Impaired Nitric Oxide–Mediated Renal Vasodilation in Rats With Experimental Heart Failure
Role of Angiotensin II
Background Congestive heart failure (CHF) is associated with a decrease in renal perfusion. Because endothelium-derived NO is important in the regulation of renal blood flow (RBF), we tested the hypothesis that an impairment in the NO system may contribute to the decrease in RBF in rats with experimental CHF.
Methods and Results Studies were performed in rats with experimental high-output CHF induced by aortocaval (AV) fistula and sham-operated controls. In controls, incremental doses of acetylcholine (ACh, 1 to 100 μg · kg−1 · min−1) increased RBF and caused a dose-related decrease in renal vascular resistance (RVR). However, the increase in RBF and decrease in RVR were markedly attenuated in rats with CHF. Likewise, the effects of ACh on urinary sodium and cGMP excretion were also diminished in CHF rats, as was the renal vasodilatory effect of the NO donor S-nitroso-N-acetylpenicillamine (SNAP). These attenuated responses to endothelium-dependent and -independent renal vasodilators in CHF rats occurred despite a normal baseline and stimulated NO2+NO3 excretion and normal expression of renal endothelial NO synthase (eNOS), as determined by eNOS mRNA levels and immunoreactive protein. Infusion of the NO precursor l-arginine did not affect baseline RBF or the response to ACh in rats with CHF. However, administration of the nonpeptide angiotensin II receptor antagonist A81988 before ACh completely restored the renal vasodilatory response to ACh in CHF rats.
Conclusions This study demonstrates that despite a significant attenuation in the NO-related renal vasodilatory responses, the integrity of the renal NO system is preserved in rats with chronic AV fistula. This impairment in NO-mediated renal vasodilation in experimental CHF appears to be related to increased activity of the renin-angiotensin system and may contribute further to the decrease in renal perfusion seen in CHF.
Reduced RBF is one of the most consistent manifestations of CHF. Traditionally, this decrease in renal perfusion has been attributed to a combination of several pathogenetic mechanisms: a reduction in the pumping capacity of the failing myocardium and a compensatory activation of neurohormonal vasoconstrictor systems such as the sympathetic nervous system and the renin-angiotensin system.1
In the past decade, evidence has been provided regarding the importance of locally released vasoactive substances in the regulation of RBF and systemic hemodynamics, in particular the endothelium-derived NO system.2 It is now accepted that NO, generated in the endothelial cells of the renal vasculature, plays an important role in the physiological regulation of RBF.3 NO is constitutively produced from its precursor, l-arginine, by the enzyme eNOS and acts on adjacent smooth muscle cells to exert vasodilatory tone on the renal microvasculature, primarily the afferent arteriole.4 In addition, NO affects renal function by modulating tubuloglomerular feedback5 and renin release6 and by altering tubular salt reabsorption.7 These actions are thought to be mediated by NO generated in mesangial cells, macula densa, and epithelial tubular cells.8,9
In view of the importance of the NO in regulating RBF, it is possible that altered activity of the NO system may be involved in the pathogenesis of the renal hypoperfusion in CHF. Indeed, several studies in recent years have clearly documented an impaired endothelium-dependent vascular response in heart failure.10–17 In most of these studies, the response to acetylcholine, an endothelium-dependent vasodilator, was found to be markedly attenuated in patients and experimental animals with CHF and in isolated vessels from animals with CHF examined in vitro.10,11,13–15 The endothelium-independent relaxation, such as that produced by nitroprusside or nitroglycerin, was found to be preserved in some10,11 but not all studies.13,16 In none of those studies, however, was the role of the impaired endothelial function addressed specifically as a cause of the altered renal perfusion in CHF. Moreover, with the exception of one study,12 no data on endothelium-dependent renal vasorelaxation in CHF have been reported.
Several mechanisms could lead to impaired activity of the NO system in CHF. These include downregulation of eNOS,18 decreased availability of the NO precursor l-arginine,19 and overriding activity of counterregulatory vasoconstrictor systems such as the renin-angiotensin and the sympathetic nervous systems. Activation of the renin-angiotensin system has been shown to play a key role in the pathogenesis of renal vasoconstriction and sodium retention in CHF.1 Recently, it has been speculated that increased expression of ACE in the vascular wall in CHF may lead to downregulation of the eNOS, secondary to a decrease in bradykinin activity.18 In addition, an antagonistic interrelationship between NO and Ang II in the regulation of renal hemodynamics has been reported.20,21 Thus, it is possible that the high circulating levels of Ang II in CHF might counteract the renal vasodilatory activity of NO and thereby reduce renal perfusion.
Whether baseline activity of the NO system is decreased in CHF is another area of controversy. Although this was the initial interpretation for the finding of a blunted response to acetylcholine, more recent studies in patients with CHF have demonstrated an increase in plasma levels of nitrates, the final metabolic product of NO.22 The idea that baseline NO activity is actually elevated in CHF and may serve as a compensatory response to the excessive vasoconstriction is further supported by the demonstration that treatment with NO synthase blockers causes an exaggerated vasoconstrictor response in patients with CHF.14,23
To shed light on some of these questions, the present study evaluated the possibility that an impairment in the activity of the NO system may be involved in the abnormal regulation of RBF in rats with experimental CHF and explored the potential mechanisms responsible for this impairment. We have shown previously that rats with chronic AV fistula display hemodynamic and neurohumoral alterations that are characteristic of patients with CHF. These include elevated central venous and cardiac pressures, reduction in glomerular filtration and RBF, renal retention of salt and water, and activation of several neurohumoral systems, especially the renin-angiotensin system. We used this model to specifically examine the renal vasodilatory responses to endothelium-dependent and -independent maneuvers.
Studies were performed on male Wistar rats (250 to 320 g) fed standard rat chow (0.6 g% sodium) and given tap water ad libitum. CHF was induced by surgical creation of an AV fistula (side to side, 0.8 to 1.0 mm) between the abdominal aorta and inferior vena cava according to the method described by Stumpe et al24 and adapted in our laboratory.25–27 After surgery, the animals were allowed to recover and then placed in individual metabolic cages for daily monitoring of urinary flow rate and sodium excretion. A matched group of sham-operated rats served as controls.
In Vivo Studies
Five to 7 days after the operation, the animals were anesthetized with Inactin (90 to 100 mg/kg IP) and prepared for clearance and renal hemodynamic studies. After a tracheotomy, the left carotid artery and right jugular vein were cannulated with polyethylene tubing (PE50) for blood pressure monitoring, periodic blood sampling, and infusion of solutions. The urinary bladder was catheterized (PE50) via a suprapubic incision for urine collections. Inulin (2% in normal saline) was infused intravenously at a rate of 1.5% of body wt/h throughout the experiment.
For measurements of renal hemodynamics, the left renal artery was exposed via a midabdominal incision, and an ultrasonic flow probe (type 1RB) connected to an ultrasonic flowmeter (model T206, Transonic Corp) was placed around the renal artery. Arterial blood pressure was continuously monitored with a pressure transducer (model 156 PC05GWL, Microswitch). RBF and MAP were continuously recorded via a computerized data acquisition system with Labtech Acquire software. RVR was calculated by the standard formula (RVR=MAP/RBF) and expressed as RU. After a 60-minute equilibration period, the following protocols were performed.
Response to Acetylcholine Administration
Control animals (n=6) and CHF rats (n=8) were prepared as previously described. After a 30-minute baseline period, acetylcholine was infused in incremental doses (1, 10, and 100 μg · min−1 · kg body wt−1) over a 30-minute period for each dose, followed by a 30- to 45-minute recovery period. Measurements of renal clearance and excretory parameters were performed in additional groups of control rats (n=8) and rats with AV fistula (n=9) that were prepared similarly, with the exception that the abdominal cavity was not opened in these experiments. Timed urine collections were obtained during baseline, experimental, and recovery periods. Urine was collected into preweighed tubes on ice, and urine volumes were determined gravimetrically. Samples of 100 μL were diluted with distilled water (1:10), frozen immediately, and kept at −70°C until assay for cGMP content. The remainder of the urine samples were used for determination of inulin, sodium, and nitrite/nitrate concentrations. Blood samples (0.3 mL) were obtained at the midpoint of each clearance period for measurement of inulin and sodium concentrations in plasma.
Response to Endothelium-Independent NO Donor
To determine the hemodynamic response to endothelium-independent NO donor, SNAP (Research Biochemical International) was administered to control (n=8) and CHF (n=7) rats. SNAP was dissolved in DMSO (2.5 mg/mL) and was further diluted in cold saline to a final concentration ranging from 60 to 180 μg/mL. The rats were prepared as described in the previous protocol. After a 30-minute baseline period, SNAP was infused in two doses of 10 and 30 μg · kg−1 · min−1, each dose for 30 minutes, followed by a recovery period in which the infusion of SNAP was omitted and replaced by infusion of saline. Recordings of renal hemodynamic parameters were performed during baseline, experimental, and recovery phases as previously described.
Urinary Excretion of NO2+NO3 in Control and CHF Rats
Daily measurements of urinary NO2+NO3 excretion were conducted in two groups of rats (n=8 to 9 in each group). All rats were placed in metabolic cages and fed standard rat chow and tap water, and urine was collected in sterile tubes containing antibiotic/antimyocotic solution (100 U/mL penicillin, 100 U/mL streptomycin, and 25 μg/mL amphotericin B). After 4 to 5 days, an AV fistula was surgically created in one group of rats, while the second group of animals underwent sham operation. After surgery, the rats were returned to the metabolic cages, and urine was collected daily. After 6 to 7 days, the rats were provided the same diet but were switched to drinking water containing 1% l-arginine, and their urine was collected as described above.
Effects of l-Arginine on Baseline and Acetylcholine-Mediated Renal Vasodilation
To test the possibility that a decrease in the availability of l-arginine, the endogenous precursor of NO, may be involved in the diminished renal vasodilatory response to acetylcholine, an additional group of rats with AV fistula (n=6) was studied. This group was prepared identically to that described in protocol 1, with the exception that after baseline hemodynamic parameters were obtained, an infusion of l-arginine (10 mg · kg−1 · min−1) was started and sustained throughout the experiment. After 15 minutes, acetylcholine was administered in three incremental doses followed by a recovery period, as described in protocol 1. Hemodynamic measurements were repeated at each dose of acetylcholine and during the recovery phase. In an additional group of rats with AV fistula (n=5), the effects of continuous infusion of l-arginine (10 mg · kg−1 · min−1 for 60 minutes) were studied.
Effects of Ang II Receptor Blockade on Acetylcholine-Mediated Renal Vasodilatation
In these experiments, the renal response to acetylcholine administration was studied in CHF rats in which the activity of the renin-angiotensin system was inhibited by a novel nonpeptide Ang II receptor antagonist, 2-(N-propyl-N[2′-[1H-tetrazol-5-yl]biphenyl-4yl)methyl]aminopyridine-3-carboxylic acid, (A-81988; Abbott Laboratories). Rats with AV fistula (n=7) were prepared for hemodynamic and clearance studies as described in protocol 1. After baseline data had been obtained, the Ang II receptor antagonist A-81988 was administered as an intravenous bolus at a dose of 0.1 mg/kg body wt, followed by a sustained infusion of 0.03 mg · kg−1 · h−1 throughout the experiment. The efficacy of this compound in blocking the effects of Ang II have been described.28 Moreover, in preliminary studies, we found that this dose effectively blocked the renal vasoconstrictor action of Ang II injected in a dose of 0.1 μg/kg body wt, which in the absence of A-81988 caused a >30 mm Hg increase in MAP in normal animals (data not shown). Thirty minutes after the initiation of the Ang II receptor antagonist infusion, acetylcholine was administered in three incremental doses, followed by a final recovery phase, as described above. Renal hemodynamic and clearance parameters were obtained at each dose of acetylcholine and during the recovery period.
To further evaluate the effects of the Ang II receptor antagonist, two additional groups of rats (n=5 in each) were studied. In one group, consisting of normal rats, the renal hemodynamic response to acetylcholine was studied as described in protocol 1, with the exception that A-81988 was administered after baseline parameters were obtained and was sustained throughout the experiment. The second group consisted of rats with AV fistula in which the Ang II receptor antagonist was administered after baseline recordings were obtained and was continuously infused throughout the experiment. However, no acetylcholine was administered in these animals. These two groups served as controls for normal rats infused with acetylcholine and for the CHF group treated with the Ang II antagonist and acetylcholine.
In Vitro Studies
Determination of eNOS mRNA
To further assess the integrity of the renal NO system, control rats and rats with CHF were decapitated, and their kidneys were removed and immediately placed in liquid nitrogen. Total RNA was extracted from the whole kidney as described by Chomczynski and Sacchi29 and was quantified by spectrophotometry. Quantitative RT followed by PCR was applied. Endothelial NOS cDNA was synthesized from 2 μg total RNA with a specific downstream primer: 5′-GCCG GATCCTCCAGGAGGGTGTCCACCGCATGGAA-3′ (bases 3038 to 3069; synthesized by Lofstrand Laboratories). Avian myeloblastosis virus reverse transcriptase (15 U per reaction; Promega) was used for RT in a volume of 20 μL. PCR was then applied with 2 μL of the resulting cDNA and the GeneAmp kit (Cetus Perkin-Elmer), with the upstream primer 5′-CCGGAATTCGAATACCAGCCTGATCCATG-3′ (bases 2456 to 2487) and the downstream primer used for RT. Each PCR reaction mixture contained 200 μmol/L each of dATP, dGTP, dTTP, and dCTP.
In a preliminary study, we found that the minimum number of PCR cycles necessary to obtain a visible product on an agarose gel was 32 for eNOS and 23 for β-actin, an internal standard. The quantity of the product yielded after the PCR cycles was directly proportional to the amount of DNA used. After an initial denaturation step at 94°C for 2 minutes, cycles of annealing at 56°C for 0.75 minute, elongation at 72°C for 1.5 minutes, and denaturation at 94°C for 0.5 minute were applied with 10% of the cDNA described above. The expected size of the eNOS PCR product is 614 bp. The β-actin primers were upstream, 5′-GACTACCTCATGAAGATCCTGACC-3′ (nucleotides encoding amino acids 210 to 217) and downstream, 5′-TGATCTTCATGGTGCTAGGAGCC-3′ (nucleotides encoding amino acids 320 to 327).
Five microliters of the PCR product was electrophoresed in a 1% agarose gel. The resulting gel was stained with ethidium bromide until clear bands were visible. Negative controls for the PCR reaction included tubes lacking either template or avian myeloblastosis virus reverse transcriptase. eNOS and β-actin mRNA were quantified by densitometric analysis (IMAGE 1.55, NIH).
Determination of eNOS Immunoreactivity Levels by Western Blot Analysis
Membranes were prepared from whole kidneys. Briefly, rats were decapitated, and their kidneys were removed, minced, and resuspended in 3 mL of 10 mmol/L sodium phosphate buffer, pH 7.4, containing 1 mmol/L NaCl, 0.02% sodium azide, 20 mg/L bestatin, 20 mg/L leupeptin, and 10 μg/L DNAse. Tissues were then homogenized for 30 seconds with polytron homogenizer (Brinkmann Instruments) at a setting of 7. The homogenate was centrifuged at 100 000g for 30 minutes. The supernatant was removed, and the precipitate was washed by resuspension in 10 mmol/L Tris, pH 7.4. Both the supernatants and pellets were stored at −70°C until analysis by methods previously described.30 Purified eNOS was used as a positive marker (Transduction Laboratory).
Inulin concentrations in plasma and urine samples were determined by the anthrone method, and GFR was equated with the clearance of inulin. Sodium concentrations in plasma and urine were measured by flame photometry. Urinary cGMP concentrations were determined by radioimmunoassay with a commercially available kit (New England Nuclear).
Urine samples were assayed for NO2+NO3 with the Griess reagent,31 and values reflect the sum of NO2+NO3 in the assayed samples. The daily excretion of NO2+NO3 was calculated by multiplying the total NO2+NO3 in each sample by the 24-hour urine volume.
One-way ANOVA for repeated or nonrepeated measures, as appropriate, followed by the Dunnett test was used for comparison of data from treatment periods with baseline values. For comparison between the different experimental groups, a two-way ANOVA for repeated measures was used. A value of P<.05 was considered statistically significant. Data are presented as mean±SEM.
In Vivo Studies
Response to Acetylcholine Administration
Fig 1⇓ summarizes the systemic and renal hemodynamic data obtained in response to acetylcholine administration. Baseline MAP was significantly lower in CHF rats than in controls (88±7 versus121±5 mm Hg, P<.05). However, the decline in MAP in response to increasing doses of acetylcholine was of similar magnitude (≈−33%) in both groups (from 121±5 to 80±7 mm Hg in controls versus from 88±7 to 57±5 mm Hg in rats with CHF).
Baseline RBF was also lower in CHF rats (4.3±0.6 mL · min−1 · g kidney−1) than in control animals (8.8±1.3 mL · min−1 · g kidney−1, P<.05.). In response to acetylcholine, RBF increased significantly with the first dose of the drug in control animals (to 10.9±1.3 mL · min−1 · g kidney−1) and remained above the baseline value despite the decrease in MAP. This increase in RBF was reflected by a significant decrease (−40%) in RVR during acetylcholine infusion (from baseline value of 14.7±1.7 to 8.8±0.6 RU during the highest dose, P<.01). The changes in MAP, RBF, and RVR induced by acetylcholine were reversible on cessation of drug infusion (Fig 1⇑, recovery phase). In contrast, the renal vasodilatory response to incremental doses of acetylcholine was markedly attenuated in rats with AV fistula, with RBF values during acetylcholine infusions not being significantly different from the baseline value (Fig 1⇑). Likewise, the maximal decrease in RVR (from 22.6±3.0 to 18.9±2.9 RU) in CHF rats in response to acetylcholine, which was obtained during infusion of the 10-μg · kg−1 · min−1 dose, was substantially impaired compared with control animals.
Table 1⇓ summarizes the renal excretory and clearance data obtained during baseline and acetylcholine infusions. Baseline GFR values were lower in rats with AV fistula than in control animals (P<.05). Administration of acetylcholine resulted in a further decrease in GFR both in control and in CHF rats, probably related to the associated decrease in renal perfusion pressure. In control animals, acetylcholine caused a significant increase in urinary sodium excretion. The maximal natriuretic response was observed during the low dose of acetylcholine (1.0 μg · kg−1 · min−1), with no further increase at the higher doses, apparently because of the hypotensive effect of acetylcholine at higher concentrations. However, in rats with AV fistula, baseline fractional sodium excretion was lower than in controls (0.4±0.2% versus 0.9±0.3%), and the natriuretic response to acetylcholine administration was markedly blunted. In control animals, acetylcholine induced a dose-related increase in urinary cGMP excretion (normalized to GFR), which was significant at the highest acetylcholine dose. A similar pattern was observed for urinary NO2+NO3 excretion (Table 1⇓). In rats with AV fistula, however, baseline urinary cGMP and NO2+NO3 excretion were significantly higher than in control animals. Moreover, in CHF rats, acetylcholine at the highest dose increased the excretion of NO2+NO3 to levels comparable to that of control rats, but in contrast to controls, it failed to cause a further increase in urinary cGMP excretion (Table 1⇓).
Hemodynamic Effects of SNAP Infusion
Table 2⇓ summarizes the effects of the endothelium-independent NO donor on MAP, RBF, and RVR. SNAP resulted in a dose-dependent decrease in MAP both in controls (124±5 to 98±6 mm Hg, P<.05) and in rats with AV fistula (101±5 to 77±5 mm Hg, P<.05). Thus, the decrease in MAP in response to SNAP was of similar magnitude (21% to 24%) in both groups. A different pattern of response, however, was observed for RBF and RVR during infusion of SNAP (Table 2⇓). Baseline RBF was significantly higher in control rats and did not change further in response to SNAP despite the reduction in MAP. RVR was significantly lower (≈19%) in control rats after SNAP (from 17.4±1.6 to 14.1±1.3 RU, P<.05). In contrast, the hypotensive effect of SNAP in rats with AV fistula was associated with a significant fall in RBF (3.9±0.3 to 3.2±0.5 mL · min−1 · g kidney weight−1, P<.05) without any change in the calculated RVR (27.2±2.0 to 26±3.2 RU, P=NS). These findings indicate that the endothelium-independent renal vasodilation, such as produced by exogenous infusion of NO donor, is also attenuated in rats with CHF.
Urinary NO2+NO3 Excretion Rate in Control and AV Fistula Rats
Urinary excretion of NO2+NO3 is shown in Fig 2⇓. Interestingly, despite blunted renal vasodilatory responses to endothelium-dependent and -independent maneuvers, urinary excretion of NO2+NO3 before and after the placement of an AV fistula was comparable in both control animals and rats with AV fistula. When animals were treated with l-arginine, urinary NO2+NO3 increased significantly and to the same extent in both groups.
Effects of l-Arginine on Baseline and Acetylcholine-Mediated Renal Vasodilation
Infusion of l-arginine for 60 minutes failed to produce any significant change in renal perfusion in rats with CHF (data not shown). This suggests that the decrease in baseline RBF in rats with CHF is not related to a decrease in the availability of l-arginine. Fig 3⇓ summarizes the response to acetylcholine in an additional group of rats with AV fistula continuously infused with l-arginine throughout the experiment. For comparison, data from Fig 1⇑ depicting the effects of acetylcholine in control and AV fistula rats (not given l-arginine) are included in this figure. The response to acetylcholine in the group of rats with AV fistula continuously infused with l-arginine did not differ significantly from the response observed in the absence of l-arginine (Fig 3⇓). Thus, the infusion of l-arginine did not significantly alter baseline RBF in rats with CHF and did not affect the blunted renal vasodilatory response to acetylcholine in this group.
Effects of Ang II Receptor Blockade on Renal Response to Acetylcholine
The effects of Ang II receptor blockade on renal and systemic responses to acetylcholine are summarized in Figs 4⇓ and 5⇓ and Table 3⇓. In normal animals treated with A-81988, acetylcholine infusion produced a response similar to that observed in normal animals infused with acetylcholine alone (Figs 4⇓ and 5⇓), consistent with a minimal degree of activation of the renin-angiotensin system in normal animals. As expected, baseline values of MAP, RBF, and RVR in rats with AV fistula, before commencement of any treatment, were substantially different from baseline values in control animals. Administration of A-81988 alone to rats with CHF caused a significant increase in RBF, from 4.2±0.4 to 5.3±0.6 mL · min−1 · g kidney−1 and decreased RVR from 26.1±1.0 to 18.0±1.6 RU These effects were sustained throughout the experiment. However, in the group of CHF rats treated with the Ang II antagonist and acetylcholine, administration of A-81988 resulted in a prompt increase in RBF and completely restored the renal vasodilatory response to acetylcholine infusion. Thus, RBF increased from a basal value of 4.8±0.6 to 6.5±0.7 mL · min−1 · g kidney−1 during A-81988 infusion and further increased to 8.3±1.1 mL · min−1 · g kidney−1 during 10 μg/kg acetylcholine. These changes were also reflected in the progressive reduction in RVR in CHF rats with Ang II receptor blockade (baseline, 23.5±3.1 RU; A-81988 alone, 15.3±1.7 RU; and A-81988+10 μg · kg−1 · min−1 acetylcholine, 9.8±0.8 RU), the latter value being not statistically different from that obtained in control animals infused with the same dose of acetylcholine (protocol 1). The restoration of the vasodilatory effect of acetylcholine in rats with CHF occurred despite a marked decrease in MAP (from 101±6 to 57±4 mm Hg). As shown in Fig 5⇓, the maximal decrease in RVR, in response to acetylcholine, was significantly attenuated in CHF rats compared with controls (−12±9% versus −37±8%, respectively, P<.05). Treatment with both A-81988 and acetylcholine in CHF rats produced a maximal decrease in RVR, which was not different from that observed in control rats treated with the same combination (CHF, −55.3±5.2%; controls, −45.8±5.6%; P=NS). Nevertheless, despite the complete restoration of the renal vasodilatory effect of acetylcholine in CHF rats pretreated with the Ang II receptor blocker, the natriuretic response to acetylcholine infusion remained blunted and was not significantly altered by Ang II receptor blockade (Table 3⇓). This finding could be related to the marked decrease in GFR, which was more pronounced in CHF rats treated with the Ang II antagonist, and the resultant fall in the filtered load of sodium.
In Vitro Studies
Screening of mRNA levels for eNOS, the major isoform of NOS in the renal tissue, was done by RT-PCR. PCR amplification detected a single band of the expected length for eNOS. The amounts of this product, compared with the β-actin quantity control, were not significantly different in kidneys of control rats and rats with CHF (Fig 6⇓).
Western Blot Analysis
A major band of 140 kD comigrating with pure eNOS was detected. The intensity of this band in both the cytosolic (soluble) and precipitate (particulate) eNOS of the whole renal tissue was similar in sham-operated controls and in rats with CHF (Fig 7⇓). Studies using specific antibodies to bNOS and inducible NOS did not reveal the presence of these isoforms of nitric oxide synthase in the renal tissues of either group of rats (data not shown).
This study provides novel information regarding the mechanism of the abnormal regulation of RBF in rats with chronic AV fistula, an experimental model of high-output CHF. First, our data demonstrate that the NO-mediated renal vasodilatory responses, both the endothelium-dependent and -independent responses, are attenuated in rats with this model of CHF. Second, this attenuation in NO-dependent renal vasodilatation occurred despite an intact baseline activity of the renal NO system, suggesting that the biological activity of NO, rather than its production, is diminished in this model of experimental CHF. Finally, perhaps most importantly, our data suggest that this impairment is related to increased activity of the renin-angiotensin system, because the impaired renal vasodilatory effect of acetylcholine could be completely restored by Ang II receptor blockade. Thus, the findings of our study extend the present knowledge by indicating that, in addition to increased activity of systemic vasoconstrictor systems, an impairment in the local, NO-mediated control of RBF may contribute to the mechanism of decreased renal perfusion in CHF. Moreover, our observations point once again to the critical role of the renin-angiotensin system in the pathophysiological alterations in CHF1,26,27,32 and emphasize the important interaction between this system and the endothelium-derived NO system in regulating RBF in this pathophysiological state.
Impaired endothelium-dependent responses in CHF have been reported by several investigators in various vascular beds but not in the kidney. Kaiser et al10 documented that acetylcholine infusion directly into the femoral artery of dogs with CHF resulted in a blunted vasodilatory response. Similar observations were later reported in other vascular beds, such as the thoracic aorta and peripheral arteries,11,13–15 by in vivo infusions of both acetylcholine and NO synthase blockers, as well as in isolated vessels studied by in vitro methodology. However, data addressing the impaired activity of NO as a cause in the abnormal regulation of RBF in CHF are scant and incomplete. Drexler and coworkers12 studied the changes in regional blood flow in response to NO synthase blockade in another model of CHF. Using the microsphere technique, those authors reported an intact vasoconstrictor response to L-NMMA in the renal vascular bed in rats with coronary ligation, suggesting normal basal activity of the NO system in that model. In the present study, we demonstrated that basal NO production is indeed preserved in rats with CHF. However, the NO-mediated renal vasodilatory response that occurred during acetylcholine or SNAP administration was significantly impaired.
The present study also documents that in addition to the renal vasodilatory response to acetylcholine being blunted, the effects of acetylcholine on urinary sodium and cGMP excretion were also attenuated in rats with CHF. In that respect, it should be noted that in response to acetylcholine administration, urinary cGMP but not NO2+NO3 excretion was also reduced in rats with CHF. Moreover, this blunted cGMP response occurred despite a fourfold increase in the baseline rate of urinary cGMP and NO2+NO3 excretion. Because baseline urinary cGMP excretion is thought to reflect the activity of another important hormone, ANP, the finding of high baseline urinary cGMP levels in rats with AV fistula may not be ascribed solely to the NO system but may also reflect the increased activity of ANP in heart failure.33 Nevertheless, the high basal NO2+NO3 excretion may indicate an increased, or at least preserved, basal activity of the NO system in CHF, as has been suggested by some investigators.14,16,22,23 This notion is further supported by the findings in the present study of normal daily NO2+NO3 excretion rates, as well as normal expression of eNOS mRNA levels and immunoreactivity in kidneys of rats with CHF. Moreover, the dissociation of the effect of acetylcholine on urinary NO2+NO3 and on cGMP excretion may indicate that the interference with the vasodilatory action of acetylcholine in CHF occurs at a step beyond NO generation. The finding that the renal vasodilatory response to SNAP, an endothelium-independent NO donor, was also attenuated in rats with CHF supports this contention.
A blunted response to endothelium-dependent vasodilators has been generally equated with a decrease in NOS activity and NO generation. The present findings argue against this notion by demonstrating an attenuated response to NO-mediated vasodilators in the face of intact renal NOS activity and preserved NO generation. A similar phenomenon, ie, a lack of response to endothelium-dependent vasodilators such as acetylcholine and bradykinin despite high baseline NOS activity, has previously been described in the renal vascular bed of rats with norepinephrine-induced ischemic acute renal failure.34
Although the mechanisms responsible for the impaired NO-dependent renal vasodilatation have not been fully elucidated, several important conclusions may be drawn from the present findings. First, the attenuated renal response to endothelium-dependent and -independent vasodilation in rats with CHF cannot be explained by a decrease in eNOS activity. Second, the possibility that a decrease in the availability of the NO precursor l-arginine may be responsible for the impaired vasodilatory response to acetylcholine in this model of experimental heart failure is also unlikely. Such a mechanism was postulated in the hypercholesterolemic rat, a model also characterized by impaired endothelium-dependent vasorelaxation, which demonstrated marked renal vasodilatation in response to l-arginine infusion.19 In the present study, however, neither baseline RBF nor the acetylcholine-induced renal vasorelaxation was modified significantly by infusion of l-arginine in rats with AV fistula.
Perhaps the most intriguing and important finding of the present study is the demonstration that the impairment in endothelium-dependent renal vasodilation is a reversible phenomenon, which may be corrected on removal of the influence of the renin-angiotensin system by a specific Ang II receptor antagonist. Beyond the important therapeutic implications, this finding extends our understanding of the mechanism of abnormal regulation of RBF in CHF. Thus, the results of the study underscore the important role of the renin-angiotensin system in mediating the diminished renal perfusion, not only by a direct vasoconstrictor action of Ang II on renal vessels but also by impairment of the normal, NO-mediated tonic vasodilatation. In fact, the concept of an antagonistic interrelationship between Ang II and the NO system in the regulation of RBF under normal conditions is well recognized.20,21 The findings in the present study suggest that this antagonistic interaction may play an important role in the abnormal regulation of RBF in CHF. Indeed, it has been shown that the activation of the renin-angiotensin system in rats with CHF due to coronary ligation precedes the defect in endothelium-dependent vasorelaxation, although no direct correlation could be found between the increase in plasma renin activity and endothelial dysfunction.15 Moreover, Mulder et al35 recently reported that chronic treatment with the ACE blocker perindopril restored the vasodilatory response to acetylcholine in isolated segments of femoral and mesenteric vessels of rats with CHF induced by coronary ligation.
The rapid restoration of the renal vasodilatory response to acetylcholine after Ang II receptor blockade observed within 30 minutes after infusion of the antagonist also supports the conclusion that the ability of the endothelium to generate NO remains intact in CHF. The fact that Ang II receptor blockade did not restore the natriuretic effect of acetylcholine to normal may be related to the marked decrease in MAP caused by the Ang II receptor antagonist and the resultant decrease in GFR. However, it may also suggest that other mechanisms, in addition to increased activity of the renin-angiotensin system, may contribute to the avid sodium retention in the rats with CHF.
A similar antagonistic interrelationship has been previously reported between the renin-angiotensin system and another vasodilatory/natriuretic hormone, ANP. Our group26,27 and others32 have demonstrated that the blunted renal natriuretic response to ANP in animals with experimental CHF may be improved by treatment with ACE inhibitors. It is noteworthy that the biological activities of both the NO system and ANP are mediated through activation of cGMP as a second messenger. It is thus conceivable that this nucleotide, although produced by different enzymes, may serve as the common site of interaction among these systems. Indeed, Smith and Lincoln36 demonstrated that Ang II, as well as other calcium-mobilizing agents, inhibited the ANP-mediated accumulation of cGMP in cultured vascular smooth cells, apparently by activating a calcium-dependent cGMP phosphodiesterase. Whether such a mechanism may also explain the blunted renal actions of acetylcholine in CHF awaits further clarification.
In summary, this study demonstrates that the integrity of the renal NO system is preserved in rats with chronic AV fistula despite a marked attenuation in the NO-mediated renal vasodilatory responses. Furthermore, this impairment appears to be related to increased activity of the renin-angiotensin system. It is suggested that, in addition to other neurohumoral vasoconstrictor factors, this overriding of the biological action of NO may contribute significantly to the decrease in renal perfusion in this model of high-output CHF.
Selected Abbreviations and Acronyms
|Ang II||=||angiotensin II|
|ANP||=||atrial natriuretic peptide|
|CHF||=||congestive heart failure|
|eNOS||=||endothelial NO synthase|
|GFR||=||glomerular filtration rate|
|MAP||=||mean arterial pressure|
|NOS||=||nitric oxide synthase|
|PCR||=||polymerase chain reaction|
|RBF||=||renal blood flow|
|RVR||=||renal vascular resistance|
This work was supported by grant 2056 from the Israel Ministry of Health to Dr Winaver and by grants from the American Heart Association and Abbott Laboratories to Dr Haramati. The authors acknowledge the expert technical assistance by E. Shuranyi, A. Kaballa, and H. Berger.
- Received February 25, 1997.
- Revision received July 23, 1997.
- Accepted August 6, 1997.
- Copyright © 1997 by American Heart Association
Baylis C, Harton P, Engels K. Endothelial derived relaxing factor controls renal hemodynamics in the normal rat kidney. J Am Soc Nephrol. 1990;1:875–881.
Wilcox CS, Welch WJ, Murad F, Gross SS, Taylor G, Levi R, Schmidt HHW. Nitric oxide synthase in macula densa regulates glomerular capillary pressure. Proc Natl Acad Sci U S A. 1992;89:11993–11997.
Mckee M, Scavone C, Nathanson JA. Nitric oxide, cGMP, and hormone regulation of active sodium transport. Proc Natl Acad Sci U S A. 1994;91:12056–12060.
Stoos BA, Carretero OA, Farhy RD, Scicili G, Garvin JL. Endothelium-derived relaxing factor inhibits transport and increases cGMP content in cultured mouse cortical collecting duct cells. J Clin Invest. 1992;89:761–765.
Schultz PJ, Schorer AE, Raij L. Effects of endothelium-derived relaxing factor and nitric oxide on rat mesangial cells. Am J Physiol. 1990;258:F162–F167.
Kaiser L, Spickard RC, Olivier NB. Heart failure depresses endothelium-dependent responses in canine femoral artery. Am J Physiol. 1989;256:H962–H967.
Kubo SH, Rector TS, Bank AJ, Williams RE, Heifetz SM. Endothelium-dependent vasodilation is attenuated in patients with heart failure. Circulation. 1991;84:1589–1596.
Drexler H, Hablawetz E, Lu W, Riede U, Christes A. Effects of inhibition of nitric oxide formation on regional blood flow in experimental myocardial infarction. Circulation. 1992;86:255–262.
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.
Arnal J, Schott C, Stoclet J, Michel J. Vascular relaxation and cyclic guanosine monophosphate in a rat model of high output heart failure. Circ Res. 1993;27:1651–1656.
Teerlink JR, Gray GA, Clozel M, Clozel J. Increased vascular responsiveness to norepinephrine in rats with heart failure is endothelium dependent: dissociation of basal and stimulated nitric oxide release. Circulation. 1994;89:393–401.
Drexler H, Holtz J. Endothelium dependent relaxation in chronic heart failure. Cardiovasc Res. 1994;28:720–721.
Bank N, Aynedjian HS. Role of thromboxane in impaired renal vasodilatation response to acetylcholine in hypercholesterolemic rats. J Clin Invest. 1992;89:1636–1642.
Ohishi K, Carmines PK, Inscho EW, Navar LG. EDRF-angiotensin II interactions in rat juxtamedullary afferent and efferent arterioles. Am J Physiol. 1992;263:F900–F906.
Ito S, Johnson CS, Carretero OA. Modulation of angiotensin II-induced vasoconstriction by endothelium-derived relaxing factor in the isolated microperfused rabbit afferent arteriole. J Clin Invest. 1991;87:1656–1663.
Winaver J, Hoffman A, Burnett JC Jr, Haramati A. Hormonal determinants of sodium excretion in rats with experimental high-output heart failure. Am J Physiol. 1988;254:R776–R784.
Abassi Z, Haramati A, Hoffman A, Burnett JC Jr, Winaver J. Effect of converting-enzyme inhibition on renal response to ANF in rats with experimental heart failure. Am J Physiol. 1990;259:R84–R89.
Lee JY, Buckner SA, Brune ME, Warner RB, Winn M, De B, Zydovsky TM, Opgenorth TJ, Kerman DJ, Debernardis JF. Characterization of antihypertensive activity of ABBOTT-81988, a nonpeptide angiotensin II antagonist in the renal hypertensive rat. J Pharmacol Exp Ther. 1994;268:427–433.
Abassi ZA, Kotob S, Golomb E, Pieruzzi F, Keiser HR. Pulmonary and renal neutral endopeptidase ec 220.127.116.11 in rats with experimental heart failure. Hypertension. 1995;25:1178–1184.
Abassi Z, Grushka E, Haramati A, Hoffman A, Burnett JC Jr, Winaver J. Atrial natriuretic peptide and renal cGMP in rats with experimental heart failure. Am J Physiol. 1991;261:R858–R864.
Conger J, Robinette J, Villar A, Raij L, Shultz P. Increased nitric oxide synthase activity despite lack of response to endothelium-dependent vasodilators in post-ischemic acute renal failure in rats. J Clin Invest. 1995;96:631–638.
Mulder P, Elefertak L, Mace B, Thuillez C. Peripheral artery structure and endothelial function in heart failure: effect on ACE inhibition. Am J Physiol. 1996;271:H469–H477.
Smith JB, Lincoln TM. Angiotensin decreases cyclic GMP accumulation produced by atrial natriuretic factor. Am J Physiol. 1987;253:C147–C150.