Evidence That Expression of Inducible Nitric Oxide Synthase in Response to Endotoxin Is Augmented in Atherosclerotic Rabbits
Abstract Atherosclerotic lesions contain monocytes/macrophages and vascular smooth muscle cells and thus may have an increased capacity for generation of nitric oxide by inducible nitric oxide synthase (NOS). We used three approaches (contractile responses, generation of l-citrulline from l-arginine, and staining with NADPH-diaphorase) to test the hypothesis that after administration of lipopolysaccharide (LPS) in vivo, generation of nitric oxide by inducible NOS is augmented in atherosclerotic arteries. New Zealand White (normal, n=18) and Watanabe heritable hyperlipidemic (atherosclerotic, n=21) rabbits were anesthetized and injected intravenously with vehicle or LPS. Contractile responsiveness of aortic segments was examined in vitro 4 hours after injection of LPS in vivo. There was a substantial (approximately fivefold) decrease in contractile sensitivity of aortas from LPS-treated atherosclerotic rabbits and a small (approximately twofold) decrease in normal rabbits. Incubation of aortic segments with aminoguanidine, which inhibits inducible NOS, restored contractile responsiveness after LPS treatment. In vitro assay of conversion of [14C]l-arginine to [14C]l-citrulline by aortic segments demonstrated marked (approximately fivefold) increase in calcium-independent conversion of [14C]l-arginine by LPS-treated atherosclerotic, but not normal, aortas. NADPH-diaphorase staining demonstrated positive cells only in the endothelium of normal rabbits and in the lesions and media of the atherosclerotic aortas in both vehicle- and LPS-treated rabbits. The general distribution of these NADPH-diaphorase–positive cells resembled that of smooth muscle cells and not macrophages. Thus, impairment of contractile responses, generation of l-citrulline, and staining with NADPH-diaphorase suggest that atherosclerotic arteries have increased capacity for generation of nitric oxide by inducible NOS.
In atherosclerosis, there is impaired endothelium-dependent relaxation by stimuli, such as acetylcholine, that activate constitutively expressed endothelial NOS.1 2 3 However, the capacity of atherosclerotic vessels to express inducible NOS and generate nitric oxide in response to an inflammatory stimulus, such as LPS, is not known.
Inducible NOS is a calcium-independent enzyme that generates relatively large amounts of nitric oxide in an unregulated fashion for several hours.4 Exposure of blood vessels to LPS may thus impair vascular contractility by overproduction of nitric oxide by inducible NOS.5 The intimal-medial lesions of atherosclerotic arteries are rich in monocytes/macrophages, and it is now apparent that these cells can produce mediators that modulate functions of other cells within the atherosclerotic vessel wall.6 Therefore, we postulated that monocytes/macrophages within the atherosclerotic lesion may confer a considerably greater capacity for vascular generation of nitric oxide in response to LPS. In support of this hypothesis, there is recent evidence for preexisting inducible NOS in nonendothelial cells of atherosclerotic aortas.7 There is also evidence that vascular smooth muscle cells in tissue culture generate greater amounts of nitric oxide in response to LPS after enrichment with cholesteryl esters.8
Thus, we used three approaches (contractile responses, generation of l-citrulline, and staining with NADPH-diaphorase) to test the hypothesis that after administration of LPS in vivo, the generation of nitric oxide by inducible NOS is augmented in atherosclerotic arteries. First, we compared the effects of LPS, administered in vivo, on contractile responses of aortas and carotid arteries from normal and atherosclerotic rabbits and assessed the in vitro effect of aminoguanidine, a reasonably selective inhibitor of inducible NOS, on responses of vessels from control and LPS-treated animals. Second, the activity of inducible NOS in isolated aortas was assessed by conversion of [14C]l-arginine to [14C]l-citrulline in the absence of calcium. Third, localization of NOS-containing cells was assessed by NADPH-diaphorase staining. Immunostaining for macrophages and vascular smooth muscle cells was performed on adjacent aortic sections to determine whether NOS-containing cells colocalized with macrophages or smooth muscle cells.
Materials and Methods
Model of Atherosclerosis
We studied homozygous WHHL rabbits of either sex from the University of Iowa colony. Although WHHL rabbits in our colony have a spontaneous elevation of plasma cholesterol to 400 to 500 mg/dL, we found only fatty streaks and relatively mild lesions in the aortas at 8 to 24 months of age, with no lesions in the carotid or other major branch arteries. To augment the severity of the atherosclerotic lesions, the diet was intermittently supplemented with 0.25% cholesterol, with 8 weeks of atherogenic diet followed by 8 weeks of normal diet. The atherogenic diet was prepared by dissolving 25 g cholesterol in 1 L corn oil under gentle heat and mixing with 1 kg normal rabbit chow (Teklad Highfiber, Harlan). At the time of study, the plasma cholesterol level was 677±99 mg/dL in WHHL rabbits and 69±12 mg/dL in normal NZW rabbits. The study was approved by the University of Iowa Animal Care and Use Review Committee.
The atherogenic diet in WHHL rabbits resulted in a generally thickened aortic intima, which consisted of a largely foam cell population. There was only a modest fibrous tissue reaction in the aorta. In contrast, the common carotid arteries remained virtually free of atherosclerotic changes. Moderate to severe lesions occurred around both the origin of the common carotid artery and the bifurcation to the internal and external carotid arteries, but we observed only an occasional early lesion within the common carotid arteries. Rings of carotid artery that were used for functional studies were not obtained from regions with lesions.
In Vivo Experimental Protocol
NZW (normal, 2.5 to 3.0 kg, n=18) and WHHL (atherosclerotic, 2.4 to 3.2 kg, n=21) rabbits of either sex were studied. Anesthesia was induced with sodium pentobarbital (40 to 60 mg/kg IV) and was maintained by supplementary administration at 15 to 30 mg/h IV. A tracheostomy was performed, and the rabbit was artificially ventilated with room air. Arterial blood pressure was recorded from a cannula in the right femoral artery, and intravenous injections were given via a cannula in the right femoral vein. Ventilation parameters were adjusted, and if necessary, oxygen was added to the inspired air to achieve arterial Po2 of 100 to 130 mm Hg, Pco2 of 32 to 36 mm Hg, and pH of 7.40 to 7.50. Rectal temperature was maintained at 37°C to 39°C with a heating pad.
After the surgery, a period of 15 minutes was allowed before 5 mL of saline (vehicle) or LPS (20 mg/kg) was injected intravenously for 1 minute. Four hours after injection of vehicle or LPS, the animal was killed by anesthetic overdose, and the thoracic aorta and left common carotid artery were removed. Vessels were placed in Krebs’ bicarbonate solution of the following composition (mmol/L): NaCl 118, KCl 4.7, KH2PO4 1.2, MgSO4 · 7H2O 1.2, d-glucose 11.1, NaHCO3 25.0, and CaCl2 · H2O 2.54.
In Vitro Protocol
Aortic and carotid arterial segments were cleaned of loose connective tissue; care was taken to avoid damage to the endothelial surface. We cut five rings from the thoracic aorta (5 to 6 mm long) and two rings from the left common carotid artery (3 to 4 mm long). Endothelium was removed from some aortic rings by gently rubbing the luminal surface with a sharpened wooden stick. The vascular rings were mounted on stainless steel hooks at resting tension (aorta, 8 g; carotid artery, 4 g) in organ baths, bathed in Krebs’ bicarbonate solution at 37°C, and gassed with 5% CO2/95% O2. Tension was periodically adjusted to the desired level during a 45-minute equilibration period. The vascular rings were then contracted twice with 80 mmol/L KCl and rinsed three times after each contraction.
When vascular tone had returned to the resting level, the rings were again contracted to a steady level with phenylephrine (0.1 to 2 μmol/L, ≈50% of maximum phenylephrine contraction), and acetylcholine (3 μmol/L, 100% of maximum acetylcholine relaxation) was added to the bath to assess functional integrity of the endothelium. Sodium nitroprusside (10 μmol/L, 100% of maximum sodium nitroprusside relaxation) was then added to assess the relaxation capacity of vascular smooth muscle. After they were rinsed three times, the vascular segments were incubated for 2 hours with or without aminoguanidine (0.3 mmol/L) and/or l-arginine (3 mmol/L). A concentration-response curve for phenylephrine was then generated, ≈7 to 8 hours after the administration of LPS in vivo. After each experiment, vascular rings were fixed in 10% formaldehyde for microscopic examination of lesions and/or confirmation of endothelial removal.
Acetylcholine chloride, aminoguanidine hemisulfate, LPS from Escherichia coli serotype 055:B5 (lyophilized powder prepared by phenol extraction), l-phenylephrine hydrochloride, and sodium nitroprusside were obtained from Sigma Chemical Co and were dissolved in normal saline.
Assay of NOS Activity by [14C]l-Citrulline Production
Segments (1 to 2 cm long) of rabbit aorta were cleaned of adherent tissue, rinsed in Krebs’ solution, and frozen under liquid nitrogen. Samples were stored at −70°C until assayed for NOS activity by measuring the conversion of [14C]l-arginine to [14C]l-citrulline. We used a modification of the method of Clavier et al.9 Tissue was homogenized in 50 mmol/L Tris-HCl (pH 7.4) and 2 mmol/L EDTA. The homogenate was centrifuged at 10 000g for 15 minutes at 4°C, and the supernatant was decanted and assayed in duplicate. The reaction was begun by adding 50 μL of supernatant to 100 μL of reaction mixture, consisting of 3 μmol/L [14C]l-arginine (NEN), 50 mmol/L Tris (pH 7.4), and 1.0 mmol/L NADPH. To determine whether citrulline production was due to NOS activity, parallel samples were processed in the presence of 1.0 mmol/L L-NAME, which caused complete (>99%) inhibition of l-arginine conversion.
The mixture was incubated at 20°C for 15 minutes, and the reaction was stopped with 1.8 mL of 30 mmol/L HEPES (pH 5.2). The mixture was eluted through a Dowex column (Na+ form; volume, 0.5 mL) and washed with 1 mL H2O. Radioactivity of the eluent was counted with a liquid scintillation counter. NOS activity was quantified by calculation of the amount of [14C]l-citrulline produced per minute per milligram protein. Protein concentration in the supernatant was measured by a modified Lowry assay (Bio-Rad).
Segments of aorta, carotid artery, ileum, and hind-limb skeletal muscle were placed in 2% paraformaldehyde for 2 hours at 4°C and then transferred to a 20% sucrose solution for 48 hours. Vessels were embedded in O.C.T. compound (Tissue-Tek, Miles) and stored frozen at −70°C. Sections were mounted on microscope slides and incubated at 37°C in 0.1 mmol/L Triton X-100 (Fischer Scientific), 1 mg/mL β-NADPH (Sigma), and 0.5 mg/mL nitro blue tetrazolium (Sigma) in 0.1 mol/L phosphate buffer (pH 7.4) for ≈4 hours in darkness. Slides were dehydrated in a series of ethanol solutions (70%, 95%, and 100%) and cleared in xylene, and a coverslip was applied.
For immunohistochemical analysis, we used RAM-11 (mouse IgG1, purchased from Dako Corp), a monoclonal antibody that recognizes a cytoplasmic antigen protein expressed by rabbit macrophages,10 and HHF-35 (mouse IgG1, purchased from Dako Corp), a monoclonal antibody that recognizes muscle-specific α-actin.10
Serial 6-μm-thick frozen sections of rabbit aorta were adhered to poly-l-lysine–coated slides, allowed to dry in room air, fixed in acetone at −20°C for 5 minutes, and incubated with RAM-11 (1:50) or HHF-35 (1:50) for 1 hour. Pilot experiments \E indicated that these concentrations of antibodies were appropriate for staining. After washing three times in phosphate-buffered saline containing 2% horse serum, a biotinylated secondary antibody (horse anti-mouse IgG) was applied, followed by avidin-biotin peroxidase complexes (ABC Elite kit, Vector Laboratories). Antibody binding was visualized with 3,3′-diaminobenzidine (Vector). Rabbit lung and spleen were used as positive controls for RAM-11. Omission of primary antibodies and staining with type- and class-matched nonimmune IgGs served as negative controls for each antibody in the present study.
All data are mean±SEM. From each experiment, the EC50 value (the concentration of phenylephrine that produced 50% of the maximum contraction) was calculated from the mean phenylephrine concentration-response data by using a computer program that assumed the curve to be linear between the two points around 50% of maximum response. Comparisons were made by using either a paired or unpaired Student’s t test. Statistical significance was accepted at P<.05.
Effect of LPS on Arterial Pressure
Arterial pressure did not change significantly after injection of vehicle in 4 normal (76±6 mm Hg before injection and 75±7 mm Hg 4 hours after injection) or 4 atherosclerotic (80±2 mm Hg before injection and 73±4 mm Hg 4 hours after injection) rabbits. Vessels were harvested from another 3 normal and 3 atherosclerotic rabbits that were killed immediately by anesthetic overdose. Responses of these vessels were similar to those from vehicle-treated rabbits and therefore were combined with data for vehicle-treated rabbits.
In preliminary experiments, we found that injection of 0.5 to 5 mg/kg IV LPS had no significant sustained effect on arterial pressure of normal rabbits over 4 hours. Thus, a higher dose (20 mg/kg IV) was chosen for the study. Injection of 20 mg/kg IV LPS produced a large transient decrease in arterial pressure that was maximal after 2 minutes and similar in magnitude in normal and atherosclerotic rabbits (from 73±2 to 34±2 mm Hg in normal rabbits and from 78±2 to 36±2 mm Hg in atherosclerotic rabbits). Arterial pressure returned to control levels 10 minutes after the LPS injection. During the next 4 hours, arterial pressure decreased gradually after LPS in both groups, and the decrease was greater in atherosclerotic rabbits (to 54±7 mm Hg in normal rabbits and to 35±3 mm Hg in atherosclerotic rabbits, P<.05). Two of 10 normal rabbits and 5 of 13 atherosclerotic rabbits developed severe hypotension and died <4 hours after administration of LPS. Thus, vessels from 8 LPS-treated rabbits were studied in each group.
Effect of LPS in Normal Rabbit Vessels
Aortic rings with endothelium, obtained from normal rabbits treated with LPS, were slightly (1.9-fold) less sensitive to contraction by phenylephrine than vessels from rabbits treated with vehicle (Fig 1A⇓). LPS did not significantly affect the sensitivity to phenylephrine of aortic rings denuded of endothelium (Fig 2A⇓) or of carotid artery rings with endothelium (log EC50 values: vehicle, −6.02±0.07; LPS, −5.93±0.12; n=5 to 7). Maximum contractions by phenylephrine were unaffected by LPS treatment in normal rabbits (Figs 1A⇓ and 2A⇓; carotid artery maximum values: vehicle, 8.6±0.0.6 g; LPS, 8.1±0.6 g; n=5 to 7). Likewise, in normal aortas, LPS had no effect on endothelium-dependent relaxation by 3 μmol/L acetylcholine (vehicle, 75±2%; LPS, 71±4%) or endothelium-independent relaxation by 10 μmol/L sodium nitroprusside (vehicle, 101±2%; LPS, 105±5%).
Effect of LPS in Atherosclerotic Rabbit Vessels
Aortic rings, with or without endothelium, from vehicle-treated atherosclerotic rabbits were slightly (2-fold) less sensitive to phenylephrine than normal aortic rings (Figs 1⇑ and 2⇑). Moreover, LPS treatment caused a substantial (4.9-fold) further decrease in aortic sensitivity to phenylephrine (Figs 1B⇑ and 2B⇑). The maximum contraction of aortas by phenylephrine also tended to be reduced by LPS (Figs 1B⇑ and 2B⇑). However, LPS treatment had no effect on responses of carotid arteries (log EC50 values: vehicle, −5.80±0.07; LPS, −5.68±0.10; maximum values: vehicle, 9.5±0.7 g; LPS, 9.0±0.5 g; n=4 to 7). In vehicle-treated WHHL rabbits, endothelium-dependent relaxation by 3 μmol/L acetylcholine was reduced in the aorta (30±2%, P<.05 vs vehicle-treated normal rabbits), and LPS did not significantly affect this response (24±8%). Endothelium-independent relaxation by 10 μmol/L sodium nitroprusside was similar to that in normal rabbit aortas (104±6%), and responses were slightly reduced by LPS (89±5%, P<.05 vs. vehicle-treated normal rabbits).
After LPS treatment in vivo, incubation with l-arginine (3 mmol/L) in vitro had no effect on phenylephrine-induced contractions of aortic rings from either normal (log EC50 values: without l-arginine, −6.58±0.06; with l-arginine, −6.58±0.09; n=5) or atherosclerotic (log EC50 values: without l-arginine, −5.92±0.08; with l-arginine, −6.06±0.32; n=3) rabbits treated with LPS, suggesting that substrate availability did not limit the production of nitric oxide by inducible NOS.
We attempted to determine whether the magnitude of LPS-induced hypotension accounted for greater impairment of vascular contractility in vitro in atherosclerotic rabbits than in normal rabbits. Subsets of normal and atherosclerotic animals treated with LPS were compared. In 3 normal rabbits in which LPS produced the greatest decrease in arterial pressure (−31±2 mm Hg), the log EC50 for aortic contraction by phenylephrine was −6.64±0.14. This value was similar to that for the other 4 LPS-treated normal rabbits (EC50, −6.65±0.07) in which arterial pressure decreased by only −10±4 mm Hg after LPS treatment. Likewise, in 3 atherosclerotic rabbits in which LPS produced the smallest decrease in arterial pressure (−34±2 mm Hg), the log EC50 value for contractions by phenylephrine was −5.94±0.10. By comparison, in the 4 other LPS-treated atherosclerotic rabbits, arterial pressure decreased by −51±4 mm Hg, and the sensitivity of the aorta to phenylephrine was similar (log EC50, −5.90±0.17). Thus, it does not seem that greater hypotension in atherosclerotic rabbits accounted for greater inhibition of contraction to phenylephrine in vitro.
Effect of Aminoguanidine
In aortic rings from vehicle-treated normal rabbits, aminoguanidine (0.3 mmol/L, 2 hours) had no effect on phenylephrine-induced contractions (respective maximum and log EC50 values: control rings, 14.8±1.2 g and −6.77±0.08; aminoguanidine-treated rings, 14.2±0.9 g and −6.68±0.12; n=5) or on acetylcholine-induced relaxations (respective maximum and log EC50 values: control rings, 76±4% and −7.21±0.06; aminoguanidine-treated rings, 74±5% and −7.16±0.05; n=4). Aminoguanidine produced a very small (1.5-fold), but significant, increase in sensitivity to phenylephrine in aortas from vehicle-treated atherosclerotic rabbits (Fig 3A⇓).
Aminoguanidine restored the aortic sensitivity to phenylephrine after LPS treatment in both normal (log EC50 values: aminoguanidine-treated aortas, −6.75±0.07; aortas without aminoguanidine treatment, −6.65±0.06; n=7; P<.05) and atherosclerotic rabbits (Fig 3B⇑). Coincubation of l-arginine with aminoguanidine prevented the increase in sensitivity to phenylephrine (Fig 3B⇑).
NOS Activity in Rabbit Aortas
Calcium-independent (inducible) NOS activity, determined by conversion of [14C]l-arginine to [14C]l-citrulline, was minimal in normal aortas after vehicle or LPS treatment (Fig 4⇓). In contrast, there was some calcium-independent NOS activity in aortas of vehicle-treated atherosclerotic rabbits (Fig 4⇓). Moreover, LPS treatment further increased calcium-independent NOS activity by 4.5-fold (12.9-fold higher than in LPS-treated normal rabbits, Fig 4⇓).
NADPH-Diaphorase Activity in Rabbit Aortas
We attempted to use immunohistochemical techniques to specifically stain aortic cells containing inducible NOS by using an antibody against murine inducible NOS but found no cross-reactivity with rabbit tissue. Therefore, we used the presence of NADPH-diaphorase staining as a less specific marker of cells containing any isoform of NOS.
NADPH-diaphorase staining was observed in endothelial cells but not in medial smooth muscle cells in both vehicle- and LPS-treated aortas of normal rabbits (Fig 5A⇓). By contrast, NADPH-diaphorase–containing cells were observed in the endothelium, the thickened intima-media, and in the medial smooth muscle layer of aortas from vehicle- and LPS-treated atherosclerotic rabbits (Fig 5B⇓). The relative intensity or frequency of staining in vehicle- and LPS-treated atherosclerotic aortas was not quantified. In samples from rabbits of all groups, there was little or no evidence of NADPH-diaphorase staining in carotid arteries (except in endothelium) or microvascular cells of ileum or skeletal muscle (data not shown).
Immunohistochemical Staining of Macrophages and Smooth Muscle Cells in Atherosclerotic Aortas
We attempted to determine whether NADPH-diaphorase–positive cells in the vessels were macrophages and/or smooth muscle cells by use of immunohistochemical staining. These studies revealed the presence of both macrophages (Fig 6A⇓) and smooth muscle cells (Fig 6B⇓) in the intimal lesions of vehicle- or LPS-treated atherosclerotic rabbit aortas. Smooth muscle cells (which stained with HHF-35) were widespread throughout the thickened intima and generally resembled the pattern of distribution of NADPH-diaphorase staining (Fig 6B⇓ and 6C⇓). In contrast, macrophages (stained with RAM-11) were most concentrated immediately beneath the endothelium (Fig 6A⇓). Double labeling of sections with antibody (RAM-11 or HHF-35) and NADPH-diaphorase was unsuccessful, because the NADPH-diaphorase reaction was weak in antibody-treated sections. In adjacent thin sections of aortas, however, we noted good correlation of regions that were stained by HHF-35 and NADPH-diaphorase but not RAM-11 (Fig 6A⇓ through 6E). It was apparent, though, that many smooth muscle cells in the intimal-medial lesion did not stain positively for NADPH-diaphorase. Moreover, we cannot exclude the possibility that some macrophages may have been stained with NADPH-diaphorase, although the regional localization within the vessel did not correspond.
The major new finding of the present study is that after administration of LPS in vivo, there is greater expression of inducible NOS in atherosclerotic aortas than in normal aortas. Evidence was obtained from functional, biochemical, and histological studies. In functional studies, we found that impairment of aortic contractility by LPS is minimal in normal rabbits and substantial in atherosclerotic rabbits. In contrast to the aorta, carotid arteries from atherosclerotic WHHL rabbits that do not contain intimal lesions are functionally unaffected by LPS treatment. There is marked elevation of calcium-independent NOS activity in LPS-treated atherosclerotic aortas. NADPH-diaphorase staining demonstrated positive cells in the lesions and media of the atherosclerotic aortas. The general distribution of these NADPH-diaphorase–positive cells resembled that of smooth muscle cells and not macrophages. Thus, our findings imply that cells (possibly smooth muscle cells) within atherosclerotic lesions contribute to enhanced generation of nitric oxide by inducible NOS in response to LPS.
Effects of LPS and Aminoguanidine on Vascular Reactivity
Injection of LPS in normal rabbits caused little or no impairment in the aortic response to phenylephrine, whereas in atherosclerotic rabbits the effect of LPS was much greater. The impaired contractility was observed in the presence or absence of endothelium. Decreased vasoconstrictor responses to α-adrenoceptor agonists occur after exposure to LPS as a result of the inhibitory influence of nitric oxide when inducible NOS is synthesized.11 12
In these experiments, impairment of phenylephrine-induced contractions by LPS was prevented by treatment of the vessels in vitro with aminoguanidine, an agent that is reported to be a reasonably selective inhibitor of inducible NOS.13 14 15 The effect of aminoguanidine was prevented by l-arginine, the substrate for nitric oxide synthesis. Aminoguanidine had no effect on acetylcholine-induced relaxation of rabbit aortas, which is mediated by endothelial NOS. Thus, our data suggest that at the concentration used, aminoguanidine is a selective inhibitor of inducible but not endothelial NOS in rabbit aortas and support the conclusion that the hyporesponsiveness of LPS-treated atherosclerotic aortas is due to increased activity of inducible NOS.
Inducible NOS Activity in Rabbit Aortas
We measured the activity of inducible (calcium-independent) NOS in rabbit aortas by using an in vitro assay of [14C]l-arginine conversion to [14C]l-citrulline. There was negligible activity of inducible NOS in aortas from vehicle-treated normal rabbits and small but measurable levels in atherosclerotic aortas. Attempts to specifically stain cells containing inducible NOS by using an antibody against murine inducible NOS were unsuccessful, presumably because of failure of cross-reactivity with rabbit tissue. However, we observed NADPH-diaphorase–containing cells in the intima-media of atherosclerotic but not normal aortas (except in endothelium). These histological findings are consistent with the finding of a slightly decreased sensitivity of vehicle-treated atherosclerotic aortas to contraction by phenylephrine and improvement of contractile sensitivity with aminoguanidine. Together, these results support previous evidence for “preexisting” inducible NOS in nonendothelial cells of atherosclerotic rabbit aortas.7
We found that LPS in vivo stimulates substantial increases in levels of inducible NOS activity in atherosclerotic aortas but comparatively low levels in normal aortas. These findings support the conclusion that excessive production of nitric oxide by inducible NOS in atherosclerotic but not normal aortas accounts for the significantly greater impairment of contractility by LPS. Basal production of nitric oxide may be elevated in atherosclerotic rabbit aortas,16 and the present data suggest that activity of inducible NOS may contribute to the enhanced basal generation of nitric oxide.
Mechanism of Enhanced Production of Inducible NOS in Atherosclerosis
Endothelial cells, vascular smooth muscle cells, and monocytes/macrophages can all potentially synthesize inducible NOS.17 Our studies of vascular rings after denudation indicate that the predominant source of inducible NOS was not endothelium, because equivalent decreases in sensitivity to phenylephrine were observed in atherosclerotic aortic rings with or without endothelium. On the basis of histological examination of the aortic rings, it appeared that there was minimal removal of the intimal lesion during endothelial denudation of atherosclerotic vessels. Our conclusion, ie, that LPS stimulated the synthesis of inducible NOS in nonendothelial cells of atherosclerotic aortas, would not be altered even if a substantial amount of lesion were removed by denudation.
There was no impairment of responses to phenylephrine in carotid arterial rings from LPS-treated atherosclerotic rabbits. Because intimal lesions were absent in the carotid arteries of the WHHL rabbits, we suggest that an atherosclerotic lesion per se may be necessary for the increased vascular activity of inducible NOS in response to LPS under the conditions of the present study.
Rabbit macrophages may not generate significant amounts of nitric oxide in response to LPS in vitro18 ; thus, a different cell type within the atherosclerotic lesion may be responsible for generation of nitric oxide. LPS-treated macrophages may, however, release cytokines, which stimulate production of inducible NOS in vascular smooth muscle cells.18 These findings18 are consistent with our present data, and we speculate that substantial amounts of cytokines and inducible NOS are synthesized within the atherosclerotic lesion in response to LPS, resulting in local generation of a large amount of nitric oxide, which profoundly inhibits vascular contractility. The presence of NADPH-diaphorase–containing cells in the atherosclerotic lesion is consistent with the possibility that cells in the intima are a source of inducible NOS. We found that the localization of NADPH-diaphorase– containing cells within the atherosclerotic lesion generally correlated with the localization of intimal vascular smooth muscle cells rather than macrophages. Thus, intimal vascular smooth muscle cells may be a significant source of inducible NOS expression in atherosclerotic arteries.
Previous authors have reported that LPS-induced hypotension is mediated at least in part by nitric oxide.19 20 The immediate transient hypotension evoked by LPS probably reflects release of endothelial nitric oxide and other vasoactive factors from systemic resistance vessels.21 Increased serum levels of nitric oxide metabolites have been reported in humans with septic shock,22 23 and NOS inhibitors restore arterial pressure toward normal.24 In the present study, the sustained LPS-induced hypotension was more severe in atherosclerotic than in normal rabbits. We did not study mechanisms that account for hypotension after LPS, but greater hypotension in atherosclerotic than normal rabbits may be due to several factors, including decreased baroreflex regulation of arterial pressure in WHHL rabbits.25 26 Because atherosclerotic lesions in these rabbits were confined largely to the aorta, it would be surprising if nitric oxide generated solely in the aorta could produce profound sustained hypotension after injection of LPS. We did not detect greater NADPH-diaphorase staining of microvascular cells in the ileum or skeletal muscle from atherosclerotic rabbits than from normal LPS-treated rabbits. It is also possible that enhanced induction of NOS in circulating leukocytes or higher levels of plasma or intravascular27 cytokines in LPS-treated atherosclerotic rabbits may contribute to exaggerated hypotension in atherosclerotic rabbits.
In the present study, we observed that LPS increases the activity of inducible NOS, especially in atherosclerotic vessels. It is of interest to note that others have reported decreased activity of endothelial NOS in response to LPS28 or cytokines.29
We suggest that the augmented vascular effect of LPS observed in WHHL compared with NZW rabbits is due to the presence of atheromatous aortic lesions and not to unrelated genetic differences. Although this latter possibility cannot be excluded completely, the lack of effect of LPS in the lesion-free carotid segments of WHHL rabbits suggests that an atherosclerotic lesion is required for the exaggerated vascular response to LPS.
In summary, the data suggest that after LPS administration, vascular expression of inducible NOS is augmented in atherosclerotic arteries. We speculate that the production of nitric oxide by inducible NOS, combined with superoxide anion generation, may lead to formation of peroxynitrite and may thus contribute to vascular injury in atherosclerosis.30 31 Cells within the atherosclerotic lesion, possibly vascular smooth muscle cells, may be a major source of vascular nitric oxide synthesis in inflammation.
Selected Abbreviations and Acronyms
|L-NAME||=||N-nitro-l-arginine methyl ester|
|NOS||=||nitric oxide synthase|
|NZW||=||New Zealand White|
|WHHL||=||Watanabe heritable hyperlipidemic|
These studies were supported by National Institutes of Health grants HL-14388, HL-16066, AG-10269, and NS-24621 and research funds from the Veterans Administration. Dr Sobey is supported by a C.J. Martin Fellowship from the National Health and Medical Research Council of Australia and is the recipient of a Michael J. Brody Fellowship in Basic Cardiovascular Research from the University of Iowa. We are grateful for the assistance of Dr Mark Armstrong in evaluation of vascular morphology, Dr Eddie Brian and Kristen Orgren in the assay of NOS activity, Dr Alan Law and Dr Steve Meller in the staining of NADPH-diaphorase, Pam Tompkins in the preparation of histological specimens and photographs, and Dr Frank Faraci for critical review of the manuscript.
- Received October 12, 1994.
- Accepted May 15, 1995.
- © 1995 American Heart Association, Inc.
Verbeuren TJ, Jordaens FH, Zonnekeyn LL, Van Hove CE, Coene MC, Herman AG. Effect of hypercholesterolemia on vascular reactivity in the rabbit. Circ Res. 1986;58:552-564.
Freiman PC, Mitchell GG, Heistad DD, Armstrong ML, Harrison DG. Atherosclerosis impairs endothelium-dependent vascular relaxation to acetylcholine and thrombin in primates. Circ Res. 1986;58:763-789.
Brady AJB, Poole-Wilson PA. Circulatory failure in septic shock: nitric oxide: too much of a good thing? Br Heart J. 1993;70:103-105.
Clavier N, Tobin JR, Kirsch JR, Izuta M, Traystman RJ. Brain nitric oxide synthase activity in normal, hypertensive, and stroke-prone rats. Stroke. 1994;25:1674-1678.
Tsukada T, Rosenfeld M, Ross R, Gown AM. Immunocytochemical analysis of cellular components in lesions of atherosclerosis in the Watanabe and fat-fed rabbit using monoclonal antibodies. Arteriosclerosis. 1986;6:601-613.
Corbett JA, Tilton RG, Chang K, Hasan KS, Ido Y, Wang JL, Sweetland MA, Lancaster JR, Williamson JR, McDaniel ML. Aminoguanidine, a novel inhibitor of nitric oxide formation, prevents diabetic vascular dysfunction. Diabetes. 1992;41:552-556.
Tilton RG, Chang K, Hasan KS, Smith SR, Petrash JM, Misko TP, Moore WM, Currie MG, Corbett JA, McDaniel ML, Williamson JR. Prevention of diabetic vascular dysfunction by guanidines: inhibition of nitric oxide synthase versus advanced glycation end-product formation. Diabetes. 1993;42:221-232.
Minor RL, Myers PR, Guerra R, Bates JN, Harrison DG. Diet-induced atherosclerosis increases the release of nitrogen oxides from rabbit aorta. J Clin Invest. 1990;86:2109-2116.
Bernard C, Szekely B, Philip I, Wollman E, Payen D, Tedgui A. Activated macrophages depress the contractility of rabbit carotids via an l-arginine/nitric oxide-dependent effector mechanism: connection with amplified cytokine release. J Clin Invest. 1992;89:851-860.
Wright CE, Rees DD, Moncada S. Protective and pathological roles of nitric oxide in endotoxin shock. Cardiovasc Res. 1992;26:48-57.
Salvemini D, Korbut R, Anggard E, Vane J. Immediate release of a nitric oxide-like factor from bovine aortic endothelial cells by Escherichia coli lipopolysaccharide. Proc Natl Acad Sci U S A. 1990;87:2593-2597.
Fleet JC, Clinton SK, Salomon RN, Loppnow H, Libby P. Atherogenic diets enhance endotoxin-stimulated interleukin-1 and tumour necrosis factor gene expression in rabbit aortae. J Nutr. 1992;122:294-305.
Graier WF, Myers PR, Rubin LJ, Adams R, Parker JL. Escherichia coli endotoxin inhibits agonist-mediated cytosolic Ca2+ mobilization and nitric oxide biosynthesis in cultured endothelial cells. Circ Res. 1994;75:659-668.
Yoshizumi M, Perrella MA, Burnett JC Jr, Lee M-E. Tumor necrosis factor downregulates an endothelial nitric oxide synthase mRNA by shortening its half-life. Circ Res. 1993;73:205-209.
Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci U S A. 1990;87:1620-1624.
White CR, Brock TA, Chang L-Y, Crapo J, Briscoe P, Ku D, Bradley WA, Gianturco SH, Gore J, Freeman BA. Superoxide and peroxynitrite in atherosclerosis. Proc Natl Acad Sci U S A. 1994;91:1044-1048.