Vascular Biology |
From the Departments of Internal Medicine and Pharmacology, University of Iowa College of Medicine and VA Medical Center, Iowa City, Iowa.
Correspondence to Carol A. Gunnett. PhD, Department of Internal Medicine, University of Iowa College of Medicine, Iowa City, IA 52242-1081. E-mail Carol-Gunnett{at}uiowa.edu
| Abstract |
|---|
|
|
|---|
Key Words: superoxide NO-dependent relaxation adenovirus acetylcholine nitroprusside
| Introduction |
|---|
|
|
|---|
See page 1259
Endothelium-dependent vasorelaxation mediated by endothelial NO synthase (eNOS) is impaired in response to several inflammatory stimuli that induce iNOS in blood vessels.12,1719 Pharmacological inhibitors of iNOS improve endothelium-dependent relaxation after LPS.12,20,21 The findings suggest that expression of iNOS is associated with decreased function of eNOS. However, pharmacological inhibitors of iNOS are not entirely selective for iNOS, and inhibition of neuronal NO synthase also improves endothelium-dependent relaxation during inflammation.20 Therefore, it is not known whether iNOS, per se, affects relaxation of blood vessels.
The goal of the present study was to use gene transfer of iNOS, in vitro and in vivo, to directly examine effects of iNOS on vasomotor function. We tested the hypothesis that expression of iNOS would be a sufficient stimulus to produce impaired contractile responses and impaired relaxation of carotid arteries. Several previous studies using iNOS gene transfer to blood vessels have evaluated antiproliferative effects,2227 and a recent study examined the effects of gene transfer of iNOS on contractile responses of tail arteries.28 Although several studies have examined the effects of gene transfer of eNOS on vascular function,2939 to our knowledge, these are the first studies to evaluate the effects of gene transfer of iNOS on NO-dependent relaxation.
| Methods |
|---|
|
|
|---|
To test whether the shuttle plasmid containing iNOS expressed functional iNOS, the plasmid was transfected to COS-1 cells, which have very low endogenous NO synthase activity. One or 2 days after transfection, NO synthase activity was assayed by measuring nitrite concentrations in the incubation media by using the Griess assay. The positive plasmid was then used to cotransfect 293 cells with a large adenovirus fragment for homologous recombination. Recombinant iNOS adenovirus was screened by polymerase chain reaction by using specific primers for mouse iNOS cDNA and by the Griess assay after infection of the 293 cells. AdiNOS was then propagated in the presence of 1 mmol/L aminoguanidine, and function was confirmed by demonstration of a large increase in nitrite concentrations in transduced COS-1 and 293 cells in the absence of aminoguanidine.
Animals
Adult male New Zealand White rabbits (2.5 to 3.0 kg) were used for gene transfer studies in vitro and in vivo. Rabbits (n=64) were euthanized with pentobarbital (150 mg/kg), and both common carotid arteries were removed.
Gene Transfer In Vitro
Carotid arteries were cut into 8 (
2- to 3-mm) segments and placed in DMEM. Viruses were diluted to final titers (reported as plaque-forming units [pfu] per milliliter) in 0.4 mL of tissue culture medium. Vessels were then incubated at 37°C for 2.5 hours in tissue culture medium plus AdCMViNOS or AdCMVBglII (a null virus with no functional transgene) or tissue culture medium alone (control). After 2.5 hours in virus or vehicle, vessels were transferred to 1.0 mL fresh DMEM with 1% penicillin/streptomycin and 2% FCS for an additional 24-hour incubation period at 37°C in an O2 incubator. After 24 hours, pairs of vessels were used for functional studies. We have previously demonstrated normal vasomotor function in carotid arteries after incubation for 24 hours.30,40
Gene Transfer In Vivo
Rabbits (n=18) were anesthetized with xylazine/ketamine (5/20 mg/kg IM), and both carotid arteries were exposed. Proximal and distal sutures were used to stop blood flow, an incision was made in the external carotid artery, and a catheter was introduced. Adenovirus (AdiNOS, 150 µL, 1x1010 pfu/mL) was instilled, and after 20 minutes, viral solution was withdrawn, the artery was sutured, and blood flow was reestablished. The procedure was repeated on the contralateral carotid artery by using control virus (AdBglII) or vehicle. Incisions were sutured with 4.0 silk. Rabbits were allowed to recover, and 2 days later, they were euthanized with sodium pentobarbital (150 mg/kg), and the carotid arteries were removed.
Immunohistochemistry
Immunohistochemistry was performed in segments of carotid artery after gene transfer of iNOS. Vessel segments were frozen and cut into serial sections (5 µmol/L thick). Frozen sections of vessels were placed on poly-L-lysinecoated slides and were allowed to dry at room temperature. Sections were fixed in acetone and 1% paraformaldehyde at 4°C for 5 minutes. Horse serum (5%) and 0.2% albumin were used for blocking nonspecific binding proteins for 20 minutes. Mouse IgG monoclonal antibody to murine iNOS (1:50, Transduction Laboratories) was then incubated for 30 minutes. After the slides were washed for 5 minutes in PBS, biotinylated goat anti-human antibody (Vector Laboratories) was applied for 30 minutes, followed by avidin-conjugated alkaline phosphatase (kit No. SK5300, Vector Laboratories) for 30 minutes. Slides were incubated in 0.05% diaminobenzidine tetrahydrochloride and 0.01% hydrogen peroxide for 5 minutes and then washed in water. Vessel sections were counterstained with nuclear fast red (Vector Laboratories) and examined for positive staining of iNOS by light microscopy.
Vasomotor Function
Arteries were placed in cold oxygenated Krebs solution, and isometric tension was recorded to assess vasomotor function, as described previously in detail.33,40
We examined the contraction of carotid rings in response to KCl, phenylephrine, and U46619 and relaxation in response to acetylcholine, ADP, A23187, and sodium nitroprusside. Vessels were preconstricted with phenylephrine to
60% of maximum constriction for subsequent dose-response curves to vasodilators. Acetylcholine41 and ADP42,43 are receptor-mediated and endothelium-dependent vasodilators. A23187 is a calcium ionophore that increases intracellular calcium and activates eNOS by a nonreceptor-mediated mechanism. Nitroprusside is a nonendothelium-dependent vasodilator that produces relaxation by direct activation of soluble guanylate cyclase in smooth muscle.
Some vessels were preincubated for 1 hour with aminoguanidine (300 µmol/L) or L-N-iminoethyl lysine (L-NIL, 100 µmol/L) before vasoconstrictor or vasodilator agents. At these concentrations, the agents are relatively selective inhibitors of iNOS,5,9,4447 and they do not impair responses to acetylcholine in our control vessels.
To determine whether cyclooxygenase (COX) enzymes or xanthine oxidase contribute to impaired function of carotid arteries after gene transfer of iNOS, some vessels were treated with indomethacin (10 µmol/L) or allopurinol (1 mmol/L) for 30 to 60 minutes before administration of vasoactive drugs. This concentration of indomethacin inhibits COX-mediated endothelium-derived contracting factors48 and vascular effects of arachidonic acid.49
Other experiments were performed in the presence of cell-permeable polyethylene glycol superoxide dismutase (PEG-SOD) or 4,5-dihydroxy-1,3-benzene disulfonic acid (Tiron), a scavenger of superoxide, to determine whether superoxide contributes to impaired vasomotor function after AdiNOS.
Chemicals
Acetylcholine, L-phenylephrine, sodium nitroprusside, A23187, ADP, aminoguanidine, L-NIL, Tiron, PEG-SOD, and indomethacin were obtained from Sigma Chemical Co and dissolved in normal saline. Allopurinol, also obtained from Sigma, was dissolved in 1 mol/L NaOH and titrated to physiological pH by using HCl. U46619 was obtained from Cayman Chemical, dissolved in ethanol, and then diluted in normal saline.
Statistical Analysis
Contractile responses are expressed in grams, and relaxation is expressed as percentage of precontraction produced by phenylephrine. Group comparisons were performed in the context of a factorial repeated-measures ANOVA, involving iNOS treatment versus BglII treatment, presence versus absence of pharmacological inhibitors, and the interaction of these 2 factors. Multiple comparisons were tested at a Bonferroni-adjusted
level. All data are expressed as mean±SEM.
| Results |
|---|
|
|
|---|
|
Contraction After Gene Transfer In Vitro
Contractile responses to phenylephrine (Figure 2, left) were similar after AdBglII or incubation in culture medium with no virus (vehicle). Contraction was similar in freshly harvested vessels and in vessels incubated in AdBglII or vehicle (data not shown). Contractile responses to phenylephrine (Figure 2, middle) and U46619 (Figure 2, right) were impaired in vessels after gene transfer of AdiNOS in vitro. Contraction to KCl also was impaired after AdiNOS. Contraction to KCl (60 mmol/L) was 2.8±0.1 g after AdiNOS and 3.5±0.2 g after AdBglII (n=11 and 6, respectively; P<0.05).
|
After gene transfer of iNOS, contraction to phenylephrine was improved by pretreatment with aminoguanidine (300 µmol/L; Figure 2, middle) or L-NIL (100 µmol/L, data not shown), which are relatively selective inhibitors of iNOS. Neither inhibitor of iNOS altered contraction of the vessels after AdBglII (data not shown). Contractile responses of carotid arteries were not altered by indomethacin (10 µmol/L, n=3) or allopurinol (1 mmol/L, n=4; data not shown).
Relaxation After Gene Transfer In Vitro
NO-dependent vasorelaxation was impaired after gene transfer of iNOS in vitro. Responses to acetylcholine were similar after AdBglII or vehicle (Figure 3, left). Responses to acetylcholine were markedly impaired in a titer-dependent manner after gene transfer using AdiNOS (Figure 3, middle). After AdiNOS, the responses of vessels to acetylcholine were restored largely to normal by aminoguanidine (Figure 3, right) or L-NIL (Figure 4, right). Neither inhibitor altered responses to acetylcholine in vessels after AdBglII (Figure 4) or vehicle (data not shown). Indomethacin had no effect on responses to acetylcholine after AdiNOS (n=3, data not shown).
|
|
PEG-SOD markedly improved responses to acetylcholine in 5 of 14 experiments and had little or no effect in 9 of 14 experiments after AdiNOS. Tiron, a second agent used to decrease superoxide levels, improved responses to acetylcholine in 5 of 18 experiments and had no effect in 13 of 18 experiments. Thus, effects of PEG-SOD and Tiron were not consistent, and group differences were not statistically significant (data not shown).
Responses to ADP (Figure 5, left) and A23187 (Figure 5, right) also were impaired after AdiNOS. Aminoguanidine improved the responses to ADP and to A23187 (Figure 5).
|
We also examined responses to nonendothelium-dependent vasodilators. Relaxation in response to nitroprusside, which produces NO, was similar in vessels after gene transfer of BglII or vehicle (data not shown). In contrast, responses to nitroprusside were impaired after gene transfer of iNOS and were improved by aminoguanidine (Figure 6, left). However, 8-bromo-cGMP and papaverine produced similar relaxation after AdiNOS or AdBglII (Figure 6, middle and right).
|
Contraction and Relaxation After Gene Transfer In Vivo
After intraluminal gene transfer of iNOS in vivo, contractile responses to phenylephrine in vitro were not significantly impaired. Maximum contraction to phenylephrine (10 µmol/L) was 6.6±0.7 g after AdiNOS and 7.2±0.3 after AdBglII (n=6, P>0.05).
After gene transfer in vivo by use of intraluminal AdiNOS, responses to acetylcholine (Figure 7, left) and nitroprusside (Figure 7, middle) were impaired compared with responses of vessels after AdBglII. Aminoguanidine improved relaxation to acetylcholine (Figure 7, left) and to nitroprusside (Figure 7, middle). Responses to papaverine were similar in vessels treated with AdiNOS or AdBglII (Figure 7, right).
|
| Discussion |
|---|
|
|
|---|
Expression of iNOS After iNOS Gene Transfer
We found extensive expression of iNOS in adventitia and endothelium after gene transfer in vitro and predominantly endothelial expression after intraluminal gene transfer in vivo. Minimal expression of iNOS was observed occasionally in adventitia of vessels after gene transfer of BglII in vitro, which suggests that some induction of endogenous iNOS may occur after infection with adenovirus. In contrast to BglII, all vessels exposed to AdiNOS in vitro showed expression of iNOS in adventitia and endothelium. Thus, gene transfer of iNOS produces expression of iNOS in blood vessels.
Effects of iNOS on Vasorelaxation
Previous studies have suggested that conditions in which expression of iNOS is observed (eg, in atherosclerosis or after LPS) are also characterized by impaired function of eNOS, but direct effects of iNOS on NO-mediated relaxation have not been described. Proinflammatory stimuli, such as tumor necrosis factor-
, which induces expression of iNOS, decrease the levels of eNOS protein and the amount of NO in cultured endothelial cells.50 LPS produces impaired eNOS-dependent relaxation in vessels,14,15,18 and expression of iNOS has been demonstrated in blood vessels after LPS.3,4,9 Responses to pharmacological inhibitors, such as aminoguanidine, suggest that iNOS mediates impaired endothelial function after LPS.12,20,21 However, inhibition of neuronal NO synthase also improves endothelium-dependent relaxation during inflammation.20 Because inhibitors of iNOS frequently have effects on >1 NO-synthase isoform, it is not known whether iNOS, per se, affects the relaxation of blood vessels. Thus, a causal relationship has not been demonstrated previously between the expression of iNOS and impaired NO-dependent relaxation.
We conclude that iNOS impairs NO-dependent relaxation because responses to acetylcholine and nitroprusside were impaired after gene transfer of iNOS. Acetylcholine is eNOS dependent in the carotid artery,41,51 and nitroprusside is an exogenous donor of NO. The finding that impairment of NO-dependent relaxation was observed after gene transfer of iNOS in vitro and in vivo reduces the possibility that impaired responses are an artifact of experiments performed in vitro.
Impaired responses to acetylcholine and nitroprusside after AdiNOS were improved by aminoguanidine and L-NIL, 2 different selective inhibitors of iNOS. Aminoguanidine and L-NIL produced no impairment of relaxation in control vessels, which suggests that the blockers do not inhibit eNOS at the concentrations used in these experiments. These results strongly suggest that iNOS mediates the observed impairment of relaxation.
We also examined responses to ADP, another receptor-mediated activator of eNOS,42 after the gene transfer of iNOS in vitro. Because 2 different receptor-mediated pathways were impaired after AdiNOS, we conclude that the effects are not limited to muscarinic receptors. Additional experiments were performed with nonreceptor-mediated activation of eNOS by the calcium ionophore A23187. Responses to A23187 also were impaired by iNOS, which suggests that impairment occurs downstream from endothelial receptors.
Responses to papaverine, a nonNO-mediated vasodilator, were similar after AdiNOS or AdBglII, suggesting that smooth muscle is capable of relaxing normally in the presence of iNOS. NO produces vasorelaxation by activating soluble guanylate cyclase in smooth muscle. To examine smooth muscle function downstream from guanylate cyclase, we tested the effects of 8-bromo-cGMP and found responses to be normal after AdiNOS. Thus, impairment of the relaxation produced by iNOS appears to be limited to NO-dependent mechanisms and occurs at the level of NO itself or at the level of soluble guanylate cyclase.
One mechanism that can reduce the effects of NO involves superoxide. NO binds readily with superoxide to form peroxynitrite.52 In the presence of elevated levels of superoxide, quantities of NO from eNOS or from nitroprusside may be insufficient to produce normal vasorelaxation. We tested whether superoxide mediates impaired responses to NO after AdiNOS by using PEG-SOD and Tiron. Responses to acetylcholine and nitroprusside were not consistently improved after AdiNOS in the presence of PEG-SOD or Tiron in a large number of studies. However, because improvement of NO-mediated relaxation by PEG-SOD or Tiron was observed in some studies, we cannot exclude the possibility that superoxide contributes to impaired responses after AdiNOS under some conditions.
Effects of iNOS on Contraction
Using pharmacological inhibitors and gene-targeted mice, we and others3,4 have provided evidence that implicates iNOS in impaired vasoconstrictor responses after acute inflammatory stimuli. Data from iNOS-deficient mice indicate that impairment of contraction is dependent on the expression of iNOS after LPS.3 However, previous studies do not define the role of iNOS, per se, because proinflammatory stimuli, such as LPS, produce the expression of many cytokines and inducible enzymes, such as COX-2, in addition to iNOS in blood vessels. The current data, with gene transfer of iNOS in vitro, are similar to responses observed by Worthington et al28 and provide evidence that the expression of iNOS in blood vessels is a sufficient stimulus to impair contraction. The finding that indomethacin did not alter responses suggests that COX-2 does not contribute to impaired contraction after gene transfer of iNOS.
We did not measure cytokines or other indices of inflammation in vessels after AdiNOS, but similar vasomotor function in vessels after AdBglII and in freshly harvested vessels provides evidence that exposure to adenovirus alone does not impair vasomotor function. Although data in the present study do not preclude the possibility of interaction of iNOS with other enzyme systems within the vessel wall, once iNOS is expressed, the data suggest that iNOS can alter vascular function independently of concurrent induction of other systemic inflammatory mediators.
Previous studies have documented expression of iNOS in adventitia after proinflammatory stimuli.6,53 The studies used pharmacological approaches to suggest a role for iNOS in adventitia in impairment of vasoconstrictor responses.53 In the present study, impaired contraction occurred only after the gene transfer of iNOS in vitro, in which extensive expression of iNOS was observed in the adventitia. Expression of iNOS was limited to endothelium after intraluminal gene transfer of iNOS in vivo. The findings in vitro are consistent with a role for adventitial iNOS in the impairment of contraction.
Summary
These findings provide direct evidence that iNOS can impair contraction and NO-dependent vasorelaxation. Receptor-mediated and nonreceptor-mediated activation of eNOS as well as responses to exogenous NO are all impaired after gene transfer of iNOS, but smooth muscle retains the ability to relax to some stimuli. Thus, impaired relaxation produced by iNOS appears to be specific for the NO-mediated pathway(s). The finding that iNOS contributes to vascular dysfunction during inflammation may have implications for atherosclerosis and several other cardiovascular diseases.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received March 19, 2001; accepted May 21, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. H. Lee, S. Xia, and L. Ragolia Upregulation of AT2 receptor and iNOS impairs angiotensin II-induced contraction without endothelium influence in young normotensive diabetic rats Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2008; 295(1): R144 - R154. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ohashi, F. Faraci, and D. Heistad Peroxynitrite hyperpolarizes smooth muscle and relaxes internal carotid artery in rabbit via ATP-sensitive K+ channels Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H2244 - H2250. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. R. Nagareddy, Z. Xia, J. H. McNeill, and K. M. MacLeod Increased expression of iNOS is associated with endothelial dysfunction and impaired pressor responsiveness in streptozotocin-induced diabetes Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H2144 - H2152. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Gunnett, D. D. Lund, F. M. Faraci, and D. D. Heistad Vascular interleukin-10 protects against LPS-induced vasomotor dysfunction Am J Physiol Heart Circ Physiol, August 1, 2005; 289(2): H624 - H630. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.A. Gunnett, D.D. Lund, A.K. McDowell, F.M. Faraci, and D.D. Heistad Mechanisms of Inducible Nitric Oxide Synthase-Mediated Vascular Dysfunction Arterioscler. Thromb. Vasc. Biol., August 1, 2005; 25(8): 1617 - 1622. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. T. Noronha, J.-M. Li, S. B. Wheatcroft, A. M. Shah, and M. T. Kearney Inducible Nitric Oxide Synthase Has Divergent Effects on Vascular and Metabolic Function in Obesity Diabetes, April 1, 2005; 54(4): 1082 - 1089. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Kimura, Y. Hirooka, Y. Sagara, K. Ito, T. Kishi, H. Shimokawa, A. Takeshita, and K. Sunagawa Overexpression of Inducible Nitric Oxide Synthase in Rostral Ventrolateral Medulla Causes Hypertension and Sympathoexcitation via an Increase in Oxidative Stress Circ. Res., February 4, 2005; 96(2): 252 - 260. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. G. Chicoine, E. Tzeng, R. Bryan, S. Saenz, M. L. Paffett, J. Jones, C. R. Lyons, T. C. Resta, L. D. Nelin, and B. R. Walker Intratracheal adenoviral-mediated delivery of iNOS decreases pulmonary vasoconstrictor responses in rats J Appl Physiol, November 1, 2004; 97(5): 1814 - 1822. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Egi, N. E. Conrad, J. Kwan, C. Schulze, R. Schulz, and S. M. Wildhirt Inhibition of inducible nitric oxide synthase and superoxide production reduces matrix metalloproteinase-9 activity and restores coronary vasomotor function in rat cardiac allografts Eur. J. Cardiothorac. Surg., August 1, 2004; 26(2): 262 - 269. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Hayashi, D. Sumi, P. A.R Juliet, H. Matsui-Hirai, Y. Asai-Tanaka, H. Kano, A. Fukatsu, T. Tsunekawa, A. Miyazaki, A. Iguchi, et al. Gene transfer of endothelial NO synthase, but not eNOS, plus inducible NOS regressed atherosclerosis in rabbits Cardiovasc Res, February 1, 2004; 61(2): 339 - 351. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Gunnett, D. D. Heistad, and F. M. Faraci Gene-Targeted Mice Reveal a Critical Role for Inducible Nitric Oxide Synthase in Vascular Dysfunction During Diabetes Stroke, December 1, 2003; 34(12): 2970 - 2974. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. S. Katusic, N. M. Caplice, and K. A. Nath Nitric Oxide Synthase Gene Transfer as a Tool to Study Biology of Endothelial Cells Arterioscler. Thromb. Vasc. Biol., November 1, 2003; 23(11): 1990 - 1994. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Zanetti, L. V. d'Uscio, I. Kovesdi, Z. S. Katusic, and T. O'Brien In Vivo Gene Transfer of Inducible Nitric Oxide Synthase to Carotid Arteries From Hypercholesterolemic Rabbits Stroke, May 1, 2003; 34(5): 1293 - 1298. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-Y. Sohn, F. Krotz, S. Zahler, T. Gloe, M. Keller, K. Theisen, T. M Schiele, V. Klauss, and U. Pohl Crucial role of local peroxynitrite formation in neutrophil-induced endothelial cell activation Cardiovasc Res, March 1, 2003; 57(3): 804 - 815. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. V. d'Uscio, S. Milstien, D. Richardson, L. Smith, and Z. S. Katusic Long-Term Vitamin C Treatment Increases Vascular Tetrahydrobiopterin Levels and Nitric Oxide Synthase Activity Circ. Res., January 10, 2003; 92(1): 88 - 95. [Abstract] [Full Text] [PDF] |
||||
![]() |
|