Tài liệu Báo cáo khoa học: A novel electron transport system for thermostable CYP175A1 from Thermus thermophilus HB27 doc

A novel electron transport system for thermostable
CYP175A1 from Thermus thermophilus HB27
Takao Mandai, Shinsuke Fujiwara and Susumu Imaoka
Nanobiotechnology Research Center and Department of Bioscience, School of Science and Technology, Kwansei Gakuin University,
Gakuen, Sanda, Japan
Cytochrome P450s are associated with a number of
physiologically essential reactions, including drug
metabolism, carbon source assimilation, and the bio-
synthesis of steroids, vitamins, prostaglandins, and
antibiotics [1]. Cytochrome P450s have great potential
to perform numerous industrially important reactions.
Indeed, cytochrome P450sca-2 from Streptomyces car-
bophilus has already been used for the production of
pravastatin, a cholesterol-lowering drug [2]. However,
low tolerance to various solvents and high temperature
has generally limited the usefulness of cyto-
chrome P450s for industrial applications. Thermophilic
cytochrome P450s possess extreme stability, and
might be used to overcome such limitations. Recently,
two thermophilic cytochrome P450s, CYP119 and
CYP175A1, were identified in Sulfolobus solfataricus
and Thermus thermophilus, respectively [3,4].
CYP119 is well characterized, and its crystal struc-
ture has been determined in the ligand-free state and
in several ligand-bound states [5,6]. As expected,
CYP119 is highly resistant to both high temperatures
(T
m
=91°C) and high pressures (up to 2 kbar) [7].
Keywords
CYP175A1; ferredoxin; ferredoxin–NAD(P)

+
reductase; Thermus thermophilus;
b-carotene hydroxylase
Correspondence
S. Imaoka, Department of Bioscience,
School of Science and Technology, Kwansei
Gakuin University, 2-1 Gakuen, Sanda
669-1337, Japan
Tel ⁄ Fax: +81 79 565 7673
E-mail:
(Received 30 January 2009, revised 15
February 2009, accepted 18 February 2009)
doi:10.1111/j.1742-4658.2009.06974.x
CYP175A1 from Thermus thermophilus is a thermophilic cytochrome P450
and has great potential for industrial applications. However, a native elec-
tron transport system for CYP175A1 has not been identified. Here, an elec-
tron transport system for CYP175A1 was isolated from T. thermophilus
HB27 by multistep chromatography, and identified as comprising ferre-
doxin (Fdx; locus in the genome, TTC1809) and ferredoxin–NAD(P)
+
reductase (FNR; locus in the genome, TTC0096) by N-terminal amino acid
sequence analysis and MALDI-TOF-MS, respectively. Although TTC0096,
which encodes the FNR, is annotated as a thioredoxin reductase in the
T. thermophilus HB27 genome database, TTC0096 lacks an active-site dithi-
ol ⁄ disulfide group, which is required to exchange reducing equivalents with
thioredoxin. The FNR reduced ferricyanide, an artificial electron donor, in
the presence of NADH and NADPH, but preferred NADPH as a cofactor
(K
m
for NADH = 2440 ± 546 lm; K

m
for NADPH = 4.1 ± 0.2 lm).
Furthermore, the FNR reduced cytochrome c in the presence of NADPH
and Fdx. The T
m
value of the FNR was 99 °C at pH 7.4. With an electron
transport system consisting of Fdx and FNR, CYP175A1 efficiently cata-
lyzed the hydroxylation of b-carotene at the 3-position and 3¢-position at
65 °C, and the K
m
and V
max
values for b-carotene hydroxylation were
14.3 ± 1.6 lm and 18.3 ± 0.6 nmol b-cryptoxanthinÆmin
)1
Ænmol
)1
CYP175A1, respectively. This is the first report of a native electron trans-
port system for CYP175A1.
Abbreviations
Fdx, ferredoxin; FNR, ferredoxin–NAD(P)
+
reductase; IPTG, isopropyl-thio-b-D-galactoside; OFOR, 2-oxoacid:ferredoxin oxidoreductase;
ONFR, oxygenase-coupled NADH–ferredoxin reductase; SD, standard deviation; TR, thioredoxin reductase; UPLC, ultra-performance liquid
chromatography.
2416 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS
The structure of CYP119 exhibits the typical cyto-
chrome P450 fold [5]. However, differences between
CYP119 and other cytochrome P450s include a rela-
tively high number of salt bridges, a low number of

Ala residues and a high number of Ile residues in the
interior of CYP119, and the presence of more exten-
sive aromatic networks [8]. It has been suggested that
these differences contribute to the thermostability of
CYP119. In particular, aromatic networks appear to
contribute significantly to the thermostability of
CYP119 [9]. On the other hand, CYP175A1 has been
only partially characterized, although its crystal struc-
ture has been determined [4]. CYP175A1 shows high
thermostability (T
m
=88°C), and its substrate-bind-
ing region is highly similar to the substrate-binding
region of cytochrome P450 BM-3, which catalyzes the
hydroxylation of saturated fatty acids [4]. However,
CYP175A1 catalyzes the hydroxylation of b-carotene
at the 3-position and 3¢-position, but does not catalyze
the hydroxylation of fatty acids [10,11].
To perform their oxidative reactions, cyto-
chrome P450s require two electrons supplied primarily
from NAD(P)H via electron transport systems, which
are composed of one or more redox proteins and are
divided into two main classes. Most bacterial and
mammalian mitochondrial cytochrome P450s utilize
the class I system, which is composed of an iron–sulfur
protein and an FAD-containing NAD(P)H-dependent
reductase [12]. Eukaryotic cytochrome P450s utilize the
class II system, composed of an NADPH-dependent
reductase containing both FAD and FMN [12]. How-
ever, recent studies have revealed a number of unusual

electron transport systems for cytochrome P450s that
cannot be described as belonging to either class I or
class II [1,12]. The electron transport system for
CYP119 is a good example of such a system. In this
case, the electron transport system is composed of
ferredoxin (Fdx) and 2-oxoacid:Fdx oxidoreductase
(OFOR), and utilizes pyruvate as an electron source
rather than NAD(P)H [13,14]. On the other hand, the
native electron transport system for CYP175A1 has
not yet been identified, although the catalytic activity
of CYP175A1 has been detected using an artificial
electron transport system for CYP101 from the meso-
philic bacterium Pseudomonas putida [11].
Most Thermus species are known to produce carot-
enoid-like pigments. CYP175A1 catalyzes the hydrox-
ylation of b-carotene at the 3-position and 3¢-position,
producing zeaxanthin via b-cryptoxanthin [10]. The
zeaxanthin produced by CYP175A1 is used as an inter-
mediate for the synthesis of thermozeaxanthins and
thermobiszeaxanthins, which are the main carotenoids
of T. thermophilus [15]. The insertion of thermozeax-
anthins and thermobiszeaxanthins into the cell mem-
brane reduces membrane fluidity and reinforces the
membrane [16], contributing to the survival of T. ther-
mophilus at high temperatures. Thus, identification of
the electron transport system for CYP175A1 is consid-
ered important not only for developing industrial
applications, but also for investigating the physiologi-
cal characteristics associated with this system.
A native electron transport system for CYP175A1

has not yet been identified, although CYP175A1 pos-
sesses great potential for industrial applications. Thus,
in this study, a native electron transport system for
CYP175A1 was isolated from the cytosol of T. thermo-
philus HB27, in order to reconstitute a high-tempera-
ture CYP175A1 catalytic system. The electron
transport system was composed of Fdx and Fdx–
NAD(P)
+
reductase (FNR), and these components
were characterized at high temperature.
Results
Isolation and identification of the components of
the CYP175A1 electron transport system
To find the electron donor of the electron transport
system for CYP175A1, we initially measured the
b-carotene hydroxylation activity in the presence of
purified CYP175A1, the cytosol of T. thermophilus,
and the electron donors NADH, NADPH, and pyru-
vate (+CoA), which are generally used in cyto-
chrome P450 systems. The catalytic activities of
CYP175A1 in the presence of NADH and NADPH
were 0.03 and 0.43 nmol b-cryptoxanthinÆmin
)1
Ænmol
)1
CYP175A1, respectively. NADPH was about 14-fold
more effective than NADH in this system. Pyruvate
(+CoA) is known to be used in the CYP119 system
[13], but was not effective in the CYP175A1 system.

Then, in order to identify electron transport proteins,
the cytosol of T. thermophilus was separated into five
fractions using an anion exchange column (DE52) by
stepwise elution with KCl (50, 100, 200, 300, and
500 mm). b-Carotene hydroxylation activity was not
detected in the presence of any single fraction, but was
detected in the presence of both the 100 mm KCl and
300 mm KCl fractions with purified CYP175A1 and
NADPH. These results suggest that the electron trans-
port system for CYP175A1 was dependent on
NADPH and composed of at least two proteins in the
100 mm KCl and 300 mm KCl fractions. The 300 mm
KCl fraction from the DE52 column was further puri-
fied using a butyl–Sepharose column and a Mono Q
column. b-Carotene hydroxylation activity was
detected in a major peak when it was reacted with
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2417
purified CYP175A1, NADPH, and the 100 mm KCl
fraction from the DE52 column (data not shown). The
peak was subjected to SDS ⁄ PAGE, and a single band
was observed (Fig. 1A). These purification steps are
summarized in Table 1. The purified protein gave a
UV–visible spectrum with a broad absorption peak at
400 nm and a peak at 280 nm (A
400
⁄ A
280
= 0.63)
(Fig. 1B). The absorption spectrum was very similar to

that of Fdx from T. thermophilus, which contains two
iron–sulfur clusters (one [4Fe–4S] cluster and one
[3Fe–4S] cluster) [17–19]. The N-terminal amino acid
sequence of the purified protein was Pro-His-Val-Ile-
X-Glu-Pro-X-Ile, which corresponds to the N-terminal
sequence of the seven-iron Fdx (locus in the genome,
TTC1809). These results suggest that Fdx is a com-
ponent of an electron transport system for CYP175A1.
The 100 mm KCl fraction from the DE52 column was
further purified using a 2¢,5¢-ADP–Sepharose column
and a Mono Q column. b-Carotene hydroxylation
activity was detected in a major peak when it was
reacted with purified CYP175A1, NADPH, and the
300 mm KCl fraction from the DE52 column (data not
shown). The peak was subjected to SDS ⁄ PAGE, and a
single band was observed (Fig. 2A). These purification
steps are summarized in Table 2. The purified protein
was analyzed by MALDI-TOF-MS. Peptide mass fin-
gerprinting was used to search the NCBInr database
using mascot. The result of the mascot search
suggested that the band was a protein encoded by
TTC0096 (locus in the genome). The molecular mass
estimated by SDS ⁄ PAGE was 33.2 kDa, which corre-
sponds to that calculated from the amino acid
sequence of the protein encoded by TTC0096
(36 176 Da). On the other hand, the molecular mass of
the purified protein under nondenaturing conditions
was determined to be 74.9 kDa by gel filtration on a
Superdex-200HR column (data not shown), suggesting
that the protein encoded by TTC0096 forms a homo-

dimer under nondenaturing conditions. Furthermore,
the protein encoded by TTC0096 gave a UV–visible
spectrum with absorption peaks at 273, 392, and
473 nm, which is characteristic of flavoproteins
(Fig. 2B). The FAD content of the protein was
0.70 mol FADÆmol
)1
subunit, suggesting that the FAD
was noncovalently bound to the protein. These results
suggest that another component of an electron trans-
port system for CYP175A1 is a protein encoded by
TTC0096, which functions as an FNR. Thus, we
concluded that the electron transport system for
CYP175A1 belongs to class I.
Characterization of recombinant FNR
The FNR and Fdx were expressed in Escherichia coli
and purified to homogeneity. The purified recombinant
FNR and Fdx had the same chromatographic, photo-
metric and catalytic properties as the native FNR and
Fdx (data not shown). Although the FNR reduced ferri-
cyanide, an artificial electron acceptor, at 25 °C and at
pH 7.4 in the presence of NADH as well as NADPH,
the K
m
value of the FNR for NADPH was about
600-fold lower than that for NADH, and the V
max
value
of the FNR with NADPH was about 55-fold higher
A

62
47.5
32.5
25
16.5
kDa
12 3 45
B
Wavelength (nm)
0.0
1.0
0.5
300 400
500 600 700
Absorbance
Fig. 1. Purification and characterization of Fdx from T. thermophilus
HB27. (A) SDS ⁄ PAGE of fractions containing Fdx at each step of
purification. SDS ⁄ PAGE was carried out on a 15% polyacrylamide
gel. Lane 1: molecular mass markers. Lane 2: cytosol of T. thermo-
philus HB27 (20 lg). Lane 3: 300 m
M KCl fraction from a DE52
column (8.3 lg). Lane 4: fraction eluted from a butyl–Sepharose
column (13.3 lg). Lane 5: fraction eluted from a Mono Q column
(4.6 lg). (B) Absorption spectrum of native Fdx purified from
T. thermophilus HB27. The absorption spectrum of purified Fdx
(25 l
M) was measured in buffer A (50 mM potassium phosphate
buffer, pH 7.4, 10% glycerol).
Thermostable electron transport system T. Mandai et al.
2418 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS

than that with NADH (Table 3). Taken together, these
results show that the FNR prefers NADPH over
NADH. Furthermore, the FNR showed 4.2-fold greater
ferricyanide reduction activity at 50 °C with saturating
concentrations of NADPH (1 mm) and ferricyanide
(1 mm) than at 25 °C (data not shown).
To determine the optimal pH of the FNR, we mea-
sured ferricyanide reduction activity at 50 °C and at a
range of pH values from 4.0 to 8.0 (Fig. 3A).
Although the intracellular pH of T. thermophilus is
known to be maintained at 6.9–7.1 [20], the FNR
unexpectedly exhibited maximal activity at pH 4.5–6.5.
The thermostability of the FNR was evaluated by
measuring the residual ferricyanide reduction activity
after incubation of the FNR for 30 min at various
temperatures (Fig. 3B). The T
m
values of the FNR at
pH 7.4 and at pH 5.0 were 99 and 95 °C, respectively.
These results indicate that the FNR is an extremely
thermostable protein at both pH 7.4 and pH 5.0.
The FNR reduced cytochrome c at 50 °C in the
presence of NADPH and Fdx, and the activity was
dependent on the concentration of Fdx (Table 4).
These results also indicate that the FNR, which is
encoded by TTC0096, transfers electrons from
NADPH to Fdx.
Characterization of the CYP175A1 system
reconstituted from its recombinant components
We attempted to reconstitute b-carotene hydroxylation

activity with the excess purified recombinant
CYP175A1, Fdx, and FNR. The reconstitution system
did support NADPH-dependent b-carotene hydroxyl-
ation, and two hydroxylated products were detected by
HPLC (Fig. 4A). Using ultra-performance liquid chro-
matography (UPLC)-MS, we confirmed that the two
products were b-cryptoxanthin and zeaxanthin
(data not shown). Furthermore, b-carotene hydroxyl-
ation products were not detected in the absence of
CYP175A1, Fdx, or FNR (data not shown). There-
fore, these results clearly indicate that the electron
transport system for CYP175A1 is composed of Fdx,
FNR, and NADPH (Fig. 4B).
All quantitative analyses were performed at 65 °C
for 2 min, to limit the production of a second metabo-
lite, zeaxanthin, and to inhibit the degradation of
b-carotene by high temperatures. The b-carotene
hydroxylation activity was determined from the
production of b-cryptoxanthin, and the production of
zeaxanthin was ignored. In order to determine the
optimal conditions for the reconstitution system, the
effects of pH, Fdx, and Tween 20 on b-carotene
hydroxylation activity were assessed. The optimal pH
for the reconstitution system was pH 5.0, which is con-
sistent with the optimal pH of the FNR (Fig. 5A). The
Fdx ⁄ CYP175A1 ratio was saturated at 8 : 1, and the
turnover rate at an Fdx ⁄ CYP175A1 ratio of 8 : 1 was
4.9-fold greater than that at a ratio of 1 : 1 (Fig. 5B).
The addition of appropriate detergents or phospholip-
ids was required to obtain maximal turnover

with other carotenoid oxygenases, such as carotenoid
dioxygenases, because detergents and phospholipids
presumably aid the solubilization of carotenoid and
thus increase its ability to access the active site of
carotenoid oxygenases [21–23]. Thus, we assessed the
effect of Tween 20 on b-carotene hydroxylation acti-
vity (Fig. 5C). Tween 20 stimulated b-carotene hydrox-
ylation activity, with maximal activity at 0.6–0.8%.
The turnover rate of the reconstitution system under
the optimal conditions was 12.4 nmol b-cryptoxan-
thinÆmin
)1
Ænmol
)1
CYP175A1. Furthermore, the K
m
and V
max
values for b-carotene hydroxylation by the
reconstitution system were determined under the opti-
mized conditions (Fig. 5D). The reaction followed
Michaelis–Menten kinetics, and the K
m
and V
max
values were 14.3 ± 1.6 lm and 18.3 ± 0.6 nmol
b-cryptoxanthinÆmin
)1
Ænmol
)1

CYP175A1, respectively.
Discussion
In the present study, we isolated an electron transport
system for CYP175A1 from T. thermophilus HB27 by
Table 1. Purification of Fdx from T. thermophilus HB27. Total activity is defined as b-carotene hydroxylation activity. Activities were
measured with reaction mixtures (total volume, 200 lL) containing CYP175A1 (0.5 l
M), b-carotene (20 lM), NADPH (1 mM), and the 100 mM
KCl fraction (10 lg) from the DE52 column in buffer A (50 mM potassium phosphate buffer, pH 7.4, and 10% glycerol). The reactions were
performed at 65 °C for 2 min.
Purification
steps Total protein (mg)
Total activity
(nmolÆmin
)1
)
Specific activity
(nmolÆmin
)1
Æmg
)1
) Purification (fold) Yield (%)
Crude extract 474.5 53.2 0.1 1 100
DE52 28.1 47.0 1.7 17 88
Butyl–Sepharose 1.7 26.4 15.4 154 50
Mono Q 0.7 24.1 34.8 348 45
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2419
multistep chromatography, and identified the electron
transport proteins. The system utilized NADPH as a
source of electrons, and was composed of Fdx

(TTC1809) and FNR (TTC0096). Thus, the electron
transport system for CYP175A1 belongs to class I,
along with electron transport systems for other bacte-
rial cytochrome P450s, and is very different from
another thermophilic cytochrome P450 (CYP119)
system. In the CYP119 system from the thermophilic
archaeon S. solfataricus, electrons are transferred from
pyruvate via OFOR and Fdx to CYP119 [13,14]. Inter-
estingly, the electron transport system for CYP175A1
did not utilize OFOR, although the T. thermophilus
HB27 genome contains the genes encoding OFOR
(TTC1591 and TTC1592) [24]. An Fdx that contains
seven irons (one [4Fe–4S] cluster and one [3Fe–4S]
cluster) was discovered more than 20 years ago in
T. thermophilus [19], but its function has remained
unclear. Thus, this is the first report to demonstrate
that a protein encoded by TTC0096 functions as an
FNR in T. thermophilus, and that the seven-iron Fdx
functions as a redox partner of CYP175A1. Further-
more, we attempted to purify native CYP175A1, and
measured reduced CO difference spectra in order to
investigate whether or not CYP175A1 would be
expressed under the culture conditions used in this
study, but we could not purify native CYP175A1 and
detect an absorption peak at 450 nm (data not shown).
Nonetheless, very low b-carotene hydroxylation acti-
vity was detected in the presence of Fdx, FNR,
NADPH, and the cytosol of T. thermophilus (data not
shown), suggesting that CYP175A1 was expressed at
very low levels under the culture conditions used in

this study.
TTC0096, which actually encodes FNR, is anno-
tated as a thioredoxin reductase (TR) in the T. thermo-
philus HB27 genome database [24]. According to a
comparison with genuine TRs, shown in Fig. 6A, the
protein encoded by TTC0096 shows significant identity
with the TRs from E. coli and Aeropyrum pernix (31%
and 34%, respectively), and possesses conserved motifs
responsible for the binding of FAD (GXGXXA and
GXFAAGD) and the binding of NADPH
(GXGXXA), whereas the protein encoded by
TTC0096 lacks a redox-active site (CXXC), which par-
ticipates in various redox reactions, such as the reduc-
tion of thioredoxin. Thus, the protein encoded by
TTC0096 will not actually function as a TR, and
TTC0096 is misannotated in the T. thermophilus HB27
genome database. A blast analysis with the FNR
from T. thermophilus revealed a high level of identity
with YumC from Bacillus subtilis (45%) and FNR
from Chlorobium tepidum (44%). Seo et al. [25,26]
have reported that YumC from B. subtilis and FNR
from C. tepidum form a homodimer, contain noncova-
lently bound FAD, and function as a FNR. Further-
more, Seo et al. [25,26] have reported that YumC from
B. subtilis and FNR from C. tepidum share high
sequence identity with genuine TRs from various
A
B
175
83

62
47.5
32.5
25
16.5
kDa
12 3 4 5
Wavelength (nm)
0.0
0.1
0.2
0.3
0.4
300 400 500 600
Absorbance
Absorbance
Wavelength (nm)
450
0.00
0.02
0.04
550350
Fig. 2. Purification and characterization of FNR from T. thermophi-
lus HB27. (A) SDS ⁄ PAGE of fractions containing FNR at each step
of purification. SDS ⁄ PAGE was carried out on a 15% polyacryl-
amide gel. Lane 1: molecular mass markers. Lane 2: cytosol of
T. thermophilus HB27 (20 lg). Lane 3: 100 m
M KCl fraction from a
DE52 column (14 lg). Lane 4: fraction eluted from a 2¢,5¢-ADP–
Sepharose column (4.6 lg). Lane 5: fraction eluted from a Mono Q

column (2.1 lg). (B) Absorption spectrum of native FNR purified
from T. thermophilus HB27. The absorption spectrum of purified
FNR (3.3 l
M) was measured in buffer A. The inset shows the
absorption spectrum between 350 and 600 nm.
Thermostable electron transport system T. Mandai et al.
2420 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS
species, but both lack the redox-active site, and that
YumC from B. subtilis and FNR from C. tepidum con-
stitute a new type of FNR. These characteristics are
similar to those of our FNR, suggesting that our FNR
belongs to this new type. A phylogenetic tree of FNRs
from different sources was constructed (Fig. 6B). As
noted by Aliverti et al. [27], FNRs could be grouped
into two families: plant-type and glutathione reduc-
tase-type FNRs. The phylogenetic analysis revealed
that our FNR as well as YumC from B. subtilis and
FNR from C. tepidum belong to a new type of FNR
among the glutathione reductase-type FNRs. To the
best of our knowledge, this is the first demonstration
that an FNR of this new type is related to a cyto-
chrome P450 system.
The rate of turnover for the reconstitution system
consisting of CYP175A1, Fdx and FNR was
12.4 nmol b-cryptoxanthinÆmin
)1
Ænmol
)1
CYP175A1
under the optimized conditions, with the exception of

temperature. This was about 54-fold greater than the
turnover rate (0.23 nmol b-cryptoxanthinÆmin
)1
Ænmol
)1
CYP175A1) reported by Momoi et al. [11], who car-
ried out reconstitution using an artificial electron
transport system, putidaredoxin and putidaredoxin
reductase from the mesophilic bacterium P. putida.
Although CYP97A4 from Oryza sativa also catalyzes
the hydroxylation of b-carotene at the 3-position and
3¢-position in E. coli [28], the activity of CYP97A4 had
not been characterized in vitro. Thus, this is the first
report to characterize a cytochrome P450-type b-caro-
tene hydroxylase with its native electron transport
system.
In this study, the turnover rate of b-carotene
hydroxylation by the reconstitution system containing
CYP175A1, Fdx and FNR was about 5000-fold lower
than that of ferricyanide reduction by the FNR. The
reason for this discrepancy is unclear, but general
class I systems such as mitochondrial cytochrome P450
systems also show a turnover rate of substrates of
cytochrome P450 that is much lower than the turnover
rate of ferricyanide reduction by FNR [29–31].
As noted above, the CYP175A1 system produces
thermozeaxanthins and thermobiszeaxanthins for rein-
forcement of the cell membrane at high temperature
[16]. Most enzymes, including CYP175A1, that are
related to the carotenoid biosynthetic pathway are

encoded on a megaplasmid, pTT27 [24]. However, the
electron transport system components, Fdx and FNR,
are encoded on a chromosome, suggesting that
the chromosome controls the carotenoid biosynthetic
pathway.
In conclusion, we have found that electrons are
transferred from NADPH via Fdx and FNR to
CYP175A1. The CYP175A1 system is composed of
extremely thermostable proteins (Fig. 4B), and the
T
m
values of CYP175A1, Fdx and FNR are 88, 114,
and 99 °C, respectively [4,32]. The thermostability of
this system may facilitate the development of novel
industrial applications of CYP175A1. In particular, the
substrate-binding region of CYP175A1 was found to
Table 2. Purification of FNR from T. thermophilus HB27. Total activity is defined as b-carotene hydroxylation activity. Activities were
measured with reaction mixtures (total volume, 200 lL) containing CYP175A1 (0.5 l
M), b-carotene (20 lM), NADPH (1 mM), and the 300 mM
KCl fraction (10 lg) from the DE52 column in buffer A. The reactions were performed at 65 °C for 2 min.
Purification
steps Total protein (mg)
Total activity
(nmolÆmin
)1
)
Specific activity
(nmolÆmin
)1
Æmg

)1
) Purification (fold) Yield (%)
Crude extract 474.5 118.3 0.2 1 100
DE52 51.0 85.4 1.7 7 72
ADP–Sepharose 1.5 69.6 46.2 185 59
Mono Q 0.4 29.9 77.8 312 25
Table 3. Kinetic parameters for the ferricyanide reduction activity
of FNR. Ferricyanide reduction activities were measured in 50 m
M
potassium phosphate buffer (pH 7.4) containing potassium ferri-
cyanide (1 m
M). The K
m
value for NADH was determined in the pre-
sence of FNR (200 n
M) and NADH (0.5–7.0 mM), and the K
m
value
for NADPH was determined in the presence of FNR (20 n
M) and
NADPH (2–100 l
M).
NADH NADPH
K
m
(lM) 2440 ± 546 4.1 ± 0.2
V
max
(nmolÆmin
)1

Ænmol
)1
of FAD) 152 ± 15 8318 ± 71
Table 4. Cytochrome c reduction activities. Cytochrome c reduc-
tion activities were measured in 50 m
M potassium phosphate
buffer (pH 7.4) containing horse heart cytochrome c (0.1 m
M), FNR
(50 n
M), Fdx (50–500 nM), and NADPH (0.5 mM)at50°C.
Ratio (FNR : Fdx)
1:0 1:1 1:2 1:5 1:10
(nmolÆmin
)1
Ænmol
)1
of FAD)
105 ± 2 150 ± 4 186 ± 1 346 ± 6 544 ± 10
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2421
be highly similar to the substrate-binding region of
cytochrome P450 BM-3 [4], whose substrates are long-
chain fatty acids. Cytochrome P450 BM-3 has been
engineered to improve activities towards substrates
such as naphthalene, propranolol, and dioxins other
than long-chain fatty acids [33–35]. Thus, this system
with Fdx, FNR and CYP175A1 engineered by site-
directed and random mutagenesis may exhibit activity
towards industrially useful compounds other than
b-carotene, even in the context of industrial envi-

ronments.
Experimental procedures
Materials
T. thermophilus HB27 was a gift from S. Kuramitsu
(Department of Biology, Graduate School of Science,
Osaka University, Osaka, Japan). KOD Plus DNA poly-
merase was purchased from Toyobo (Osaka, Japan). Emul-
gen 911 was a gift from Kao Chemical (Tokyo, Japan).
NADPH, NADH and NADP
+
were purchased from
Oriental Yeast (Tokyo, Japan). a-Cyano-4-hydroxycinnamic
acid was obtained from Bruker Daltonics GmbH (Bremen,
Germany). Molecular mass standards for gel filtration
(MW-GF-200), glucose 6-phosphate and cytochrome c were
purchased from Sigma Chemical Co. (St Louis, MO, USA).
b-Carotene, glucose-6-phosphate dehydrogenase from yeast,
potassium ferricyanide, chloramphenicol, ampicillin, isopro-
pyl-thio-b-d-galactoside (IPTG) and phenylmethanesulfonyl
fluoride were obtained from Wako Pure Chemical indus-
tries (Osaka, Japan). Tween 20 was purchased from
Bio-Rad Laboratories (Hercules, CA, USA).
Cloning, expression and purification of CYP175A1
T. thermophilus HB27 was cultured at 70 °CinThermus
medium (4 g of tryptone, 2 g of yeast extract and 1 g of
NaCl per liter, pH 7.5). T. thermophilus HB27 genomic
DNA was extracted using the Wizard Genomic DNA Puri-
fication Kit (Promega, Madison, WI, USA). CYP175A1
(locus in the genome, TT_P0059) was amplified by PCR
using genomic DNA as a template and two oligonucleotide

primers, 5¢-GGAATTCCATATGAAGCGCCTTTCCCTG-
3¢ (forward primer) and 5¢-CCAAGCTTTCACGCCCGCA
CCTCCTCCCTAG-3¢ (reverse primer). PCR was carried
out at 94 °C for 5 min, and this was followed by 30 cycles
of 94 °C for 30 s, 55 °C for 30 s, and 68 °C for 1 min,
using KOD Plus DNA polymerase. After the PCR product
had been digested with NdeI and HindIII, the fragment was
ligated into the expression vector pET-21a (Novagen, Mad-
ison, WI, USA), and the construct was designated as pET–
CYP175A1. E. coli BL21 (DE3) Codon Plus cells were
transformed with pET–CYP175A1. The transformant was
grown in 2 · YT medium containing chloramphenicol and
ampicillin at 37 °CuptoanD
600
of 1.0, and CYP175A1
expression was induced by treatment with 0.5 mm IPTG
for 24 h at 25 °C. Cells were harvested by centrifugation at
5000 g for 20 min. The pellet was suspended in
buffer A (50 mm potassium phosphate buffer, pH 7.4, and
10% glycerol) containing 1 mm phenylmethanesulfonyl
fluoride, 0.1 mm EDTA, and 0.1% Emulgen 911. Lysozyme
was added to a final concentration of 1 mgÆmL
)1
, and the
Ferricyanide reduction activity
(µmol·min
–1
·nmol of FAD
–1
)

pH
4 5 6 7 8
20
40
60
A
B
0
Residual activity (%)
100
80
60
40
20
0
Temperature (°C)
40 60 80 100
Fig. 3. Characterization of FNR. (A) Effect of pH on the activity of
FNR. The buffers used in this experiment were 50 m
M potassium
acetate buffer of pH range 4.0–6.0 (closed circles and solid line)
and 50 m
M potassium phosphate buffer of pH range 6.0–8.0 (open
circles and dotted line). Ferricyanide reduction assays were
performed in each buffer containing 1 m
M potassium ferricyanide,
FNR (30 n
M) and 1 mM NADPH at 50 °C. The values represent the
mean ± standard deviation (SD) of triplicate experiments. (B) Ther-
mostability of FNR. FNR (60 n

M) was incubated at various tempera-
tures (40–110 °C) for 30 min at pH 7.4 (closed circles and solid
line) or pH 5.0 (open circles and dotted line). The residual ferricya-
nide reduction activity was measured in 50 m
M potassium phos-
phate buffer (pH 7.4) or 50 m
M potassium acetate buffer (pH 5.0)
containing 1 m
M potassium ferricyanide, heat-treated FNR and
1m
M NADPH at 25 °C. The values represent the mean ± SD of
triplicate experiments.
Thermostable electron transport system T. Mandai et al.
2422 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS
solution was stirred at 4 °C for 30 min. The cell suspension
was disrupted by sonication, and the cell debris was then
removed by centrifugation at 50 000 g for 30 min at 4 °C.
The cytosolic fraction was fractionated with ammonium
sulfate as described previously [11]. The pellet was sus-
pended in buffer A, and the solution was diluted two-fold
with buffer A containing 2.0 m ammonium sulfate. The
diluted sample was loaded onto a butyl-Sepharose 4 Fast
Flow column (Amersham Biosciences, Chalfont St Giles,
UK) equilibrated with buffer A containing 1.0 m ammo-
nium sulfate. The column was washed, and CYP175A1 was
eluted with a stepwise gradient of ammonium sulfate (0.5,
0.2, and 0 m) in buffer A. The fraction containing
CYP175A1 was dialyzed against 50 mm potassium phos-
phate buffer (pH 6.3) containing 10% glycerol. The dia-
lyzed solution was loaded onto a Mono S HR5 ⁄ 5 column

(Pharmacia). After the column had been washed with
50 mm potassium phosphate buffer (pH 6.3) containing
10% glycerol and 100 mm KCl, CYP175A1 was eluted with
a linear gradient of 100–600 mm KCl in 50 mm potassium
phosphate buffer (pH 6.3) containing 10% glycerol, at a
flow rate of 1.0 mLÆmin
)1
. Fractions exhibiting a ratio of
absorbance at 418 ⁄ 280 nm above 1.3 were pooled, dialyzed
against buffer A, and stored at )80 °C until use. The con-
centration of purified CYP175A1 was determined with an
extinction coefficient of 104 mm
)1
Æcm
)1
at 418 nm [4].
Approximately 5 mg of purified CYP175A1 was obtained
per 1 L of culture, and a single band was observed on
SDS ⁄ PAGE.
Purification of an electron transport system for
CYP175A1 from T. thermophilus HB27
T. thermophilus HB27 was cultured in Thermus medium
(total volume: 6 L) at 70 °C overnight. T. thermophilus
HB27 was harvested by centrifugation at 5000 g for
20 min. All purification steps were performed at room tem-
perature. The pellet was suspended in buffer B (20 mm
potassium phosphate buffer, pH 7.7, and 10% glycerol)
containing 1 mm phenylmethanesulfonyl fluoride and
0.1 mm EDTA, and the cell suspension was disrupted by
sonication. The cell debris was removed by centrifugation

at 100 000 g for 90 min at 4 °C, and the cytosolic fraction
was then loaded onto a DE52 (Whatman, Maidstone, UK)
column (column volume: 30 mL) equilibrated with
buffer B. The column was washed with buffer B, and the
proteins bound to it were eluted with a stepwise gradient
of KCl (50, 100, 200, 300, and 500 mm) in buffer B. The
300 mm KCl fraction from the DE52 column was diluted
two-fold with buffer A containing 3.0 m ammonium sulfate.
The diluted sample was loaded onto a butyl-Sepharose 4
Fast Flow column equilibrated with buffer A containing
1.5 m ammonium sulfate. After the column had been
washed with buffer A containing 1.5 m ammonium sulfate,
the proteins were eluted with a stepwise gradient of ammo-
nium sulfate (1.0, 0.5, and 0 m) in buffer A. The 1.0 m
ammonium sulfate fraction from the butyl–Sepharose col-
umn was concentrated and desalted on a Bio-Gel P6 DG
column (Bio-Rad Laboratories, Hercules, CA, USA) equili-
brated with buffer D (20 mm potassium phosphate buffer,
pH 6.5, 10% glycerol). The desalted solution was loaded
onto a Mono Q HR5 ⁄ 5 column (Pharmacia) equilibrated
with buffer D. After the column had been washed with buf-
fer D containing 200 mm KCl, the proteins were eluted
with a linear gradient of 200–600 mm KCl at a flow rate
of 1.0 mL Æmin
)1
. The purified protein was desalted on a
Bio-Gel P6 DG column equilibrated with buffer A, and
stored at )80 °C. The 100 mm KCl fraction from the DE52
column was dialyzed against buffer C (20 mm potassium
phosphate buffer, pH 7.4, 10% glycerol, and 0.1 mM

EDTA), and the dialyzed solution was then loaded onto a
2¢,5¢-ADP–Sepharose column (Amersham Biosciences)
equilibrated with buffer C. After the column had been
washed with buffer C containing 150 mm KCl, the proteins
were eluted with buffer C containing 150 mm KCl and
1mm NADP
+
. The fraction eluted from the 2¢,5¢-ADP–
Sepharose column was dialyzed against buffer B. The
β-carotene
β-cryptoxanthin
Retention time (min)
10 20 300
A
454
0
20
40
60
80
100
A
B
Zeaxanthin
NADPH
NADP
+
e
–
FNR

Fdx
CYP175A1
heme
FAD
(99 °C)
a
(114 °C)
b
(88 °C)
c
β-carotene
β-cryptoxanthin
zeaxanthin
Fig. 4. (A) HPLC profiles of the metabolites produced by the recon-
stitution system consisting of excess CYP175A1, Fdx, and FNR.
The reaction mixtures contained CYP175A1 (0.4 l
M), Fdx (0.8 lM),
FNR (0.4 l
M) and b-carotene (30 lM) in buffer A (total volume,
200 lL). The reactions were performed at 65 °C for 5 min without
(solid line) or with (dotted line) 1 m
M NADPH, and the products
were then extracted with ice-cold acetonitrile (1.0 mL). The
extracted products were analyzed by RP-HPLC. The HPLC analysis
was performed as described in Experimental procedures. (B)
Scheme of the electron transport system for CYP175A1. The num-
bers in parentheses indicate the T
m
value of each protein at neutral
pH.

a
Data from this study.
b
Data from Griffin et al. [32].
c
Data from
Yano et al. [4].
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2423
dialyzed solution was loaded onto a Mono Q HR5 ⁄ 5
column equilibrated with buffer B, and the column was
washed with buffer B containing 50 mm KCl. The proteins
were eluted with a linear gradient of 50–200 mm KCl in
buffer B at a flow rate of 1.0 mLÆmin
)1
. The purified pro-
tein was dialyzed against buffer A and stored at )80 °C.
Identification of the purified electron transport
proteins
The electron transport protein purified from the 300 mm KCl
fraction eluted from the DE52 column was identified by
determining the N-terminal amino acid sequence of the puri-
fied protein, which was analyzed by an automated amino
acid sequencer (PPSQ-21A; Shimadzu, Kyoto, Japan),
according to the manufacturer’s instructions. The electron
transport protein purified from the 100 mm KCl fraction
eluted from the DE52 column was identified by MALDI-
TOF-MS. The purified protein was electrophoresed with
an SDS ⁄ polyacrylamide gel, and stained with Coomassie
Brilliant Blue R-250. The band containing the purified

protein was excised from the gel, dehydrated, and then
digested with Trypsin Gold (Promega), according to the
method reported by Wang et al. [36]. The concentrated
peptides were mixed with a-cyano-4-hydroxycinnamic acid in
60% acetonitrile and 0.1% trifluoroacetic acid, and analyzed
8
pH
45 67
0
1
2
3
4
5
A
C
B
D
Turnover rate
(nmol·min
–1
·nmol of CYP175A1
–1
)
Turnover rate
(nmol·min
–1
·nmol of CYP175A1
–1
)

Tween 20 (%)
0.0
0
5
10
15
0.5 1.0 1.5 2.0
Turnover rate
(nmol·min
–1
·nmol of CYP175A1
–1
)
Turnover rate
(nmol·min
–1
·nmol of CYP175A1
–1
)
Fdx (n
M
)
0
300 600 900 1200
0
2
4
6
8
β-carotene (µ

M
)
0 20406080100
0
5
10
15
20
Fig. 5. Characterization of the reconstitution system. (A) Effect of pH on b-carotene hydroxylation activity. The reactions were performed at
the indicated pH value in the presence of CYP175A1 (30 n
M), Fdx (60 nM), FNR (30 nM), 20 lM b-carotene (containing 0.1% Tween-20) and
NADPH (1 m
M)at65°C for 2 min. The buffers used in this experiment were 50 mM potassium acetate buffer containing 10% glycerol of
pH range 4.0–6.0 (closed circles and solid line) and 50 m
M potassium phosphate buffer containing 10% glycerol of pH range 6.0–7.4 (open
circles and dotted line). (B) Effect of Fdx on b-carotene hydroxylation activity. The reaction mixtures contained CYP175A1 (30 n
M), Fdx
(30–960 n
M), FNR (30 nM), 20 lM b-carotene (containing 0.1% Tween-20) and NADPH (1 mM)in50mM potassium acetate buffer (pH 5.0)
containing 10% glycerol (total volume, 200 lL). The reactions were performed at 65 °C for 2 min. (C) Effect of Tween-20 on b-carotene
hydroxylation activity. The reaction mixtures contained CYP175A1 (30 n
M), Fdx (240 nM), FNR (30 nM), Tween-20 (0.1–1.6%), 20 lM b-caro-
tene (containing 0.1% Tween-20) and NADPH (1 m
M)in50mM potassium acetate buffer (pH 5.0) containing 10% glycerol (total volume,
200 lL). The reactions were performed at 65 °C for 2 min. (D) Kinetic analysis of b-carotene hydroxylation by the reconstitution system. The
reaction mixtures contained CYP175A1 (30 n
M), Fdx (240 nM), FNR (30 nM), 0.8% Tween-20, b-carotene (1–80 lM) and NADPH (1 mM)in
50 m
M potassium acetate buffer (pH 5.0) containing 10% glycerol (total volume, 200 lL). The reactions were performed at 65 °C for 2 min.
The reaction products were extracted with 25-fold volumes of ice-cold acetonitrile. In all cases, HPLC of the reaction products was carried

out as described in Experimental procedures, and the values represent the mean ± SD of triplicate experiments.
Thermostable electron transport system T. Mandai et al.
2424 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS
T. thermophilus 1 MAADHTDVLIVGAGPAGLFAGFYVGMRGLSFRFVDPLPEPGGQLTAL 47
E.
coli 1 MGTTKHSKLLILGSGPAGYTAAVYAARANLQPVLITGM-EKGGQLTTT 47
A.
pernix 1 MPLRLSAVRAPKIPRGEEYDTVIVGAGPAGLSAAIYTTRF-LMSTLIVSM-DVGGQLNLT 58
. :*:*:****
* *. * :: : : ****.
1
T.
thermophilus 48 YPEKYIYDVAG-FPKVYAKDLVKGLVEQVAPFNPVYSLGERAETLE-REGDLFKVTTSQG 105
E.
coli 48 T EVENWPGDPNDLTGPLLMERMHEHATKFETEIIFD-HINKVD-LQNRPFRLNGDNG 102
A.
pernix 59 N WIDDYPG-MGGLEASKLVESFKSHAEMFGAKIVTGVQVKTVDRLDDGWFLVRGSRG 114
: : .* : . *:: : .:. * . . : :.:: :. * : *
T.
thermophilus 106 NAYTAKAVIIAAGVGAFEPRRIGAPGEREFEGRGVYYAVKSKA-EFQGK-RVLIVGGGDS 163
E.
coli 103 -EYTCDALIIATGASA RYLGLPSEEAFKGRGVSACATCDG-FFYRNQKVAVIGGGNT 157
A.
pernix 115 LEVKARTVILAVGSRR RKLGVPGEAELAGRGVSYCSVCDAPLFKGKDAVVVVGGGDS 171
::*:*.* * :* *.* : **** . .
* : * ::***::
2 3
T.
thermophilus 164 AVDWALNLLDTARRITLIHRRPQFRAHEASVKELMKAHEEGRLEVLTPYELRRVEGDER- 222
E.

coli 158 AVEEALYLSNIASEVHLIHRRDGFRAEKILIKRLMDKVENGNIILHTNRTLEEVTGDQMG 217
A.
pernix 172 ALEGALLLSGYVGKVYLVHRRQGFRAKPFYVEEARKK-PNIEFILDS IVTEIRGRDR- 227
*
:: ** * . . .: *:*** ***. ::. . : .: : : : .: * :
T.
thermophilus 223 VRWAVVFHNQTQEELA-LEVDAVLILAGYITKLGPLANWGLALEKNKIK VDTTMA 276
E.
coli 218 VTGVRLRDTQNSDNIESLDVAGLFVAIGHSPNT-AIFEGQLELENGYIKVQSGIHGNATQ 276
A.
pernix 228 VESVVVKNKVTGEEKE-LRVDGIFIEIGSEPPK-ELFEA-IGLETDSMG NVVVDEWMR 282
* . : . :: * * .::: * . : : : ** : .
T.
thermophilus 277 TSIPGVYACGDIVTYPGKLPLIVLGFGEAAIAANHAAAYAN-PALKVNPGHSSEKAAPGT 335
E.
coli 277 TSIPGVFAAGDVMDHI YRQAITSAGTGCMAALDAERYLD GLADAK 321
A.
pernix 283 TSIPGIFAAGDCTSMWPGFRQVVTAAAMGAVAAYSAYTYLQEKGLYKPKPLTGLK 337
*****::*.**
: . . :** * * : .*
1
C. tepidum FNR
B. subtilis YumC
T. thermophilus FNR
M. tuberculosis FNR
Pseudomonas sp. BphA4
P. putida PDR
S. cerevisiae ADR
M. tuberculosis FprA
H. sapiens ADR

Nostoc sp. PCC 7120 FNR
Z. mays FNR
S. oleracea FNR
E .coli FNR
R. capsulatus FNR
A. vinelandii FNR
Plant-type FNRs
Plastidic-type
Bacterial-type
New type
A
B
ADR-like
ONFR-like
GR-type FNRs
Fig. 6. (A) Multiple alignment of the amino acid sequences of FNR from T. thermophilus HB27, TR from E. coli, and TR from A. pernix.
Accession numbers (NCBI) are: FNR from T. thermophilus HB27, YP_004071; TR from E. coli, NP_415408; and TR from A. pernix,
NP_147693. Asterisks indicate identical amino acid residues. Colons indicate conservative replacements, and single dots indicate less
conservative replacements. Underlines 1, 2 and 3 indicate the FAD-binding site, the redox-active site, and the NADPH-binding site,
respectively. (B) Phylogenetic tree of FNR from different sources. The phylogenetic tree was constructed using the program
CLUSTALW
( The accession numbers are: FNR from Spinacia oleracea, AAA34029; FNR from Nostoc sp. PCC 7120, NP_488161;
FNR from Zea mays, NP_001105568; FNR from E. coli, NP_418359); FNR from Azotobacter vinelandii, ZP_00417949; FNR from Rhodobact-
er capsulatus, AAF35905; ADR from Homo sapiens, AAB59498; adrenodoxin reductase from Saccharomyces cerevisiae, AAB64812; FprA
from Mycobacterium tuberculosis, O05783; BphA4 from Pseudomonas sp. KKS102, BAA04112; putidaredoxin reductase from P. putida,
AAA25758; FNR from M. tuberculosis H37Rv, NP_215202; YumC from B. subtilis, CAB15201; and FNR from C. tepidum, NP_662397.
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2425
by MALDI-TOF-MS (Ultraflex; Bruker Daltonics GmbH).
Protein identification was carried out by a search of the data-

base NCBInr using mascot (Matrix Science, Boston, MA,
USA).
Cloning, expression, and purification of Fdx
Fdx (TTC1809) was amplified by PCR using genomic DNA
as a template and two oligonucleotide primers, 5¢-GGA
ATTCCATATGCCGCACGTGATCTGCGAG-3¢ (forward
primer) and 5¢-CGCGGATCCTTACTCTAGGCCCGC
GAGCT-3¢ (reverse primer). The PCR product was inserted
into the pET-21a vector, and the construct was designated
pET–Fdx. E. coli BL21(DE3) Codon Plus cells were trans-
formed with pET–Fdx. The transformant was grown in
2 · YT medium containing chloramphenicol and ampicillin
at 37 °CuptoanD
600
of 1.0, and Fdx expression
was induced by treatment with 1.0 mm IPTG for 24 h at
25 °C. Cells were harvested by centrifugation at 5000 g for
20 min, and the pellet was suspended in buffer A contain-
ing 1 mm phenylmethanesulfonyl fluoride. The crude
extract of E. coli was prepared as described above. The
extract was incubated at 80 °C for 30 min, and then centri-
fuged at 20 000 g for 30 min at 4 °C to remove denatured
proteins. The heat-treated supernatant was diluted two-fold
with buffer A containing 3.0 m ammonium sulfate. The
diluted solution was purified with a butyl-Sepharose 4 Fast
Flow column and a Mono Q column under the conditions
described above. The purified Fdx was desalted on a Bio-
Gel P6 DG column equilibrated with buffer A and stored
at )80 °C. The concentration of the purified Fdx was deter-
mined using a molar extinction coefficient of

29.0 mm
)1
Æcm
)1
at 408 nm [13].
Cloning, expression and purification of FNR
FNR (TTC0096) was amplified by PCR using geno-
mic DNA as a template and two oligonucleotide prim-
ers, 5¢-GGAATTCCATATGGCGGCGGAC CACACGGA
CGT-3¢ (forward primer) and 5¢-CGCGGATCCTAGG
TCCCGGGGGCGGCCTTCTC-3¢ (reverse primer). The
PCR product was inserted into the pET-21a vector, and the
construct was designated pET–FNR. E. coli BL21(DE3)
Codon Plus cells were transformed with pET–FNR. The
transformant was grown in 2 · YT medium containing
chloramphenicol and ampicillin at 37 °CuptoanD
600
of
1.0, and FNR expression was induced by treatment with
1.0 mm IPTG for 5 h at 37 °C. Cells were harvested by cen-
trifugation at 5000 g for 20 min, and the pellet was sus-
pended in buffer B containing 1 mm phenylmethanesulfonyl
fluoride and 0.1 mm EDTA. The crude extract of E. coli
was prepared as described above. The extract was incu-
bated at 70 °C for 30 min, and then centrifuged at 20 000 g
for 30 min at 4 °C. The heat-treated supernatant was puri-
fied with a DE52 column, a 2¢,5¢-ADP–Sepharose column
and a Mono Q column under the conditions described
above. The purified FNR was dialyzed against buffer A,
and stored at )80 °C. The concentration of the purified

FNR was determined using a molar extinction coefficient
of 12.5 mm
)1
Æcm
)1
at 473 nm.
Characterization of FNR
The molar extinction coefficient of the purified FNR was
determined by extracting the total enzyme-bound FAD.
The purified FNR was incubated at 110 °C for 30 min in
an aluminum block in 50 mm potassium phosphate buffer
(pH 7.4), and denatured proteins were then removed by
centrifugation at 10 000 g for 10 min. The concentration
of free flavin was determined from its absorption coeffi-
cient of 11.3 mm
)1
Æcm
)1
at 450 nm [37]. The molecular
mass of the purified FNR under nondenaturing condi-
tions was determined by gel filtration on a Superdex-
200HR column, which was calibrated using molecular
mass standards and equilibrated with buffer A containing
150 mm KCl. Gel filtration was carried out with buffer A
containing 150 mm KCl at a flow rate of 0.4 mLÆmin
)1
.
Measurement of b-carotene hydroxylation activity
b-Carotene was dissolved in chloroform (4 mm).
The solution (5 lL) was diluted 10-fold with acetone

containing 2.3% Tween 20 (v ⁄ v), and then mixed
vigorously and vacuum-dried. The resulting residue was
dissolved in 99 lL of the reaction buffer (200 lm b-caro-
tene solution). All reactions were carried out in 2 mL
tubes with caps.
To purify the electron transport system for CYP175A1,
b-carotene hydroxylation reactions with CYP175A1
(0.5 lm) and the electron transport system were per-
formed in buffer A containing 20 lm b-carotene (total
volume, 200 lL). The reaction mixtures were incubated at
65 °C for 3 min, and the reactions were initiated by the
addition of 2 lL of 100 mm NADPH. After 2 min at
65 °C, ice-cold acetonitrile ⁄ chloroform [4 : 1 (v ⁄ v),
1.0 mL] was added to extract the reaction products. The
tubes were placed on ice for 5 min, and then centrifuged
at 13 000 g for 10 min. The supernatant was directly ana-
lyzed by RP-HPLC. The HPLC analysis was performed
using an HPLC system (Prominence; Shimadzu, Kyoto,
Japan) equipped with an ODS-100S column (150 ·
4.6 mm; Tosoh, Tokyo, Japan), and acetonitrile ⁄ metha-
nol ⁄ isopropanol (85 : 10 : 5) was used as the mobile
phase, at a flow rate of 1 mLÆmin
)1
. To determine the
optimal reaction conditions, we assessed the effects of pH
(4.0–7.4), Fdx (30–960 nm) and Tween 20 (0.1–1.6%) on
b-carotene hydroxylation activity. The b-carotene hydrox-
ylation reactions were carried out under the conditions
described above, and the products were extracted with
ice-cold acetonitrile (1.0 mL). For the kinetic analysis,

Thermostable electron transport system T. Mandai et al.
2426 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS
b-carotene was completely solubilized, with minor modifi-
cations. A 4 mm b-carotene solution (5 lL) was diluted
10-fold with acetone containing 4.5% Tween 20 (v ⁄ v).
The solution was mixed vigorously, and then vacuum-
dried. The resulting residue was dissolved in 98 lLof
50 mm potassium acetate buffer (pH 5.0) containing 10%
glycerol (200 lm b -carotene solution). Other reaction
conditions were identical to those described above, and
the reaction products were extracted with 25-fold volumes
of ice-cold acetonitrile.
The reaction products were identified by UPLC-MS. The
products were extracted with chloroform (400 lL), and dis-
tilled water (1 mL) was then added to the organic phase to
remove Tween 20. The organic phase was dried, dissolved
in methanol, and analyzed by UPLC-MS. The reaction
products were separated using an UPLC system (ACQUITY
UPLC system; Waters, Milford, MA, USA) equipped with
an ACQUITY UPLC BEH C18 column (1.7 lm,
2.1 · 150 mm; Waters, Ireland), and the same mobile phase
as described above was used at a flow rate of 0.2 mLÆmin
)1
.
MS was carried out using a NanoFrontier LD mass spec-
trometer (Hitachi, Tokyo, Japan). The MS parameters were
as follows: atmospheric pressure chemical ionization
(APCI) positive ion mode; spray potential, 3500 V; N
2
gas

flow, 5 LÆmin
)1
;N
2
gas temperature, 250 °C.
Ferricyanide and cytochrome c reduction assay
Unless otherwise stated, ferricyanide reduction assays
were performed in 50 mm potassium phosphate buffer
(pH 7.4) containing potassium ferricyanide (1 mm)at
25 °C (total volume, 500 lL). For the kinetic analysis,
the concentration of NADPH was kept constant by
regeneration with glucose 6-phosphate and glucose-6-
phosphate dehydrogenase from yeast. Cytochrome c
reduction assays were performed in 50 mm potassium
phosphate buffer (pH 7.4) containing horse heart cyto-
chrome c (0.1 mm), NADPH (0.5 mm), FNR (50 nm)
and Fdx (50–500 nm)at50°C (total volume,
500 lL). Ferricyanide reduction activity was calculated
from the decrease in absorbance at 420 nm (e
420 nm
=
1.02 mm
)1
Æcm
)1
). Cytochrome c reduction activity was
calculated from the increase in absorbance at 550 nm
(e
550 nm
= 21.0 mm

)1
Æcm
)1
).
Thermostability of FNR
Purified FNR (60 nm) was incubated for 30 min at various
temperatures (40–110 °C) in buffer A or 50 mm potassium
acetate buffer (pH 5.0) containing 10% glycerol. The heat
treatment was stopped by placing the sample on ice for
5 min, and the denatured proteins were then removed by
centrifugation at 10 000 g for 10 min at 4 °C. The residual
ferricyanide reduction activity of each sample was then
measured as described above.
Construction of a phylogenetic tree
A phylogenetic tree was constructed with FNRs from
different sources, using the neighbor-joining method of
clustalw ( />Acknowledgements
This study was partially supported by a Grant-in-Aid
for Exploratory Research from the Japan Society for
the Promotion of Science and a special Grant-in-Aid
of the Advanced Program of High Profile Research for
Academia-Industry Cooperation, sponsored by the
Ministry of Education, Science, Culture, Sports and
Technology of Japan.
References
1 Hannemann F, Bichet A, Ewen KM & Bernhardt R
(2007) Cytochrome P450 systems – biological variations
of electron transport chains. Biochim Biophys Acta
1770, 330–344.
2 Serizawa N (1996) Biochemical and molecular

approaches for production of pravastatin, a potent cho-
lesterol-lowering drug. Biotechnol Annu Rev 2, 373–389.
3 Wright RL, Harris K, Solow B, White RH & Kennelly
PJ (1996) Cloning of a potential cytochrome P450 from
the archaeon Sulfolobus solfataricus. FEBS Lett 384,
235–239.
4 Yano JK, Blasco F, Li H, Schmid RD, Henne A &
Poulos TL (2003) Preliminary characterization and
crystal structure of a thermostable cytochrome P450
from Thermus thermophilus. J Biol Chem 278, 608–
616.
5 Yano JK, Koo LS, Schuller DJ, Li H, Ortiz de Montel-
lano PR & Poulos TL (2000) Crystal structure of a
thermophilic cytochrome P450 from the archaeon Sulfo-
lobus solfataricus. J Biol Chem 275, 31086–31092.
6 Park SY, Yamane K, Adachi S, Shiro Y, Weiss KE,
Maves SA & Sligar SG (2002) Thermophilic cyto-
chrome P450 (CYP119) from Sulfolobus solfataricus :
high resolution structure and functional properties.
J Inorg Biochem 91, 491–501.
7 McLean MA, Maves SA, Weiss KE, Krepich S & Sligar
SG (1998) Characterization of a cytochrome P450 from
the acidothermophilic archaea Sulfolobus solfataricus.
Biochem Biophys Res Commun 252, 166–172.
8 Chang YT & Loew G (2000) Homology modeling,
molecular dynamics simulations, and analysis of
CYP119, a P450 enzyme from extreme acidothermophil-
ic archaeon Sulfolobus solfataricus. Biochemistry 39,
2484–2498.
9 Puchkaev AV, Koo LS & Ortiz de Montellano PR

(2003) Aromatic stacking as a determinant of the
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2427
thermal stability of CYP119 from Sulfolobus solfatari-
cus. Arch Biochem Biophys 409, 52–58.
10 Blasco F, Kauffmann I & Schmid RD (2004)
CYP175A1 from Thermus thermophilus HB27, the first
b-carotene hydroxylase of the P450 superfamily. Appl
Microbiol Biotechnol 64, 671–674.
11 Momoi K, Hofmann U, Schmid RD & Urlacher VB
(2006) Reconstitution of b-carotene hydroxylase activity
of thermostable CYP175A1 monooxygenase. Biochem
Biophys Res Commun 339, 331–336.
12 McLean KJ, Sabri M, Marshall KR, Lawson RJ,
Lewis DG, Clift D, Balding PR, Dunford AJ,
Warman AJ, McVey JP et al. (2005) Biodiversity of
cytochrome P450 redox systems. Biochem Soc Trans
33, 796–801.
13 Puchkaev AV & Ortiz de Montellano PR (2005) The
Sulfolobus solfataricus electron donor partners of ther-
mophilic CYP119: an unusual non-NAD(P)H-depen-
dent cytochrome P450 system. Arch Biochem Biophys
434, 169–177.
14 Puchkaev AV, Wakagi T & Ortiz de Montellano PR
(2002) CYP119 plus a Sulfolobus tokodaii strain 7 ferre-
doxin and 2-oxoacid:ferredoxin oxidoreductase consti-
tute a high-temperature cytochrome P450 catalytic
system. J Am Chem Soc 124, 12682–12683.
15 Yokoyama A, Sandmann G, Hoshino T, Adachi K,
Sakai M & Shizuri Y (1995) Thermozeaxanthins, new

carotenoid-glycoside-esters from thermophilic eubacteri-
um Thermus thermophilus. Tetrahedron Lett 36, 4901–
4904.
16 Hara M, Yuan H, Yang Q, Hoshino T, Yokoyama A
& Miyake J (1999) Stabilization of liposomal mem-
branes by thermozeaxanthins: carotenoid-glucoside
esters. Biochim Biophys Acta 1461, 147–154.
17 Sato S, Nakazawa K, Hon-Nami K & Oshima T (1981)
Purification, some properties and amino acid sequence
of Thermus thermophilus HB8 ferredoxin. Biochim
Biophys Acta 668, 277–289.
18 Hille R, Yoshida T, Tarr GE, Williams CH Jr, Ludwig
ML, Fee JA, Kent TA, Huynh BH & Mu
¨
nck E (1983)
Studies of the ferredoxin from Thermus thermophilus.
J Biol Chem 258, 13008–13013.
19 Ohnishi T, Blum H, Sato S, Nakazawa K, Hon-nami K
& Oshima T (1980) A stable Thermus thermophilus
iron–sulfur protein containing only one binuclear and
one tetranuclear cluster. J Biol Chem 255, 345–348.
20 Laptenko O, Kim SS, Lee J, Starodubtseva M, Cava F,
Berenguer J, Kong XP & Borukhov S (2006) pH-depen-
dent conformational switch activates the inhibitor of
transcription elongation. EMBO J 25, 2131–2141.
21 Marasco E, Vay K & Schmidt-Dannert C (2006) Identi-
fication of carotenoid cleavage dioxygenases from
Nostoc sp. PCC 7120 with different cleavage activities.
J Biol Chem 281, 31583–31593.
22 Liu B & Lee YK (2001) In vitro biosynthesis of xantho-

phylls by cell extracts of a green alga Chlorococcum
.
Plant Physiol Biochem 39, 147–154.
23 Goodman DS, Huang HS, Kanai M & Shiratori T
(1967) The enzymatic conversion of all-trans-beta-caro-
tene into retinal. J Biol Chem 242, 3543–3554.
24 Henne A, Bru
¨
ggemann H, Raasch C, Wiezer A, Hart-
sch T, Liesegang H, Johann A, Lienard T, Gohl O,
Martinez-Arias R et al. (2004) The genome sequence of
the extreme thermophile Thermus thermophilus. Nat Bio-
technol 22, 547–553.
25 Seo D & Sakurai H (2002) Purification and character-
ization of ferredoxin-NAD(P)(+) reductase from the
green sulfur bacterium Chlorobium tepidum. Biochim
Biophys Acta 1597, 123–132.
26 Seo D, Kamino K, Inoue K & Sakurai H (2004) Purifi-
cation and characterization of ferredoxin-NADP+
reductase encoded by Bacillus subtilis yumC. Arch
Microbiol 182, 80–89.
27 Aliverti A, Pandini V, Pennati A, De Rosa M & Zanetti
G (2008) Structural and functional diversity of ferre-
doxin-NADP(+) reductases. Arch Biochem Biophys
474, 283–291.
28 Quinlan RF, Jaradat TT & Wurtzel ET (2007) Escheri-
chia coli as a platform for functional expression of plant
P450 carotene hydroxylases. Arch Biochem Biophys 458,
146–157.
29 Chun YJ, Shimada T, Sanchez-Ponce R, Martin MV,

Lei L, Zhao B, Kelly SL, Waterman MR, Lamb DC
& Guengerich FP (2007) Electron transport pathway
for a Streptomyces cytochrome P450: cytochrome P450
105D5-catalyzed fatty acid hydroxylation in Strepto-
myces coelicolor A3(2). J Biol Chem 282, 17486–
17500.
30 Sielaff B & Andreesen JR (2005) Kinetic and binding
studies with purified recombinant proteins ferredoxin
reductase, ferredoxin and cytochrome P450 comprising
the morpholine mono-oxygenase from Mycobacterium
sp. strain HE5. FEBS J 272, 1148–1159.
31 Mitani F (1979) Cytochrome P450 in adrenocortical
mitochondria. Mol Cell Biochem 24, 21–43.
32 Griffin S, Higgins CL, Soulimane T & Wittung-Stafsh-
ede P (2003) High thermal and chemical stability of
Thermus thermophilus seven-iron ferredoxin. Linear
clusters form at high pH on polypeptide unfolding. Eur
J Biochem 270, 4736–4743.
33 Carmichael AB & Wong LL (2001) Protein engineering
of Bacillus megaterium CYP102. The oxidation of poly-
cyclic aromatic hydrocarbons. Eur J Biochem 268,
3117–3125.
34 Otey CR, Bandara G, Lalonde J, Takahashi K &
Arnold FH (2006) Preparation of human metabolites of
propranolol using laboratory-evolved bacterial cyto-
chromes P450. Biotechnol Bioeng 93, 494–499.
Thermostable electron transport system T. Mandai et al.
2428 FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS
35 Sulistyaningdyah WT, Ogawa J, Li QS, Shinkyo R,
Sakaki T, Inouye K, Schmid RD & Shimizu S (2004)

Metabolism of polychlorinated dibenzo-p-dioxins by
cytochrome P450 BM-3 and its mutant. Biotechnol Lett
26, 1857–1860.
36 Wang W, Sun J, Nimtz M, Deckwer W & Zeng A
(2003) Protein identification from two-dimensional gel
electrophoresis analysis of Klebsiella pneumoniae by
combined use of mass spectrometry data and raw
genome sequences. Proteome Sci 1, 6, doi:10.1186/1477-
5956-1-6.
37 Jeon S-J & Ishikawa K (2002) Identification and charac-
terization of thioredoxin and thioredoxin reductase from
Aeropyrum pernix K1. Eur J Biochem 269, 5423–5430.
T. Mandai et al. Thermostable electron transport system
FEBS Journal 276 (2009) 2416–2429 ª 2009 The Authors Journal compilation ª 2009 FEBS 2429