| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Division of Hormone Research, Departments of Cell Biology, Pharmacology & Neuroscience, Georgetown University Medical Center, Washington, DC 20007
Address all correspondence and requests for reprints to: Vassilios Papadopoulos, Georgetown University Medical Center, Department of Cell Biology, 3900 Reservoir Road, Northwest, Washington, D.C. 20007. E-mail: papadopv{at}gunet.georgetown.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, ß/
, and
(3).
In addition to their effect on the liver, peroxisome proliferators also
induce pathological changes in a number of target organs, including the
testis. In 1983, Olson et al. reported that the testis is a
target organ for the 10-carbon perfluorinated carboxylic acid PFDA
(perfluorodecanoic acid; nonadecafluoro-n-decanoic acid;
CF3(CF2)3COOH)
toxicity in the rat, hamster and guinea pig (11, 12). In the rat
testis, early degenerative changes first seen 8 days after PFDA
treatment progressed to tubular atrophy, calcification, and necrosis by
16 days (12). In 1988, Reddy and co-workers reported that treatment of
rats with the peroxisome proliferator ciprofibrate resulted in
increased levels of peroxisomal ß-oxidation and catalase genes in the
Leydig cells of the testis (13). No changes in the peroxisomal volume
or activities were shown in the seminiferous tubule compartment of the
testis. More recently, Peterson and colleagues reported that PFDA
treatment of rats markedly decreased plasma testosterone and DHT levels
with an ED50 of 30 mg/kg (14). Associated with
the decrease in plasma testosterone levels were decreased accessory
organ weights due to atrophic changes characteristic of decreased
plasma androgen concentrations. PFDA had no effect on accessory organs
when the androgenic deficiency was prevented by castrating and
implanting PFDA-treated rats with testosterone. In these studies, rats
were dosed with PFDA and then killed 7 days later. Moreover, PFDA
treatment did not affect plasma clearance of testosterone and plasma LH
levels, suggesting a direct effect on testicular androgen production.
Decapsulated testes from PFDA-treated rats incubated with hCG secreted
far less testosterone into the medium that did the control testis (14).
Thus, these data present direct evidence that the peroxisome
proliferator PFDA targets specifically the interstitial Leydig cell of
the testis where it may inhibit the hormone-stimulated steroidogenesis.
In addition, Wahli and co-workers recently reported that the Leydig
cells and the ovarian granulosa and theca cells express the PPAR and
PPARß proteins (15). PPAR and PPARß were also found in Sertoli
cells but not in the germ cell line. These findings collectively
indicated that the peroxisome proliferator PFDA, probably acting via
PPAR inhibits the hormone-stimulated Leydig cell steroidogenesis. In
this manuscript, we examined the effect of PFDA on Leydig cell
function. We identified the outer mitochondrial membrane
peripheral-type benzodiazepine receptor (PBR) protein to be the target
of PFDA. PFDA-induced decrease in PBR levels leads to decreased
cholesterol transport and steroid synthesis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-32P]dCTP (Sp. Act., 3000
Ci/mmol), [32P]NAD (Sp. Act., 800 Ci/mmol) and
[1,2,6,7-N-3H]progesterone (Sp. Act., 94.1
Ci/mmol) were obtained from NEN Life Science Products
(Wilmington, DE). PK 11195 and aminoglutethimide were obtained from
Research Biochemicals International, Inc. (Natick, MA).
Nitrocellulose (0.45 µm) was from Hoefer Scientific (San
Francisco, CA). Trilostane
(4,5-epoxy-17-hydroxy-3-keto-5-androstan-2-carbonitrile)
was donated by Sterling-Winthrop (Surrey, UK).
22R-hydroxycholesterol, dibutyryl cyclic AMP and PFDA were
purchased from Sigma-Aldrich Corp. (St. Louis, MO). Cell
culture supplies were purchased from Life Technologies, Inc.(Grand Island, NY) and cell culture plasticware was from
Corning, Inc. (Corning, NY). Electrophoresis reagents and
materials were supplied from Bio-Rad Laboratories, Inc.
(Richmond, CA). All other chemicals used were of analytical grade and
were obtained from various commercial sources.
Cell culture and treatments
MA-10 mouse Leydig tumor cells were a gift from Dr. Mario Ascoli
(University of Iowa) and were plated at low density for 24 h in
modified Waymouth’s MB 752/1 medium containing 20 mM
HEPES, 1.2 g/liter NaHCO3, and 15% horse serum,
pH 7.4 (16). After 24 h the media were changed and fresh media
containing the indicated amounts of PFDA or its solvent propylene
glycol (PEG) were added for 24, 48, or 72 h. At the end of the
incubation the cells were washed with serum-free medium and stimulated
for 2 h with saturating amounts of hCG (1 nM) in
serum-free medium. This concentration of hCG was previously shown (16)
to maximally stimulate steroid synthesis by these cells. At the end of
the 2-h incubation period the cell media were saved for progesterone
determination, and the cells were either dissolved in 0.1 N
NaOH for protein determination or processed for protein (Western) or
messenger RNA (mRNA) (Northern) blot analysis or other biochemical
assays.
Testicular interstitial cells were prepared by collagenase dissociation (17) of testes obtained from adult Sprague Dawley (300 g) rats. This preparation contained 20–30% 3ß-hydroxysteroid deshydrogenase positive cells (Leydig cells). Leydig cells were further purified using discontinuous Percoll gradient centrifugation as previously described (17). The preparations obtained contained 75–85% Leydig cells as shown by the histochemical staining for 3ß-hydroxysteroid deshydrogenase (17). Isolated rat Leydig cells were cultured for 48 h with various concentrations of PFDA under the same conditions as described above for the MA-10 cells. At the end of the incubation the cells were washed with serum-free medium and stimulated for 2 h with saturating amounts of hCG (1 nM) in serum-free medium. Media were collected for testosterone measurements.
Measurement of cholesterol transport in mitochondria
For these studies, MA-10 cells cultured in 150-mm culture dishes
were treated as described above for 48 h with or without the
indicated concentrations of PFDA. At the end of the incubation cells
were washed with serum-free medium and treated with 1 nM
hCG in the presence of 0.76 mM aminoglutethimide, a
specific inhibitor of the P450scc. At the indicated time points, media
were removed, the cells were resuspended in PBS and mitochondria were
isolated in buffers containing aminoglutethimide as described
previously (18). The rate of pregnenolone formation, reflecting the
amount of cholesterol available to the P450scc, was determined as
previously described in aminoglutethimide-free buffer (18). To examine
the cytochrome P450 side chain cleavage activity (P450scc),
mitochondria from control and PFDA-treated cells were incubated with
22R-hydroxycholesterol in the presence of 1 µM
trilostane, an inhibitor of 3ß-hydroxysteroid dehydrogenase enzyme
(16, 19). Pregnenolone formation was measured as an index of the
P450scc activity.
RIAs
Progesterone, testosterone and pregnenolone were measured by
means of RIA (16, 18) using antibodies specific for each steroid
(ICN Pharmaceuticals, Inc., Costa Mesa, CA). The data were
analyzed using the MultiCalc software from EGG-Wallac, Inc. (Gaithersburg, MD).
Electron microscopy and morphological analysis
MA-10 cells were plated at low density and after 48 h, the
media were replaced with fresh medium alone (control cells) or
containing the indicated amounts of PFDA (10-5
to 10-7 M) and further incubated for
24, 48, or 72 h. At the end of incubation, both cell groups
(control and PFDA-treated) were washed with serum-free medium and then
stimulated for 2 h with 1 nM hCG in serum-free medium.
Cells were processed for electron microscopic morphometry as we
previously described (20). Briefly, cells were fixed in 2%
glutaraldehyde and 1% paraformaldehyde solution prepared in 3
mM phosphate buffer (pH 7.4) at 4 C for 20 min, washed with
buffer A (3 mM phosphate buffer, pH 7.4) for up to 2 days,
and postfixed for 20 min in 1% OsO4 solution mixed with 0.2
M s-collidine/HCl buffer (pH 7.4). The samples were then
washed with buffer A, stained in 2% uranyl acetate solution in water
for 15 min, washed, dehydrated in upgraded (20–100%) ethanol
solutions, embedded in Epon:alcohol mixture 1:1 for 3 h and then
in 100% Epon for an additional 3 h, and then overnight in the
oven. Thin sections (90–95 nm) of blocks were treated with 1%
solution of lead citrate, placed on copper grids, and morphometrically
analyzed for the surface density of cytoplasmic lipid inclusions. All
observations were made with a JEOL-2000 transmission electron
microscope operated at 50–60 kV. Up to seventy sectioned cells, with a
cell:nucleus ratio of less than 2:1, from each control and experimental
protocol were subjected to morphometric analysis to determine the
surface density of cytoplasmic lipids. The analysis of sectioned
cellular profiles were made on images magnified X20,000–30,000 with
the intersection point count method. Vertical and horizontal separation
of the adjacent points was 0.3 µm. The proportion of cellular surface
covered with the cytoplasmic lipid inclusions were first established by
a simple comparison of numbers of intercept points of test lines
coinciding with the inclusions and with the rest of the cell. The
surface density value of cytoplasmic lipid pools (Svl) was subsequently
calculated as: Svl = (2 x Il)/Lt, where Il is the number of
intercepts of test lines with lipid pools, and Lt = 1/2 x
Pt x z, where Pt is the number of end points of test lines z
(21). This two-dimensional evaluation of the intracellular free lipid
content, rather than the three-dimensional assessment of the volumetric
presentation of the substrate, was selected by necessity because of the
unknown absolute volume of this cellular population.
Analysis of cell viability and mitochondrial integrity
Cell viability was analyzed using the trypan blue exclusion
method. At the end of the incubation, the cells were washed three times
with PBS and incubated for 15 min with 0.1% trypan blue stain. After
washing, stained cells (dead cells) were counted by light microscopy.
Cell viability and mitochondrial integrity were also evaluated
following the above described cell treatment protocol, by measuring the
levels of cellular 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl
tetrazolium bromide (MTT) reduction (22) using the TACS MTT
proliferation assay (Trevigen, Inc., Gaithersburg, MD).
Total protein labeling
106 MA-10 cells were treated without and
with PFDA for 48 h. Cells were then translabeled with
35S-met/cys (0.6 mCi/well) in met/cys free medium
for 4 h. Total protein labeling was examined in aliquots
precipitated by 10% trichloroacetic acid. The precipitates were heated
at 90 C for 20 min, cooled, filtered on glass fiber filters, rinsed
with 5% trichloroacetic acid, and washed with ethanol. Filters were
air-dried and radioactivity was measured by liquid scintillation
spectrometry.
Analysis of DNA damage
The possibility that PFDA may cause DNA damage to the cells was
examined using the poly(ADP-ribose) polymerase (PARP) assay (23,24;
Trevigen, Inc., Gaithersburg, MD). Cells were collected and resuspended
in 50 mM Tris [pH 8.0] buffer supplemented with 25
mM MgCl2 and 0.1 mM PMSF.
Cells were sonicated, centrifuged, and the supernantants were used to
determine the PARP activity by measuring the incorporation of
radiolabeled poly(ADP-ribose) into their control PARP or cellular PARP.
The reaction was stopped using 20% ice-cold TCA and the precipitates
were collected on GF/C filters. Filters were air-dried and
radioactivity was measured by liquid scintillation spectrometry.
Radioligand binding assays
Radioligand binding assays were performed as previously
described (16). In brief, cells were scraped from culture flasks into
PBS, dispersed by trituration, and centrifuged at 1200 x
g for 5 min. The cells were resuspended in PBS to a final
concentration of 5 to 10 µg protein/100 µl. Saturation binding
studies were performed on 10 µg protein for the MA-10 cells in a
final volume of 300 µl in the presence of the radioligand at the
indicated concentrations. Nonspecific binding was determined in the
presence of 6 µM of the homologous
nonradioactive ligand. After 2 h incubation at 4 C, the assays
were stopped by filtration through GF/B filters (Brandel, Gaithersburg,
MD) equilibrated in 0.1% polyethyleneimine and washed with 40 ml
ice-cold PBS. Radioactivity trapped on the filters was determined by
liquid scintillation counting. The dissociation constant
(Kd) and the binding capacity were defined
following Scatchard analysis of the specific binding using the LIGAND
program (25). In some experiments the direct effect of PFDA on PK 11195
ligand binding to isolated MA-10 Leydig cell mitochondria was
examined.
SDS-PAGE and immunoblot analysis
Mitochondria were isolated as we previously described (26).
Mitochondrial proteins were fractionated by one dimensional SDS-PAGE
and electro-transferred onto nitrocellulose (27). The nitrocellulose
was subjected to immunoblot analysis using anti-PBR (1:1000) or
anti-StAR (1:1000) antisera and goat antirabbit IgG alkaline
phosphatase-conjugated secondary antibody followed by the substrate
nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate for
detection (27). Densitometric analysis of the immunoreactive protein
bands was performed using the Sigmagel software (Jandel Scientific, San
Rafael, CA). The production and characterization of the anti-PBR and
anti-StAR antisera, against internal peptide sequences of the proteins
have been previously described (27).
RNA (Northern) blot analysis
Total cellular RNA was isolated by the acid guanidinium
thiocyanate-phenol-chloroform extraction method (28) using the RNAzol B
reagent (Tel-Test Inc., Friendswood, TX). RNA
electrophoresis, transfer, probe labeling, and membrane hybridization
were performed as previously described (18). The blots were hybridized
against the [32P] complementary DNA (cDNA)
probe for PBR or cyclophilin (18). Screen enhanced autoradiography was
performed by exposing Kodak X-OMAT AR films to the blots
at -80 C for 48 h. Densitometric analysis of the immunoreactive
protein bands was performed using the Sigmagel software (Jandel
Scientific).
Transient transfections
The MA-10 cells were transiently transfected using the
electroporation method. Briefly, each Genepulser cuvette (0.4 cm-gap,
Bio-Rad Laboratories, Inc.) contained 8 x
106 cells in 350 µl of antibiotic-free complete
Waymouth’s growth medium, plus 50 µl of 0.1 x TE containing
the plasmid preparations. The cells were cotrasfected with 30 µg of
the pPBR (-1219/+36) CAT construct, and 5 µg per
transfection of the pSV promoter-ß-galactosidase plasmid
(Stratagene, La Jolla, CA) to check the transfection
efficiency. The 1219 bp promoter region fragment contained the first 36
bp of exon 1 of rPBR subcloned into the polylinker of the pCAT-Basic
(Promega Corp., Madison, WI) and was provided by Dr.
Krueger (29, Georgetown University). As control, MA-10 cells were also
transiently co-transfected with 30 µg of the pCAT-basic plasmid
(Promega Corp.) and 5 µg per transfection of the pSV
promoter-ß-galactosidase plasmid (Stratagene). Cells in
electroporation cuvettes were electro-shocked using 330V and a
capacitance of 950 µFd generated from Genepulser II (Bio-Rad Laboratories, Inc.). The cells were kept at room temperature for
10 min before plating onto six-well dishes, and further allowed to grow
in complete Waymouth’s medium for 16 h. Then, the media were
changed into serum-free Waymouth’s and cells were treated with or
without 30 µM of PFDA. Twenty-four hours later, the cells
were harvested for CAT-ELISA (5-Prime
3-Prime Inc., Boulder, CO) and
ß-galactosidase (Promega Corp.) activity
measurements.
Measurement of PBR mRNA stability
To examine the effects of PFDA on PBR mRNA stability we treated
MA-10 cells for 48 h with the indicated concentrations of PFDA and
then measured the decay of PBR mRNA by incubating the cells with
actinomycin D (10 µg/ml) for an additional 15, 30, 60, and 180 min.
Longer exposures to actimomycin D were toxic to the cells. Total
cellular RNA was isolated as described above and PBR mRNA levels were
measured by RNA (Northern) blot analysis.
Protein measurement
Protein was measured by the method of Bradford (30), using BSA
as a standard.
Statistics
Statistical analysis was performed by one-way ANOVA and unpaired
Student’s t test using the INSTAT 3.00 package from
GraphPad Software, Inc. (San Diego, CA). In the case of
morphological analyses the Fisher’s PLSD test was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). This effect was
statistically significant (ANOVA, P < 0.001) at all
treatment periods. This inhibitory effect of PFDA is shown to be
significant after 24 h of treatment with PFDA at concentrations
higher than 3 x 10-5
M and with an ED50 of
10-5 M. Treatment of the
cells for longer periods of time reduced the ED50
to 3 x 10-6 M for
the 48 h treatment and 10-6
M for the 72 h treatment. PEG had no effect
on the hCG-stimulated progesterone production.
|
The results presented above on the MA-10 mouse tumor cell line were
confirmed using isolated adult rat Leydig cells (Fig. 1B
). 48 h
treatment with PFDA resulted in the inhibition of the subsequent
hCG-stimulated testosterone production with and
ED50 of 3 x 10-6
M, similar to that seen in MA-10 cells. These results
validate the use of the MA-10 cell model for identifying the mechanism
of PFDA inhibitory action on Leydig cell steroidogenesis.
PFDA does not affect Leydig cell viability, mitochondrial
integrity, total protein synthesis and DNA
Concentrations of PFDA up to 10-4
M do not affect cell viability as attested by the trypan
bleu exclusion test. However at 3 x 10-4
M 30–40% of the cells had taken up the trypan blue dye,
indicating that at this concentration the cells were dying. These
results were validated using the MTT assay. In Fig. 2, the effect of 48 h treatment with
increasing concentrations of PFDA on cell cultures where different
numbers of cells were plated is shown. It is clear again that PFDA up
to 10-4 M does not affect cell
viability and more specifically mitochondrial integrity because the MTT
assay is based on the MTT reduction by the mitochondrial diaphorase
enzyme, an index of mitochondrial integrity (22).
|
) but it did not
affected total protein synthesis, used as additional index of toxicity
and determined by incubating the cells with
35S-methionine/cysteine (Fig. 3). Moreover, PFDA did not induce DNA
damage of the cells (Fig. 4), as
determined by measuring the activity of the poly(ADP-ribose)
polymerase, the enzyme involved in DNA repair after its damage.
|
|
.3) and
PFDA-treated cells (data not shown). In Fig. 5.3, a cross-section of a
MA-10 cell illustrates the required cellular profile for the
morphometric analysis. In addition to the usual organellar profiles
(nucleus N, mitochondria M, Golgi zone G, reticulum R), this cell
contains several lipid/phospholipid inclusions (indicated by an
asterisk). In this section, all inclusions appear to contain
phospholipids organized in micellar, labile mode that, unless
immediately metabolized, will assume a stable bilayer configuration. We
noticed that treatment with PFDA induced changes in the configuration
of the lipids within the lipid droplets. For statistical analysis of
differences in the cytoplasmic lipid/phospholipid content the
inclusions were characterized either as intact (entirely micellar
configuration; Fig. 5.1) or transitional (either combination of
micellar and bilayered configuration, or entirely in bilayer aligned
phospholipids; Fig. 5.2). Total surface density of phospholipid
inclusions was also ascertained for each cellular section. Thirty-five
to 70 representative sections, made approximately at the midnuclear
level (Fig. 5.3), for each cellular category were analyzed for the
lipid/phospholipid surface density within the cytoplasm and the average
values and the standard errors were derived as shown in Table 1. Corresponding values were analyzed
between the control and experimental cellular categories for the
statistical significance of difference in the cytoplasmic
lipid/phospholipid surface density using Fisher’s PLSD test at 5%
significance level. The results of this investigation show that
PFDA-treated cells have higher surface density value of transitional
and total lipid/phospholipid inclusions but lower of micellar type
inclusions.
|
|
Table 2 shows that 48 h treatment
with 10 µM PFDA inhibits the cAMP-stimulated progesterone
production suggesting that the effect of PFDA was localized beyond the
cAMP production. We then examined whether PFDA affected the first step
in steroid biosynthesis, the transformation of cholesterol into
pregnenolone by the P450scc enzyme located within the inner
mitochondria membrane. MA-10 cells were treated for 48 h with 10
µM PFDA and then incubated in the presence of
22R-hydroxycholesterol (10 µM), the
hydrosoluble form of cholesterol, which enters readily the inner
mitochondria membrane. Progesterone production was identical in both
control and PFDA-treated cells, suggesting that the site of action of
PFDA is not at the P450scc or other steroidogenic enzyme level but
rather at an earlier step of the pathway (Table 2). In agreement with
these results, mitochondria isolated from control and PFDA-treated
MA-10 cells and incubated with 22R-hydroxycholesterol also
produced similar amounts of pregnenolone (data not shown). Thus, it
seems like the inhibitory effect of PFDA on steroidogenesis may be due
to an impairment of the multistep process of cholesterol mobilization,
transport into mitochondria and loading onto P450scc in response to hCG
stimulation. To determine whether the hCG-induced translocation of
cholesterol to the inner mitochondria membrane was affected by PFDA
treatment, we treated cells with aminoglutethimide before and during
the 2 h hCG stimulation. In this experiment, aminoglutethimide was
used to inhibit the transformation of cholesterol to pregnenolone by
P450scc, thus we could determine the amount of endogenous cholesterol
accumulated within the inner mitochondrial membrane upon hCG
stimulation. Fig. 6 shows that upon
washing out of the aminoglutethimide from the isolated mitochondria,
pregnenolone production by mitochondria isolated from MA-10 cells
treated with PFDA was significantly inhibited compared with control
cells (P < 0.001 by ANOVA). This data indicates that
PFDA inhibits the transfer or loading of cholesterol to the inner
mitochondrial membrane P450scc. We then examined the effect of PFDA
treatment on the two known proteins shown to be involved in the
cholesterol translocation to the inner mitochondria membrane, PBR (31)
and the steroidogenic acute hormone regulator protein (StAR; 32).
|
|
). No effect on the receptor affinity
was seen. Interestingly, as with progesterone synthesis, 48 h
treatment with 10 µM PFDA, followed by washing
and further incubation of the cells in PFDA-free medium for 48 h,
PBR levels returned to control levels (data not shown). It should be
noted that PFDA, tested at concentrations up to 100
µM, did not displace radiolabeled PK 11195 in
direct ligand binding studies performed using isolated MA-10
mitochondria.
|
).
Densitometric analysis of the immunoreactivity suggested a 60%
decrease of the 18-kDa PBR expression in PFDA-treated cells (Fig. 8).
In addition, PFDA treatment reduced PBR mRNA levels by 50% (Fig. 9).
|
|
).
PFDA does not affect PBR mRNA transcription but accelerates its
decay
The results presented above suggest that PFDA exerts a direct
effect on PBR mRNA stability or DNA transcription. Preliminary nuclear
run-on experiments failed to provide conclusive evidence on the effect
of PFDA on PBR mRNA transcription. Thus, we decided to extend these
studies by examining the effects of PFDA treatment on PBR gene
transcription. It has been shown previously that the -1219/+36 DNA
region upstream from the transcription start site of the PBR gene (29),
contains the regulatory elements which are necessary and sufficient for
the expression of PBR in MA-10 mouse Leydig tumor cells. The eventual
role of the promoter elements present within this region implicated in
the down-regulation of the PBR gene transcription by PFDA was examined,
with transient transfection experiments. The -1219/+36 fragment within
the PBR promoter region subcloned into a CAT-reporter plasmid, was
used. Promotorless vector was used as internal control. MA-10 mouse
Leydig tumor cells transfected with the pPBR (-1219/+36)-CAT
construct, were treated with and without 30 µM PFDA. PFDA
was dissolved in 50% PEG. Cells treated with the corresponding amounts
of PEG were used also as controls demonstrating that PEG did not have
any effect either in the cells function or PBR expression. When the
transfected cells were treated with PFDA, no significant changes in the
expression of CAT activity were observed (Fig. 10).
|
suggest that
PFDA treatment accelerates the PBR mRNA decay. However, PFDA treatment
did not affect the mRNA levels of the housekeeping gene cyclophilin
(Fig. 11).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has been shown that the testis is one of the target organs affected by peroxisome proliferators. The peroxisome proliferator perfluorinated carboxylic acid PFDA was reported to induce early degenerative changes that progressed to tubular atrophy (11, 12). The same peroxisome proliferator was also reported to suppress plasma testosterone levels, without affecting plasma LH concentrations, thus inducing a decrease in accessory organ weights (13). Because implants of testosterone reversed the effect of PFDA and testes from PFDA-treated rats were far less responsive to LH than control testes, it can be concluded that this peroxisome proliferator exerts a direct effect on the hormone-stimulated testosterone biosynthesis. The peroxisome proliferators phthalates were also shown to reduce in vivo and in vitro testosterone levels and induce testicular atrophy (34, 35). Moreover, it has been shown that, in testis, only the Leydig cell is a target for a fibrate hypolipidemic drug, also member of the peroxisome proliferator family of chemicals (13). In addition, it was recently reported that peroxisome proliferators inhibit progesterone synthesis by a human choriocarcinoma cell line in culture (36). Thus, these data suggest that peroxisome proliferators target specifically steroidogenic cells, such as the Leydig cell, where they may control the process of steroid synthesis and thus gonadal function.
Indeed, the data presented herein clearly demonstrates that PFDA inhibits in a dose- and time-dependent manner the hCG-stimulated progesterone production by the MA-10 mouse tumor Leydig cells. PFDA inhibited with the same potency the hCG-stimulated testosterone production by isolated rat Leydig cells, indicating that the results seen with the MA-10 cells were not specific to a tumor Leydig cell line. PFDA also inhibited the cAMP-stimulated progesterone synthesis by MA-10 cells, but did not affect the 22R-hydroxycholesterol-supported steroid formation by these cells. 22R-hydroxycholesterol is an hydrosoluble form of cholesterol, which enters readily the inner mitochondria membrane, and an intermediate of the P450scc reaction (37). This result suggests that PFDA did not act on the P450scc or other steroidogenic enzymes involved in the biosynthesis of the final steroid product. Moreover, these data, together with the reversibility data, suggest that treatment with PFDA for times up to 48 h does not exert a toxic effect on Leydig cells. These results were corroborated by a series of assays performed to assess the potential toxic effect of PFDA on mitochondrial integrity, total protein synthesis, DNA, and cell viability. All tests performed clearly showed that PFDA, under the conditions, concentrations and treatment time periods used, does not exert any toxic effect on the cells. Moreover, a detailed morphologic analysis of the PFDA-treated cells failed to demonstrate any morphological alterations of the cells, compared with control, with one exception, the state of the lipid/phospholipid configuration in the lipid droplets of the cells. In conclusion, the peroxisome proliferator PFDA exerts a specific antisteroidogenic effect on Leydig cells. If this effect of peroxisome proliferators is conserved among species, concern is raised about the risk in human populations exposed for long term to peroxisome proliferators. Thus, characterization of the molecular determinants involved in the response of Leydig cells to peroxisome proliferators such as PFDA will be useful in developing a predictive model of risk to humans.
Because PFDA did not directly affect P450scc enzyme activity, it was clear that the steroid biosynthetic pathway was not affected by the treatment but it was rather the substrate (cholesterol) availability, which it was limiting. The rate-determining step in steroid biosynthesis, however, is not the rate of metabolism of cholesterol to pregnenolone by the P450scc but rather the transport of the precursor cholesterol from intracellular stores to the inner mitochondrial membrane where the P450scc is located (38, 39). Morphologic analysis of the cells treated with and without PFDA indicated that the treatment induced a build-up of cytoplasmic lipid/phospholipid pools that could be due to either reduced utilization and/or release from the cell of this substrate. PFDA-treated cells appeared to have a faster rate of transition of lipids/phospholipids from the micellar to bilayered configuration than the corresponding control cells. Consequently, the treated cells contained significantly greater amount of bilayer-configured lipids/phospholipids in the cytoplasm than control cells. Because hCG has been shown to reduce the number of lipid droplets in Leydig cells in vivo (40) in a time-frame close to the hCG-induced steroid production, it has been suggested that cholesterol esters form, at least part, of the content of these lipid droplets. De-esterified cholesterol from these lipid droplets is part of the substrate used for testosterone formation by Leydig cells. The fact that the transitional configuration of substrate, the only configuration released from the cell (20), was found in the greater quantity in the PFDA-treated cells suggests that most of the cytoplasmic substrate was not transported to the mitochondria and instead accumulated in the droplets. These results further suggest that PFDA’s site of action is at the level of cholesterol transport into the mitochondria.
Biochemical studies using the inhibitor of P450scc aminoglutethimide, which allows for the accumulation of transported cholesterol into the inner mitochondrial membrane by blocking its further metabolism, followed by release of the inhibition and measurement of the synthesized pregnenolone, demonstrated that indeed PFDA inhibited the transfer of cholesterol to the P450scc. Because PFDA did not affect the Leydig cell mitochondrial integrity, we then examine its effect on the expression of molecular entities involved in cholesterol transport in steroidogenic cells. As we noted earlier, two such regulatory elements of the steroidogenic machinery have been identified: PBR (31) and StAR (32).
PBR, although present in all tissues examined, was found to be particularly high in steroid producing tissues, where it was primarily localized in the outer mitochondrial membrane (31). Numerous in vitro and in vivo studies demonstrated that PBR is a functional component of the steroidogenic machinery (16, 31, 41, 42, 43) mediating cholesterol delivery from the outer to the inner mitochondrial membrane (26). An 18-kDa isoquinoline-binding protein was identified as PBR, cloned and expressed (31, 41). Strong evidence was presented suggesting that PBR is a multimeric complex (41) located at the junctions between outer and inner mitochondrial membranes (contact sites) where the PBR complex could function as a pore, thus allowing the translocation of cholesterol from the outer to the inner mitochondrial membrane (44). Further support for this hypothesis was provided by first, molecular modeling studies of the human and mouse 18-kDa PBR protein where it was clearly shown that PBR is a cholesterol channel (44, 45) and second, in vitro reconstitution studies in bacteria where the expression of the 18-kDa PBR protein induced cholesterol transport into these cells (46). We demonstrate herein that a decrease in PBR expression after treatment with PFDA results in reduced hCG-stimulated steroid formation by Leydig cells. The cause-effect relation between the effect of PFDA on PBR expression and steroid synthesis was further supported by the observation that both PBR levels and steroid synthesis returned to normal upon the end of the treatment.
Thus, we focused our studies on examining the effect of PFDA on PBR. We did not see any direct effect of PFDA on PBR ligand binding studies, which excludes the possibility that this compound may act as a PBR ligand. We then demonstrated that PFDA reduced Leydig cell PBR ligand binding, protein, and mRNA levels, resulting in reduced ability of the cells to synthesize progesterone in response to hCG. Although it is difficult to quantify accurately PBR protein and mRNA levels, the comparison of the PFDA-induced reduction of steroid formation with PBR ligand binding, 18-kDa PBR protein, and mRNA levels, as determined by image analysis, shows a close correlation between the message, protein, activity, and function. This observation further supports the cause-effect relation between PBR levels and the ability of Leydig cells to form steroids in response to hCG.
It has been shown that the 30-kDa StAR is expressed in response to trophic hormones, in parallel with steroid production, and it has been suggested to be the key mediator of the hormone-stimulated steroidogenesis (32, 47). No changes in hCG-induced StAR levels were observed upon treatment of the cells with PFDA. Considering that, in the in vivo treatment studies with PFDA no changes in circulating LH levels were reported (14), these data suggest that the antiandrogenic effect of PFDA was not due to an effect on the hormone-induced 30-kDa StAR levels.
The regulation of PBR levels by a number of factors, including hormones and stress, has been reported (31, 48, 49) and more recently both positive and negative regulatory transcriptional elements have been identified in the sequence upstream the first exon of the rat PBR gene (29). Thus, we can speculate that PFDA may control the expression, affect the PBR mRNA stability, or regulate the activity of one or more of these transcriptional elements.
Peroxisome proliferator-responsive genes were found to contain
PPREs consisting of a near-perfect direct repeat (DR) of two copies of
the sequence TGA/TCCT separated by one bp (DR-1). However, only half of
the nucleotides within each repeat sequence are strictly conserved,
whereas others vary considerably (3, 8, 50). A fragment spanning from
36 bp into the first exon of PBR with 1,219 bp of upstream genomic
sequence has been isolated (29). This upstream sequence was analyzed
for transcriptional promoter activity and revealed the presence of
three regulatory elements. A closer look at the negative acting
regulatory element indicated the presence of a 14-bp sequence with high
homology to the previously reported PP response elements (PPREs),
suggesting that the molecular determinant required for PP action on
gene expression is present in the promotor sequence of the PBR gene.
Moreover, recent publications identified peroxisome proliferator
activated receptor immunoreactive proteins (PPAR and PPARß) in the
Leydig cells of the testis and the granulosa and theca cells of the
ovary (15, 51). In addition, it is known that Leydig cells respond to
retinoids, suggesting that they may express the retinoic X receptor
(RXR). The dimerization partner RXR (3) potentiates gene regulation by
PPAR. Taken together these results suggest that indeed the
antiandrogenic effect of the peroxisome proliferator PFDA may be due to
a direct effect on PBR mRNA expression.
In vitro transfection studies of the PBR promotor into Leydig cells followed by PFDA treatment of the cells indicated that PFDA did not affect the PBR gene expression directly. However, we cannot exclude the possibility that PFDA may have a small effect on PBR expression, which is not obvious, because it overlaps with the effect of other positively regulated elements in the PBR promotor. Further studies on PBR mRNA stability indicated that PFDA might act by accelerating the PBR mRNA decay.
Considering the potential health risk issue associated with PFDA toxicity the question arises whether the levels required to see the effect of PFDA on PBR mRNA levels and steroidogenesis in vitro are likely to be achieved in vivo. To answer this question one should consider that 1) in vivo dosing of the rats with PFDA for 7 days decreased plasma T with an ED50 of 30 mg/kg (13); and 2) the in vitro 48 h treatment with PFDA inhibited the hormone-stimulated steroidogenesis with ED50 of 10 µM (5.14 mg/liter). Considering the blood volume of the adult rat, and without taking into account the PFDA tissue distribution, pharmacokinetics and excretion, we can estimate that the in vitro system is at least 6-fold more sensitive than the in vivo model and provides a sensitive system to test for the effects of PFDA. In addition, the facts that 1) organic fluorine, derived from perfluorocarboxylic acids, persists in the blood of industrial workers exposed to these substances (13), 2) in industrial areas the density of peroxisome proliferators is high (52); and 3) the levels of these compounds in food samples reach the levels of 20 mg/kg (53) indicate that PFDA, as all peroxisome proliferators, is a putative health hazard.
In conclusion, PFDA is shown to inhibit the expression of the PBR protein, which controls the amount of cholesterol transported to the P450scc and consequently the amount of steroids formed by Leydig cells. Ultimately, these changes will affect spermatogenesis (14) and fertility. Industrial and pharmaceutical peroxisome proliferator compounds, some of which are extensively used in humans, may exert similar antiandrogenic effects by inhibiting PBR expression, cholesterol transport, steroid biosynthesis and hence testicular function. Whether the effect of PFDA on Leydig cell PBR expression is unique or representative of other classes of peroxisome proliferators is under investigation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Endocrinologie Moléculaire de la
Reproduction, UPRES-A CNRS 6026, Université de Rennes I, 35042
Rennes Cedex, France.
3 Current address: Department of Cell Biology & Biochemistry, Texas
Tech University Health Sciences Center, Lubbock, Texas 79430.
Received March 23, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
: central mediator of peroxisome
proliferator toxicity. CIIT Activities, Chemical Industry Institute of
Toxicology 17:1–9
, -ß, and - in the adult rat. Endocrinology 137:354–366[Abstract]
messenger ribonucleic acid and
protein in human and rat testis. Endocrinology 140:2968–2975This article has been cited by other articles:
 |
|
 |
|
  X. Cheng and C. D. Klaassen Critical Role of PPAR-{alpha} in Perfluorooctanoic Acid- and Perfluorodecanoic Acid-Induced Downregulation of Oatp Uptake Transporters in Mouse Livers Toxicol. Sci., November 1, 2008; 106(1): 37 - 45. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-Y. Zhang, Y. Ito, O. Yamanoshita, Y. Yanagiba, M. Kobayashi, K. Taya, C. Li, A. Okamura, M. Miyata, J. Ueyama, et al. Permethrin May Disrupt Testosterone Biosynthesis via Mitochondrial Membrane Damage of Leydig Cells in Adult Male Mouse Endocrinology, August 1, 2007; 148(8): 3941 - 3949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Shi, H. Zhang, Y. Liu, M. Xu, and J. Dai Alterations in Gene Expression and Testosterone Synthesis in the Testes of Male Rats Exposed to Perfluorododecanoic Acid Toxicol. Sci., July 1, 2007; 98(1): 206 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gravel and M. M. Vijayan Salicylate Disrupts Interrenal Steroidogenesis and Brain Glucocorticoid Receptor Expression in Rainbow Trout Toxicol. Sci., September 1, 2006; 93(1): 41 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gazouli, Z.-X. Yao, N. Boujrad, J. C. Corton, M. Culty, and V. Papadopoulos Effect of Peroxisome Proliferators on Leydig Cell Peripheral-Type Benzodiazepine Receptor Gene Expression, Hormone-Stimulated Cholesterol Transport, and Steroidogenesis: Role of the Peroxisome Proliferator-Activator Receptor {alpha} Endocrinology, July 1, 2002; 143(7): 2571 - 2583. [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 |
