate in a cell sus-
pension of 10 g cdw/L. 4.1 mm trans-pinocarveol 12
was formed within 5 h, with a yield of 82%. This
[a]
Substrate (mM) Product
Conversion [%]
0
.5h 1h 2h 5h 20h
5
5
7
7
9
9
(5.0)
(25)
(5.0)
(25)
(5.0)
(25)
6
6
8
8
10
10
9.3
5.0
21
6.8
33
22
8.0 14
35
13
54
17
45
68
24
92
40
96
47
88
56
98
68
99
83
59
20
70
31
9.1
[
a]
Biotransformation was performed with substrate 5, 7 and
9
1
in 5 g cdw/L of cell suspension of E. coli (P450pyr) in
0 mL KP buffer (50 mM; pH 8.0) containing glucose
(2%, w/v) at 300 rpm and 258C for 20 h.
Scheme 3.
3384
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3380 – 3390

Regio- and Stereoselective Biohydroxylations with a Recombinant Escherichia coli
Table 3. Regio- and stereoselective biohydroxylation of (À)-
b-pinene 11 with E. coli (P450 ).
pyr
[a]
Entry Substrate 11 Cell density
Product 12 at 5h
Conc. (mM)
A
H
U
G
R
N
U
G
Conc. (mM) Conv. (%)
1
2
3
5.0
10
5.0
5.0
5.0
10
3.4
2.9
4.1
68
29
82
[
a]
Biotransformation was performed with substrate 11 in
cell suspension of E. coli (P450pyr) in 10 mL KP buffer
(50 mM; pH 8.0) containing glucose (2%, w/v) at
3
00 rpm and 258C for 5 h.
Scheme 4.
with the possibility for direct hydroxylation of the
cheaper norbornane 13 into norbornaneol, but gave
mixtures of endo- and exo-norbornaneol and norbor-
nanone.
We examined the potential of hydroxylation of nor-
bornane 13 with E. coli (P450 ) (Scheme 4). Here
pyr
again, the biotransformation was performed with rest-
ing cells in closed system with several parallel shaking
flasks. As shown in Table 4 (entry 1), 1.2 mM exo-nor-
bornaneol 14 was formed after 5 h incubation of
5
mM norbornane 13 in a cell suspension of 5 g cdw/
L, with an exo/endo-selectivity of 95%. When increas-
ing the cell density to 10 g cdw/L (entry 4), 1.6 mM
exo-norbornaneol 14 was obtained with an exo/endo-
selectivity of 95%. Increasing the initial substrate con-
centration to 20 mM (entry 3) resulted in an increase
of the final product concentration to 3.2 mM at 5 h,
with a specific activity within the first 30 min of 6.0 U/
Figure 5. GC chromatograms of samples taken from biohy-
droxylation of (À)-b-pinene 11 (5 mM) in 10 mL cell suspen-
sion (10 g cdw/L) in KP buffer (50 mM; pH 8.0) containing
glucose (2%, w/v) at 300 rpm and 258C. A) 0 min; B) 5 h.
product concentration is over 200 times higher than g cdw. These results suggested E. coli (P450 ) is a
pyr
that obtained with the best biocatalytsts known thus much better catalyst than any other known catalytic
[14b]
far.
As shown in Figure 5, there is only one prod- system for the hydroxylation of norbornane 13 to nor-
uct peak at 12.6 min in the GC chromatogram for the bornaneol 14, with very good exo/endo selectivity.
sample taken at 5 h. This result demonstrated the
clean biohydroxylation of b-pinene 11. E. coli
(
P450 ) is thus definitely much better than any other Regioselective Biohydroxylation of Tetralins 15 and
pyr
known catalyst for the biohydroxylation of pinene to 17
trans-pinocarveol, with high yield, high product con-
centration, and clean reaction.
2-Tetralols 16 and 18 are useful synthetic intermedi-
ates. For instance, they can be used to prepare phar-
macologically active 2-aminotetralins which show af-
[
17]
Stereoselective Biohydroxylation of Norbornane 13
finity to the melatonin receptor. Several synthetic
routes have been developed via the reduction of 2-tet-
[
18a–c]
[18d]
Both endo- and exo-norbornaneol are high-value ralone
or cycloalkylation of arylalkyl epoxide.
compounds. While the former is a key synthon for the However, no direct hydroxylation of the easily acces-
[
15a]
preparation of novel cannabionoid analogues
potassium-channel openers,
and sible and much cheaper tetralins has been reported
the latter serves as an for the preparation of 2-tetralols, either by chemical
intermediate for the production of A adenosine re- or enzymatic method.
[
15b]
1
[15c]
ceptor agonists.
The synthesis of norbornaneol has
been widely studied in recent years, mainly via the such type of reaction. Hydroxylation of tetralin 15
The recombinant E. coli (P450 ) was explored for
pyr
[
16]
[
16a–e]
reduction of 2-norbornanone.
phyrin
Only ironAHCUTNGTERNNUNG
[16g]
[16f]
and a purified liver P450
Adv. Synth. Catal. 2010, 352, 3380 – 3390
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3385

Wei Zhang et al.
FULL PAPERS
Table 4. Selective biohydroxylation of norbornane 13, tetralin 15, and 6-methoxytetralin 17 with E. coli (P450pyr).
[a]
Entry
Substrate Substrate concen- Cell density Product Product concentration Conversion Hydroxylation
tration (mM)
A
H
U
G
R
N
N
(g cdw/L)
at 5h (mM)
at 5h (%)
selectivity (%)
[
b]
b]
b]
b]
c]
c]
c]
d]
d]
d]
1
2
3
4
5
6
7
8
9
13
13
13
13
15
15
15
17
17
17
5.0
10
20
5.0
5.0
10
5.0
5.0
10
5.0
5.0
5.0
10
5.0
5.0
10
5.0
5.0
10
14
14
14
14
16
16
16
18
18
18
1.2
1.9
3.0
1.6
1.1
1.0
1.7
1.2
1.2
2.1
24
19
15
32
22
10
34
24
12
42
95[
95[
95[
95
[
84[
83
{
84[
83
[
84[
1
0
5.0
83
[
a]
Biotransformation was performed with substrate 13, 15 or 17 in cell suspension of E. coli (P450pyr) in 10 mL KP buffer
(50 mM; pH 8.0) containing glucose (2%, w/v) at 300 rpm and 258C for 5 h.
b]
[
[
[
The exo/endo selectivity.
c]
The 2-tetralol/1-tetralol selectivity.
The 7-methoxy-2-tetralol/6-methoxy-2-tetralol selectivity.
d]
atoms. As shown in Table 4 (entries 5 to 10), 2-tetra- achieved by protein engineering as well as process en-
lols were formed as the major product with 83–84% gineering.
regioselectivity and the specific hydroxylation activity
reached 5.1 U/g cdw and 7.3 U/g cdw for 2-tetralols
1
6 and 18, respectively. Hydroxylation of 5 mM sub- Conclusions
strate 15 at the cell density of 10 g cdw/L gave 2-tetra-
lol 16 in 1.7 mM at 5 h. 2.1 mM 7-methoxy-2-tetralol A recombinant E. coli expressing cytochrome P450pyr
1
8 was formed within 5 h by hydroxylation of 5 mM monooxygenase from Sphingomonas sp. HXN-200
substrate 17 in a cell suspension of 10 g cdw/L. In all has been engineered, with specific activity of 4.1 U/g
the cases, no ketone was formed according to GC cdw and 9.9 U/g cdw for the hydroxylation of N-ben-
analysis. The identification of bioproduct 16 was zylpyrrolidin-2-one 1 and N-benzyloxycarbonylpyrro-
straightforward by comparing its retention time in lidine 3, respectively. E. coli (P450 ) grows easily,
pyr
GC chromatogram with that of the commercially demonstrates a similar high hydroxylation activity as
available standard 16. On the other hand, the identifi- the wild-type strain, and shows no side reaction. Hy-
cation of the bioproduct from hydroxylation of 6-me- droxylation of N-benzylpyrrolidin-2-one 1 with resting
thoxytetralin 17 was difficult, since the desired prod- cells of E. coli (P450 ) produces 10.8 mM (S)-N-
pyr
uct 7-methoxy-2-tetralol 18 is not commercially avail- benzyl-4-hydroxypyrrolidin-2-one 2, a 2.6-fold in-
able, and the hydroxylation may create theoretically crease of the product concentration compared with
four different regioisomers. Fortunately, all other the wild-type strain. The hydroxylations of N-tert-bu-
three regioisomers are available, and they have differ- toxycarbonylpiperidin-2-one 5, N-benzylpiperidine 7,
ent retention times in GC analysis: 21.1 min for 7-me- and N-tert-butoxycarbonylazetidine 9 are also highly
thoxy-1-tetralol, 22.5 min for 6-methoxy-1-tetralol, regioselective and active. Moreover, hydroxylation of
and 22.8 min for 6-methoxy-2-tetralol, respectively. (À)-b-pinene 11 with E. coli (P450 ) cells shows ex-
pyr
The bioproduct from hydroxylation of 17 has a reten- cellent regio- and stereoselectivity and gives (1R)-
tion time of 21.3 min which is different from those of trans-pinocarveol 12 in 82% yield and 4.1 mM, thus
the three known isomers, thus it is a different com- being over 200 times higher than that obtained with
pound. In addition, GC-MS analysis of the bioproduct the best biocatalyst known thus far. E. coli (P450 ) is
pyr
showed an MW of 178, corresponding to the mass of also able to hydroxylate other types of substrates with
a hydroxylated 6-methoxytetralin. Therefore, the unique selectivity: hydroxylation of norbornane 13
structure of the bioproduct was deduced to be 7-me- gives exo-norborneol 14 with exo/endo selectivity of
thoxy-2-tetralol 18. Although the product concentra- 95%, and hydroxylayion of tetralin 15 and 6-methoxy-
tions and conversions of these hydroxylations are rel- tetralin 17 demonstrates, for the first time, a regiose-
atively low, our results represent the first example of lectivity of 83–84% at the non-activated 2-position.
hydroxylation of tetralins on the non-activated carbon
atoms at 2-position. Further improvement of the pro-
ductivity of P450pyr for these hydroxylations might be
3386
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3380 – 3390

Regio- and Stereoselective Biohydroxylations with a Recombinant Escherichia coli
Experimental Section
The concentrations of (À)-b-pinene 11, trans-pinocarveol
2, norbornane 13, exo-norbornaneol 14, tetralin 15, 2-tetra-
1
lol 16, 6-methoxytetralin 17, 7-methoxy-2 tetralol 18 and
their corresponding products were analyzed by an Agilent
GC 6890 on an Agilent HP-5 column (25 m by 0.32 mm)
with inlet temperature of 2808C and detector temperature
of 3008C; temperature program: 608C increase to 1008C at
Chemicals
Ampicillin (>99%), kanamycin solution (50 mg/mL),
NADH (>99%), N-benzylpyrrolidin-2-one 1 (98%), (À)-b-
pinene (ꢀ99%) 11, norbornane 13 (98%), tetralin 15 (99%),
(
À)-trans-pinocarveol (ꢀ96%), exo-norborneol 14 (98%),
À1
a rate of 58Cmin , increase to 1168C at a rate of
endo-norborneol (96%), and dl-dithiothreitol (ꢀ99%) were
purchased from Sigma–Aldrich. 2-Tetralol 16 (97%) was ob-
tained from Acros Organics. 6-Methoxy-2-tetralol (>96%)
was purchased from JW-Pharmlab. 6-Methoxy-1-tetralol
À1
2
8Cmin , hold at 1168C for 2 min, then increase to 2808C
À1
at a rate of 308Cmin , and hold at 2808C for 3 min; reten-
tion times: 4.2 min for 13, 7.8 min for 14, 7.9 min for 11,
8.0 min for endo-norborneol, 12.6 min for 12, 13.5 min for
15, 19.9 min for 1-tetralol, 20.4 min for 16, 20.9 min for 17,
21.1 min for 7-methoxy-1-tetralol, 21.3 min for 18, 22.3 min
(
98%) and 7-methoxy-1-tetralol (98%) were bought from
Aurora Building Blocks. Isopropyl b-d-thiogalactopyrano-
side (ITPG,>99%) was bought from 1 BASE. 6-Methoxy-
tetralin 17 (85%, Sigma) was further purified by column
chromatography on silica gel (R =0.32, n-hexane/ethyl ace-
st
for n-hexadecane (internal standard), 22.5 min for 6-me-
thoxy-1-tetralol, and 22. 8 min for 6-methoxy-2-tetralol.
f
tate, 40:1) to a purity of >98% (GC). N-Benzyloxycarbonyl-
[
8b]
[8e]
pyrrolidine 3,
N-tert-butoxycarbonylpiperidin-2-one 5,
[8d]
N-benzylpiperidine 7, and N-tert-butoxycarbonylazetidine
[
8d]
Genetic Engineering of E. coli (P450pyr)
9
were prepared according to the published procedures.
[10]
Plasmid pCom8-P F200R1500 was used as a PCR template
A7
Strain and Biochemicals
to amplify the coding region of cytochrome P450pyr (P ),
ferredoxin (F200), and ferredoxin reductase (R1500). The
P450pyr gene was amplified using the forward primer P450fwd
A7
Escherichia coli BL21 (DE3), DH5a, plasmids pETDuet
and pRSFDuet were purchased from Novagen. Failsafe
PCR 2X Premix G was purchased from Epicentre Biotech-
nologies. Phusion DNA polymerase, T4 DNA ligase and the
restriction enzymes NcoI, NdeI, KpnI, and AflII were pur-
chased from New England Biolabs. Medium components
tryptone and yeast extract were purchased from Biomed Di-
agnostics.
(
3
5’-TTA ACT ACT CCA TGG AAC ATA CAG GAC AAA GCG CGG-
’, the NcoI restriction site is underlined) and reverse
primer P450rev (5’-TTA ACT ACT GGT ACC CTA CGC GTG GAC
GCG AAC- 3’, the KpnI restriction site is underlined). The
F200 gene was amplified using the forward primer Fdxfwd (5’-
TTA ACT ACT CCA TGG ca ACA GTG ACC TAT GTT GAA ATA
AAT GGC AC-3’, the NcoI restriction site is underlined, the
second codon of the F200 gene, in small caps, was changed
from the original cca to gca) and reverse primer Fdxrev (5’-
TTA ACT ACT CTT AAG TCA ATG CTG CGC GAG AGG AAG- 3’,
the AflII restriction site is underlined). The R1500 gene was
amplified using the forward primer R1500fwd (5’-TTA ACT
ACT CAT ATG ATC CAC ACC GGC GTG AC-3’, the NdeI re-
Analytical Methods
The concentrations of N-benzylpyrrolidin-2-one 1, N-benzyl-
4-hydroxypyrrolidin-2-one 2, N-benzyloxycarbonylpyrroli-
dine 3, N-benzyloxycarbonyl-3-hydroxypyrrolidine 4, N-tert-
butoxycarbonylpiperidin-2-one 5, N-tert-butoxycarbonyl-4-
hydroxypiperidin-2-one 6, N-benzylpiperidine 7, N-benzyl-4-
hydroxypiperidine 8, N-tert-butoxycarbonylazetidine 9, and
N-tert-butoxycarbonyl-3-hydroxyazetidine 10 were analyzed
via Shimadzu Prominence HPLC on a Hypersil BDS-C18
column (125 mmꢁ4 mm) at 258C; UV detection: 210 nm;
striction site is underlined) and the reverse primer R1500rev
(
3
5’-TTA ACT ACT GGT ACC TTA GAG GGA GGT TGG GGA CG-
’, the KpnI restriction site is underlined). PCR amplified
genes were digested with their respective enzymes. The di-
gested P450pyr fragment was ligated into the pRSFDuet
vector that was digested with NcoI and KpnI restriction en-
zymes. The digested F200 fragment was ligated into the mul-
tiple cloning site (MCS) 1 of the pETDuet vector that was
digested with NcoI and AflII. The resulting plasmids from
the ligations were separately transformed into chemical
competent E. coli DH5a cells which were then plated on
À1
flow rate: 1 mLmin ; eluent: acetonitrile/water (20:80), re-
tention time: 2.8 min for 2, and 7.8 min for 1; eluent: aceto-
nitrile/water (25:75), retention time: 2.7 min for 10, 4.7 min
for 4, 9.2 min for 9, and 11.4 min for 3; eluent: acetonitrile/
water (30:70), retention time: 1.8 min for 6, and 4.3 min for
5
; eluent: acetonitrile/water (15:85), retention time: 3.0 min
À1
Luria-Bertani (LB) agar plates containing 50 mgmL kana-
for 8, and 5.2 min for 7.
À1
mycin (for pRSFDuet P450pyr) and 100 mgmL ampicillin
The ees of N-benzyl-4-hydroxypyrrolidin-2-one 2, N-ben-
zyloxycarbonyl-3-hydroxypyrrolidine 4, and N-tert-butoxy-
carbonyl-4-hydroxypiperidin-2-one 6 were determined on a
Shimadzu Prominence HPLC at 258C. UV detection:
(for pETDuet F200), respectively. Subsequently, the digested
R
1500 fragment was ligated into the MCS 2 of the same pET-
Duet vector, containing the F200 gene, digested with NdeI
and KpnI restriction enzymes. The resulting plasmid from
the ligation was transformed into chemical competent E.
2
2
10 nm; for 2 and 4: Chiralpak AS-H column (150 mmꢁ
.1 mm); flow rate: 1 mLmin ; eluent: n-hexane/2-propanol
À1
(
(
1
4:1), retention time: 20.3 min for (S)-2 and 30.5 min for
R)-2; eluent: n-hexane/2-propanol (10:1), retention time:
coli DH5a cells which were then plated on LB agar plates
À1
containing 100 mgmL
ampicillin. Finally, the purified
3.5 min for (S)-4, and 15.4 min for (R)-4; for 6: Chiralcel
pRSFDuet P450pyr and pETDuet F200 plasmids were trans-
À1
OB-H column (250 mmꢁ4.6 mm); flow rate: 0.5 mLmin ;
eluent: n-hexane/2-propanol (95:5); retention time: 28.3 min
for (S)-6, and 30.8 min for (R)-6.
formed into electrocompetent E. coli BL21 (DE3) and
À1
plated on LB agar plates containing 50 mgmL kanamycin
À1
and 100 mgmL ampicillin.
Adv. Synth. Catal. 2010, 352, 3380 – 3390
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3387

Wei Zhang et al.
FULL PAPERS
Growth and Hydroxylation Activity of E. coli
for the molar extinction increment at 450 nm for cyto-
chrome P450 was used to calculate the P450 concentra-
(
P450 )
pyr
pyr
[11]
tion.
The recombinant E. coli (P450pyr) was inoculated on an LB
agar plate (10 g tryptone, 5 g yeast extract, and 5 g NaCl in
Kinetics of Biohydroxylation of N-Benzylpyrrolidin-2-
one 1 and N-Benzyloxycarbonylpyrrolidine 3 with
1
L deionized water with 2% agar) containing ampicillin
À1
À1
(
100 mgmL ) and kanamycin (50 mgmL ) and grown over-
night at 378C. A single colony from the LB agar plate was CFE or Resting Cells of of E. coli (P450pyr
)
inoculated into 30 mL of LB medium containing ampicillin
À1
À1
To 925 to 933 mL of the CFE (1.5 g protein/L) of E. coli
(
3
100 mgmL ) and kanamycin (50 mgmL ) and grown at
00 rpm and 378C for 12 h, giving an OD600 of 4.0 (1.6 g
(
P450pyr) in KP buffer (50 mM; pH 8.0) in a 2 mL Eppen-
dorf tube were added 40 mL of glucose stock solution (50%,
w/v), 25 mL of NADH stock solution (0.2M), and 2 to 10 mL
of stock solution (0.2M or 0.5M in methanol) of N-benzyl-
pyrrolidin-2-one 1 or N-benzyloxycarbonylpyrrolidine 3 to a
final volume of 1 mL. The mixtures were shaken at
cdw/L). 2 mL of this preculture were added in 100 mL LB
À1
medium containing ampicillin (100 mgmL ) and kanamycin
À1
(
3
50 mgmL ), and the mixture was shaken at 300 rpm and
78C. IPTG (0.5 mM) was added at 2 h, the temperature
was changed to 308C, and the mixture was shaken at
00 rpm. Samples were taken at different time points for
OD measurement, activity test, and CFE preparation. The
cells were harvested after 3 h induction with an OD600 of 4.0
1.6 g cdw/L) by centrifugation at 10,375 g for 10 min. The
cell pallets were used for the activity test and biotransforma-
tion.
To determine the specific hydroxylation activity of the re-
combinant strain, the above freshly harvested cells were re-
suspended in KP buffer (50 mM; pH 7.5) to a density of
.5 g cdw/L. To 950 mL of the cell suspension in a 2 mL Ep-
pendorf tube were added 40 mL of glucose stock solution
50%, w/v) and 10 mL of stock solution of N-benzylpyrroli-
din-2-one 1 or N-benzyloxycarbonylpyrrolidine 3 (0.2M in
methanol). The mixture was shaken at 1200 rpm and 308C
with a Bioer mixing block MB-102 for 30 min and then
mixed with 1 mL cold methanol to quench the reaction. An
analytical sample was prepared by centrifugation, and the
1
1
200 rpm and 258C with for 15 min and then mixed with
mL cold methanol to quench the reaction. Analytical sam-
3
ples were prepared using the same method described above
and analyzed by HPLC. The initial velocities were calculat-
ed from the product 2 and 4 formed within the first 15 min,
(
and the plot 1/[v] vs. 1/[S] was made to determine K and
m
[12]
Vmax
.
Biohydroxylation of 1 or 3 with resting cells of E. coli
P450pyr) was performed for 15 min with 950 to 958 mL of
(
cell suspensions (1.5 g cdw/L) in KP buffer (50 mM;
pH 8.0), 40 mL of glucose stock solution (50%, w/v), and 2
to 10 mL of stock solution (0.2M or 0.5M in methanol) of 1
or 3. The product formation was determined by HPLC, the
initial velocities with different substrate concentrations were
calculated, and the plot 1/[v] vs. 1/[S] was used to establish
1
(
K and Vmax.
m
aqueous solution was analyzed by reverse phase HPLC to General Procedure for the Biohydroxylation of N-
determine the concentration of product 2 or 4. The specific Benzylpyrrolidin-2-one 1 with Resting Cells of E. coli
hydroxylation activity was expressed in U/g cdw, and 1 U
(
P450 )
pyr
was defined as the formation of 1 mmol product per min.
To 10 mL cell suspensions (5 g cdw/L) of E. coli (P450pyr) in
KP buffer (50 mM; pH 8.0) were added 400 mL of glucose
stock solution (50%, w/v) and 8.8 to 43.8 mg N-benzylpyrro-
lidin-2-one 1 (final concentration of 5 to 25 mM). The mix-
ture was shaken at 258C and 300 rpm. An appropriate
amount of 1.0 M NaOH solution was added at several time
points to keep pH around 8.0. At different time points, ali-
quots (100 mL) were taken, mixed with cold methanol
Protein Gel and CO Difference Spectrum of CFE of
E. coli (P450pyr)
5
mL cell suspension (10 g cdw/L) of above freshly harvest-
ed cells in KP buffer (50 mM; pH 7.5) were passed through
a homogenizer (Constant Cell Disruption System) twice at
2
1
0 kpsi. The cell debris was removed by centrifugation at
3,000 g at 48C for 30 min. The supernatant was diluted 10
(
100 mL), and subjected to centrifugation. The supernatants
were analyzed by HPLC to determine the product concen-
tration.
times with KP buffer (50 mM; pH 7.5), and the protein con-
centration was determined by Bradford protein content
[19]
assay with bovine serum albumin (BSA) as standard.
SDS-PAGE was performed by loading 15 mg CFE on a gel
containing 0.1% sodium dodecyl sulfate and 14% acrylam-
ide, staining with a 0.1% solution of Commassie brilliant
blue R-250 in methanol/acetic acid/water (4:1:5; v/v/v), and
destaining by soaking in deionized water overnight.
General Procedure for the Biohydroxylation of (À)-
b-Pinene 11 with Resting Cells of E. coli (P450 )
pyr
To 5 different 100 mL shaking flasks containing 10 mL cell
suspensions (5 to 10 g cdw/L) of E. coli (P450pyr) in KP
buffer (50 mM; pH 8.0) were added 400 mL of glucose stock
solution (50%, w/v) and 6.8–13.6 mg (À)-b-pinene 11 (final
concentration of 5–10 mM). The flasks were then covered
with glass stoppers and tightly sealed with parafilm. The
mixtures were shaken at 300 rpm and 258C. Individual
flasks were removed from the shaker at different time
points, put into ice for 10 min, opened, and mixed with same
volume of ethyl acetate containing 2 mM n-hexadecane. The
flask was kept on ice for another 10 min, mixed violently on
The CO difference spectra of CFE of E. coli (P450pyr
)
were measured at room temperature (258C) in a Hitachi
UV-Vis ratio beam spectrophotometer U-1900, with the use
of quartz cuvette with a 1 cm light path. 3 mL CFE (1.7 g
protein/L) in KP buffer (50 mM; pH 7.5) in the quartz cuv-
ette were reduced by addition of a few milligrams of solid
dithiothreitol and saturated with CO by bubbling the gas for
about 2 min. The absorbance at 450 nm of CFE was mea-
À1
sured before and after the CO bubbling. A value of 91 cm
3388
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3380 – 3390

Regio- and Stereoselective Biohydroxylations with a Recombinant Escherichia coli
vortex, and followed by phase separation. The organic phase
was dried over anhydrous Na SO and analyzed by GC to
determine the product concentration.
Feichtenhofer, H. Griengl, I. Kopper, A. Lehmann, H.
Weber, Angew. Chem. 1999, 111, 2946–2949; Angew.
Chem. Int. Ed. 1999, 38, 2763–2765; j) A. Raadt, H.
Griengl, H. J. Weber, Chem. Eur. J. 2001, 7, 27–31.
[3] a) K. E. Liu, C. C. Johnson, M. Newcomb, S. J. Lippard,
J. Am. Chem. Soc. 1993, 115, 939–947; b) M. Merkx,
D. A. Kopp, M. H. Sazinsky, J. L. Blazyk, J. Muller, S. J.
Lippard. Angew. Chem. 2001, 113, 2860–2888; Angew.
Chem. Int. Ed. 2001, 40, 2782–2807.
2
4
General Procedure for the Biohydroxylation of
Norbornane 13 with Resting Cells of E. coli (P450pyr)
The biotransformation of norbornane 13 at 5 mM (4.8 mg,
0.05 mmol) to 60 mM (57.6 mg, 0.6 mmol) with resting cells
of E. coli (P450pyr) was performed according to the same
procedure as for the hydroxylation of (À)-b-pinene 11 de-
scribed above.
[4] a) J. B. van Beilen, J. Kingma, B. Witholt, Enzyme
Microb. Technol. 1994, 16, 904–911; b) R. T. Ruetting-
er, G. R. Griffith, M. J. Coon, Arch. Biochem. Biophys.
1
967, 183, 528–537; c) J. A. Peterson, D. Basu, M. J.
General Procedure for the Biohydroxylation of
Tetralin 15 and 6-Methoxytetralin 17 with Resting
Cells of E. coli (P450pyr)
Coon, J. Biol. Chem. 1966, 241, 5162–5164.
[5] a) L. L. Wong, Curr. Opin. Chem. Biol. 1998, 2, 263–
268; b) D. Werck-Reichhart, R. Feyereisen, Adv.
Genome Biol. 2000, 1, 3003; c) T. L. Poulos, B. C.
Finzel, A. J. Howard, J. Mol. Biol. 1987, 195, 687–700;
d) K. G. Ravichandran, S. S. Boddupalli, C. A. Hase-
mann, J. A. Peterson, J. Deisenhofer, Science 1993, 261,
To 10 mL suspensions of resting cells (5 to 10 gcdw/L) of E.
coli (P450pyr) in KP buffer (50 mM; pH 8.0) were added
400 mL of glucose stock solution (50%, w/v) and 6.6–13.2 mg
tetralin 15 or 8.1–16.2 mg 6-methoxytetralin 17 (final con-
centration of 5–10 mM). The mixtures were shaken at
7
1
31–736; e) L. O. Narhi, A. J. Fulco, J. Biol. Chem.
986, 261, 7160–7169; f) H. Koga, B. Rauchfuss, I. C.
300 rpm and 258C. Aliquots (200 mL) were taken out at dif-
Gunsalus, Biochem. Biophys. Res. Commun. 1985, 130,
12–417; g) O. Narhi, L. P. Wen, A. J. Fulco, Mol. Cell.
ferent time points, mixed with 200 mL of ethyl acetate con-
taining 2 mM n-hexadecane as internal standard, and centri-
fuged. The organic phase of the supernatants were separat-
ed, dried over anhydrous Na SO , and analyzed by GC to
4
Biochem. 1988, 79, 63–71; h) J. B. van Beilen, R. Hol-
tackers, D. Luscher, U. Bauer, B. Witholt, W. A. Duetz,
Appl. Environ. Microbiol. 2005, 71, 1737–1744; i) L. P.
Wen, A. J. Fulco, J. Biol. Chem. 1987, 262, 6676–6687;
j) T. Mouri, J. Michizoe, H. Ichinose, N. Kamiya, M.
Goto, Appl. Microbiol. Biotechnol. 2006, 72, 514–520.
6] a) M. W. Peters, P. Meinhold, A. Glieder, F. H. Arnold,
J. Am. Chem. Soc. 2003, 125, 13442–13450; b) E. T.
Farinas, U. Schwaneberg, A. Glieder, F. H. Arnold,
Adv. Synth. Catal. 2001, 343, 601–606; c) J. P. Jones,
E. J. OꢂHare, L. L. Wong, Eur. J. Biochem. 2001, 268,
2
4
determine the product concentration.
[
Acknowledgements
This work was financially supported by Science & Engineer-
ing Research Council of A*STAR, Singapore, through a re-
search grant (project No. 0621010024) and Singapore-MIT
Alliance through a CPE Flagship research program.
1
460–1467; d) R. Weis, M. Winkler, M. Schittmayer, S.
Kambourakis, M. Vink, J. D. Rozzell, A. Glieder, Adv.
Synth. Catal. 2009, 351, 2140–2146; e) M. Landwehr, L.
Hochrein, C. R. Otey, A. Kasrayan, J. E. Bꢃckvall,
F. H. Arnold, J. Am. Chem. Soc. 2006, 128, 6058–6059;
f) T. Kubo, M. W. Peters, P. Meinhold, F. H. Arnold,
Chem. Eur. J. 2006, 12, 1216–1220.
References
[1] a) J. A. Labinger, J. E. Bercaw, Nature 2002, 417, 507–
[
7] a) A. B. Carmichael, L. L. Wong, Eur. J. Biochem. 2001,
5
14; b) M. Costas, K. Chen, L. Que Jr, Coord. Chem.
2
68, 3117–3125; b) C. F. Harford-Cross, A. B. Carmi-
Rev. 2000, 200–202, 517–544; c) A. E. Shilov, G. B.
Shul’pin, Activation and Catalytic Reactions of Saturat-
ed Hydrocarbons in the Presence of Metal Complexes,
Kluwer Academic, Dordrecht, 2000; d) A. Sen, Acc.
Chem. Res. 1998, 31, 550–557; e) A. E. Shilov, G. B.
Shul’pin, Chem. Rev. 1997, 97, 2879–2932; f) D. H. R.
Barton, D. Doller, Acc. Chem. Res. 1992, 25, 504–512.
2] a) Z. Li, D. Chang, Curr. Org. Chem. 2004, 8, 1647–
chael, F. K. Allan, P. A. England, D. A. Rouch, L. L.
Wong, Protein Eng. 2000, 13, 121–128; c) Q. S. Li, J.
Ogawa, R. D. Schmid, S. Shimizu, Appl. Environ. Mi-
crobiol. 2001, 67, 5735–5739.
[
8] a) Z. Li, H. J. Feiten. , J. B. van Beilen, W. Duetz, B.
Witholt, Tetrahedron: Asymmetry 1999, 10, 1323–1333;
b) Z. Li, H. J. Feiten, D. Chang, W. A. Duetz, J. B. van
Beilen, B. Witholt, J. Org. Chem. 2001, 66, 8424–8430;
c) D. Chang, B. Witholt, Z. Li, Org. Lett. 2000, 2, 3949–
3952; d) D. Chang, H. J. Feiten, K. H. Engesser, J. B.
van Beilen, B. Witholt, Z. Li, Org. Lett. 2002, 4, 1859–
1862; e) D. Chang, H. J. Feiten, B. Witholt, Z. Li, Tetra-
hedron: Asymmetry 2002, 13, 2141–2147.
[
1
658; b) Z. Li, J. B. Beilen, W. A Duetz, A. Schmid, A.
Raadt, H. Griengl, B. Witholt, Curr. Opin. Chem. Biol.
2
2
002, 6, 136–144; c) S. G. Burton, Trends Biotechnol.
003, 21, 543–549; d) A. Raadt, H. Griengl, Curr.
Opin. Biotechnol. 2002, 13, 537–542; e) H. L Holland,
H. K Weber, Curr. Opin. Biotechnol. 2000, 11, 547–
5
53; f) H. L. Holland, Steroids 1999, 64, 178–186; g) Y.
[9] a) M. Miyauchi, R. Endo, M. Hisaoka, H. Yasuda, I.
Kawamoto, J. Antibiot. 1997, 50, 429–439; b) J. F. Al-
meida, J. Anaya, N. Martin, M. Grande, J. R. Moran,
M. C. Caballero, Tetrahedron: Asymmetry 1992, 3,
1431–1440.
Chen, J. Xu, X. Xu, Y. Xia, H. Lin, S. Xia, L. Wang,
Tetrahedron: Asymmetry 2007, 18, 2537–2540; h) H. L.
Holland , T. A. Morris , P. J. Nava, M. Zabic, Tetrahe-
dron 1999, 55, 7441–7460; i) G. Braunegg, A. Raadt, S.
Adv. Synth. Catal. 2010, 352, 3380 – 3390
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3389

Wei Zhang et al.
FULL PAPERS
[
10] J. B. van Beilen, E. G. Funhoff, A. Van Loon, A. Just,
L. Kaysser, M. Bouza, R. Holtackers, M. Rçthlisberger,
Z. Li, B. Witholt, Appl. Environ. Microbiol. 2006, 72,
J. Org. Chem. 2001, 2235–2243; c) P. D. de Koning, M.
Jackson, I. C. Lennon, Org. Process Res. Dev. 2006, 10,
1054–1058; d) I. Kamiya, A. Ogawa, Tetrahedron Lett.
2002, 43, 1701–1703; e) F. Yoshizako, A. Nishimura,
M. Chubachi, M. Kirihata, J. Ferment. Bioeng. 1996, 82,
601–603; f) T. G. Traylor, K. W. Hill, W. P. Fann, S.
Tsuchiya, B. E. Dunlap, J. Am. Chem. Soc. 1992, 114,
1308–1312; g) J. T. Groves, G. A. McClusky, R. E.
White, M. J. Coon, Biochem. Biophys. Res. Commun.
1978, 81, 154–160.
5
9–65.
11] T. Omura, R. Sato, J. Biol. Chem. 1964, 239, 2379–
385.
12] H. Lineweaver, D. Burk, J. Am. Chem. Soc. 1934, 56,
58–666.
[
[
2
6
[
[
13] J. Nowicki, Molecules 2000, 5, 1033–1050.
14] a) D. Busmann, R. G. Berger, Z. Naturforsch. C:
Biosci. 1994, 49, 545–552; b) M. Lindmark-Henriksson,
D. Isaksson, T. Vanek, I. Valterova, H. E. Hogberg, K.
Sjodin, J. Biotechnol. 2004, 107, 173–184.
15] a) R. I. Duclos Jr, D. Lu, J. Guo, A. Makriyannis, Tetra-
hedron Lett. 2008, 49, 5587–5589; b) M. Eda, T. Take-
moto, S. Ono, T. Okada, K. Kosaka, M. Gohda, S.
Matzno, N. Nakamura, C. Fukaya, J. Med. Chem. 1994,
[17] S. Copinga, P. G. Tepper, C. J. Grol, A. S. Horn, M. L.
Dubocovich, J. Med. Chem. 1993, 36, 2891–2898.
[18] a) G. de Gonzalo, I. Lavandera, K. Durchschein, D.
Wurm, K. Faber, W. Kroutil, Tetrahedron: Asymmetry
2007, 18, 2541–2546; b) F. Zaccheria, N. Ravasio, M.
Ercolic, P. Allegrini, Tetrahedron Lett. 2005, 46, 7743–
7745; c) J. S. Yadav, G. S. K. K. Reddy, G. Sabitha,
A. D. Krishna, A. R. Prasad, H. U. R. Rahaman, K. V.
Rao, A. B. Rao, Tetrahedron: Asymmetry 2007, 18,
717–723; d) J. A. Elings, R. S. Downing, R. A. Sheldon,
Eur. J. Org. 1999, 837–846.
[
[
3
7, 1983–1990; c) T. D. Ashton, S. P. Baker, Sally A.
Hutchinson, P. J. Scammells, Bioorgan. Med. Chem.
008, 16, 1861–1873.
16] a) F. Alonso, P. Riente, M. Yus, Tetrahedron Lett. 2008,
2
4
9, 1939–1942; b) A. Clerici, N. Pastori, O. Porta, Eur.
[19] M. M. Bradford, Anal. Biochem. 1976, 72, 248-254.
3390
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3380 – 3390