Potential Functional Significance of Brain-Type and Muscle-Type Nitric Oxide Synthase I Expressed in Adventitia and Media of Rat Aorta
Abstract—Skeletal muscle and myocardium express μNOS I, an elongated splice variant of neuronal-type nitric oxide (NO) synthase (NOS I), and NOS III, endothelial-type NO synthase, respectively. This study was designed to elucidate whether vascular smooth muscle also contains a constitutively expressed NO synthase isoform. In the rat, μNOS I contains an insert of 102 nucleotides after nucleotide 2865 of the cDNA, yielding a protein of 164 kd. Reverse transcription-polymerase chain reaction with primers flanking this insert and with insert-specific primers indicated that endothelium-denuded rat aorta expresses both brain-type NOS I and μNOS I. RNase protection analyses with an antisense RNA probe overlapping the μNOS I insert detected significant amounts of NOS I mRNA and lesser amounts of μNOS I mRNA in endothelium-denuded aorta. Western blots using a specific polyclonal antibody recognizing NOS I and μNOS I showed a major band of the 160-kd NOS I and a lesser band of a slightly larger protein in endothelium-denuded aorta. Immunohistochemistry demonstrated low levels of NOS I/μNOS I immunoreactivity in the medial layer of rat aorta, whereas the endothelium expressed only NOS III immunoreactivity. When the adventitia also was removed, NOS I and μNOS I mRNA decreased markedly but remained detectable in the medial layer. In functional experiments with endothelium-denuded rat aortic rings (that contained no NOS III), contractions induced by KCl were markedly increased in the presence of the NOS inhibitor NG-nitro-l-arginine. These data demonstrate that 2 subforms of NOS I are expressed in nonendothelial components of rat aorta: NOS I and lesser amounts of μNOS I. Under certain conditions, this NOS I/μNOS I expression could serve as a backup system to the functionally predominant NOS III.
- smooth muscle
- potassium chloride
- RNase protection analysis
- Western blot
- Received February 9, 1998.
- Accepted March 25, 1999.
Nitric oxide (NO) is a gaseous molecule with important biological functions in the cardiovascular system and elsewhere. It is produced by 3 isoforms of NO synthase (NOS), namely, neuronal-type NOS I, inducible-type NOS II, and endothelial-type NOS III (for review see Reference 11 ). NOS I and NOS III are usually expressed constitutively. NOS II is not normally found in resting cells and tissues (with the exception of skeletal muscle and intestine2 ) but can be induced in many cell types by bacterial lipopolysaccharide (LPS) and cytokines.
In skeletal muscle, all 3 isoforms of NOS have been described. NOS I is localized at the sarcolemma and especially enriched at the neuromuscular end plate.2 3 4 In the rat, skeletal muscle expresses an alternatively spliced form of NOS I, namely, μNOS I, that contains a 102-nucleotide (nt)/34–amino acid insert after nucleotide 2865 of the NOS I cDNA.5 NOS II has been found to be constitutively expressed in type I fibers of guinea pig skeletal muscle.2 The expression could be further increased with LPS.2 Immunohistochemical evidence indicating the presence of NOS III in rat skeletal muscle has also been published.6 This isoform has been localized mainly to mitochondria.6 Skeletal muscle–derived NO is likely to participate in the regulation of contractile function.2 3
Cardiac myocytes, another type of muscle cells, have been reported to express NOS III constitutively.7 This isoform seems to mediate the muscarinic cholinergic attenuation of contractile responsiveness to β-adrenergic agonists.7 However, the same study failed to detect NOS I in ventricular myocytes.7 NOS II has been shown to be induced in cardiac myocytes by cytokines.8 9
Recently, NOS I immunoreactivity has been described in the media of the carotid artery of both normotensive and spontaneously hypertensive rats.10 In hypertensive rats, NO produced by this enzyme negatively modulated the contractions induced by angiotensin II.10 Exposure of cultured human aortic smooth muscle cells to fluid flow has been shown to result in an increased NO production.11 Western blots demonstrated that the only isoform expressed by these cells was NOS I.11 Another study using Western blot analysis showed a low expression of NOS III protein in endothelium-denuded rat aorta.12 This expression increased when animals were treated with monocrotaline to induce heart failure.12 NOS II is not normally present in vascular smooth muscle cells but can be induced by cytokines or LPS.13 14
These somewhat controversial findings prompted us to investigate whether vascular smooth muscle and adventitia of an intact artery, namely, rat aorta, express a NOS isoform constitutively, to identify the isoform(s), and to test the possible functional significance of this expression for the modulation of vascular tone. We demonstrate that NOS III expression is clearly restricted to the endothelium. The nonendothelial components of rat aorta show a clearly detectable expression of NOS I>μNOS I. Functional studies with endothelium-denuded rat aortic rings indicate that this enzyme can modulate vascular contractility under conditions in which intracellular calcium levels are elevated (eg, after KCl depolarization).
Materials and Methods
Preparation of Rat Aortas and Control Tissues
Female Sprague-Dawley rats (200 to 300 g) were killed by a blow to the head and exsanguination. The thorax was immediately opened, and a large segment of the aorta (thoracic and abdominal) was placed in ice-cold PBS. The aorta was dissected free of fat and connective tissue and opened longitudinally. Depending on the experiment performed, the artery was left intact, or the endothelium with or without the adventitia was removed mechanically. As positive controls of tissues expressing NOS I and μNOS I, respectively, the cerebellum and the gastrocnemius muscle were removed from the rats as well. For RNA and protein isolation, the tissues were rapidly frozen in liquid nitrogen. For immunohistochemistry, aortas were frozen in 2-methylbutane at −80°C.
RNA Isolation and RT-PCR
Total RNA was isolated from aorta (intact, endothelium-denuded, or medial layer alone), rat cerebellum, and gastrocnemius muscle by acid guanidinium thiocyanate/phenol/chloroform extraction.15 RNA (2 μg) was annealed with 40 ng random hexamer primers and reverse-transcribed with SuperScript reverse transcriptase according to the manufacturer’s instructions. Reverse transcription (RT)-generated cDNAs encoding for rat NOS I, μNOS I, and γ-actin were amplified by polymerase chain reaction (PCR) as described previously (40 cycles).16 The primer pairs specific for NOS I, μNOS I, and γ-actin were selected by analysis of the published cDNA sequences (Table⇓).5 17 18 PCR products were separated on 1.8% agarose gels in Tris borate/EDTA buffer containing 0.5 μg/mL ethidium bromide.
Cloning of Rat NOS I/μNOS I and γ-Actin cDNA Fragments
The amplified cDNA fragments (μNOS I, 380 base pairs [bp]; γ-actin, 353 bp) were cloned into the EcoRV site of vector pCR-Script with a SureClone Ligation Kit, generating the cDNA clones pCR-μNOS I-rat and pCR-γ-actin-rat. μNOS I and γ-actin cDNAs were sequenced by use of the dideoxy-mediated chain termination method (T7Sequencing Kit, Pharmacia).
Preparation of Antisense RNA Probes and RNase Protection Analyses
To generate radiolabeled antisense RNA probes for RNase protection analyses, the cDNA clones pCR-μNOS I-rat and pCR-γ-actin-rat were linearized with EcoRI and AspI, respectively, extracted with phenol/chloroform, and concentrated by ethanol precipitation. These DNAs (0.5 μg) were in vitro–transcribed for 60 minutes at 37°C by using T3/T7 RNA polymerase and [α-32P]UTP. Then, the template DNA was degraded with DNase I (RNase free, 10 U/μL) for 45 minutes at 37°C, and the labeled RNA was precipitated with ethanol.
RNase protection analyses were performed with the above [α-32P]UTP-labeled probes as described.19 Briefly, after a denaturation step for 10 minutes at 85°C, 20 μg of total RNA (isolated as described above) was hybridized for 14 hours at 51°C with a 200 000 cpm–labeled NOS I/μNOS I cRNA probe (435 nt) and a 30 000 cpm–labeled γ-actin cRNA probe (183 nt) in hybridization buffer (40 mol/L PIPES [pH 6.7], 400 mmol/L NaCl, 1 mmol/L EDTA, and 50% formamide). Then, the hybridization mixture was incubated for 30 minutes at 30°C with 300 μL digestion buffer (10 mmol/L Tris HCl [pH 7.4], 300 mmol/L NaCl, and 5 mmol/L EDTA) containing 3.5 μg RNase A and 25 U RNase T1. The reaction was stopped by addition of 70 μL of a buffer (10 mmol/L Tris HCl [pH 7.8], 5 mmol/L EDTA, and 2.85% SDS) containing 70 μg proteinase K. After incubation for 15 minutes at 37°C and phenol/chloroform extraction, the samples were concentrated by ethanol precipitation and analyzed by electrophoresis using 6% polyacrylamide/8 mol/L urea gels. Gels were dried and exposed to x-ray films for ≈3 days. Densitometric analyses were performed using the PhosphoImager system (Bio-Rad). The protected RNA fragments for NOS I, μNOS I, and γ-actin were 302, 377, and 110 nt, respectively.
For Western blotting, the aorta (intact, endothelium-denuded, or medial layer alone), cerebellum, and gastrocnemius muscle were homogenized on ice. Combined soluble and CHAPS-solubilized particulate fractions (100 μg each) were separated on 5% or 7.5% SDS-polyacrylamide gels and electroblotted to nitrocellulose membranes (Schleicher & Schuell).20 21 Blots were blocked for 60 minutes at room temperature in Tris-buffered saline (TBS, consisting of 10 mmol/L Tris HCl [pH 7.4] and 154 mmol/L NaCl) containing 5% (wt/vol) nonfat dry milk and 0.05% (wt/vol) Tween 20. They were then incubated overnight at 4°C with rabbit polyclonal antibodies to NOS I (1:2000)22 or NOS II (1:1000)23 or a mouse monoclonal antibody to NOS III (1:500) in PBS containing 1% (wt/vol) bovine serum albumin and 0.1% (wt/vol) Tween 20. After 3 washes with TBS containing 5% (wt/vol) nonfat dry milk and 0.05% (wt/vol) Tween 20, the blots were incubated for 60 minutes at room temperature with a goat anti-rabbit or a horse anti-mouse alkaline phosphatase–conjugated secondary antibody. After 3 washes with TBS containing 0.05% (wt/vol) Tween 20, bands were visualized with 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium chloride.
To detect NOS I in rat aorta, serial sections of 10 μm were made with a cryostat (Leica). The sections were fixed for 5 minutes in 4% (wt/vol) paraformaldehyde at 4°C. Immunohistochemistry was performed as described previously.16 Briefly, after blocking of endogenous peroxidase and biotin, sections were washed for 10 minutes in antibody incubation medium (4% [wt/vol] nonfat dry milk and 0.3% [vol/vol] Triton X-100 in PBS). They were then incubated for 30 minutes in a 1:10 dilution of normal serum from the species in which the secondary antibody was generated (goat or horse). Sections were incubated overnight at 4°C with one of the following primary antibodies: a rabbit polyclonal antibody to NOS I (1:500) and a mouse monoclonal antibody to NOS III (1:100). Sections were washed in incubation medium and then incubated for 60 minutes with the secondary biotinylated antibody (1:100, a goat anti-rabbit IgG or a horse anti-mouse IgG). After several washes in PBS, cells were exposed for 20 minutes to an avidin DH-biotinylated horseradish peroxidase H complex (1:100 in PBS, Vectastain Elite ABC Kit). The chromogen reaction was performed with 3,3′-diaminobenzidine (DAB peroxidase substrate tablet set). Cells were washed in distilled water, mounted in glycerol gelatin, and coverslipped. In control experiments, the primary antibodies were replaced with rabbit preimmune serum or mouse IgG.
Measurement of Vascular Responses
Rat aortic segments were obtained as described above. Then, vascular rings of 3-mm length were prepared. Endothelium was removed by gently rubbing the intimal surface with a stainless steel wire. The rings were incubated at 37°C in 5-mL organ baths in modified Krebs’ solution of the following composition (mmol/L): NaCl 118.0, KCl 4.7, KH2PO4 1.2, CaCl2 · 2H2O 2.5, MgSO4 · 7H2O 1.2, NaHCO3 25.0, and glucose 10.6. The solution was aerated continuously with 95% O2 and 5% CO2. The pH was 7.4. Isometric tension of the aortic rings was recorded with force transducers coupled to DC amplifiers and a pen recorder. The initial load on the tissues was adjusted to 1 g. Before the actual experiment, preparations were allowed to stabilize for ≈1 hour. Rings were precontracted several times with 0.1 μmol/L norepinephrine (NE) until the contractile response was reproducible. The absence of endothelium was verified by a complete loss of relaxation to acetylcholine (1 μmol/L). Then, concentration-response curves were generated with either KCl (5 to 90 mmol/L) or NE (0.1 nmol/L to 1 μmol/L). When the KCl concentration was increased, the appropriate amounts of NaCl were omitted to maintain osmolarity. After a washout period, rings were incubated for 30 minutes with 300 μmol/L NG-nitro-l-arginine (L-NNA). Then, a second concentration-response curve was generated with KCl and NE in the presence of the NOS inhibitor.
Data represent mean±SEM. Statistical differences were determined by factorial ANOVA followed by the Fisher protected least significant difference test for comparison of multiple means.
The DAB peroxidase substrate tablet set, goat anti-rabbit antibody conjugated to alkaline phosphatase, horse anti-mouse antibody conjugated to alkaline phosphatase, mouse immunoglobulin G, LPS from Escherichia coli (serotype 055:B5), L-NNA, and (−)-norepinephrine bitartrate salt were obtained from Sigma Chemical. SuperScript reverse transcriptase was from Life Technologies. EcoRI, EcoRV, random hexamer primer, the SureClone Ligation Kit, Taq DNA polymerase, Taq polymerase reaction buffer, and the T7Sequencing Kit were from Pharmacia. pCR-Script was from Stratagene. AspI, DNase I, RNase A, RNase T1, proteinase K, and T3 RNA polymerase were from Boehringer-Mannheim. [α-32P]UTP was from ICN. Biotinylated goat anti-rabbit IgG, biotinylated horse anti-mouse IgG, goat normal serum, horse normal serum, and the Vectastain Elite ABC Kit were from Vector Laboratories. Mouse monoclonal antibody to NOS III was from Transduction Laboratories.
Detection of NOS I and μNOS I mRNA in Rat Aorta
For the detection of NOS I and μNOS I mRNA in rat aorta, 2 different RT-PCR approaches were used. In a first set of experiments, primers flanking the μNOS I insert were used (Figure 1⇓). With RNA from rat cerebellum, this yielded a 340-bp NOS I fragment, whereas with RNA from rat gastrocnemius muscle, only a 442-bp μNOS I fragment was amplified (Figure 1⇓). RNA from intact aorta or endothelium-denuded rat aorta or from the medial layer of the aorta showed the 340-bp fragment (Figure 1⇓).
In a second approach, the same antisense primer was used but was combined with a sense primer specific for the μNOS I insert (Figure 2⇓). This yielded a 380-bp fragment in gastrocnemius muscle (Figure 2⇓). This fragment was amplified to a lesser extent from RNAs from cerebellum and aorta (intact, endothelium-denuded, or medial layer alone; Figure 2⇓). The amplified 380-bp cDNA fragment from endothelium-denuded rat aorta was cloned into the EcoRV site of vector pCR-Script. Its sequence was found to be 100% identical to the sequence reported for rat skeletal muscle,5 thereby demonstrating that μNOS mRNA is expressed in the endothelium-denuded aorta.
An antisense RNA probe generated from this μNOS-specific sequence was used in subsequent RNase protection analyses. Because this probe overlaps the μNOS insert, it recognizes both NOS I and μNOS I mRNAs. As expected, RNA from gastrocnemius muscle showed mainly the 377-nt fragment for μNOS I (95.3±3.7% of total, n=5; Figure 3⇓). RNA from cerebellum produced predominantly the 302-nt fragment of brain-type NOS I, but contained 9.6±1.9% μNOS I mRNA (n=5, Figure 3⇓). Also, the endothelium-denuded aorta contained brain-type NOS I and μNOS I mRNA with brain-type NOS I representing 92.1±2.8% of total (n=5). In the medial layer alone, the overall amount of both NOS I subforms was less than in total aorta, but the relative amounts of NOS I and μNOS I remained about the same (with μNOS I representing 8.7±1.5% of total, n=5; Figure 3⇓).
Detection of NOS Protein in Rat Aorta
Western blots (from 5% SDS-polyacrylamide gels) using a specific polyclonal antibody to NOS I and μNOS I demonstrated the 160-kd NOS I protein in the combined soluble and CHAPS-solubilized particulate fractions of the cerebellum, the endothelium-denuded aorta, and the medial layer alone (Figure 4A⇓). A slightly larger protein (probably μNOS I) was found in gastrocnemius muscle. This larger protein was also seen in cerebellum, endothelium-denuded aorta, and the medial layer alone (Figure 4A⇓). On 7.5% SDS-polyacrylamide gels, the 2 protein bands could not be distinguished (n=3, data not shown). Endothelium-intact and endothelium-denuded aortas did not differ in their NOS I/μNOS I content (n=3, data not shown). Homogenates from total aorta showed a 135-kd NOS III band with the anti–NOS III antibody (Figure 4B⇓). This band disappeared when the endothelium or the endothelium plus adventitia was removed (Figure 4B⇓). Western blots using a polyclonal anti–NOS II antibody demonstrated the 130-kd NOS II protein in LPS-induced aortas but not in normal aortas (n=4, data not shown).
To determine the cellular location of the NOS isoforms in rat aorta, sections were first stained with a polyclonal antibody recognizing NOS I and μNOS I. NOS I/μNOS I immunoreactivity was found in the smooth muscle layer of the aorta (Figure 5A⇓). NOS I/μNOS I immunoreactive material was also detected in the adventitia (Figure 5B⇓). The endothelium showed no NOS I/μNOS I immunoreactivity (Figure 5A⇓). Immunoreactivity was lost when the anti–NOS I antibody was replaced with preimmune serum (Figure 5D⇓ and 5E⇓). Staining with a monoclonal antibody to NOS III was clearly restricted to the endothelium (Figure 5C⇓); Figure 5F⇓ shows a negative control.
Functional Significance of NOS I/μNOS I in Rat Aorta for Regulation of Vascular Tone
When endothelium-denuded aortic rings were constricted with increasing concentrations of KCl (5 to 90 mmol/L), the addition of the NOS inhibitor L-NNA (300 μmol/L) produced a marked increase in the contractile response (Figure 6A⇓). In contrast, when tension was induced with NE (0.1 nmol/L to 1 μmol/L), the maximum contraction remained unchanged in the presence of 300 μmol/L L-NNA (Figure 6B⇓). Only at low concentrations of NE was a small increase in the contractile response detectable with the NOS inhibitor (Figure 6B⇓).
The present study demonstrates that nonendothelial components of rat aorta express 2 subforms of NOS I: the brain-type enzyme and the skeletal muscle-type μNOS I, the former being the predominant form. The overall expression levels of NOS I/μNOS I in endothelium-denuded aortas are low, and only part of this NOS I/μNOS I is found in the smooth muscle layer. The remainder is likely to reside in nitrergic neurons within the adventitia.
Controversial information has been published concerning the presence of NOS in vascular smooth muscle. Recently, it has been shown by Western blot that the smooth muscle layer of rat aorta expresses ≈22% of NOS III protein of the total aorta.12 This expression increased when rats were treated with monocrotaline to induce heart failure.12 On the other hand, Western blots of protein from cultured human aortic smooth muscle cells demonstrated a specific NOS I signal but showed no immunoreactivity for NOS II and NOS III.11 While the present study was under review, another study appeared demonstrating NOS I protein in the media of the rat carotid artery.10 Our Western blot and immunohistochemistry data (Figures 4⇑ and 5⇑) clearly indicate that NOS III is expressed only in the endothelium and not in the medial layer of rat aorta. On the other hand, Western blots and staining of aortic sections with our anti–NOS I antibody showed specific immunoreactivity in only the smooth muscle layer and not in the endothelium of rat aorta. Therefore, our results agree with the data involving NOS I expression in cultured human aortic smooth muscle cells and carotid arteries.10 11
NOS I is subject to alternative splicing.1 In the rat, an additional 102-bp insert between exons 16 and 17 corresponds to an alternatively spliced isoform in skeletal muscle, referred to as μNOS I.5 A μNOS I–specific antibody detected the resulting 164-kd protein in differentiated skeletal muscle and heart.5 μNOS I has been reported to have the same activity as brain-type NOS I.5 Another group cloned the same μNOS I cDNA from rat penile corpora cavernosa.24 This subform was expressed alone in rat penis, urethra, prostate, and skeletal muscle and was coexpressed with brain-type NOS I in the pelvic plexus, bladder, and cerebellum.
Therefore, we decided to investigate whether rat vascular smooth muscle, the third muscle cell type besides skeletal muscle and heart, also expresses μNOS I alone or together with brain-type NOS I. We compared NOS I/μNOS I RNA and protein expression in rat aorta with that of cerebellum (positive control for brain-type NOS I) and with gastrocnemius muscle (positive control for μNOS I). Preliminary studies with rat aorta and cultured endothelial cells have clearly indicated that NOS I/μNOS expression is absent from endothelial cells. Two different RT-PCR approaches and RNase protection analyses have clearly demonstrated that μNOS I mRNA is expressed as a minor form together with brain-type NOS I in the adventitia and the medial layer of the rat aorta (Figures 1 to 3⇑⇑⇑). PCR is nonquantitative by definition. If 2 competing templates are present in different amounts, it is not unusual that only one dominant band is amplified. In Figures 1⇑ and 2⇑, this is shown for cerebellum and aorta. Therefore, only RNase protection analyses have been used for quantification (see Figure 3⇑). Additional evidence for the expression of μNOS I came from the sequencing of a μNOS I fragment cloned from deendothelialized rat aorta. Interestingly, the NOS I/μNOS I ratio in the aortic wall is similar to that in cerebellum, which also expresses μNOS I as a minor splice variant (Figures 1 to 3⇑⇑⇑).
Western blot analyses of gastrocnemius muscle showed a protein that was slightly larger than the 160-kd NOS I protein in cerebellum (Figure 4⇑). The small difference in molecular weight became apparent only when the proteins were separated on 5% SDS-polyacrylamide gels. It seems likely that the larger protein is μNOS I. The larger protein band was also detected in rat cerebellum, where, however, the 160-kd protein prevailed. The endothelium-denuded aorta also expressed the 160-kd NOS I protein together with the slightly larger protein. Removal of the adventitia resulted in a reduction of brain-type NOS I expression, whereas the larger protein remained largely unimpaired.
To demonstrate a potential functional significance of NOS I/μNOS I in the regulation of vascular tone, endothelium-denuded rat aortic rings were constricted with the physiological vasoconstrictor NE. Blocking NOS activity with L-NNA produced only a small increase in tension at lower NE concentrations. NE is known to induce only a transient increase in free intracellular calcium levels in smooth muscle as well as phosphorylation of myosin light chain.25 26 However, free intracellular calcium levels rapidly return to basal or even below basal levels while tone is maintained (for review see Reference 2727 ) Therefore, the calcium-regulated NOS I is probably inactive again at the time when a stable tension plateau is reached. Consequently, in a second approach, we induced tension with the membrane-depolarizing agent KCl. Under these conditions, free intracellular calcium levels stay elevated for a longer period of time, and L-NNA produces a marked increase in contraction. This suggests that under these conditions, NOS I/μNOS I is active and negatively contributes to the tension reached. Thus, under conditions of elevated free calcium levels in vascular smooth muscle (eg, in hypertension), the NOS I/μNOS I system, in addition to the predominant endothelial NOS III, may be relevant in controlling vascular tone.
We conclude from our data that 2 subforms of NOS I are expressed in nonendothelial structures of rat aorta: the brain-type NOS I and the skeletal muscle–type μNOS I, with NOS I being the predominant form. Their distribution in the vascular wall differs from that of skeletal muscle (where μNOS I prevails) and resembles that of the cerebellum (which expresses mainly the NOS I). NOS I/μNOS I may serve as a backup system to the predominant endothelial NOS III in certain pathophysiological conditions.
This study was supported by the Collaborative Research Center SFB 553, Project A1 (to UF) from the Deutsche Forschungsgemeinschaft, Bonn, Germany.
Förstermann U, Kleinert H, Gath I, Schwarz P, Closs EI, Dun NJ. Expression and expressional control of nitric oxide synthases in various cell types. In: Ignarro L, Murad F, eds. Nitric Oxide: Biochemistry, Molecular Biology, and Therapeutic Implications. San Diego, Calif: Academic Press; 1995:171–186.
Gath I, Closs EI, Gödtel-Armbrust U, Schmitt S, Nakane M, Wessler I, Förstermann U. Inducible NO synthase II and neuronal NO synthase I are constitutively expressed in different structures of guinea pig skeletal muscle: implications for contractile function. FASEB J. 1996;10:1614–1620.
Silvagno F, Xia HH, Bredt DS. Neuronal nitric-oxide synthase-mu, an alternatively spliced isoform expressed in differentiated skeletal muscle. J Biol Chem. 1996;271:11204–11208.
Balligand JL, Kobzik L, Han XQ, Kaye DM, Belhassen L, O’Hara DS, Kelly RA, Smith TW, Michel T. Nitric oxide-dependent parasympathetic signaling is due to activation of constitutive endothelial (type III) nitric oxide synthase in cardiac myocytes. J Biol Chem. 1995;270:14582–14586.
Balligand JL, Ungureanu-Longrois D, Simmons WW, Pimental D, Malinski TA, Kapturczak M, Taha Z, Lowenstein CJ, Davidoff AJ, Kelly RA, Smith TW, Michel T. Cytokine-inducible nitric oxide synthase (iNOS) expression in cardiac myocytes: characterization and regulation of iNOS expression and detection of iNOS activity in single cardiac myocytes in vitro. J Biol Chem. 1994;269:27580–27588.
Boulanger CM, Heymes C, Benessiano J, Geske RS, Levy BI, Vanhoutte PM. Neuronal nitric oxide synthase is expressed in rat vascular smooth muscle cells: activation by angiotensin II in hypertension. Circ Res. 1998;83:1271–1278.
Papadaki M, Tilton RG, Eskin SG, McIntire LV. Nitric oxide production by cultured human aortic smooth muscle cells: stimulation by fluid flow. Am J Physiol. 1998;274:H616–H626.
Mitchell JA, Kohlhaas KL, Matsumoto T, Pollock JS, Förstermann U, Warner TD, Schmidt HH, Murad F. Induction of NADPH-dependent diaphorase and nitric oxide synthase activity in aortic smooth muscle and cultured macrophages. Mol Pharmacol. 1992;41:1163–1168.
Brown CW, McHugh KM, Lessard JL. A cDNA sequence encoding cytoskeletal gamma-actin from rat. Nucleic Acids Res. 1990;18:5312.
Kleinert H, Euchenhofer C, Ihrig-Biedert I, Förstermann U. In murine 3T3 fibroblasts, different second messenger pathways resulting in the induction of NO synthase II (iNOS) converge in the activation of transcription factor NF-κB. J Biol Chem. 1996;271:6039–6044.
Schwarz PM, Gierten B, Boissel J-P, Förstermann U. Expressional down-regulation of neuronal-type nitric oxide synthase I by glucocorticoids in N1E-115 neuroblastoma cells. Mol Pharmacol. 1998;54:258–263.
Schmidt HHHW, Gagne GD, Nakane M, Pollock JS, Miller MF, Murad F. Mapping of neural nitric oxide synthase in the rat suggests frequent co-localization with NADPH diaphorase but not with soluble guanylyl cyclase, and novel paraneural functions for nitrinergic signal transduction. J Histochem Cytochem. 1992;40:1439–1456.
Tracey WR, Xue C, Klinghofer V, Barlow J, Pollock JS, Förstermann U, Johns RA. Immunochemical detection of inducible NO synthase in human lung. Am J Physiol. 1994;266:L722–L727.
Horowitz A, Menice CB, Laporte R, Morgan KG. Mechanisms of smooth muscle contraction. Physiol Rev. 1996;76:967–1003.