0013-7227/99/$03.00/0
Endocrinology
Copyright © 1999 by The Endocrine Society
Vol. 140, No. 11
Printed in U.S.A.
Changes in Testosterone Metabolism Associated with the
Evolution of Placental and Gonadal Isozymes of Porcine
Aromatase Cytochrome P450*
C. J. CORBIN, J. M. TRANT, K. W. WALTERS,
AND
A. J. CONLEY
Department of Population Health and Reproduction, School of Veterinary Medicine, University of
California (C.J.C., K.W.W., A.J.C.), Davis, California 95616; and the Center of Marine Biotechnology,
University of Maryland Biotechnology Institute (J.M.T.), Baltimore, Maryland 21202
ABSTRACT
Differences in the catalytic activity of the placental and gonadal
isozymes of porcine aromatase cytochrome P450 (P450arom) were
examined in cell lines exhibiting stable expression of recombinant
enzyme. Cell lines were selected that expressed high, but similar,
immunodetectable levels of each isozyme based on Western analysis.
Aromatase activity varied with growth in culture, decreasing at confluence from a peak reached between 50 – 80% cell density. Cells
expressing the placental isozyme had 3–5 times higher catalytic activity (per mg protein) than those expressing the gonadal isozyme.
The P450arom inhibitor fadrazole (1 mM) inhibited more than 97% of
this activity, whereas another imidazole, etomidate (1 mM), selectively
inhibited gonadal P450arom activity by 92%. Estrogen synthesis from
androstenedione and testosterone was determined by RIA and confirmed by HPLC analysis, which also identified the accumulation of
the 19-hydroxy and 19-oxo intermediates of the respective substrates.
A
ROMATASE cytochrome P450 (P450arom) is the catalytic component of the aromatase enzyme complex
responsible for the synthesis of estrogens from androgens. As
such, this enzyme complex plays a unique role in maintaining a physiological balance between androgens and estrogen,
which is critical for reproductive development and function
in vertebrates. Furthermore, estrogen synthesis, through aromatase expression, contributes significantly to the normal
growth and well-being of both males and females (1, 2).
Consistent with a fundamental contribution to reproduction
and fertility, P450arom is highly conserved, demonstrating
50 –90% peptide sequence identity among mammalian,
avian, and fish enzymes (3–14). Despite the recent explosion
of sequence information on the vertebrate aromatases, most
of what is understood concerning the catalytic properties of
this important enzyme has been derived from studies in
human tissues, principally the placenta (15), which expresses
unusually high levels of activity compared with placentas of
other primates or domestic animal species (16). Much less is
known concerning the functional biochemistry of P450arom
expressed in other tissues or other species.
Aromatase cytochrome P450 is encoded by a single gene
Received June 8, 1999.
Address all correspondence and requests for reprints to: Dr. A. J.
Conley, VM-PHR, School of Veterinary Medicine, University of California, Davis, California 95616. E-mail: [email protected].
* This work was supported in part by USDA 98 –35203-6439 and
HD-36913– 01 (to A.J.C.) and USDA 94 –37203-0761 (to J.M.T.).
There was no evidence of other steroid metabolites accumulating in
the media of cell lines expressing either isozyme. Tritiated water
formed during aromatization of substrates 3H labeled at the C1 and
C2 positions was stereo-selective for the b orientation, but less so for
testosterone than androstenedione during metabolism by the porcine
placental (and human) isozyme than the gonadal isozyme. Testosterone showed a higher affinity for the porcine placental P450arom
than the gonadal P450arom, but both isozymes had similar affinities
for androstenedione. Testosterone was also aromatized more slowly
than androstenedione by the porcine gonadal P450arom. These data
suggest that catalytic differences have arisen in the substrate binding
pocket during the evolution of isozymes of porcine P450arom that
affect androgen metabolism, particularly the aromatization of testosterone. The physiological significance of these differences to the
reproductive biology of the pig remains to be determined. (Endocrinology 140: 5202–5210, 1999)
in humans, even though it is expressed in a broad array of
tissues, including many that are steroidogenic in the classical
sense as well as others generally considered nonsteroidogenic (15). For instance, in addition to the placenta and the
gonads of both men (17) and women (18), P450arom is expressed in brain, liver (19), adipose tissue (20), and a number
of tissues in the fetus (21). Additionally, tumors from numerous sites, such as breast (22, 23), gonads (24), prostate
(19), and even myeloid leukemia cells (25), have been shown
to express P450arom. An elaborate mechanism involving
alternative splicing of untranslated first exons and associated
promoter elements differentially regulates tissue-specific expression in these sites (15). The ovary is notable in this regard
because it appears to be the only tissue expressing P450arom
in which splicing is not invoked. Instead, the first untranslated exon is contiguous with exon II (26). Hinshelwood et al.
(27) showed that this characteristic of human ovarian
P450arom expression is apparently shared in equine and
porcine species, even at the nucleotide level. Moreover, this
contrasts the lack of sequence conservation in the 59-untranslated region among placental P450arom transcripts from
these species (27) and even P450arom transcripts expressed
in the porcine testis (28). Thus, this feature of the basic structure of the CYP19 gene regulating P450arom expression appears conserved only in the female gonad. These observations are consistent with the suggestion that P450arom may
have first evolved in the vertebrate ovary (15) and that expression in the placenta is a more recent event among eu-
5202
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
therian mammals. How the functional properties of
P450arom might have changed to accommodate the physiological needs of the developing fetus in utero for both androgen metabolism and/or estrogen synthesis is an interesting, but unanswered, question.
As noted above, and in contrast to the wealth of information on human P450arom, many fewer studies have been
conducted on the aromatase enzyme system of other mammals. Recent investigations in this laboratory and others have
determined that there are essential differences in the biology
of P450arom in the pig. Unlike any other mammal investigated to date, the pig expresses functionally distinct
isozymes in a tissue-specific manner (12). Evidence continues
to accumulate that these are encoded by completely duplicated genes clustered on chromosome 1 (29 –31). Our studies
have shown that, unlike other mammals, a gonadal isozyme
is expressed in the theca interna as well as the stratum granulosum of the ovarian follicle (32), the Leydig cell of the testes,
and, interestingly, the zona reticularis of the adrenal gland
of the pig (28). The porcine placenta expresses a second
isozyme (12, 29), and a third is expressed in the early, preattachment porcine blastocyst (13). However, the roles
served by multiple isozymes of P450arom in the pig remain
unclear and are difficult to elucidate given such a broad
range of cellular environments possibly influencing catalytic
function, in addition to other factors, including levels of
expression. In fact, few studies conducted to date have compared the function of isozymes of P450arom; none has done
so in the same cellular background or examined isozymes
that have evolved in the same species. Kinetic experiments
conducted on recombinant enzyme generated by transient
transfection often suffer from low, unpredictable levels of
expression. Therefore, the catalytic characteristics of the placental and gonadal isozymes of porcine P450arom were compared under conditions of stable recombinant expression in
the hope of gaining insight into the evolution of function of
this important enzyme system. The existence of gonadal and
placental isozymes of porcine P450arom provides a unique
opportunity to explore the functional evolution of a highly
conserved enzyme critical for reproductive success in
mammals.
Materials and Methods
Materials
Tissue culture medium, geneticin, and Lipofectamine were obtained
from Life Technologies, Inc. (Gaithersburg, MD), and plasmids were
obtained from Invitrogen (Carlsbad, CA). Tritiated tracers were purchased from New England Nuclear (Wilmington, DE) with the following
specific activities: [1b-3H]androstenedione, 23.1 Ci/mmol; [1b,2b3
H]androstenedione, 42.0 Ci/mmol; [1a,2a-3H]androstenedione, 52.1
Ci/mmol; [7-3H]androstenedione, 24.5 Ci/mmol; [1b,2b-3H]testosterone, 40.3 Ci/mmol; [1a,2a-3H]testosterone, 60.0 Ci/mmol; and
[7-3H]testosterone, 30.0 Ci/mmol. Authentic steroids were purchased
from Steraloids, Inc. (Wilton, NH) and Sigma Chemical Co. (St. Louis,
MO). RIA reagents were purchased from Diagnostics Systems Laboratories, Inc. (Webster, TX). HPLC solvents were of ultrapure quality from
Burdick and Jackson (Muskegon, MI). Cell protein concentrations were
estimated using bicinchoninic acid protein assay reagent kits (Pierce
Chemical Co., Rockford, IL). Immunoblots were conducted on polyvinyl
membranes (Immobilon-P, Millipore Corp., Bedford, MA), and proteins
were detected by chemiluminescence (ECL, Amersham Pharmacia Biotech, Arlington Heights, IL). The imidazole enzyme inhibitors fadrazole
5203
(CGS16949A) and etomidate were gifts from Ciba-Geigy (Summit, NJ)
and Abbott Laboratories (Abbott Park, IL), respectively. All other chemicals were of the highest quality available.
Transient transfection
Experiments were conducted in which the parent 293 cell line was
used for recombinant P450arom expression by transient transfection.
Full-length complementary DNAs (cDNAs) were constructed in the
pCMV expression plasmid. Conditions including levels of plasmid DNA
(1.5 mg), Lipofectamine (5 ml), and time (6 h) were carefully optimized
for 5 3 105 cells cultured in six-well plates. Kinetic analyses of aromatase
activity using androstenedione and testosterone were conducted 48 h
posttransfection.
Establishment of cell lines exhibiting stable expression of
porcine aromatases
Cultures of the human 293 fetal kidney cell line were maintained
in DMEM high glucose with 10% FCS, 10 mm HEPES, and increasing
concentrations (100–800 mg/ml) of geneticin (G418). All cells were killed
at 600 mg/ml during 6 days of culture. Both placental and gonadal porcine
P450arom cDNAs were subcloned into the pcDNA3 vector (Invitrogen,
Carlsbad, CA) containing the NeoR ORF gene. Transfection of 293 cells was
carried out on 1 3 106 cells using 30 ml Lipofectamine and 5 mg linearized
plasmid DNA for 5 h. Cells were placed into medium without G418 for 48 h,
after which time they were split 1:15 and transferred to selective medium
containing 600 mg/ml G418 thereafter. On day 14, individual colonies were
lifted (24 for each transfection) using sterilized circles of Whatman filter
paper (Clifton, NJ) soaked in trypsin solution and transferred to 24-well
plates. Cells were grown and subcultured under the same, continuous
antibiotic selection pressure until enough cells were obtained to cryopreserve aliquots of 2.5 3 106 cells/ml, and separate lines expressing the
gonadal and placental isozymes were stored. Each line was recovered from
liquid nitrogen and cultured in G418 (600 mg/ml) to verify viability.
P450arom activities were determined at different stages of confluence and
passage number by the tritiated water assay using [1b-3H]androstenedione,
as described below. Protein expression was further investigated by Western
immunoblot analysis, also described below. All subsequent studies were
conducted on lines designated G2 (gonadal P450arom isozyme) and P11
(placental P450arom isozyme).
Aromatase activity by tritiated water assay
Screening cells in monolayer culture for P450arom activity required
measuring the incorporation of tritium into 3H2O from [1b-3H]androstenedione. Incubations were carried out at 37 C in 5% CO2 with 150 nm
substrate (20% labeled, 80% radioinert) in duplicate wells. Incubation
times varied from 30 min for the placental line to 2 h for the gonadal line.
Inhibition of P450arom activity was achieved with the addition of fadrazole or etomidate (each at 1 mm) to the culture medium along with the
substrate. Reactions were stopped by the addition of 0.5 vol cold 30%
trichloroacetic acid and were extracted with 2 vol chloroform. A portion
of the aqueous phase was mixed with an equal volume of 5% charcoal0.5% dextran suspension, centrifuged for 30 min at 2000 3 g, and quantified by liquid scintillation counting. Background values for 3H2O released in the absence of cells were subtracted from each point. Substrate
saturation and kinetic studies were carried out as described above from
duplicate or triplicate wells using either [1b-3H]androstenedione or
[1b,2b-3H]testosterone as substrate.
Stereo-specific hydrogen loss
These measurements employed the use of [1b-3H]-, [1b,2b-3H]-, and
[1a,2a-3H]androstenedione and [1b,2b-3H]- and [1a,2a-3H]testosterone
as substrates. Culture medium (2 ml) containing substrate was added to
1 3 106 cells/35-mm dish. Aliquots were removed at numerous time
points from triplicate wells (100 ml for P450arom activity measurement
by tritiated water assay as described above, and 50 ml for estrone or
estradiol determination by RIA). The amount of 3H2O and the amount
of estrogen formed with each labeled substrate were used to calculate
5204
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
the distribution label as the percentage retained in the estrogen product
or released as 3H2O. This percentage, estimated at each time point in
replicated experiments, was used in the calculation of a mean for each
labeled substrate for each isozyme.
Estrone and 17b-estradiol RIA
Estrogens were measured in culture medium without extraction using a double antibody system with 125I-labeled estrone or estradiol
tracers, enabling assays to be performed on samples containing 3Hlabeled estrogens. Standard curves (0.2–500 pg/tube) for 17b-estradiol
and estrone were prepared in assay buffer and culture medium equivalent to the diluted sample medium, and antibody addition was adjusted
to approximately 40% binding. Assays were sensitive to 1.0 and 0.2 pg
for estrone and estradiol, respectively, and the intra- and interassay
coefficients of variation were less than 10% and 15%, respectively, in
each case.
Western immunoblot analysis
Cells were homogenized in PBS containing 1% sodium cholate and
0.1% SDS and sonicated for 3 sec. Equal amounts of protein (50 mg) were
subjected to SDS-PAGE (8% gel) in buffer containing 50 mm Tris, 383 mm
glycine, 0.1% SDS, and 0.4 mm EDTA. Separated proteins were transferred by electroblot onto membranes and immunoblotted with antisera
(1:1000) raised against recombinant human P450arom protein (courtesy
of Dr. N. Harada, Fujita Health University, Aichi, Japan).
Steroid product analysis by HPLC
Medium from cell lines incubated with 7-3H-labeled tracers were
removed after 30 min to 2 h of culture. A mixture of authentic steroids
(2.5 nmol each of testosterone, 19-hydroxytestosterone, 19-nortestosterone, estradiol, androstenedione, 19-hydroxyandrostenedione, 19-norandrostenedione, 19-oxoandrostenedione, and estrone) was added and
equilibrated with each sample to facilitate extraction and to serve as
standards during analysis. The steroids were extracted from the medium
with a 10-fold volume of methylene chloride, and the organic phase was
transferred and evaporated under reduced pressure. The steroidal residue was subjected to HPLC analysis (computer-controlled model
HP1100 with photo diode array detector, Hewlett-Packard Co., Palo
Alto, CA), and the radioactivity of the eluant was monitored by an in-line
radioisotope detector (Ramona 2000, RayTest, New Castle, DE). The
quarternary mobile phase was made up of water, methanol, acetonitrile,
and tetrahydrofuran (solvents A–D, respectively) flowing at 1 ml/min.
The column (4.6 3 25 mm, 5m-Ultrasphere ODS, Beckman Coulter, Inc.,
Fullerton, CA) was maintained at 30 C and initially equilibrated to 80%
A, 10% B, 6% C, and 4% D. These conditions were maintained for the
first 3 min of the gradient, after which the solvent composition changed
linearly over 15 min to 63% A, 20% B, 11% C, and 6% D. Over the next
14 min, the solvent composition linearly changed to 45.5% A, 25.6% B,
15.9% C, and 13% D. The solvent system quickly (1 min) changed to 10%
B, 83% C, and 4% D, and these conditions were held for an additional
2 min. The column was reequilibrated to the initial solvent conditions
for 15 min before the next injection of sample.
Endo • 1999
Vol 140 • No 11
strate affinities for both androstenedione and testosterone
were estimated after transient transfection of expression
plasmids in the parent 293 cells. Porcine placental P450arom
exhibited a 3-fold lower Km for testosterone than for androstenedione (24 vs. 70 nm, respectively). Estimates of Km in
transfections with the gonadal P450arom were problematic
due to the low, variable levels of expression, slower rates of
substrate metabolism, and correspondingly longer incubation times coupled with the higher background typical of the
[1b,2b-3H]testosterone label (33). This being so, the Km estimates for androstenedione and testosterone were calculated at 50 and 68 nm, respectively.
Cell lines exhibiting stable expression of the porcine placental and gonadal isozymes of P450arom were developed to
overcome the problems encountered with transient transfection, especially with the gonadal P450arom. Multiple independent cell lines expressing the gonadal (n 5 6) and
placental (n 5 11) isozymes of porcine P450arom were isolated from 293 colonies under constant antibiotic selection
pressure. Consistently, cells lines expressing the gonadal
P450arom isozyme plated more efficiently, formed simple
monolayers, and reached confluence earlier than lines expressing the placental isozyme. Cell lines expressing the
placental P450arom isozyme not only grew more slowly, but
also tended to grow in distinct clusters around the dish rather
than as a more uniform monolayer. These growth characteristics were not seen to change in the presence of the
P450arom inhibitor fadrazole (1 mm). Lines were also
screened for expression of P450arom by immunoblot analysis. Similar levels of immunodetectable P450arom were
demonstrated in cell lines P11 and G2, expressing the placental and gonadal isozymes, respectively. The apparent molecular size was approximately 48 kDa, and no expression
was observed in the nontransfected parent 293 cells (Fig. 1).
Aromatase enzyme activity was further investigated in
these cell lines by the release of tritiated water from [1b3
H]androstenedione, which was examined at different stages
of confluence and after passage (Fig. 2). Consistently higher
activity was exhibited by the cell line expressing the placental
isozyme than by that expressing the gonadal isozyme of
porcine P450arom. Activity varied from 25– 41 pmol/mgz2 h
in the gonadal cell line and from 86 –239 pmol/mgz2 h in the
placental cell line as confluence progressed from 40 –100%
Statistical analysis
Product formation was analyzed by linear regression. Kinetic constants were evaluated by ANOVA using the general linear models and
Kruskal-Wallis nonparametric test procedures (SAS Institute, Inc., Cary,
NC).
Results
Preliminary studies performed on the nonsteroidogenic
human 293 kidney tumor cell line demonstrated that neither
androstenedione nor testosterone was metabolized to any
detectable degree over 6 h or more of culture. This suggested
that the expression of other enzymes, such as 17b-hydroxysteroid dehydrogenase, that might otherwise compete with
P450arom for substrate was negligible in this cell line. Sub-
FIG. 1. Western immunoblot analysis for P450arom expression in
293 cell lines. Single bands of approximately 48 kDa in size were
detected in lines exhibiting stable recombinant expression of the
porcine placental (P11) and gonadal (G2) isozymes of P450arom. No
band was detected in parent 293 cells (lane 3). Each lane contained
50 mg total protein from cell homogenates.
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
before and after passage (Fig. 2). The lowest catalytic rates
were observed when cells had reached 100% confluence, and
all subsequent experiments were conducted on cells at a
density of around 80% to maximize the yield of the recombinant P450arom. Greater than 97% of the aromatase activity
was inhibited by fadrazole (1 mm) addition in both cell lines.
Gonadal P450arom activity was also inhibited by 92% in the
presence of the other imidazole inhibitor, etomidate (1 mm),
which marginally inhibited placental P450arom activity by
less than 10%.
The time course of conversion of both androstenedione
and testosterone to estrone and estradiol was further evaluated by the tritiated water release assay and RIA of medium
for each estrogen. Cells stably expressing porcine P450arom
isozymes demonstrated progressive accumulation of estrogens with time, as shown in Fig. 3. The rate of estrogen
formation was generally greater from androstenedione than
FIG. 2. Aromatase activity was measured by the tritiated water assay using [1b-3H]androstenedione as substrate, as described in Materials and Methods. Cells from the P11 and G2 lines were cultured
and passaged. Activity (picomoles per mg/2 h) was measured at different cell densities (percent confluence).
5205
from testosterone and considerably faster for the placental
P450arom than the gonadal isozyme, consistent with the
results of activity analyses. Product formation was confirmed in parallel cultures by HPLC analysis, which demonstrated the presence of estrone, 19-oxoandrostenedione,
and 19-hydroxyandrostenedione (Fig. 4). Similarly, estradiol,
19-oxotestosterone, and 19-hydroxytestosterone were observed in cultures incubated with testosterone (data not
shown). No evidence of other steroid products was noted
either early (,10%) or late (.50%) in the metabolism of either
of the substrates. These results also further confirmed the
lack of interconversion of either androgen substrates or estrogen products by endogenous oxidoreductases during the
course of these experiments.
The stereo-specific loss of hydrogen from the C1 and C2
positions during the aromatization of androstenedione and
testosterone was examined using appropriately labeled substrates. Tritiated water formation was monitored during the
metabolism of 1b-3H-, 1b,2b-3H-, and 1a,2a-3H-labeled steroids and expressed as a percentage of estrone and estradiol
synthesis determined in medium by RIA after various times
of incubation. An example is illustrated in Fig. 5, which
presents the results of experiments examining [1b-3H]- and
[1b,2b-3H]androstenedione metabolism by the placental
P450arom. The results of metabolism of all tracers by both
isozymes are summarized in Table 1. These data demonstrated a preferential loss of label from [1b-3H]androstenedione to form 3H2O (Table 1) which represented about 90% of
the estrone formed. However, the rate of 3H2O formation
from tracers labeled in both the 1b- and 2b-positions was
lower for the placental isozyme than for the gonadal
P450arom, and this difference was particularly marked for
[1b,2b-3H]testosterone (56% vs. 84%, respectively). Similar
studies conducted with transiently expressed, human
P450arom demonstrated rates of 3H2O release very similar to
those obtained with the porcine placental isozyme (Table 1).
Only low rates of 3H2O formation were observed from the
1a,2a-3H-labeled tracers.
FIG. 3. Estrone (left panels) and estradiol (right panels) formation from androstenedione and testosterone metabolism by the P11 (lower panels)
and G2 (upper panels) cell lines expressing the placental and gonadal isozymes of porcine P450arom, respectively. Estrogen formation was
measured by RIA of aliquots of medium removed at various times. Assays were performed in duplicate as described in Materials and Methods.
The calculated linear regression coefficients are shown.
5206
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
Endo • 1999
Vol 140 • No 11
FIG. 5. Estrogen accumulation measured by RIA and tritiated water
release from [1b-3H]androstenedione or [1b,2b-3H]androstenedione
metabolism by cells expressing the placental isozyme of porcine
P450arom. Cells were incubated with this and other substrates (150
nM), and aliquots were removed at various times, as shown. Tritium
released as 3H2O and estrone formation were measured at each of
these times in replicated experiments. Estimates of the percentage of
label retained in the estrogen product or otherwise released as 3H2O
were made at each time point, all of which were subsequently used in
the calculation of means (see Table 1) for each of the substrates and
each P450arom isozyme.
FIG. 4. HPLC analysis of radiolabeled products of [7-3H]androstendione metabolized by P11 cells expressing the placental isozyme of
porcine P450arom. Radioactive steroids (A) coeluted with authentic
19-hydroxyandrostenedione (19-OHA4), 19-oxoandrostenedione (19oxoA4), androstenedione (A4), and estrone (E1). The patterns of elution of authentic radioinert steroids were simultaneously detected at
240 nm wavelength (B) and 280 nm wavelength (C). The times of
elution of authentic 19-hydroxytestosterone (19-OHT4), 19-norandrostenedione (19-norA4), testosterone (T4), and estradiol (E2) are also
shown in B and C.
Kinetic studies were initiated by examining substrate vs.
velocity relationships of the placental and gonadal isozymes
with both substrates, and subsequent experiments used substrate concentrations below and above those that yielded
approximately half the maximum velocity (Vmax). Incubation
times were chosen to limit substrate conversion to between
10 –30%. The results of these experiments were analyzed by
double reciprocal plots (Fig. 6) from which apparent Km
values were calculated (Table 2). The Km for both substrates
tended to be lower for the placental isozyme than for the
gonadal isozyme and significantly lower (P , 0.05) for testosterone than for androstenedione metabolism by the porcine placental P450arom isozyme. There was little difference
in Vmax for androstenedione and testosterone aromatization
by the placental isozyme (244 and 198 pmol/mgzh), but a
3-fold higher Vmax was observed for the metabolism of androstenedione vs. testosterone for the gonadal isozyme (209
and 69 pmol/mgzh, respectively).
Discussion
These data document the establishment of cell lines stably
expressing placental and gonadal isozymes of porcine
P450arom that enable studies of substrate kinetics and stereospecific loss of hydrogen during aromatization. P450arom
expression was verified by immunodetection of the expressed protein in cell lysates as well as by measurement of
aromatase activity and estrogen formation, as confirmed by
HPLC analysis. The levels of aromatase activities exhibited
by each cell line were comparable to those measured in
porcine placental and gonadal homogenates (12, 28), although direct comparisons were not made by Western analysis. Evidence was presented for catalytic differences between these isozymes, which included sensitivity to
inhibition by the imidazole etomidate, the stereo-specific loss
of hydrogen from the C1 and C2 positions of testosterone, and
apparent affinity constants for testosterone. Specifically, the
gonadal isozyme, expressed in ovarian thecal and granulosa
cells (12, 32, 34), testes, and adrenal gland (28) of the pig, was
inhibited by etomidate; the placental isozyme was not. A
lower proportion of 3H was released as 3H2O during the
metabolism of testosterone by the placental and human enzymes compared with the porcine gonadal P450arom
isozyme. Finally, the placental isozyme of porcine P450arom
exhibited a higher affinity for testosterone than the gonadal
form of the enzyme. Collectively, these results demonstrate
fundamental catalytic differences in androgen metabolism
by isozymes of P450arom expressed in the placenta and
gonads of the pig.
The possible existence of isozymes of P450arom in any
species was first recognized in the pig by Corbin et al. (12).
Subsequent studies in this laboratory and others have confirmed the existence of multiple genes encoding the porcine
P450arom isozymes (29 –31). Although still a unique finding
among mammals, at least two genes are also known to encode different isozymes of P450arom in the goldfish (14, 35),
a polyploid species. The results of the present study, con-
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
5207
TABLE 1. Distribution of tritium (expressed as a percentage) in labeled C19 substrates (determined by New England Nuclear) and in the
products, estrogen and water, during metabolism by porcine (gonadal and placental) and human isozymes of aromatase cytochrome P450
(see Materials and Methods for details)
Isotope distribution
Substrate
Product
%a
%b
% Estrogen
% Water
Pig gonadal
Pig placental
Human
1b-A4
1b-A4
1b-A4
26
26
26
74
74
74
14
7
10
86
93
90
Pig gonadal
Pig placental
Human
1b,2b-A4
1b,2b-A4
1b,2b-A4
23
23
23
77
77
77
19
27
18
81
73
82
Pig gonadal
Pig placental
Human
1b,2b-T4
1b,2b-T4
1b,2b-T4
27
27
27
73
73
73
16
44
39
84
56
61
Pig gonadal
Pig placental
Human
1a,2a-A4
1a,2a-A4
1a,2a-A4
87
87
87
13
13
13
82
88
83
18
12
17
Pig gonadal
Pig placental
1a,2a-T4
1a,2a-T4
87
87
13
13
71
85
29
15
Androstenedione (A4) and testosterone (T4) were labeled in either the a or b configurations at the C1 or C2 position of the steroid A ring
as shown. Percentages shown for estrogen and water represent the means (n 5 5–7 determinations) of estimates that were calculated from the
ratios derived at each time point for each P450arom/substrate combination (see Fig. 5).
sistent with the preliminary observations (12), indicate that
there are distinct functional differences between the placental and gonadal isozymes of porcine P450arom. Cell lines
exhibiting stable expression of each isozyme, at similar levels
based on immunoblot analysis, demonstrated slower aromatization of testosterone, particularly so for the gonadal
isozyme, which, unlike the placental P450arom, was also
susceptible to inhibition by etomidate (12). The differential
sensitivity to inhibition by etomidate, generally considered
an inhibitor of 11b-hydroxylase cytochrome P450, is consistent with functionally important, structural variations in the
substrate binding pockets of the placental and gonadal
isozymes. Product analysis verified the formation of intermediates (19-hydroxy- and 19-oxoandrogens), apparently released from the catalytic site during oxidation before estrogen formation. Earlier studies of aromatase with porcine
ovarian microsomes (36, 37) as well as others studying human, rat, and equine aromatases (38 – 40) demonstrated a
significant release of intermediates. No evidence was found
for the formation of 19-norandrostenedione or 19-nortestosterone, even though both steroids are resolved by the HPLC
system that was developed and used in the current studies.
This contrasts the results of Khalil et al. (41, 42), but is consistent with the report by Garrett et al. (43); both investigators
used porcine granulosa cells as the source of P450arom.
Garrett et al. (43) suggested that nonenzymatic conversion
during sample processing might explain the appearance of
19-norandrogens in medium from cultured porcine granulosa cells. If enzymatic synthesis does occur, the conditions
for formation of 19-nor products were not reproduced in that
or the present study.
The establishment of cell lines expressing the placental and
gonadal isozymes of porcine P450arom also facilitated an
examination of stereo-specific loss of hydrogen from the b
orientation of positions C1 and C2 on the steroid A ring of
both androgen substrates. This forms the basis of the widely
used radiometric procedure known as the tritiated water
assay for aromatase activity (33). The exact nature of the
stereo-selective hydrogen loss, chemical or biological, remains unclear and largely unexplored in other species, except the rat (40). The data presented here suggest that the
tendency for selective loss of the b-oriented hydrogen from
the C1 and C2 positions of androgen substrates differed for
the placental and gonadal isozymes, particularly in the case
of testosterone. Previous studies by Swinney et al. (40) suggested that the rate of release of b-hydrogen differed between
rat ovarian and human placental P450arom enzymes. Specifically, less 3H2O was recovered during estrogen synthesis
from either [1b-3H]- or [1b,2b-3H]androstenedione by rat
ovarian P450arom than from human placental P450arom.
Isotope effects were considered unlikely, but the researchers
were unable to determine whether the differences reflected
tissue (ovarian vs. placenta) or isozyme (rat vs. human) effects. Our data on the percentage of label appearing as 3H2O
in cells transiently expressing human P450arom correspond
well with those reported by Swinney et al. (40) for the four
labels examined (90% vs. 90% for [1b-3H]androstenedione;
82% vs. 85% for [1b,2b-3H]androstenedione; 61% vs. 49% for
[1b,2b-3H]testosterone; and 17% vs. 20% for [1a,2a-3H]androstenedione). However, based on P450arom isozymes expressed in the same cell background, the results of the
present study suggest that the P450 itself, whether species or
tissue derived, contributes to the observed differences in the
rates of b-hydrogen loss. Thus, quantitative comparisons of
aromatase activity among species at least, if determined by
tritiated water release, should consider differential loss of
label particularly if dual 1b,2b-3H-labeled tracers are used.
Our data suggest that the porcine placental isozyme is
catalytically very similar to human P450arom, but that the
gonadal isozyme is more distinct. These differences must
involve important conformational changes in the substrate
binding pocket of the gonadal isozyme that affect substrate
5208
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
Endo • 1999
Vol 140 • No 11
FIG. 6. Double reciprocal plots (1/[S] vs. 1/V) of [1b-3H]androstenedione (upper panels) and [1b,2b-3H]testosterone (lower panels) metabolism by cells
expressing the placental (left panels) and gonadal (right panels) isozymes of porcine P450arom. Cells were cultured, and activities were measured
as described in Materials and Methods. Shown are the results of experiments performed in triplicate. Insets show substrate saturation curves.
orientation causing a tighter fit of the steroid in the active site.
Graham-Lorence et al. (44) hypothesized that hydrogen abstraction and enolization occurred via a proton shuttle mechanism
involving aspartate 309, a highly conserved, acidic residue that
in structurally determined P450s projects into the pocket. This
residue was predicted to lie within 2A of the substrate C1 and
C2. The stereo-specific loss of tritium from [1b-3H]androstenedione seen for both porcine isozymes and human and rat
P450arom (40) suggest that this substrate is probably oriented
with the b face more exposed to the proton acceptor. The less
selective loss of tritium during aromatization of [1b,2b-3H]testosterone suggests that this substrate must be slightly rotated
relative to the proton acceptor, exposing the a and b labels more
equally. The stereo-selectivity shown with the a-labeled substrates is consistent with the preferential removal of hydrogen,
which has a higher zero point energy than tritium. Whether the
characteristics of the stereo-selective loss of hydrogen during
aromatization correlates with other aspects of catalytic activity
of different isozymes of P450arom remains an interesting question. However, it seems from these data and those reported by
Swinney et al. (40) that hydrogen abstraction and enolization of
androstenedione in both isozymes and of testosterone in the
gonadal isozyme are protein-facilitated reactions. Moreover,
these reactions might not necessarily proceed in the same way
for androstenedione and testosterone because of differences in
orientation in the substrate binding pocket that depend on the
species- or tissue-specific form of the P450arom in question.
Catalytic differences between the porcine placental and
gonadal P450arom isozymes were also examined in kinetic
experiments to estimate affinities for the C19 steroid substrates androstenedione and testosterone. Similar Km estimates were obtained by transient transfection and in stable
expressing cell lines for androstenedione and testosterone in
the case of the placental isozyme. Stable expression was
necessary to obtain reliable estimates of Km for testosterone
metabolism by the gonadal P450arom. Collectively, these
data indicate that the two isozymes have similar affinities for
androstenedione, but that the gonadal isozyme has a significantly lower affinity for testosterone than the placental
P450arom. The apparent Km for androstenedione (77 and 104
nm for placental and gonadal isozymes, respectively) is in the
range reported for human P450arom by numerous researchers (3, 40, 45– 47). Apparent Km values reported for testosterone are more variable, sometimes similar to those for
androstenedione (3, 47), slightly higher (46), or markedly
higher (45), which may reflect different sources of enzymes
or methods of purification. Our results suggest that there are
functional differences in affinity for testosterone between the
porcine placental and gonadal isozymes (33 vs. 116 nm, respectively) stably expressed in 293 cells, which, in the case of
the placental isozyme, is also less than half the Km for androstenedione (77 nm). McPhaul et al. (4) reported slightly
higher Km values derived from testosterone metabolism by
chicken P450arom (150 nm) expressed in ovarian microsomes
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
TABLE 2. Apparent Km (nanomolar concentrations) for
aromatization of androstenedione and testosterone by cells
expressing the placental or gonadal isozymes of porcine P450arom
Placental
Gonadal
Androstenedione
Testosterone
77 6 19
104 6 19
33 6 10
116 6 33
Shown are the mean 6 SEM of five independent experiments, each
performed in duplicate or triplicate.
and for the recombinant enzyme expressed in vitro. Although
the microsomal or in vitro environment may influence P450
activity (46), the differences noted between the porcine
isozymes examined here are unlikely to suffer from such
potential confounding factors.
The results of the present study also indicated that testosterone was aromatized more slowly than androstenedione, as
suggested by the slower accumulation of estradiol and by estimates of the apparent Vmax from kinetic experiments. However, a valid comparison of catalytic rates depends directly on
equal levels of P450arom expression in each cell line, which our
data suggest may change with the stage of confluence. Although immunoblot analysis suggested that levels were generally similar in each cell line, this also assumes that the antiserum used recognizes the isozymes with equal affinity. In fact,
Hinshelwood et al. (9) considered that a single amino acid
change rendered bovine P450arom essentially undetectable by
immunoblot analysis, consistent with the possibility that small
changes in structure might have marked effects on epitope
recognition. For this reason, we recently overexpressed each
porcine isozyme in insect cells using a baculovirus vector. This
has enabled estimates to be made of P450arom content in preparations of both isozymes by difference spectroscopy and analysis of androgen metabolism on an equal molar basis. Complete
kinetic experiments using these preparations are in progress.
However, preliminary results confirm that the placental
isozyme of porcine P450arom is 3 times as active in aromatizing
androgen as the gonadal isozyme and is more efficient in doing
so at lower testosterone concentrations (Corbin, C. J., and A. J.
Conley, unpublished observations). These data compare well
with the 3- to 5-fold differences in catalytic rates observed in the
stable cell lines, suggesting that P450arom expressions in cell
lines are similar, but that differences in activity arise from the
catalytic characteristics inherent in the P450 isozymes they express. It also suggests that there are no major differences in
immunorecognition by the antisera used for immunoblot analysis. Collectively, the data suggest that the greatest differences
in catalytic function between the porcine gonadal and placental
P450arom isozymes relate to their metabolism of testosterone.
Moreover, despite the potential for differences arising from
expression levels, variations with confluence or peculiarities
arising during the development of the cell lines specifically
investigated, the evidence points strongly to P450arom
isozymes themselves as the source of the functional divergence
reported here.
Based on the data presented above, we conclude that the
evolution of porcine P450arom expression in the placenta
was associated with an increase in the efficiency of androgen
metabolism, potentially involving both the affinity and rate
of aromatization. These changes are consistent with the
lower androgen concentrations likely to be presented to the
5209
enzyme expressed in utero compared with P450arom in the
ovarian follicle (32, 48, 49). The apparent Km values typical
of mammalian aromatases are considerably lower than those
commonly reported for other steroid hydroxylases such as
P450c17 (50), suggesting that P450arom may be adapted for
expression in cells or tissues not also synthesizing androgen
substrates (51). Equally, the greater affinity for testosterone
may have important adaptive significance in terms of the
physiological protection of female fetuses developing adjacent to males during sexual differentiation. It is perhaps
noteworthy that there was no significant effect of the proximity of male fetuses on the development of females in pigs
(52), whereas an effect is obvious in rodents (53), a species
that lacks P450arom in the placenta (54). In other words, the
lack of placental P450arom may leave the female fetuses of
some mammalian species susceptible to androgenization in
utero, a concept consistent with the pseudohermaphroditism
observed in women suffering congenital P450arom deficiency (55, 56). The similarity between the porcine placental
and human P450arom isozymes suggests that similar functional shifts to more efficient testosterone metabolism may
have occurred in other species exhibiting placenta P450arom
expression. In fact, the reported bovine CYP19 pseudogene
(57) might represent the disappearance of an ancestral CYP19
gene encoding P450arom after evolution of expression in the
bovine placenta (9). Further studies comparing P450arom
function across species will add to our understanding of the
roles of androgens and estrogens in vertebrate development
and reproduction. To what degree this involves essential
changes in regions of the P450 mediating substrate binding,
membrane anchoring, redox partner association, or other
physical features remains to be determined.
Acknowledgments
The authors thank Dr. David Swinney for kindly sharing tracers, Drs.
Larry Hjelmeland and K. C. Dunn for 293 cells, Carla Berard for technical
assistance, and Drs. Marios Georgiadis and Bill Sisco for help with
statistical analysis. The authors also acknowledge the contributions of
Dr. Sandy Graham, whose unusual insight into structure-function relationships of P450arom formed the foundation of these aspects of the
discussion appearing in this paper.
References
1. Federman DD 1994 Life without estrogen. N Engl J Med 331:1088 –1089
2. Bulun SE 1996 Aromatase deficiency in women and men: would you have
predicted the phenotypes? J Clin Endocrinol Metab 81:867– 871
3. Corbin CJ, Graham-Lorence S, McPhaul M, Mason JI, Mendelson CR, Simpson ER 1988 Isolation of a full-length cDNA insert encoding human aromatase
system cytochrome P-450 and its expression in non-steroidogenic cells. Proc
Natl Acad Sci USA 85:8948 – 8952
4. McPhaul MJ, Noble JF, Simpson ER, Mendelson CR, Wilson JD 1988 The
expression of a functional cDNA encoding the chicken cytochrome P-450arom
(aromatase) that catalyses the formation of estrogen from androgen. J Biol
Chem 263:16358 –16363
5. Hickey GJ, Krasnow JS, Beattie WG, Richards JS 1990 Aromatase cytochrome
P450 in rat ovarian granulosa cells before and after luteinization: adenosine 39,59monophosphate-dependent and independent regulation. Cloning and sequencing of rat aromatase cDNA and 59 genomic DNA. Mol Endocrinol 4:3–12
6. Lephart ED, Peterson KG, Noble JF, Wilson JD, George FW, McPhaul MJ
1990 The structure of cDNA clones encoding the aromatase P450 isolated from
a rat Leydig cell tumor line demonstrates differential processing of aromatase
mRNA in rat ovary and a neoplastic cell line. Mol Cell Endocrinol 70:31– 40
7. Terashima M, Toda K, Kawamoto T, Kuribayashi I, Ogawa Y, Maeda T,
Shizuta Y 1991 Isolation of a full-length cDNA encoding mouse aromatase
P450. Arch Biochem Biophys 285:231–237
8. Tanaka M, Telecky TM, Fukada S, Adachi S, Chen S, Nagahama Y 1992
Cloning and sequence analysis of a cDNA encoding P-450 aromatase
5210
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
FUNCTIONAL EVOLUTION OF PORCINE AROMATASES
(P450arom) from a rainbow trout (Onchorhynchus mykiss) ovary: relationship
between the amount of P450arom mRNA and the production of estradiol-17b
in the ovary. J Mol Endocrinol 8:53– 61
Hinshelwood MM, Corbin CJ, Tsang PCW, Simpson ER 1993 Isolation and
characterization of a complimentary deoxyribonucleic acid insert encoding
bovine aromatase cytochrome P450. Endocrinology 133:1971–1977
Shen P, Campagnoni CW, Kampf K, Schlinger BA, Arnold AP, Campagnoni
AT 1994 Isolation and characterization of a zebra finch aromatase cDNA: in
situ hybridization reveals high aromatase expression in brain. Mol Brain Res
24:227–237
Trant JM 1994 Isolation and characterization of the cDNA encoding the channel catfish (Ictalurus punctatus) form of cytochrome P450arom. Gen Comp
Endocrinol 95:155–168
Corbin CJ, Khalil MW, Conley AJ 1995 Functional ovarian and placental
isoforms of porcine aromatase. Mol Cell Endocrinol 113:29 –37
Choi I, Simmen, RCM, Simmen FA 1996 Molecular cloning of cytochrome P450
aromatase complementary deoxyribonucleic acid from preimplantation porcine
and equine blastocysts identifies multiple novel 59-untranslated exons expressed
in embryos, endometrium and placenta. Endocrinology 137:1457–1467
Tchoudakova A, Callard GV 1998 Identification of multiple CYP19 genes
encoding different cytochrome P450 aromatase isozymes in brain and ovary.
Endocrinology 139:2179 –2189
Simpson, ER, Mahendroo MS, Means GD, Kilgore MW, Hinshelwood MM,
Graham-Lorence S, Amarneh B, Ito Y, Fisher CR, Michael MD, Mendelson
CR, Bulun SE 1994 Aromatase cytochrome P450, the enzyme responsible for
estrogen synthesis. Endocr Rev 15:342–355
Jurke MH, Czekala NM, Jurke S, Hagey LR, Lance VA, Conley AJ, FitchSnyder H 1998 Monitoring pregnancy in twinning Pygmy loris (Nycticebus
pygmaeus) using fecal estrogen metabolites. Am J Primatol 46:173–183
Brodie A, Inkstar S 1993 Aromatase in human testis. J Steroid Biochem Mol
Biol 44:549 –556
Tamura T, Kitawaki J, Yamamoto T, Osawa Y, Kominami S, Takemori S,
Okada H 1992 Immunohistochemical localization of 17a-hydroxylase/C17–20
lyase and aromatase cytochrome P-450 in the human ovary during the menstrual cycle. J Endocrinol 135:589 –595
Harada N, Utsumi T, Takagi Y 1993 Tissue-specific expression of the human
aromatase cytochrome P450 gene by alternative use of multiple exons 1 and
promoters, and switching of tissue-specific exons 1 in carcinogenesis. Proc Natl
Acad Sci 90:11312–11316
Ackerman GE, Smith ME, Mendelson CR, MacDonald PC, Simpson ER 1981
Aromatization of androstenedione by human adipose tissue stromal cells in
monolayer culture. J Clin Endocrinol Metab 53:412– 417
Toda K, Simpson ER, Mendelson CR, Shizuta Y, Kilgore MW 1994 Expression of the gene encoding aromatase cytochrome P450 (CYP19) in fetal tissues.
Mol Endocrinol 8:210 –217
Bulun SE, Simpson ER 1994 Breast cancer and the expression of aromatase in
breast adipose tissue. Trends Endocrinol Metab 5:113–120
Lu Q, Nakmura J, Savinov A, Yue W, Weisz J, Dabbs DJ, Wolz G, Brodie A
1996 Expression of aromatase protein and messenger ribonucleic acid in tumor
epithelial cells and evidence of functional significance of locally produced
estrogen in human breast cancers. Endocrinology 137:3061–3068
Bulun SE, Rosenthal IM, Brodie AM, Inkster SE, Zeller WP, DiGeorge AM,
Frasier SD, Kilgore MW, Simpson ER 1994 Use of tissue-specific promoters
in the regulation of aromatase cytochrome P450 gene expression in human
testicular and ovarian sex cord tumors, as well as in normal fetal and adult
gonads. J Clin Endocrinol Metab 78:1616 –1621
Jacob F, Homann D, Seufert J, Schneider D, Kohrle J 1995 Expression and
regulation of aromatase cytochrome P450 in THP 1 human myeloid leukaemia
cells. Mol Cell Endocrinol 110:27–33
Means GD, Kilgore MW, Mahendroo MS, Mendelson CR, Simpson ER 1991
Tissue-specific promoters regulate aromatase cytochrome P450 gene expression in human ovary and fetal tissues. Mol Endocrinol 5:2005–2013
Hinshelwood MM, Liu Z, Conley AJ, Simpson ER 1995 Demonstration of
tissue-specific promoters in non-primate species which express aromatase
P450 (P450arom) in placentae. Biol Reprod 53:1151–1159
Conley AJ, Corbin CJ, Hinshelwood MM, Liu Z, Simpson ER, Ford JJ,
Harada N 1996 Functional aromatase expression in porcine adrenal gland and
testes. Biol Reprod 54:497–505
Choi I, Troyer DL, Cornwell DL, Kirby-Dobbels KR, Collante WR, Simmen
FA 1997 Closely related genes encode developmental and tissue isoforms of
porcine cytochrome P450 aromatase. DNA Cell Biol 16:769 –777
Conley AJ, Corbin CJ, Smith T, Hinshelwood MM, Liu Z, Simpson ER 1997
Porcine aromatases: studies on tissue-specific, functionally distinct isozymes
from a single gene? J Steroid Biochem Mol Biol 61:407– 413
Lahbib-Mansais Y, Barbosa A, Yerle M, Parma P, Milan D, Pailhoux E, Gellin
J, Cotinot C 1997 Mapping in pig of genes involved in sexual differentiation:
AMH, WT1, FTZF1, SOX2, SOX9, AHC, and placental and embryonic CYP19.
Cytogenet Cell Genet 76:109 –114
Endo • 1999
Vol 140 • No 11
32. Conley AJ, Howard HJ, Slanger WD, Ford JJ 1994 Steroidogenesis in the
preovulatory porcine follicle. Biol Reprod 51:655– 661
33. Lephart ED, Simpson ER 1991 Assay of aromatase activity. Methods Enzymol
206:477– 483
34. Garrett WM, Guthrie HD 1997 Steroidogenic enzyme expression during preovulatory follicular maturation in pigs. Biol Reprod 56:1424 –1431
35. Callard GV, Tchoudakova A 1997 Evolutionary and functional significance of
two CYP19 genes differentially expressed in brain and ovary of goldfish. J
Steroid Biochem Mol Biol 61:387–392
36. Kautsky MP, Thurman GW, Hagerman DD 1975 A convenient biological
synthesis of 19-hydroxy[4-14C]4-androstene-3,17-dione from [4-14C]4-androstene-3,17-dione with sow ovarian 19-hydroxylase. J Chromatogr 114:473– 475
37. Kautsky MP, Hagerman DD 1980 Kinetic properties of steroid 19-hydroxylase
and estrogen synthetase from porcine ovarian microsomes. J Steroid Biochem
13:1283–1290
38. Thompson EA, Siiteri PK 1974 Utilization of oxygen and reduced nicotinamide adenine dinucleotide phosphate by human placental microsomes during aromatization of androstenedione. J Biol Chem 249:5364 –5372
39. Gaillard J-L 1991 Equine testicular aromatase: substrates specificity and kinetic characteristics. Comp Biochem Physiol 100B:107–115
40. Swinney DC, Watson DM, So O-Y 1993 Accumulation of intermediates and
isotopically sensitive enolization distinguish between aromatase (cytochrome
P450 CYP19) from rat ovary and human placenta. Arch Biochem Biophys
305:61– 67
41. Khalil MW, Walton JS 1985 Identification and measurement of 4-oestren3,17-dione (19-norandrostenedione) in porcine ovarian follicular fluid using
high performance liquid chromatography and capillary gas chromatographymass spectrometry. J Endocrinol 107:375–381
42. Khalil MW, Morley P, Glasier MA, Armstrong DT, Lang T 1989 Formation
of 4-oestrene-3,17-dione (19-norandrostenedione) by porcine granulosa cells in
vitro is inhibited by the aromatase inhibitor 4-hydroxyandrostenedione and
the cytochrome P-450 inhibitors aminoglutethamide and ketoconazole. J Endocrinol 120:251–260
43. Garrett WM, Hoover DJ, Shackleton CHL, Anderson LD 1991 Androgen
metabolism by porcine granulosa cells during the process of luteinization in
vitro: Identification of 19-oic-androstenedione as a major metabolite and possible precursor for the formation of C18 neutral steroids. Endocrinology
129:2941–2950
44. Graham-Lorence S, Amarneh B, White R, Peterson JA, Simpson ER 1995 A
three-dimensional model of aromatase cytochrome P450. Protein Sci 4:1065–1080
45. Kellis JT, Vickery LE 1987 Purification and characterization of human placental aromatase cytochrome P450. J Biol Chem 262:4413– 4420
46. Sethumadhavan K, Bellino FL 1991 Human placental etrogen synthetase
(aromatase). Effect of environment on the kinetics of protein-protein interactions and the production of 19-oxygenated androgen intermediates in the
purified reconstituted cytochrome P450 system. J Steroid Biochem Mol Biol
39:381–394
47. Zhou D, Korzekwa KR, Poulos T, Chen S 1992 A site-directed mutagenesis
study of human placental aromatase. J Biol Chem 267:762–768
48. McNatty KP 1981 Hormonal correlates of follicular development in the human
ovary. Aust J Biol Sci 34:249 –268
49. Wise T 1987 Biochemical analysis of bovine follicular fluid: albumin, total
protein, lysosomal enzymes, ions, steroids and ascorbic acid content in relation
to follicular size, rank, atresia classification and day of estrous cycle. J Anim
Sci 64:1153–1169
50. Conley AJ, Bird IM 1997 The role of P450 17a-hydroxylase and 3b-hydroxysteroid dehydrogenase in the integration of gonadal and adrenal steroidogenesis via the D4 and D5 pathways of steroidogenesis. Biol Reprod 56:789 –799
51. Conley AJ Walters KW 1999 Aromatization. In: Knobil E, Neill JD (eds)
Encyclopedia of Reproduction. Academic Press, New York, vol 1:280 –291
52. Rohde Parfet KA, Lamberson WR, Rieke AR, Cantley TC, Ganjam VK, vom
Saal FS, Day BN 1990 Intrauterine position effects in male and female swine
subsequent survivability, growth rate, morphology and semen characteristics.
J Anim Sci 68:179 – 85
53. vom Saal FS 1981 Variation in phenotype due to random intra-uterine position
of male and female fetuses in rodents. J Reprod Fertil 62:633– 650
54. Towsend L, Ryan KJ 1970 In vitro metabolism of pregnenolone-7a-3H, progesterone-4-14C and androstenedione 4-14C by rat placental tissue. Endocrinology 87:151–155
55. Shozu M, Akasofu K, Harada N, Kubota Y 1991 A new cause of female
pseudohermaphrodism: placental aromatase deficiency. J Clin Endocrinol
Metab 72:560 –566
56. Conte FA, Grumbach MM, Ito Y, Fisher CR, Simpson ER 1994 A syndrome
of female pseudohermaphrodism, hypergonadotropic hypogonadism, and
multicystic ovaries associated with missense mutations in the gene encoding
aromatase (P450arom). J Clin Endocrinol Metab 78:1287–1292
57. Furbass R, Vanselow J 1995 An aromatase pseudogene is transcribed in the
bovine placenta. Gene 154:287–291