| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Urology, UCLA School of Medicine, Harbor-UCLA Medical Center, Torrance, California 90509
b Department of Urology, The Johns Hopkins Hospital, Baltimore, Maryland 21205
ABSTRACT
Penile erection is mediated by nitric oxide (NO) synthesized by the neuronal nitric oxide synthase (nNOS). In the rat penis, the main nNOS mRNA variant, PnNOS, differs from cerebellar nNOS (CnNOS) by a 102 base pair insert encoding a 34-amino acid sequence. In the mouse, two nNOS mRNAs have been identified: nNOS
, encoding a 155-kDa protein, and an exon 2-deletion variant, nNOSß, encoding a 135-kDa protein that lacks a domain where a protein inhibitor of nNOS (PIN) binds. We wished to determine whether PnNOS and ß are expressed in the rat penis and are located in the nerves and whether the ß form persists in the potent nNOS knock-out mouse (nNOS
). A PnNOS antibody against the insert common to both PnNOS and ß detected the expected 155-kDa protein in PnNOS-transfected cells. This antibody, and the one common to PnNOS/CnNOS, showed (on Western blots) the 155- and 135-kDa nNOS variants in rat penile tissue during development and aging. PnNOS mRNA and its subvariants were found as the main nNOS in the penile corpora, the cavernosal nerve, and the pelvic ganglia, with lower levels of PnNOSß mRNA. In tissue sections, PnNOS protein was immunodetected in the penile nerve endings in the rat and in the nNOS wild-type and nNOS mice. An antibody against the sequence encoded by exon 2 did not react (on Western blots) with the 135-kDa band, which confirms that this protein is the ß form. In conclusion, both PnNOS and ß are expressed in the rat penis at all ages and are located in the nerves. The ß form may allow nitric oxide synthesis during erection to be partially insensitive to PIN. The residual expression of PnNOS, and possibly CnNOS, in the penis of the nNOS mouse occurs through transcription of the ß mRNA, and this may explain the retention of erectile function when the expression of nNOS is disrupted.
gene regulation, glutamate, male sexual function, nitric oxide, penis
INTRODUCTION
The neuronal nitric oxide synthase (nNOS) is the main NOS isoform responsible for the synthesis of the physiological mediator of penile erection, nitric oxide (NO) [1–4]. Antibodies against peptide sequences encoded by the nNOS cDNA cloned from rat and human cerebellum (CnNOS) have allowed the immunohistochemical detection of nNOS protein in the innervation of the rat, human, and mouse penis and in the pelvic ganglion, where their neuronal bodies are located [5–10]. The nerve endings were stained in the corpora cavernosa. Western blot assays of extracts from the rat and human penis with these antibodies have detected a 155-kDa band, the size of which is compatible with the amino acid sequence encoded by the nNOS mRNA, as well as a smaller 135-kDa band of unknown origin [11–16].
The stimulation of nNOS in the nonadrenergic-noncholinergic postsynaptic nerve terminals of the penis [17] appears to be the process that triggers penile erection, either as a result of a sexually evoked central stimulus or through a direct electrical stimulation of the cavernosal nerve [1–4, 11–21]. A similar NO-dependent mechanism is involved in the central control of the erectile and ejaculatory responses in the paraventricular nucleus and medial preoptic area of the hypothalamus and in the sacral spinal cord [22, 23]. Therefore, the regulation of nNOS enzyme activity through cycles of activation and inhibition of NO synthesis is crucial for male reproductive function. In the brain and other tissues, nNOS activation occurs through Ca2+/calmodulin binding close to the active site of the enzyme. This response may be inhibited or stimulated by the binding of additional factors [4], such as the protein inhibitor of NOS (PIN) [24, 25], the NMDA receptor (NMDAR) [26], or the carboxy-terminal PDZ ligand of nNOS (CAPON) [27], to an N-terminal PDZ sequence in nNOS encoded by exon 2 [28]. These nNOS modulators are also expressed in the rat penis ([29, 30], and unpublished data, respectively).
The regulation of nNOS activity is to some extent organ specific, since it may depend on the type of transcriptional and posttranscriptional processing affecting nNOS expression in the central nervous system versus peripheral neural tissue or in the skeletal muscle. Several alternative splicing and promoter usage variants of nNOS mRNA have been identified, mainly in the skeletal muscle and the brain [31–35]. At least two of these variants, nNOSß and nNOS
, lack the region encoded by exon 2, a region that is present in the full-length transcript nNOS. Although the form is enzymatically inactive, the recombinant ß variant is active. However, this form cannot be modulated by PIN or CAPON, and it potentially cannot be modulated by any other factor associated with the missing region containing the PDZ domain. Both the and ß forms can be expressed as mRNA subvariants according to the site of transcription initiation upstream of exon 1 [36], although single nNOS and ß proteins are predicted to result from the three subvariant mRNAs. No nNOSß or mRNA could be detected in different rat tissues [36].
Because of the lack of exon 2, transcription of an mRNA encoding the nNOSß protein in the nNOS knock-out mouse is theoretically possible through bypass of the blockade at exon 1 that impedes translation into the form [37]. Alternative splicing would then lead to the expression of the shorter but functional nNOSß protein. In the case of the penis, the presence of this nNOS variant would help to explain why the nNOS mouse is fertile, since an adequate erectile function would be sustained by NO synthesis generated by nNOSß [37–39], with or without a putative up-regulation of endothelial NOS (eNOS). Supporting this assumption, both nNOSß mRNA and a fully active 135-kDa protein supposedly arising from this transcript and have been detected in the brain of the nNOS mouse [28]. The nNOSß mRNA has also been found in the penis from this animal [38, 40], but no detection of the penile nNOSß protein has so far been reported. The evidence in the rat is confusing, since no nNOSß mRNA could be found in several tissues [36], despite the fact that the abundant 135-kDa nNOS protein seen on Western blots in penile tissue is of the size expected for nNOSß [11–16].
Another potentially important variant of nNOS cDNA has been cloned from the rat penis [41]; this variant is termed penile nNOS (PnNOS). It differs from the CnNOS cDNA based on the presence of a 102-base pair (bp) insert between exons 16 and 17 that encodes an "in frame" 34-amino acid sequence. Compared to CnNOS, PnNOS has several base sequence changes. PnNOS mRNA appears to be the only nNOS form present in the rat penis, since only a little CnNOS mRNA can be detected by reverse transcription-polymerase chain reaction (RT-PCR), but both variants are expressed in the rat pelvic ganglion. The rat PnNOS sequence insert is identical to the one present in mouse nNOSµ mRNA, a variant found in skeletal and cardiac muscle [42], but it differs by a few amino acids from the nNOSµ found in the human penis [43]. It is not known whether PnNOS and nNOSµ are derived from the same gene as identical splicing products with some base polymorphisms or whether they originate from different genes. The 34-amino acid stretch contains a putative phosphorylation/regulatory site, and it is adjacent to the autoinhibitory region of nNOS [44]. Although the function of this insert is unknown, it is assumed to confer some type of specificity to the modulation of nNOS activity in the penis and therefore to be important for the control of penile erection. The nNOSµ protein has been detected in skeletal and cardiac muscle extracts [42], but no information is available on whether there are ß and forms of PnNOS mRNA or, because of the lack of a specific antibody, on the demonstration of PnNOS protein in any tissue.
The clarification of which PnNOS variant mRNAs and proteins are present in the penis is important not only so that we can approach the study of their functional significance for the erectile mechanism but also so that we can understand how alternative nNOS promoter usage, mRNA splicing, and potential posttranslational modification may regulate NOS expression during development and aging. The activation of some variants, like PnNOS and the ß forms of both PnNOS and CnNOS, may be regulated differentially from CnNOS. This knowledge may have implications for the therapy of erectile dysfunction in aging, since in rats this condition has been shown to be associated with an imbalance of NO synthesis in relation to penile smooth muscle compliance and since it has been shown to be correctable with NOS gene transfer to the penis [20, 21, 45]. In the present work, we have investigated whether the full-length PnNOS protein is present in the rat penis and whether it is accompanied by significant amounts of the PnNOSß variant, whether both forms are differentially expressed during the life span of the animal, and whether their tissue levels are consistent with those of their respective mRNA species. We have also determined whether the PnNOS protein is located in penile nerves in both the rat and the mouse and whether the ß form of PnNOS protein is expressed in the nNOS mouse despite the blockage of the expression of the form.
MATERIALS AND METHODS
Materials
Protease inhibitors, NADPH, N-
-nitro L-arginine methyl esther (L-NAME), and other reagents were from Sigma Chemical Co. (St. Louis, MO). L-[2,3,4,5-3H]Arginine mono hydrochloride (specific activity: 35–70 Ci/mmol) was purchased from Amersham Corporation (Arlington Heights, IL) and was purified in our laboratory by column chromatography. Vector plasmids pZErO-1 and pcDNA3 were from Invitrogen (Carlsbad, CA). Human embryonic kidney cells (HEK-293) were purchased from ATCC (Rockville, MD). Lipofectamine and Trizol reagent were from GIBCO BRL (Grand Island, NY). The anti-human nNOS monoclonal antibody was from Transduction Laboratories (Lexington, KY); the anti-human synaptophysin polyclonal antibody was from Zymed Laboratories (San Francisco, CA); the biotinylated goat anti-rabbit secondary antibody and the avidin-biotin horseradish peroxidase antibody complex were from Vector Laboratories (Burlingame, CA). DAB was from DAKO Corporation (Carpinteria, CA). The horse- and goat-radish peroxidase-linked secondary antibodies (anti-mouse and anti-rabbit immunoglobulin [IgG], repectively), the Hybond ECL membrane and Western blotting kits were from Amersham Pharmacia Biotech (Piscataway, NJ). The BCA protein assay kit was from Pierce (Rockford, IL).
Animals and Tissue Processing
Male Fischer 344 rats of different ages were used, and in certain cases, experiments with adult and aged rats were repeated with Brown Norway rats. Rats were purchased from Harlan Sprague-Dawley, Inc. (San Diego, CA) and Charles River Laboratories (Wilmington, MA); rats were maintained under controlled lighting and were treated according to National Institutes of Health regulations. The mouse nNOS and nNOSwt strains were as described [37–39] and were bred and maintained at the Johns Hopkins animal facility. Rats were anesthetized with thiopental, and the penile shaft and bulb were carefully dissected from any surrounding tissue. In certain cases, the cavernosal nerve and the pelvic ganglion were also excised. The cerebellum was obtained from some animals. In the case of mice, penile tissue did not include the bulb. Tissues were frozen in liquid nitrogen and stored at -80°C, or for immunohistochemistry, small pieces of tissue were fixed and treated with O.C.T. compound prior to freezing.
Preparation of Anti-PnNOS Antibodies
In order to study PnNOS protein expression, an antibody against the 34-amino acid insert of PnNOS was obtained. A 16-amino acid peptide (YPEPLRFFPRKGPSLS) within this sequence was selected based on hydrophilicity analysis (using the MacVector program) and was synthesized by Fmoc solid-phase methods utilizing MAP resin technology. The peptide was used to elicit polyclonal antibodies in rabbits (Bethyl Laboratories, Inc., Montgomery, TX). The titer was estimated by ELISA, and the affinity-purified antibody was obtained using sepharose-linked peptide columns. A similar procedure was applied to generate a second antibody against an amino acid sequence (PTMKSTKANLQDIGEHC) encoded by exon 2 in both CnNOS and PnNOS.
Transfection of PnNOS into HEK Cells
A construct of the full-length rat PnNOS was prepared by cutting both 5’ and 3’ of the partially overlapping fragments previously described [41] with Bst11071 and XhoI restriction enzymes, purifying the fragments, and ligating them into pZErO-1 vector to give a clone named pZRPnNOS. The PnNOS fragment was then subcloned into eukaryotic expression vector pcDNA3, thereby originating a construct named pcDNA3-RPnNOS. Cultures of HEK-293 cells were transfected on 6-well plates with pcDNA3-RPnNOS (0.5–2 µg) and lipofectamine (2–6 µl/µg DNA) for 3–5 h [46]. Cells were grown in the presence of Dulbecco's modified Eagle's medium (10%) fetal calf serum for 3 days, and then total homogenate (TH) extracts were prepared by adding a boiling buffer (100–200 µl) containing 1% SDS; 1 mM sodium vanadate-protease inhibitors (3 µM leupeptin, 1 µM pepstatin A, 1 mM phenylmethyl sulfonylfluoride); and 10 mM Tris-HCl (pH 7.4); the extracts were then microwaved for 30 sec and clarified for 5 min at 12 500 x g. In cases in which NOS activity was measured, cells were trypsinized and then centrifuged in the presence of serum and trypsin inhibitor for further processing.
Preparation of Tissue Protein Extracts
Unless otherwise stated, the rat penile crura (bulb) and shaft were processed together, but the mouse penile tissue was restricted to the shaft. Each tissue was weighed, and homogenates were prepared at 4°C in 10 volumes of medium containing 0.32 M sucrose per 20 mM Hepes (pH 7.2) per 0.5 mM EDTA per 1 mM dithiothreitol (DTT) and protease inhibitors (as above) using the Polytron homogenizer [20]. The postmitochondrial (PM) and particulate fractions were separated by centrifugation at 15 000 x g for 60 min, and the PM fractions were stored at -80°C. The particulate fraction was re-extracted with a buffer as described above, but this buffer contained 2% SDS instead of 0.32 M sucrose; this was followed by heating at 95°C for 5 min and centrifuging for 5 min at 15 000 x g (pellet extract). When tissue TH was needed, the procedure used for cell TH was applied to small pieces (10–20 mg) of tissue. Protein was estimated by a modified Lowry-Bradford procedure.
Western Blots
Unless stated, equal amounts of protein (40–80 µg) of the tissue (PM or TH) and cell (TH only) extracts were run on 7.5% polyacrylamide gels under denaturing-reducing conditions, and the proteins were transferred to a nitrocellulose membrane for 16 h at 30 V, followed by 1 h at 100 V [11–16]. Prestained protein markers (48–199 kDa) were run in each gel. The immunodetection on the Western blot was carried out for 1 h at room temperature with an affinity chromatography-purified primary antibody consisting of either the mouse monoclonal antibody against a 22.3-kDa fragment of the carboxy terminus of human CnNOS (1:500 dilution) or our rabbit anti-PnNOS or anti-nNOS exon 2 polyclonal antibodies (1:1000). The secondary antibodies were a rabbit anti-mouse or a horse anti-rabbit IgG linked to horseradish peroxidase, and the incubations (1:10 000 or 1:2000 dilutions, respectively) were carried out for 1 h.
The reactive bands were detected with a luminol-based kit. The PM fraction from rat cerebellum or a standard TH fraction from the rat penis was used as a positive control. The NOS identity of the bands was assessed with mouse and rabbit control IgGs at the same concentrations as the primary antibody, or else we omitted this antibody. The quantitative determination of band intensities was carried out by submitting each luminol-reacted membrane to several x-ray exposure times and by selecting the one(s) that fell within the film response range. The films were then scanned, and when necessary, each band density was evaluated by densitometry using an adequate program.
NOS Activity Determinations
Nitric oxide synthase activity in cell extracts was determined as previously described [20, 47]. Briefly, the PM fraction was passed through Dowex AG50WX-8 (Na+) resin to remove endogenous arginine, and 50-µl aliquots were incubated in triplicate for 30 min at 37°C, as indicated, in the presence of 2 µCi/ml resin-purified [3H]L-arginine, 2 mM NADPH, 100 µM L-arginine, and 0.45 mM Ca2+, with or without 2 mM L-NAME. When Ca2+ was omitted, EGTA was added to 5 mM. After eliminating the residual [3H]L-arginine through the resin, [3H]citrulline was counted in the trichloroacetic acid ether-extracted supernatant. Determinations were conducted in triplicate. All values were corrected by the radioactivity eluted in time-zero incubations and were expressed per milligram of soluble protein or per gram of original tissue.
RNA Analysis
Total RNA was isolated from tissue samples (20–100 mg) by a standard protocol using a commercial kit (Trizol) [48]. For RT-PCR determinations [41], total RNA was reverse transcribed (1 µg) in a 20-µl reaction with 200 units of Superscript II RNase H- reverse transcriptase and 0.25 µg of oligo(dT)16 in 50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 10 mM DTT, and 0.5 mM deoxynucleotide triphosphates (dNTPs). Polymerase chain reaction was carried using a 2-µ RT aliquot, with 0.13 units of KlenTAQ DNA polymerase and 0.2 µM selected primers in a 25-µl reaction with 3.5 mM MgCl2, 0.25 mM dNTPs, 16 mM NH4SO4, and 0.15 mg/ml bovine serum albumin. The reaction was conducted for 35 cycles consisting of 35 sec at 94°C, 45 sec at 58°C, and 80 sec at 72°C, preceded by a single 2-min denaturation step (94°C); the process was terminated with a 7-min extension step (72°C). The forward (F) and reverse (R) PCR 20-mer primers were based on the rat nNOS exon sequences [36, 49] and the PnNOS insert [41] and were as follows: 1A: AGCGGGATCCACAGCCCTGGAACT; 1B: GACTGAGGGGCGACACTACCATGC; 1C: CACCACAGCCTCTG GAATGAAAGA; N3:TCAGGGGCAGCAACGGGATGTGTC; EX3.2:CAGGGGCGGAGCTTTGTGCGATTT; B4: CTTGGTGGGAGACTGTTTCC; B6: TCCTTTGTGCGGACATCTTC; RCI3: GTACCCGGAACCCTT GCGTT; and RCI4: TGTCACGGGCAGCAACCAGA.
Fragment sizes were determined by agarose gel electrophoresis. When sequencing was needed, the selected amplified fragments were eluted from the corresponding gel slices and cloned into the SrfI site of the PCR Script-Amp vector [48]. Sequencing by an automatic DNA sequenator was carried out in both directions using the T3 and T7 vector primers.
In certain cases, corpora cavernosa RNA (20 µg) was run on denaturing formaldehyde 1% agarose gels and submitted to Northern blot hybridization [47, 48] with cDNA probes corresponding to either a 1 kilobase (kb) HindIII fragment common to CnNOS and PnNOS or to the 102-bp PnNOS insert. Probe labeling was performed with [32P]dCTP by random priming [47].
Immunohistochemistry
Sections (6 µm) were obtained with a cryostat from the O.C.T. frozen tissue and were fixed for 15 min on Colorfrost-plus microscope slides with 4% paraformaldehyde in Sorensen solution (0.1 M phosphate buffer) [9]. Sections were washed with PBS, permeabilized with 0.4% Triton X-100, blocked with 1.5% normal goat serum in PBS, and incubated overnight in a humidified chamber at 4°C with the primary anti-PnNOS (1:500) or anti-human synaptophysin (1:100) antibodies in blocking solution. Biotinylated goat anti-rabbit secondary antibody (1:200) was applied and incubated for 30 min, followed by 30 min with avidin-biotin-horseradish peroxidase complex. Antibody binding was visualized with DAB solution, and sections were dehydrated and coverslips mounted.
Statistical Analysis
Values were expressed as mean ± SEM. The normality distributions of the data were established using the Wilk-Shapiro test. The outcome measures in the different groups were compared using a one-way ANOVA. The differences among groups was considered significant when P < 0.05. Pairwise multiple comparison procedures between two groups were performed using the Tukey test.
RESULTS
Validation of the Anti-PnNOS Antibody with PnNOS cDNA-Transfected Cells
In order to test the antibody elicited against rat PnNOS, HEK-293 cells were transfected with a construct of rat PnNOS cDNA in the eukaryotic expression vector pcDNA3. When protein extracts were prepared from the transfected HEK cells, run on polyacrylamide gels, and submitted to Western blots with the commercial antibody raised against sequences common to both CnNOS and PnNOS, the expected single 155- to 160-kDa band was detected at 3 h of transfection (Fig. 1). This band coincided with the one in the control rat cerebellum, was even more intense at 5 h of transfection, and the signal exceeded the one corresponding to endogenous nNOS in the nontransfected HEK cells. When the anti-PnNOS antibody was applied to a second gel run with the same cell and tissue extracts, no band was visible in the HEK cells, indicating that the endogenous nNOS is exclusively the cerebellar-type variant. A 155-kDa band was visible in the cerebellum and was very intense in the PnNOS-transfected cells, thus confirming the ability of the antibody to detect PnNOS. The smaller band observed at the 4:1 lipofectamine:pDNA ratio is probably an artifact, since it was not detected with the CnNOS/PnNOS common antibody.
|
The cells transfected with the PnNOS cDNA exhibited a three- to fourfold increase in NOS activity, by the L-arginine/citrulline conversion assay, over the basal level in the HEK-293 cells transfected with the empty vector (not shown). Although the activity value was lower than the endogenous one in the rat cerebellar extracts, it indicated that the protein encoded by our PnNOS cDNA construct and detected by our anti-PnNOS antibody was functional.
Expression of PnNOS in the Rat Penis
In order to determine whether the level of the nNOS variants in the rat penis is subject to age-related regulation, the PM fractions were isolated from the penile shaft and crura of rats that ranged from 8 days to 26 mo of age and were submitted to Western blot analysis with the antibody common to CnNOS and PnNOS. Figure 2 shows that in general, the 155- and 135-kDa nNOS proteins were present in approximately equal amounts, with a slight predominance of the larger form. The only exception was in the penis of the oldest rats, in which the 135-kDa protein was significantly more abundant than the 160-kDa protein. The total nNOS, representing the sum of both variants, did not change significantly throughout the range of different ages.
|
The anti-PnNOS antibody detected both bands reacting with the CnNOS/PnNOS common antibody (Fig. 3, top panel), and although the ratio between the 155- and 135-kDa forms was more variable, the smaller protein was also more abundant in the 26-mo-old animals. Both nNOS variants were detectable with the nNOS antibody in the particulate extract (PE) from the shaft and the crura, representing membrane-bound nNOS (Fig. 3, bottom left panel). The PnNOS antibody detected essentially the 155-kDa band and another larger band running at 230–250 kDa, a band of unknown origin (Fig. 3, bottom right panel). Because of the differences in sensitivity and specificity of both antibodies, it was not possible to assess by Western blot analysis what fraction of the nNOS in a given penile extract detected by the common CnNOS/PnNOS antibody is specifically PnNOS.
|
A more adequate comparison of both antibodies—and hence, of the distribution, and, by inference, the relative amounts of PnNOS and CnNOS—can be obtained by immunocytochemistry. Adjacent frozen cross sections of rat penile shaft from an adult rat were immunoreacted separately with the antibody common for CnNOS and PnNOS, the antibody against PnNOS, and an antibody against the neural marker synaptophysin (Fig. 4). At low magnification (Fig. 4, left panels), it is apparent that the anti-nNOS antibody stained what appeared to be nerve endings in the corpora cavernosa and around the tunica albuginea; these areas coincide with the similar areas that were detected by the nerve marker. Most of the tissue regions stained with the CnNOS/PnNOS common antibody also reacted with the PnNOS antibody. In some cases, other cell types seemed to be stained by both nNOS antibodies, such as the lining of cavernous spaces and the epithelium of the spongiosum urethra. The co-localization of nNOS and synaptophysin staining was more evident at high magnification (Fig. 4, right panels). This indicates that a substantial fraction of nNOS protein in the rat penis is PnNOS, and it is primarily located in the nerves.
|
PnNOS mRNA Variant Expression in the Rat Penis
In order to corroborate the expression of PnNOS in penile nerves, the cavernosal nerves of three rats were dissected and pooled, RNA was isolated, and an RT-PCR reaction was performed with primers priming inside the 102-bp sequence specific for PnNOS. A single 92-bp band was obtained, corresponding to the expected product (Fig. 5, left panel). Although a similar procedure had been used for detecting and cloning PnNOS mRNA from the whole penile shaft [41], no direct assessment of the size of PnNOS or nNOS transcripts had been available previously. This assessment would have allowed us to determine whether PnNOS mRNA is expressed as ß or other truncated nNOS transcripts; it would also have allowed us to determine the expected full-length form. Therefore, total RNA was isolated individually from three rat penile shafts and was submitted to Northern blot hybridization with a probe common to CnNOS and PnNOS. Figure 5 (right panel) shows a transcript with a size (9–10 kb) that is approximately similar to the one seen in the cerebellum for the full-length CnNOS [28, 48]. No smaller transcript with a size corresponding to the ß or form or that was compatible with a putative transcript for the 135-kDa protein band was detected by this method. The PnNOS-specific probe is smaller (by a 10-fold measure) and difficult to label, so the radioactive probe was insufficient to obtain an adequate signal with total RNA (not shown).
|
The apparent absence of shorter CnNOS or PnNOS variants in the Northern blot analysis prompted a search for the ß form by PCR-amplifying regions of cDNA generated by RT of total cavernosal nerve and pelvic ganglion RNA with oligo(dT) primer. Primers spanning exon 2 (by anchoring on exons 1 [sense] and 3 [antisense]) or within exon 2 (by anchoring on exons 1 [sense] and 2 [antisense]) were used to generate DNA fragments corresponding to the and ß forms (Fig. 6). Each exon 1 variant (a, b, and c) was amplified by a specific sense primer on exon 1 (1A, 1B, and 1C, respectively). In RNA obtained from the cavernosal nerve, the three variants were detected on the ethidium bromide-stained gel by using as antisense the N3 primer pairing to exon 2, which is absent in the ß forms. When the 1B sense was combined with an antisense primer on exon 3 (EX3.2), the expected 1132-bp fragment was amplified, thus confirming the presence of the form in the cavernosal nerve. In the pelvic ganglion, the N3 antisense generated the expected bands with primers 1B and 1C, and the combination EX3.2/1A was particularly efficient in the amplification of the expected fragment, thus indicating that the a, b, and c variants are expressed. However, no traces of the ß form were observed in the cavernosal nerve and pelvic plexus.
|
Since the failure to detect the ß form could be due to the inadequacy of the specific primer (EX3.2) applied above, new sets of exon 3 primers were used on RNA extracted from the penile shaft and bulb, as shown on Figure 7 (top panel). Primers B4 and B6 were able to amplify the 217- and 373-bp fragments expected for the ßb forms, thus suggesting that the shorter nNOS splicing product is actually produced in the penis (Fig. 7, bottom panel). The amount of the nNOS mRNA appears to be in excess of that of the putative ß form, as indicated by the 1288-bp band that originated from the 1B-B4 combination in comparison with the corresponding putative ß fragment. However, with the combination 1B-B6, the amounts of the putative nNOSß and the 1473-bp nNOS were approximately similar. To confirm that the short fragments generated with the combination of 1A and 1B sense and B4 antisense primers do indeed arise from nNOS ß splicing, the amplified cDNA bands were excised from the gel, purified, cloned, and sequenced. In both bands, the obtained sequence was completely homologous to the sequence that would be obtained from a splicing event in the published sequence of rat CnNOS [36, 49] between exons 1a and 1b, respectively, and exon 3, which would delete exon 2.
|
Persistence of PnNOS Protein and the ß Variant Protein in the Penis of the nNOS Knock-Out Mouse
The reported expression of the nNOSß mRNA in the penis of the nNOS mouse [38] provides an explanation for the persistence of fertility—and hence, the persistence of erectile function—in the male animal, but no direct demonstration of the ß nNOS protein in the penis of this animal has been published, and it is not known whether the residual nNOS protein is the CnNOS or PnNOS variant. In order to elucidate this question, cross sections of the penises of the mouse wild-type and transgenic animals were immunostained independently with antibodies common to CnNOS/PnNOS, with antibodies against PnNOS only, and with antibodies against synaptophysin. Figure 8 shows that the antibody selective for PnNOS gives a pattern of staining of the wild-type penis that resembles the one observed in the rat; it also shows that in the nNOS
/ penis, essentially the same regions were stained, except that the intensity was lower. The staining with antibodies against synaptophysin and the common sequence in CnNOS/PnNOS coincided, in the knock-out mouse penile sections, with the location of PnNOS reactivity (not shown), which indicates that PnNOS is primarily in the nerves and represents a fraction of the nNOS that survives in the penis.
|
In order to determine whether the nNOSß mRNA (CnNOS or PnNOS) in the rat and mouse penis is indeed expressed into the 135-kDa protein detected in the rat penile extracts (by Western blot), we developed an antibody against an amino acid sequence encoded by rat nNOS exon 2. Figure 9 (top panel) shows that the 135-kDa band that is visible with the carboxy-end CnNOS/PnNOS antibody in the shaft and bulb extracts from the rat penis and that is faintly expressed in the cerebellum (Fig. 9, left panel) is not detected by the CnNOS/PnNOS exon 2 antibody (Fig. 9, right panel). As expected, this antibody detects the 155-kDa band, corresponding to the exon 2-containing form. When the carboxy-end CnNOS/PnNOS antibody was used with mouse tissue extracts, the 155-kDa protein was not visible in the penile shaft from the knock-out strain, despite a protein input that was fourfold higher than that found in the wild-type extract, and only the 135-kDa ß band remained (Fig. 9, bottom panel). A similar situation occurred in the cerebellum, where an additional smaller band of unknown origin was detected in the nNOS/ extract. A 10-fold higher protein input was run in the case of the nNOS compared to the nNOSwt extract.
|
DISCUSSION
In the current work we have characterized the expression of the PnNOS protein in the rodent penis with the application of a new custom-made antibody against a peptide encoded by the 102-bp insert in PnNOS. The efficacy of the new antibody was substantiated by its ability to detect the recombinant protein encoded by a rat PnNOS cDNA construct, in comparison to an antibody against a region common to both CnNOS and PnNOS. By a combination of Western blot analysis and immunohistochemical detection with both antibodies, it has been shown for the first time that PnNOS is the major nNOS variant protein expressed in the rat and mouse penis and that it is located in the nerves. Two forms of PnNOS proteins that differed in size were detected throughout the life span of the rat; these forms corresponded to the 155- and 135-kDa nNOS bands reported previously in penile homogenates from adult rats. During this period, no significant variation in the absolute or relative amounts of PnNOS proteins occurred, with the exception of the relative predominance of the shorter protein in very old animals.
The analysis of nNOS mRNA expression by Northern blot and RT-PCR detected nNOS/1a, 1b, and 1c alternative transcripts in the rat penis, cavernosal nerve, and pelvic ganglion. But in some cases, these assays detected few similar nNOSß variants. The imbalance between the amounts of both nNOS mRNAs, despite the similarity in the abundance of both nNOS proteins, indicates that the PnNOSß mRNA is less stable than the nNOS transcripts. We have confirmed that the 155- and 135-kDa proteins are encoded by the and ß nNOS mRNAs, respectively, by applying the second custom-made antibody against exon 2. The neural location of PnNOS agrees with the hypothesis that the synthesis of NO required for penile erection is essentially sustained by PnNOS activity, an assumption that is supported by the demonstration of substantial PnNOS in the penile nerve terminals of the knock-out mouse. Our results on the persistence of the ß variant protein in this animal also agree with the previous interpretation from nNOS mRNA analysis [28, 38], which postulates that the survival of erectile function in the transgenic animal is due to alternative splicing that bypasses the engineered blockade of gene expression.
No previous reports are available on the expression in the penis of the PnNOS protein or of the homologous nNOSµ, although several papers have described the detection of the respective mRNAs in organs as diverse as the penis, prostate, bladder, pelvic ganglion, skeletal muscle, and heart [41, 42]. Recently, a novel human nNOSµ cDNA was cloned from skeletal muscle and was shown to have a 24% divergence in the 34-amino acid insert from the mouse nNOSµ and the rat PnNOS [43]. This form, as well as other variants having additional 42- and 67-bp insertions in the splice site between exons 16 and 17, are expressed in the penis, but if translated, they would give rise to truncated forms of nNOS, forms that lacked the C-terminal electron-donating domain. In contrast, the recombinant PnNOS protein is nearly as active in vivo and in vitro as is the regular CnNOS [41, 50]. Since CnNOS mRNA is not found in substantial amounts in the rat penis, it may be inferred that the PnNOS protein is the main nNOS variant responsible for erectile function in this animal. In the nNOS mouse, the residual expression of PnNOS in the penis appears to sustain male reproduction capacity, but the contribution by CnNOS expression cannot be evaluated. The contribution of eNOS up-regulation [39] cannot be excluded either, although the role of this isoform in penile erection is not clear [4].
The question of whether PnNOS and nNOSµ are identical alternative splicing products from the same nNOS gene that gives rise to CnNOS is still unresolved, although the position of the insert between exon/intron junctions suggests that it is a cryptic exon designated 16.1 [42, 43]. However, until the intronic sequences surrounding the 102-bp insert are determined and compared with the ones published for CnNOS, and until the resulting DNA probes are applied to decide whether they map in the same region of chromosome 12, the possibility that PnNOS and nNOSµ/CnNOS are products of different genes cannot be discarded.
It is not known whether the preferential expression of PnNOS or nNOSµ in the penis and other organs has any functional tissue specificity. Studies with recombinant nNOSµ have shown that the 34-amino acid insertion decreases by 50% the rate of cytochrome c and molecular oxygen reduction, as compared to CnNOS [50], and it does so without affecting the L-arginine oxidizing activity. The insert in the rat PnNOS and mouse nNOSµ has a putative phosphorylation site, and the human nNOSµ contains a potential myristoylation site [43]. In addition, the insert is located in the vicinity of an autoinhibitory control element that is displaced by Ca2+ and calmodulin binding to nNOS during activation of the enzyme [44]. This region is absent in iNOS, an isoform that binds calmodulin tightly and that is not subject to Ca2+ regulation. So far, these features are insufficient to propose a role for the 34-amino acid sequence in controlling NO synthesis during smooth or skeletal muscle relaxation. Further studies are needed to decide whether an important functional distinction is provided by the insert, such as for the binding of protein modulators different from the ones acting on the PDZ domain of exon 2. However, an interesting characteristic is the rapid degradation of recombinant and endogenous nNOSµ, in comparison to CnNOS, by calpain, a Ca2+-dependent protease [50].
Another issue that is still undecided involves the functional significance of the shorter splicing variants of nNOS mRNA. The /1a, 1b, and 1c exon 1 spliced RNAs have been described in brain, kidney, heart, intestine, and embryo [36], but since the respective first exons 1a, 1b, and 1c are located upstream of the coding sequence, these sequence alterations do not modify the coding region for the 155-kDa nNOS. However, their transcripts are likely to result from activation of different promoters [33]. We do not know whether the detection of these three splicing variants in the adult rat penis may be linked to a developmental specific expression pattern, as has been suggested to be the case with other organs [36].
In turn, the ß form of nNOS mRNA has been described in mouse skeletal muscle, brain, and penis [28, 38], but it was not found in rat organs [36]. The corresponding protein lacking the PDZ domain has been found in skeletal muscle and brain [28], and the endogenous and recombinant proteins are fully active. The other nNOS alternative splicing mRNA variant, the form, has been detected in the mouse penis [38]. Our finding of the nNOSß mRNA in the rat penis suggests that this splice product is expressed in many organs, albeit at low levels. This does not reconcile with the relative abundance of the 135-kDa PnNOS protein in relation to the 155-kDa band. The discrepancy may result from a rapid degradation of PnNOSß mRNA or a slow turnover of the 135-kDa PnNOS protein. We have conclusively shown that the 135-kDa nNOS protein lacks the region encoded by exon 2. The fact that a large fraction of PnNOS protein in the penis is the ß form indicates that part of PnNOS activity is insensitive to modulation by PIN, CAPON, or NMDA receptor, likely as a result of the absence of the PDZ and PIN binding domains [32]. The persistence of the CnNOS/PnNOSß form(s) in the knock-out mouse would then provide an adaptive response to the obliteration of the expression of the full-length nNOS form by making the residual nNOS resistant to inhibition by PIN/CAPON. The finding of a substantial nNOS activity in penile homogenates from the nNOS knock-out mouse—obtained by measuring the L-arginine/citrulline conversion (unpublished data)—agrees with this assumption, with the caveat that the complex in vivo regulation of nNOS activity may not be mimicked in vitro.
In conclusion, our demonstration that PnNOS protein is considerably expressed in the rat penile nerves and survives in the knock-out mouse nNOS as well as our finding that the ß form mRNA is present and encodes a 135-kDa PnNOS protein variant that is resistant to exon 2-binding modulators may have functional significance for penile erection. The insertion of a 34-amino acid sequence and the deletion of an amino terminus domain may provide alternative or complementary ways for a specific modulation of NOS expression/activity, and hence of NO synthesis, that is tailored to the needs of smooth muscle or skeletal muscle relaxation. Should this be the case, PnNOS and its variants may be targets for gene-therapy approaches to erectile dysfunction [4, 51].
FOOTNOTES
1 Supported by National Institutes of Health grants RO1DK53069 (to N.G.C.) and RO1 DK 02568 (to A.L.B). 
2 Correspondence: Nestor F. Gonzalez-Cadavid, Harbor-UCLA Medical Center, Division of Urology, Urology Research Laboratory, 1000 West Carson St., Torrance, CA 90509. FAX: 310 222 1914;ncadavid{at}ucla.edu
Accepted: April 11, 2000.
Received: December 31, 1999.
REFERENCES
1-syntrophin mediated by PDZ domains. Cell 1996; 84:757–767.[CrossRef][Medline]-deficient transgenic mice. J Urol 1998; 159:89 (A341). and ß are closely related calpain-sensitive proteins. Mol Pharmacol 1998; 54:305–312.This article has been cited by other articles:
|
|
 |
|
  K. J. Hurt, S. F. Sezen, H. C. Champion, J. K. Crone, M. A. Palese, P. L. Huang, A. Sawa, X. Luo, B. Musicki, S. H. Snyder, et al. Alternatively spliced neuronal nitric oxide synthase mediates penile erection PNAS, February 28, 2006; 103(9): 3440 - 3443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Bivalacqua, M. F. Usta, H. C. Champion, P. J. Kadowitz, and W. J. G. Hellstrom Endothelial Dysfunction in Erectile Dysfunction: Role of the Endothelium in Erectile Physiology and Disease J Androl, November 1, 2003; 24(6_suppl): S17 - S37. [Full Text] [PDF]
|
|
|
|
|
|
M. A. Palese, J. K. Crone, and A. L. Burnett A Castrated Mouse Model of Erectile Dysfunction J Androl, September 1, 2003; 24(5): 699 - 703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Toda and T. Okamura The Pharmacology of Nitric Oxide in the Peripheral Nervous System of Blood Vessels Pharmacol. Rev., June 1, 2003; 55(2): 271 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.R. Magee, M.G. Ferrini, H.H. Davila, C.B. Zeller, D. Vernet, J. Sun, R. Lalani, A.L. Burnett, J. Rajfer, and N.F. Gonzalez-Cadavid Protein Inhibitor of Nitric Oxide Synthase (NOS) and the N-Methyl-D-Aspartate Receptor Are Expressed in the Rat and Mouse Penile Nerves and Colocalize with Penile Neuronal NOS Biol Reprod, February 1, 2003; 68(2): 478 - 488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Burnett Nitric Oxide Regulation of Penile Erection: Biology and Therapeutic Implications J Androl, September 1, 2002; 23(5): S20 - S26. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Maas, E. Schwedhelm, J. Albsmeier, and R. H Boger The pathophysiology of erectile dysfunction related to endothelial dysfunction and mediators of vascular function Vascular Medicine, August 1, 2002; 7(3): 213 - 225. [Abstract] [PDF]
|
|
|
|
|
|
T. R. Magee, M. Ferrini, H. J. Garban, D. Vernet, K. Mitani, J. Rajfer, and N. F. Gonzalez-Cadavid Gene Therapy of Erectile Dysfunction in the Rat with Penile Neuronal Nitric Oxide Synthase Biol Reprod, July 1, 2002; 67(1): 20 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-E. Andersson Pharmacology of Penile Erection Pharmacol. Rev., September 1, 2001; 53(3): 417 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ferrini, T. R. Magee, D. Vernet, J. Rajfer, and N. F. González-Cadavid Aging-Related Expression of Inducible Nitric Oxide Synthase and Markers of Tissue Damage in the Rat Penis Biol Reprod, March 1, 2001; 64(3): 974 - 982. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
