| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Physiology and Obstetrics and Gynecology, University of Western Ontario, London, Ontario, Canada N6A 5C1
Address all correspondence and requests for reprints to: Dr. David K. Pomerantz, Department of Physiology, University of Western Ontario, Medical Sciences Building, London, Ontario, Canada N6A 5C1. E-mail: dpomer{at}physiology.uwo.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(50 U/ml) and then bacterial
lipopolysaccharide (LPS; 100 ng/ml) for 24 h each. At various
times after initiation of activation, selected wells were harvested for
identification of messenger RNA for inducible NO synthase by RT-PCR.
Amplicons of the predicted 651-bp product were isolated, cloned, and
sequenced to validate the PCR procedure. Such amplicons first appeared
between 2–4 h after exposure to LPS, and staining increased in
intensity for the rest of the study. Nitrite accumulation followed a
similar time course. Similarly treated wells were washed after 24-h
activation and cocultured with purified LC for a final 24-h incubation
in the absence of interferon- and LPS. Basal and LH-stimulated
production of androgen was estimated by RIA. In some experiments the NO
synthase inhibitor
N
-nitro-L-arginine methyl
ester or the NO scavenger
2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide
(C-PTIO) was added during activation and coculture. Coculture of LC
with quiescent macrophages altered neither basal nor LH-stimulated
androgen production. Coculture with either type of activated macrophage
did not alter basal, but significantly reduced (by 50%) LH-stimulated,
androgen production.
N-Nitro-L-arginine methyl
ester and C-PTIO blocked the inhibitory effect. The NO donor
S-nitroso-N-acetyl penicillamine at
concentrations greater than 10-5 M
significantly inhibited LH-stimulated androgen production by purified
LC (P < 0.01). The inhibitory effect of
S-nitroso-N-acetyl penicillamine was
evident when exposure exceeded 4 h. Intermediates of
steroidogenesis were added to elucidate the site of NO inhibition. The
enzymatic inhibition occurred at least in part at
17
-hydroxylase/C17/20 lyase (P450c17). Enzyme inhibition
was reversed by C-PTIO. Northern blot analysis indicated that
accumulation of messenger RNA for P450c17 was not significantly
altered. Therefore, activation of macrophages results in decreased
androgen production by cocultured LC. The inhibition is mediated in
part by macrophage-derived NO acting directly on the LC via inhibition
of at least one of the P450 steroidogenic enzymes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, -ß, and - (IFN, -ß, and -); and
tumor necrosis factor- inhibit androgen secretion (5). As macrophage
function becomes better understood, it is clear that other products of
this cell may also be candidates for modulators of testosterone
secretion; nitric oxide (NO) is such a molecule. NO is a simple and relatively unstable radical under physiological conditions. It has been characterized as a widely distributed, biologically produced molecule that can mediate intracellular and intercellular communication. NO acts as an endothelium-derived relaxation factor, as a neurotransmitter, and as an important molecule mediating the cytotoxic actions of the immune system. The molecule is synthesized from L-arginine in a reaction catalyzed by NO synthase (NOS), an enzyme that exists in several isoforms. Macrophages respond to activation by cytokines with increased synthesis of an inducible form of the enzyme (iNOS) that is neither calcium nor calmodulin dependent. The NO produced as a result of iNOS induction may mediate some of the tumoricidal or antimicrobial effects of macrophages (6, 7, 8).
Several recent reports have provided evidence that NO may directly influence gonadal steroidogenesis. Progesterone and estradiol secretion by human granulosa/luteal cells is inhibited by NO (9), as is secretion of progesterone by mouse MA10 tumor cells and testosterone secretion by rat LC (10). These studies raised the possibility that the inhibitory effect of macrophages on androgen secretion reported by others could be mediated by NO.
In this report, we established that in vitro activation of
peritoneal macrophages or a macrophage cell line was associated with
elevated steady state levels of messenger RNA (mRNA) for iNOS and
nitrite accumulation. When such activated cells were cocultured with
normal mouse LC, inhibition of androgen production occurred, and this
effect was blocked by inhibition of iNOS activity or blockade of NO
action. Finally, we have demonstrated that an NO donor can act directly
on normal mouse LC to inhibit steroidogenesis at least in part by
decreasing the activity of the enzyme
17-hydroxylase/C17/20 lyase (P450c17).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of macrophages
Peritoneal macrophages were obtained from exudates elicited by
ip injection of 1 ml sterile 4% Brewer’s thioglycolate medium (Difco,
Detroit, MI). Animals were killed 3–5 days later, and using sterile
procedures, cells were obtained by peritoneal lavage with calcium- and
magnesium-free Hanks’ Balanced Salt Solution. After two washes with
RPMI 1640 supplemented with 100 U penicillin and 100 µg
streptomycin/ml (Life Technologies, Grand Island, NY), the cell
suspension was allowed to adhere to plastic culture dishes for 1 h
at 34 C. Nonadherent cells were discarded; adhered cells were lifted
with a plastic scraper and resuspended in supplemented RPMI. Cell
numbers were determined with a hemocytometer, and viability, assessed
by trypan blue exclusion after 3 min, exceeded 95%. Purity was
estimated by immunohistochemical identification of the cell surface
marker F4/80 (11), and for the studies reported here, 70–80% of
peritoneal cells were positive for this marker. The identities of the
contaminating cells are not known, but they are assumed to be a mixture
of lymphocytes and/or other migratory cells.
To use a pure macrophage preparation, the murine cell line RAW 264.7 (American Type Culture Collection, Rockville, MD) was grown in the same supplemented RPMI 1640 described above with the addition of 10% FBS. Cells were passaged two or three times weekly and harvested for activation at various times after passage. The F4/80 marker was consistently present on more than 97% of the RAW cells.
The peritoneal or RAW cells were plated in serum-free RPMI in six-well
tissue culture plates (Falcon, VWR Scientific, Toronto, Canada) at
5 x 105 viable cells/cm2. Twenty-four
hours later, nonadherent cells were aspirated, and the remaining
macrophages were primed with 50 U/ml recombinant murine IFN
(Genzyme, Cambridge, MA) for 24 h. Macrophages were then activated
by exposure to 100 ng/ml Escherichia coli lipopolysaccharide
(LPS; serotype 0111:B4, Sigma Chemical Co., Oakville, Canada). IFN
increases the rate of iNOS mRNA transcription in response to LPS (12).
At various time intervals after activation, medium was collected, and
Griess reagent was used to determine nitrite accumulation (13). The
cells were then lifted from plates and sonicated briefly. The sonicate
was centrifuged for 15 min at 12,000 x g, the
supernatant was discarded, and the pellet was resuspended 1 ml GTC
buffer [4 M guanidinium thiocyanate, 0.1 M
Tris-HCl (pH 7.4), and 1% ß-mercaptoethanol; Sigma], 20 µg
E. coli ribosomal RNA (Boehringer Mannheim, Laval, Canada)
in 5 µl were added, and total RNA was extracted (14). The RNA was
reverse transcribed by resuspending 2 µg total RNA and 0.6 µg
3'-primer in diethylpyrocarbonate H2O (10 µl). The
mixture was incubated for 10 min at 70 C, placed on ice for 4 min, and
then centrifuged for 15 sec at 12,000 x g before the
addition of 10 µl reverse transcriptase mix (1 µl RNA guard, 4 µl
5 x first strand buffer, 2 µl 0.1 M dithiothreitol,
2 µl 200 µM deoxy-NTPs, and 1 µl Superscript reverse
transcriptase; Life Technologies) to each tube. Samples were incubated
at 43 C for 90 min, followed by a 5-min incubation at 94 C. The
upstream and downstream primer sequences were
5'-CAACAGGAACCTACCAGCTCA-3' and 5'-GATGTTGTAGCGCTGTGTGTCA-3',
respectively. These sequences have been used by others (15) and produce
a predicted product of 651 bp that has greater than 97% homology with
the rat form of iNOS. PCR was conducted in a reaction mixture
containing 1 U Taq DNA polymerase in a final volume of 50
µl, consisting of 5 µl 10 x PCR buffer (200 mM
Tris-HCl, pH 8.4, and 500 mM KCl), 1.5 mM
MgCl2, 0.2 mM dNTPs (all reagents from Life
Technologies), and 2 µM of each primer; 5-µl aliquots
of complementary DNA (cDNA) were used. The mixture was overlaid with
mineral oil and amplified by PCR for 40 cycles in a DNA thermal cycler
(Perkin-Elmer/Cetus 480, Norwalk, CT), with each cycle consisting of
denaturing temperature of 94 C for 1 min, reannealing of primers to
target sequence at 55 C for 30 sec, and extension at 72 C for 1 min.
PCR products were resolved on 2% agarose gels (Bio-Rad, Richmond, CA)
containing 0.5 µg/ml ethidium bromide. Lanes with a 100-bp DNA ladder
(Life Technologies) served as a standard. Controls are described in
Results.
The identities of the amplified transcripts were established by sequence analysis. The PCR product was directly ligated into the PCR 2.1 TA vector (Invitrogen, San Diego, CA) following the instructions provided by the manufacturer. Clones of transformed cells were selected and grown overnight in Luria-Bertoni broth, and DNA was extracted using the FlexiPrep Kit (Pharmacia Biotech, Uppsala, Sweden). Automated sequencing was provided by a commercial service at the University of Guelph (Guelph, Ontario, Canada).
Preparation of purified LC and androgen assay
Testes were aseptically removed, decapsulated, and minced with
scissors. Dispersion of the tissue was carried out in medium 199 with
Hanks’ salts (Life Technologies) containing 12 ng/ml deoxyribonuclease
(Sigma Chemical Co., St. Louis, MO) by gentle swirling on a magnetic
stirrer for 10 min at 37 C. The suspension was repeatedly triturated
with a fire-polished Pasteur pipette for the first minute to break up
large clumps, filtered through a double layer of nylon mesh (70 µm),
and centrifuged at 400 x g at 4 C for 15 min. The
cellular pellet was resuspended in 1–4 ml (depending on the number of
testes processed) medium 199 with Hanks’ salts at pH 7.4 supplemented
as follows: sodium bicarbonate: 0.7 g/liter; HEPES, 25 mM
(Life Technologies); penicillin, 100 IU/ml; streptomycin, 100 µg/ml
(Life Technologies); and BSA, 2 mg/ml (Sigma, St. Louis, MO).
Isosmotic, continuous density gradients of Percoll (Pharmacia) were
prepared by a modification of the method reported by Browning et
al. (16). Nine parts Percoll were mixed with 1 part (vol/vol)
10 x Hanks’ buffered medium 199 containing 20 mg/ml BSA. The pH
was brought to 7.4 with 1 N HCl. Gradients were prepared by
centrifuging 8.5 ml of this mixture in 15-ml tubes at 4 C for 1 h
at 20,000 x g. One half milliliter of the suspension
of testicular cells was layered onto each gradient and then centrifuged
at 800 x g for 20 min at 4 C. The layer containing LC,
a 2.0-ml band to whose center 1.062 g/ml density calibration beads
migrate, was washed three times with medium, and the cells were
resuspended in medium 199 with Earle’s salts supplemented with 2.2
g/liter sodium bicarbonate at the same pH and at the concentrations of
insulin, penicillin, streptomycin, and BSA indicated above.
Cell viability was determined using trypan blue dye exclusion and constituted greater than 90% of the cells observed. Cells were stained for 3ß-hydroxysteroid dehydrogenase (3ßHSD) as a marker for steroidogenic cells according to the method described by Wiebe (17); etiocholanol was used as a substrate. The preparations were found to contain 95–99% 3ßHSD-positive cells. LC were incubated at 33–34 C in 24-well polystyrene culture plates (Costar, Cambridge, MA) at a concentration of 10,000 viable, 3ßHSD-positive cells/0.5 ml·well. The incubations were carried out in 95% air and 5% CO2 in a water-saturated atmosphere, and treatments were replicated in four to six wells. After 20 h, LH or blank medium was added to the appropriate wells as described for the individual experiments, and 4 h later, the experiments were terminated. Medium was collected and centrifuged at 2000 x g at 4 C for 10 min; supernatant was aspirated and then stored at -20 C until assayed for androgen. The data are expressed as androgen accumulation per million LC/24 h.
An established RIA was used to determine androgen accumulation (18). The data are reported as androgen released per million LC during the final 4 h of incubation. The sensitivity of the assay was 2.9 pg/tube, and the intra- and interassay coefficients of variation were 3.2% and 5.7%, respectively, for the experiments reported here.
Preparation of macrophages for coculture with LC
One million viable peritoneal macrophages or RAW cells were
plated in serum-free RPMI in 24-well tissue culture plates. Adherence,
priming, and activation were carried out as described above. After
aspiration of the IFN/LPS-containing activation medium, 10,000
purified, viable, murine LC, prepared as described above, were added to
each well. After a further 20 h, additional control medium or
medium containing ovine LH (to achieve 200 pg NIDDK oLH-S26/ml) was
added for the final 4 h of incubation. Medium was then aspirated,
cells were removed by centrifugation, and the supernatants were frozen
at -20 C until RIA for androgen. In several experiments the effect of
inhibition of NO synthesis or scavenging of NO was evaluated. In each
case, compounds were added to the culture wells at the same time as LPS
and maintained for the duration of the experiment. The competitive
inhibitor of NOS,
N-nitro-L-arginine methyl ester
(L-NAME; Sigma, Oakville, Canada) was used at a final concentration of
1 mM. This concentration effectively eliminates NO
production in vitro (19). We also tested the effect of a
NO scavenger,
2–4-(-carboxyphenyl-4,4,5,5,-tetramethylimidazoline-1-oxyl-3-oxide
(C-PTIO; Precision Biochemicals, Vancouver, Canada). C-PTI0 (250
µM) was added to the appropriate wells at the same time
as the LC. This concentration was arrived at after preliminary
experiments in our laboratory and consideration of published data
(20).
Site of action of NO on purified LC
To more directly address the contribution of NO to the
inhibitory effects of macrophages on androgen production, we evaluated
the effects of the NO donor S-nitroso-N-acetyl
penicillamine (SNAP; Sigma, Oakville, Canada), at various
concentrations (10-6–10-3 M) and
for various lengths of time (4, 8, 12, 16, and 24 h). In the
remaining experiments, a technical variation was introduced. At the
onset of the final 4-h incubation (except for incubations of 4-h
duration), appropriate wells were sampled to determine the background
androgen release that had occurred until that point in the experiment.
LH-S26 or blank medium was then added to each remaining well to achieve
the appropriate gonadotropin concentration. At the end of incubation,
medium was again collected and stored until assay for androgen content.
The background value was subtracted from the final value, and the data
are reported as androgen released per million LC during the final
4 h of incubation.
Incubation with various saturating concentrations (21) of
precursors of testosterone was employed to determine the site in the
steroidogenic pathway of inhibition caused by SNAP. Cells were cultured
for 20 h with or without SNAP, and then one of the following
precursors was added for the final 4 h to achieve the
concentrations indicated: 22R-hydroxycholesterol (5
µg/ml), pregnenolone, progesterone, 17-hydroxypregnenolone,
17-hydroxyprogesterone, dehydroepiandrosterone, androstenedione, and
androstenediol (all at 500 ng/ml). Wells containing each precursor, but
devoid of LC, provided medium that was used as a blank in the androgen
assay.
To confirm that the SNAP effect we observed was primarily due to NO released into the medium, cells were cultured as described above with the addition of the NO scavenger used above. C-PTIO (250 µM) was added to the appropriate wells immediately before SNAP.
Extraction of total RNA and quantification of mRNA levels
To obtain information about the mechanism of the inhibitory
effect of NO on the activities of selected steroidogenic enzymes, net
accumulation of mRNA for 3ßHSD (as a noncytochrome P450 enzyme) and
P450c17 (as a P450-associated enzyme) was assessed by Northern blot
analysis. One million LC were incubated for 24 h with or without
10-4 M SNAP. During the final 4 h of the
experiment, cells received control medium, or LH was added to achieve
200 pg/ml. Medium was then aspirated, and total RNA was extracted using
the technique described above for macrophages. Five micrograms of total
RNA from each treatment group were resuspended in 50% formamide, 2.2
M formaldehyde, 20 mM
3[N-morpholino]propane sulfonic acid (MOPS), 50
mM acetate, and 10 mM EDTA, pH 7.0, and
subjected to electrophoresis in 1.1% agarose containing 2.2
M formaldehyde, 20 mM MOPS, 50 mM
sodium acetate, and 10 mM EDTA, pH 7.0. RNA was transferred
to Hybond-N nylon membrane (Amersham, Arlington Heights, IL) by
overnight capillary blotting in 3 M NaCl and 0.3
M Na citrate, pH 7.0. The blots were cross-linked in a UVC
500 UV cross-linker (Hoefer, San Francisco, CA) at 120
mJ/cm2 and prehybridized for 30 min at 65 C in Church
buffer [0.5 M Na2PO4, 1
mM EDTA, 1% (wt/vol) BSA, and 0.25 M SDS].
The blots were then hybridized overnight in hybridization buffer at 60
C with 106 cpm/ml 32P-labeled cDNA. DNA probes
were radiolabeled with [-32P]deoxy-CTP using the
Random Primer DNA labeling system (Life Technologies). After
hybridization, the blots were washed in 0.04 M
Na2PO4-0.1% SDS three times at 65 C for 15
min. Radioactivity was visualized by autoradiography and quantitated by
densitometry.
Statistical analysis
All experiments reported here were repeated at least three
times, and the data were pooled. If heterogeneity of variance was
detected by Bartlett’s test, this was reduced by logarithmic
transformation of the data before analysis. These data were then
subjected to multifactor ANOVA, the Tukey-Kramer test for multiple
comparisons, and Student’s t test for comparing two means
(22). P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). ANOVA
detected a significant (P < 0.01) interaction between
time and activation state of the RAW cells, because quiescent cells did
not cause increased nitrite accumulation, but, beginning 4 h after
exposure to LPS, sensitized RAW cells produced increasing amounts of
nitrite for the duration of the experiment. Similar data (not shown)
were obtained from peritoneal macrophages. RT-PCR of mRNA obtained from
similarly treated RAW cells produced an amplicon of the predicted size
of approximately 650 bp (Fig. 2). This
material was also present when RNA from quiescent cells was used,
although the signal strength was weak. Isolation, cloning, and
sequencing of this amplicon indicated a cDNA whose base sequence had
93.5% homology with the appropriate portion of iNOS cDNA reported by
Xie et al. (23).
|
|
inhibits androgen production by mouse LC (24). For this
reason, all of the data shown here were obtained from incubations in
which macrophage-conditioned medium was replaced by fresh medium,
without IFN/LPS, before the introduction of LC at the onset of the
final 24-h incubation. Thus, any effects noted are due to macrophage
activities that are maintained in vitro after removal of
IFN/LPS.
In the absence of macrophages the basal accumulation of androgen was
0.31 ± 0.05 µg/million LC·24 h. Exposure to 200 pg LH/ml
increased accumulation to 2.5 ± 0.4 µg/million cells (Fig. 3). When LC were cocultured with
quiescent cells for 24 h, neither basal nor LH-stimulated androgen
accumulation was altered. However, activation of macrophages produced a
significant (P < 0.02) interaction because basal
androgen release was unaltered by macrophage activation, whereas
LH-stimulated androgen release was significantly (P <
0.01) reduced.
|
). A final test of NO
involvement in inhibition of androgen secretion used the NO scavenger
C-PTIO. This compound itself had a modest, but statistically
insignificant, inhibitory effect on basal and LH-stimulated androgen
accumulation. As with L-NAME, the inhibitory effect of activated
macrophages on androgen accumulation was completely reversed by
scavenging NO with C-PTIO (Fig. 5).
|
|
and 4; similar results were obtained. Quiescent RAW cells had no effect
on basal or stimulated androgen production, whereas significant
(P < 0.02) inhibition of androgen production occurred
when the RAW cells were activated with sequential IFN/LPS before the
addition of LC. This inhibition was prevented by inhibition of iNOS
activity with L-NAME (Fig. 6).
|
The effect of 10-4 M SNAP for 24 h on
androgen production in response to various concentrations of LH is
shown in Fig. 7. Multifactorial ANOVA
detected significant interaction between LH concentration and SNAP
treatment (P < 0.001); this was due to LH eliciting a
concentration-dependent response (P < 0.001), which
was completely blocked by SNAP at all concentrations of LH.
|
. Purified LC were incubated with
or without SNAP (10-4 M) for 4, 8, 16, and
24 h. LH (100–1600 pg/ml) or blank medium was added the last
4 h of each incubation. Cells that did not receive SNAP showed an
increase in androgen production that was dependent on the concentration
of LH, and this was observed for all durations of incubation. The
response to the maximal concentration of LH did not change during the
entire experiment, and the response to 100 and 400 pg/ml LH actually
increased (P < 0.01). Thus, there was no diminution of
the LC capacity to produce androgen in response to LH during 24-h
incubation under the conditions of this study. Separate 2 x 3
ANOVA was carried out on the data from each duration of incubation. At
4 h, there was no interaction between the LH concentration and the
presence of SNAP (P > 0.3), indicating that the
release of androgen in response to each concentration of LH was
unaltered by exposure to SNAP for 4 h. However, exposure to SNAP
for 8, 16, or 24 h caused significant (P < 0.001)
interaction between the LH concentration and the presence of SNAP. This
effect was attributed to inhibition of LH-stimulated androgen
accumulation at all concentrations of LH. Although a low level of
androgen production could be detected after 8-h exposure to SNAP,
inhibition was complete at 16- and 24-h exposure.
|
4- or 5-precursors of
testosterone were added to incubation wells for the final 4 h of
24-h incubations during which cells had been exposed to control medium
or medium containing 10-4 M SNAP. Each data
pair was evaluated by t test. It was found that
androstenedione, dehydroepiandrosterone, and androstenediol completely
overcame the inhibition caused by SNAP (P < 0.01),
whereas 22R-hydroxycholesterol, pregnenolone, progesterone,
17-hydroxypregnenolone, and 17-hydroxyprogesterone did not (Figs. 9 and 10).
|
|
).
|
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that activation of macrophages with IFN/LPS was associated with an increase in nitrite accumulation and a concomitant increase in the signal strength of a 651-bp amplicon produced by RT-PCR of iNOS mRNA. Isolation, cloning, and sequencing of this amplicon validated the RT-PCR procedure. Although our RT-PCR approach cannot be treated as a quantitative measure, the results are consistent with earlier data from Northern blot analysis in which steady state levels of iNOS mRNA increased between 2–4 h after activation of RAW cells, and the accumulation of message and nitrite remained elevated for at least 12 h (12). In addition, other studies have shown macrophage activation to be associated with a prolonged increase in NO production that is optimal between 24–48 h after activation with LPS (29).
Coculture of quiescent macrophages with LC had no effect on LH-stimulated androgen production. However, activation of either type of macrophage consistently caused inhibition of stimulated androgen production during the period of optimal NO production. There are unresolved inconsistencies in the existing literature concerning the in vitro effect of macrophages on androgen production by LC. NO is unstable in aqueous media and is probably dissipated from macrophage-conditioned medium. It is likely that NO exerts its biological effects near its site of production. Thus, only studies in which macrophages and LC were cocultured can be properly compared with our data. Our studies agree in part with those of Sun et al. (30), who also found no effect of coincubation with quiescent rat testicular macrophages on basal androgen secretion. However, in contrast to our findings, they observed inhibition of LH-stimulated steroidogenesis. A report of coculture of nontesticular macrophages and LC used LPS treatment of both cell types (31). No effect on basal secretion of androgen was noted, and LH-stimulated LC were not studied. The differences are probably due to significant variation in experimental design, such as duration of coculture and omission of IFN sensitization before LPS treatment.
Our studies are the first to directly demonstrate a role for NO in the inhibitory effect that cocultured macrophages exert on steroidogenesis in LC. We found that inhibition of iNOS activity with L-NAME completely reversed the inhibitory effect of activated cells without directly effecting the LC. Putative effects of NO have been studied by use of a binding compound such as glutathione or hemoglobin to reduce or block the observed effect. However, results are equivocal because such agents may actually serve as competitors with target molecules of NO or as NO carriers rather than scavengers (32). The availability of imidazolineoxyl N-oxide derivatives has provided an improved approach. C-PTIO is a potent and highly selective NO scavenger that has been used in vivo to inhibit hypotension and endotoxic shock induced by LPS and in vitro to prevent NO-induced vascular relaxation (20, 33, 34). We demonstrated here that C-PTIO completely prevented the inhibitory effect of activated peritoneal macrophages on androgen production. When considered with the data obtained with L-NAME, we conclude that NO release is involved in the mechanism(s) by which activated macrophages inhibit LH-stimulated androgen secretion.
Our coculture experiments cannot be used to eliminate other macrophage
products from consideration as modulators of LC steroidogenesis, nor do
they address whether the NO produced by macrophages acts directly on LC
or has an autocrine effect, which, in turn, produces a cytokine
directed at the LC. Previous workers have concluded that
macrophage-derived interleukin-1, -1ß, and -2; IFN; and tumor
necrosis factor- all are capable of inhibiting LH-stimulated
androgen synthesis (see Ref. 5 for review). These reports include
studies in which macrophage-conditioned medium and/or purified
cytokines were used.
Therefore, we undertook an additional series of experiments using the NO donor SNAP and highly purified mouse LC. We demonstrated that NO can inhibit androgen production by an action directly on the LC. This assertion is supported by the finding that the NO donor SNAP caused significant inhibition of both basal and LH-stimulated production of androgen in purified LC cultures. The inhibitory effect of SNAP became significant between 4–8 h after application to the LC and remained complete at 16 and 24 h. It is unlikely that this inhibitory effect is the consequence of nonspecific toxicity, because functional viability remained after 24-h exposure to SNAP when several testosterone precursors were shown to overcome the inhibitory effect of SNAP on production of androgen. C-PTIO completely reversed the inhibitory effect of SNAP on LH-stimulated androgen production, but had no effect of its own. These data further support our contention that the inhibitory effect of SNAP on androgen production by the LC of the mouse is due to released NO and not to another by-product of SNAP breakdown. Other workers (10) were unable to detect inhibition of progesterone secretion by MA-10 cells when SNAP was applied at 10-4 M, and inhibition was about 50% of the control value with 10-3 M, whereas these two concentrations produced 94% and 100% inhibition, respectively, in our hands. A major difference in the two studies was their duration (3.5 vs. 24 h), suggesting an interaction between the duration of the incubation and the concentration of NO in the manifestation of the inhibitory effect. Our findings are more compatible with those of Van Voorhis et al. (9), who demonstrated that SNAP (10-3 M for 16 h) caused 80% and 90% reductions in progesterone and estradiol production, respectively, by cultured human granulosa/luteal cells. Furthermore, in those studies it was shown that native NO, SNAP, as well as an additional NO donor produced identical results.
Despite observing a biologic effect and preservation of cell viability, we cannot be certain whether the concentration of NO produced by 10-4 M SNAP in this study or those used in previous studies with other NO donors as well as SNAP at similar concentrations in a variety of experimental systems represent physiologically meaningful amounts for the interstitium of the testis. Rubbo et al. (35) used physical methods in vitro to estimate that the concentration of SNAP used here would produce a peak of 8 µM NO during a 30-min incubation. Although this concentration exceeds the amounts of NO thought to be present in normal tissue, it does approach values indirectly estimated for NO release by cytokine-activated nontesticular macrophages (36) and decidual mononuclear cells from sites of early embryo loss in the uterus of the mouse (37). If it is eventually found that transient or resident macrophages, the LC itself, or other testicular cells are sources of NO, it is possible that the local concentrations of NO achieved in our studies will fall within limits for the testicular interstitium. Further, the apparent increase in effectiveness of lower doses of SNAP after 24-h exposure may be taken to suggest that lower concentrations in vivo are effective only after a prolonged presence. Definitive resolution of this issue must await further study.
NO elicits many of its biological actions by reacting with
Fe2+ heme proteins to form NO-heme complexes. This
interaction with NO translates into activation for some enzymes, such
as soluble guanylyl cyclase and cyclooxygenase, while for others, such
as lipoxygenases, catalase, and peroxidase, NO binding has been shown
to be inhibitory (38). Cytochrome P450-associated enzymes, which also
contain Fe heme groups, catalyze several reactions in the synthesis of
testosterone. Dehydroepiandrosterone, androstenedione, and
androstenediol completely overcame the inhibitory effect of SNAP; thus,
it appears that NO has a negligible effect on 3ßHSD and
17ß-hydroxysteroid dehydrogenase activities in the LC of the adult
mouse. The failure of 22R-hydroxycholesterol, pregnenolone,
and progesterone and their 17-derivatives to overcome the inhibitory
effect of SNAP is consistent with inhibition of side-chain cleavage
activity and/or P450c17 activity, both of which are cytochrome P450
enzymes. That side-chain cleavage cytochrome P450 activity may be a
site of action is supported by the observations that NO inhibited the
secretion of progesterone by ovarian cells (9) and that NO inhibition
of progesterone production could be overcome by pregnenolone, but not
by 22R-hydroxycholesterol in MA-10 cells (10). More distal
enzymes in the normal steroidogenic pathway of the LC were not studied.
The latter data also suggest that transmitochondrial movement of
cholesterol is not altered by NO. Our data suggest that the activity of
P450c17 is also inhibited, because neither pregnenolone, progesterone,
nor their 17-derivatives had a significant effect on the inhibition
of androgen synthesis caused by SNAP.
The mechanism by which NO exerts its inhibitory effect on P450 steroidogenic enzymes is not known. NO increases intracellular cGMP in many cells, presumably through the activation of soluble guanylate cyclase (6). This mechanism is not likely to account for our observations because a correlation between cGMP and NO effects on steroidogenesis in the testis and granulosa/luteal cells has not been found (9, 10, 28). Furthermore, increased intracellular cGMP in the mouse LC is associated with an increase in androgen production (39). Stradler et al. (40) demonstrated that SNAP not only bound to but also decreased the expression of message for liver cytochrome P4501A. Under the conditions of our experiments, we could not detect a significant effect of NO on the expression of mRNA for key P450-associated and P450-independent steroidogenic enzymes. Given the profound inhibition of P450c17 activity with negligible decrease in message, it appears most likely that in the LC, NO interacts directly with the Fe heme of the P450 enzymes and alters their catalytic properties.
The argument was presented above that disease states associated with macrophage activation are often associated with decreased circulating levels of androgen. Bosmann and co-workers (41) recently showed that LPS-induced endotoxemia in mice resulted in a greater than an 80% decrease in circulating androgen when assessed 24 h later. Our studies may be taken to suggest that NO contributes to the hypogonadal state in this model. It is possible that NO released by circulating macrophage/monocytes passing through the testis is the source of this inhibition. It is also possible that these disease states result in activation of testicular macrophages. It is well known that agents that activate peripheral macrophages are capable of exerting similar, but not always identical, effects on macrophages isolated from the rat and mouse testes (5, 42, 43). Consistent with this suggestion are the results reported by Welch and co-workers (28) that L-NAME could increase hCG-stimulated androgen production by an unpurified interstitial preparation from rat testes.
Although our initial view was that macrophages were the source of the NO that acted on LC in vivo, there are other potential sites for testicular production of NO. Cultured Sertoli cells from immature rats express the inducible form of NOS (44), and the constitutive form has been immunohistochemically localized in human Sertoli cells (45). Burnett et al. (25) reported that calcium-dependent NOS activity was confined to the testicular vascular endothelium in the rat, but Davidoff and co-workers (46) localized NOS-1 (the calcium-dependent, brain form of the enzyme) in some, but not all, human LC as well as in mouse MA10 and TM3 LC lines. Localization of endothelial type NOS in the human LC has recently been reported by Zini and co-workers (45). Our data allow the possibility of an autocrine effect of NO produced by the LC on steroidogenesis. Direct assessment of iNOS activity in testicular cells will be required to unequivocally define the source(s) of induced NO production in the testis.
In this report, we established that in vitro activation of
peritoneal macrophages or a macrophage cell line was associated with
elevated steady state levels of mRNA for iNOS and nitrite accumulation.
When such activated cells were cocultured with normal mouse LC,
inhibition of LH-stimulated androgen production occurred, and this
effect was blocked by inhibition of iNOS activity or blockade of NO
action. Finally, we demonstrated that a NO donor can act directly on
normal mouse LC to inhibit steroidogenesis at least in part by
decreasing the activity of the enzyme
17-hydroxylase/C17/20 lyase. These findings show that NO
may be capable of acting within the gonad to regulate androgen
production and thus may be viewed as an additional regulator of male
fertility and sexual function, especially in disease states associated
with immune activation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received July 28, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and lipopolysaccharide. J Biol
Chem 268:1908–1913This article has been cited by other articles:
 |
|
 |
|
  C. Mondillo, R. M. Pagotto, B. Piotrkowski, C. G. Reche, Z. J. Patrignani, C. B. Cymeryng, and O. P. Pignataro Involvement of Nitric Oxide Synthase in the Mechanism of Histamine-Induced Inhibition of Leydig Cell Steroidogenesis via Histamine Receptor Subtypes in Sprague-Dawley Rats Biol Reprod, January 1, 2009; 80(1): 144 - 152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kondo, T. Ishikawa, K. Yamaguchi, T. Yada, and M. Fujisawa Oral Administration of Tetrahydrobiopterin Attenuates Testicular Damage by Reducing Nitric Oxide Synthase Activity in a Cryptorchid Mouse Model J Androl, March 1, 2008; 29(2): 153 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Andric, M. M. Janjic, N. J. Stojkov, and T. S. Kostic Protein kinase G-mediated stimulation of basal Leydig cell steroidogenesis Am J Physiol Endocrinol Metab, November 1, 2007; 293(5): E1399 - E1408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Herman and C. Rivier Activation of a Neural Brain-Testicular Pathway Rapidly Lowers Leydig Cell Levels of the Steroidogenic Acute Regulatory Protein and the Peripheral-Type Benzodiazepine Receptor while Increasing Levels of Neuronal Nitric Oxide Synthase Endocrinology, January 1, 2006; 147(1): 624 - 633. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Weissman, E. Niu, R. Ge, C. M. Sottas, M. Holmes, J. C. Hutson, and M. P. Hardy Paracrine Modulation of Androgen Synthesis in Rat Leydig Cells by Nitric Oxide J Androl, May 1, 2005; 26(3): 369 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Asirvatham and B. C. Bruot Cell-Cell Interactions in the Testis of Adjuvant-Induced Arthritic Rat J Androl, July 1, 2002; 23(4): 491 - 497. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, D. C. Newton, T. L. Miller, A.-M. Teichert, M. J. Phillips, M. S. Davidoff, and P. A. Marsden An Alternative Promoter of the Human Neuronal Nitric Oxide Synthase Gene Is Expressed Specifically in Leydig Cells Am. J. Pathol., January 1, 2002; 160(1): 369 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Peterson, F. Moran, A. J. Conley, and I. M. Bird Zonal Expression of Endothelial Nitric Oxide Synthase in Sheep and Rhesus Adrenal Cortex Endocrinology, December 1, 2001; 142(12): 5351 - 5363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. O'Bryan, S. Schlatt, O. Gerdprasert, D. J. Phillips, D. M. de Kretser, and M. P. Hedger Inducible Nitric Oxide Synthase in the Rat Testis: Evidence for Potential Roles in Both Normal Function and Inflammation-Mediated Infertility Biol Reprod, November 1, 2000; 63(5): 1285 - 1293. [Abstract] [Full Text]
|
|
|
|
|
|
C. J. Hanke and W. B. Campbell Endothelial cell nitric oxide inhibits aldosterone synthesis in zona glomerulosa cells: modulation by oxygen Am J Physiol Endocrinol Metab, October 1, 2000; 279(4): E846 - E854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Hanke, T. O'Brien, K. A. Pritchard Jr, and W. B. Campbell Inhibition of Adrenal Cell Aldosterone Synthesis by Endogenous Nitric Oxide Release Hypertension, January 1, 2000; 35(1): 324 - 328. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. H. Zwain and S. S. C. Yen Neurosteroidogenesis in Astrocytes, Oligodendrocytes, and Neurons of Cerebral Cortex of Rat Brain Endocrinology, August 1, 1999; 140(8): 3843 - 3852. [Abstract] [Full Text]
|
|
|
|
 |
|
 E. Xiao, L. Xia-Zhang, and M. Ferin Stress and the Menstrual Cycle: Short- and Long-Term Response to a Five-Day Endotoxin Challenge during the Luteal Phase in the Rhesus Monkey J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 623 - 626. [Abstract] [Full Text]
|
|
|
|
|
|
I. H. Zwain and S. S. C. Yen Dehydroepiandrosterone: Biosynthesis and Metabolism in the Brain Endocrinology, February 1, 1999; 140(2): 880 - 887. [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 |
