P2 Purinoceptor Saturation by Adenosine Triphosphate Impairs Renal Autoregulation in Dogs

A growing body of evidence indicates that extracellular ATP exerts paracrine influences and has substantial effects on the vascular function of various tissues and organs, including the kidney (1,2,3). The vascular effects of extracellular ATP seem to be mediated via stimulation of specific P2 purinoceptors (3,4,5). These receptors have been subclassified as P2X and P2Y purinoceptors (3, 6,7,8,9,10). Renal vasodilation evoked by P2Y purinoceptor activation involves the release of nitric oxide (NO) from the vascular endothelium (5, 10, 11). The direct renal vasoconstrictor action of ATP is mediated predominantly by activation of P2X receptors on vascular smooth muscle cells and has been demonstrated on preglomerular microvascular elements (3, 12). Interestingly, postglomerular vessel diameters are not altered during exposure to ATP (3, 12).

Recent microvascular studies have suggested that P2 purinoceptors may play a role in renal autoregulatory responses (13, 14). Using the in vitro blood-perfused rat juxtamedullary microvascular preparation, it was shown that inhibition of P2 purinoceptor-dependent responses to ATP administration impaired normal pressure-mediated autoregulatory adjustments in afferent arteriolar diameter (14). In that study, receptor desensitization, saturation, and blockade were used to interfere with P2 purinoceptor activation and were shown to impair autoregulatory behavior in renal afferent arterioles. In an in vivo micropuncture study in rats (13), it was also observed that tubuloglomerular feedback (TGF)-mediated decreases in stop-flow pressure, in response to increases in loop perfusion rate, were markedly attenuated during peritubular capillary infusion of high doses (10 mM) of ATP or a nonmetabolizable ATP analog (β,γ-methylene-ATP), indicating that P2 purinoceptor saturation inhibits TGF responses.

On the basis of the findings of previous studies (12,13,14), we hypothesized that renal afferent arteriolar vasoconstrictor responses to increases in arterial pressure are mediated by the local release of ATP, which acts on P2 receptors located on the vascular smooth muscle cells. Because the macula densa could be a potential source of ATP (2, 15), it is possible that acute increases in arterial pressure alter glomerular dynamics and flow to the macula densa, causing increased release of ATP that causes afferent arteriolar vasoconstriction via its action on vascular P2 receptors (3). It has also been reported that ATP is released from endothelial cells in response to increases in shear stress on the vessel walls (16, 17). Such an increase in shear stress on endothelial cells of afferent arterioles would be predicted during increases in arterial pressure (18, 19). Therefore, it seems that locally produced extracellular ATP could be involved in both TGF and myogenic responses that mediate renal autoregulatory mechanisms. However, this issue has not been examined in studies evaluating overall renal blood flow (RBF) autoregulatory efficiency.

This investigation was performed to evaluate the possible role of P2 purinoceptors in mediating or modulating autoregulatory changes in renal vascular resistance (RVR) in response to changes in arterial pressure in anesthetized dogs. The efficiency of RBF autoregulation was examined before and during P2 purinoceptor saturation by intra-arterial administration of high doses of ATP. The rationale for these experiments is that if pressure-induced adjustments in afferent arteriolar diameter are mediated by the local release of ATP, then delivery of excess exogenous ATP should occupy the available P2 receptors and leave none to respond to ATP released in response to the pressure stimulus. The unavailability of P2 receptors to bind locally released ATP would be expected to blunt or abolish pressure-dependent autoregulatory responses. Although maximal saturation of P2 receptors in the renal vasculature can be achieved with infusion of high doses of ATP, this also activates P2 receptors on endothelial cells, causing NO release and vasodilatory responses (11). In pilot experiments, we observed that intrarenal infusion of high doses of ATP in intact dogs caused substantial renal vasodilation, as well as systemic hypotension resulting from peripheral vasodilation caused by the overflow of ATP into the circulation, thus precluding accurate assessment of RBF autoregulatory efficiency. To avoid these NO-mediated responses to intra-arterial ATP infusion, these experiments were performed in dogs that had been pretreated with the NO synthase inhibitor nitro-L-arginine (NLA).

Materials and Methods

Procedures

Experiments were performed with mongrel dogs (approximately 20 kg body wt), which were given supplemental amounts of sodium chloride (1.5 g/kg body weight per d for 3 d) added to the normal laboratory diet so that they achieved a sodium-replete state. The dogs were anesthetized with pentobarbital sodium (30 mg/kg, intravenously), with additional amounts being given as needed to maintain an appropriate level of anesthesia. An endotracheal tube was inserted and connected to an artificial respirator set to provide ventilation with room air at a rate of 18 strokes/min, with a stroke volume of 15 ml/kg body weight. Body temperature was monitored with a rectal temperature probe (Yellow Springs Instrument Co., Yellow Springs, OH), and temperature was maintained within the normal range (99 to 101 °F) using an electric blanket. Systemic arterial pressure was continually monitored with a Statham pressure transducer (P23 DC; Statham, Hatorey, Puerto Rico), via a catheter placed in the right femoral artery, and was recorded on a polygraph (model 7D; Grass Instruments, Quincy, MA). The left femoral artery was cannulated for collection of blood samples. The jugular and femoral veins were cannulated for administration of an inulin solution and additional anesthetic as necessary. Dogs were given a continuous infusion of isotonic saline solution, at a rate of 30 ml/h, via the catheter in the right femoral vein.

The left kidney was exposed via a flank incision and was denervated by cutting all visible renal nerves. The renal artery was isolated from the surrounding tissue, and an electromagnetic flow probe (Carolina Medical Electronics, King, NC) was placed around the renal artery to measure RBF. An adjustable plastic clamp was placed around the renal artery distal to the flow probe, to produce reductions in renal arterial pressure (RAP). A curved, 23-gauge needle cannula was inserted into the renal artery distal to the plastic clamp and was connected to a pressure transducer to measure RAP. Another catheter was also connected to this needle cannula, for continuous infusion of heparinized saline or drug solutions at a rate of 0.4 ml/min. Urine was collected from a catheter placed in the left ureter.

After completion of all surgical procedures, a 2.5% solution of inulin in normal saline was administered into the jugular vein at least 45 min before the start of the experimental protocol. An initial inulin dose of 1.6 ml/kg was followed by a continuous infusion of 0.03 ml/kg per min. At least 45 min before the start of the experimental protocol, the right common carotid artery was occluded and the left common carotid artery was partially constricted to elevate the basal mean arterial pressure to approximately 150 mmHg. This was performed to facilitate evaluation of the arterial pressure-blood flow relationship over a wider range of arterial pressures.

After a 45-min stabilization period, the experimental protocol was started with urine collections for two consecutive 10-min periods, at the spontaneous arterial pressure. An arterial blood sample (2 ml) was collected at the midpoint of each urine collection, for measurement of plasma inulin, sodium, and potassium concentrations. Figure 1 illustrates the experimental protocol used in this study. After control measurements were obtained, the adjustable clamp was used to reduce RAP in steps of approximately 10 mmHg (approximately 2 min at each step) until RBF was reduced to nearly zero. After the last reduction in RAP, the clamp was released to reestablish control RAP and RBF. NLA (Aldrich Chemical Co., Milwaukee, WI) was then continuously infused (intra-arterially, at a rate of 50 μg/kg per min) until the end of the experimental procedure (11, 18, 19). After 30 min of NLA infusion, two consecutive 10-min urine samples were collected. The pressure-flow relationship was then reevaluated by gradual reductions in RAP (NLA/pre-ATP period), as performed during the control period. After the clamp had been released and steady-state RAP and RBF were reestablished, a continuous infusion of ATP (Sigma Chemical Co., St. Louis, MO) was added to the NLA infusion at a rate of 1 mg/kg per min, which was calculated to provide an estimated concentration of 300 μM ATP in plasma. After 10 min of ATP infusion, two consecutive 10-min urine samples were collected. Then the evaluation of the arterial pressure-blood flow relationship by gradual reductions in RAP, using the vascular clamp, was repeated. ATP infusion was then discontinued, and steady-state RAP and RBF were restored by releasing the vascular clamp. The efficiency of RBF autoregulation was reexamined after discontinuation of ATP infusion (NLA/post-ATP period).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Schematic diagram of the experimental protocol, illustrating the control and experimental sequence. NLA, nitro-L-arginine; NE, norepinephrine.

To determine whether the ATP-induced vasoconstrictor effect itself contributed to the renal autoregulatory responses to ATP infusion, norepinephrine (5 μg/kg per min) was added to the NLA infusion line in four of the dogs after cessation of the ATP infusion. The RBF autoregulatory efficiency was then examined during infusion of norepinephrine.

At the end of each experiment, the electromagnetic flow probe was calibrated in situ by collection of timed blood samples into a graduated cylinder, at different flow rates, via a catheter placed in the renal artery. The kidney was removed, stripped of surrounding tissues, blotted dry, and weighed, so that the calculated parameters could be expressed per gram of net kidney mass. Flame photometry (Instrumentation Laboratory, Lexington, MA) was used to determine the sodium and potassium concentrations in plasma and urine. Inulin concentrations in plasma and urine samples were determined by the anthrone colorimetric technique, using a spectrophotometer (Gilford Co., Oberlin, OH).

Results

Effects on RBF Autoregulation

Figure 2A illustrates the effects of ATP infusion on RBF autoregulatory efficiency. As reported earlier (11, 18, 19), the autoregulatory efficiency of RBF remained intact during NLA infusion, although there was a suppression of the plateau of the pressure-flow relationship. The slopes of the pressure-flow relationships within the autoregulatory range of RAP were not significantly different from zero during control and NLA infusion periods (-0.002 ± 0.001 and 0.003 ± 0.001 ml/min per g per mmHg). There was also a rightward shift in the slope of the linear portion of the pressure-flow relationship curve at RAP below the autoregulatory range (0.10 ± 0.008 to 0.05 ± 0.004 ml/min per g per mmHg), as reported previously (11, 18), which indicates that NO primarily influences the autoregulation-independent component of RVR. As reported previously (19), NLA infusion led to increases in RVR at all levels of altered pressure but did not alter the basic pattern (Figure 2B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Effects of intra-arterial ATP infusion on renal blood flow (RBF; Panel A) and renal vascular resistance (RVR; Panel B) responses to changes in renal arterial pressure (RAP) in NLA-treated dogs (n = 9). During NLA infusion alone (NLA/pre-ATP period; •), the autoregulatory efficiency remained intact as in control period (○). During infusion of ATP in these NLA-treated dogs (▪), there was marked impariment of the autoregulatory efficiency, which was restored after discontinuation of ATP infusion (NLA/post-ATP period; ▵).

During ATP administration in these NLA-treated dogs, there was marked impairment of RBF autoregulatory efficiency (Figure 2A). The slope of the pressure-flow relationship curve within the autoregulatory range of RAP was increased to 0.02 ± 0.002 ml/min per g per mmHg (P < 0.001), which was not significantly different from the slope of the linear portion of the curve during ATP infusion (0.03 ± 0.005 ml/min per g per mmHg). The usual decreases in RVR during stepwise decreases in RAP within the autoregulatory range were absent during ATP administration (Figure 2B), demonstrating a loss of autoregulatory capability. After cessation of the ATP infusion, this loss of RBF autoregulatory efficiency was restored to that in the pre-ATP period (Figure 2A). The slopes of the autoregulatory and linear portions of the pressure-flow relationships after discontinuation of ATP (NLA/post-ATP period) were restored to 0.006 ± 0.002 and 0.05 ± 0.004 ml/min per g per mmHg, respectively, which were not significantly different from the respective values before ATP infusion (NLA/pre-ATP period). The RAP-RVR relationship curve also returned to the position observed before ATP infusion (Figure 2B).

In four of the NLA-treated dogs, norepinephrine was administered intra-arterially, at a rate of 5 μg/kg per min, after cessation of ATP infusion. RBF was decreased from 3.3 ± 0.3 ml/min per g to 2.5 ± 0.3 ml/min per g during norepinephrine infusion. This reduction in RBF was similar to the reduction in RBF observed during ATP infusion (to 2.8 ± 0.4 from 3.3 ± 0.3 ml/min per g). RVR was increased to 64.4 ± 8.3 from 47.5 ± 3.7 mmHg/ml per min per g during norepinephrine infusion in these NLA-treated dogs. Although there were similar reductions in RBF during ATP and norepinephrine influsions, there was no loss of RBF autoregulatory efficiency during norepinephrine infusion (Figure 3A). The slope of the pressure-flow relationship within the autoregulatory range of RAP was 0.004 ± 0.003 ml/min per g, which was not significantly different from zero. The slope of the linear portion of the curve during norepinephrine infusion was 0.04 ± 0.002 ml/min per g per mmHg, which was also similar to values measured (0.03 ± 0.001 ml/min per g per mmHg) during the period of NLA infusion alone. The usual relationship between RAP and RVR (i.e., a decrease in RVR in response to a decrease in RAP) remained intact during norepinephrine infusion, although it was shifted upward because of the norepinephrine-induced increases in RVR.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Effects of ATP and norepinephrine infusions on RBF (A) and RVR (B) autoregulation in NLA-treated dogs (n = 4). As shown in Figure 2, ATP infusion in these NLA-treated dogs (▪) caused impairment of autoregulatory efficiency. However, autoregulatory efficiency remained intact during norepinephrine infusion (⋄), although reductions in basal RBF were similar with ATP and norepinephrine infusions. •, treatment with NLA alone.

As mentioned earlier, we avoided experiments with infusions of ATP in intact dogs because of the technical difficulty of assessing RBF autoregulatory efficiency, due to renal vasodilation as well as systemic hypotension caused by ATP-mediated NO release. However, we conducted experiments in intact dogs using a lower dose of ATP (0.33 mg/kg per min; estimated plasma concentration, 100 μM) that minimized the decreases in systemic arterial pressure and produced a reasonably stable preparation for assessment of RBF autoregulatory efficiency. In these experiments (n = 3), we also observed impairment of RBF autoregulatory efficiency, inasmuch as the control slope of the autoregulatory portion of the pressure-flow relationship was significantly altered (0.001 ± 0.002 to 0.02 ± 0.004 ml/min per g per mmHg, P < 0.01) during low-dose infusion of ATP. However, the impairment of autoregulation in intact animals was quantitatively less complete than in dogs treated with the high dose of ATP, suggesting only partial saturation of P2 receptors with the low dose.

Discussion

This study demonstrates that renal P2 purinoceptor saturation by intra-arterial administration of ATP results in marked impairment of RBF autoregulatory efficiency in response to reductions in RAP in anesthetized dogs. If pressure-dependent adjustments in renal afferent arteriolar diameter are mediated by the release of ATP, then providing an excess concentration of exogenous ATP would be expected to blunt the pressure-dependent autoregulatory response. Administration of excess exogenous ATP would minimize the effects of endogenously released ATP by shifting the agonist/receptor competition to favor binding of exogenously administered ATP. The results of this study are consistent with this hypothesis. Marked impairment of the ability of the kidney to autoregulate RBF was clearly evident in this study, at an estimated renal plasma ATP concentration of approximately 300 μM. This concentration of ATP is substantially greater than the concentration needed to cause a maximum vasoconstricting effect on rat juxtamedullary afferent arterioles (12). Therefore, it would be expected that the P2 purinoceptors in the renal vasculature would be saturated with this high dose of ATP, rendering them unresponsive to local fluctuations in ATP levels. In addition, the impaired autoregulatory capability observed in this study is in complete agreement with the compromised autoregulatory behavior observed during P2 receptor saturation in rats (14). The effects of ATP administration on RBF autoregulatory efficiency were reversible, because autoregulatory capability was quickly restored after discontinuation of ATP infusion (Figure 2). Furthermore, the effects of ATP on autoregulatory efficiency cannot be ascribed simply to the ATP-mediated increase in RVR, because norepinephrine caused similar reductions in RBF without impairment of RBF autoregulation (Figure 3). Similar results have been reported for angiotensin II-mediated increases in renal microvascular tone (14). These data demonstrate that impairment of autoregulatory behavior during ATP infusion cannot be attributed to enhancement of renal vascular tone, and they suggest that the impairment of renal autoregulation is a direct result of excess P2 receptor occupancy by infused ATP and consequent blunting of responsiveness to endogenously released ATP.

It is recognized that ATP is metabolized rapidly, so that the actual ATP concentration at the effector site is much less than the calculated value. It is also possible that the observed effects could be attributable, in part, to metabolites. However, it is unlikely that the hydrolyzed products of ATP, such as adenosine, ADP, and AMP, are responsible for the impairment of RBF autoregulatory efficiency seen with ATP. It has been shown that adenosine receptor blockade does not alter renal autoregulatory behavior in dogs (20), and the renal microvasculature is essentially unresponsive to ADP (21) and AMP (22). In a previous micropuncture study in rats (13), it was observed that an adenosine receptor antagonist actually prevented the waning effect of stop-flow pressure responses to peritubular ATP infusion, indicating that adenosine was not responsible for the vasoconstriction. The findings in this in vivo study in dogs are in close agreement with our earlier observations in an in vitro rat preparation (14), that P2 purinoceptor saturation significantly decreased the afferent arteriolar diameter and prevented pressure-mediated autoregulatory adjustments in the afferent arteriolar diameter. As in this study, impairment of autoregulatory efficiency did not occur when a similar degree of vasoconstrictor tone was imposed with a different vasoconstrictor. Marked attenuation of autoregulatory responses in afferent arterioles was also observed during P2 purinoceptor inactivation by receptor desensitization or by treatment with receptor antagonists (14). For this study, it was not possible to use similar receptor antagonists, because of the excessively large amounts that would be required to achieve adequate receptor blockade in dogs.

It is possible that high levels of extracellular ATP prevented renal autoregulatory responses through mechanisms not associated with P2 purinoceptor saturation. However, this possibility seems unlikely, because previous in vitro studies (14) showed that impairment of autoregulatory responses in renal afferent arterioles was observed not only with P2 purinoceptor saturation, but also with receptor desensitization, as well as with receptor blockade. We are not aware of any pharmacologic actions of extracellular ATP other than its receptor-mediated actions. The purpose of studies with norepinephrine (Figure 3) was simply to demonstrate that a similar increase in RVR caused by a structurally different vasoconstrictor would not elicit the loss of autoregulatory responses. In previous in vitro studies (14), it was also shown that P2 purinoceptor desensitization prevented autoregulatory adjustments in afferent arteriolar diameter, whereas responsiveness to other receptor-dependent (norepinephrine) and receptor-independent (potassium chloride) vasoconstrictor stimuli was retained in those vessels. Therefore, it seems most unlikely that ATP exerted nonspecific pharmacologic actions other than P2 purinoceptor activation to influence autoregulatory adjustments in renal preglomerular resistance vessels.

ATP infusion in dogs in this study also caused reductions in GFR. In an earlier study by Harvey (23), using an isolated dog kidney preparation, it was shown that ATP injected into the renal artery produced a decrease in GFR, but the study failed to observe any significant changes in the autoregulatory behavior of renal plasma flow and GFR. However, ATP was administered at a much lower dose (0.1 to 0.5 mg/min) in that in vitro study, compared with the dose used in this in vivo study and, at the time of those experiments, the confounding effect of endothelial NO release produced by ATP infusion was not considered.

The hypothesis that P2 purinoceptors are involved in the mechanism of renal autoregulation stems from earlier observations (2,3,13,14,21,22,24) that the renal microvascular responses to extracellular ATP exhibit a striking resemblance to the responses of afferent arterioles subjected to an acute increase in renal perfusion pressure. In addition, ATP-mediated renal microvascular responses are limited primarily to afferent arterioles and depend on activation of voltage-gated calcium channels (5). Because the TGF mechanism is one major mediator of autoregulatory responses, extracellular ATP might be involved in the transmission of feedback signals from the macula densa to modulate afferent arteriolar resistance. Results from in vivo micropuncture studies in rats (13) demonstrated marked attenuation of TGF-mediated effects on proximal tubular stop-flow pressure during peritubular capillary infusion of ATP. This observation is consistent with a role for ATP as a mediator of TGF responses. Moreover, renal microvascular actions of ATP observed in vitro are also consistent with those required for a feedback mediator (3,12). Feedback-dependent regulation of glomerular capillary pressure occurs predominantly at afferent arterioles. It has been reported that the preglomerular renal vasculature expresses abundant P2X₁ receptors, whereas efferent arterioles seem to be devoid of such receptors (25). Myogenic adjustments in afferent arteriolar vascular resistance during changes in RAP also occur with a very rapid response time, and it is possible that shear stress-related release of ATP from the endothelial cells contributes to the myogenic response. Collectively, the results of the experiments presented here and those of previous studies (13,14) support the hypothesis that extracellular ATP is possibly the effector agent responsible for the afferent arteriolar resistance adjustments in response to alterations in perfusion pressure, through its interaction with P2 purinoceptors on the vascular smooth muscle cells.

The possible mechanisms by which extracellular ATP participates in autoregulatory adjustments in RVR are not yet clearly understood. It is possible that signals originating as a consequence of alterations in RAP initiate a sequence of events that alter the rate of ATP production. The notion that macula densa cells may be a source of ATP during changes in RAP is supported by previous reports showing that these cells have relatively low Na⁺/K⁺-ATPase activity although they are rich in mitochondria (2,3,15). Therefore, they seem to possess a high ATP-generating capacity while exhibiting only a limited need for metabolic energy. Endothelial cells and vascular smooth muscle cells, as well as epithelial cells, can release ATP into the extracellular environment (3,15,16,26). It has been reported that increases in flow-induced shear stress on vessel walls can stimulate ATP release from mesenteric (16) and aortic (17) endothelial cells. The possibility that hemodynamic changes can induce alterations in ATP release in the renal circulation is supported by a recent study by Bohmann et al. (27), who demonstrated that during low-frequency renal nerve stimulation, ATP was present in the renal venous effluent, which was of non-neuronal origin. However, additional studies are needed to establish a relationship between changes in RAP and ATP release by macula densa and/or microvascular cells. In conclusion, the results of this investigation suggest that extracellular ATP, activating P2 purinoceptors on the vascular smooth muscle cells of renal afferent arterioles, may be involved in the mechanism of autoregulatory adjustment in RVR.