The role of angiotensin II (Ang II) in the short- and long-term control of renal function and arterial pressure is well established (see (1-3) for review). It has been reported that the renal Ang II effects are mediated by activation of AT1 receptors (4,5) and modulated by endogenous levels of nitric oxide (NO) and prostaglandins (PG) (6-11). It is well known that acute Ang II effects on renal function are potentiated when cyclooxygenase (COX) activity is reduced (8,9). It also has been reported that the long-term increment of Ang II levels induces an increase in the renal production of vasodilator and vasoconstrictor COX-derived metabolites (12,13), and that thromboxane is required for full expression of Ang II hypertension in rats (14). However, it is unknown whether the prolonged COX inhibition modifies the renal changes induced by a long-term increment in Ang II levels.
Considering that long-term infusion of Ang II induces an increase in vasodilator and vasoconstrictors prostanoids (12,13), it is difficult to predict whether the Ang II-induced changes in renal function will be enhanced or reduced by COX inhibition. The aim of this study was to examine the renal functional and hemodynamic changes elicited by the prolonged COX inhibition, when Ang II levels are elevated for several days. To accomplish our objective, meclofenamate was infused during 4 consecutive days in conscious dogs, in which Ang II levels were elevated with the prolonged infusion of either an initially subpressor dose or a pressor dose of this peptide. The results obtained may have clinical implications that could contribute to understanding the renal functional and hemodynamic changes induced by a COX inhibitor in situations in which there is an activation of the renin-angiotensin system. Initially nonpressor and pressor Ang II doses have been administered because the mechanisms involved in the hypertension induced by such Ang II doses seem to be different (3). When Ang II is administered at initially subpressor doses, the increase of arterial pressure is gradual and modest. However, higher doses of Ang II induce an acute increase of arterial pressure. There are also important differences between the effects of initially subpressor and pressor doses of Ang II on sympathetic activity, renal excretory ability, vascular structure, and secretion of several vasoactive hormones (3).
MATERIALS AND METHODS
Experiments were performed, as previously described (15,16), in 31 female conscious permanently instrumented dogs (16-22 kg). Surgery was performed under aseptic conditions, and experiments were designed according to the Guiding Principles in the Care and Use of Laboratory Animals, approved by the Council of the American Physiological Society. The dogs were surgically instrumented under anesthesia induced with pentobarbital (30 mg/kg) and maintained with a 1.5-2% halothane/O2 mixture. Tygon catheters were inserted through the femoral vessels into the abdominal aorta, distal to the origins of the renal arteries, and into the inferior vena cava. An arterial catheter was used for arterial pressure monitoring and blood-sample collection. A venous catheter was used for infusion of various solutions. A transit-time flow probe (4R; Transonic Systems, Ithaca, NY, U.S.A.) was implanted on the left renal artery for the renal blood flow (RBF) measurement. The catheters and cable connected to the probe were tunneled subcutaneously, exteriorized between the scapulae, and placed in neck collars. Dogs were allowed to recover from surgery for ≥2 weeks, before any experiments were performed. It has been observed by our group (15,16) that 2 weeks of recovery is enough time for the dogs to be in steady-state conditions. Antibiotic prophylaxis was initiated before surgery and maintained throughout the experiment.
Seven days before the experiments were started, the dogs were housed in individual metabolic cages and fitted with harnesses that contained arterial pressure transducers, mounted at heart level, and a connector to the flow line. The arterial pressure and flow lines were connected to an analog-to-digital data-collection system (T208; Transonic), and the data obtained was analyzed using a personal computer. During the experiments, the data from the recorder were obtained every minute and averaged over a 20-h period (12:00 p.m. to 8:00 a.m.). It has been reported that the continuous whole-day measurement of blood flow provides the best day-by-day reproducibility of the average daily blood flow, and suggested that this measurement is a sensitive method to detect small changes in blood flow (17).
Dogs were given a low-sodium chow (H/D; Hill Pet Products, Topeka, KS, U.S.A.) and infused intravenously with isotonic saline (0.34 ml/min) to provide a fixed sodium load (≈80 mEq/day) over the study period. Saline was pumped through a disposable filter (0.22 μm, Cathivex; Millipore, Bedford, MA, U.S.A.) to prevent formation of air bubbles and possible contaminants from entering the venous system. Filters were changed frequently throughout the study. The infusion tubing and arterial pressure and flow lines were protected with a flexible vacuum hose that was attached to the harness. The dogs were allowed to move freely in the cage but were unable to turn around completely.
Group 1 (n = 6). Captopril (10 μg/kg/min) and Ang II (1 ng/kg/min) were intravenously infused during 5 consecutive days, after a control period of 3 days. A recovery period of 2 days was allowed after both infusions were finished. Captopril infusion was started 60 min before that of Ang II, to minimize fluctuations in the endogenous generation of this vasoactive peptide. Twenty-four-hour urine samples were measured between 8:30 and 9:00 a.m. each day. Blood samples for the measurement of glomerular filtration rate (GFR), and plasma sodium and potassium concentrations were withdrawn daily, 22 h after the last feeding.
In preliminary experiments (n = 8), it was observed that captopril infusion to control dogs induced a decrease (p < 0.05) in mean arterial pressure (MAP; 86 ± 4 vs. 100 ± 3 mm Hg during control period) and an increase (p < 0.05) in urinary sodium excretion (UNaV) (92 ± 12 vs. 65 ± 7 mEq/day during control period). Captopril infusion also induced a slight increase (NS) in RBF (268 ± 25 vs. 237 ± 25 ml/min during control period) and urine flow rate (UV; 873 ± 86 vs. 819 ± 130 ml/day during the control period).
Group 2 (n = 6). A protocol similar to that of group 1 was performed, with the difference that a meclofenamate infusion (5 μg/kg/min, i.v.) was started 24 h after those of captopril and Ang II. Meclofenamate was infused during 4 days because the objective of this study was to evaluate the renal hemodynamic and excretory changes induced by prolonged COX inhibition when Ang II levels are already elevated. In previous experiments (15), it was found that this infusion of meclofenamate reduces the urinary excretion of PGE2 and 6-keto-PGF1α, by 70% and 90%, respectively. Meclofenamate also was administered during 4 days because no side effects (diarrhea, vomiting, etc.) were observed in preliminary experiments, during the first 5 days of meclofenamate infusion.
Group 3 (n = 7). The protocol was similar to that performed in group 1, with the difference that the Ang II dose was greater in this group (5 ng/kg/min).
Group 4 (n = 6). A protocol similar to that of group 2 was performed, with the difference that the Ang II dose was greater in this group (5 ng/kg/min).
Group 5 (n = 6). Only isotonic saline was infused throughout the experiment.
Sodium and potassium concentrations in urine and plasma were measured by flame photometry (Instrumentation Laboratories). GFR was determined by the clearance of endogenous creatinine. Creatinine was measured using a photocolorimetric method (Boehringer Mannheim, Germany).
Data are expressed as mean ± SEM. Significance of differences in values of each day, with respect to the control period, was evaluated using a one-way analysis of variance for repeated measurements and the Fisher's test for multiple comparisons. The significant difference between the same experimental day in different groups was calculated with a two-way analysis of variance and the Duncan test. A value of p < 0.05 was considered significant. Statistical comparisons made throughout are with respect to the mean value obtained during the control period in the same group. Only when specified, the statistical comparisons are made between the same experimental day in different groups.
Results are presented in figures as percentage changes from the mean value found during control period. Mean values of MAP, GFR, RBF, and UNaV during the 3-day control period in each group are presented in Table 1. Significant differences, within and between groups, are similar when the statistics are analyzed using raw numbers or percentage changes from the control period.
Figure 1 shows the changes in MAP, GFR, and RBF induced by the lower Ang II dose (1 ng/kg/min). MAP increased by 12 ± 2% (p < 0.05) during the last 3 days of Ang II infusion and returned to control levels thereafter. GFR did not change throughout the experiment. RBF was not altered the first day, but decreased by 25 ± 6% (p < 0.05) the second day of Ang II infusion. RBF remained decreased the last 3 days of Ang II administration and returned to control levels during recovery period (Fig. 1).
Figure 2 (top) shows the UNaV changes in response to the infusion of the lower Ang II dose. UNaV decreased (p < 0.05) the first 2 days of infusing Ang II, and returned to control levels thereafter. It also shows that UNaV increased (p < 0.05) the first day of the recovery period. UV changes were similar to those of UNaV. Plasma sodium and potassium concentrations did not change throughout the experiment.
Figure 1 shows the changes induced by the lower Ang II dose (1 ng/kg/min) with the simultaneous COX inhibition. Contrary to what was observed in dogs, in which COX activity was not reduced (group 1), MAP increased slightly (8 ± 2%; p < 0.05) only on day 3 of Ang II infusion. It also can be seen in Fig. 1 that no significant changes in GFR and RBF were found in this group. No significant differences also were found between the RBF values in this group and group 1, in which Ang II was infused without inhibiting COX activity. However, the increment in renal vascular resistance (RVR) found during the last 3 days of infusing Ang II alone (0.65 ± 0.02 vs. 0.45 ± 0.01 mm Hg/ml/min during the control period) was greater (p < 0.05) than that found when the same Ang II dose was simultaneously infused with the COX inhibitor (0.54 ± 0.01 vs. 0.44 ± 0.01 mm Hg/ml/min, during the control period).
UNaV decreased (p < 0.05) by 22 ± 6% the first day of Ang II infusion, and by 79 ± 6% mEq/day the first day that COX was inhibited in the Ang II-infused dogs (Fig. 2, top). During the following 3 days that Ang II and meclofenamate were simultaneously administered, UNaV remained decreased (p < 0.05), not only with respect to the control period, but also in relation to the values found during the first day of Ang II infusion. UNaV was lower (p < 0.05) in this group than in group 1, in which the same Ang II dose was administered but without reducing PG synthesis. The continuous decrease of UNaV in this group led to a positive cumulative sodium balance of 189 ± 40 mEq, greater (p < 0.05) than that found in group 1 (26 ± 24 mEq). UNaV was not significantly modified during the recovery period in group 2. Changes in UV were similar to those observed in UNaV. Plasma sodium concentration did not change throughout the experiment, but plasma potassium increased during meclofenamate infusion (4.5 ± 0.3 mEq/L) and the first day of the recovery period (4.5 ± 0.3 mEq/L), with respect to the level found in the control period (3.9 ± 0.2 mEq/L).
The effects of the higher Ang II dose (5 ng/kg/min) are shown in Figs. 2 (bottom) and 3. MAP increased progressively during Ang II infusion (Fig. 3). The increments were 12 ± 2% (p < 0.05) the first day, and 25 ± 4% (p < 0.05) the last day of Ang II infusion. During the recovery period, MAP returned to control levels. Ang II induced a slight decrease in GFR that was significant only during the second day. RBF decreased by 24 ± 5% (p < 0.05) on the first day, and remained diminished until the end of the Ang II infusion. During the recovery period, RBF increased to levels not different from those found during the control period.
UNaV decreased (p < 0.05) the first 2 days of infusing the higher Ang II dose and returned to control levels thereafter (Fig. 2, bottom). UNaV increased (p < 0.05) on the first day of the recovery period. Changes in UV were similar to those of UNaV. Plasma sodium and potassium levels did not change throughout the experiment.
Figure 3 shows that MAP increased, during the simultaneous administration of Ang II and meclofenamate, to levels similar to those found in group 3, in which only the same Ang II dose (5 ng/kg/min) was infused. It also can be observed in Fig. 3 that the renal vasoconstriction induced by the prolonged infusion of this Ang II dose was not modified by the simultaneous COX inhibition.
UNaV decreased (p < 0.05) the first day of Ang II infusion and during both day 1 and 2 of simultaneous Ang II and meclofenamate infusion (Fig. 2, bottom). This reduction in renal excretory ability led to a positive cumulative sodium balance (116 ± 40 mEq), not significantly different from that found in group 3 (45 ± 36 mEq), where the same Ang II dose was infused alone. The response of UV in this group was similar to that of UNaV. Plasma sodium and potassium concentrations did not change throughout the experiment.
As expected, MAP, GFR, and RBF did not change when only isotonic saline was infused. No significant changes were observed throughout the experiment in UNaV, UV, and plasma sodium and potassium concentrations.
This study provides new information about the renal functional and hemodynamic changes induced by prolonged administration of a COX inhibitor when Ang II levels are elevated for several consecutive days. Many studies have proposed that COX-dependent metabolites modulate or potentiate the renal Ang II effects (1-3,12,14). However, it has not been evaluated whether COX inhibition modifies the renal functional and hemodynamic changes induced by long-term exogenous infusion of different Ang II doses. Because the increment in Ang II elicits an elevation in vasodilator and vasoconstrictor COX metabolites (12,13), it was difficult to predict whether COX inhibition would enhance or reduce the renal changes induced by Ang II.
In this study, Ang II levels were elevated by infusing either an initially subpressor or a pressor dose, because there are differences in the mechanisms involved in the hypertension induced by Ang II at such doses (3). The infusion of captopril (given simultaneous with Ang II) intended to minimize fluctuations in the endogenous generation of Ang II. The exogenous infusion of this peptide restored plasma Ang II to levels that could not be further modified by any suppressive effect on renin release elicited by COX inhibition or changes in sodium balance. It is well known that Ang II increases sodium reabsorption and that both sodium retention and COX inhibition reduce renin release (1,2). To differentiate between the renal effects of initially subpressor or pressor Ang II doses may be important because (a) long-term administration of initially subpressor Ang II doses mimics the development of human hypertension to a greater extent than does infusion of pressor doses (1,3); and (b) prolonged infusion of initially subpressor Ang II doses develop a progressive hypertension similar to that observed in two-kidney, one-clip hypertension (1,3).
The role of endogenous PG in regulating renal function has been extensively evaluated by the administration of COX inhibitors (see 1,2 for review). In a previous study (15), it was found that prolonged COX inhibition induced an increase in RVR, a transitory decrease in sodium and water excretion, and no changes in MAP. The renal functional and hemodynamic effects induced by prolonged increments in Ang II also have been extensively examined (see 1-3 for review), and the results obtained in this work are similar to those reported. From these results, it can be suggested that (a) renal function is more sensitive than arterial pressure to long-term increments in Ang II; and (b) the Ang II-induced changes in sodium reabsorption can be secondary to the renal hemodynamic and tubular effects elicited by this peptide.
This study presents new evidence supporting the concept that COX inhibition does not enhance the renal vasoconstriction induced by prolonged increments in Ang II levels. It was found that the hypertension and renal vasoconstriction induced by a pressor Ang II dose were similar in dogs with and without simultaneous COX inhibition. It also was observed that the hypertension and renal vasoconstriction induced by a lower dose of Ang II (that is initially subpressor) were significantly reduced by the meclofenamate administration. These results were partly unexpected because many studies have proposed that vasodilator PG modulates the Ang II-induced constriction of the systemic and renal vasculatures (1-3,8,9,12). However, based on the reported increments in the urinary excretion of thromboxane B2 (TxB2) induced by long-term Ang II infusion (13,14), it could be proposed that the renal vasoconstriction observed in our study during administration of the lower dose of Ang II is mediated by a pressor prostanoid. This idea also is supported by studies (18) showing that COX-derived vasoconstrictor metabolites are of particular importance in the two-kidney, one-clip model of renovascular hypertension. There is not an easy explanation for the results obtained during the simultaneous infusion of the higher Ang II dose and the COX inhibitor. Based on the reported increments of PGI2 and TXA2, in response to pressor Ang II doses (12,14), it can be speculated that the Ang II-induced vasoconstriction was not modified because the COX inhibition prevented the increases in the endogenous production of these metabolites, with opposite vasoactive effects.
In a previous study (15), the prolonged COX inhibition induced only a small effect on sodium excretion. However, meclofenamate significantly potentiated the renal antinatriuretic effect induced by the lower Ang II dose infusion. This change in sodium excretion may be secondary to a direct tubular effect. In support of a direct tubular effect, it has been demonstrated that Ang II and PG play an important role in regulating tubular sodium and water reabsorption (1,8,11). It is expected that the sodium retention would increase MAP if COX inhibition is maintained for a longer period, simultaneous with the lower Ang II dose infusion.
The most frequent explanation given for the hyperkalemia elicited by the prolonged administration of a COX inhibitor is a suppression of renin and aldosterone secretion (1,19). This notion is supported by the results obtained in this study, during infusion of meclofenamate and the higher dose of Ang II, and those showing that our dose of meclofenamate decreases plasma renin activity (15). However, the decrease in Ang II levels and possible aldosterone suppression does not seem to be the only explanation of the hyperkalemia elicited by the administration of COX inhibitors. In our study, it was found that meclofenamate also induces an increase in plasma potassium in dogs with a continuous infusion of Ang II, at a dose that is initially subpressor but high enough to induce a renal vasoconstriction. It only can be speculated that the greater plasma levels of Ang II and aldosterone, during the infusion of the higher Ang II dose, are probably the reason that meclofenamate induced an increase of plasma potassium only in dogs infused with the lower Ang II dose. Another mechanism proposed to explain the hyperkalemia induced by prolonged COX inhibition is a decrease in distal sodium delivery or an activation of high-conductance K+ channels, described in the collecting tubules (19).
In summary, the results of this study show that the hypertension and renal vasoconstriction induced by the prolonged administration of a low dose of Ang II that is initially subpressor are significantly reduced by the simultaneous infusion of a COX inhibitor. The results obtained in this study demonstrate, for the first time, that the prolonged COX inhibition does not potentiate the renal vasoconstriction elicited by the long-term infusion of a pressor Ang II dose. This study also presents new evidence that the administration of a COX inhibitor may reduce the renal excretory ability during several consecutive days, when Ang II levels are chronically elevated.
Acknowledgment: This study was supported by grants from the Fondo de Investigaciones Sanitarias (FIS, 98/1309 and 99/1024) of Spain. Juan D. González was supported by a grant from Fundación Mapfre (Spain). Maria T. Llinás was supported by a postdoctoral grant from University of Murcia (Spain). Carol Moreno was supported by a grant from the Spanish Ministerio de Educación y Ciencia (ref. PN-95). Francisca Rodríguez was supported by a grant from the FIS (98/1309).
Parke-Davis Labs kindly provided meclofenamate.
1. Navar LG, Inscho EW, Majid DSA, Imig JD, Harrison-Bernard LM, Mitchell KD. Paracrine regulation of the renal microcirculation. Physiol Rev
2. Romero JC, Knox FG. Mechanisms underlying pressure-related natriuresis: the role of the renin-angiotensin and prostaglandins
systems: state of the art lecture. Hypertension
3. Simon G, Abraham G, Cserep G. Pressor and subpressor angiotensin II
administration: two experimental models of hypertension
. Am J Hypertens
4. Van der Mark J, Kline RL. Altered pressure natriuresis in chronic angiotensin II hypertension
in rats. Am J Physiol
5. Wang C-T, Zou L-X, Navar LG. Renal responses to AT1
blockade in angiotensin II
-induced hypertensive rats. J Am Soc Nephrol
6. Llinás MT, González JD, Salazar FJ. Interactions between angiotensin and nitric oxide in the renal response to volume expansion. Am J Physiol
7. Chin SY, Wang CT, Majid DSA, Navar LG. Renoprotective effects of nitric oxide in angiotensin II
in the rat. Am J Physiol
8. Llinás MT, González JD, Nava E, Salazar FJ. Role of angiotensin II
in the renal effects induced by nitric oxide and prostaglandin synthesis inhibition. J Am Soc Nephrol
9. Pinilla JM, Alberola A, González JD, Quesada T, Salazar FJ. Role of prostaglandins
on the renal effects of angiotensin and interstitial pressure during volume expansion. Am J Physiol
10. Hennington BS, Zhang H, Miller MT, Granger JP, Reckelhoff JF. Angiotensin II
stimulates synthesis of endothelial nitric oxide synthase. Hypertension
11. Salazar FJ, Linás MT, González JD, Pinilla JM. Role of prostaglandins
and nitric oxide in mediating renal response to volume expansion. Am J Physiol
12. Luft FC, Wilcox CS, Unger T, et al. Angiotensin-induced hypertension
in the rat: sympathetic nerve activity and prostaglandins
13. Wilcox CS, Welch WJ, Snellen H. Thromboxane mediates renal hemodynamic response to infused angiotensin II
. Kidney Int
14. Keen HL, Brands MW, Smith Jr. MJ, Shek EW, Hall JE. Thromboxane is required for full expression of angiotensin hypertension
in rats. Hypertension
15. González JD, Llinás MT, Nava E, Ghiadoni L, Salazar FJ. Role of nitric oxide and prostaglandins
in the long-term control of renal function. Hypertension
16. Salazar FJ, Alberola A, Pinilla JM, Romero JC, Quesada T. Salt-induced increase in arterial pressure during nitric oxide synthesis inhibition. Hypertension
17. Montani JP, Mizelle HL, Van Vliet BN, Adair TH. Advantages of continuous measurement of cardiac output 24 h a day. Am J Physiol
18. Hilmmelstein SI, Klotman PE. The role of thromboxane in two-kidney, one-clip Goldblatt hypertension
in rats. Am J Physiol
19. Schlondorff D. Renal complications of nonsteroidal anti-inflammatory drugs. Kidney Int