Reduced Gene Expression of Vascular Endothelial NO Synthase and Cyclooxygenase-1 in Heart Failure
Abstract Endothelium-dependent responses are depressed in coronary and peripheral blood vessels after the onset of pacing-induced heart failure in dogs and heart failure of various etiologies in humans. The present study was designed to examine whether these responses were due to decreases in the expression of endothelial cell NO synthase (ecNOS) and cyclooxygenase-1 (COX-1). After 1 month of left ventricular pacing, 8 mongrel dogs were monitored for heart failure as defined by clinical signs and left ventricular end diastolic pressures >25 mm Hg. Total RNA and protein were isolated from endothelial cells scraped from the thoracic aorta and analyzed by Northern and Western blotting, respectively. Blots probed with 32P-labeled cDNAs for ecNOS and COX-1 were quantified densitometrically, and results were normalized against GAPDH or von Willebrand factor (vWF). In arbitrary units, the ratios of ecNOS to GAPDH were 2.66±0.77 (mean±SEM, n=17) and 1.12±0.37 (n=6), and the ratios of COX-1 to GAPDH were 1.52±0.52 and 0.56±0.15 before and after heart failure, respectively. These represent 56% to 64% (P<.05) reductions in ecNOS and COX-1 gene expression. There was no change in the ratios of either COX-1 or ecNOS to vWF. There was also a marked reduction in ecNOS protein after heart failure, estimated at 70%. A marked reduction in nitrite production, a measure of enzyme activity, from thoracic aortas in response to stimulation by either acetylcholine or bradykinin also occurred. To determine whether ecNOS and COX-1 could be independently regulated, an orally active NO-releasing agent, CAS 936, was given to 7 normal dogs for 7 days, and aortic ecNOS and COX-1 mRNAs were analyzed. The ratio of ecNOS to GAPDH was depressed by 52% (P<.05) in aortas from these dogs, whereas the ratio of COX-1 to GAPDH was unchanged. Similar results were found when data were normalized to vWF. These results suggest that at least two endothelial vasodilator gene products are reduced in heart failure, as opposed to a selective defect in NO synthase gene expression.
Nitric oxide (NO) is recognized as an important endothelium-derived vasodilator in several vascular beds.1 Impaired endothelium-dependent coronary and peripheral vasodilation has been reported during heart failure in both humans and animals.2 3 4 5 6 7 For instance, Wang et al4 have shown that large coronary artery dilation is reduced after heart failure in dogs, and Drexler et al7 have shown that the endothelium-dependent dilation in the forearm of humans is depressed after the development of heart failure. Endothelial dysfunction may contribute to the vascular pathophysiology of heart failure, eg, increased total peripheral resistance, and could reflect reductions in the generation of NO and/or vasodilatory prostanoids such as prostacyclin (prostaglandin I2).8
Recent studies in our laboratory indicate that severe (pacing-induced) congestive heart failure in dogs is associated with alterations in endothelium-derived NO and prostaglandin-related regulation in the large coronary artery.4 Coronary microvessels from failing hearts exhibit reduced acetylcholine-dependent production of nitrite (the stable metabolite of NO in aqueous solution).4 In addition, increases in coronary artery diameter that are induced by brief occlusion, acetylcholine, or arachidonic acid are all depressed in heart failure.4 In contrast, the responses of the coronary artery diameter to nitroglycerin and exogenous prostacyclin are unaffected. Thus, heart failure is characterized by an apparent reduction in the release of NO and other vasodilators in the coronary vasculature.
The objective of the present study was to elucidate the molecular basis for the coronary endothelial dysfunction that we have found previously in heart failure. Perhaps endothelial dysfunction involves specific alterations in gene expression for ecNOS.9 10 Alternatively, heart failure may be associated with a nonspecific depression in ecNOS and other enzymes that produce vasodilators (eg, constitutive COX-1) by means of an alteration in endothelial function. To determine whether ecNOS gene expression could be selectively regulated, the changes in endothelial gene expression in dogs with HF were compared with those in normal dogs chronically administered an NO-releasing agent, CAS 936, with the goal of selectively downregulating ecNOS gene expression.
Materials and Methods
Surgical Preparation and Hemodynamic Measurements in Conscious Dogs
Eight adult mongrel dogs (body weight, 23 to 31 kg) were used in the present study. The dogs were sedated with acepromazine (0.3 mg/kg IM, Ayerst) and anesthetized (sodium pentobarbital, 25 mg/kg IV). A thoracotomy was performed in the left fifth intercostal space. A Tygon catheter was placed in the descending aorta, and a second catheter was inserted in the left atrial appendage. A solid-state pressure gauge (P6.5, Konigsberg Instruments) was placed in the apex of the LV. A human screw-type unipolar myocardial pacing electrode (manufactured in our laboratory) was placed on the LV. The wires and catheters were run subcutaneously to the intrascapular region. The chest was closed in layers, and the pneumothorax was reduced. The dogs were allowed to fully recover. Antibiotics were given after surgery. Heart rate and body temperature were monitored daily. After 10 days, the dogs were trained to lie quietly on the laboratory table.
At the beginning and end of the present study, cardiovascular responses were obtained in dogs used for pacing-induced heart failure. The following hemodynamic variables were measured when the dogs were lying quietly on a table: LV systolic and end-diastolic pressure, systemic arterial pressure, LV dP/dt, and heart rate, as described previously.4 11 12 The previously implanted catheters were attached to P23ID strain-gauge transducers (Statham Instruments) for the measurement of arterial and atrial pressures. LV pressure was measured with the solid-state pressure gauge. The data were recorded on a 14-channel tape recorder (Bell and Howell 3700B) and played back on a direct-writing oscillograph (Gould 2800s). Mean values were derived for pressures using 2-Hz resistance-capacitance filters. Heart rate was measured with a cardiotachometer (model 9857B, Beckman Instruments) from the LV pressure pulse interval. The first derivative of LV pressure, LV dP/dt, was derived with an operational amplifier (National Semiconductor 324). Triangular wave signals with known slopes were substituted for the pressure signals to calibrate the differentiators directly. The tape recording system and strip-chart recorder were calibrated periodically during the experiment to eliminate electronic drift.
The protocols were approved by the Institutional Animal Care and Use Committee of New York Medical College and conform to the guiding principles for the use and care of laboratory animals of the American Physiological Society and the National Institutes of Health.
ecNOS and COX-1 During Heart Failure
Dogs were paced at 210 bpm for 3 weeks, and the pacing was increased to 240 bpm for an additional week with an external pacemaker (model EV4543, Pace Medical), which the dog carried in a vest. Hemodynamic measurements in all 7 dogs were made before and at 4 to 5 weeks after chronic LV pacing when the pacer was turned off and with the heart in spontaneous rhythm. We have used these techniques previously.4 11 12 The mRNAs for ecNOS, COX-1, GAPDH, and vWF were isolated from the aorta of each dog. Protein was isolated from the abdominal aorta for Western analysis. In addition, to determine the enzyme activity from the aorta, pieces of the thoracic aorta were incubated in buffer with increasing doses of acetylcholine or bradykinin, and nitrite was measured by using the Greiss reaction. Nitrite is the hydration product of NO. To ensure that nitrite reflected NO production, the highest dose of each agonist was repeated after preincubation with nitro-l-arginine (100 μmol/L), which blocks NO synthase. We have used these techniques previously.4 13
Administration of an NO-Releasing Agent
In another group of 7 dogs, the orally active NO-releasing agent, CAS 936, was given twice a day at a dose of 20 mg/kg for 1 week. Only arterial pressure and heart rate were determined in those dogs. We have used this compound previously to cause NO-dependent vasodilation in the normal dog and dogs with pacing-induced heart failure.12 After 1 week, the dogs were killed, the aortas were removed, and endothelium was processed for measurement of mRNA.
ecNOS and COX-1 mRNA During Heart Failure and CAS 936
Total RNA was isolated from the endothelium of thoracic aortas as previously described13 from 7 dogs with chronic LV pacing, 7 dogs after 1 week of CAS 936 administration, and 17 historic control dogs (ie, dogs similarly instrumented). The vessel was flushed in situ with aerated sterile medium 199 with HBSS, and a 10- to 12-cm segment of the thoracic aorta was dissected free and opened lengthwise. Endothelium was removed by gently scraping the lumen with a scalpel blade (No. 10) after prewetting the surface with guanidine isothiocyanate to harvest total RNA.13 Total RNA (10 to 2 μg) was denatured by heating (65°C) in 50% (vol/vol) formamide and 4.4 mol/L formaldehyde, electrophoresed through a 1% agarose gel containing 2.2 mol/L formaldehyde, and transferred by capillary blotting to a nylon membrane (BioRad Zeta-Probe). The RNA was cross-linked to the blot by UV irradiation (Stratagene).
cDNA probes were labeled with [32P]dCTP (1 to 3×109 cpm/μg) by random priming (Ambion). The cDNA for ovine prostaglandin H synthase-1/COX-1 (catalogue No. R74) was obtained from Oxford Biomedical, and a 1.1-kb fragment of the cDNA for human GAPDH was from Clonetech; a full-length bovine ecNOS cDNA8 9 was used as in a previous study.13 After a 2-hour prehybridization period, nylon blots were hybridized overnight with radiolabeled cDNAs probes under high-stringency conditions as indicated below. Hybridizations with COX-1 and GAPDH were carried out at 42°C in 50% formamide, 5× Denhardt’s solution, 28 mmol/L sodium phosphate (pH 7.4), 375 mmol/L NaCl, 1% N-lauroylsarcosine, 0.5 mg/mL heparin, and 0.2 mg/mL salmon sperm DNA. These blots were washed at 55°C for 30 minutes each in 2× SSC/0.5% SDS, 1× SSC/0.5% SDS, and 0.5× SSC/0.25% SDS. Hybridization of blots with ecNOS was carried out at 65°C under high-stringency conditions as described previously9 ; these blots were washed twice in 2× SSC/0.1% SDS for 15 minutes at room temperature, followed by two washes in 0.4× SSC/0.1% SDS for 15 minutes at 65°C. Blots were dried and exposed to Kodak OMAT x-ray film in the presence of intensifying screens at −80°C for 3 (COX-1) to 10 (ecNOS) days.
Approximately 15 μg of endothelial total RNA was typically recovered from a single aorta. Five separate Northern blots were prepared for the heart failure study, and three were prepared for the NO-releaser study. Each blot contained samples from 1 or 2 dogs with heart failure along with 3 to 5 control dogs or from 2 or 3 dogs infused with CAS 936 along with 3 or 4 control dogs. Blots were stripped after probing with one cDNA and then rehybridized with another probe. All of the blots were sequentially probed for COX-1, GAPDH, ecNOS, and vWF13 mRNAs. Previous studies13 and unpublished data from this laboratory validated the use of either vWF (an endothelium-specific gene) or GAPDH as a denominator for RNA loading/transfer of canine aortic endothelial RNA. Optical densities of hybridization signals on several x-ray film exposures were quantified by laser-scanning densitometry (LKB Ultrascan) to determine steady state RNA levels.
ecNOS Protein During Heart Failure
Protein was also isolated from the abdominal aorta of each of the dogs used to study heart failure and saved for Western analysis of ecNOS. Protein was centrifuged for 2 minutes at 4°C at 12 000g, the PBS was decanted, and the pellet was resuspended in 60 to 100 μL of detergent containing lysis buffer.14 The composition of the lysis buffer was 10 mmol/L HEPES, pH 7.45, 315 mmol/L sucrose, 10% glycerol, 1% NP-40 detergent, 0.1 mmol/L EDTA, 1 mmol/L dithiothreitol, 2 μg/mL aprotinin, and 10% glycerol/mL of leupeptin, soybean trypsin inhibitor, and phenylmethylsulfonyl fluoride. The endothelial lysate was stored at −80°C until used for protein quantification and subsequent precipitation by TCA (5% final TCA) for SDS-PAGE of crude endothelial lysates. Between 20 and 100 μg of total endothelial protein scrapings (lysed in detergent buffer) were recovered per abdominal aorta segment. Protein recovery following detergent solubilization of the lysate was ≈50%. In three initial Western blots (each with 2 or 3 control and 2 or 3 HF samples, total aortic protein before solubilization), individual dogs were analyzed (25 to 100 μg TCA-precipitated protein per lane).
Two methods were used to quantify ecNOS protein: (1) Western blotting of isolated protein from individual dogs and (2) affinity purification of pooled NADPH binding proteins from several dogs by use of ADP-Sepharose followed by Western blotting. The first method and three different blots with individual lanes were used for 7 HF and 7 normal dogs, and ecNOS protein (135-kD band) was found to be generally more intense in control samples compared with HF samples. However, in two of the three experiments, ecNOS was barely detectable (even with the sensitive chemiluminescence secondary antibody detection method) and made our conclusions unreliable. Therefore, we used the second method and increased the amount of the ecNOS protein by almost an order of magnitude by combining detergent-solubilized protein from a number of dogs (7 HF and 9 normal dogs) and again performed Western blot analysis.
The second method involved gentle mixing of the lysate for 16 hours at 4°C followed by 10-minute centrifugation to remove detergent-insoluble protein. Soluble protein was pooled from several animals and quantified, and comparable total amounts (≈300 μg in 0.5 mL; see Fig 4⇓) from control and HF groups were further incubated for 2 hours at 4°C with 30 μL of a 50% suspension of ADP-Sepharose as described previously.14 ecNOS quantitatively binds to this resin and allows for concentration of ecNOS protein and other NADPH binding proteins. Protein samples were placed in 50 μL Laemmli sample buffer before SDS–8% PAGE. Similar procedures were performed on protein isolated from cultured human umbilical vein endothelial cells grown in culture and used as a method for determining the sensitivity of the method.
Proteins were transferred to PVDF membranes (Millipore) in 10 mmol/L CAPS buffer. The PVDF membrane was blocked for 1.5 hours at room temperature in 5% BSA in 1× TBS with 0.05% Tween 20, followed by an overnight incubation at 4°C with the primary ecNOS antibody diluted 1:1000 in TBS-T/1% BSA. The membrane was washed four times in TBS-T/1% BSA, incubated 1.5 hours with the secondary antibody (1:10 000), and finally washed four times in TBS-T/BSA before signal detection with the chemiluminescence ECL method (Amersham). The 135-kD ecNOS band was detected by scanning densitometry of film autoradiograms as previously described. The primary antibody was rabbit polyclonal anti-bovine ecNOS peptide antibody (Affinity Bioreagents); rabbits were immunized with a C-terminal peptide (amino acids 599 to 613) coupled to keyhole limpet hemocyanin.
Nitrite Production by Thoracic Aorta After Heart Failure
The thoracic aorta was cut into 30- to 50-mg pieces, and each piece was incubated in PBS for 20 minutes (control) or in the presence of increasing doses of acetylcholine or bradykinin for 20 minutes. After 20 minutes, the buffer was removed, and the amount of nitrite produced was quantified by using the Greiss reaction as we have done previously.4 13 In all of the studies, nitro-l-arginine (100 μmol/L) was added to the aorta 10 minutes before the addition of the highest dose of either bradykinin or acetylcholine to ensure that the production of nitrite reflected NO production.
Results are expressed as the mean±SEM from n dogs. The data from Northern blots were analyzed by an unpaired Student’s t test; a paired t test was used in dogs to compare hemodynamics in the same animal before and after heart failure or chronic CAS 936 treatment. A value of P<.05 was considered statistically significant. All figures were produced by using Slidewrite Plus for Windows.
Mongrel dogs were subjected to a ventricular pacing regime for 4 weeks, which produced overt heart failure as evidenced by ascites, edema, and dyspnea, as previously reported by us.4 11 12 Hemodynamic characteristics for these animals before and after chronic pacing are summarized in the Table⇓. LV dP/dt, LV systolic pressure, and mean arterial blood pressure were significantly decreased 4 weeks after pacing by 44%, 21%, and 17%, respectively, while spontaneous heart rate increased by 82% (all P<.05 from prepacing). The most notable change in heart failure was a 491% increase in LV end-diastolic pressure (P<.05 from the control value).
ecNOS and COX-1 mRNA During Heart Failure
Total RNA was isolated from endothelium scraped from the thoracic aortas of dogs subjected to chronic pacing and control dogs. Steady state mRNA levels for ecNOS and COX-1 were evaluated in these samples by Northern blotting (Fig 1⇓). Under the hybridization conditions used, each radiolabeled cDNA probe identified one major mRNA species. The bovine ecNOS cDNA hybridized to a 4.4-kb ecNOS mRNA and did not detect a cytokine-inducible-INOS.13 Similarly, the ovine COX-1 cDNA bound to a 2.8-kb COX-1 mRNA and did not interact with a larger message (4.2 kb, near 28S ribosomal RNA), which would be expected for the mitogen-responsive COX-2.15 16 17 Scanning densitometry of COX-1 or ecNOS mRNA signals was normalized against that for GAPDH (Fig 2⇓) and vWF (Fig 3⇓) to quantify differences between control and HF groups. Heart failure was associated with a significant reduction (P<.05) in both mRNA levels, with a significant (P<.05) decrease in both the ratio of COX-1 to GAPDH (64%) and the ratio of ecNOS to GAPDH (56%) (Fig 2⇓). Because there was also a reduction in vWF after pacing-induced heart failure, the ratios of COX-1 to vWF or ecNOS to vWF were not altered (Fig 3⇓).
ecNOS Protein During Heart Failure
In preliminary experiments, the signal-to-noise ratio for detection of the 135-kD ecNOS protein was poor when material from individual dogs was evaluated. We subsequently pooled protein from 7 dogs with HF and 9 normal dogs and used ADP-Sepharose affinity chromatography to enrich ecNOS. There was a marked reduction in ecNOS protein after pacing-induced heart failure as illustrated by Western blotting (Fig 4⇓). Four of seven of the animals used in the HF group for data in Fig 4⇓ were also used for ecNOS gene expression in Fig 2⇑. Scanning densitometry showed an ≈70% reduction in ecNOS protein in the pooled dogs with HF compared with control dogs.
Nitrite Production by Thoracic Aorta After Heart Failure
There was a marked reduction in acetylcholine- or bradykinin-induced nitrite production in the thoracic aorta after pacing-induced heart failure as shown in Fig 5⇓. When nitro-l-arginine was incubated with aorta before the addition of the highest dose of either bradykinin or acetylcholine, the increase in nitrite production was totally eliminated. This indicates that nitrite production reflects NO production.
Administration of an NO-Releasing Agent
There were no significant changes in mean arterial blood pressure or heart rate during the chronic infusion of CAS 936, consistent with the results of our previous studies.11 In contrast to the concurrent reductions in ecNOS and COX-1 in our model of heart failure, there were marked differences in the expression of these genes in dogs with NO infusion. There was a selective 52% decrease in the ratio of ecNOS to GAPDH (P<.05 versus the control value) and no change in the ratio of COX-1 to GAPDH (Fig 6⇓). Unlike the response during heart failure, there was also a depression in the ratio of ecNOS to vWF but not in the ratio of COX-1 to vWF (Fig 7⇓).
The most striking finding of the present study is the downregulation of the mRNA for two enzymes necessary for the constitutive synthesis of endothelium-dependent vasodilators after the development of pacing-induced cardiomyopathy in dogs. The decrease in ecNOS mRNA also resulted in a reduction in ecNOS protein and reduced nitrite production in vitro from the aorta, supporting the reduced production of nitrite from coronary microvessels described in our previous study.4 This downregulation was not due to some autocrine link between NO and prostaglandin production in blood vessels but was a defect due to heart failure. This is supported by the finding that chronic exposure to an NO-releasing agent produced a selective reduction in ecNOS gene expression. Thus, although COX-1 and ecNOS can be independently regulated, after the development of overt heart failure, there is a depression of a number of endothelial cell gene products.
The hemodynamic changes that occurred in dogs with either pacing-induced cardiomyopathy or chronic NO infusion were similar to those found in our previous studies.4 11 12 Most important, pacing-induced heart failure was characterized by increases in LV end-diastolic pressure and reductions in myocardial contractile state, mean arterial pressure, and resting tachycardia. In addition, there was an increase in LV diameter and some diffuse fibrosis in the LV.18 In other studies using this model, the reduction in cardiac output occurred only after a prolonged period of pacing in conscious dogs,19 which is, by definition, the onset of overt heart failure. Thus, we were careful to assess changes in endothelial function in dogs with prolonged LV pacing. In previous studies using this model, we have also shown that the production of nitrite, the hydration product of NO metabolism in vitro, was reduced and that endothelium-dependent dilation in vivo was almost abolished after the development of pacing-induced heart failure.4
In dogs with chronic administration of the NO-releasing agent CAS 936, there was no change in heart rate or mean arterial pressure. These results are consistent with many previous studies indicating that organic nitrates dilate primarily the venous circulation and large arteries.20 21 These results are also consistent with our previous studies indicating that CAS 936 reduces preload in normal dogs and dogs with HF, perhaps by dilating veins, and increases large coronary artery diameter without affecting coronary or peripheral vascular resistance.11 The selective large vessel–dilating effects of this NO-releasing agent suggested to us that this agent caused such dilation by an NO-dependent mechanism, making it likely that the aorta would also be a target for this compound.
The reductions in ecNOS and COX-1 gene expression that we measured after pacing-induced heart failure are consistent with our previous studies and may be the mechanism responsible for the reduced NO-dependent dilation that we measured in vivo and the reduced nitrite production that we found in vitro.4 Furthermore, the present investigation supports our conclusions that the heart failure–associated defect in nitrite production in vitro was not due to reduced availability of l-arginine, since the addition of l-arginine in vitro did not enhance nitrite production in coronary microvessels from either normal or dogs with HF. Since ecNOS protein levels and nitrite production were also reduced, these further suggest a reduction in NO synthesis. This may be analogous to the impaired peripheral vasodilator responses in humans with congestive heart failure,5 some of which our data suggest may be attributed to reduced release of NO.
The mechanisms responsible for the downregulation of the constitutive NO synthase gene expression after heart failure are not known. In our previous studies, we have found that 3 weeks of pacing results in an increase in endothelium-dependent dilation of large coronary arteries in vivo, and only after the development of overt heart failure do EDRF-dependent responses disappear.4 22 These data are supported by a recent study by O’Murchu et al,23 although there is some difference in the interpretation of those data. In a recent study from our laboratory, Zhao et al24 also showed that cholinergic reflex coronary NO-dependent vasodilation is reduced after 4 weeks of chronic pacing (ie, after the development of overt congestive heart failure) but not after 3 weeks of pacing in the conscious dog (ie, during compensated cardiac dysfunction but no clinical signs of heart failure). This is consistent with preliminary studies demonstrating no obvious reduction in ecNOS gene expression in extracts prepared from a single dog paced for 3 weeks before the development of overt heart failure. In addition, we have recently found that after 1 week of chronic exercise, there is an increase in EDRF-dependent responses, increased nitrite production in vitro, and upregulation of ecNOS mRNA levels in the aorta.13 This is probably due to a biochemical mechanism that is coupled to the increase in blood flow and vessel shear stress that occurs during exercise. In our pacing model, there is no increase in cardiac output with pacing, since the end-diastolic filling time will be reduced at the high heart rates that we used and since there will be no alteration in the regulation of venous return by the peripheral circulation.19 Thus, there is no obvious change in aortic blood flow or cardiac output. However, there will be a marked alteration in the “pulsatility,” since heart rate and stretch of the aorta will be 210 to 240 times per minute as long as the pacemaker is turned on. Pulsatility is believed to be a potent stimulus for EDRF production in blood vessels both in vivo25 and in vitro,26 and cyclic strain on cultured endothelial cells increases ecNOS activity and nitrite release.27 Alternatively, there may be alterations in local or circulating hormones that occur with the development of heart failure, such as angiotensins.28
However, our data indicate that the alteration in NO synthase gene expression is not selective, since cyclooxygenase gene expression is also decreased. This is further supported by the finding that the ratio of either COX-1 or ecNOS to vWF is not altered after heart failure. vWF gene expression is also a marker for endothelial cells. Thus, whatever mechanism is responsible for the reduction in ecNOS in heart failure may also be responsible for the reduction in COX-1, although these two genes could be independently regulated.
Another provocative finding in the present study is that infusion of an NO donor selectively downregulated ecNOS gene expression. To our knowledge, this is the first demonstration that exogenous NO can influence steady state ecNOS mRNA levels and complements the inhibitory action of NO on NO production. Recently Pilz et al29 showed that both cGMP and NO can regulate a number of transcriptional factors, including TPA response element–regulated genes. The molecular mechanism of NO inhibition of ecNOS is unknown and under investigation. A previous study in cultured cells by Buga et al30 suggested that chronic exposure to NO could inhibit NO production, presumably by inhibiting enzyme activity. The present study provides evidence of an additional mechanism for the NO control of NO synthase by the control of gene expression. Furthermore, these data suggest that NO synthase and COX-1 gene expression are separable and that they can be independently regulated.
There is increasing evidence that myocytes can make NO, especially after exposure to cytokines.31 32 There is also some evidence that circulating tumor necrosis factor increases in patients with heart failure and that peripheral venous nitrate, the product of NO metabolism in blood, is increased in heart failure.33 34 Together, these studies suggest that NO production may be enhanced after heart failure. Our previously published in vitro data from dogs4 and from humans35 suggest that this enhanced NO production does not come from blood vessels. We have examined, in a preliminary fashion, blood vessels for the expression of an inducible form of NO synthase. The basal production of NO by these tissues was not as high as would be expected. Thus, we did not systematically evaluate the expression of the inducible genes, although we do not believe that there is enhanced production of NO or prostaglandins in blood vessels from the dogs in the present study.
In summary, we have found that both ecNOS and COX-1 mRNA are reduced in aortic endothelial cells from dogs with HF and that ecNOS protein and nitrite production, reflecting enzyme activity, are also reduced. The reduced gene expression for these enzymes may be responsible for the reduced NO production that we have previously observed during heart failure in dogs and, as we recently reported, in human coronary microvessels.4 35 Furthermore, since two endothelial vasodilator systems are depressed, although they can be independently regulated by chronic exposure to NO in vivo, these alterations are most likely indicative of altered regulation of multiple endothelial cell genes after heart failure.
Selected Abbreviations and Acronyms
|ecNOS||=||endothelial cell NO synthase|
|EDRF||=||endothelium-dependent relaxing factor|
|HF||=||pacing-induced heart failure (used in conjunction with dog groups and samples from those groups)|
|LV||=||left ventricle (left ventricular)|
|vWF||=||von Willebrand factor|
This study was supported by Biomedical Research Support grant RR05398 to New York Medical College from the Division of Research Resources, National Institutes of Health (Dr Smith); by grant 94-324, American Heart Association, New York State Affiliate, Inc (Dr Smith); and by National Institutes of Health, National Heart, Lung, and Blood Institute grants PO-1 HL-43023, HL-50142, and HL-53053 (Dr Hintze). Dr Smith was the 1994 Tarnower Scholar, American Heart Association, Westchester Chapter of the New York State Affiliate, Inc. Dr Jie Wang assisted with pilot studies as a 1993 Research Fellow of the American Heart Association, New York State Affiliate, Inc. Dr Alan Springer (Cell Biology and Anatomy) and Dr Theresa Burke-Wolin (Experimental Pathology) kindly provided access to Image Analysis software for autoradiogram graphics. Patricio Villalon (Experimental Pathology) provided the human umbilical vein endothelial cell protein for Western blotting experiments.
Preliminary reports of these findings were presented at the 66th (November 8-11, 1993, in Atlanta, Ga) and 67th (November 14-17, 1994, in Dallas, Tex) Scientific Sessions of the American Heart Association.
This manuscript was sent to Leslie A. Leinwand, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
- Received December 21, 1994.
- Accepted August 29, 1995.
- © 1996 American Heart Association, Inc.
Kaiser L, Spickard RC, Olivier NB. Heart failure depresses endothelium-dependent responses in canine femoral artery. Am J Physiol.. 1989;256:H962-H967.
Wang J, Seyedi N, Xu X, Wolin MS, Hintze TH. Defective endothelium-mediated control of coronary circulation in conscious dogs after heart failure. Am J Physiol.. 1994;266:H670-H680.
Katz SD, Schwarz M, Yuen J, LeJemtel TH. Impaired acetylcholine-mediated vasodilation in patients with congestive heart failure: role of endothelium-derived vasodilating and vasoconstricting factors. Circulation. 1993;88:55-61.
Treasure CB, Vita JA, Cox DA, Fish RD, Gordon JB, Mudge GH, Colucci WS, St John Sutton MG, Selwyn AP, Alexander RW, Ganz P. Endothelium-dependent dilation of coronary microvasculature is impaired in dilated cardiomyopathy. Circulation. 1990;81:772-779.
Drexler H, Hayoz D, Munzel T, Just H, Zelis R, Brunner HR. Endothelial dysfunction in chronic heart failure: experimental and clinical studies. Arzneimittelforschung. 1994;44(suppl I):455-458.
Busse R, Hecker M, Fleming I. Control of nitric oxide and prostacyclin synthesis in endothelial cells. Arzneimittelforschung. 1994;44(suppl I):392-396.
Sessa WC, Harrison JK, Barber CM, Zeng D, Durieux ME, D’Angelo DD, Lynch KR, Peach MJ. Molecular cloning and expression of a cDNA encoding endothelial cell nitric oxide synthase. J Biol Chem.. 1992;267:15274-15276.
Wang J, Zhao G, Shen W, Ochoa M, Moore D, Hubbard JW, Hintze TH. Effects of an orally active NO-releasing agent, CAS 936, and its active metabolite, 3754, on cardiac and coronary dynamics in normal conscious dogs and after pacing-induced heart failure. J Cardiovasc Pharmacol. 1993;22(suppl 7):S51-S58.
Belloni FL, Wang J, Hintze TH. Adenosine causes bradycardia in pacing-induced heart failure. Circulation. 1992;85:1118-1124.
Sessa WC, Pritchard K, Seyedi N, Wang J, Hintze TH. Chronic exercise in dogs increases coronary vascular nitric oxide production and endothelial cell nitric oxide synthase gene expression. Circ Res.. 1994;74:349-353.
Liu J, Sessa WC. Identification of covalently bound amino-terminal myristic acid in endothelial nitric oxide synthase. J Biol Chem. 1994;269:11691-11694.
Inase N, Levine TM, Lazarus SC. Mitogen-inducible prostaglandin G/H synthase is expressed in canine mastocytoma cells. Am J Respir Cell Mol Biol.. 1993;9:526-532.
Pritchard KA Jr, O’Banion MK, Miano JM, Vlasic N, Bhatia UG, Young DA, Stemerman MB. Induction of cyclooxygenase-2 in rat vascular smooth muscle cells in vitro and in vivo. J Biol Chem.. 1994;269:8504-8509.
Szoke E, Zhang X, Wang J, Anversa P, Hintze TH. Myocardial fibrosis and chronic tachycardia induced heart failure in dogs. Circulation. 1993;88(suppl I):I-541. Abstract.
Shannon RP, Komamura K, Stambler BS, Bigaud M, Manders WT, Vatner SF. Alterations in myocardial contractility in conscious dogs with cardiomyopathy. Am J Physiol. 1991;260:H1903-H1911.
Hintze TH, Vatner SF. Comparison of the effects of nifedipine and nitroglycerin on large and small coronary arteries and cardiac function in conscious dogs. Circ Res. 1983;52(suppl 1):139-146.
Hintze TH, Wang J, Seyedi N, Wolin MS. Role of EDRF/NO in chronic high coronary blood flow states during myocardial dysfunction and failure. In: Bevan J, Kaley G, Rubanyi G, eds. Flow Velocity Control of the Circulation. New York, NY: Oxford University Press; 1995.
O’Murchu B, Miller VM, Perella MA, Burnett JC. Increased production of nitric oxide in coronary arteries during congestive heart failure. J Clin Invest. 1994;93:165-171.
Zhao G, Shen W, Xu X, Ochoa M, Bernstein R, Hintze TH. Selective impairment of vagally mediated nitric-oxide-dependent coronary vasodilation in conscious dogs after pacing induced heart failure. Circulation. 1995;91:2655-2663.
Canty JM Jr, Schwartz JS. Nitric oxide mediates flow-dependent epicardial coronary vasodilation to changes in pulse frequency but not mean flow in conscious dogs. Circulation. 1994;89:375-384.
Rubanyi G, Romero JC, Vanhoutte PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol. 1986;250:H1145-H1149.
Holtz J, Goetz RM. The endothelium and the renin-angiotensin system. Arzneimittelforschung. 1994;44(suppl I):397-402.
Pilz RB, Sushasini M, Idriss S, Meinkoth JL, Boss GR. Nitric oxide and cGMP analogs activate transcription from AP-1-responsive promoters in mammalian cells. FASEB J. 1995;9:552-558.
Buga GM, Griscavage JM, Rogers NE, Ignarro LJ. Negative feedback regulation of endothelial cell function by nitric oxide. Circ Res.. 1993;73:808-812.
Brady AJB, Poole-Wilson PA, Harding SE, Pickering T. Nitric oxide production within cardiac myocytes reduced their contractility in endotoxemia. Am J Physiol. 1993;265:H176-H182.
Balligand JL, Ungureanu D, Kelly RA, Smith T. Abnormal contractile function due to induction of nitric oxide synthesis in rat cardiac myocytes follows exposure to activated macrophage conditioned medium. J Clin Invest. 1993;91:2314-2319.
Kichuk MR, Seyedi N, Marboe C, Addonizio L, Michler R, Hintze TH. Reduced EDRF production from coronary microvessels from the failing explanted human heart. FASEB J. 1994;8:A826.