| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
-Hydroxysteroid Dehydrogenase Activity of Rat Leydig Cells as Type II Retinol Dehydrogenase1
Population Council and The Rockefeller University (D.O.H., R.-S.G., J.F.C., M.P.H.), New York, New York 10021; and the Department of Pharmacology (Y-t.H., T.M.P.), University of Pennsylvania, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: Dianne O. Hardy, Population Council, 1230 York Avenue, New York, New York 10021. E-mail: d-hardy{at}popcbr.rockefeller.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reductase, 2) conversion of DHT to
5-androstane-3,17ß-diol (3-DIOL) by the reductive activity
of 3-hydroxysteroid dehydrogenase (3-HSD), and 3) conversion of
3-DIOL by an oxidative 3-HSD activity. While the type I 3-HSD
enzyme (3-HSD1 or AKR1C9) is an oxidoreductase in
vitro and could theoretically be responsible for factors 2 and
3, we have shown previously that rat Leydig cells have two 3-HSD
activities: a cytosolic NADP(H)- dependent activity, characteristic of
3-HSD1, and a microsomal NAD(H)-dependent activity. The two
activities were separable by both developmental and biochemical
criteria, but the identity of the second enzyme was unknown. To
identify the microsomal NAD(H)-dependent 3-HSD in rat Leydig cells,
degenerate primers were used to amplify a number of short-chain alcohol
dehydrogenases. Sequence analysis of cloned PCR products identified
retinol dehydrogenase type II (RoDH2) as the prevalent species in
purified Leydig cells. RoDH2 cDNA was subcloned into expression vectors
and transiently transfected into CHOP and COS-1 cells. Its properties
were compared with transiently transfected 3-HSD1. When measured in
intact CHOP and COS-1 cells, RoDH2 cDNA produced a protein that
catalyzed the conversions of 3-DIOL to DHT and androsterone to
androstanedione, but not the reverse reactions. Therefore, the 3-HSD
activity of RoDH2 was exclusively oxidative. In contrast, type I
3-HSD cDNA produced a protein that was exclusively a 3-HSD
reductase. In cell homogenates and subcellular fractions, RoDH2
catalyzed both 3-HSD oxidation and reduction reactions that were
NAD(H) dependent, and the enzyme activities were located in the
microsomes. Type I 3-HSD also catalyzed both oxidation and
reduction, but was located in the cytosol and was NADP(H) dependent. We
conclude that type I 3-HSD and RoDH2 have distinct 3-HSD
activities with opposing catalytic directions, thereby controlling the
rates of DHT production by Leydig cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reductase; DHT then binds to the androgen receptor with a
dissociation constant (Kd) of 2.0 x
10-10 M (2, 3). A major
catabolic pathway for DHT is catalyzed by 3-hydroxysteroid
dehydrogenase (3-HSD), which converts DHT into a weak androgen,
5-androstane-3,17ßdiol (3-DIOL), which has a
Kd of 10-6
M (3, 4) for the androgen receptor. Reduction of
5-androstanedione to androsterone is also catalyzed by 3-HSD.
Androsterone is readily glucuronidated and excreted (5). The
interconversion of DHT to 3-DIOL catalyzed by 3-HSD is
potentially reversible.
Leydig cells are the major site for T production and also secrete
abundant amounts of DHT and 3-DIOL during pubertal development
(6, 7, 8). Production of DHT can occur through 5-reduction of T and
through 3-HSD oxidation of 3-DIOL. If 3-HSD reductive activity
predominates, DHT is converted into 3-DIOL. However, if 3-HSD
oxidative activity predominates, then 3-DIOL from the circulation
can be converted to DHT.
The most thoroughly characterized 3-HSD is the enzyme expressed in
rat liver (9). This enzyme, designated as type I 3-HSD (3-HSD1 or
AKR1C9) (10), was first purified from liver, and its cDNA was cloned
and shown to be a member of the aldo-keto reductase (AKR) superfamily
(10, 11, 12, 13, 14). Analysis of 3-HSD expression and activity (15, 16, 17) in
Leydig cells revealed that developmental decreases in type I 3-HSD
messenger RNA (mRNA) levels were correlated with a loss in 3-HSD
reductive activity. In addition to 3-HSD1, Leydig cells also contain
a microsomal NAD+-dependent enzyme that catalyzes
the oxidation of 3-DIOL to DHT, and we suggested that, in intact
cells, 3-HSD1 may act as a reductase and the microsomal enzyme as an
oxidase (17). Importantly, the oxidative activity did not correlate
with type I mRNA expression. In the current study we sought to identify
the gene responsible for the microsomal oxidative 3-HSD activity. A
number of enzymes, including human 11-cis -retinol
dehydrogenase (11-cis-RoDH) and mouse
cis-retinol/3-hydroxysterol short-chain dehydrogenase
(CRAD), catalyze unidirectional oxidation of 3-DIOL to DHT when
measured in intact transfected cells (18, 19). As the primary sequences
of these enzymes are conserved, degenerate primers were designed to
amplify mRNAs of short-chain dehydrogenase/reductases (SDRs), in Leydig
cells, that could be responsible for the conversion of 3-DIOL to
DHT. We then cloned and sequenced PCR products to identify which
enzymes were present. Only clones identical to retinol dehydrogenase
type II (RoDH2) were identified in Leydig cells, clearly implicating
this gene as the source of the oxidative activity. Transfection studies
showed that RoDH2 was exclusively oxidative in intact cells, whereas
3-HSD1 was only reductive. These two enzymes, RoDH2 and 3-HSD1,
control the level of DHT synthesized by the Leydig cell.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-[9,11-3H(N)]androstane-3,17ß-diol,
and [1,2-3H(N)]androsterone were purchased from
NEN Life Science Products (Boston, MA). Nonradioactive
steroids were purchased from Sigma (St. Louis, MO) or
Steraloids (Wilton, MA).
Cell isolation
Leydig cells from 90-day old rats were purified as described
previously (20). Cell preparations were judged to be more than 95%
pure by histochemical staining for the Leydig cell-specific marker
3ß-hydroxysteroid dehydrogenase (21).
First strand synthesis
Human kidney poly(A)+ selected RNA (0.5 µg), human liver
poly(A)+ selected RNA (0.5 µg), rat testis poly(A)+ selected RNA (2
µg), rat kidney total RNA (5.5 µg), and rat adult Leydig cell total
RNA (5 µg) served as templates for DNA synthesis by Moloney murine
leukemia virus (M-MLV) reverse transcriptase (M1701,
Promega Corp., Madison, WI) in 40 µl reactions
containing 1 µM oligo(dT), 0.5 mM
deoxyribonucleotides, 50 mM Tris-HCl (pH 8.3 at 25 C), 75
mM KCl, 3 mM MgCl2, and
10 mM dithiothreitol.
Primer selection
All primers in this study were selected with the assistance of
Primer Software (The Whitehead Institute for Biomedical Research,
Cambridge, MA) following guidelines for internal stability (22).
Oligonucleotides were synthesized by The Rockefeller University’s DNA
and Protein Sequencing Facility, Genosys (The Woodlands, TX), or
Biosource International (Camarillo, CA). For amplification
of 11-cis-RoDH sequences in the rat, human (23) and mouse
(24) sequences were aligned using FASTA software (25).
Oligonucleotide sequences corresponding to regions of exact match were
chosen for high thermal stability (Tm > 70 C).
The forward 11-cis-RoDH primer was
5'-dGTTTGGCCTGGAGGCCTTCTCTGN-3'; the reverse primer was
5'-dCAGTCAGGGCATGCTCCAGGCN3-'. The expected product size was 285 bp.
The CRAD-specific primers were based on the mouse sequence (19), which
was the only CRAD sequence available from the database.
Oligonucleotides with Tm = 72 C were selected,
with 5'-dCTGACCGGTGTGACCAGTAGTGCCAGAN-3' as the forward primer and
5'-dGAGGGCTTTTTCAGGCTTCAGGGAAGTN-3' as the reverse primer. The expected
product size for CRAD PCR amplification was 324 bp. Generalized primers
were designed by alignment of several protein sequences, including
11-cis-RoDH (23, 24), CRAD (18), RoDH (types I, II and III)
(26, 27, 28), 17ß-HSD VI (29), and human 3-HSD oxidative enzyme (30)
using BLAST software (National Center for Biotechnology Information,
National Library of Medicine, National Institutes of Health). Some
codons in the primers were degenerate to allow for unknown codon usage.
The range of Tm for both primers was 70-76 C.
The sequence of the forward primer was 5'-dACGGGCTGTGACTCNGGCTTYGG-3',
and the reverse was 5'-dCTTGGCATCCCAGCCNGGTGARTA-3', where N =
A+C+G+T, Y= C+T, and R = A+G. Thus each primer had a degeneracy of
8. The expected product size was 767 bp. To obtain a PCR product
containing the entire coding sequence for RoDH2, the forward primer was
5'-dTGTCCCTCTGCTTGTCTTCT-3', and the reverse primer was
5'-dGATCCCCTCCCTAACACTGT-3'.
PCR
Buffer conditions for amplification of 11-cis-RoDH
were: 50 mM KCl, 10 mM
Tris-HCl (pH 9.0 at 25 C), 1.0% Triton X-100, 1.5
mM MgCl2, 50
µM dATP, deoxycytidine triphosphate, dGTP, and
deoxythymidine triphosphate, and each primer at 1
µM. The thermal cycle parameters were 94 C for
15 sec (denaturing), 65 C for 30 sec (annealing), and 72 C for 15 sec
(extension) for 30 cycles. Buffer conditions for amplification of CRAD
were the same except the concentration of MgCl2
was 1.5, 3.0, 4.5, or 6.0 mM. The thermal cycle
parameters were 94 C for 15 sec and 70 C for 40 sec
(annealing/extension) for 20 cycles with a decrease of 1 C per cycle in
the annealing/extension step, followed by 94 C
for 15 sec, 50 C for 30 sec, and 72 C for 15 sec for 20 cycles. Buffer
conditions using mixed consensus primers were the same except the
concentration of MgCl2 was 2.0
mM and the concentration of forward or reverse
primer was 2.0 µM. Thermal cycle parameters
were 94 C for 15 sec, 65 C for 30 sec, and 72 C for 48 sec for 35
cycles. For amplification of the entire coding region of RoDH2, thermal
cycle parameters were 94 C for 15 sec, 55 C for 30 sec, and 72 C for 1
min for 30 cycles, with an additional 9-min extension in the final
cycle. Reaction volumes of 50 µl were equilibrated to 94 C before
addition of 1 U of Taq DNA polymerase (in storage buffer B:
Promega Corp.). PCR products were analyzed by
electrophoresis on 5% polyacrylamide gels in Tris-borate EDTA buffer
followed by ethidium bromide staining.
Direct cloning and sequence analysis
PCR products were directly inserted into linearized pCR3.1
vectors with unmatched 3'-deoxythymidine residues using the Original TA
Cloning Kit, or for expression plasmids, the pCR3.1 vector using the
Eukaryotic TA Cloning Kit (Unidirectional) (Invitrogen,
San Diego, CA). Plasmids were purified using either Wizard Plus
Minipreps DNA Purification System (Promega Corp.) or
QIAprep Spin Miniprep Kit (QIAGEN Inc., Valencia, CA).
Samples were submitted to The Rockefeller University Protein DNA
Technology Center for automated sequence analysis. Expression plasmids
in pCR3.1 were subcloned into pcDNA1.1 for episomal replication in CHOP
cells. An expression plasmid containing the complete coding region of
type I 3-HSD in the vector pRc/CMV was constructed for transient
3-HSD expression (31).
Transient transfection
CHOP cells (32) were maintained in -MEM (Life Technologies, Inc., Gaithersburg, MD) supplemented with 10% FCS
(Life Technologies, Inc.) and 5%
CO2 at 37 C. COS-1 cells were maintained in DMEM
(Life Technologies, Inc.) supplemented with 10% FCS and
5% CO2 at 37 C. For transfection, 100,000 cells
were seeded per well in a six-well plate and cultured for 24 h in
media supplemented with charcoal-stripped FCS to obtain 50–80%
confluence. Transfection was performed using the FuGENE 6 Transfection
Reagent (Roche Molecular Biochemicals, Indianapolis, IN)
according to the manufacturer’s protocol. Titration experiments using
0, 0.5, 1, 2, and 4 µg DNA per well showed maximal efficiency at 1
µg DNA; therefore, this quantity was used in the transfection
assays.
Enzyme assays
For assays of enzyme activity in intact transfected cells,
24 h after transfection the cell culture media were replaced with
4 ml media containing 1 µM steroid substrates (with 10%
radiolabeled substrates). Aliquots of media (0.5 ml) were collected at
timed intervals ranging from 1 h through 6 h for COS-1 cells
and 1–24 h for CHOP cells. For assays of enzyme activity in
transfected cell lysates and subcellular fractions, cells were scraped
from dishes 24 h after transfection into ice-cold PBS and
collected by centrifugation. Cell pellets were homogenized in 10
mM potassium phosphate, pH 7.0, 150 mM KCl, 1
mM EDTA, and 0.25 M sucrose with a Dounce
tissue grinder (Wheaton, Millville, NJ). Microsomal and cytosolic
fractions were harvested after subsequent centrifugation at 10,000
x g for 1 h and twice at 105,000 x g
for 1 h. The protein concentrations in cell lysates and
subcellular fractions were determined using a kit (No. 500-0006,
Bio-Rad Laboratories, Inc., Hercules, CA) with BSA as a
standard. Mixtures (500 µl) containing 20 µg protein, 500
µM cofactors, and 0.01–5
µM radiolabeled steroid substrates were
incubated at 37 C for 10 min. The steroids were extracted from spent
media or reaction mixture with 2 ml of ice-cold ethyl acetate, and the
organic layer was evaporated under nitrogen gas. The steroids were
separated chromatographically on thin layer plates (Baker-flex,
Phillipsburg, NJ) in diethyl ether and acetone (98:2). The
radioactivity was measured with a scanning radiometer (System
200/AC3000, Bioscan, Inc., Washington DC). The conversion
of steroid to product was calculated as a percentage of the total
radioactivity found in the product. Kinetic constants for steroid
substrates were calculated by conventional Lineweaver-Burke analysis.
All assays were repeated in triplicate or quadruplicate from different
transfection experiments.
Data analysis
In each experiment, data were obtained from triplicate assays
and the results were expressed as the mean ± SEM.
Statistical analysis of the changes in 3-HSD oxidative and reductive
activities was performed by Kruskal-Wallis ANOVA followed by multiple
comparisons testing to identify significant differences between groups
(33).
Kinetic analysis was performed by fitting initial velocity data as a function of substrate concentration to a Michaelis-Menten equation of the form: v = [VmaxS/(Km + S)], where S is substrate concentration and Km and Vmax are the parameters. Parameters and their estimated SD values were determined by weighted nonlinear least-squares curve fitting using the SAAM II optimizer (34). The appropriate model was selected by assessing goodness of fit to the hyperbolic equation using multiple criteria including the principal of parsimony (Akaike information criterion and Schwarz criterion), and F-tests for comparison between models (34).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-HSD
oxidative activity in adult Leydig cells-HSD
oxidative activity (18, 23), and in initial attempts to identify the
SDR present in rat Leydig cells, we analyzed Leydig cell RNA samples
for the presence of 11-cis-RoDH and CRAD mRNAs by RT-PCR.
For 11-cis-RoDH, we selected primers based on the common
sequences found in human (23) and mouse (24) 11-cis-RoDH
cDNA. The PCR primers amplified products of the expected size of 285 bp
based on the 11-cis-RoDH sequence using human, mouse, and
rat kidney templates as positive controls, but this product was not
detected in templates from rat adult Leydig cells. For detection of
CRAD, we selected primer sequences from the mouse CRAD cDNA sequence
(18). The expected CRAD product of 324 bp was amplified by PCR using
mouse kidney cDNA template as a positive control, but it did not
amplify cDNA made from rat kidney, human kidney, or rat adult Leydig
cells. These data eliminated 11-cis RoDH and CRAD as the
source of 3-HSD oxidation in adult rat Leydig cells.
PCR identification of membrane-bound short-chain alcohol
dehydrogenase
Given the difficulty of identifying which of the SDR family
members were present in adult rat Leydig cells by gene-specific PCR, we
devised an approach that would amplify several family members expressed
in Leydig cells. The resultant PCR products could then be identified by
cloning and sequencing. SDR protein sequences were examined to identify
similar regions in their putative cytosolic domains, and degenerate
primers were designed. The sequence of the forward primer relative to
sequences in other members of the membrane-bound SDR superfamily is
shown schematically in Fig. 1
. The
reverse primer was constructed by an analogous procedure. These mixed
consensus primers were used for PCR amplification of sequences present
in adult rat Leydig cell total RNA. This approach yielded a product of
the expected size of 767 bp for an SDR. As this PCR product could
contain single or multiple SDR sequences of similar size, the amplified
DNA was cloned by ligation into a plasmid vector. Eleven clones were
sequenced, all of which were identical to RoDH2 cDNA, indicating that
RoDH2 was the predominant, if not the sole, SDR expressed in rat Leydig
cells.
|
.
|
-HSD activities of RoDH2 and 3-HSD1 in intact
CHOP and COS-1 cells-HSD oxidative activity in Leydig cells, its
ability to convert 3-DIOL to DHT was examined by measuring enzyme
activity in transiently transfected CHOP cells. CHOP cells are a
variant of CHO cells that are stably transfected with polyoma virus and
were selected for transfection experiments because of their frequent
use in expression cloning studies of steroidogenic enzymes. A
full-length RoDH2 cDNA was constructed by PCR amplification of mRNA
template from adult Leydig cells; its identity was confirmed by DNA
sequencing analysis, and it was inserted into the vector pcDNA1.1. The
expression vector pcDNA1.1 contains the polyoma origin of replication
necessary for efficient replication in CHOP cells. The activities of
two independent clones were the same; it was therefore unlikely that
incorporation of mismatched bases by Taq DNA polymerase had
produced RoDH2 with altered activity. The expression vector itself was
used in a transfection assay as a negative control. An expression
vector containing full- length rat 3-HSD1 cDNA (22) was used for a
comparative analysis. In each paradigm we determined the direction of
3-HSD catalysis, cofactor preference, and intracellular
distribution. Four substrates were assayed: 3-DIOL or androsterone
for oxidation to DHT or androstanedione, and DHT or androstanedione for
reduction to 3-DIOL or androsterone, respectively.
When 3-DIOL was supplied as the substrate to intact CHOP cells that
had been transfected with RoDH2 expression vector, conversion to DHT
could be measured at 2 h; the amount converted to DHT increased
with time until 8 h, and at this endpoint 75% of the substrate
was converted (Fig. 2A). Culture media
from CHOP cells that had been transfected with pRc/CMV-3-HSD or
pcDNA1.1 did not form detectable amounts of DHT. Similar results were
obtained using androsterone as substrate since no androstanedione was
produced (Fig. 2C). For measurements of reductive activity, DHT was
supplied as substrate to transfected CHOP cells. Under these
conditions, cells that had been transfected with RoDH2 expression
vectors converted DHT to 3-DIOL, but the amount was exactly equal to
that achieved in cells transfected with pcDNA1.1 vector alone (Fig. 2B). The detectable levels of 3-HSD reduction from 8 to 24 h
was attributed to the endogenous 3-HSD reductive activity in CHOP
cells. In separate experiments (data not shown) in which velocity
measurements were taken at a single time point using the cytosolic
fractions of homogenates from CHOP cells alone or from CHOP cells that
had been transfected with pRc/CMV-3-HSD, the amount of activity
attributed to endogenous activity was 8.8% in the presence of NADPH
(reduced form of nicotinamide adenine dinucleotide) and DHT
substrate. CHOP cells transfected with type I 3-HSD had conversion
rates for DHT to 3-DIOL that were well above the background levels
attributable to endogenous 3-HSD activity; by 2 h, 35% of the
DHT substrate was converted and 85% was converted by 24 h. In the
linear range of reductive activity, 9% of the total activity was from
endogenous enzyme. When androstanedione was used as substrate, 10%
conversion occurred by 2 h and 70% by 24 h (Fig. 2D).
Endogenous enzyme accounted for 14% of the total 3-HSD activity.
These results indicate that RoDH2 has an exclusive 3-HSD oxidative
activity, whereas type I 3-HSD is an exclusive reductase in intact
transfected CHOP cells.
|
-HSD
reductase, COS-1 cells were assayed for endogenous 3-HSD activities
by addition of 3-DIOL or DHT to the culture medium. COS-1 cells did
not have endogenous 3-HSD activity but did convert 3-DIOL to
androsterone and therefore had weak endogenous 17ß-HSD oxidative
activity. The amount of RoDH2 expression plasmid was systematically
varied (0, 0.5, 1, 2, and 4 µg of DNA per assay well), and 24 h
after substrate addition, oxidative conversion of both 3-DIOL and
androstanedione occurred with as little as 0.5 µg of DNA. No
reductive conversion of DHT or androsterone occurred under any
conditions (Fig. 3, A and B). The amount
of conversion measured did not increase significantly when more than 1
µg of DNA was used per assay well. This amount of DNA was used in all
subsequent assays.
|
). This represents the background
oxidation by the endogenous 17ß-HSD in COS-1 cells. Against this
background, the amount of oxidative activity after the introduction of
either of RoDH2/pCR 3.1 cloned plasmids was 40% by 2 h, and 80%
by 4 h. No increase in oxidative activity was seen after
transfection with type 1 3-HSD (Fig. 4A). The endogenous 17ß-HSD
oxidase was not a consideration in androstanedione formation because
androsterone is not a substrate for this enzyme. Introduction of RoDH2
DNA resulted in 60% conversion of androsterone to androstanedione by
2 h, and 90% conversion by 8 h (Fig. 4C). The 3-HSD
reductive activity did not increase after transfection with RoDH2 cDNA.
In contrast, significant increases in 3-HSD reduction were observed
after transfection of COS-1 cells with type 1 3-HSD (Fig. 4, B and
D). These results corroborated the results in CHOP cells showing that
in intact cells RoDH2 is an oxidase and 3-HSD1 is a reductase.
|
-HSD and type I 3-HSD-HSD activity in Leydig cells, it would be
NAD(H)-dependent (17). Therefore, we measured the 3-HSD activities
of RoDH2 in the presence of different cofactors. The 3-HSD oxidative
activity of RoDH2, measured as the production of androstanedione from
androsterone by homogenates of COS-1 cells transfected with RoDH2, was
17.5 nmol/h/mg protein in the presence of NAD+
and less than 1 nmol/h/mg in the presence of
NADP+ or in the absence of cofactors (Fig. 5A). Reductive activity, measured as the
production of androsterone from androstanedione by homogenates of
transfected cells, was 4.5 nmol/h/mg protein for RoDH2 in the presence
of NADH and 1 nmol/h/mg protein in the presence of NADPH or no added
cofactor. These results indicate that, in contrast to its function in
retinol metabolism, RoDH2 uses NAD(H) as cofactors in 3
oxidation.
|
-HSD is NAD(P)(H) dependent (9). As a positive control
for NADP(H)-dependent activities, 3-HSD1 activities were measured in
transfected cell lysates in the presence of cofactors. The oxidative
activity of 3-HSD1 was 4 nmol/h/mg protein in the presence of
NADP+ and zero (a background rate of 0.5
nmol/h/mg protein or lower) in the presence of
NAD+ or in the absence of cofactor. The reductive
activity of type I 3-HSD was 4 nmol/h/mg protein in the presence of
NADPH and 1 nmol/h/mg protein in the presence of NAD or no added
cofactor (Fig. 4B
). These results indicate that, unlike 3-HSD1,
RoDH2 uses NAD(H) as cofactors in the whole cell
environment.
Subcellular fractionation of RoDH2 and 3-HSD1
In Leydig cells, NAD(H)-dependent 3-HSD activities are
localized to the microsomal fraction (17). Therefore, NAD(H)-dependent
3-HSD activities of RoDH2 were measured in subcellular fractions of
transfected COS-1 cells. NAD+-dependent 3-HSD
oxidative activity was 35 nmol/h/mg protein in microsomal fractions and
0.45 nmol/h/mg in cytosolic fractions after transfection with RoDH2
cDNA. NAD+-dependent oxidation was undetectable
in COS-1 cells treated with the transfection agent alone (Fig. 6A). NADH-dependent 3-HSD reductive
activity was 12.5 nmol/h/mg in microsomal fractions and 0.64 in
cytosolic fractions after transfection with RoDH2 (Fig. 6B). Therefore,
transfection with RoDH2 resulted in 3-HSD activities only in the
microsomal fraction.
|
-HSD oxidative
activity after 3-HSD1 transfection of type I 3-HSD was 3.9 nmol/h
in cytosolic fractions and 1.2 nmol/h in the microsomal fraction (Fig. 5C). The NADPH-dependent 3-HSD1 reductive activity was 5.5 nmol/h/mg
protein in cytosolic fractions and 1.4 nmol/h in the microsomal
fractions (Fig. 5D). This showed that the type I 3-HSD activity was
present predominantly in cytosolic fractions.
Kinetic parameters of RoDH2/3-HSD in transfected COS-1 cells
The kinetic characteristics of 3-HSD activities of RoDH2,
determined by incubating COS-1 cell lysates with appropriate steroid
substrates and cofactors, are presented in Table 2. The Michaelis-Menten constants
(Km) with substrates DHT and 3-DIOL were
similar to those in Leydig cells as previously reported (17). These
results suggest that RoDH2 is responsible for oxidation of 3-DIOL to
DHT in Leydig cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-HSD1
and that this enzyme behaves as a unidirectional NADPH-dependent
reductase in intact cells (17). In the prepubertal rat, 3-HSD1 is
the only 3-HSD present. In adult rats, a microsomal NAD(H)-dependent
3-HSD activity is also expressed. We inferred that the 3
oxidative activity was a separate gene product from 3-HSD1 because
of the temporal differences that exist in the detection of this enzyme
activity and the expression of 3-HSD1 mRNA and the different
subcellular locations of oxidative and reductive activities and their
different cofactor preferences. Subsequently, we were able to identify
the 3-HSD oxidative activity present in adult rat Leydig cells, and
these results are presented in this report.
By using a PCR cloning approach aimed at amplifying members of the SDR
family, the only member of this family that we found in purified adult
Leydig cells was RoDH2. We show that RoDH2 is an exclusive 3-HSD
oxidase in intact cells. Furthermore, type I 3-HSD activity was
exclusively reductive in intact cells, which contrasts the activity of
the purified enzyme and its properties in cell lysates. It appears that
both of these oxidoreductases, RoDH2 and rat type 1 3-HSD, are
unidirectional in intact cells but bidirectional in cell homogenates.
RoDH2 expression in adult Leydig cells and its activity as a 3-HSD
oxidase support the previous hypothesis that two distinct 3-HSD
enzymes are present during Leydig cell development (17).
The two 3-HSD enzymes may have specific functions during Leydig cell
development. The rat type 1 3-HSD is present in progenitor and
immature Leydig cells at higher levels compared with adult Leydig cells
(17) and is exclusively a reductase, metabolizing DHT to 3-DIOL
(15, 16, 17). This may account for high circulating concentrations of
3-DIOL in midpubertal rats (6, 7, 8). The oxidative 3-HSD in
immature and adult Leydig cells is distinguishable from the 3-HSD1
in progenitor Leydig cells. The latter enzyme has the same
characteristics as the enzyme in rat liver (9). It is cytosolic,
dependent on NADPH (9, 35), and exclusively reductive when its activity
was assayed in intact progenitor cells (17). We emphasize that the
directionality and cofactor preference observed by transfection in CHOP
cells and COS cells does not reflect the properties of purified type 1
3-HSD enzyme in vitro, which is bidirectional and uses
either pyridine nucleotide cofactor. Although type 1 3-HSD can use
NAD(P)(H) as cofactors as previously reported (9), NAD(H)-dependent
activities were far lower compared with NADP(H)-dependent activities,
especially in the presence of high concentrations of cofactors (seen in
the present study). Type 1 3-HSD is more dependent on NADP(H),
possibly because this cofactor lowers the Km for
the steroid substrate (9). In contrast to progenitor cells, an
enzymatic activity in immature and adult Leydig cells was microsomal
and dependent on NAD+ and was predominantly
oxidative when assayed in intact adult cells (17). The present study
showed that RoDH2 was microsomal, NAD+ dependent,
and catalyzed the oxidative formation of DHT from 3-DIOL.
Identification of the gene responsible for 3-hydroxysteroid
oxidation activity in Leydig cells is notable because the presence of
an oxidative activity will convert the weak androgen, 3-DIOL, to the
most potent endogenous androgen, DHT. Identification of a separate gene
could also explain the changes in 3-HSD oxidation that occur during
Leydig cell development. Candidates for new genes that might catalyze
3-HSD oxidation were sought among the SDR family. The family has
more than 50 members (36) and many, such as 11ß-HSD types I and II
(37, 38, 39, 40) and RoDH types I, II, and III (26, 27, 28) are membrane
bound. Members of this family share conserved primary and tertiary
structures and catalyze diverse biochemical reactions with different
steroids (41). Several SDRs, including 11-cis-RoDH (25) and
CRAD (18), have 3-HSD oxidative activities. To determine whether
11-cis-RoDH and CRAD were present in Leydig cells and
catalyze 3-HSD oxidation, primers were designed based on human and
mouse 11-cis-RoDH cDNA (23, 42) and on mouse CRAD cDNA (17).
The primers detected 11-cis-RoDH in rat kidney, but not in
rat adult Leydig cells. Therefore, either adult Leydig cells do not
contain 11-cis-RoDH or they might have
11-cis-RoDH that diverged in sequence from the kidney
enzyme. Another candidate, CRAD, has only been cloned from mouse kidney
(18). Degenerate primers based on the CRAD cDNA sequences were unable
to detect products in tissues other than mouse kidney. Since neither
11-cis-RoDH nor CRAD were found in rat Leydig cells, another
approach was required to identify the oxidative activity.
The membrane-bound members of the SDR family have extended stretches of hydrophobic amino acids resembling classic transmembrane domains (36). The structural model proposed by Gough et al. (43) for members of this family uses hydropathy plots to predict the cytosolic domains, transmembrane domains, and domains facing the lumen of the endoplasmic reticulum. In targeting primer design, we chose regions of similarity in the cytosolic domains since they might be required to maintain catalytic activity. Using degenerate primers representing all variants, several sequences were detected in liver, kidney, prostate, and Leydig cells. These degenerate primers were useful in identifying members of the SDR family when it was not known which ones were expressed in a given tissue. Using this approach, we identified three SDR members in rat liver by sequencing 9 clones, and we identified two SDR members in rat prostate by sequencing only 5 clones. The presence of 17ß-HSD and RoDH sequences in these liver and prostate clones was consistent with the relative prevalence of those enzymes reported originally (26, 27, 28, 29). While the combination of RT-PCR and subcloning is not a quantitative method to determine relative expression of SDR family members, we emphasize that it works as an identification tool. A limitation of this technique is that if one member is far more prevalent than others, as RoDH2 is in rat kidney, many clones may need to be sequenced before less prevalent enzymes can be identified. RoDH2 was the only mRNA detected in Leydig cells when 11 clones were sequenced. Therefore, RoDH2 mRNA was the most prevalent species present and must encode the unknown enzyme present in the immature and adult Leydig cells.
There are three isoforms of RoDH (types I, II, and III). RoDH2, cloned
from rat kidney (29), is involved in retinoid metabolism. Its cofactor
preference for conversion of retinol to retinal was shown to be NADPH
(26). RoDH1 is also NADPH dependent when catalyzing retinoid metabolism
and is known to convert 3-DIOL to DHT (29). It shares 80% sequence
identity with RoDH2 (26). These results indicate RoDH2, like RoDH1, may
also convert 3-DIOL to DHT.
Full-length expression clones of RoDH2 were constructed and analyzed
using RNA from adult Leydig cells as the template. PCR mutations did
not affect the activity of two independent expression clones. Both
oxidative and reductive activities were measured above background
levels seen in CHOP cells. The endogenous 3-HSD1 in CHOP cells was a
minor portion of the total activity in transfected cells, yet can
contribute to product formation (or back conversion) after longer
incubations. To further test the hypothesis that RoDH2 is a 3-HSD
oxidase and that 3-HSD1 is a reductase, and to eliminate the effects
of endogenous 3-HSD in CHOP cells, we also conducted experiments
with COS-1 cells. Full-length RoDH2 sequences were directly cloned into
an expression vector (pCR 3.1) containing SV40 promoter with an origin
for episomal replication. Although COS-1 cells had endogenous
17ß-HSD, which converted DHT product to androstanedione, the total
amount of 3-DIOL converted from substrate by 3-HSD oxidase
was measured as the sum of DHT and androstanedione formed. These
studies validated our conclusion that Leydig cell RoDH2 is a 3-HSD
oxidase.
The timing of DHT production is a critical factor in the development of
reproductive organs. In humans with functional androgen receptors, the
deficiencies of type I and II 5-reductases (2, 44) cause androgen
insufficiency syndromes. Treatment of such individuals with DHT allows
them to proceed through puberty. In rats, the role of DHT in the testis
has not been established, but one possibility is maintenance of
spermatogenesis when testicular T concentrations are low (45, 46). The
relationship between the expression level of RoDH2 and concentration of
DHT in Leydig cells is not straightforward because DHT can be made from
T by 5-reductase, and DHT can be interconverted with 3-DIOL by
the actions of 3-HSD1 and RoDH2. In the testis of prepubertal rats,
RoDH2 activity is not measurable (17), but 5-reductase is present
(47), and in immature rats, 5-reductase and RoDH2 are both present.
In adult rat testes, the levels of 5-reductase become barely
detectable (47), and since 3-HSD1 does not have oxidase activity in
intact cells, RoDH2 would be the only source of DHT in the adult
testes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received November 29, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reductase 2 deficiency. Endocr Rev 14:577–593-hydroxysteroid
dehydrogenase. Steroids 62:101–111[CrossRef][Medline]
-hydroxysteroid
dehydrogenase of rat liver cytosol and its inhibition by
anti-inflammatory drugs. Biochem J 222:601–611[Medline]
-hydroxysteroid dehydrogenase. A member of the oxidoreductase gene
family. J Biol Chem 266:15253–15257-hydroxysteroid/dihydrodiol dehydrogenase. J Biol Chem 266:8820–8825-hydroxysteroid dehydrogenase. Mol
Endocrinol 5:823–828-hydroxysteroid dehydrogenase
of the rat liver active against C19-, C20-, C23-, C24-, C25- and
C26-steroids. Hoppe Seylers Z Physiol Chem 362:511–520[Medline]
-hydroxysteroid dehydrogenase oxidative and
reductive activities in rat Leydig cells during pubertal development.
Biol Reprod 60:855–860-hydroxysterol short-chain
dehydrogenase. J Biol Chem 272:33125–33131-hydroxysteroid (oxidative) dehydrogenase activity. Program of the
10th International Congress on Hormonal Steroids, Québec City,
Canada, 1998 (Abstract 192A)
-hydroxysteroid
dehydrogenases from rat and human prostate. J Biol Chem 272:15959–15966-hydroxysteroid dehydrogenase
that differs from 20-hydroxysteroid dehydrogenase by seven amino
acids. Biochem Biophys Res Commun 228:474–479[CrossRef][Medline]
-hydroxysteroid
dehydrogenase. Steroids 62:101–111
-hydroxysteroid
dehydrogenases [published erratum appears in Steroids 1997
May;62(5):455–456]. Steroids 61:508–523[CrossRef][Medline]
-hydroxysteroids. J Biol Chem 273:19778–19785-reductase type 1 caused by estrogen
excess. Mol Endocrinol 11:917–927-reductase activity impairs
the testosterone-dependent restoration of spermiogenesis in adult rats.
Endocrinology 137:2703–2710[Abstract]
-reductase
inhibitor. J Androl 20:109–117-reductase messenger ribonucleic acid levels and immunocytochemical
localization of the type 1 protein in the rat testis during postnatal
development. Endocrinology 136:5409–5415[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  T. L. Rizner, H. K. Lin, D. M. Peehl, S. Steckelbroeck, D. R. Bauman, and T. M. Penning Human Type 3 3{alpha}-Hydroxysteroid Dehydrogenase (Aldo-Keto Reductase 1C2) and Androgen Metabolism in Prostate Cells Endocrinology, July 1, 2003; 144(7): 2922 - 2932. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Ivell, M. Balvers, R. J. K. Anand, H.-J. Paust, C. McKinnell, and R. Sharpe Differentiation-Dependent Expression of 17{beta}-Hydroxysteroid Dehydrogenase, Type 10, in the Rodent Testis: Effect of Aging in Leydig Cells Endocrinology, July 1, 2003; 144(7): 3130 - 3137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. D. Griffin and S. H. Mellon Biosynthesis of the Neurosteroid 3{alpha}-Hydroxy-4-pregnen-20-one (3{alpha}HP), a Specific Inhibitor of FSH Release Endocrinology, November 1, 2001; 142(11): 4617 - 4622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. Theobald, B. L. Roman, T.-M. Lin, S. Ohtani, S.-W. Chen, and R. E. Peterson 2,3,7,8-Tetrachlorodibenzo-p-dioxin Inhibits Luminal Cell Differentiation and Androgen Responsiveness of the Ventral Prostate without Inhibiting Prostatic 5{alpha}-Dihydrotestosterone Formation or Testicular Androgen Production in Rat Offspring Toxicol. Sci., December 1, 2000; 58(2): 324 - 338. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
