| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
-Hydroxylase/C-17,20-Lyase Cytochrome P450 in Rat Liver1
Department of Biology, University of Padova (S.V., L.D.V., L.C.), Padova, Italy; and the Department of Biochemistry, Vanderbilt University School of Medicine (M.R.W.), Nashville, Tennessee 37232
Address all correspondence and requests for reprints to: Dr. Silvia Vianello, Dipartimento di Biologia, Università di Padova, via U. Bassi 58/B, 35121 Padova, Italy.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/C-17,20-lyase cytochrome P450 (cytochrome
P450c17) in the liver, stomach, duodenum, and testis of rats from day
18 of pregnancy to adulthood. In the male liver, the enzyme became
detectable at birth (135 pmol/mg protein·min) at a level comparable
to that in the testis (188 pmol/mg protein·min). The activity then
increased dramatically, reaching a peak at 8 days (691 pmol/mg
protein·min), which was more than 4-fold the testicular levels in
rats of the same age or in adults. Thereafter it declined steadily,
becoming undetectable from puberty onward. The hepatic peak followed a
depression in testicular activity (58 pmol/mg protein·min) on day 6.
Northern and immunoblot analyses showed a good temporal correlation
between enzyme activity and the occurrence of P450c17 messenger RNA
(mRNA) and protein. The same patterns of mRNA and protein occurrence
were observed in female rat liver, indicating that the hepatic CYP17
expression is not sexually dimorphic. Sequencing confirmed a complete
identity in the coding region between hepatic and gonadal mRNAs.
Hepatic P450c17 mRNA, however, was 150–200 bases longer than the
gonadal counterparts. No significant expression of mRNAs encoding
P450scc and P450arom was observed in liver of either sex at any age. In
stomach and duodenum, enzyme activity was much lower (maxima at 25 and
14 pmol/mg protein·min, respectively) than that in liver, but
persisted from the time of weaning onward. It is suggested that the
hepatic peak in P450c17 activity may serve to convert circulating
progestogens into androgens for gonadal aromatization during Sertoli
and granulosa cell proliferation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
17-Hydroxylase/C-17,20-lyase cytochrome P450 (cytochrome P450c17),
on the other hand, is practically absent in the brain of adult rats
(3), but is expressed at significant levels in the peripheral nervous
system at least during fetal and neonatal life (8) and in several
peripheral organs (5, 9, 10, 11, 12, 13, 14). This microsomal steroid hydroxylase
catalyzes two enzymatic steps: the conversion of PREG and progesterone
(PROG) to their 17-hydroxylated products (17-hydroxylase
activity) and the subsequent C-17,20 bond scission (C-17,20-lyase
activity) with the production of dehydroepiandrosterone and
androstenedione (AD), respectively (15). The rat is a
corticosterone-secreting species, and P450c17 is expressed in the
gonads of adult animals, but not in the adrenal. Peripheral P450c17
activity was initially demonstrated in rat submandibular (5, 10) and
sublingual glands (10) and small intestine (9). More recently,
17-hydroxylation or androgen formation from PREG has been reported
in rat kidney, skin, esophagus, stomach, duodenum, and colon (11, 12, 13, 14).
In the stomach, P450c17 is selectively localized in parietal cells
(13). The coding region of P450c17 mRNA in both rat stomach and
duodenum is identical to that of the corresponding testicular mRNA
(12).
In rodents, this enzyme shows interesting developmental changes in its peripheral expression. Transient occurrence in the subcortical plate of the fetal mouse brain was suggested in the synthesis of neuroactive steroids guiding cortical axon growth (8). In rat stomach, immunostaining for P450c17 begins in parietal cells at the time of weaning (days 16–21), increases on day 31, and plateaus during adulthood (13). Temporal studies are likely to provide some insight into the functional role of P450c17 and other steroidogenic enzymes in peripheral tissues. Thus, we have examined the developmental expression of P450c17 in rat liver, where previous reports indicated that the enzyme might be present (16, 17), and compared this expression pattern to those in the stomach, duodenum, and testis. Our data demonstrate that the expression of hepatic P450c17 activity undergoes dramatic changes during the postnatal period, reaching very high levels by 7–8 days and disappearing before adulthood in both males and females.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxyprogesterone
(17OH-PROG), [4-14C]17OH-PROG,
[1,2-3H]AD, [4-14C]AD,
[1,2-3H]testosterone (TST), [4-14C]TST, and
[1,2-3H]11-deoxycorticosterone (DOC) were purchased from
New England Nuclear Corp. (Boston, MA). Steroids were purified before
use by two-dimensional TLC (silica gel 60 F254, Merck, Darmstadt,
Germany) with benzene-acetone (4:1, vol/vol). Reference steroids were
obtained from Steraloids (Pawling, NY) and Sigma Chemical Co. (St.
Louis, MO). Restriction endonucleases were purchased from Promega
(Madison, WI) and Boehringer Mannheim (Indianapolis, IN).
Animals
Male and female Wistar rats of different ages, nontimed cycling
females, and timed pregnant females were purchased from Charles River
(Calco, Italy), Morini (Reggio Emilia, Italy), and Harlan
Sprague-Dawley (Indianapolis, IN). Animals were killed by cervical
dislocation or CO2-induced asphyxia. Gastric and duodenal
contents were removed by thoroughly rinsing with ice-cold 0.15
M KCl. Blood was withdrawn in EDTA by cardiac puncture. For
enzymatic assays and microsomal fractioning, tissues were used
immediately, whereas for RNA extraction, they were frozen in liquid
N2 and stored at -80 C until use. Animals were treated in
accordance with European Community regulations (decree 87–848).
Enzymatic assays
Pools of each tissue from different animals of the same age were
minced and homogenized in a balanced salt medium [100 mM
KCl, 16 mM K2HPO4, 4 mM
KH2PO4, 1 mM K EDTA, 1
mM dithiothreitol, and 2 µg/ml each of leupeptin and
pepstatin A (Sigma), pH 7.4]. After centrifugation (four times) at
1000 x g for 15 min at 4 C, the supernatant protein
content was determined by Bradford assay. After preincubation (10 min
at 37 C), 800 µl homogenized tissue (stomach, 0.16–1.5 mg protein;
duodenum, 0.3–1.1 mg; liver, 0.03–0.35 mg; testis, 0.03–0.25 mg)
were added to 200 µl salt medium containing
[1,2-3H]PROG (stomach, duodenum, and testis, 0.9–2
µCi; liver, 2–2.6 µCi) plus unlabeled PROG (final concentrations
in stomach and duodenum, 1–1.5 µM; liver, 4–4.5
µM; testis, 1–2 µM), and an
NADPH-generating system (final concentrations, 0.1 mM
NADPH, 1 mM NADP+, 10 mM
glucose-6-phosphate, 2 IU/ml glucose-6-phosphate dehydrogenase; Sigma).
Reactions at 37 C in a metabolic shaker were stopped by adding 2 ml
water-saturated ethyl acetate in a dry ice-ethanol bath. Different
incubation times (1–5 min) in duplicate or triplicate were used for
each experiment. For one incubation time, two protein concentrations
were used. In some cases, additional longer incubations (10–20 min)
were also performed. Metabolic patterns and validation of
chromatographic procedures were assessed by preliminary incubations
with [4-14C]PROG.
Tissues boiled for 30 min before incubation were used as a blank.
Extraction, purification, and characterization of steroids
Carrier (5 µg each) and 4-14C- or
1,2-3H-labeled tracer steroids (PROG, 17OH-PROG, AD,
TST, and DOC) were added for both correction of recovery and metabolite
identification by isotope ratio. Steroids were extracted four times
with 2 vol water-saturated ethyl acetate and evaporated to dryness. To
check the reliability of the steroid separation method, half of the
steroids extracted from incubates with [4-14C]PROG were
separated by two-dimensional TLC using cyclohexane-ethyl acetate (9:1,
vol/vol, two or three runs) in the first direction for fat removal,
benzene-acetone (4:1, vol/vol, one run) in the second direction, and
cyclohexane-ethyl acetate (1:1, vol/vol, one run) again in the first
direction. Chromatoplates were autoradiographed for 7 days at 4 C, and
radioactive zones corresponding to either steroid standards (visualized
under UV light at 254 nm) or unknown radioactive products were eluted
with 20 ml acetone. They were then dried, resuspended in 20 µl
ethanol, and separated by HPLC using a 4 x 250-mm long, 5-µm
particle, C2/C18-bonded silica reverse phase
column (SuperPac Pep-S, Pharmacia Biotech, Uppsala, Sweden). The mobile
phase was a discontinuous solvent gradient of methanol-water (1:1,
vol/vol) and acetonitrile (14). Retention times were: PROG, 51 min;
17OH-PROG, 36 min; TST, 33 min; and AD, 30 min. Fractions
corresponding to the central portion of each steroid peak were counted
by liquid scintillation using a dual label program. Steroids were
conclusively identified by recrystallization in acetone-water (for AD
and TST) or pyridine-water (for 17OH-PROG) with the corresponding
reference compounds until a constant 3H/14C
ratio (within ±5% of mean) was reached in three consecutive crystal
crops and the final mother liquor. It was confirmed by HPLC that the
retention times of the unknown radioactive products were not
overlapping those of the identified steroids. As a further
methodological check, the remaining half of the extracted
14C-labeled steroids were separated by HPLC alone with the
same solvent gradient as that described above. Steroid yields were then
compared with those obtained after steroid separation by either TLC
alone or TLC plus HPLC.
After this validation, steroids extracted from incubates with
[1,2-3H]PROG were resuspended in 1 ml water and applied
to a C18 Extract-Clean cartridge (85% total recovery;
Alltech, Deerfield, IL) to remove interfering lipids before HPLC. The
central portion of each steroid peak was counted, and radiochemical
identity was occasionally confirmed. Percent conversions, corrected for
procedural losses and blank values, were calculated from the
3H/14C ratio of individual steroids and
transformed into the corresponding steroid concentrations and steroid
formation rates. As yields of TST were always low at short incubation
times, and TST could not be freed in HPLC from minor contamination by
DOC, TST formation rates were not included in the calculation of total
P450c17 activity. Moreover, yields of unidentified metabolites were
also disregarded in this calculation, although it is possible that some
were derivatives of 17OH-PROG or AD. Thus, the reported P450c17
activities may be underestimated.
Immunoblot analysis
Microsomes from rat liver, testis, and spleen were prepared
according to the method of Trzaskos et al. (18), using
phenylmethylsulfonylfluoride (115 µg/ml), leupeptin, pepstatin A, and
apoprotinin (2 µg/ml each; Boehringer Mannheim) as protease
inhibitors. Bovine adrenal microsomes were prepared as previously
described (19). Immunoblot analysis was carried out with antihuman
P450c17 antibody as previously described (20).
Northern blot analysis
Total RNA was extracted from pools of frozen tissues (3) and
polyadenylated RNA [poly(A)+] was isolated using batch
affinity chromatography on oligo(deoxythymidine)-cellulose (Life
Technologies, Gaithersburg, MD) with one round of purification (21).
Total RNA (20 µg) or poly(A)+ RNA (15 µg) was analyzed
by Northern analysis, as previously described (21), using random primed
32P-labeled complementary DNA (cDNA) probes. Rat P450c17
cDNA probe was a mixture of two fragments, one covering the region -29
to +257 (286 bp), and another covering the region from +932 to +1243
(312 bp) (22). A 1200-bp EcoRI fragment from the 3'-end of
rat ovary P450scc cDNA (23) and a 700-bp
EcoRI/Xho fragment of rat Leydig cell tumor
P450arom cDNA (clone 19) (24) were also used as probes.
Reverse transcription-PCR (RT-PCR) and nucleotide sequencing
Total RNA was used for RT-PCR. Forty cycles of amplification
were used, with an annealing temperature of 60 C for P450c17 and
P450scc cDNAs, and 64 C for P450arom cDNA. Primers are shown in Table 2
. Rat spleen, muscle, and blood were used as negative controls for the
P450c17 amplification, and muscle was used for the P450scc and P450arom
amplifications. Blanks were carried out in the absence of template
RNA.
|
). To obtain the
complete sequence of male rat liver P450c17 cDNA, two additional
primers were employed (sense, 5'-AGCCAGATCAGTTCATGCCTGAA-3'; antisense,
5'-AGCACCTCCCTGGCCAGCTGAT-3'). Sequence reactions used T7 DNA
polymerase (T7 Sequencing kit, Pharmacia Biotech; Sequenase version 2.0
DNA sequencing kit, Pharmacia). |
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/C-17,20-lyase
activity in male rat stomach, duodenum, liver, and testis-hydroxylase/C-17,20-lyase activities. A preliminary experiment
was carried out to determine the metabolic pattern of each organ and to
check the reliability of the steroid separation method. For this
purpose, low speed supernatants of homogenized tissues from 16-day-old
(liver only) and 30-day-old (all four organs) male rats were incubated
with [4-14C]PROG for only 1–2 min. Half of the extracted
steroids were separated by two-dimensional TLC and analyzed by
autoradiography. Different PROG metabolic patterns were observed; the
liver showed the most complex one (Fig. 1). In addition
to 17OH-PROG and AD, two radioactive metabolites with the same
reference front as TST and DOC were detected by autoradiography, except
that DOC was not present in the duodenum, nor was TST in the duodenum
and testis. Under the short term incubation conditions used, these
compounds were only 2–10% of the total 17OH-PROG plus AD. At least
eight unknown metabolites of varying polarity were found in the liver
(Fig. 1), whereas duodenum formed one unknown product, and stomach and
testis formed four (data not shown). When all TLC eluates were further
purified by HPLC, none of the unknown metabolites had the same
retention time as 17OH-PROG or AD. TST and DOC were separated from
all other steroids, but were not resolved from each other.
|
OH-PROG and AD were very
similar to those obtained by TLC plus HPLC (±5–15%), indicating the
equivalence of the two chromatographic procedures. The radiochemical
purity of 17OH-PROG and AD, separated by both TLC and HPLC, was
confirmed by recrystallization to a constant
3H/14C ratio (Table 1). PROG
metabolism was completely abolished in boiled tissues. After these
preliminary control assays, the developmental pattern of
17-hydroxylase/C-17,20-lyase activity was examined by incubating
[1,2-3H]PROG with homogenates of the four tissues from
male rats ranging in age from 18 days of gestation to more than 400
days after birth.
|
-hydroxylase/C-17,20-lyase activity in liver and testis is
shown in Fig. 2A, and that in stomach and duodenum is
presented in Fig. 2B. 17-Hydroxylase/C-17,20-lyase activity appears
to be differentially regulated during development in these four organs.
In the liver, it is not detectable before birth, but it is present in
newborn animals (135 pmol/mg protein·min) at levels comparable to
those in the testis (188 pmol/mg protein·min). Activity then
increased dramatically, reaching a peak at 8 days of age (691 pmol/mg
protein·min) that was more than 4-fold the level in the testis at the
same age or in adult rats. Thereafter it steadily declined until late
puberty and adulthood, when it became very low or undetectable. The
peak in liver activity corresponded to a reproducible depression
of 17-hydroxylase/C-17,20-lyase activity in testis (58 pmol/mg
protein·min at 6 days) during the transition from fetal to adult
Leydig cell steroidogenesis (26).
|
OH-PROG and AD was also absent before birth in
both stomach and duodenum. In the stomach, it was detected at a low
level (0.2 pmol/mg protein·min) at birth and remained low in neonatal
and juvenile rats (1.4–3.4 pmol/mg protein·min), until reaching a
maximun (25 pmol/mg protein·min) at 29 days, just after the weaning
period. It then decreased, but remained detectable throughout life. In
the duodenum, the activity after birth was very low or undetectable
(0–4.0 pmol/mg protein·min) until 16 days, increased thereafter
until adulthood (14.1 pmol/mg protein·min), and remained at about
this level throughout life. As TST was hardly detectable in short term incubations with [1,2-3H]PROG, longer incubations (10–20 min) were also carried out under nonsaturating conditions. Production of TST could be demonstrated in liver, stomach, duodenum, and testis by HPLC separation and recrystallization. In the liver and duodenum, the quantity of TST was consistently lower than that of AD, whereas in the stomach, TST increased with incubation time at the expense of AD.
Developmental pattern of expression of P450c17 protein in male and
female rat livers
In the rat testis and ovary, cytochrome P450c17 is a single
polypeptide chain of 507 amino acids, with a calculated molecular
weight of 57.2 kDa (27). To support the inference that the observed
17-hydroxylase/C-17,20-lyase activities are due to cytochrome
P450c17 and not to a different enzyme having gratuitous P450c17-like
activity, the presence of the P450c17 enzyme was assessed in male rat
liver by immunoblot analysis of microsomal protein from prenatal (18
days postcoitum), postnatal (7 days), juvenile (21 days), and adult
(85–90 days) animals. Microsomes from adult rat testis and bovine
adrenal gland were used as positive controls, and those from the spleen
of 21-day-old male rats were used as negative controls. P450c17 was
expressed in the liver of 7- and 21-day-old male rats at a higher level
than in adult rat testis, but was undetectable in the liver before
birth and in adulthood (Fig. 3A). This pattern of
expression correlates well with the temporal pattern of P450c17
activity in male rat liver (Fig. 2A), although the decrease in activity
observed between 7–21 days was not matched by a corresponding decrease
in protein content. Immunoblot analysis of microsomal proteins from
female rat liver at the same time points showed the same temporal
pattern of expression as in the male (Fig. 3B). The enzyme was absent
from the liver of adult females even during pregnancy (18 days of
gestation) and lactation (end of the first week) despite the hepatic
hypertrophy induced by the enhanced metabolic demand (28) (data not
shown). Thus, the expression of P450c17 in rat liver is not sexually
dimorphic. P450c17 activity, and protein and mRNA expression in adult
rat stomach and duodenum have been previously reported (11, 12, 13, 14).
|
2 kilobases) in the rat testis (22, 27).
Hybridization of 32P-labeled rat testis P450c17 cDNA to
total RNA from male and female rat livers at different ages is shown in
Fig. 4 (upper panel). Total mRNAs extracted
from both adult testis and 21-day-old ovary were used as positive
controls, and adult female spleen mRNA was used as a negative control.
A strong and unique signal was present in both male and female livers
at 7 and 21 days of age. A faint signal was detected in adult female
liver (68–74 days), whereas no signal was observed in either sex
before birth (18 days gestation) or in adult male liver (68–74 days).
The marked age-dependent presence of P450c17 mRNA correlates with that
of P450c17 protein in both male and female rat livers. An exception is
adult female liver, in which a very small quantity of mRNA was present,
but no protein could be detected. However, the amount of hepatic
P450c17 mRNA was greater at 21 days than at 7 days in both sexes,
whereas no such difference could be detected at the protein level.
|
Characterization of the cDNA encoding rat liver cytochrome
P450c17
To determine whether the difference in size between liver and
gonadal transcripts was due to sequence differences within the coding
region, liver P450c17 cDNA from 29-day-old male rat liver was amplified
with oligo pairs 1, 2, and 3 (Table 2) in three
overlapping pieces covering the full sequence (nucleotides -29 to
+1571) (22), and then completely sequenced in both directions. The
sequence was identical to that reported for rat testis (22), stomach,
and duodenum (12) cDNAs. Thus, the size difference between hepatic and
gonadal mRNAs is probably due to longer 5'- and/or 3'-untranslated
regions of the liver mRNA.
To confirm the presence of hepatic mRNA encoding authentic rat
cytochrome P450c17 in both sexes at different ages, total liver RNAs
from prenatal (18 days postcoitum), postnatal (7 days), juvenile (21
days), and adult (85–90 days) male and female rats were amplified by
RT-PCR with oligo pair 4 (Table 2). A unique fragment of the expected
size was detected by ethidium bromide staining in each liver sample,
but not in spleen, muscle, or blood from adult animals used as negative
controls (not shown). The presence of an amplified fragment at ages
that were negative by Northern or Western blot analyses can be
attributed to the greater sensitivity of RT-PCR. Approximately 100 bp
of the 744-bp amplificate were directly sequenced in all samples and
found to be identical to rat testis P450c17 cDNA (22).
Expression and characterization of mRNA encoding P450scc and
P450arom during male and female rat liver development
To establish whether the liver could autonomously produce the
steroid substrates (PREG or PROG) for P450c17, and whether estrogens
could be additional products of liver steroidogenesis, the expression
of P450scc and P450arom was examined in male and female livers during
development. Hepatic poly(A)+ mRNAs extracted from fetal
(18 days postcoitum), neonatal (7 days), juvenile (21 days), and adult
(68–74 days) male and female rats were examined by Northern blot
analysis. No evidence of expression of either gene was observed at any
age examined. When total hepatic mRNAs extracted from the same age
rats were analyzed by RT-PCR, a unique amplified fragment of the
correct size for P450scc cDNA was visualized on the ethidium
bromide-stained gel in fetal male and female livers (data not shown).
The identity of this amplificate with ovarian rat P450scc cDNA (23) was
verified by direct sequencing of about 100 bp. Expression of P450arom
mRNA was detected and confirmed by direct sequencing (30) in female
liver before birth (data not shown). In male liver, a faint band of the
correct size was detected before birth and at 7 days, but the quantity
of the amplified product was not sufficient for direct sequencing.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/C-17,20-lyase activities, such as stomach, duodenum
(11, 12, 13, 14), esophagus, and colon (12), or 17-hydroxylase activity
alone, such as submandibular and sublingual glands (10), kidney, and
skin (11). Kinetic analysis showed that Km values for
17-hydroxylase/C-17,20-lyase activities of rat duodenal P450c17 in
both sexes are comparable to those in the testis, whereas androgen
synthesis is more efficient in the duodenum (14). However, in all of
these previously investigated extraglandular sites, only low levels of
enzymatic activities were observed, suggesting local, rather than
systemic, function. In the rat liver, on the other hand, P450c17
specific activity attains a level during the second postnatal week that
exceeds not only those of other extraglandular sites but even that of
the testis. In point of fact, the level at 8 days after birth was more
than 4-fold greater than that found in the testis at the same age or in
adult animals and more than 80 times higher than the average activity
found in stomach and duodenum.
Although CYP17 gene expression in rat liver has not been
previously demonstrated, two studies actually contain cursory evidence
suggestive of its presence. Swinney et al. (17), analyzing
the regioselective hydroxylation of PROG by rat liver microsomes,
observed a weak 17-hydroxylation in 3-week-old males, but not in
adult males and females. This finding, however, was disregarded because
among 11 purified hepatic cytochromes P450, none had any
17-hydroxylase activity. Imaoka et al. (16) sequenced the
NH2-termini of 6 hepatic cytochromes P450 from immature
female rats and reported a 12-amino acid sequence identical to that of
rat P450c17, although this was not pointed out by the authors.
It is well established that a single CYP17 gene is present in the rat genome (27). Our Northern analyses demonstrate that both male and female livers contain a P450c17 mRNA about 150–200 bases longer than that in the gonads. This difference in size is not accounted for by the coding region, whose sequence was found to be identical to the testicular one (22). Thus, it could result from a differential 3'-polyadenylation, the use of different transcriptional start sites, or alternative splicing at the 5'-end. The only evidence regarding the use of different CYP17 promoters comes from a study in the pig, in which transcription starts at different sites in gonads and trophoblasts (31). The possibility that extraglandular expression of rat CYP17 is via an alternate promoter is intriguing because it suggests the possibility of tissue-specific developmental regulation. However, further analyses of the 5'- and 3'-termini of P450c17 transcripts in rat extraglandular tissues is necessary before conclusions can be drawn.
Based on analysis of steroid metabolites, immunoblot analysis, Northern
analysis, and cDNA sequence determination, it is evident that hepatic
17-hydroxylase/C-17,20-lyase activities result from the presence of
P450c17, the same enzyme that is present in steroidogenic organs. This
is very different from hepatic C21-hydroxylation in the rat, where
P450c21 is absent and P4502C6 supports this activity (32, 33). The
physiological significance of the strictly temporal and intense
expression of P450c17 in the rat liver is not obvious. Nevertheless, a
coherent interpretation can be proposed by framing our data within the
vast body of evidence found in the literature on the steroid
endocrinology of postnatal rats. The following points should be
considered.
First of all, the postnatal rat liver cannot be regarded as an
autonomous source of steroid hormones because expression of P450scc was
not detectable after birth in either sex. Secondly, the postnatal liver
does not seem to be a site of substantial androgen aromatization
because expression of P450arom after birth was extremely low, if
present at all. Hence, androgens synthesized by hepatic P450c17 and
17ß-hydroxysteroid oxidoreductase (34) would be terminal secretory
products, if they were not substrates for other catabolic hepatic
enzymes. Thirdly, a hepatotropic action of hepatic androgens is
unlikely, inasmuch as the rat liver is insensitive to androgens from a
few days after birth until puberty (<35–40 days) in both sexes due to
minimal or no expression of the androgen receptor gene (35), which is
not even inducible by androgen treatment during this period (36). In
particular, sexual dimorphism in hepatic steroid-transforming enzymes,
such as 3ß-hydroxysteroid dehydrogenase type III (37, 38), 16- and
6ß-hydroxylases (male-specific), and 15ß-hydroxylase
(female-specific) (39, 40, 41), cannot be ascribed to hepatic androgen
action. Rather, it appears to be due to androgen imprinting by fetal
and neonatal testis via the induction of a male pattern of GH secretion
(40) because it is abolished by neonatal castration and cannot be
restored by androgen administration after day 5 (41), and because the
pattern of expression of hepatic P450c17 peaks during the second
postnatal week and is not sexually dimorphic (this study).
Thus, if a paracrine/intracrine role of hepatic androgens is ruled out, then products of P450c17 activity may act extrahepatically either as hormones on androgen-sensitive targets or as aromatizable prohormones. Although the first alternative remains to be explored, there are consistent indications suggesting that hepatic androgens may actually participate in gonadal cellular differentiation after conversion to estrogens. The elements in favor of this hypothesis should be examined separately for the testis and the ovary, even though this type of gonadotropic action would be similar in both organs and, thus, not intrinsically related to gonadal sexual differentiation.
In the rat testis the following events take place during the first 4
postnatal weeks. 1) The fetal Leydig cells, which appear at 15–16 days
of pregnancy, undergo a process of involution during the first
postnatal week and are replaced by adult Leydig cells proliferating
from the second week up to the time of puberty (about 60 days) (26, 42). 2) This renewal in Leydig cell populations is characterized by a
drop in total Leydig cell number (42), a lowered concentration of
hCG/LH receptors (26), and a decrease in 3ß-hydroxysteroid
dehydrogenase (42), 17ß-hydroxysteroid dehydrogenase (34), and
17-hydroxylase/C-17,20-lyase activities (Ref. 43 and this work) and
P450c17 in immunodetectable protein (44) during the first 2 weeks. 3)
Sertoli cells proliferate during the first 2 weeks concomitantly with a
steep increase in FSH receptors (26) and high aromatase activity (45).
This converts exogenous androgens into 17ß-estradiol, whose mitogenic
action sustains Sertoli cell proliferation and whose circulating levels
continue to increase throughout the second week (46). As circulating
androgen levels remain high during this period (46) despite the shift
in Leydig cell populations and their presumable functional impairment
caused by involution or proliferation, it is suggested that Sertoli
cells are actually using androgens of hepatic origin for aromatization.
This would explain the peak of hepatic P450c17 expression and activity
during the first 2 weeks.
For the ovary, the evidence in favor of an analogous ovarian-hepatic interaction is based on the following facts. 1) At birth, the ovary consists of undifferentiated stroma and primary follicles; granulosa cells are already proliferating by day 5, whereas proliferation and differentiation of thecal cells start om day 9. Preantral follicles are visible by day 7, and antral follicles by day 12 (47). 2) A sharp peak in ovarian aromatase activity occurs from days 4–16 (maximum on day 10) (43, 47) and is matched by coincident surges of circulating 17ß-estradiol (46, 48) and FSH levels, peaking in the second week and declining completely during the third week (46, 49). 3) Cytochrome P450scc and P450c17 enzymatic activities are undetectable in the ovary until day 8 and increase significantly from day 12 onward (47); 3ß- and 17ß-HSD activities are instead constitutively present from the time of birth (47); hence, ovarian androgen synthesis from C27 or C21 precursors to sustain aromatization is impaired before day 12, although some androgen interconversions are possible. 4) Circulating androgen levels remain constant from birth to puberty (46), although a transient increase has been noticed from days 8–16 (50). 5) We found that P450c17 mRNA and protein are present in the female liver during the second and third weeks and are undetectable before birth and in the adult, thus showing the same pattern of expression as in the male liver. Therefore, it is tempting to speculate that granulosa cells, under the influence of FSH, aromatize hepatic androgens into estrogens, which promote their proliferation. However, a direct measurement of the time course of P450c17 specific activity in both the liver and ovary of female rats is needed.
A further point of consideration concerns the possibility of supplying the liver with progestogens during the first 3 postnatal weeks. As a matter of fact, during this period, serum progesterone was found to remain at 2–4 ng/ml in both sexes (46, 49), a concentration higher than that of TST (0.5–1 ng/ml) or estradiol (0.1–0.2 ng/ml) (46). Considering that the adrenal cortex of both sexes expresses P450scc from day 16.5 of pregnancy to adulthood (46), and that the plasma corticosterone concentration is extremely low from birth to day 12 despite good ACTH responsiveness during this period (51, 52), it should be determined whether postnatal corticosteroidogenesis is temporarily curtailed in the rat, resulting in substantial intermediate progestogen release. Because the binding capacity of corticosteroid-binding globulin is relatively depressed during the first 2 weeks (52), most circulating progesterone would be in a free form and thus readily available for androgen synthesis in the liver. The extremely high level reached by hepatic P450c17 during the first 2 weeks would be required to overcome nonandrogenic progesterone metabolism by other hepatic P450 cytochromes. The high level of P450c17 would compete favorably with these other activities, thus assuring androgen synthesis during this period.
It is interesting to note that this hypothetical adrenal (or any other
progestogen-secreting organ)-liver-gonadal steroidogenic
complementarity is reminiscent of fetal adrenal-fetal liver-placental
cooperation in estriol synthesis during human pregnancy (53) and
skin-liver-kidney interplay in 1,25-dihydroxycholecalciferol
formation (54). In these well established systems, the liver performs
an intermediate role restricted to a single hydroxylation step (16-
and 25-hydroxylation, respectively), as would be the case in the
proposed model.
Finally, when the expression of P450c17 in liver is compared with that in stomach and duodenum, it is evident that marked differences exist in both activity level and peak timing. In fact, gastric and duodenal enzyme specific activities were low compared with hepatic activities and increased at a later time during weaning (day 16 onward). This suggests that extraglandular P450c17 may be involved in different regulatory circuits in the various organs. Although it may sustain an endocrine function in postnatal hepatocytes, a paracrine commitment seems more likely in stomach and duodenum. In particular, in the stomach, P450c17 was found to be selectively expressed in parietal cells (13), which are the site of HCl production. It has been suggested that androgens released by these cells might counteract locally the detrimental effect of acidic secretion by supporting the intense regenerative turnover of the gastric mucosa (14). As in the duodenum there are no parietal cells, other cell types and regulatory mechanisms should be associated with its P450c17 activity. We conclude that investigation of extraglandular steroidogenesis in the rat may reveal unsuspected regulatory networks operating serially or in parallel with those of glandular steroidogenesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 9, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/17,20-lyase messenger ribonucleic acid and protein,
and metabolism of pregnenolone and progesterone by parietal cells of
the rat gastric mucosa. Endocrinology 136:1744–1752[Abstract]
-hydroxylase from calf
testis. J Steroid Biochem Mol Biol 40:431–435[CrossRef][Medline]
-hydroxylase/17,20-lyase (P450c17) in Escherichia coli.
J Biol Chem 268:19681–19689 cDNA: the deduced amino acid sequence,
expression and secondary structural configuration. Biochem Biophys Res
Commun 157:705–712[CrossRef][Medline]
from
testis: characterization of a full-length cDNA encoding a unique
steroid hydroxylase capable of catalyzing both 
4- and
5-steroid-17,20-lyase reactions. Mol Endocrinol 3:968–9755-4-isomerase activity in
rat liver by sex steroids and pituitary hormones. Endocrinology 131:3034–3044This article has been cited by other articles:
 |
|
 |
|
  L. L. Grasfeder, S. Gaillard, S. R. Hammes, O. Ilkayeva, C. B. Newgard, R. B. Hochberg, M. A. Dwyer, C.-y. Chang, and D. P. McDonnell Fasting-Induced Hepatic Production of DHEA Is Regulated by PGC-1{alpha}, ERR{alpha}, and HNF4{alpha} Mol. Endocrinol., August 1, 2009; 23(8): 1171 - 1182. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Shi, M. D. Schonemann, and S. H. Mellon Regulation of P450c17 Expression in the Early Embryo Depends on GATA Factors Endocrinology, February 1, 2009; 150(2): 946 - 956. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Aizawa, M. Iemitsu, T. Otsuki, S. Maeda, T. Miyauchi, and N. Mesaki Sex differences in steroidogenesis in skeletal muscle following a single bout of exercise in rats J Appl Physiol, January 1, 2008; 104(1): 67 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Mellon, S. R. Bair, C. Depoix, J.-L. Vigne, N. B. Hecht, and P. B. Brake Translin Coactivates Steroidogenic Factor-1-Stimulated Transcription Mol. Endocrinol., January 1, 2007; 21(1): 89 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Pignatelli, F. Xiao, A. M Gouveia, J. G Ferreira, and G. P Vinson Adrenarche in the rat. J. Endocrinol., October 1, 2006; 191(1): 301 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Browne, N J Place, J D Vidal, I T Moore, G R Cunha, S E Glickman, and A J Conley Endocrine differentiation of fetal ovaries and testes of the spotted hyena (Crocuta crocuta): timing of androgen-independent versus androgen-driven genital development. Reproduction, October 1, 2006; 132(4): 649 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bhusari, L. B. Hearne, D. E. Spiers, W. R. Lamberson, and E. Antoniou Effect of fescue toxicosis on hepatic gene expression in mice J Anim Sci, June 1, 2006; 84(6): 1600 - 1612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Liu, Z.-X. Yao, and V. Papadopoulos Cytochrome P450 17{alpha} Hydroxylase/17,20 Lyase (CYP17) Function in Cholesterol Biosynthesis: Identification of Squalene Monooxygenase (Epoxidase) Activity Associated with CYP17 in Leydig Cells Mol. Endocrinol., July 1, 2005; 19(7): 1918 - 1931. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Nakamura, M. Yoshimura, M. Nakayama, T. Ito, Y. Mizuno, E. Harada, T. Sakamoto, Y. Saito, K. Nakao, H. Yasue, et al. Possible Association of Heart Failure Status With Synthetic Balance Between Aldosterone and Dehydroepiandrosterone in Human Heart Circulation, September 28, 2004; 110(13): 1787 - 1793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Boverhof, K. C. Fertuck, L. D. Burgoon, J. E. Eckel, C. Gennings, and T. R. Zacharewski Temporal- and dose-dependent hepatic gene expression changes in immature ovariectomized mice following exposure to ethynyl estradiol Carcinogenesis, July 1, 2004; 25(7): 1277 - 1291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Borlak, I. Schulte, and T. Thum ANDROGEN METABOLISM IN THYMUS OF FETAL AND ADULT RATS Drug Metab. Dispos., June 1, 2004; 32(6): 675 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. V. Solovyeva, M. Hayashi, K. Margi, C. Barkats, C. Klein, A. Amsterdam, A. J.W. Hsueh, and A. Tsafriri Growth Differentiation Factor-9 Stimulates Rat Theca-Interstitial Cell Androgen Biosynthesis Biol Reprod, October 1, 2000; 63(4): 1214 - 1218. [Abstract] [Full Text]
|
|
|
|
|
|
X. Chu, C. J. Corbin, M. A. Kaminski, and A. J. Conley Unique Regulation of CYP17 Expression in the Trophectoderm of the Preattachment Porcine Blastocyst Endocrinology, February 1, 1999; 140(2): 632 - 640. [Abstract] [Full Text]
|
|
|
|
|
|
M.A. Kaminski, C.J. Corbin, and A.J. Conley Development and Differentiation of the Interstitial and Tubular Compartments of Fetal Porcine Testes Biol Reprod, January 1, 1999; 60(1): 119 - 127. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
