Metabolism of substrates incorporated into phospholipid vesicles by mouse 25-hydroxyvitamin D3 1α-hydroxylase (CYP27B1)

Article abstract of DOI:10.1016/j.jsbmb.2010.02.022

CYP27B1 catalyzes the 1α-hydroxylation of 25-hydroxyvitamin D3 to 1α,25-dihydroxyvitamin D3, the hormonally active form of vitamin D3. To further characterize mouse CYP27B1, it was expressed in Escherichia coli, purified and its activity measured on substrates incorporated into phospholipid vesicles, which served as a model of the inner mitochondrial membrane. 25-Hydroxyvitamin D3 and 25-hydroxyvitamin D2 in vesicles underwent 1α-hydroxylation with similar kinetics, the catalytic rate constants (k_cat) were 41 and 48mol/min/mol P450, respectively, while K_m values were 5.9 and 4.6mmol/mol phospholipid, respectively. CYP27B1 showed inhibition when substrate concentrations in the membrane were greater than 4 times K_m, more pronounced with 25-hydroxyvitamin D3 than 25-hydroxyvitamin D2. Higher catalytic efficiency was seen in vesicles prepared from dioleoyl phosphatidylcholine and cardiolipin than for dimyristoyl phosphatidylcholine vesicles. CYP27B1 also catalyzed 1α-hydroxylation of vesicle-associated 24R,25-dihydroxyvitamin D3 and 20-hydroxyvitamin D3, and 25-hydroxylation of 1α-hydroxyvitamin D3 and 1α-hydroxyvitamin D2, but with much lower efficiency than for 25(OH)D3. This study shows that CYP27B1 can hydroxylate 25-hydroxyvitamin D2 and 25-hydroxyvitamin D3 associated with phospholipid membranes with the highest activity yet reported for the enzyme. The expressed enzyme has low activity at higher concentrations of 25-hydroxyvitamin D in membranes, revealing that substrate inhibition may contribute to the regulation of the activity of this enzyme.

Full text of DOI:10.1016/j.jsbmb.2010.02.022

Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
Contents lists available at ScienceDirect
Journal of Steroid Biochemistry and Molecular Biology
journal homepage: www.elsevier.com/locate/jsbmb
Metabolism of substrates incorporated into phospholipid vesicles by mouse
25-hydroxyvitamin D3 1␣-hydroxylase (CYP27B1)
Edith K.Y. Tang, Kimberley J.Q. Voo, Minh N. Nguyen, Robert C. Tuckey∗
School of Biomedical, Biomolecular and Chemical Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
CYP27B1 catalyzes the 1␣-hydroxylation of 25-hydroxyvitamin D3 to 1␣,25-dihydroxyvitamin D3,
the hormonally active form of vitamin D3. To further characterize mouse CYP27B1, it was expressed
in Escherichia coli, purified and its activity measured on substrates incorporated into phospholipid
vesicles, which served as a model of the inner mitochondrial membrane. 25-Hydroxyvitamin D3
and 25-hydroxyvitamin D2 in vesicles underwent 1␣-hydroxylation with similar kinetics, the cat-
alytic rate constants (kcat) were 41 and 48 mol/min/mol P450, respectively, while Km values were 5.9
and 4.6 mmol/mol phospholipid, respectively. CYP27B1 showed inhibition when substrate concentra-
tions in the membrane were greater than 4 times Km, more pronounced with 25-hydroxyvitamin D3
than 25-hydroxyvitamin D2. Higher catalytic efficiency was seen in vesicles prepared from dioleoyl
phosphatidylcholine and cardiolipin than for dimyristoyl phosphatidylcholine vesicles. CYP27B1 also
catalyzed 1␣-hydroxylation of vesicle-associated 24R,25-dihydroxyvitamin D3 and 20-hydroxyvitamin
D3, and 25-hydroxylation of 1␣-hydroxyvitamin D3 and 1␣-hydroxyvitamin D2, but with much lower
efficiency than for 25(OH)D3. This study shows that CYP27B1 can hydroxylate 25-hydroxyvitamin D2 and
25-hydroxyvitamin D3 associated with phospholipid membranes with the highest activity yet reported
for the enzyme. The expressed enzyme has low activity at higher concentrations of 25-hydroxyvitamin
D in membranes, revealing that substrate inhibition may contribute to the regulation of the activity of
this enzyme.
Received 31 December 2009
Received in revised form 23 February 2010
Accepted 23 February 2010
Keywords:
Vitamin D
CYP27B1
Cytochrome P450
Phospholipid vesicles
25-Hydroxyvitamin D3
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
ularly in relation to anti-proliferative and pro-differentiation
actions.
25-Hydroxyvitamin D 1␣-hydroxylase (CYP27B1) hydroxylates
25-hydroxyvitamin D3 (25(OH)D3) in the 1␣-position produc-
ing 1␣,25-dihydroxyvitamin D3 (1,25(OH)2D3), the hormonally
active form of vitamin D [1,2]. 1,25(OH)2D3 stimulates cal-
cium absorption, and to a lesser extent phosphorous absorption,
from the small intestine [1]. 1,25(OH)2D3 also has anticar-
cinogenic properties, affecting proliferation, differentiation and
apoptosis in cells of different lineages [1,3,4]. The major loca-
tion of CYP27B1 is in mitochondria of the proximal renal tubule,
but it is also found in a number of extra-renal sites includ-
ing skin, brain, colon, prostate and breast [3,5,6]. This gives
rise to autocrine and paracrine roles of 1,25(OH)2D3, partic-
Despite its importance, CYP27B1 has been poorly studied to
date due to its very low concentration in kidney mitochondria
and its lability once extracted with detergents [7,8]. Hiwatashi et
al. [7] reported purifying a small amount of bovine CYP27B1 to
electrophoretic homogeneity. More recently, Uchida et al. [9] have
reported the purification of mouse CYP27B1 expressed in E. coli.
CYP27B1 belongs to the mitochondrial (Type-1) cytochrome
P450 class which receive electrons to support their hydroxylation
reactions from NADPH via adrenodoxin reductase and adreno-
doxin [10]. Two of the mitochondrial P450s, CYP27A1 and CYP11A1,
appear to be anchored to the mitochondrial membrane primar-
ily by a region involving the F-G loop [11–14]. These P450s
can hydroxylate both vitamin D and cholesterol [12,15–18], and
substrates appear to reach the active site from the membrane
phase [11,13,14,19]. Since CYP27B1 belongs to the same family as
CYP27A1, is associated with the inner mitochondrial membrane
and uses relatively hydrophobic hydroxyvitamin D metabolites as
substrates, it is reasonable to predict that it is also anchored to the
membrane by its F-G loop region and accesses substrate from the
membrane phase. Its substrate, 25(OH)D3, has been shown to par-
tition efficiently (95%) into the membrane phase of liposomes made
Abbreviations: 1(OH)D2, 1␣-hydroxyvitamin D2; 1(OH)D3, 1␣-hydroxyvitamin
D3; 20(OH)D3, 20-hydroxyvitamin D3; 25(OH)D2, 25-hydroxyvitamin D2;
25(OH)D3, 25-hydroxyvitamin D3; 1,25(OH)2D2, 1␣,25-dihydroxyvitamin D2;
1,25(OH)2D3, 1␣,25-dihydroxyvitamin D3; 24,25(OH)2D3, 24R,
25-dihydroxyvitamin D3; cyclodextrin, 2-hydroxypropyl-␤-cyclodextrin.
∗
Corresponding author. Tel.: +61 864883040; fax: +61 864881148.
E-mail address: rtuckey@cyllene.uwa.edu.au (R.C. Tuckey).
0960-0760/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jsbmb.2010.02.022

172
E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
from egg phosphatidylcholine [20]. In the present study, we have
investigated the ability of expressed mouse CYP27B1 to metabo-
lize substrates incorporated into phospholipid vesicles. We show
that 25(OH)D3 in vesicles is efficiently metabolized by CYP27B1 at
low substrate concentrations, whereas at high substrate concen-
trations, marked substrate inhibition is observed.
Affi-Gel 15 coupled with bovine adrenodoxin was added. After 3 h
at 4 ◦C the gel was placed in a column and washed with 20 mM
NaCl in buffer B. The CYP27B1 was eluted with 1 M NaCl in buffer
B and the eluant applied to a Fast Flow octyl Sepharose column
(0.75 cm × 3 cm) equilibrated with buffer C (buffer A plus 0.1 mM
EDTA and 0.1 mM DTT). The column was washed with 20 mL buffer
C plus 0.2% CHAPS and the CYP27B1 eluted with 1.0% CHAPS in
buffer C. Fractions with an absorbance ratio 416 nm/280 nm ≥ 0.5
were pooled, dialyzed against 100 volumes buffer C without NaCl
for 3 h and concentrated to 1–2 ␮M for storage at −80 ◦C.
2. Materials and methods
2.1. Materials
2.4. Bacterial expression and purification of mouse adrenodoxin
20(OH)D3 was produced enzymatically by the action of
CYP11A1 on vitamin D3 and purified by TLC and reverse-phase
HPLC as described before [18]. Vitamin D3, 1,25(OH)2D2, dioleoyl
phosphatidylcholine and bovine heart cardiolipin were from Sigma
(St. Louis, MO). 1(OH)D2, 1(OH)D3, 25(OH)D3, 25(OH)D2 and
CHAPS were from Merck (Darmstadt, Germany). 1,25(OH)2D3 and
24,25(OH)2D3 were a gift from Dr Milan Uskokovic (Hoffmann-La
Roche, Nutley, NJ). [3H]25(OH)D3 and [4-14C]cholesterol were from
PerkinElmer Life Science (Boston, MA). The plasmid pGro7 was from
Takara Bio Inc. (Shiga, Japan). Affi-Gel 15 was from Bio-Rad (CA).
Bovine adrenodoxin reductase and adrenodoxin were purified from
bovine adrenal mitochondria as before [21].
cDNA encoding mature mouse adrenodoxin [25] was chemically
synthesized by GenScript and cloned into pTrc99A using NcoI and
KpnI sites in a similar fashion to that described for human adreno-
doxin [26]. Bacterial cultures were also grown and the adrenodoxin
purified as for human adrenodoxin [26].
2.5. Preparation of phospholipid vesicles
Dioleoyl
phosphatidylcholine
(1.08 ␮mol),
cardiolipin
(0.19 ␮mol) and hydroxyvitamin
D
substrates (as required),
were placed in tubes and the ethanol solvent removed under
nitrogen. To prepare large vesicles, 1.0 mL of vesicle buffer (20 mM
HEPES pH 7.4, 100 mM NaCl, 0.1 mM DTT and 0.1 mM EDTA)
was added, tubes vortexed and the mixture extruded through a
1.0 ␮m membrane 12 times using a Mini-Extruder (Avanti Polar
Lipids, Inc., Alabaster, AL). The resulting vesicles have a diameter
of approximately 1.0 ␮m. Small unilamellar vesicles were pre-
pared by adding 0.5 mL vesicle buffer to the dry lipid mixture
and sonicating for 10 min in a bath-type sonicator as described
before [19]. Vesicles prepared by this method have a diameter of
approximately 40–50 nm [27].
2.2. Expression of mouse CYP27B1 in Escherichia coli
The cDNA for mature mouse CYP27B1 [22], encoding a 4-His-tag
at the C-terminus, was chemically synthesized by GenScript Cor-
poration (Piscataway, NJ). The site of cleavage of the mitochondrial
target sequence was assumed to be between Ser-32 and Val-33, as
predicted by alignment with CYP24 [23]. In our construct, Ser-32
was replaced by Met as the first amino acid of the mature sequence
and Val-33 was replaced by Ala. Silent mutations were made in
the first 6 codons to enhance the AT-richness of the region [24].
The cDNA was ligated into the expression vector, pTrc99A, using
NcoI (5ꢀ-end) and KpnI (3ꢀ-end) restriction sites. Competent E. coli
(JM109) cells containing pGro7 plasmid (encoding the chaperones,
GroEL/ES, under the control of the araB promoter) were trans-
formed with the pTrc99A-CYP27B1 construct. A 1:100 dilution of
an overnight culture was made into 500 mL Terrific Broth (pH 7.5)
containing ampicillin (50 ␮g/mL) and chloramphenicol (20 ␮g/mL),
and incubated at 37 ◦C until the absorbance at 600 nm was
0.6. l-Arabinose (4 mg/mL), isopropyl-␤-d-thiogalactopyranoside
(1 mM) and ␦-aminolevulinic acid (0.5 mM) were added and the
cultures incubated with shaking (220 rpm) for a further 46 h at
26 ◦C, followed by 22 ◦C for 3 h.
2.6. Measurement of 25(OH)D3 association with phospholipid
vesicles
Large vesicles were made from dioleoyl phosphatidylcholine
and cardiolipin by extrusion as described above except that
the initial mixture of lipid also contained [4-14C]cholesterol
(7 nCi), [3H]25(OH)D3 (20 nCi) and unlabelled 25(OH)D3
(0.0031–0.125 nmol). Vesicles (0.9 mL) were diluted to 6.0 mL with
vesicle buffer and centrifuged at 107 000 × g for 6 h at 20 ◦C. The
supernatant was removed and the 14C and 3H radioactivity deter-
mined by scintillation counting using a dual isotope program. The
vesicle pellet was resuspended in 1.0 mL vesicle buffer and the
radioactivity similarly measured. The percentage of vesicles sed-
imented (greater than 99.5%) was determined from the proportion
of [4-14C]cholesterol in the pellet. The degree of association of
25(OH)D3 with the vesicles was determined from the proportion
of [3H]25(OH)D3 in the pellet.
2.3. Purification of CYP27B1
Bacterial cells from a total of 2 L of culture (above) were har-
vested by centrifugation (4 ◦C) and resuspended in 100 mL of
100 mM potassium phosphate pH 7.4, 0.1 mM EDTA, 0.1 mM DTT,
0.1 mM PMSF, 20% glycerol (v/v) and 1% CHAPS (w/v). Cells were
then sonicated on ice using a Branson sonicator with four 2-
min pulses with 2 min cooling intervals. Particulate material was
removed by centrifugation (107 000 × g for 60 min) and the super-
natant dialyzed against 100 volumes buffer A (50 mM sodium
phosphate pH 7.4, 0.3 M NaCl, 0.1 mM PMSF and 20% glycerol (v/v)),
then applied to a nickel-affinity column (1.5 cm × 3 cm, Ni-NTA His-
Bind Resin) equilibrated with the same buffer. The column was
washed with 100 mL buffer A containing 20 mM imidazole and the
CYP27B1 eluted with an imidazole gradient (20–500 mM). The elu-
ant was dialyzed against 100 volumes buffer B (10 mM potassium
phosphate pH 7.4, 0.1 mM PMSF, 0.1 mM EDTA, 0.1 mM DTT and
20% glycerol (v/v)) and concentrated to 2 mL. This solution was
incubated with 10 ␮M 25(OH)D3 for 30 min on ice then 1.5 mL of
2.7. Measurement of CYP27B1 activity
Phospholipid vesicles containing hydroxyvitamin D substrates
(as required) were prepared as described above. Activity mea-
surements were performed in a similar way to that described
for CYP11A1 [18,28,29]. The incubation mixture comprised vesi-
cles (510 ␮M phospholipid), CYP27B1 (8–200 nM), 15 ␮M mouse
adrenodoxin, 0.4 ␮M bovine adrenodoxin reductase, 2 mM glu-
cose 6-phosphate, 2 U/mL glucose 6-phosphate dehydrogenase and
50 ␮M NADPH, in the buffer used for preparation of vesicles.
Following dialysis and enzyme dilution the remaining CHAPS con-
centration in the incubation was typically below 0.06 ␮M. Samples
were preincubated for 8 min, reactions started by the addition of

E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
173
NADPH and incubations carried out at 37 ◦C in a shaking water bath.
Typical incubation volumes were 0.2–1.0 mL. Incubation times
were kept short (typically 2.0 min) in experiments designed to
measure the kinetic constants of CYP27B1 to ensure initial rates
were measured. Reactions were stopped by the addition of 2 mL
ice-cold dichloromethane and the aqueous phase extracted with
dichloromethane [29]. Reverse phase HPLC of samples and mea-
surement of CYP27B1 activity in 0.45% cyclodextrin were also
carried out as before [28,29].
2.8. Analysis of kinetic data
Kinetic parameters were determined by fitting the Michaelis–
Menten equation to the experimental data using Kaleidagraph 3.6.
For metabolism of 25(OH)D3 and 25(OH)D2 where substrate inhi-
bition was observed, we also fitted the following equation to the
data:
Fig. 1. Analysis of the purity of expressed mouse CYP27B1 and mouse adreno-
doxin. CYP27B1 samples were run on a 12.5% SDS-polyacrylamide gel (A). Lane
1, molecular weight markers; lane 2, CHAPS extract of E. coli cells (4 ␮g); lane 3,
CYP27B1 after nickel-affinity chromatography (4 ␮g); lane 4, CYP27B1 following
nickel- and adrenodoxin-affinity chromatography (2 ␮g); lane 5, CYP27B1 following
nickel affinity-, adrenodoxin affinity- and octyl Sepharose-chromatography (2 ␮g).
Purified mouse adrenodoxin was run on a 4–20% gradient gel (B). Lane 1, purified
mouse adrenodoxin (2 ␮g); lane 2, molecular weight markers. Arrows indicate the
protein of interest. CYP, CYP27B1; Adx, adrenodoxin.
ꢀ
ꢃ
ꢁ
ꢂ
2
[S]
Ki
Km
V = kcat[E] 1 +
+
[S]
This equation is derived from a model where two molecules of
substrate (S) binds to the enzyme–substrate complex (ES) simul-
taneously with equal Ki values to form an ES3 complex that is
catalytically inactive. This equation provides the Km and kcat as well
as the inhibitor constant for substrate (Ki).
minus reduced difference spectroscopy directly in the CHAPS
extract of the bacterial cells. Despite using a similar procedure to
Uchida et al. [9], including co-expression with chaperones GroES
and GroEL, and optimizing expression time, our expression level
was considerably lower. CYP27B1 expression levels were simi-
lar for the pTrc99A-CYP27B1 construct in E. coli strains JM109
and DH5␣, the latter being the strain used by Uchida et al. [9].
Due to the low expression, Ni-affinity chromatography alone was
insufficient to provide reasonably pure enzyme for activity stud-
ies, so adrenodoxin-affinity chromatography and octyl Sepharose
chromatography were employed to further purify the cytochrome
(Fig. 1A). The band at 56 kDa (Fig. 1A), clearly visible after Ni-affinity
chromatography and subsequent steps, was close to the predicted
mass for CYP27B1 (53.4 kDa) and was confirmed to be CYP27B1
by electrospray ionization mass spectrometry following trypsin
digestion. The final preparation of CYP27B1 was predominantly low
spin with a 417 nm/280 nm absorbance ratio of 0.6, similar to that
reported by Uchida et al. [9], and was relatively free of cytochrome
P420 (Fig. 2A).
2.9. Other procedures
The concentration of cytochrome CYP27B1 was determined
from the CO-reduced minus reduced difference spectrum using
an extinction coefficient of 91 000 M−1cm−1 for the absorbance
difference between 450 and 490 nm [30]. The concentration of
hydroxyvitamin D3 was determined using an extinction coefficient
of 18 000 M−1cm−1 at 263 nm [7]. SDS-polyacrylamide gel elec-
trophoresis was carried out as before [31]. Protein was determined
by the Ponceau S procedure [32]. For the measurement of the Km
and kcat for adrenodoxin, the concentration of free adrenodoxin was
calculated as described by Hanukoglu and Jefcoate [33].
3. Results
3.1. Expression and purification of mouse CYP27B1 and
adrenodoxin
CYP27B1 proved to be labile during purification despite the
presence of 20% glycerol as a stabilizing agent. It was found impor-
tant to avoid prolonged standing of the enzyme at 4 ◦C, including
overnight dialysis, as this caused marked loss of the holo-enzyme
(up to 50%). Losses were minimized by carrying out the purifica-
The level of expression of mouse CYP27B1 was 20 nmol/L cul-
ture. This was measured after Ni-affinity chromatography since
the low level of expression and the presence of some cytochrome
P420 prevented measurement of its concentration via CO-reduced
Fig. 2. Spectral analysis of purified mouse CYP27B1 and adrenodoxin. (A) CO-reduced minus reduced difference spectrum of mouse CYP27B1 following nickel affinity-,
adrenodoxin affinity- and octyl Sepharose-chromatography. (B) Absolute absorbance spectrum of purified mouse adrenodoxin.

174
E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
Table 1
Association of 25(OH)D3 with phospholipid vesicles.
25(OH)D3/phospholipid
(mol/mol)
Vesicles
sedimented (%)
25(OH)D3
sedimented (%)
0.0025
0.020
0.100
99.6
99.6
99.8
97.5
97.4
97.0
Phospholipids vesicles were prepared by extrusion of an aqueous suspension of
phospholipid, [3H]25(OH)D3 and [14C]cholesterol through a 1.0 ␮m membrane and
then sedimented by ultracentrifugation. The proportion of vesicles sedimented was
determined from the amount of [14C]cholesterol in the vesicle pellet.
tion procedures quickly at 4 ◦C and storing the enzyme at −80 ◦C
between steps, as required.
In contrast to CYP27B1, mouse adrenodoxin expressed at high
levels providing 3.0 ␮mol of pure protein per litre culture. The
final preparation ran as a single band (14.5 kDa) on an SDS-
polyacrylamide gel stained with Coomassie blue (Fig. 1B), close to
the predicted size of 13.7 kDa. It displayed a typical adrenodoxin
absorption spectrum with a 414 nm/280 nm absorbance ratio of
0.88 (Fig. 2B).
Fig. 3. Substrate-induced difference spectrum of CYP27B1. CYP27B1 was added
to phospholipid vesicles alone (reference cuvette) or to phospholipid vesicles
containing 25(OH)D3 (test cuvette, 25(OH)D3/phospholipid = 0.05 mol/mol) and
the difference spectrum recorded. The phospholipid vesicles were prepared from
prepared from dioleoyl phosphatidylcholine and cardiolipin and were used at a
phospholipid concentration of 510 ␮M. The CYP27B1 concentration was 0.7 ␮M.
3.2. Association of 25(OH)D3 with membranes
Merz and Sternberg [20] have shown that 95.4% of 25(OH)D3
associates with egg phospholipid liposomes. We tested the
degree of association of 25-[26,27-3H]-hydroxyvitamin D3
([3H]25(OH)D3) with large phospholipid vesicles (liposomes) pre-
pared from dioleoyl phosphatidylcholine and cardiolipin, which
resemble the inner mitochondrial membrane. These large vesicles,
produced by extrusion through a 1.0 ␮m membrane, were quan-
titatively (greater than 99%) sedimented by ultracentrifugation.
At all three ratios of 25(OH)D3 to phospholipid tested, between
97.0% and 97.5% of the radiolabelled 25(OH)D3 was sedimented
with the vesicles (Table 1) showing that it strongly partitions into
the membrane bilayer.
3.3. Binding and metabolism of 25(OH)D3 in phospholipid
vesicles by CYP27B1
The addition of purified CYP27B1 to vesicles containing
25(OH)D3 caused a change towards the high spin state indicat-
ing that CYP27B1 can access substrate from the vesicle membrane
(Fig. 3). When adrenodoxin, adrenodoxin reductase and NADPH
were added to the vesicles, 0.5 ␮M CYP27B1 was able to metab-
olize 95% of the 25(OH)D3 in the vesicle membrane in 2 min
(Fig. 4). The product was identified as the expected 1,25(OH)2D3
based on an identical retention time to authentic standard by HPLC
(Fig. 4) and identical Rf values by TLC (not shown). Thus, CYP27B1
can rapidly access essentially all the 25(OH)D3 in both halves of
the phospholipid bilayer. A similar result was observed for 25-
hydroxyvitamin D2 (25(OH)D2) with the product being identified
as 1␣,25-dihydroxyvitamin D2 (1,25(OH)2D2) from authentic stan-
dard (not shown).
Time courses for the metabolism of 25(OH)D3 in phospholipid
vesicles by CYP27B1 were determined for large vesicles prepared by
extrusion through a 1.0 ␮m membrane and small unilamellar vesi-
cles prepared by sonication (Fig. 5). The time courses were linear for
2–3 min with the initial rate of product formation for small vesicles
being 50% higher than for large vesicles. The loss of linearity after
2–3 min of incubation appears to be due to the instability of the
CYP27B1. CO-reduced minus reduced difference spectra recorded,
prior to and at the end of a 10 min incubation at 37 ◦C, showed a
50% reduction in P450 concentration by the end of the incubation.
Fig. 4. Chromatograms showing metabolism of 25(OH)D3 in phospholipid vesicles
by CYP27B1. Small phospholipid vesicles were prepared from dioleoyl phosphatidyl-
choline and cardiolipin with 25(OH)D3 present at a molar ratio of 0.05 mol/mol
phospholipid, by sonication. Vesicles (510 ␮M phospholipid) were incubated with
CYP27B1 (0.5 ␮M) in the presence of bovine adrenodoxin reductase (0.4 ␮M), mouse
adrenodoxin (15 ␮M) and NADPH (50 ␮M) for 2 min at 37 ◦C, then the reaction
stopped and secosteroids extracted with dichloromethane. Substrate and products
were separated by reverse phase HPLC on a C18 column. (A) Zero time control where
the reaction mixture was extracted prior to starting the reaction with NADPH; (B)
test reaction.

E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
175
tion was observed (Fig. 6A). The data gave a good fit to a hyperbolic
curve up to a substrate concentration of 0.025 mol 25(OH)D3 per
mol phospholipid. This permitted estimation of the Km and kcat val-
ues which are listed in Table 2. When a similar kinetic analysis was
done using 25(OH)D2 as substrate, inhibition was also observed at
high substrate concentrations, but was much less marked than for
25(OH)D3 (Fig. 6B). The Km and kcat values from a hyperbolic curve
fitted to the data for 25(OH)D2 are listed in Table 2 and are similar
to the values for 25(OH)D3. The inhibition at high substrate concen-
trations was seen in three replicate experiments for both 25(OH)D3
and 25(OH)D2 and with three separate preparations of CYP27B1.
In contrast to phospholipid vesicles, no substrate inhibition was
seen when 25(OH)D3 was dissolved in 0.45% 2-hydroxypropyl-␤-
cyclodextrin (cyclodextrin) (Fig. 6C). In this system a good fit to a
hyperbolic curve was seen with substrate concentrations as high as
8Km. The kcat for CYP27B1 in the cyclodextrin system was approxi-
mately 30% of that seen in vesicles. We have previously shown that
the cyclodextrin concentration markedly affects both the kcat and
Km for metabolism of vitamin D3 and cholesterol by CYP11A1 [29].
The cyclodextrin forms a complex with the vitamin D substrate and
holds it in solution and catalytic activity is influenced by the rela-
tive binding of the substrate to the cyclodextrin and the active site
of the enzyme [29].
Substrate inhibition has been observed previously for microso-
mal P450s with a number of substrates [34–36]. Equations based
on one catalytic and one inhibitory site for substrate binding have
been derived to describe this inhibition but gave a poor fit to our
experimental data. A reasonable fit was given by an equation we
developed which assumes one catalytic and two inhibitory sites
on the P450 for substrate (see Section 2). The curve fitted to the
data for 25(OH)D3 using this equation (Fig. 6A) gave a Ki value
approximately 4Km (Table 2). Km and kcat values were not signif-
icantly different from those provided by the hyperbolic curve fit
(Table 2) due to the large standard errors accompanying the sub-
strate inhibition model. The substrate inhibition model fitted the
data for 25(OH)D2 very well (Fig. 6B) giving Km and kcat values
close to those determined from the hyperbolic curve fit, and a high
Ki relative to Km (Table 2).
Fig. 5. Comparison of time courses for hydroxylation of 25(OH)D3 in small and
large phospholipid vesicles by CYP27B1. Phospholipid vesicles were prepared from
dioleoyl phosphatidylcholine and cardiolipin with 25(OH)D3 present at a molar ratio
of 0.15 mol/mol phospholipid, by sonication (small vesicles) or extrusion through a
1.0 ␮m membrane (large vesicles). Vesicles were incubated with 12.5 nM CYP27B1
in a reconstituted system as in Fig. 4 and the amount of 1,25(OH)2D3 formed mea-
sured by HPLC.
3.4. Comparison between mouse and bovine adrenodoxin for
supporting the activity of mouse CYP27B1
Since other researchers have used bovine adrenodoxin to mea-
sure the catalytic activity of mouse CYP27B1 [9,23,41], in contrast
to our study employing expressed mouse adrenodoxin, we com-
pared the activity of mouse CYP27B1 supported by the two different
species of adrenodoxin. Mouse and bovine adrenodoxin did dis-
play similar abilities to support CYP27B1 activity. At a 25(OH)D3
concentration of 0.035 mol/mol phospholipid, mouse adrenodoxin
gave Km and kcat values of 3.2 0.3 ␮M and 34.2 0.8 mol/min/mol
CYP27B1, respectively, while Km and kcat values for bovine adren-
odoxin were 4.6 1.2 ␮M and 35.8 2.5 mol/min/mol CYP27B1,
respectively.
We also determined the kinetics of metabolism of two
other hydroxyvitamin
dihydroxyvitamin D3 (24,25(OH)2D3) was metabolized to
1␣,24R,25-trihydroxyvitamin D3 with similar kcat to that
for 25(OH)D3, but with a Km 4.4-fold higher and therefore a
correspondingly lower kcat/Km No substrate inhibition was
D compounds by CYP27B1. 24R,25-
a
3.5. Kinetics of 1˛-hydroxylation of different substrates by
CYP27B1
.
observed at the maximum substrate concentration tested (2Km).
20-Hydroxyvitamin D3 (20(OH)D3) is a product of CYP11A1
action on vitamin D3 [15,17,18,29] and is biologically active on
When the hydroxylation of 25(OH)D3 was measured as a
function of increasing 25(OH)D3 concentration the reaction rate
increased, but as substrate became near-saturating strong inhibi-
Fig. 6. The effect of substrate concentration on the rate of hydroxylation of 25(OH)D3 and 25(OH)D2 by CYP27B1 in phospholipid vesicles. Reaction rates were determined
for 25(OH)D3 (A) and 25(OH)D2 (B) using substrates incorporated into small phospholipid vesicles prepared by sonication of dioleoyl phosphatidylcholine and cardiolipin, in
a 2 min incubation as described in the Experimental Procedures. The activity of CYP27B1 was also measured for 25(OH)D3 in 0.45% cyclodextrin (C). Data for high substrate
concentrations where inhibition was observed were excluded from the Michaelis–Menten fits (solid lines). A curve for a model describing substrate inhibition was also fitted
to the data for vesicles as described in Section 2 (dashed lines).

176
E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
Table 2
Kinetic parameters for metabolism of different substrates in phospholipid vesicles by CYP27B1.
Substrate
Km (mmol/mol PL)
Ki (mmol/mol PL)
kcat (min−1
41
)
kcat/Km ((mmol/mol PL)−1 min−1
)
25(OH)D3a
25(OH)D3b
25(OH)D2a
25(OH)D2b
24,25(OH)2D3a
20(OH)D3a
5.9 1.2
9.7 5.3
4.6 0.8
4.8 0.7
3
6.9 1.9
5.7 4.6
10.4 2.2
10.2 1.9
2.1 0.6
37 10
55 15
48
49
54
2
2
5
204 36
26
5
49 21
8.5 1.7
0.17 0.11
CYP27B1 activity was determined with substrates in small unilamellar phospholipid (PL) vesicles as described in Fig. 7. Data are mean SE of the curve fit for representative
experiments.
a
Kinetic parameters determined by Michaelis–Menten curve fit.
Kinetic parameters determined by the kinetic model with substrate binding at two inhibitory sites.
b
Table 3
The effect of vesicle composition on the kinetic parameters for metabolism of 25(OH)D3.
Phospholipid
Km (mmol/mol PL)
kcat (min−1
)
kcat/Km ((mmol/mol PL)−1 min−1
)
DOPC
DOPC + cardiolipin
DMPC
8.2 2.0
4.6 1.4
15.8 2.1
49
41
11
4
4
1
5.9 1.9
8.9 3.6
0.70 0.16
Vesicles were prepared by sonication and kinetic parameters determined from Michaelis–Menten curve fits, excluding data above 4Km where substrate inhibition was
observed. DOPC, dioleoyl phosphatidylcholine. DMPC, dimyristoyl phosphatidylcholine. Cardiolipin, when present, was 15% of the total moles of phospholipid.
skin cells working as a partial receptor agonist for the vitamin
D3 receptor [37,38]. It was metabolized by CYP27B1 to 1␣,20-
dihydroxyvitamin D3, identified from its identical retention time
with authentic standard by HPLC. It displayed a Km 8-fold higher
than for 25(OH)D3 and a kcat 5-fold lower giving it a 40-fold lower
kcat/Km value (Table 2). No substrate inhibition was observed at
the maximum substrate concentration tested (2Km).
4. Discussion
We obtained relatively low expression of mouse CYP27B1 com-
pared to that reported by Uchida et al. [9] using a similar procedure.
These workers reported a 10-fold increase in expression to levels
as high as 300 nmol/L culture by co-expression with chaperones
GroEL/ES but in our hands these chaperones only marginally
increased CYP27B1 expression. Despite the low expression we were
able to obtain a reasonably pure preparation of enzyme for catalytic
studies from a combination of Ni affinity-, adrenodoxin affinity-
and octyl Sepharose-chromatography. Mouse adrenodoxin was
expressed in E. coli, purified and used for the first time as the elec-
tron donor in assays of mouse CYP27B1 activity.
3.6. The effect of vesicle phospholipid composition on metabolism
of 25(OH)D3 by CYP27B1
The value for kcat/Km for 25(OH)D3 in vesicles made from
dioleoyl phosphatidylcholine was slightly lower than in vesi-
cles comprising dioleoyl phosphatidylcholine plus 15% cardiolipin
(Table 3). This was due to cardiolipin causing at 44% decrease
in the Km for 25(OH)D3 with little effect on kcat. In contrast,
vesicles prepared from dimyristoyl phosphatidylcholine gave a
12.7-fold lower kcat/Km for 25(OH)D3 metabolism than that seen
with dioleoyl phosphatidylcholine and cardiolipin, caused by an
increase in Km and a decrease in kcat. Substrate inhibition at high
substrate concentrations (above 4Km) was observed for all three
vesicle compositions analyzed.
Our study shows that mouse CYP27B1 efficiently utilizes sub-
strates incorporated into the bilayer of phospholipid vesicles of a
composition reflecting the inner mitochondrial membrane where
the enzyme is located in vivo [39,40]. The kcat of CYP27B1 for
membrane-associated 25(OH)D3 in our study (41–55 min−1) is
4–20-fold higher than previous reports of the kcat for the purified
enzyme measured in detergent micelles [8,9,41]. Higher CYP27B1
activity was seen with small phospholipid vesicles prepared by son-
ication than with large vesicles prepared by extrusion, as we have
observed for CYP11A1 [42]. Vesicle size has previously been shown
to affect the activity of CYP11A1 in a similar fashion and is sug-
gested to be due to the high radius of curvature of small vesicles
causing looser packing of lipids in the outer half of the bilayer [42].
Attempts to address whether purified CYP27B1 permanently or
transiently interacts with the vesicle membrane were made using
gel filtration and centrifugation to separate free cytochrome from
vesicle-bound cytochrome, but proved unsuccessful because of the
lability of the enzyme. Phospholipid composition was observed
to strongly influence CYP27B1 activity with a lower kcat and
higher Km for 25(OH)D3 being observed in vesicles made from
phospholipids containing saturated fatty acid chains (dimyristoyl
phosphatidylcholine) than those with mono-unsaturated chains
(dioleoyl phosphatidylcholine). The presence of the mitochon-
drial phospholipid, cardiolipin, caused a decrease in the Km for
25(OH)D3. Similarly, under comparable conditions cardiolipin has
been shown to reduce the Km for cholesterol of CYP11A1, another
mitochondrial P450, by 2–3-fold, with no effect on kcat [40,43].
This study is the first to compare the metabolism of 25(OH)D3
and 25(OH)D2 by purified CYP27B1. Results show that they are both
metabolized with similar kcat and Km values with the major differ-
3.7. Testing metabolism of vitamin D and 1˛-hydroxyvitamin D
by CYP27B1
Since both 1␣-hydroxyvitamin D3 (1(OH)D3) and vitamin D3
itself have been reported to serve as substrates for CYP27B1 in
a detergent reconstituted system [9], we tested the metabolism
of these two substrates plus their vitamin D2 counterparts, in
phospholipid vesicles. Incubations were carried out for 20 min
with vesicles prepared from dioleoyl phosphatidylcholine and
cardiolipin using a high concentration of CYP27B1 (0.2 ␮M). No
metabolism of either vitamin D2 or D3 was observed (not shown).
Both 1(OH)D2 and 1(OH)D3 were hydroxylated in the 25-position
producing 1,25(OH)2D2 and 1,25(OH)2D3, respectively (Fig. 7).
While a detailed kinetic analysis was not carried out, we mea-
sured the initial rate of hydroxylation of both 1(OH)D2 and 1(OH)D3
at molar ratios to phospholipid of 0.025 in a 2 min incubation.
Rates were 0.73 and 0.28 mol/min/mol CYP27B1 for 1(OH)D3 and
1(OH)D2, respectively, both low compared to 1␣-hydroxylation of
25(OH)D2 and 25(OH)D3.

E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
177
Fig. 7. Chromatograms showing metabolism of 1(OH)D3 and 1(OH)D2 by CYP27B1. Vesicles were prepared from dioleoyl phosphatidylcholine and cardiolipin with 1(OH)D3
(A, C, E) or 1(OH)D2 (B, D, F) present at ratio to phospholipid of 0.15 mol/mol. Vesicles were incubated with CYP27B1 (0.2 ␮M) for 20 min as in Fig. 4 and the products analyzed
by HPLC. No adrenodoxin was present for control reactions (A, B). Test reactions contained adrenodoxin (C, D). The identity of products was confirmed by spiking test reaction
extracts prior to HPLC with 0.3 nmol standard 1,25(OH)2D3 (E) or 0.09 nmol 1,25(OH)2D2 (F).
ence being that more pronounced substrate inhibition is observed
for 25(OH)D3. Our data for the relative rates of 25(OH)D2 and
25(OH)D3 hydroxylation by CYP27B1 are consistent with early
studies with kidney mitochondria from both the rat and chick,
where comparable rates of metabolism of 25(OH)D2 and 25(OH)D3,
presumably by CYP27B1, were reported [44]. 24,25(OH)2D3 is
a metabolite of vitamin D3 found in the bloodstream result-
ing from the action of CYP24 on 25(OH)D3 and may have some
direct biological activity on bone or cartilage [2]. Our study
shows that 24,25(OH)2D3 in vesicles is metabolized with a sim-
ilar kcat to that for 25(OH)D3 but with a 4-fold higher Km, and
hence with a lower kcat/Km. Thus, at low substrate concentrations
CYP27B1 will hydroxylate 25(OH)D2 and 25(OH)D3 in prefer-
ence to 24,25(OH)2D3. In contrast, Sakaki et al. [23] reported that
24,25(OH)2D3 has a higher kcat/Km than 25(OH)D3 for expressed
CYP27B1 in E. coli membranes. Presumably this difference is due to
the different reconstituted systems used between the two stud-
ies. In the vesicle system, the presence of a hydroxyl group at
either C20 or C24 markedly reduces the catalytic efficiency of 1␣-
hydroxylation suggesting that interaction of the side chain of the
substrate with the enzyme plays an important role in positioning
C1 near the heme group for optimal hydroxylation at this position.
However, the presence of a hydroxyl group at C20 or C24 does not
change the position of hydroxylation from the 1␣-position.
ing cancer. This study shows that 20(OH)D3 is a poor substrate for
CYP27B1 with a kcat/Km value only 2.5% of that for 25(OH)D3. Thus,
this vitamin D derivative will compete poorly with 1,25(OH)2D3 for
metabolism by CYP27B1 and is likely to undergo 1␣-hydroxylation
very slowly in vivo. It is interesting to note that the product,
1,20(OH)2D3 does show some calcemic activity, although less than
1,25(OH)2D3 [45].
The current study on CYP27B1 was done on the mouse enzyme
because of the reported difficulties in expressing and solubiliz-
ing human P450scc [46,47]. Human CYP27B1 has been expressed
in E. coli and its activity measured in E. coli membranes, but it
has never been successfully extracted or purified [47]. Preliminary
studies in our laboratory confirm this difficulty of solubiliz-
ing the expressed human enzyme in active form for catalytic
studies.
We have previously shown that there is a species difference
in the ability of adrenodoxin to support the activity of human
CYP11A1, with human adrenodoxin producing higher activity than
bovine adrenodoxin [43]. The current investigation shows that
CYP27B1 activity is not influenced by the adrenodoxin species, with
mouse and bovine adrenodoxin being about equally effective in
supporting 1␣-hydroxylase activity of mouse CYP27B1. Our Km for
mouse adrenodoxin with mouse CYP27B1 is 4-fold higher than the
value we observed for bovine adrenodoxin with bovine CYP11A1
under comparable conditions [21]. In tissues expressing a relatively
high level of CYP11A1 such as the adrenal cortex, corpus luteum or
placenta, adrenodoxin levels are high andare believed to be saturat-
ing, or at least near-saturating, for CYP11A1 activity [10,48–50]. The
relatively high Km of mouse CYP27B1 for adrenodoxin and the low
concentrations of adrenodoxin in kidney mitochondria [51] sug-
gest that adrenodoxin concentration may limit CYP27B1 activity in
vivo. In support of this, Eto et al. [52] have shown that the activity
20(OH)D3 is the major product of CYP11A1 action on vitamin
D3 [15,17,18,29] and is biologically active on skin and other cells
acting as an inhibitor of proliferation and a promoter of differen-
tiation [37,38]. It acts as a partial receptor agonist for the vitamin
D receptor and unlike 1,25(OH)2D3, it only weakly stimulates the
expression of the CYP24 gene and does not raise serum calcium
levels in rats [37,38,45]. These properties make this compound of
interest for the treatment of hyperproliferative disorders includ-

178
E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
of CYP27B1 in rat kidney mitochondria is greatly increased when
the mitochondria are solubilized with detergent and supplemented
with exogenous adrenodoxin reductase and adrenodoxin.
[4] R. Bouillon, G. Eelen, L. Verlinden, C. Mathieu, G. Carmeliet, A. Verstuyf, Vitamin
D and cancer, J. Steroid Biochem. Mol. Biol. 102 (1–5) (2006) 156–162.
[5] D.D. Bikle, M.K. Nemanic, E. Gee, P. Elias, 1,25-Dihydroxyvitamin D3 production
by human keratinocytes, J. Clin. Invest 78 (2) (1986) 557–566.
Uchida et al. [9] reported that CYP27B1 in detergent micelles can
hydroxylate vitamin D3 to metabolites identified as 1(OH)D3 and
1,25(OH)2D3 based on HPLC retention times. They also reported
that the CYP27B1 can hydroxylate 1(OH)D3 in the 25-position.
Thus, they proposed that CYP27B1 alone can convert vitamin D3
to the hormonally active form, 1,25(OH)2D3. We could not detect
any metabolism of either vitamin D3 or D2 in vesicles by CYP27B1.
Since we used a high CYP27B1 concentration, we can exclude any
metabolism at a rate above our minimum detection level of approx-
imately 0.03 mol/min/mol of CYP27B1 or 0.075% of the kcat seen
with 25(OH)D3. In agreement with Uchida et al. [9], we observed
that CYP27B1 can slowly hydroxylate 1(OH)D3 in the 25-position
and we have further shown that this reaction can also occur with
1(OH)D2. Given that we could not detect 1␣-hydroxylation of vita-
min D3 by CYP27B1 and that 1(OH)D3 has not been detected as a
circulating form of vitamin D3 [9], the physiological relevance for
this 25-hydroxylation of 1(OH)D3 is questionable.
[6] M.F. Holick, Vitamin D: a millennium perspective, J. Cell Biochem. 88 (2) (2003)
296–307.
[7] A. Hiwatashi, Y. Nishii, Y. Ichikawa, Purification of cytochrome P-450D1a (25-
hydroxyvitamin D3-1␣-hydroxylase) of bovine kidney mitochondria, Biochem.
Biophys. Res. Commun. 105 (1) (1982) 320–327.
[8] Y. Nakamura, T. Eto, T. Taniguchi, K. Miyamoto, J. Nagatomo, H. Shiotsuki, H.
Sueta, S. Higashi, K. Okuda, T. Setoguchi, Purification and characterization of 25-
hydroxyvitamin D3 1␣-hydroxylase from rat kidney mitochondria, FEBS Lett.
419 (1) (1997) 45–48.
[9] E. Uchida, N. Kagawa, T. Sakaki, N. Urushino, N. Sawada, M. Kamakura, M. Ohta,
S. Kato, K. Inouye, Purification and characterization of mouse CYP27B1 over-
produced by an Escherichia coli system coexpressing molecular chaperonins
GroEL/ES, Biochem. Biophys. Res. Commun. 323 (2) (2004) 505–511.
[10] R.C. Tuckey, Progesterone synthesis by the human placenta, Placenta 26 (4)
(2005) 273–281.
[11] M.J. Headlam, M.C.J. Wilce, R.C. Tuckey, The F-G loop region of cytochrome
P450scc (CYP11A1) interacts with the phospholipid membrane, Biochim. Bio-
phys. Acta Biomembr. 1617 (1–2) (2003) 96–108.
[12] I.A. Pikuleva, Cholesterol-metabolizing cytochromes P450, Drug Metab. Dispos.
34 (4) (2006) 513–520.
[13] I.A. Pikuleva, N. Mast, W. Liao, I.V. Turko, Studies of membrane topology of
mitochondrial cholesterol hydroxylases CYPs 27A1 and 11A1, Lipids 43 (12)
(2008) 1127–1132.
The mechanism by which high concentrations of 25(OH)D3,
and to a lesser degree 25(OH)D2, cause substrate inhibition is
unclear. It is unlikely that it is a non-specific mechanism, such
as perturbation of the phospholipid membrane, since the struc-
turally similar 25(OH)D3 and 25(OH)D2 display markedly different
potencies for inhibition, with Ki values of 37 and 204 mmol/mol
phospholipid, respectively. Furthermore, 20(OH)D3 did not cause
any inhibition at a molar ratio to phospholipid of 0.1, where strong
inhibition was seen with 25(OH)D3. Substrate inhibition of micro-
somal drug-metabolizing P450s is well established, and in some
cases is caused by two substrates occupying the active site, with one
being inhibitory [34–36]. One-inhibitor site models gave a poor fit
to our experimental data. The best fit was given by a two-inhibitory
site model with each site being filled simultaneously with equal Ki
values. Some involvement of the phospholipid bilayer in the mech-
anism is apparent since substrate inhibition was not seen when the
25(OH)D3 was dissolved in cyclodextrin. No substrate inhibition
has been reported for 25(OH)D3 in detergent micelles [9,41,46,53],
however, it is unclear whether substrate concentrations as high as
the required 6Km were tested in these studies.
The concentrations of 25(OH)D3 and 25(OH)D2 in the inner
mitochondrial membrane of kidney mitochondria (and other mito-
chondria that contain CYP27B1) are unknown, but are likely to be
low. The low Km that CYP27B1 displays for these substrates in the
membrane is compatible with the enzyme being active with low
substrate concentrations. As 25(OH)D3 concentrations rise to near-
saturating levels, the resulting substrate inhibition may provide a
mechanism to ensure excessive amounts of the hormonally active
1,25(OH)2D3 are not generated. This mechanism may be impor-
tant for the short-term regulation of 1,25(OH)2D3 levels, providing
a more rapid response than the well characterized transcriptional
regulation of the CYP27B1 and CYP24 genes [2].
[14] K. Storbeck, P. Swart, A.C. Swart, Cytochrome P450 side-chain cleavage: insights
gained from homology modeling, Mol. Cell Endocrinol. 265–266 (2007) 65–70.
[15] O. Guryev, R.A. Carvalho, S. Usanov, A. Gilep, R.W. Estabrook, A pathway for the
metabolism of vitamin D3: unique hydroxylated metabolites formed during
catalysis with cytochrome P450scc (CYP11A1), Proc. Natl. Acad. Sci. U.S.A. 100
(25) (2003) 14754–14759.
[16] M.N. Nguyen, A. Slominski, W. Li, Y.R. Ng, R.C. Tuckey, Metabolism of vitamin
D2 to 17,20,24-trihydroxyvitamin D2 by cytochrome P450scc (CYP11A1), Drug
Metab. Dispos. 37 (4) (2009) 761–767.
[17] A. Slominski, I. Semak, J. Zjawiony, J. Wortsman, W. Li, A. Szczesniewski, R.-C.
Tuckey, The cytochrome P450scc system opens an alternate pathway of vitamin
D3 metabolism, FEBS J. 272 (16) (2005) 4080–4090.
[18] R.C. Tuckey, W. Li, J.K. Zjawiony, M.A. Zmijewski, M.N. Nguyen, T. Sweatman,
D. Miller, A. Slominski, Pathways and products for the metabolism of vitamin
D3 by cytochrome P450scc (CYP11A1), FEBS J. 275 (10) (2008) 2585–2596.
[19] J.D. Lambeth, S.E. Kitchen, A.A. Farooqui, R. Tuckey, H. Kamin, Cytochrome P-
450scc-substrate interactions. Studies of binding and catalytic activity using
hydroxycholesterols, J. Biol. Chem. 257 (4) (1982) 1876–1884.
[20] K. Merz, B. Sternberg, Incorporation of vitamin D3-derivatives into liposomes
of different lipid types, J. Drug Target. 2 (5) (1994) 411–417.
[21] R.C. Tuckey, P.M. Stevenson, Properties of ferredoxin reductase and ferredoxin
from the bovine corpus luteum, Int. J. Biochem. 16 (5) (1984) 489–495.
[22] K. Takeyama, S. Kitanaka, T. Sato, M. Kobori, J. Yanagisawa, S. Kato, 25-
Hydroxyvitamin D3 1␣-hydroxylase and vitamin D synthesis, Science 277
(5333) (1997) 1827–1830.
[23] T. Sakaki, N. Sawada, K.I. Takeyama, S. Kato, K. Inouye, Enzymatic properties
of mouse 25-hydroxyvitamin D3 1␣-hydroxylase expressed in Escherichia coli,
Eur. J. Biochem. 259 (3) (1999) 731–738.
[24] A. Wada, M.R. Waterman, Identification by site-directed mutagenesis of two
lysine residues in cholesterol side chain cleavage cytochrome P450 that are
essential for adrenodoxin binding, J. Biol. Chem. 267 (32) (1992) 22877–22882.
[25] M. Stromstedt, M.R. Waterman, A full-length cDNA encoding mouse adreno-
doxin, Biochim. Biophys. Acta 1261 (1) (1995) 126–128.
[26] S.T. Woods, J. Sadleir, T. Downs, T. Triantopoulos, M.J. Headlam, R.C. Tuckey,
Expression of catalytically active human cytochrome P450scc in Escherichia
coli and mutagenesis of isoleucine-462, Arch. Biochem. Biophys. 353 (1) (1998)
109–115.
[27] J.D. Lambeth, H. Kamin, D.W. Seybert, Phosphatidylcholine vesicle reconsti-
tuted P-450scc. Role of the membrane in control of activity and spin state of
the cytochrome, J. Biol. Chem. 255 (17) (1980) 8282–8288.
[28] R.C. Tuckey, Z. Janjetovic, W. Li, M.N. Nguyen, M.A. Zmijewski, J.K. Zjawiony,
A. Slominski, Metabolism of 1␣-hydroxyvitamin D3 by cytochrome P450scc to
biologically active 1␣,20-dihydroxyvitamin D3, J. Steroid Biochem. Mol. Biol.
112 (4–5) (2008) 213–219.
Acknowledgements
[29] R.C. Tuckey, M.N. Nguyen, A. Slominski, Kinetics of vitamin D3 metabolism by
cytochrome P450scc (CYP11A1) in phospholipid vesicles and cyclodextrin, Int.
J. Biochem. Cell Biol. 40 (11) (2008) 2619–2626.
[30] T. Omura, R. Sato, The carbon monoxide binding pigment of liver microsomes.
I. Evidence for its hemoprotein nature, J. Biol. Chem. 239 (7) (1964) 2370–2378.
[31] R.C. Tuckey, J.F. Holland, Comparison of pregnenolone synthesis by cytochrome
P-450scc in mitochondria from porcine corpora lutea and granulosa cells of
follicles, J. Biol. Chem. 264 (10) (1989) 5704–5709.
This work was supported by the University of Western Australia
Research Grants Scheme. We thank Prof. Paul Attwood for help in
deriving kinetic equations to describe substrate inhibition.
References
[32] M.A. Pesce, C.S. Strande, A new micromethod for determination of protein in
cerebrospinal fluid and urine, Clin. Chem. 19 (11) (1973) 1265–1267.
[33] I. Hanukoglu, C.R. Jefcoate, Mitochondrial cytochrome P-450scc. Mechanism of
electron transport by adrenodoxin, J. Biol. Chem. 255 (7) (1980) 2057–3061.
[34] K.R. Korzekwa, N. Krishnamachary, M. Shou, A. Ogai, R.A. Parise, A.E. Ret-
tie, F.J. Gonzalez, T.S. Tracy, Evaluation of atypical cytochrome P450 kinetics
with two-substrate models: evidence that multiple substrates can simulta-
[1] M.F. Holick, Vitamin D: importance in the prevention of cancers, type 1 dia-
betes, heart disease, and osteoporosis, Am. J. Clin. Nutr. 79 (3) (2004) 362–371.
[2] D.E. Prosser, G. Jones, Enzymes involved in the activation and inactivation of
vitamin D, Trends Biochem. Sci. 29 (12) (2004) 664–673.
[3] D. Bikle, Nonclassic actions of vitamin D, J. Clin. Endocrinol. Metab. 94 (1) (2009)
26–34.

E.K.Y. Tang et al. / Journal of Steroid Biochemistry & Molecular Biology 119 (2010) 171–179
179
neously bind to cytochrome P450 active sites, Biochemistry 37 (12) (1998)
4137–4147.
[35] Y. Lin, P. Lu, C. Tang, Q. Mei, G. Sandig, A.D. Rodrigues, T.H. Rushmore, M. Shou,
Substrate inhibition kinetics for cytochrome P450-catalysed reactions, Drug
Metab. Dispos. 29 (4) (2001) 368–374.
[45] A.T. Slominski, Z. Janjetovic, B. Fuller, M.A. Zmijewski, R.C. Tuckey, M.N. Nguyen,
T. Sweatman, W. Li, J. Zjawiony, D. Miller, T.C. Chen, G. Lozanski, M.F. Holick,
Products of vitamin D3 or 7-dehydrocholesterol metabolism by cytochrome
P450scc show anti-leukemia effects, having low or absent calcemic activity,
PloS ONE (2010), in press.
[36] S. Zhou, Drugs behave as substrates, inhibitors and inducers of human
cytochrome P450 3A4, Curr. Drug Metab. 9 (4) (2008) 310–322.
[37] Z. Janjetovic, M.A. Zmijewski, R.C. Tuckey, D.A. DeLeon, M.N. Nguyen,
L.M. Pfeffer, A.T. Slominski, 20-Hydroxycholecalciferol, product of vita-
min D3 hydroxylation by P450scc, decreases NF-␬B activity by increasing
[46] K. Yamamoto, E. Uchida, N. Urushino, T. Sakaki, N. Kagawa, N. Sawada, M.
Kamakura, S. Kato, K. Inouye, S. Yamada, Identification of the amino acid residue
of CYP27B1 responsible for binding of 25-hydroxyvitamin D3 whose muta-
tion causes vitamin D-dependent rickets type 1, J. Biol. Chem. 280 (34) (2005)
30511–30516.
I␬B␣ levels in human keratinocytes, PLoS ONE
doi:10.1371/journal.pone.0005988.
4
(6) (2009) e5988,
[47] N. Sawada, T. Sakaki, S. Kitanaka, K. Takeyama, S. Kato, K. Inouye, Enzymatic
properties of human 25-hydroxyvitamin D3 1␣-hydroxylase. Coexpression
with adrenodoxin and NADPH-adrenodoxin resuctase in Escherichia coli, Eur. J.
Biochem. 265 (1999) 950–956.
[48] I. Hanukoglu, Z. Hanukoglu, Stoichiometry of mitochondrial cytochromes P-
450, adrenodoxin and adrenodoxin reductase in adrenal cortex and corpus
luteum. Implications for membrane organization and gene regulation, Eur. J.
Biochem. 157 (1) (1986) 27–31.
[49] R.C. Tuckey, Z. Kostadinovic, P.M. Stevenson, Ferredoxin and cytochrome
P-450scc concentrations in granulosa cells of porcine ovaries during follic-
ular cell growth and luteinisation, J. Steroid Biochem. 31 (2) (1988) 201–
205.
[50] R.C. Tuckey, S.T. Woods, M. Tajbakhsh, Electron transfer to cytochrome P-
450scc limits cholesterol side-chain-cleavage activity in the human placenta,
Eur. J. Biochem. 244 (3) (1997) 835–839.
[51] N. Maruya, A. Hiwatashi, Y. Ichikawa, T. Yamano, Purification and characteri-
zation of renal ferredoxin form bovine renal mitochondria, J. Biochem. 93 (5)
(1983) 1239–1247.
[52] T.A. Eto, Y. Nakamura, T. Taniguchi, K. Miyamoto, J. Nagatomo, Y. Maeda,
S. Higashi, K. Okuda, T. Setoguchi, Assay of 25-hydroxyvitamin D3 1␣-
hydroxylase in rat kidney mitochondria, Anal. Biochem. 258 (1) (1998) 53–
58.
[53] N. Urushino, S. Nakabayashi, M.A. Arai, A. Kittaka, T.C. Chen, K. Yamamoto, K.
Hayashi, S. Kato, M. Ohta, M. Kamakura, S. Ikushiro, T. Sakaki, Kinetic stud-
ies of 25-hydroxy-19-nor-vitamin D3 and 1a,25-dihydroxy-19-nor-vitamin D3
hydroxylation by CYP27B1 and CYP24A1, Drug Metab. Dispos. 35 (9) (2007)
1482–1488.
[38] B. Zbytek, Z. Janjetovic, R.C. Tuckey, M.A. Zmijewski, T.W. Sweatman, E. Jones,
M.N. Nguyen, A.T. Slominski, 20-hydroxyvitamin D3, a product of vitamin D3
hydroxylation by cytochrome P450scc, stimulates keratinocytes differentia-
tion, J. Invest. Dermatol. 128 (2008) 2271–2280.
[39] N.S. Cunningham, B.S. Lee, H.L. Henry, The renal mitochondrial metabolism of
25-hydroxyvitamin D-3: a possible role for phospholipids, Biochim. Biophys.
Acta 881 (1) (1986) 480–488, 64.
[40] R.C. Tuckey, P.M. Stevenson, Purification and analysis of phospholipids in the
inner mitochondrial membrane fraction of the bovine corpus luteum, and prop-
erties of cytochrome P-450scc incorporated into vesicles prepared from these
phospholipids, Eur. J. Biochem. 148 (2) (1985) 379–384.
[41] N. Urushino, K. Yamamoto, N. Kagawa, S. Ikushiro, M. Kamakura, S. Yamada,
S. Kato, K. Inouye, T. Sakaki, Interaction between mitochondrial CYP27B1 and
adrenodoxin: role of arginine 458 of mouse CYP27B1, Biochemistry 45 (14)
(2006) 4405–4412.
[42] R.C. Tuckey, H. Kamin, Kinetics of the incorporation of adrenal cytochrome
P-450scc into phosphatidylcholine vesicles, J. Biol. Chem. 257 (6) (1982)
2887–2893.
[43] P. Kisselev, R.C. Tuckey, S. Woods, T. Triantopolous, D. Schwarz, Enzymatic
properties of vesicle-reconstituted human cytochrome P450scc (CYP11A1).
Differences in functioning of the mitochondrial electron transfer chain using
human and bovine adrenodoxin and activation by cardiolipin, Eur. J. Biochem.
260 (3) (1999) 768–773.
[44] G. Jones, H. Schnoes, H.F. DeLuca, Isolation and identification of 1,25-
dihydroxyvitamin D2, Biochemistry 14 (6) (1975) 1250–1256.