Vascular smooth muscle and nitric oxide synthase ¨ CKER,§ IGOR B. BUCHWALOW,*,†,1 THOMAS PODZUWEIT,‡ WERNER BO ,† ‡ 储 VERA E. SAMOILOVA,* SYLVIA THOMAS, MAREN WELLNER, HIDEO A. BABA,§ ¨ RGEN SCHNEKENBURGER,* AND MARKUS M. LERCH*,† HORST ROBENEK,¶ JU *Department of Medicine B, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, D-48149 Mu¨nster, Germany; † Central Ultrastructure Research Unit, Interdisciplinary Centre of Clinical Research, c/o Gerhard Domagk Institute of Pathology, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, D-48149 Mu¨nster, Germany; ‡Experimental Cardiology, Max Planck Institute for Clinical Research, D-61231 Bad Nauheim, Germany; §Gerhard Domagk Institute of Pathology, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, D-48149 Mu¨nster, Germany; 储Franz Volhard Clinic, Medical Faculty of the Charite, Humboldt University of Berlin, D-13125 Berlin, Germany; and ¶Institute for Arteriosclerosis Research, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, D-48149 Mu¨nster, Germany The concept of endothelium-derived relaxing factor (EDRF) put forward in 1980 by Furchgott and Zawadzki implies that nitric oxide (NO) produced by NO synthase (NOS) in the endothelium diffuses to the underlying vascular smooth muscle, where it modulates vascular tone as well as vascular smooth muscle cell (VSMC) proliferation by increasing cGMP formation with subsequent activation of cGMP-dependent protein kinase. According to this concept, VSMC do not express NOS by themselves. This attractive, simple scheme is now under considerable debate. To address this issue, we designed this study with the use of a novel supersensitive immunocytochemical technique of signal amplification with tyramide and electron microscopic immunogold labeling complemented with Western blotting, as in our recent studies demonstrating NOS in the myocardial and skeletal muscles. We provide the first evidence that, in contrast to the currently accepted view, VSMC in various blood vessels express all three NOS isoforms depending on the blood vessel type. These findings suggest an alternative mechanism by which local NOS expression may modulate vascular functions in an endothelium-independent manner.— Buchwalow, I. B., Podzuweit, T., Bo¨cker, W., Samoilova, V. E., Thomas S., Wellner, M., Baba, H. A., Robenek, H., Schnekenburger, J., Lerch, M. M. Vascular smooth muscle and nitric oxide synthase. FASEB J. 16, 500 –508 (2002)
ABSTRACT
Key Words: vascular smooth muscle cells 䡠 NOS 䡠 VSMC
Nitric oxide (NO) acts as an inter- and intracellular messenger in various cell types. NO originates via the oxidative l-arginine pathway catalyzed by a family of enzymes termed NO synthase (NOS) isozymes. Neuronal and endothelial NOS (also designated NOS1 and NOS3) were identified in and cloned from neuronal and endothelial cells, respectively; inducible NOS was isolated originally from activated macrophages and was therefore called macrophage NOS (also designated NOS2). 500
Until recently, the expression patterns of NOS isoforms indeed appeared to be cell specific. With the advent of more powerful immunohistochemical techniques based on signal amplification with tyramide (1– 4) increasing antigen detectability ⬍ 1000-fold, which we subsequently used for NOS detection (5–7), the expression patterns of NOS isoforms were found to be less exclusive. The conventional classification of NOS isoforms into neuronal, endothelial, and inducible NOS seemed to reflect only characteristics of the original tissues in which the enzymes were first described. Today, this classification is important only historically and may cause confusion. In the cardiovascular system, all three NOS isoforms have the same implications, physiological or pathophysiological (8). Applying a tyramide signal amplification (TSA) immunocytochemical technique, we earlier demonstrated that cardiomyocytes constitutively express all three NOS isoforms, including inducible NOS (NOS2) (6, 9, 10), formerly believed to be expressed only in response to pathogenic factors. Our data (7) combined with relevant data from other groups (11–14) indicated that skeletal muscles also contain all three isoforms of NOS. Myocardial and skeletal muscle-derived NO is likely to participate in the regulation of contractile function and energy production (6, 13, 14). In the third muscle cell type—vascular smooth muscle cells (VSMC)–NO apparently has similar signaling mechanisms that can potentially control vascular force generation (15). Besides its well-documented role as a vasodilator, NO has also been implicated in VSMC growth arrest (16). However, until recently NO generation in the vascular wall was believed to be restricted to the endothelium, and the vascular smooth muscle was regarded as 1
Correspondence: Department of Medicine B and Central Ultrastructure Research Unit, Interdisciplinary Centre of Clinical Research, c/o Gerhard Domagk Institute of Pathology, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, Domagkstr. 17, D-48149 Mu¨nster, Germany. E-mail:
[email protected] 0892-6638/02/0016-0500 © FASEB
a passive recipient of NO from endothelial cells (17). The concept of endothelium-derived relaxing factor (EDRF), put forward by Furchgott and Zawadzki (18), implied that arterial smooth muscle cell relaxation in response to acetylcholine depended on the anatomical integrity of the endothelium. It was suggested later that EDRF and NO are identical (19). Thus, the EDRF concept implied that NO diffuses passively from the cytosol of its cell of origin into distant target cells to elicit a response (20). More recently, this attractive simple scheme became subject to considerable debate (21, 22). Several observations were inconsistent with EDRF concept. For instance, it was shown that NO was concentrated on the surface of endothelial cells (23) and that NO does not pass as freely into the arterial media from the adventitial side as from the intimal side (24, 25). Thus, several questions remained to be answered, in particular, do VSMC express NOS, does the vascular smooth muscle generate NO by themselves, is the NO generation by VSMC physiologically relevant? The relevant literature is highly controversial (17, 20, 21, 26 –34). We thought this controversy might reflect the diversity of experimental approaches and insufficient sensitivity of the methods used. Therefore, this study designed to elucidate whether VSMC constitutively express NOS was carried out using a novel and ultrasensitive immunocytochemical amplification technique as in our earlier studies demonstrating NOS in myocardial and skeletal muscles (6, 7, 10). Using this powerful immunocytochemical method complemented with electron microscopic immunogold labeling and Western blotting, we examined NOS expression in the media of arteries and veins of Landrace pigs and humans and found that VSMC in various blood vessels do indeed express NOS constitutively under physiological conditions. These findings suggest an alternative mechanism by which local NOS expression may modulate vascular functions in an endothelium-independent manner. NOS has received attention as a possible therapeutic target (35). In view of a potential pathological role for NOS, a better understanding of the NOS regulatory networks in the VSMC may contribute to the development of novel drug and gene therapies for treatment of cardiovascular diseases.
MATERIALS AND METHODS Tissue samples The right carotid artery and circumflex coronary artery were obtained from Landrace pigs. Animals were anesthetized, tracheotomized, and ventilated via endotracheal tubus. The chest was opened by midsternal thoracotomy. Arteries were dissected, excised, and flushed with saline to remove trapped blood. One-half of arteries was denuded by scraping off the endothelium; the other half was left intact. Both halves were cut into 3 mm-thick slices, embedded in Tissue-Tek, and snap-frozen in liquid nitrogen. Cryosections (7 m thick) of VASCULAR SMOOTH MUSCLE AND NOS
porcine arteries were cut at ⫺20°C, mounted on Polysine microslides (Menzel Gla¨ ser, Braunschweig, Germany), thoroughly air-dried, fixed in acetone for 10 min at room temperature (RT), air-dried, and stored at ⫺20°C until use. The remaining arterial rings were pooled and analyzed by Western blot. Experiments were performed in accordance with the Helsinki Declaration and The Guiding Principles in the Care and Use of Animals. For human studies, paraffin-embedded tissue specimens of aorta, Arteria radialis, Arteria mammaria, kidney, heart, and pancreas were obtained from 5 individuals and classified as normal based on clinical features and histological findings. Cardiac biopsies were from male patients suffering from dilatative cardiomyopathy. The procedure for the use of human tissue from surgical pathology was approved by the institutional review board and performed in accordance with the ethical standards of the regional committee on human studies. For immunocytochemical analysis, routine paraffin sections of formaldehyde-fixed human biopsies were mounted on Polysine microslides. Antibodies and immunohistochemical staining For antigen retrieval, dewaxed and rehydrated sections of human biopsies were immersed in 10 mmol/L citric acid, pH 6.0, and boiled in a pressure cooker at an operating pressure of ⬃103 kPa/15 psi for 2 min. After cooling the pressure cooker until all pressure was released, the slides were removed, quickly washed in running distilled water, and transferred to PBS solution. Subsequently, sections were encircled with a water-repellent PAP-pen (Dianova, Hamburg, Germany) and rinsed with PBS. Cryosections of porcine arteries were processed without antigen retrieval. PBS was used for all dilutions and washing steps. After blocking nonspecific binding sites (Fc receptors) with BSA-c basic blocking solution (1:10 in PBS; Aurion, Wageningen, The Netherlands), sections were immunoreacted overnight at 4°C with rabbit primary polyclonal antibodies (AB) recognizing NOS1, NOS2, and NOS3. Anti-NOS1 AB were raised against polypeptides corresponding to NOS1 epitope at the carboxyl terminus (a. a. residues 1095–1289, Transduction Laboratories, Lexington, KY; a. a. residues 1400 –1419, Santa Cruz Biotechnology, Santa Cruz, CA). Anti-NOS2 AB were raised against polypeptides corresponding to NOS2 epitope at either the amino terminus (a. a. residues 3–22, Santa Cruz Biotechnology) or the carboxyl terminus (a. a. residues 1126 –1144, Santa Cruz Biotechnology; a. a. residues 961-1144, Transduction Laboratories; a. a. residues 1131–1144, Biomol Feinchemikalien GmbH, Hamburg, Germany). Anti-NOS3 AB were raised against polypeptides corresponding to NOS3 epitope at the carboxyl terminus (a. a. residues 1030 –1209, Transduction Laboratories). Primary AB were diluted to a final concentration of 0.25– 0.5 g/ml. As in our previous study of myocardial muscle (6), immunolabeling of NOS in blood vessels with AB purchased from these manufacturers provided similar results irrespective of the mode of visualization: bright-field or fluorescent microscopy. After immunoreacting with primary AB and washing in PBS, the sections were treated for 10 min with methanol containing 0.6% H2O2 to quench endogenous peroxidase. For fluorescent visualization of bound primary AB, sections were treated for 1 h at RT with horseradish peroxidase (HRP) -conjugated AffiniPure goat anti-rabbit IgG (H⫹L) (Dianova) at a dilution of 1:100 in PBS. These secondary AB purchased from Dianova were passed through solid-phase immunoadsorbent gels to remove IgG, which cross-react with human and pig serum proteins. The HRP label was amplified with FITC-conjugated tyramine. FITC-tyramine conjugate was synthesized from tyramine-HCl (Sigma) and FITC-succinimidyl 501
ester (NHS-FITC, Pierce, Rockford, IL) in DMSO. Incubation with FITC-tyramine conjugate was carried out at a dilution of 1:300 in PBS in the presence of 0.02% H202 for 10 min. For fluorescent microscopy, samples were counterstained for 15 s with DAPI (5 g/ml PBS; Sigma) and mounted with Vectashield (Vector Laboratories, Burlingame, CA). For bright-field microscopy, bound primary AB were visualized with double amplification using HRP-avidin-biotin complex (Vectastain ‘Elite’ ABC kit, Vector Laboratories) in combination with biotin-labeled tyramine (NEN Life Science Products, Boston, MA) following guidelines recommended by the manufacturers. To visualize the HRP label, samples were incubated with diaminobenzidine-H2O2 mixture, counterstained with Ehrlich hematoxylin for 30 s, and mounted with an aqueous mounting medium GelTol (Immunotech, Marseille, France). To visualize NOS simultaneously with either calponin (marker for VSMC) or CD31 (marker for endothelial cells), some sections were immunoreacted with anti-NOS AB in the presence of anti-human CD31 mouse monoclonal AB (PharMingen, Hamburg, Germany) or anti-human calponin mouse monoclonal AB (DAKO, Hamburg). Bound anticalponin and anti-CD31 AB were detected with Cy3-conjugated AffiniPure goat anti-mouse IgG (H⫹L) (minimal crossreaction with human serum proteins; Dianova). The controls were 1) omission of incubation with primary AB; 2) substitution of primary AB by rabbit IgG (Dianova) at the same final concentration; and 3) incubation in media containing primary AB that had been preincubated at RT for 2 h with a 10-fold molar excess of corresponding control peptides (Santa Cruz). Immunogold labeling of ultrathin cryosections Tissue blocks (1⫻1 mm) of the human aorta and coronary artery were fixed with 5% formaldehyde in 0.2 M piperazine1,4-bis(2-ethanesulfonic acid) (PIPES), pH 7.0, for 2 h at RT. After washing in PIPES buffer, the fixed tissue blocks were infused with a mixture containing 1.6 M sucrose and 25% (w/v) polyvinylpyrrolidone (Mr⫽10 000), snap-frozen in liquid nitrogen, and mounted on silver pins (36). Ultrathin frozen sections of the specimens were cut at ⫺100°C (chamber) and ⫺90°C (knife) with a Reichert-Jung FC 4 E with a diamond knife. Sections were picked up on formvar-coated copper grids (150 mesh) and immunolabeled at RT. Immunocytochemical labeling was performed by floating grids serially, section side down, on 5–20 l droplets. Steps included washing with PBS, incubation with anti-NOS AB (2.5 g/ml; Transduction Laboratories) for 1 h, washing with PBS, incubation with protein A gold for 1 h, and washing with PBS. Protein A gold particles were prepared according to the protocol of Roth et al. (37) using gold sols, with 10 nm average diameters prepared according to the method of Frens (38). After labeling, sections were washed intensively with distilled water. Subsequent staining included a 10 min uranyl acetate oxalate (pH 7.4) stain, three 1 min rinses in distilled water, and a 10 min 2% uranyl acetate stain in distilled water. The grids were then floated on drops of 1% tylose in 0.5% uranyl acetate three times for 1 min each at 0°C. After thorough washing in distilled water, the grids were air-dried and examined under a Philips 410 electron microscope. Microscopy and image processing Immunostained sections were examined on a Zeiss Axioskop microscope equipped with appropriate filters. Separate images for DAPI staining, fluorophore (FITC or Cy3) immuno502
Vol. 16
April 2002
labeling, and autofluorescence of elastic laminae and erythrocytes were captured digitally into color-separated components using an AxioCam digital microscope camera and AxioVision multi-channel image processing (Carl Zeiss Vision GmbH, Germany). The blue (for DAPI), red (for Cy3 or elastic laminae autofluorescence), and green (for FITC) components were merged and composite images were imported as BMP files into PhotoImpact 3.0 (Ulead Systems, Inc. Torrance, CA) for analysis on Power PC, followed by printing on a color printer Hewlett Packard DeskJet 970Cxi. Bright-field microscopy images were captured using AxioCam 12-bit camera and AxioVision single-channel image processing. Images shown are representative of 6 independent experiments that gave similar results. Western blotting Porcine carotid artery rings (500 mg) were homogenized in 500 l homogenization buffer (10 mM Tris-HCl pH 7.4, 100 mM NaCl, 3 mM MgCl2, 300 mM sucrose, 1 mM EDTA, 0.2 mM PMSF, 1 g/ml leupeptin, 0.5% Triton X 100, 0.02% Na azide) for 10 min on ice. The protein concentration was determined using Bradford reagent. For proteins with a mass of ⬃130 kDa, 6% acrylamide gels were used; 50 g of protein was loaded. The electrophoresis was carried out at 120 V. Protein transfer onto PVDF membranes was performed using a semidry blotting apparatus (Millipore, Bedford, MA) with 2.5 mA/cm2 for 1 h. Blocking was carried out in 5% nonfat skim milk (Bio-Rad, Mu¨ nchen, Germany) for 1 h at RT. Antibodies were diluted in 5% nonfat skim milk TBS-T (NOS1 1:500, NOS2 1:500, NOS3: 1:250; from Transduction Laboratories, and NOS2 1:5000 from Biomol) and incubated overnight at 4°C. Positive controls for NOS1, NOS2, and NOS3 were from Transduction Laboratories. The signal detection was performed using ECL-system (Amersham, Braunschweig, Germany).
RESULTS Immunocytochemical demonstration of NOS isoforms in porcine blood vessels The TSA immunocytochemical technique combined with the avidin-biotin complex (ABC), a so-called ‘blast’ amplification with diaminobenzidine as chromogen (Fig. 1), as well as TSA with FITC-tyramide (Fig. 2), revealed the presence of all three NOS isoforms in both the intima and media of the pig carotid artery (elastic type). Similar coexpression of all three NOS isoforms could be seen in the muscular-type circumflex coronary artery (not shown). Reducing the concentration of primary AB down to 0.25– 0.5 g/ml, we brought background labeling to a negligible level in bright-field and fluorescent microscopy. However, autofluorescence of elastic lamellae in the blood vessel wall strongly limits the detectability of fluorescent probes. This autofluorescence is visible in Fig. 2d representing the IgG control. We overcame this technical problem by triple exposure of the specimen to illumination with three filters exciting the fluorescence in the blue, green, and red spectra (365, 450, and 546 nm, respectively). Separate images were captured digitally into color-separated components using a digi-
The FASEB Journal
BUCHWALOW ET AL.
human pancreas. The same expression pattern in venous blood vessels was found for NOS2 (Fig. 3D) and NOS3 (not shown). In arterioles, all three NOS isoforms were coexpressed in both the VSMC and endothelium, as shown for NOS1 in Fig. 3C. In contrast to arterioles, arteria as a rule did not reveal positive immunoreaction for NOS1 and NOS2 in the intima, whereas the medial cells in arteries of the muscular type exhibited an equally strong coexpression of all three NOS isoforms, as shown for NOS1 in an artery from the human pancreas (Fig. 4). A similar pattern of NOS expression was observed in arterioles and arteries of the muscular type in tissue probes from myocardia and kidney (not shown). NOS expression in VSMC in the human blood vessels of the elastic type was markedly lower than in muscular-type arteries. In the aorta, only a small fraction of VSMC immunoreacted positively with anti-NOS AB, as shown for NOS1 in Fig. 5. In another blood vessel of the elastic type, Arteria mammaria, VSMC also ex-
Figure 1. DAB-HRP immunostaining of NOS1 (a), NOS2 (b), and NOS3 (c) in porcine carotid artery. The chromogen reaction with DAB (brown color) visualizes the HRP label amplified with biotinylated tyramine and ABC method (‘blast’ amplification). d) A control performed with substitution of primary AB with rabbit IgG at the same concentration. Nuclei are counterstained with Mayer’s hematoxylin. Nomarski contrast. 50 m scale bar for entire layout.
tal microscope camera and multi-channel image processing (see Materials and Methods). The red, blue, and green components were merged and the elastic lamellae appeared in a pseudo-lilac color. Varying the ratio of exposure times and using different filters, it is possible to represent autofluorescent components in different colors (e.g., red). This approach allowed us to discriminate between specific immunolabeling and nonspecific autofluorescence emitted by elastic lamellae and erythrocytes, as shown in Figs. 3 and 5. Immunocytochemical demonstration of NOS isoforms in the human blood vessels To substantiate NOS expression in both VSMC and endothelium, we carried out simultaneous immunolabeling of NOS with either a VSMC marker, calponin, or an endothelial marker, CD31. Figure 3A, B demonstrates localization of NOS1 in VCMC and the coexpression of this isoform with CD31 in the endothelium of a venous blood vessel in VASCULAR SMOOTH MUSCLE AND NOS
Figure 2. Immunofluorescent demonstration of NOS in the porcine carotid artery. FITC-tyramide (green fluorescence) marks NOS1 (a), NOS2 (b), and NOS3 (c) in the media and intima. Lilac color of elastic laminae (a– c) is the sum of three images picked up under illumination with three filters exciting fluorescence in the green, red, and blue spectra. d) IgG control: a single image captured with the filter exciting fluorescence in the green spectrum. Separate images were captured digitally into color-separated components using a AxioCam 12-bit and AxioVision multi-channel image processing (see Materials and Methods). This approach allows discrimination between specific immunolabeling and nonspecific autofluorescence emitted by elastic laminae as visible in the control preparation (d). 50 m scale bar for entire layout. 503
Western blotting Specificity of anti-NOS AB purchased from Transduction Laboratories, Santa Cruz, and Biomol was earlier confirmed by us with Western blotting for myocardial and skeletal muscles (6, 7). Anti-NOS AB from Transduction Laboratories also positively reacted with characteristic bands for NOS1, NOS2, and NOS3 on Western blots of the human pancreas and kidney (n⫽3, data not shown). Here (Fig. 8) Western blotting of the porcine carotid artery with anti-NOS AB (Transduction Laboratories) shows the presence of characteristic immunoreacting protein bands for NOS1, NOS2, and NOS3 not only in the intact carotid artery (Fig. 8, lane A), but also in the carotid artery devoid of endothelium (Fig. 8, lane B), indicating the coexpression of these three isoforms in the VSMC. The endothelium also showed a strong immunoreactivity with all three anti-
Figure 3. Immunofluorescent demonstration of NOS in the blood vessel wall of normal human pancreas. A) Colocalization of NOS1 and CD31 in the endothelium of a venous blood vessel is marked by orange (arrows), formed by a mixture of green (tyramide-FITC for NOS1) and red fluorescence (Cy3 for CD31). *Erythrocytes eliciting a strong red autofluorescence. B) Same as panel A, but with a 25 pixel shift in the red component of the composite image to the left. C) An arteriole exhibits strong NOS1 expression (green label) in both the endothelium and VSMC. Red is accounted for by autofluorescence of the lamina elastica interna and that of erythrocytes. D) Localization of NOS2 (green label) in the media and intima of a venous blood vessel. Autofluorescent erythrocytes in the lumen are in red. Nuclei are counterstained with DAPI. 20 m scale bar for entire layout.
pressed all three NOS isoforms (Fig. 6a– c). NOS1 expression was more pronounced than that of NOS2 and NOS3, but generally still lower than in VSMC of muscular-type arteries. The endothelium in this artery immunoreacted positively only with NOS3 AB (Fig. 6c). Electron microscopic immunogold labeling Immunogold labeling of the human coronary artery and aorta revealed NOS targeting to the contractile fibers in VSMC. As shown for NOS1, immunolabeling of contractile fibers in VSMC in the coronary artery (Fig. 7a) was markedly more intensive than in the aorta (Fig. 7b), which is in accord with a more pronounced immunofluorescent NOS labeling of VSMC in muscular-type arteries vs. that of VSMC in elastic-type arteries (cf. Figs. 4 – 6). 504
Vol. 16
April 2002
Figure 4. Double immunofluorescent labeling of NOS1 and calponin in an artery (muscular type) in the human pancreas. a) A composite image exhibiting a hybrid orange color resulting from a coincidence of two labels: Cy3 for calponin (red) and FITC-tyramide for NOS1 (green). A pure green accounts for autofluorescence of the lamina elastica interna, which can also be seen in panels b and c. b) Same as panel a, but with a 300 pixel shift of the red component of the composite image (a) downward in the direction of the lumen. c) Autofluorescence image of the lamina elastica interna in the control section incubated with rabbit IgG instead of primary AB at the same concentration of 0.5 g/ml. Nuclei are counterstained with DAPI. 50 m scale bar for entire layout.
The FASEB Journal
BUCHWALOW ET AL.
VSMC of the porcine carotid artery and complemented these findings with Western blotting of endotheliumdeprived preparations. In human blood vessels, VSMC were also found to contain all three NOS isoforms. We found that NOS expression by VSMC in blood vessels of the elastic type (human aorta and Arteria mammaria) was markedly lower than in arteries of the muscular type and in arterioles. This is in accord with indications in the literature that blood vessels may possess different physiological characteristics and respond differently to established relaxants such as acetylcholine even in the endothelium-independent mode (39 – 41). It seems that the expression level of definite NOS isoforms varies depending on the blood vessel type. A correlation between the NOS expression and blood vessel classification deserves further study and is now in progress. The main goal of the present investigation was to demonstrate that the NOS is an essential constituent of the enzyme setting in VSMC. These data complement our previous reports on NOS expression by two other muscle types: myocardial (6, 9, 10) and skeletal (7). However, it remains a matter of debate whether NO
Figure 5. Immunofluorescent labeling of NOS1 in the human aorta. a, b) Green fluorescent FITC-tyramide marks NOS1 in only a few VSMC in the media, mainly in the vicinity of the lumen. Most VSMC reveal no staining. Red autofluorescence of elastic fibers was captured with a filter exciting the autofluorescence in red spectrum under a longer exposure than with the filter exciting fluorescence in the green spectrum. Nuclei are counterstained with DAPI.
NOS AB (lane C). Similar results for the porcine carotid artery were obtained with anti-NOS2 AB (purchased from Biomol; n⫽3, data not shown). No immunoreactivity was detected when anti-NOS AB were replaced with rabbit IgG at the same concentration. NOS1 appears as a prevailing NOS isoform in the blood vessel media and seems to be represented by two subforms. The upper minor band probably corresponds to NOS1, a slightly larger protein found earlier in gastrocnemius muscle and rat aorta (32). NOS3 AB immunoreacts with a double protein band. NOS3 seems to be a minor NOS isoform expressed by VSMC, which agrees with immunocytochemical images of the carotid artery (Figs. 1, 2).
DISCUSSION We have shown in this study that VSMC of various blood vessels express NOS (NOS1, NOS2, and NOS3) constitutively under physiological conditions. Increasing antigen detectability with the TSA technique, we demonstrated the coexpression of three NOS isoforms in the VASCULAR SMOOTH MUSCLE AND NOS
Figure 6. Immunofluorescent labeling of NOS isoforms in the human Arteria mammaria. a) NOS1 in most VSMC. b) NOS2 in rare VSMC in the vicinity of the lamina elastica interna. Blood cells are visible in the lumen. c) Weak immunostaining of NOS3 in rare VSMC and strong immunostaining of the intima cells. Nuclei are counterstained with DAPI. 50 m scale bar for entire layout. 505
Figure 7. Immunogold labeling of NOS 1 in the human coronary artery (a) and aorta (b). Immunogold labeling of NOS1 targeted contractile fibers of VSMC in the coronary artery (a) much more intensively than in aorta (b). No background labeling in the nuclei (N) or extracellular matrix (asterisks). 0.5 m scale bar for entire layout.
production by VSMC is physiologically relevant in contractile functions. Zehetgruber et al. (42) demonstrated that NO is released by VSMC as well as by the endothelium, but the amount of NO released by the smooth muscle was insufficient to relax endothelium-deprived vascular preparations of bovine pulmonary arteries. It was later found that medial cells of bovine carotid and human renal arteries (43) and media of the rat aorta (32) do generate NO in amounts sufficient to modulate vascular contractility. The inconsistent results might be explained by different characteristics of experimental models; smooth muscle of different blood vessels respond differently to the relaxant effect of EDRF (39, 44). It should be noted that destruction of the vascular wall integrity during endothelial denudation increases vascular superoxides (45– 47), impairing vasodilator responses to endogenous and exogenous nitrovasodilators (48). Known as NO scavengers, superoxides drastically reduce NO bioactivity and NO bioavailability (10, 49), whereas the intact endothelium protects VSMC from the superoxide attack (50, 51). Moreover, it has been proposed that endothelium-derived hyperpolarizing factor (EDHF) contributes to microvascular dilation more than does NO (52). Background-K⫹ channel activation and myoendothelial gap junctional communications in VSMC play a major role in EDHF-mediated 506
Vol. 16
April 2002
relaxations (53), but the endothelial denudation impairs K⫹ vasorelaxation as well (54). As a result, the increase in vascular superoxides through endothelial denudation (48), the protective role of the endothelium against the superoxide attack (47, 48), reduction of NO bioactivity by superoxides (10, 45), and the abolishment of EDHF by endothelial denudation (54), when taken together, imply that the EDRF concept is not free of assumptions and oversimplifications. The cornerstone of this concept—inability of the smooth muscle to express NO-synthase constitutively—now appears to be imaginary. NOS isoforms present in VSMC are likely to play a role in the global regulation of blood pressure. In our study, NOS expression in VSMC of various blood vessels was demonstrated by bright-field and fluorescent immunolabeling, and by electron microscopic immunogold staining. Immunogold labeling revealed NOS targeting to contractile fibers of VSMC. Localization of NOS in VSMC along contractile fibers, as in the myocard (6), together with data on the inverse correlation between the expression of NOS and contractile force of skeletal muscles (11, 14), allows us to postulate that the endogenously derived NO regulates VSMC contractility as well. In skeletal muscles, this effect has been ascribed to reactive oxygen intermediates, which are thought to occur through reactions with regulatory thiols (14, 55). The same mechanism can be postulated for vascular smooth muscle. To summarize, our data allowed us to draw several conclusions. In contrast to the commonly accepted view, VSMC express all three NOS isoforms constitutively, which may represent an alternative mechanism whereby local NOS expression modulates vascular func-
Figure 8. Western blotting of NOS isoforms in the porcine carotid artery. A: Carotid artery with endothelium; B: carotid artery devoid of endothelium; C: endothelium of carotid artery; D: positive control. These gels are representative of 3 experiments with similar results.
The FASEB Journal
BUCHWALOW ET AL.
tions in an endothelium-independent manner. The expression pattern varies depending on the type of blood vessels; NOS expression by VSMC in larger blood vessels of the elastic type appears to be lower than in muscular-type arteries and arterioles. NOS targeting to contractile fibers implies an important regulatory role for endogenously generated NO in the control of the vasodilatory functions of VSMC. We developed a digitally processed approach that allows us to discriminate between specific immunolabeling and nonspecific autofluorescence emitted by elastic lamellae in blood vessels. Demonstrating constitutive NOS expression in VSMC may lead to a better understanding of the NOS regulatory networks and contribute to the development of novel drug and gene therapies for the treatment of cardiovascular diseases. The expert technical assistance of Mrs. Jana Czychi is greatly acknowledged.
13. 14.
15.
16.
17. 18. 19. 20. 21.
REFERENCES 1.
2. 3.
4.
5. 6.
7.
8. 9.
10.
11.
12.
Bobrow, M. N., Harris, T. D., Shaughnessy, K. J., and Litt, G. J. (1989) Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Methods 125, 279 –285 Adams, J. C. (1992) Biotin amplification of biotin and horseradish peroxidase signals in histochemical stains. J. Histochem. Cytochem. 40, 1457–1463 von Wasielewski, R. Mengela, M., Gignaca. S., Wilkensa, L., Wernera, M., and Georgiia, A. (1997) Tyramine Amplification Technique in Routine Immunohistochemistry. J. Histochem. Cytochem. 45, 1455–1460 Hopman, A. H. N., Ramaekers, F. C. S., and Speel, E. J. M. (1998) Rapid synthesis of biotin-, digoxigenin-, trinitrophenyl-, and fluorochrome-labeled tyramides and their application for in situ hybridization using CARD amplification. J. Histochem. Cytochem. 46, 771–777 Buchwalow, I. B. (2001) Increasing the power of immunohistochemistry. Proc. Roy. Microsc. Soc. 36, 57–59 Buchwalow, I. B., Schulze, W., Karczewski, P., Kostic, M. M., Wallukat, G., Morwinski, R., Krause, E. G., Muller, J., Paul, M., Slezak, J., Luft, F. C., and Haller, H. (2001) Inducible nitric oxide synthase in the myocard. Mol. Cel. Biochem. 217, 73– 82 Punkt, K., Zaitsev, S., Park, J. K. Wellner, M., and Buchwalow, I. B. (2001) Nitric oxide synthase isoforms I, III and protein kinase-Cq in skeletal muscle fibres of normal and streptozotocin-induced diabetic rats with and without ginkgo biloba extract treatment. Histochem J. 33, 213–219 Li, H. G., and Forstermann, U. (2000) Nitric oxide in the pathogenesis of vascular disease. J. Pathol. 190, 244 –254 Buchwalow, I. B., Schulze, W., Kostic, M. M., Wallukat, G., and Morwinski, R. (1997) Intracellular localization of inducible nitric oxide synthase in neonatal rat cardiomyocytes in culture. Acta Histochem. 99, 231–240 Buchwalow, I. B., Schulze, W., Slezak, J., Luft, F. C., and Haller, H. (1998) Biochemical and physical factors involved in reducing nitric oxide bioactivity in hypertensive heart and kidney. Circulation 98, 604 Gath, I., Loss, I., Godtelarmbrust, U., Schmitt, S., Nakane, M., Wessler, I., and Fo¨ rstermann, U. (1996) 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. 10, 1614 –1620 Gath, I., Ebert, J., Godtel-Armbrust, U., Ross, R., Reske-Kunz, A. B., and Fo¨ rstermann, U. (1999) NO synthase II in mouse skeletal muscle is associated with caveolin 3. Biochem. J. 340, 723–728
VASCULAR SMOOTH MUSCLE AND NOS
22.
23. 24. 25.
26.
27. 28.
29.
30.
31.
32.
33.
Kobzik, L., Reid, M. B., Bredt, D. S., and Stamler, J. S. (1994) Nitric oxide in skeletal muscle. Nature (London) 372, 546 –548 Kobzik, L., Stringer, B., Balligand, J. L., Reid, M. B., and Stamler, S. (1995) Endothelial type nitric oxide synthase in skeletal muscle fibers: mitochondrial relationships. Biochem. Biophys. Res. Commun. 221, 375–381 Ignarro, L. J. (1990) Haem-dependent activation of guanylate cyclase and cyclic GMP formation by endogenous nitric oxide: a unique transduction mechanism for transcellular signaling. Pharmacol. Toxicol. 67, 1–7 Sandirasegarane, L., Charles, R., Bourbon, N., and Kester, M. (2000) NO regulates PDGF-induced activation of PKB but not ERK in A7r5 cells: implications for vascular growth arrest. Am. J. Physiol. 279, C225–C235 Luscher, T. F., and Barton, M. (1997) Biology of the endothelium. Clin. Cardiol. 20, 3–10, Suppl. 2 Furchgott, R. F., and Zawadzki, J. V. (1980) The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature (London) 288, 373–376 Palmer, R. M. J., Ferridge, A. G., and Moncada, S. (1987) Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature (London) 327, 524 –526 Gaston, B., Drazen, J. M., Loscalzo, J., and Stamler, J. S. (1994) The biology of nitrogen oxides in the airways. Am. J. Respir. Cell. Mol. Biol. 149, 538 –551 Van Zwieten, P. A. (1997) Endothelial dysfunction in hypertension. A critical evaluation. Blood Press Suppl. 2, 67–70 Boulanger, C. M., Heymes, C., Benessiano, J., Geske, R. S.; Levy, B. I., and Vanhoutte, P. M. (1998) Neuronal nitric oxide synthase is expressed in rat vascular smooth muscle cells: activation by angiotensin II in hypertension. Circ. Res. 28, 1271–1278 Malinski, T., and Taha, Z. (1992) Nitric oxide release from a signal cell measured in situ by a porphyrinic-based microsensor. Nature (London) 358, 676 – 678 Steinhorn, R. H., Morin, F. C. I., and Russell, J. A. (1994) The adventitia may be a barrier specific to nitric oxide in rabbit pulmonary artery. J. Clin. Invest. 94, 1883–1888 Kozlov, A. V., Sobhian, B., Costantino, G., Nohl, H., Redl, H., and Bahrami, S. (2001) Experimental evidence suggesting that nitric oxide diffuses from tissue into blood but not from blood into tissue. Biochim. Biophys. Acta 1536, 177–184 Banning, A. P., Groves, P. H., Buttery, L. D. K., Wharton, J., Rutherford, R. A. D., Black, P., Winkler, F., Polak, J. M., Lewis, M. J., and Drexler, H. (1999) Reciprocal changes in endothelial and inducible nitric oxide synthase expression following carotid angioplasty in the pig. Atherosclerosis 145, 17–32 Fleming, I., and Busse, R. (1999) NO: the primary EDRF. J. Mol. Cell. Cardiol. 31, 5–14 Mohaupt, M. G., Elzie, J. L., Ahn, K. Y., Clapp, W. L., Wilcox, C. S., and Kone, B. C. (1994) Differential expression and induction of mRNAs encoding two inducible nitric oxide synthases in rat kidney. Kidney Int. 46, 653– 665 Comini, L., Bachetti, T., Gaia, G., Pasini, E., Agnoletti, L., Pepi, P., Ceconi, C., Curello, S., Ferrari, R. (1996) Aorta and skeletal muscle NO synthase expression in experimental heart failure. J. Mol. Cell. Cardiol. 28, 2241–2248 Rajasekaran, M., Mondal, D., Agrawal, K., Chen, I. L., Hellstrom, W., and Sikka, S. (1998) Ex vivo expression of nitric oxide synthase isoforms (eNOS/iNOS) and calmodulin in human penile cavernosal cells. J. Urol. 160, 2210 –2215 Trovati, M., Massucco, P., Mattiello, L., Costamagna, C., Aldieri, E., Cavalot, F., Anfossi, G., Bosia, A., and Ghigo, D. (1999) Human vascular smooth muscle cells express a constitutive nitric oxide synthase that insulin rapidly activates, thus increasing guanosine 3⬘: 5⬘-cyclic monophosphate and adenosine 3⬘: 5⬘-cyclic monophosphate concentrations. Diabetologia 42, 831– 839 Schwarz, P. M., Kleinert, H., and Forstermann, U. (1999) Potential functional significance of brain-type and muscle-type nitric oxide synthase I expressed in adventitia and media of rat aorta. Arteriosclerosis Thromb. Vasc. Biol. 19, 2584 –2590 Mizutani, K., Ikeda, K., Ito, T., Tamaki, K., Nara, Y., and Yamori, Y. (2000) Protective effect of inducible type nitric oxide synthase against intracellular oxidative stress caused by advanced glycation end-products in vascular smooth muscle cells from
507
34.
35. 36. 37.
38. 39. 40.
41.
42. 43. 44. 45.
508
stroke-prone spontaneously hypertensive rats. J. Hypertens. 18, 1071–1079 Kahn, A. M., Allen, J. C., Seidel, C. L., Lichtenberg, D. S., Song, T., Zhang, S. (2000) Insulin increases NO-stimulated guanylate cyclase activity in cultured VSMC while raising redox potential. Am. J. Physiol. 278, E627–E633 Nathan, C. (1997) Inducible nitric oxide synthase: What difference does it make. J. Clin. Invest. 100, 2417–2413 Tokuyasu, K. T. (1980) Immunocytochemistry of ultrathin frozen sections. Histochem. J. 12, 381– 403 Roth, J., Bendayan, M., and Orci, L. (1982) The protein-A-gold (pAG) technique—a qualitative and quantitative approach for antigen localization on thin sections. In Techniques in Immunocytochemistry (Bullock, G. R., and Petrusz, P., eds) pp. 107–133, Academic Press, London Frens, G. (1973) Controlled nucleation for the regulation of the particle size in monodisperse gold solutions. Nature Phys. Sci. 241, 20 –22 Shikano, K., Ohlstein, E. H., and Berkowitz, B. A. (1987) Differential selectivity of endothelium-derived relaxing factor and nitric oxide in smooth muscle. Br. J. Pharmacol. 92, 483– 485 Pesic, S., Grbovic, L., Jovanovic, A., and Radenkovic, M. (1999) Acetylcholine-induced contraction in the porcine internal mammary artery: Possible role of muscarinic receptors. J. Vet. Med. Ser. A 46, 509 –515 Teoh, H., Leung, S. W. S., Quan, A., Huang, M., Man, G. S. K., and Man, R. Y. K. (2000) Acetylcholine-mediated relaxation in rat thoracic aorta is enhanced following acute exposure to physiological concentrations of 17-estradiol. Mol. Cell. Biochem. 207, 65–70 Zehetgruber, M., Conforto, A., and Bing, R. J. (1993) Vascular smooth muscle and nitric oxide. Life. Sci. 52, 1397–1406 Brophy, C. M., Knoepp, L., Xin, J. D., and Pollock, J. S. (2000) Functional expression of NOS 1 in vascular smooth muscle. Am. J. Physiol. 278, H991–H997 Griffith, T. M., and Henderson, A. H. (1989) EDRF and the regulation of vascular tone. Int. J. Microcirc. Clin. Exp. 8, 383–96 Azevedo, L. C. P., Pedro, M. D., Souza, L. C., de Souza, H. P., Janiszewski, M., da Luz, P. L., and Laurindo, F. R. M. (2000) Oxidative stress as a signaling mechanism of the vascular
Vol. 16
April 2002
46.
47. 48.
49. 50. 51. 52.
53. 54.
55.
response to injury: The redox hypothesis of restenosis. Cardiovasc. Res. 47, 436 – 445 Ozel, S. K., Yuksel, M., Haklar, G., Durakbasa, C. U., Dagli, T. E., and Aktan, A. O. (2001) Nitric oxide and endothelin relationship in intestinal ischemia/reperfusion injury (II). Prostaglandins Leukot. Essent. Fatty Acids 64, 253–257 Pernow, J., Gonon, A. T., and Gourine, A. (2001) The role of the endothelium for reperfusion injury. Eur. Heart J. Suppl. 3, C22–C27 Heitzer, T., Wenzel, U., Hink, U., Krollner, D., Skatchkov, M., Stahl, R. A. K., Macharzina, R., Brasen, J. H., Meinertz, T., and Munzel, T. (1999) Increased NAD(P)H oxidase-mediated superoxide production in renovascular hypertension: Evidence for an involvement of protein kinase C. Kidney Int. 55, 252–260 Napoli, C., and Ignarro, L. J. (2001) Nitric oxide and atherosclerosis. Nitric Oxide Biol. Chem. 5, 88 –97 Linas, S. L., and Repine, J. E. (1997) Endothelial cells protect vascular smooth muscle cells from H2O2 attack. Am. J. Physiol. 272, F767–F773 Walia, M., Sormaz, L., Samson, S. E., Lee, R. M. K. W., and Grover, A. K. (2000) Effects of hydrogen peroxide on pig coronary artery endothelium. Eur. J. Pharmacol. 400, 249 –253 Miura, H., Liu, Y. P., and Gutterman, D. D. (1999) Human coronary arteriolar dilation to bradykinin depends on membrane hyperpolarization— contribution of nitric oxide and Ca2⫹-activated K⫹ channels. Circulation 99, 3132–3138 Ding, H., and Triggle, C. R. (2000) Novel endothelium-derived relaxing factors—identification of factors and cellular targets. J. Pharmacol. Toxicol. Methods 44, 441– 452 Harris, D., Martin, P. E. M., Evans, W. H., Kendall, D. A., Griffith, T. M., and Randall, M. D. (2000) Role of gap junctions in endothelium-derived hyperpolarizing factor responses and mechanisms of K⫹ relaxation Eur. J. Pharmacol. 402, 119 –128 Stuehr, D. J., Cho, H. J., Kwon, N. S., Weise, M. F., and Nathan, C. F. (1991) Purification and characterization of the cytokineinduced macrophage nitric oxide synthase: An FAD- and FMNcontaining flavoprotein. Proc. Natl. Acad. Sci. USA 88, 7773–7777
The FASEB Journal
Received for publication November 6, 2001. Accepted for publication December 14, 2001.
BUCHWALOW ET AL.
