| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Laboratoire de Neurobiologie des Fonctions Végétatives (A.S., C.M.), Laboratoire de Physiologie Sensorielle (C.B.), Laboratoire de Biologie Cellulaire et Moléculaire (E.D.), Institut National de la Recherche Agronomique, F-78352 Jouy-en-Josas Cedex, France
Address all correspondence and requests for reprints to: Dr. Alain Schirar, Laboratoire de Neurobiologie des Fonctions Végétatives, INRA, F-78352 Jouy-en-Josas Cedex, France. E-mail: schirar{at}jouy.inra.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-dihydrotestosterone (5DHT) or left empty.
After 4 days, the number of neurons intensely stained for
NADPH-diaphorase as well as those giving a NOS I signal in in
situ hybridization experiments increased in castrated rats
treated with testosterone by 31% and 42%, respectively, relative to
those in untreated castrated rats. This suggests that the increase in
NADPH-diaphorase activity resulted from enzyme synthesis and was due to
a modification of NOS I messenger RNA (mRNA) accumulation. After 7
days, Northern blot analysis showed that castration produced a decrease
in the amount of NOS I mRNA relative to that of ribosomal RNA. This
decrease was almost prevented by T treatment. No significant
differences were observed by reverse transcriptase-PCR between 7-day
and 28-day treatments. However, in 7-day castrated rats treated with
5DHT, NOS I signals relative to those of hypoxanthine
phosphoribosyltransferase, taken as reference, were significantly
higher than those in castrated rats and resembled those in
sham-castrated rats, suggesting that 5DHT was probably more potent
than testosterone in preventing the decrease in NOS I mRNA levels
elicited by castration. These results show that NOS I can be positively
regulated by androgens and are consistent with the suggestion that
these steroids play a role in the physiological processes of penile
erection. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NOS I is widely distributed in the central and peripheral as well as enteric nervous systems (2, 3, 4). It induces NO synthesis, which is strongly involved in the autonomic control of the urogenital tract, including the structures required for male reproductive function. For example, a key mechanism in penile erection is smooth muscle relaxation of the penile arterial bed and the walls of cavernous spaces. This relaxation leads to an increase in penile blood flow, resulting in blood engorgement of the corpora cavernosa. This engorgement causes tumescence and then penile rigidity (5). Many pharmacological and immunohistological data in man and animals have identified NO from a neural source as the major physiological mediator of penile erection (5, 6).
Castration induces a marked reduction in androgen levels and leads to a
reduction in erectile function, which is reliably prevented or reversed
by the administration of testosterone (T) or its active metabolite,
5-dihydrotestosterone (5DHT) (7). Androgens may exert their
effects by acting directly on penile intrinsic and extrinsic
musculature or on specific nuclei in the brain and lumbosacral spinal
cord that contain neurons involved in the control of sexual functions
(7). However, androgens may also act through alternative structures,
such as neurons of the pelvic plexus, designated the major pelvic
ganglion (MPG) in the rat. In a previous study on the modulation of
penile erection by T, we hypothesized that one probable site of action
of androgens was situated peripherally to the spinal cord, on MPG
neurons innervating the corpora cavernosa (8). Validation of such a
hypothesis, however, requires the demonstration of two prerequisites.
The first is that the androgen receptor must be expressed within
NOS-containing neurons of the MPGs. The second is that NOS I, which is
considered a purely constitutive enzyme, be regulatable. Localization
of the androgen receptor in nitric oxide synthase-containing neurons of
the MPG has been recently demonstrated (9). On the other hand, there
are some indications for expressional regulation of NOS I in neurons of
the central nervous system, especially by sex steroids (10, 11).
We, therefore, decided to investigate whether T and 5DHT could
regulate NOS I gene expression in MPG neurons. NADPH-d activity and NOS
I messenger RNA (mRNA) levels were thus evaluated in MPG neurons of
sham-castrated rats and castrated rats supplemented or not with
androgen. The results showed that 1) in castrated rats, both NADPH-d
activity and NOS I mRNA levels are increased by T supplementation; 2)
NOS I mRNA levels are significantly decreased by castration; and 3)
androgen replacement at the time of surgery precludes this change;
5DHT is seeming more efficient than T in preventing this
reduction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DHT (Sigma) were
implanted sc in the upper back region for 4, 7, and 28 days. These
SILASTIC capsules have been shown to maintain physiological serum T or
DHT levels in adult castrated rats (12, 13). Several groups of five
rats each, were designated and treated as follows: group 1, rats were
sham castrated and treated with empty capsules (intact controls); group
2, rats were castrated and treated with empty capsules (castrated
controls); group 3, rats were castrated and treated with T (2.2-cm
length capsule); and group 4, rats were castrated and treated with
5DHT (1-cm length capsule). These groups of rats were used in three sets of experiments. For histochemical reaction and in situ hybridizations, rats from groups 2 and 3 were treated as described above for 4 days. After deep anesthesia (50–60 mg/kg, ip), both MPGs were quickly removed and frozen on dry ice. The rats then were killed. Frozen MPGs were cut in a cryostat in serial sections of 8 or 10 µm, thaw-mounted onto slides, and stored at -80 C until used. For isolation and measurement of relative levels of NOS I mRNA, rats from all groups were treated as described above for 7 or 28 days. After deep anesthesia, the two MPGs were quickly removed, placed in Nunc cryotubes (Polylabo, Strasbourg, France), frozen in liquid nitrogen, and then stored at -80 C. As a tissue control, the cerebellum and seminal vesicles were treated in the same way. Rats were then killed.
NADPH-diaphorase (NADPH-d) histochemistry
Mounted serial sections of both MPGs were defrosted, cold
air-dried, and fixed in 4% paraformaldehyde-0.1 M sodium
phosphate buffer (PB), pH 7.4, for 5 min. They were thoroughly rinsed
in PB, then processed on slides for NADPH-d activity as previously
described (14). Sections were incubated in 0.1 M Tris-HCl,
pH 8.0, containing 0.3% Triton X-100, 0.5 mg/ml nitro blue tetrazolium
(Sigma) and 1.0 mg/ml ß-NADPH (reduced form; Sigma) at 37 C for 45
min in a moist chamber. As a control, some incubations were carried out
in the absence of ß-NADPH or nitro blue tetrazolium. At the end of
incubation, the sections were rinsed in PB, air-dried, and
coverslipped. Neurons intensely stained for NADPH-d were counted in
every fourth section of both MPGs of each animal from groups 2 and 3
(6–10 sections/rat). Cell counts are given as the average number of
cell profiles per section, and the results are not corrected for
possible double counting. All neuronal staining was absent in control
incubations.
In situ hybridization
For in situ hybridization of NOS I mRNA, a synthetic
45-mer oligodeoxynucleotide complementary to rat NOS I mRNA
corresponding to amino acids 151–164 of the rat NOS I protein (15) was
synthesized (Appligene, Illkirch, France). The probe was 3'end-labeled
with [-33P]deoxy-ATP (SA, 1000–3000 Ci/mmol;
Isotopchim, Ganagobie, France) using terminal deoxynucleotidyl
transferase (Boehringer Mannheim, Meylan, France) and then purified on
a minicolumn (NACS PREPAC, Life Technologies, Cergy-Pontoise, France).
In situ hybridization was performed as follows. Mounted
sections were defrosted, cold air-dried, fixed with 4%
paraformaldehyde in 1 x PBS (2.6 mM KCl, 1.4
mM KH2PO4, 136 mM NaCl,
and 8 mM Na2HPO4, pH 7.4) for 20
min at room temperature and sequentially rinsed for 5 min at room
temperature, once in 3 x PBS, and twice in 1 x PBS. The
fixed sections were then incubated in a solution of predigested pronase
(2.3 U/ml) for 10 min; proteolytic activity was stopped by immersing
the sections for 30 sec in 1 x PBS containing glycine (2 mg/ml)
and then in 1 x PBS. Sections were then dehydrated through 60%,
80%, and 95% ethanol to absolute alcohol (5 min each) and then left
to air-dry at room temperature. Finally, sections were hybridized
overnight at 42 C in a hybridization buffer [50% deionized formamide,
10% dextran sulfate, 0.6 M NaCl, 1 x Denhardt’s
solution, 1 mM EDTA, 10 mM Tris-HCl (pH 7.4),
and 1.2 mg/ml yeast transfer RNA] containing the
33P-labeled oligonucleotide probe (2.5–4.5 x
106 cpm/pmol) at a final concentration of 1 pmol/ml. After
hybridization, sections were washed four times (1 h each) in a buffer
containing 0.6 M NaCl, 10 mM Tris-HCl (pH 7.4),
and 1 mM EDTA at 50 C. After dehydration in 70%
ethanol-300 mM ammonium acetate (5 min) and then in 95%
ethanol-300 mM ammonium acetate (5 min), sections were left
to air-dry before being apposed to Hyperfilm ß-Max (Amersham, Les
Ulis, France) for 8 days at -80 C. After Hyperfilm autoradiography,
sections were dipped in water-diluted (1:1) NTB2 emulsion (Eastman
Kodak, Rochester, NY) and exposed for an additional 4 weeks in
light-tight boxes at 4 C to visualize NOS I cellular signal. Films and
sections were developed in Kodak D19 (5 min at 20 C) and fixed in
Illford Hypam (5 min). After developing, sections were counterstained
lightly with cresyl violet, dehydrated, and coverslipped in DePeX (BDH
Ltd., Poole, UK). Signal specificity was assessed by increasing the
washing temperature, by ribonuclease treatment of sections before
hybridization, and by competing the radiolabeled probe with a 100-fold
excess of unlabeled specific oligonucleotide. In all control
experiments the hybridization signal was suppressed.
Changes in the relative amounts of mRNA in the MPGs were assessed by measuring the mean optical density (OD) of autoradiographic films using a mrag image analysis system (Biocom, les Ulis, France). First, the outlines of both MPG sections from the same animal were drawn from the corresponding stained sections, then the mean optical densities of these areas were determined on the film. The mean ODs of two or three areas outside the sections were also measured, and these background values were subtracted from the value of each MPG section. The mean ODs of all MPG sections were determined and for each animal mean OD values are presented as the average mean OD per section.
When NADPH-d histochemistry was combined with in situ hybridization, the histochemical reaction was performed first, then the sections were hybridized.
RNA isolation and Northern blot analysis
Total RNA was isolated from MPGs, cerebellum, and seminal
vesicles by a guanidium-thiocyanate-phenol-chloroform extraction, as
previously described (16). Glycogen (35 µg; Appligene Oncor,
Illkirch, France) was introduced into MPG RNA. All RNA samples were
analyzed by Northern blot in 1.5% (wt/vol) agarose gels containing 2.2
M formaldehyde (17). Fractionated RNA was denatured with 50
mM NaOH, 50 mM sodium phosphate (18),
blot-transferred to nylon membranes (Biohylon Z+, Bioprobe Systems,
Montreuil, France) as described by the manufacturer, and UV
autocross-linked. Equal loading and transfer of seminal vesicle and
cerebellum RNA were checked by ethidium bromide staining. Membranes
were prehybridized for at least 2 h at 65 C in 0.5 M
phosphate buffer (pH 7.2), 1 mM EDTA, and 7% SDS (19).
Hybridization was carried out overnight at 65 C in the same buffer
containing labeled cDNA probe. Membranes were sequentially hybridized
with the NOS I probe, with a glyceraldehyde 3-phosphate dehydrogenase
(GAPDH) probe used as an internal control for mRNA quality, and
with an 18S probe as a control for equal loading and transfer of RNA.
DNA probes (1 x 106 to 3 x 106
cpm/ml; SA, 1.5 x 109 cpm/pg) were labeled with
[-32P]deoxy-CTP (SA, 3000 Ci/mmol; Amersham, les Ulis,
France) by random priming (20). The following probes were used: a
5057-bp fragment isolated by EcoRI digestion of a NOS I
plasmid (15), a 1300-bp fragment isolated by PstI digestion
of a rat GAPDH plasmid (21), a 1875-bp fragment isolated by
SalI and EcoRI digestion of a mouse 18S ribosomal
plasmid (22), and a 510-bp fragment amplified from seminal vesicle
total RNA with primers 5'-CGTCGGGTTTAGGAATCTC-3' and
5'-GACAGAAGCAGCCGCAAGGA-3' corresponding to rat seminal vesicle
secretion VI (SVS VI) cDNA as previously described (23). Hybridization
signals were quantified with an ImageQuant PhosphorImager (Molecular
Dynamics, Sunnyvale, CA). Alternatively, membranes were exposed to
x-ray films (MP films, Amersham) at -80 C in the presence of two
intensifying screens. The resulting autoradiographs were scanned using
the ImageMaster 1-D (Pharmacia Biotech, Orsay, France). Results of
hybridizations with GAPDH and 18S ribosomal RNA (rRNA) probes were used
for normalization.
Reverse transcription-PCR (RT-PCR)
Total RNAs from both MPGs from each rat were dissolved in
50 µl distilled water, and cDNA was synthesized using 6 µl RNA in a
20-µl reaction for 60 min at 37 C, using 500 ng
p(dT)12–18 primer (Pharmacia Biotech, Orsay, France), 5 U
ribonuclease inhibitor (Life Technologies, Cergy-Pontoise, France), and
100 U Moloney murine leukemia virus reverse transcriptase (Life
Technologies) as recommended by the manufacturer (RT+ reaction). DNA
contamination of RNA was evaluated in similar reactions performed
without reverse transcriptase (RT- reaction). RT was followed by a
5-min incubation at 94 C. PCRs were then carried out on 2 µl RT+ as
well as 2 µl RT- products, in the presence of primers specific for
rat NOS I cDNA (product size, 440 bp) or primers specific for rat
hypoxanthine phosphoribosyltransferase (HPRT) cDNA (product size, 329
bp; as internal control) and in the presence of 0.5U of Taq
polymerase (Appligene Oncor) as recommended by the manufacturer. The
NOS I-specific PCR was performed for 25 amplification cycles at 95 C
for 1 min, 55 C for 1 min, and 72 C for 1 min in the presence of 25
pmol of each of the following primers: (5'-CACGTGGTCCTCATTCTGAG-3') and
(5'-TCTCTGTCCACCTGGATTCC-3'). These primers were actually described as
NOS I specific in human NOS I (22), but were totally conserved in rat
NOS I. The HPRT-specific PCR was performed for 28 cycles at 95 C for 1
min, 60 C for 1 min, and 72 C for 1 min with 20 pmol of each of the
following primers (5'-CCTGCTGGATTACATTAAAGCACTG-3') and
5'-GTCAAGGGCATATCCAACAACAAAC-3'), as previously described (24). In both
PCRs, the last cycle was followed by a 10-min incubation at 72 C.
RT-PCR products (15 µl of each reaction) were subjected to
electrophoresis on 1% agarose gels, and signals were quantified with
the ImageMaster 1-D (Pharmacia Biotech, Orsay, France). The ratios of
NOS I to HPRT signals were calculated for each individual rat and
expressed as a percentage of the mean value obtained for sham-operated
rats. The amount of input RNA and the number of amplification cycles
were previously optimized in this laboratory to ensure that the PCR
products were quantified during the exponential phase of amplification
(25). No signals were observed in the control PCR reactions when
reverse transcriptase was omitted in RT reactions. Signals were
observed in almost all RT+ reactions. The linearity of PCR reactions
for an increasing amount of RT products and an increasing number of
cycles was checked as well as the reproducibility of PCR reactions from
RT products.
Statistical analysis
Results are the mean ± SEM of values obtained,
unless otherwise specified, from groups of five rats. The significance
of differences was determined with two-tailed unpaired Student’s
t test. Statistical significance was established at
P 
0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Some neurons were intensely stained, whereas others
showed moderate, light, or no staining. Staining intensity and
diaphorase activity can fully reflect the NADPH-d content of each
neuron (26, 27), as it has been shown that neuronal NADPH-d and NOS I
are identical (26, 28). Thus, these differences in staining intensity
may reflect variable intracellular concentrations of NOS I, possibly
resulting from the difference in the NOS I synthesis rate. To study NOS
I synthesis in the MPG neurons, in situ hybridization of a specific NOS
I probe was performed. NOS I mRNA was not detected in lightly stained
NADPH-d neurons. This light staining was, therefore, not considered to
be due to NOS I synthesis in MPG neurons. Only intensely stained and a
proportion of moderately stained NADPH-d neurons contained a
substantial amount of NOS I mRNA (Fig. 1B), demonstrating NOS I
synthesis in these neurons.
|
). To confirm the nature of this mRNA, total
RNAs from cerebellum and MPGs were reverse transcribed, and the
resulting cDNAs were amplified by PCR with primers specific for NOS I.
A PCR product of 440 nucleotides was observed in the cerebellum as well
as in MPG RT-PCR reactions (Fig. 3). The PCR product
from MPGs was sequenced. The sequence corresponded to the rat NOS I
published sequence, from nucleotides 799-1020 (15). The RNA detected by
Northern blotting with the NOS I probe thus is NOS I mRNA. The signal
detected for NOS I mRNA was slightly stronger in the amount of total
RNAs extracted from MPGs than the signal detected in 5 µg rat
cerebellum total RNAs (Fig. 2A), whereas 18S signals were equivalent
for both tissues (Fig. 2C). Interestingly, the signal detected for the
same blot with the GAPDH probe was slightly weaker in MPGs than in
cerebellum (Fig. 2B). NOS I mRNA levels were twice as high in MPG as in
cerebellum when treating the 18S rRNA level as a standard and were even
more different when considering the GAPDH mRNA level as a standard. The
NOS I/GAPDH ratio, calculated using results from three independent
experiments, was 3- to 6-fold higher in MPGs than in the cerebellum. We
concluded that NOS I mRNA, compared with 18S rRNA or GAPDH mRNA, was
expressed more in MPG than in cerebellum.
|
|
). This increase in the number of intensely stained
neurons was seen in both small and large neurons throughout the MPGs.
It reflected a T-dependent increase in NOS I protein resulting from
increased NOS I synthesis, as shown by in situ hybridization
studies which revealed a significant increase (42%; P
< 0.01) in the relative amount of NOS I mRNA in the MPGs of castrated
rats supplemented with T compared with the relative amount of NOS I
mRNA signal in these ganglia in castrated rats (Figs. 4, A–F, and 5).
|
|
). After correction relative to 18S rRNA, NOS I
mRNA levels were significantly lower in castrated rats (69%) compared
with those in sham-castrated rats (P < 0.01) and were
maintained at the level in sham-castrated rats by T supplementation
(Fig. 6B). Similar variations in NOS I mRNA levels among the three
groups of rats were seen when NOS I mRNA levels were quantified
relative to GAPDH mRNA (Fig. 6D), but as GAPDH mRNA levels were
significantly decreased by castration (38%) compared with levels in
sham-castrated rats (P < 0.05; Fig. 6C) and maintained
at the level in sham-castrated rats by T treatment (Fig. 6C), the
corrected values of NOS I mRNA in castrated and T-supplemented rats
failed to differ statistically (Fig. 6D).
|
DHT) on NOS I mRNA levels in
MPG neurons were also analyzed in another set of experiments. The
amounts of total RNA extracted from the MPGs of rats that were either
sham castrated or castrated and supplemented or not with T or 5DHT
were analyzed by RT-PCR with NOS I-specific primers as described above,
and results were compared on postoperative days 7 and 28. Variations in
the amounts of NOS I-amplified products were corrected for the amount
of RNA introduced in RT-PCR reactions by amplifying a housekeeping gene
mRNA. GAPDH was first used, but in some samples, RT- reactions gave
signals similar to those obtained from RT+ reactions. This amplified
product was probably due to amplification of a GAPDH pseudogene present
in the rat genomic DNA that contaminated some samples (29). As the
amount of RNA extracted from both MPGs isolated from single rats was
very low, instead of eliminating DNA from these samples, we preferred
to standardize PCR reactions by evaluating the gene expression of
another housekeeping gene, HPRT. Under this condition, no signals were
observed in RT- reactions (data not shown).
HPRT signals varied among individual rats, but did not differ
significantly between the groups of rats. On postoperative day 28, NOS
I signals were equivalent to HPRT signals in sham-castrated rats, they
were relatively weaker than HPRT signals in castrated rats, and they
were almost equivalent to HPRT levels in castrated rats supplemented
with T (Fig. 7A). These variations confirm the above
results and those obtained by Northern analysis on postoperative day 7
for NOS I mRNA levels relative to GAPDH (Fig. 6A
). Signals were
quantified and NOS I signals evaluated relative to HPRT for individual
samples. No significant differences were observed between 7- and 28-day
treatments. Whereas in both cases, castration could significantly
reduce NOS I relative signals, in neither case could T supplementation
maintain these levels at those observed in sham-castrated rats (Fig. 7B). However, in 7-day castrated rats treated with 5DHT (28-day
castrated treated rats were not tested), NOS I relative signals were
significantly higher (P < 0.05) than those in
castrated rats (Fig. 8) and showed no difference from
those in sham-castrated rats, suggesting that 5DHT was the efficient
androgen in preventing the decrease in NOS I mRNA levels that followed
castration.
|
|
, A and B, demonstrate that the expression of SVS VI
mRNA dramatically decreased by 7 days after castration (and even more
so by 28 days; data not shown), whereas it was maintained at the value
in sham-castrated rats by both T and 5DHT (data not shown)
treatments with equal efficiency.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DHT; the latter compound seems to
be more effective than T in this respect. This regulation of the
enzyme, which was previously considered constitutive, has been shown to
occur in sex steroid-concentrating neurons in the central nervous
system after castration followed by hormonal supplementation (10, 11, 30). We have demonstrated that this regulation also occurs in the
autonomic neurons of the pelvic plexus. NADPH-d activity has been extensively characterized in the central and peripheral nervous systems, where it colocalizes with NOS (26, 28). In this histochemical staining, NOS oxidative activity in fixed tissue sections reduces nitro blue tetrazolium, the electron acceptor in the presence of NADPH, forming an insoluble blue purple cytoplasmic precipitate in reactive neurons. In fact, NADPH-d staining can be elicited by any NADPH-dependent oxidative enzyme that survives fixation (31). Moreover, all isoforms of NOS possess NADPH-d activity (32), and isoforms other than NOS I can be expressed in neurons (1). Thus, it was important to verify that the observed staining was due to the presence of NOS I. The combination of NADPH-d histochemistry with in situ hybridization of a synthetic oligonucleotide probe against NOS I mRNA demonstrates clearly that only intensely (and a percentage of moderately) stained neurons contain NOS I mRNA. Similar results were found when NADPH-d histochemistry was combined with NOS I immunocytochemistry using rabbit antibodies (gift from P. Alm, K. E. Andersson, and B. Larsson) produced against a specific sequence of the rat cloned cerebellar NOS; an almost one to one colocalization of strong immunostaining and intensely NADPH-d staining in MPG neurons was observed (3) (Schirar, A., unpublished results). Thus, these combined studies prove conclusively the well founded suggestion of Santer and Symons (33) that (at least in the MPG) only the intensely stained NADPH-d neurons are involved in NOS I expression. On the basis of the identical pattern of staining produced by NOS I antibodies and the intense NADPH-d staining in MPG neurons, nerve cell bodies stained by either method were referred to as NOS I-containing neurons. However, to conserve the limited amounts of antibodies available against NOS I, we preferred, in the course of this study, to use the quick histochemical method rather than a long, expensive, immunohistochemical procedure.
Northern blotting showed that NOS I expression was higher in MPG neurons than in cerebellum. This finding is original. However, there is nothing surprising about it if one considers, that 1) the MPGs contain a large proportion of nitrergic neurons and nerve fibers innervating most of the pelvic and genital organs (34, 35); and 2) NO is strongly involved in the functioning of the different parts of the urogenital tract (5, 36). A recent study reported that the only NOS mRNA expressed in the rat penis was a penile NOS I (pNOS) mRNA differing from the cerebellar NOS I mRNA by the presence of a stretch of nucleotides resulting from the transcription of part of intron 16. This new species of pNOS mRNA was claimed to be poorly expressed in pelvic plexus (37). Our Northern analysis and RT-PCR, using the same primers (NO1 and NO2; data not shown) (37) bordering the cryptic exon, confirmed that the major species of NOS I mRNA found in the MPG was not the pNOS mRNA recently described, but a species of mRNA that we could not distinguish from that of the NOS I cloned originally from the rat cerebellum.
We found that cellular NADPH-d staining in MPG of castrated rats changed markedly with T treatment. Others have shown that the number of NADPH-d-stained neural fibers in the rat corpora cavernosa was significantly lowered by castration and increased by T supplementation (38). As a large proportion of these fibers have their cell bodies in the MPG (14), our results are consistent with their reports. The conclusion of our combined histochemical and in situ hybridization studies suggests that increased cellular diaphorase staining after T treatment is a consequence of up-regulated NOS I synthesis. Diaphorase staining intensity has been shown to reflect the NOS content of each neuron (26). Moreover, an increase in NADPH-d staining (11) and a parallel increase in NOS I immunoreactivity (30) or NOS activity (10) and NOS I mRNA levels in response to T (10) or estrogen (10, 11, 30) supplementation have been previously reported in different locations of the central nervous system. The present results suggest that androgens can modulate NOS I in autonomic neurons of the rat MPG. Furthermore, these post ganglionic neurons that innervate the pelvic organs, including the urogenital tract, have been previously shown to contain a population of androgen-concentrating neurons exhibiting NOS immunoreactivity (9). Thus, the results of the present study are consistent with an effect of androgens in regulating NOS I mRNA in the MPG.
We clearly showed that castration was followed by a significant
reduction of NOS I mRNA in the MPG neurons. It is not, however,
possible to determine at this stage whether the decrease in mRNA is
caused by a decreased rate of transcription of the NOS I gene or
altered stabilization of NOS I mRNA. In the same way it is not possible
from the present results to determine whether androgen treatment
results in enzyme induction. We did not address these issues
specifically. The failure of the NOS I/GAPDH mRNAs ratio to be
maintained at steady state levels (i.e. those in
sham-castrated rats) in castrated rats treated with T was not due to
insufficient serum T levels. Others have shown that SILASTIC capsules
similar to those used in this study were able to restore physiological
serum T or 5DHT levels in castrated rats (12, 13). Moreover, we
checked the efficiency of hormonal treatment by using rat seminal
vesicles, which are an excellent model for study of the mechanisms of
androgen action. These vesicles contain many secretory proteins that
have been used extensively as markers for androgen action (39). In the
present study, the SILASTIC capsules delivering either T or 5DHT
were able to prevent (7 as well as 28 days after castration) the
dramatic decrease in SVS VI mRNA levels that followed castration,
demonstrating that these capsules operated well and that they released
physiological amounts of gonadal steroid. In fact, GAPDH mRNA,
initially taken as an internal standard, varied with castration and
hormonal supplementation in the MPG, but not in the seminal vesicles.
GAPDH is a key enzyme in the control of glycolysis, and its gene is
considered a housekeeping gene. It is well recognized that these genes
have a continuous, low rate of transcription, but this does not rule
out possible regulation. Such regulation could exist in MPG
neurons.
When HPRT mRNA was used as an internal standard, no significant
variation in HPRT signals was seen among the individual samples of MPG
or seminal vesicles extracted from the three experimental groups of
rats (Figs. 7 and 9, A and B). The NOS I/HPRT mRNAs ratio was
maintained at steady state levels in castrated rats supplemented with
5DHT, but not in rats receiving T supplementation. It is well
established that in the central nervous system and in seminal vesicles,
the active compound is not T on its own, but, rather, its 5-reduced
metabolite, 5DHT. It is likely that 5DHT is more effective in
increasing NOS I mRNA levels; however, based on the present data, this
assertion cannot be firmly established in the absence of administration
of multiple doses of androgens and assessment of plasma steroid
levels.
Finally, the present results suggest that androgen may regulate NOS I
synthesis and thus may control the generation of NO in the MPG neurons
and their terminals. We have previously demonstrated that an androgen
receptor is present in nitrergic MPG neurons, providing anatomical
evidence for the direct association of androgen with NOS I activity and
synthesis (9). This regulation may be of physiological importance,
especially in maintenance of the neurohemodynamic mechanisms
responsible for penile erection. Penile erection is a phenomenon that
is androgen dependent. This androgen dependency has been previously
demonstrated in our (8) and other laboratories (40, 41, 42, 43) by showing that
the penile erectile response to electrical stimulation of the cavernous
nerve is reduced by castration. This change is believed to be dependent
on androgen because the impairment of the erectile response is
prevented if the castrated rats are given androgen (T or 5DHT) at
the time of surgery. Furthermore, we have previously shown that the
effect of T or 5DHT is specific for androgen (and not a general
property of all sex steroids) because 17ß-estradiol given to
castrated rats does not prevent the decrease in erectile response to
electrostimulation (45). Besides, recent evidence has shown that
androgen withdrawal by castration is accompanied by a decrease in both
NOS I protein and activity within the rat penis and that these changes
are prevented if androgen is supplemented at the time of surgery (42, 44). Our results point to the possibility that androgens, especially
5DHT (which is the active androgen in the prevention of erectile
failure seen in castrated rats) (42), by stimulating neuronal NOS
synthesis, might allow strong NO release in the penis and thus play a
role in the physiological processes of penile erection. Moreover,
androgens have a trophic effect on the urogenital tract. Growth factor,
such as nerve growth factor, produced in pelvic and genital organs may
respond to changes in androgen levels and influence NOS I biosynthesis
in MPG neurons and the release of NO in nerve terminals in the penis in
an indirect manner. However, it is unlikely that such a mechanism may
operate over the short term.
In females of different species (rat, guinea pig, and human), all three known NOS isoforms (i.e. NOS I, NOS II, and NOS III) are present in the uterus and its vasculature. Differential regulation of NO has been shown to occur in the uterus and cervix during gestation and labor. Interestingly, it appears likely that in both the pregnant and the nonpregnant state, the regulation of NO-generating mechanisms by the uterus is modulated by steroid hormones, but the exact mechanisms controlling the differential expression of NOS enzymes are still unknown (46).
In conclusion, changes in NADPH-d activity and NOS I mRNA levels have been reported in central, peripheral, and autonomic neurons (this study) not only in response to sex steroid treatment (Refs. 10, 30, 46, and 47 and this study), but also in response to axotomy (48, 49) and salt loading (50). The finding of a transcriptional induction of the constitutive NOS I by various stimuli suggests that the present classifications of NOSs, as described in the introduction, are imprecise and need to be reconsidered (1, 10).
|
|
Acknowledgments |
|---|
Received February 17, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-dihydrotestosterone. Int J Androl 15:355–364[Medline]
-réductase pour le
comportement sexuel masculin. Andrologie 5:230–235
This article has been cited by other articles:
 |
|
 |
|
  V. Rochira, A. Balestrieri, B. Madeo, A. R. M. Granata, and C. Carani Sildenafil Improves Sleep-Related Erections in Hypogonadal Men: Evidence From a Randomized, Placebo-Controlled, Crossover Study of a Synergic Role for Both Testosterone and Sildenafil on Penile Erections J Androl, March 1, 2006; 27(2): 165 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Secondo, A. Pannaccione, M. Cataldi, R. Sirabella, L. Formisano, G. Di Renzo, and L. Annunziato Nitric oxide induces [Ca2+]i oscillations in pituitary GH3 cells: involvement of IDR and ERG K+ currents Am J Physiol Cell Physiol, January 1, 2006; 290(1): C233 - C243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Traish, P. Toselli, S.-J. Jeong, and N. N. Kim Adipocyte Accumulation in Penile Corpus Cavernosum of the Orchiectomized Rabbit: A Potential Mechanism for Veno-occlusive Dysfunction in Androgen Deficiency J Androl, March 1, 2005; 26(2): 242 - 248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Burnett Neurophysiology of Erectile Function: Androgenic Effects J Androl, November 1, 2003; 24(6_suppl): S2 - S5. [Full Text] [PDF]
|
|
|
|
|
|
M. E. DiSanto Corpus Cavernosum Smooth Muscle Physiology: A Role for Sex Hormones? J Androl, November 1, 2003; 24(6_suppl): S6 - S16. [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Wingard, J. A. Johnson, A. Holmes, and A. Prikosh Improved erectile function after Rho-kinase inhibition in a rat castrate model of erectile dysfunction Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2003; 284(6): R1572 - R1579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Traish, R. Munarriz, L. O'Connell, S. Choi, S. W. Kim, N. N. Kim, Y.-H. Huang, and I. Goldstein Effects of Medical or Surgical Castration on Erectile Function in an Animal Model J Androl, May 1, 2003; 24(3): 381 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. K. Bachir, J.-N. Laverriere, and R. Counis Isolation and Characterization of a Rat Nitric Oxide Synthase Type I Gene Promoter that Confers Expression and Regulation in Pituitary Gonadotrope Cells Endocrinology, November 1, 2001; 142(11): 4631 - 4642. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-C. Lacroix, E. Devinoy, S. Cassy, J.-L. Servely, M. Vidaud, and G. Kann Expression of Growth Hormone and Its Receptor in the Placental and Feto-Maternal Environment during Early Pregnancy in Sheep Endocrinology, December 1, 1999; 140(12): 5587 - 5597. [Abstract] [Full Text]
|
|
|
|
|
|
A. M. Traish, K. Park, V. Dhir, N. N. Kim, R. B. Moreland, and I. Goldstein Effects of Castration and Androgen Replacement on Erectile Function in a Rabbit Model Endocrinology, April 1, 1999; 140(4): 1861 - 1868. |
