ThefirstcytochromeP450in ferns
Evidence foritsinvolvementinphytoecdysteroid biosynthesis
in Polypodium vulgare
Daniel Canals, Josep Irurre-Santilari and Josefina Casas
Department of Biological Organic Chemistry, IIQAB-CSIC, Barcelona, Spain
Ecdysteroids are involved at every stage inthe life-
cycle of arthropods, regulating molting, metamor-
phosis, development, reproduction and many of the
physiological ⁄ biochemical processes associated with
those events. In insects, the prothoracic glands secrete
ecdysone (E), which is subsequently transformed into
the physiologically more active ecdysteroid 20-hydroxy-
ecdysone (20E) [1]. Moreover, ecdysteroids are also
present in plants (phytoecdysteroids; PE), but, cur-
rently, no phylogenetic pattern classifies plants accord-
ing to PE content, because PEs are found widely in
ferns, gymnosperms and angiosperms [2].
Several authors have suggested that PE may partici-
pate inthe defence of plants against nonadapted phy-
tofagous invertebrates [3]. Ingested PE would conduct
to an anomalous hormonal profile that may cause mal-
formations, sterility or insect death. Moreover, it has
been shown that ecdysteroid accumulation in spinach
can be induced by mechanical or insect damage to
roots, which fits with a role of PE in plant defence [4].
Because PE are apparently nonhazardous to verteb-
rate species, but are toxic to insects, they have been
considered as attractive insecticides, without environ-
mental pollution implications.
Polypodium vulgare is a fern with a worldwide distri-
bution that usually grows in dark and humid areas,
and produces PE, the highest content being in rhi-
zomes (up to 0.4% dry weight). Different culture lines
of prothalli and calli from this plant have been
obtained and their PE content has been described by
our group [5–10]. Among those models, a calli line
with undetectable levels of PE, but that is able to
transform E into 20E, was selected for this study.
Eight different PEs have been found to date in
P. vulgare (Fig. S1), which mainly differ inthe extent
and position of hydroxylation. These oxidative reac-
tions take place over the steroidal backbone and are
similar to those described for brassinosteroids, which
are catalysed by cytochromeP450 enzymes [11,12]. In
insects, ecdysone 20-hydroxylase is a cytochrome
P450 enzyme that may be located inthe microsomes
and ⁄ or mitochondria, depending on the species and
Keywords
cytochrome P450; ferns; phytoecdysteroids;
sterol pathways
Correspondence
J. Casas, Department of Biological Organic
Chemistry, IIQAB-CSIC, Jordi Girona,
18-26, 08034 Barcelona, Spain
Fax: +34 93 204 5904
Tel: +34 93 400 6100
E-mail: jcbqob@cid.csic.es
(Received 14 March 2005, revised 19 July
2005, accepted 3 August 2005)
doi:10.1111/j.1742-4658.2005.04897.x
The fern Polypodiumvulgare is a phytoecdysteroid (PE)-producing plant.
Cultures of P. vulgare prothalus produce PE, whereas prothalus-derived
callus cultures do not. However, this callus line can transform topically
applied ecdysone (E) to 20-hydroxyecdysone (20E), which is the last step in
the biosynthetic pathway of the main plant PE. This hydroxylation is cata-
lysed by a cytochromeP450 enzyme. E treatment of the callus line results
in an increased amount of P450, showing a linear correspondence between
the amount of P450 and in vivo E 20-hydroxylation activity, estimated by
measuring the bioconversion of E to 20E. This activity can be inhibited by
molecules that bind to the P450-heme group. E shows a P450-substrate-
binding spectrum with microsomes that overexpress theP450 protein.
Finally, a P450 protein was purified from E-treated calli, this being the first
P450 to be described inthe pterydophyte group.
Abbreviations
E, ecdysone, 2b,3b,14a,22R,25-pentahydroxy-5b-cholest-7-en-6-one; 20E, 20-hydroxyecdysone; PEs, phytoecdysteroids.
FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS 4817
tissue [13]. Likewise, it has been reported that micro-
somes from spinach (a PE producer plant) catalyse the
hydroxylation of ecdysone to 20E. This hydroxylation
is dependent on NADPH and molecular oxygen, and is
inhibited by carbon monoxide, suggesting the involve-
ment of a cytochromeP450 on E hydroxylation [14].
Several hundred P450, located inthe microsomal
fraction, are currently known in plants, but no mito-
chondrial P450 has been described to date [15]. Plant
P450s are clustered in CYP71–CYP99 and CYP701–
CYP999 families [16]. Almost all plant P450 sequences
have been identified by DNA alignment with other
plant, animal or fungus P450 DNA sequences, and
their functions have been inferred by homology with
sequences of known function. Some of these assign-
ments have been demonstrated using enzymatic assays
or other methods, such as gene silencing [17].
In this context, only a few plant P450s have been
characterized and isolated directly from the plant [18–
22]. This study describes the isolation of a new P450,
which is thefirst to be reported in ferns. Moreover, a
putative function for this P450 is also presented.
Results
Cytochrome P450 expression by E treatment
Treatment of P. vulgare calli cultures with E induced
the production of a cytochromeP450 protein, which
was present inthe microsomal subcellular fraction.
This protein was detected and measured using differ-
ence spectroscopy [23,24], and the amount was depend-
ent on the dose of E applied to the calli and the
incubation time. This showed that levels of P450 pro-
tein measured by CO-binding spectra increased as a
linear function of the E amount applied to the calli up
to 55 pmol P450 per mg protein, with 60 nmol E per
100 mg of tissue (Fig. 1A).
By contrast, maximal amounts of P450 protein
(50 pmolÆmg protein
)1
) were produced 24 h after E
treatment, and began to decrease thereafter, probably
due to E diminution (Fig. 1B).
Under the highest P450 expression conditions
(60 nmol EÆ100 mg callus
)1
for 24 h) theP450 content
in P. vulgare is below that in all plant tissues reported
previously [21]. Microsomal P450 content from E-trea-
ted calli was compared with rat liver microsomal P450,
and other plant P450. It was found that theP450 con-
centration in P. vulgare (0.025 nmolÆg fresh weight
)1
)
is much lower than in rat liver (13.0 nmolÆg fresh
weight
)1
) and avocado mesocarp (Persea americana,
0.29 nmolÆg fresh weight
)1
). The low amounts of P450
in P. vulgare did not allow us to measure any enzy-
matic activity in microsomes, but this activity could be
calculated in vivo by measuring the transformation of
exogenously applied E to 20E.
Relationship between amount of P450 and the
bioconversion of ecdysone to 20-hydroxyecdysone
Ecdysone 20-hydroxylase activity in P. vulgare calli
was estimated using HPLC to quantify the transforma-
tion of topically applied E to 20E. Based on the results
shown in Fig. 1A, the transformation was measured
over time after treatment with three different E doses:
(a) 1.9 pmol EÆ100 mg calli
)1
, which does not promote
P450 induction; (b) 8.6 nmol EÆ100 mg calli
)1
, which
lies within the range of linearity inthe dose–response
curve, and (c) 86 nmol EÆ100 mg calli
)1
, which is
above the dose of E that induces maximum P450 pro-
duction. The results are shown in Fig. 2 and indicate
the amount of 20E formed or bioconversion rate, cal-
culated as described in Experimental procedures.
Fig. 1. Microsomal P450 content of P. vulgare calli after ecdysone
(E) treatment. (A) Different E concentrations were applied to calli
and incubated for 24 h. Microsomes were prepared and their P450
content was measured. (B) We applied 60 nmol E per 100 mg to
the calli and P450 was measured at different times. P450 content
was measured by CO-binding differential spectrum. Values shown
are mean ± SD for triplicate experiments.
Cytochrome P450inferns D. Canals et al.
4818 FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS
In P. vulgare calli, E is oxidized to 20E up to 80%
of transformation (Fig. 2), which is independent of the
amount of E applied (1.9 pmol, 8.6 nmol and 86 nmol
EÆ100 mg calli
)1
).
After E treatment the bioconversion rate of E to
20E (Fig. 2) increases quickly (Fig. 2A, 2 h; Fig. 2B,
14 h; and Fig. 2C, 24 h), overlapping with the period
of P450 induction. Later, the biotransformation rate
decreases rapidly (Fig. 2A, 9 h; Fig. 2B, 24 h; and
Fig. 2C, 48 h), which could be modulated by the
amount of induced P450, product (20E) inhibition and
substrate disappearance.
In order to study the influence of 20E on the meta-
bolism of E in P. vulgare, calli were treated with
several amounts of 20E (0, 0.01, 0.05, 0.1, 0.5 and
1 lmol) and the biotransformation of E to 20E was
monitored by HPLC. As shown in Fig. 3, 20E pro-
duced a dose–response inhibition, suggesting that the
bioconversion of E is partially regulated by 20E.
The bioconversion rate is likely influenced by both
the time required forP450 induction, and the amount
of E (inducing P450 protein) and 20E (inhibiting P450
activity) present inthe tissues at this time. When E is
topically applied to calli, P450 is expressed and the
bioconversion rate increases; subsequently, the 20E lev-
els are augmented, resulting in a decrease inthe bio-
conversion rate. When the induced P450 reaches its
maximum concentration, the bioconversion rate tends
to be constant. Finally, as E decreases below the high-
est induction effect, P450 levels similarly decrease. This
decrease, along with the 20E inhibitory effect, stops
biotransformation all together. Below a 10% transfor-
mation of E to 20E (when no 20E inhibitory effects
are detected), the bioconversion rate is proportional
to the amount of P450 present inthe sample, and
the E 20-hydroxylase activity could be estimated as
0.0025 nmol 20EÆh
)1
Æpmol P450
)1
.
Effect of P450 inhibitors on the bioconversion
of ecdysone to 20-hydroxyecdysone
If E 20-hydroxylase is a P450 enzyme, it is reasonable
to expect that known P450 inhibitors would inhibit its
activity. Some compounds are described as inhibitors
of P450 proteins by binding to their heme group,
particularly effective are imidazole-containing agents:
ketoconazole, miconazole and flutrimazole [20,25–27].
Fig. 2. Bioconversion of ecdysone (E) to 20-hydroxyecdysone (20E)
by P. vulgare calli. Calli were treated with (A) 1.9 pmol, (B)
8.6 nmol and (C) 86 nmol EÆ100 mg calli
)1
for different times, and
after methanolic extraction, ecdysteroids were determined by
HPLC. The amount of 20E is represented by rhombus, and the bio-
conversion rate (20 E per h) is represented by squares. Values are
shown as mean ± SD for triplicate experiments.
Fig. 3. Inhibition of ecdysone (E) bioconversion to 20-hydroxyecdy-
sone (20E) by 20E. 20E (0.01, 0.05, 0.1, 0.5 and 1 lmol) was
applied to calli (100 mg fresh weight). After 6 h treatment
88.6 nmol [
3
H]-E was also applied (0.3 lCi, 3.38 CiÆmol
)1
), and the
ratio of [
3
H]20E to [
3
H]E was analysed after 24 h using HPLC. The
biotransformation ratio was normalized to untreated calli, assigned
a relative activity of 1. Values are shown as mean ± SD for tripli-
cate experiments.
D. Canals et al. CytochromeP450in ferns
FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS 4819
Aminoglutethimide has also been described as a P450
inhibitor [28]. Ketoconazole was selected forthe dose–
response studies because it has been used to block P450
in bacteria [29], fungi [30], plants [31] and mammals
[32]. Also, a crystal structure of ketoconazole com-
plexed with P450eryF is available, showing direct inter-
action between ketoconazole and the heme group [29].
We studied the ketoconazole dose–response effect
in P. vulgare calli treated with radiolabelled E. As
reported for other plant P450s, we found a dose–
response between ketoconazole concentration and E
bioconversion in vivo. An 80% inhibition of hydroxy-
lase reaction occurred when 17.8 nmol of ketoconaz-
ole was applied topically to calli. This dose was
compared with several plant P450 inhibitors on the
same reaction calli. Control calli, and P450 inhibitor-
treated calli, were treated with E and after 24 h the
transformation of E to 20E was evaluated using
HPLC. As shown in Fig. 4, all theP450 inhibitors
investigated decreased the bioconversion of E to 20E,
thus the E 20-hydroxylase activity was decreased, giv-
ing support to the participation of a P450 enzyme in
this enzymatic step.
Ecdysone as a P450 substrate
Cytochrome P450 can show several difference spectra
depending on the type of compound bound. Some
plant P450s show typical binding spectra between P450
and their natural substrates (type I spectrum), or some
inhibitors (type II spectrum) [33]. P. vulgare callus
microsomal fraction from induced calli was obtained
and a differential binding spectrum between micro-
somes and E was registered, showing a typical type I
spectrum (maximum absorbance at 429 nm and mini-
mum absorbance at 390 nm, Fig. 5). Other plant P450
substrates tested (monoterpens geraniol and nerol) did
not show any differential spectrum.
A
B
Fig. 4. Inhibition of ecdysone (E) bioconversion to 20-hydroxyecdy-
sone (20E) by P450 inhibitors. (A) Different amounts of ketoconaz-
ole (0.78, 1.78, 17.8 and 50 nmol) were applied to calli (100 mg
fresh weight). After 6 h treatment 0.3 lCi [
3
H]E was also applied,
and the [
3
H]20E to [
3
H]E ratio was analysed after 24 h by HPLC.
(B) Inhibitors were applied (17.8 nmol in dimethylsulfoxide) on
100 mg callus tissue. After 6 h treatment, 0.3 lCi [
3
H]E was also
applied and the [
3
H]20E to [
3
H]E ratio was analysed after 24 h
using HPLC. (C) Untreated calli. The biotransformation ratio was
normalized to untreated calli, assigned a relative activity of 1. A,
aminoglutethimide; M, miconazole; K, ketoconazole; F, flutrimazole.
Values are given as mean ± SD for triplicate experiments.
Fig. 5. Differential binding spectra of ecdysone (E) with P450-
induced microsomes from P. vulgare calli. The same concentration
of microsomal protein was placed in two cuvettes (sample and ref-
erence) and the differential spectra were registered after several
amounts of E (2, 6, 12, 16, 25 and 40 l
M) had been added to the
sample cuvette.
Cytochrome P450inferns D. Canals et al.
4820 FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS
Protein purification
Purification of the E-induced cytochromeP450 was
accomplished by solubilization of the total P. vulgare
calli P450 microsomal content in zwitterionic deter-
gent, followed by two subsequent fractionations of
anion exchange and hydrophobic column chromato-
graphy (Table 1). Several ionic columns were tested,
but theP450 fraction did not bind to DEAE columns
unless the green pigments were previously removed
from the sample. Using a similar P450 protocol that
has been described in animals, fungi and angiosperms
[19,21,33–37], we obtained a single protein band cor-
responding to a P450 present only in E-treated calli. A
major degree of purification was obtained on DEAE
columns, using a sodium acetate gradient as the eluent.
A second type of selective column was a hydroxyapa-
tite column which allowed us to isolate a single
electrophoretic band (Fig. 6A, track H) with a CO-dif-
ferential spectrum of a cytochrom P420 (Fig. 6B).
Discussion
PEs are natural products that can be found in several
tissues of some plant species [2]. We reported the pres-
ence of PEs in P. vulgare, mainly in sporophytes, rhi-
zomes and protallus tissues [10].
Callus cultures grown from protalli retain PE produc-
tion capacity, because some callus lines without detect-
able amounts of PE were able to transform E to 20E.
This result suggests that these calli lines lack the biosyn-
thetic machinery required forthefirst steps of PE bio-
synthesis. However, some of the enzymes involved in E
oxidation, such as E 2-, 20- and 25-hydroxylases, can
also be expressed [9].
All attempts to detect ecdysone hydroxylase activity
in both P. vulgare microsomes and the liposome-
reconstituted system were unsuccessful because of their
low specific P450 content. However, the transforma-
tion of E to 20E in vivo could be considered to be a
measure of E 20-hydroxylation activity, and is a good
method to assess P450 expression inthe calli. The lin-
ear correlation found between the amount of E applied
to the calli and the resulting P450 content suggests that
E is a P450 gene expression inducer, and could be the
endogenous enzyme substrate.
We have shown that a set of unspecific P450 inhibi-
tors used in fungi, plants and mammals to block P450
pathways inhibit E bioconversion in P. vulgare calli. A
dose–response relation between ketoconazole concen-
tration and a decrease inthe E to 20E biotransforma-
tion ratio was observed, and a high degree of
inhibition forthe other P450 inhibitors tested at a sin-
gle dose was also determined. In addition, 20E causes
inhibition of this enzyme activity, which it could be a
feedback regulation forthe 20E balance in plant tis-
sues, as depicted in Fig. 7.
In vivo E bioconversion measured in callus culture
was a linear function of theP450 microsomal protein
content, and was inhibited by P450 inhibitors. This is
in agreement with a P450 protein being responsible for
this enzymatic step. This relationship is reinforced by
the observed binding spectrum between induced P450
microsomes and E, which was a typical P450 substrate
binding spectrum, suggesting that E is the natural sub-
strate of theP450 protein.
In vitro, E 20-hydroxylase activity has been described
in spinach microsomes [14] and microsomes from sev-
eral Ajuga tissues [38]. Nevertheless, purification of the
protein has not been reported, mainly because of its high
instability. All attempts to demonstrate P450 enzymatic
activity in P. vulgare microsomes have been unsuccess-
ful because of the low specific P450 content. However,
we successfully purified theP450 protein to homogeneity
and observed that the protein shows a characteristic
P450 CO-binding spectrum.
As shown for all known plant P450s, thefirst reported
fern P450 was located in P. vulgare callus microsomes,
but only after E-treatment. ForP450 purification, green
pigments were removed, in order to avoid interferences
with theP450 spectroscopic signal, and furthermore, to
improve P450 binding to the DEAE column [39]. These
Table 1. Purification of microsomal P450 from ecdysone-treated calli of P. vulgareThe amount of P450 was estimated from the CO-binding
spectrum. The amount of protein was determined using the Bradford procedure or was estimated from silver-stained SDS ⁄ PAGE, using dif-
ferent amounts of bovine serum albumin for calibration.
Total protein
(mg)
P450
(nmol)
Yield
(%)
Specific content
(nmol P450Æmg protein
)1
)
Purification
(fold)
Microsomes 50 0.5 100 0.01 1
Triton X-114 15 0.34 68 0.02 2
DEAE – Sephadex 0.13 0.14 28 1.07 107
Hi-Trap DEAE Sepharose Fast Flow 0.03 0.06 12 2 200
Hydroxyapatite 0.006 0.024 4.8 4 400
D. Canals et al. CytochromeP450in ferns
FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS 4821
effects are more drastic in P. vulgare than in other plants
because of the lower amount of P450inthe fern micro-
somes. Pigment extraction was carried out with Tri-
ton X-114 inthe presence of boric acid, as reported
previously [40]. The high percentage of detergent needed
in this step resulted inthe partial denaturation of P450,
which showed a maximum CO-binding spectrum shifted
to 420 nm. In a few purification steps, P450 was
obtained as a single electrophoretic band of 55 kDa,
from E-treated P. vulgare calli.
Few P450 proteins have been purified from plants,
of these, a small number have shown enzymatic activ-
ity [18,19,21,34,37,41]. All these purified P450s belong
to the angiosperm group, and although many angio-
sperm and gymnosperm P450 DNA and cDNA
sequences are known, their identities have been conclu-
ded by DNA sequence homology to known animal,
fungi or plant P450 sequences.
The amount of P450 protein isolated in this study
from E-treated calli was less than that needed to
reconstruct enzyme activity or protein sequencing pro-
cedures. The use of large-scale E-induced calli cultures
is time-consuming and costly, however, cloning the
P450 protein from P. vulgare callus RNA induced by
E may be an alternative method for isolating enough
P450 to monitor enzyme activity.
In short, this study describes thefirst fern P450 with
demonstrated functional activity. We have also presented
a purification protocol. We expect that a battery of P450
enzymes will be found in P. vulgare tissues, and other
phytoecdysteroids producer plants, responsible for the
hydroxylations inthe extensive phytoecdysterois family,
as found inthebiosynthesis of brassinosteroids [11].
Experimental procedures
Callus cultures were grown as previously described [10].
E was obtained from Northern Biochemical Co. (Syktyv-
kar, Russia). Radiolabelled [23,24-
3
H(N)]E was obtained
from NEN
TM
Life Science Products, Inc. (Boston, MA).
Fig. 7. Proposed scheme of ecdysone 20-hidroxylase control. Ecdy-
sone (E) is a P450 protein inducer, and that enzyme would trans-
form E to 20-hydroxyecdysone (20E). E shows a substrate-type
binding spectra with the induced P450, and 20E is a strong inhibitor
of that reaction. Some P450 inhibitors tested show also a strong
inhibitory effect.
A
B
0.016
0.015
0.014
0.013
0.012
0.011
400 450
500
Wavelength (nm)
423nm
Abs
Fig. 6. Isolated P450in P. vulgare. (A) SDS ⁄ PAGE of cytochrome
P450 containing fractions purified from P. vulgare. MW, molecular
mass standards; M, microsomes,10 lg; TX, Triton X-114 poor
phase, 10 lg; D1, proteins eluted from thefirst DEAE-column,1 lg;
D2, proteins eluted from the second DEAE-column (0.5 lg); H, pro-
tein eluted from the hydroxyapatite column (0.5 lg). (B) Carbon
monoxide difference spectrum of cytochromeP450 purified from
P. vulgare, eluted from the hydroxyapatite column. The spectrum
was recorded with sodium dithionite reduced protein in each
cuvette after flushing the sample cuvette with CO (20 bubbles).
Cytochrome P450inferns D. Canals et al.
4822 FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS
Spectroscopic methods
Quantitative determination of cytochromeP450 was made
according to the method described by Omura and Sato
[23], with a Cary300 BIO UV-VIS-instrument using an
extinction coefficient of 91 mm
)1
Æcm
)1
(A
450)490
) for P450
and 111 mm
)1
Æcm
)1
(A
420)490
) forthe cytochrome
P420 [23,24]. When both P450 forms were present, the total
P450 content was estimated by addition the amounts calcu-
lated using their respective extinction coefficients.
In vitro binding of E to cytochromeP450 was monit-
ored by differential spectroscopy [33] of P. vulgare micro-
somes containing P450 (0.01 nmol P450Æmg protein
)1
). E
was added to the sample cuvette (from 2 to 40 lm Ein
1 lL dimethylsulfoxide) containing oxidized microsomes
(1 mg proteinÆmL
)1
). An identical volume of solvent was
added to the reference cuvette. Both cuvettes were mixed
well and the absorption spectra were recorded from 350 to
500 nm.
Bioconversion assays
In vivo E 20-hydroxylase activity was estimated by measur-
ing the bioconversion of E to 20E in P. vulgare callus cul-
ture. Radioactive E (1.9 pmol [23,24-
3
H]E 52.2 CiÆmmol
)1
,
8.5 nmol [23,24-
3
H]E 11.5 mCiÆmmol
)1
or 86 nmol
[23,24-
3
H]E 1.15 mCiÆmmol
)1
) was applied in 2 lLof
H
2
O ⁄ MeOH (v ⁄ v) 1 : 1 solution. At different times, calli
were lyophilized and extracted with MeOH (4 · 5 mL) in a
sonication bath. After pellet sedimentation at 1500 g, meth-
anolic extracts were evaporated to dryness and solved in
1mLH
2
O. The water ecdysteroid fraction was loaded onto
aC
18
reversed-phase Sep-Pak cartridge, previously activa-
ted with 10 mL H
2
O. After washing with H
2
O (5 mL) and
15 : 85 MeOH ⁄ H
2
Ov⁄ v (10 mL), ecdysteroids were eluted
with 85 : 15 MeOH ⁄ H
2
Ov⁄ v (5 mL).
E metabolism was followed by HPLC with a radioactivity
detector (Berthold LB 507 B, Bad Wildbad, Germany), using
a Tracer Spherisorb ODS column (2.5 lm, 25 cm · 4.6 mm;
Barcelona, Spain), in isocratic conditions of H
2
O ⁄ MeOH ⁄
isopropanol (v ⁄ v ⁄ v; 85 : 11 : 4) at 1 mLÆmin
)1
and 22 °C,
because no other phytoecdysteroids were present inthe sam-
ple. Retention times for 20E and E were 10 and 15 min,
respectively.
Bioconversion rate at any time was calculated using the
equation: rate
t2
¼ (20E
t2
) 20E
t1
) ⁄ (t
2
) t
1
), where 20E
t2
and 20E
t1
correspond to the amounts of 20E form at two
consecutives times, t
2
and t
1
, respectively.
Inhibition assays
The compounds tested (miconazole, ketoconazole, flutri-
mazole and aminoglutethimide) have been previously repor-
ted as plant P450 inhibitors [37]. Several amounts of
ketoconazole (0.78, 1.78, 17.8 and 50 nmol) or a single dose
of P450 inhibitors (17.8 nmol) were applied to callus culture
(100 mg fresh weight) in a total volume of 2 lL of dimeth-
ylsulfoxide. After 6 h, 0.3 lCi [23,24-
3
H]E (specific activity
88.6 CiÆmmol
)1
)in2lL of EtOH was added and 24 h later
the 20E ⁄ E ratio was analysed by HPLC as described above.
When 20E was tested as an inhibitor of E bioconversion
0.01, 0.05, 0.1, 0.5 and 1 lmol in a total volume of 2 lL
ethanol were used.
Enzyme purification
All operations were carried out on ice or at 4 °C.
One hundred P. vulgare calli (4 weeks old, 100 mg per
callus) were treated with 60 nmol E per callus in 5 lL
EtOH. After 24 h calli were homogenized in 0.1 m Mops ⁄
NaOH pH 7.0, 5 mm EDTA, 0.3 m sucrose, 0.05% (w ⁄ v)
cysteine and 1 mm phenylmethylsulfonyl fluoride, using an
Osterizer blender (Rye, NY), in a total volume of 40 mL of
buffer.
The homogenate was centrifuged at 10 000 g for 30 min
and the supernatant was centrifuged 100 000 g for 1 h. The
pellet was suspended inthe same buffer containing 50%
(v ⁄ v) glycerol to a final protein concentration of
1–2 mgÆmL
)1
and stored at )80 °C until use. Microsomes
from 100 g of P. americana mesocarp and from 17 g of rat
liver were obtained using the same protocol.
Microsomes were centrifuged at 100 000 g and the pellet
was resuspended in 30 mm boric acid ⁄ KOH pH 8.6, 15%
(v ⁄ v) glycerol, 1 mm, and Triton X-114 was added very
slowly to 1% (w ⁄ v). The suspension was stirred for 1 h at
4 °C and centrifuged at 1500 g at 20 °C for 20 min. Three
phases were formed: the upper detergent-rich phase con-
tained 99% of green pigments, 60% total microsomal protein
and no cytochrome P450. The lower phase contained 30%
total protein and 56% microsomal P450. A pellet was recov-
ered containing 2–3% total P450.
The lower phase was carefully separated and dialysed
against 10 mm Tris ⁄ acetate pH 7.4, 20% glycerol, 1% Tri-
ton X-100 and 5 mm EDTA. It was then loaded onto a
DEAE–Sephadex column (5 mL) equilibrated with the
same Tris–acetate buffer, washed with buffer at pH 8.0
(buffer A), and eluted stepwise with buffer A containing
increasing amounts of 8 m sodium acetate (buffer B, 5%
increased each step. Akta prime system, Amersham-Phar-
macia, Bucks, UK). P450 was detected in steps between
10 and 25% of buffer B. P450 fractions were concentrated
and desalted in Amicon Ultra-15 Centrifugal Filter Units
(Billerica, MA), and then loaded onto a Pharmacia
Hi-Trap-DEAE column (1 mL, 0.5 mLÆmin
)1
) equilibrated
in buffer A with 10% buffer B. TheP450 pool was eluted
with a linear gradient (10–25%) of 8 m sodium acetate,
1 mL fractions. Fractions were screened forP450 content
after Amicon concentration to 50 lL.
Fractions containing P450 were dialysed overnight
against 10 mm sodium phosphate, pH 7.4, 20% glycerol,
D. Canals et al. CytochromeP450in ferns
FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS 4823
1% Triton X-100 and 5 mm EDTA (buffer C) and were
loaded onto a hydroxyapatite column (1 mL, 0.5 mLÆ
min
)1
). TheP450 protein was eluted with a gradient of
KCl (0–5 m for 1 h, buffer D). All fractions (1 mL), were
concentrated to 50 lL and screened forthe presence of
P450.
The amount of protein was determined using the Brad-
ford procedure with bovine serum albumin as a standard
[42]. The protein concentration of partially or total purified
P450 protein was estimated from silver-stained SDS ⁄
PAGE, using different amounts of bovine serum albumin
for the calibration.
SDS ⁄ PAGE was performed as described by Laemmli [43]
on 12.5% polyacrylamide gels, stained with silver nitrate
using the procedure of Morrissey [44] adapted to a Phast-
system equipment (Pharmacia Biotech).
Acknowledgements
Financial support from the Ministerio de Educacio
´
ny
Ciencia (Spain) projects AGL2001-2285 and AGL
2004-05252 is acknowledged. DC thanks Generalitat
de Catalunya for a predoctoral fellowship. The authors
also thank G. Fabrias for critically reading the manu-
script and making valuable suggestions.
References
1 Buszczak M & Segraves W (2000) Insect metamorpho-
sis: out with the old, in with the new. Curr Biol 10,
830–833.
2 Dinan L (2001) Phytoecdysteroids: biological aspects.
Phytochemistry 57, 325–339.
3 Adler JH & Adler G (1999) Occurrence, biosynthesis,
and putative role of ecdysteroids in plants. Crit Rev
Biochem Mol Biol 34, 253–264.
4 Schmelz EA, Grebenok RJ, Ohnmeiss TE & Bowers
WS (2002) Interactions between Spinacea olearcea and
Bradysia impatiens: a role for phytoecdysteroids. Arch
Insect Biochem Phys 51, 204–221.
5 Camps F, Claveria E, Coll J, Marco MP, Messeguer J &
Mele
´
E (1990) Ecdysteroid production in tissue cultures
of Polypodium vulgare. Phytochemistry 29, 3819–3821.
6 Coll J, Reixach N, Sa
´
nchez-Baeza F, Casas J & Camps
F (1994) New ecdysteroids from Polypodium vulgare.
Tetrahedron 50, 7247–7252.
7 Reixach N, Irurre-Santilari J, Casas J, Mele
´
E, Messe-
guer J & Camps F (1996) Biosynthesis of ecdysteroids
in in vitro prothalli cultures of Polypodium vulgare.
Phytochemistry 43, 597–602.
8 Reixach N, Irurre-Santilari J, Camps F, Mele
´
E,
Messeguer J & Casas J (1997) Phytoecdysteroid
overproduction inPolypodiumvulgare prothalli. Phyto-
chemistry 46, 1183–1187.
9 Reixach N, Lafont R, Camps F & Casas J (1999) Bio-
transformations of putative phytoecdysteroid biosyn-
thetic precursors in tissue cultures of Polypodium
vulgare. Eur J Biochem 266, 608–615.
10 Messeguer J, Mele
´
E, Reixach N, Irurre-Santilari J &
Casas J (1998) Polypodiumvulgare L. (wood fern):
in vitro cultures and the production of phytoecdyster-
oids. In Biotechnology in Agriculture and Forestry (Bajaj
YPS ed.), pp. 333–348. Springer-Verlag, Berlin.
11 Bishop GJ & Koncz C (2002) Brassinosteroids and plant
steroid hormone signaling. Plant Cell 14, S97–S110.
12 Fujioka S & Yokota T (2003) Biosynthesis and metabo-
lism of brassinosteroids. Annu Rev Plant Physiol Plant
Mol Biol 54, 137–164.
13 Petryk A, Warren JT, Marques G, Jarcho MP, Gilbert
LI, Kahler J, Parvy JP, Li Y, Dauphin-Villemant C &
O’Connor MB (2003) Shade is the Drosophila P450
enzyme that mediates the hydroxylation of ecdysone to
the steroid insect molting hormone 20-hydroxyecdysone.
Proc Natl Acad Sci USA 25, 13773–13778.
14 Grebenok RJ, Galbraith DW, Benveniste I & Feyereisen
R (1996) Ecdysone 20-monooxygenase, a cytochrome
p450 from spinach, Spinacia oleracea. Phytochemistry
42, 927–933.
15 Schuler MA & Werck-Reichhart D (2003) Func-
tional genomics of P450s. Annu Rev Plant Biol 54,
629–667.
16 Nelson DR (1999) CytochromeP450 and the individual-
ity of species. Arch Biochem Biophys 369, 1–10.
17 Burger C, Rondet S, Benveniste P & Schaller H (2003)
Virus-induced silencing of sterol biosynthetic genes:
identification of a Nicotiana tabacum L. obtusifoliol-14’-
demethylase (CYP51) by genetic manipulation of the
sterol biosynthetic pathway in Nicotiana benthamiana L.
J Exp Bot 54, 1675–1683.
18 Kochs G & Grisebach H (1989) Phytoalexin synthesis in
soybean: purification and reconstitution of cytochrome
P450 3,9-dihydropterocarpan 6a-hydroxylase and
separation from cytochromeP450 cinnamate 4-hydroxy-
lase. Arch Biochem Biophys 273, 543–553.
19 O’Keefe D & Leto KJ (1989) Cytochrome P-450 from
the mesocarp of avocado (Persea americana). Plant
Physiol 89, 1141–1149.
20 Werck-Reichhart D, Gabriac T & Durst F (1990) Two
cytochrome P450 isoforms catalysing O-deethylation of
ethoxycoumarin and ethoxyresorufin in higher plants.
J Biochem 270, 729–735.
21 Gabriac B, Werck-Reichhart D, Teutsch H & Durst F
(1991) Purification and immunocharacterization of a
plant cytochrome P450: the cinnamic acid 4-hydroxy-
lase. Arch Biochem Biophys 288, 302–309.
22 Clemens S & Barz W (1996) Cytochrome P450-depen-
dent methylenedioxy bridge formation in Cicer arieti-
num. Phytochemistry 41, 457–460.
Cytochrome P450inferns D. Canals et al.
4824 FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS
23 Omura T & Sato R (1964) The carbon monoxide-bind-
ing pigment of liver microsomes. I. Evidencefor its
hemoprotein nature. J Biol Chem 239, 2370–2378.
24 Omura T & Sato R (1964) The carbon monoxide-bind-
ing pigment of liver microsomes. II. Solubilization, puri-
fication and properties. J Biol Chem 239, 2379–2385.
25 Hamerski D & Matern U (1988) Elicitor-induced bio-
synthesis of psoralens in Ammi majus L. suspension cul-
tures. Microsomal conversion of demethylsuberosin into
(+) marmesin and psoralen. Eur J Biochem 171, 369–375.
26 Tanahashi T & Zenk MH (1990) New hydroxylated
benzo[c]phenanthridine alkaloids from Eschscholtzia
californica cell suspension cultures. J Nat Prod 53,
579–586.
27 Tang YX & Suga T (1994) Biotransformation of alpha-
and beta-ionones by immobilized cells of Nicotiana
tabacum. Phytochemistry 37, 737–740.
28 Kammerer L, de-Eknamkul W & Zenk MH (1994)
Enzymic 12-hydroxylation and 12-O-methylation of
dihydrochelirubine in dihydromacarpine formation by
Thalictrum bulgaricum. Phytochemistry 36, 1409–1416.
29 Cupp-Vickery JR, Garcia C, Hofacre A & McGee-
Estrada K (2001) Ketoconazole-induced conformational
changes inthe active site of cytochrome P450eryF.
J Mol Biol 311, 101–110.
30 Kelly SL, Lamb DC, Cannieux M, Greetham D, Jack-
son CJ, Marczylo T, Ugochukwu C & Kelly DE (2001)
An old activity inthe cytochome P450 superfamily
(CYP51) and a new story of drugs and resistance.
Biochem Soc Trans 29, 122–128.
31 Asami T, Nakano T, Nakashita H, Sekimata K &
Shimada Y (2003) The influence of chemical genetics on
plant science: shedding light on functions and mechan-
ism of action of brassinosteroids using biosynthesis inhi-
bitors. J Plant Growth Regul 22, 336–349.
32 Hasselberg L, Grøsvik BE, Goksøyr A & Celander MC
(2005) Interactions between xenoestrogens and ketoco-
nazole on hepatic CYP1A and CYP3A, in juvenile
Atlantic cod (Gadus morhua). Comp Hepatol 4, 2–17.
33 Hallahan DL, Nugent JHA, Hallahan BJ, Dawson GW,
Smiley DW, West JM & Wallsgrove RM (1992) Interac-
tions of avocado (Persea americana) cytocrhome P450
with monoterpenoids. Plant Physiol 98, 1290–1297.
34 Higashi K, Ikeuchi K, Obara M, Karasaki Y, Hirano
H, Gotoh S & Koga Y (1985) Purification of a single
major form of microsomal cytochromeP450 from tulip
bulbs (Tulipa gesneriana L.). Agric Biol Chem 49, 2399–
2405.
35 Song WC & Brash AR (1991) Purification of an allene
oxide synthase and identification of the enzyme as a
cytochrome P450. Science 25, 781–784.
36 Kochs G, Werck-Reichhart D & Grisebach H (1992)
Further characterization of cytochromeP450 involved
in phytoalexin synthesis in soybean: cytochrome P450
cinnamate 4-hydroxylase and 3,9-dihydroxypterocarpan
6a-hydroxylase. Arch Biochem Biophys 293, 187–194.
37 Sibbensen O, Koch B & Halkier M (1994) Isolation of
the heme-thiolate enzyme cytochrome P-450TYR, which
catalizes the committed step inthebiosynthesis of the
cyanogenic glucoside dhurring in Sorgum bicolor (L)
Moench. Proc Natl Acad Sci USA 91, 9740–9744.
38 Alekseeva L (2004) Ecdysone 20-monooxygenase activ-
ity of cytochromeP450in plants and cell culture of
Ajuga reptans L. Prikl Biokhim Mikrobiol (Russia) 40,
159–164.
39 Werck-Reichhart D, Benveniste I, Teutsch H, Durst F
& Gabriac B (1991) Glycerol allows low-temperature
phase separation of membrane proteins solubilized in
Triton X-114: application to the purification of plant
cytochromes P-450 and b5. Anal Biochem 197, 125–131.
40 Andersen MD & Moller BL (1998) Double Triton
X-114 phase purification forthe plant cytochromes
P450 and removal of green pigments. Protein Expr Purif
13, 366–372.
41 Kahn RA, Bak S, Olsen CE, Svendsen I & Moller BL
(1996) Isolation and reconstitution of the heme-thiolate
protein obtusifoliol 14’-demethylase from Sorghum
bicolor (L.) Moench. J Biol Chem 271, 32944–32950.
42 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein util-
izing the principle of protein–dye binding. Anal Biochem
72, 248–254.
43 Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227, 680–685.
44 Morrisey JH (1981) Silver staining for protein in poly-
acrylamide gels: a modified procedure with enhanced
uniform sensitivity. Anal Biochem 117, 307–310.
Supplementary material
The following material is available online:
Fig. S1. Chemical structure of phytoecdysteroids isola-
ted from P. vulgare prothalus culture.
D. Canals et al. CytochromeP450in ferns
FEBS Journal 272 (2005) 4817–4825 ª 2005 FEBS 4825
. The first cytochrome P450 in ferns
Evidence for its involvement in phytoecdysteroid biosynthesis
in Polypodium vulgare
Daniel Canals,. 20E.
The bioconversion rate is likely in uenced by both
the time required for P450 induction, and the amount
of E (inducing P450 protein) and 20E (inhibiting