点击显示 收起
【关键词】 arteriole
Department of Pharmacology and Therapeutics, Smooth Muscle Research Group, University of Calgary, Calgary, Alberta, Canada
ABSTRACT
We previously demonstrated that stimulation of proteinase-activated receptor-2 (PAR-2) by SLIGRL-NH2 elicits afferent arteriolar vasodilation, in part, by elaborating nitric oxide (NO), suggesting an endothelium-dependent mechanism (Trottier G, Hollenberg M, Wang X, Gui Y, Loutzenhiser K, and Loutzenhiser R. Am J Physiol Renal Physiol 282: F891F897, 2002). In the present study, we characterized the NO-independent component of this response, using the in vitro perfused hydronephrotic rat kidney. SLIGRL-NH2 (10 μmol/l) dilated afferent arterioles preconstricted with ANG II, and the initial transient component of this response was resistant to NO synthase (NOS) and cyclooxygenase inhibition. This NO-independent response was not prevented by treatment with 10 nmol/l charybdotoxin and 1 μmol/l apamin, a manipulation that prevents the endothelium-derived hyperpolarizing factor (EDHF)-like response of the afferent arteriole to acetylcholine, nor was it blocked by the addition of 1 mmol/l tetraethylammonium (TEA) or 50 μmol/l 17-octadecynoic acid, treatments that block the EDHF-like response to bradykinin. To determine whether the PAR-2 response additionally involves the electrogenic Na+-K+-ATPase, responses were evaluated in the presence of 3 mmol/l ouabain. In this setting, SLIGRL-NH2 induced a biphasic dilation in control and a transient response after NOS inhibition. The latter was not prevented by charybdotoxin plus apamin or by TEA alone but was abolished by combined treatment with charybdotoxin, apamin, and TEA. This treatment did not prevent the NO-dependent dilation evoked in the absence of NOS inhibition. Our findings indicate a remarkable redundancy in the signaling cascade mediating PAR-2 -induced afferent arteriolar vasodilation, suggesting an importance in settings such as inflamation or ischemia, in which vascular mechanisms might be impaired and the PAR system is thought to be activated.
nitric oxide; SLIGRL-NH2; potassium channels; ouabain-sensitive Na+-K+-ATPase; tetrathylammonium; apamin; charybdotoxin; 17-octadecynoic acid; endothelium-dependent hyperpolarizing factor
PROTEINASE ACTIVATED RECEPTORS (PARs) are a novel class of G protein-linked receptors that are activated by proteolytic unmasking of a tethered ligand amino acid sequence within the NH2-terminal domain (6, 18). Specific tethered ligand sequences have been identified for each of the four family members (PAR-14) (18, 27, 36). PAR-2 is unique in that it is activated by trypsin, but not by thrombin (18, 27), and is thought to play a prominent role in physiological or pathophysiological processes such as inflamation (reviewed in Refs. 5, 36, and 42). The synthetic peptide SLIGRL-NH2, based on the rodent PAR-2-tethered ligand sequence, has been found to be a highly selective PAR-2 agonist that can activate PAR-2 with the absence of proteolysis and without affecting other receptor systems (18). The role of PAR-2 in the kidney is largely unknown, but this area is of considerable interest as the expression levels of PAR-2 in the kidney are particularly high (3, 35). PAR-2 is expressed in renal epithelia and has been shown to activate a chloride conductance in cortical collecting duct cells (2). However, PAR-2 is also highly expressed in renal vascular endothelial cells and vascular smooth muscle cells (2). Previous studies from our laboratory using the isolated perfused rat kidney model (15) and the in vitro perfused hydronephrotic rat kidney preparation (41) have demonstrated a potential role for PAR-2 in the regulation of renal hemodynamics, as PAR-2 activation in these preparations exerts a potent renal vasodilatory response.
PAR-2 is suggested to play a prominent role in the cardiovascular system, impacting on tissue perfusion and angiogenesis (30, 31). Both trypsin and the selective PAR-2-activating peptide SLIGRL-NH2 induce endothelium-dependent vasodilation in isolated rat aorta, mouse mesenteric, and renal arteries (21, 29, 39). In some vessels, but not all, a component of the endothelium-dependent response is resistant to treatment with cyclooxygenase (COX) and nitric oxide synthase (NOS) inhibition, prompting speculation that PAR-2 activation causes the release of an endothelium-derived hyperpolarizing factor (EDHF) (15, 21, 29, 30, 41). In the isolated, perfused rat kidney, both SLIGRL-NH2 and trypsin elicit renal vasodilation that is only partially attenuated by COX and NOS inhibition (15). Using an isolated, perfused hydronephrotic rat kidney preparation, we have previously shown that the NO-independent afferent arteriolar dilation evoked by SLIGRL-NH2 is prevented by elevated extracellular K+, consistent with the proposed role of an EDHF (41) but also consistent with a direct smooth muscle action.
Recent findings from our laboratory indicate that multiple pathways contribute to the afferent arteriolar responses attributed to EDHF, in that we observed distinct EDHF-like responses of this vessel to acetylcholine and bradykinin (43, 44). The EDHF-like response to acetylcholine is fully abolished by the combined administration of charybdotoxin plus apamin (43), a treatment that eliminates EDHF responses in a wide variety of vessel types (reviewed in Refs. 11 and 28). However, we found that this treatment, although necessary, was not sufficient to abolish the EDHF-like response of the afferent arteriole to bradykinin (44). To eliminate the bradykinin response, it was necessary to use a combination of charybdotoxin and apamin plus either 1 mmol/l tetraethylammonium (TEA) or 50 μmol/l 17-octadecynoic acid (17-ODYA) (44). Either treatment alone was ineffective. These findings are a clear indication of the complex nature of the responses attributed to EDHF in the afferent arteriole.
The present study was undertaken to characterize the NO-independent vasodilator actions of PAR-2 activation on the afferent arteriole. Because, in other vascular beds, this NO-independent component is endothelium dependent and has been ascribed to an EDHF, we anticipated that the characteristics of this response might be similar to that previously reported for acetylcholine or bradykinin. Our findings were surprising in that the characteristics of the NO-independent response to SLIGRL-NH2 were clearly distinct from the EDHF pathways associated with either acetylcholine or bradykinin. Moreover, our study revealed that a remarkably redundant and complex mechanism mediates the afferent arteriolar vasodilation evoked by PAR-2 activation. The reason for this redundancy is not readily apparent, but this finding may implicate an underlying importance in settings where normal vascular responses might be impaired.
METHODS
The effects of the PAR-2-activating peptide agonist SLIGRL-NH2 on the renal afferent arterioles was investigated using the in vitro perfused hydronephrotic kidney. The left ureters of 6- to 7-wk-old (80100 g) male Sprague-Dawley rats were ligated under halothane-induced anesthesia to induced hydronephrosis to facilitate direct observations of the afferent arterioles. After 68 wk, the hydronephrotic kidney was harvested. The left renal artery was cannulated in situ, and the kidney was excised and transferred to a heated chamber on the stage of an inverted microscope, without disruption of perfusion. The renal perfusate consisted of DMEM (Sigma, St. Louis, MO) containing 30 mmol/l bicarbonate, 5 mmol/l glucose, and 5 mmol/l HEPES. The perfusate was equilibrated with 95% air-5% CO2. Temperature and pH were maintained at 37°C and 7.40, respectively.
Medium was pumped on demand through a heat exchanger to a pressurized reservoir connected to the renal arterial cannula. Perfusion pressure was monitored within the renal artery (26) and maintained at 80 mmHg in all experiments. Kidneys were allowed to equilibrate for at least 1 h before study. A fiber-optic probe was used to stabilize and transilluminate a portion of the membranous renal cortex for observations of renal microvascular responses. Afferent arteriolar diameters were measured by online image processing (25). Afferent arteriolar diameter measurements were obtained at each pixel and were averaged over the entire segment length ( 20 μm). These data were collected at a scanning rate of 3 Hz. Thus obtained, mean diameters over the plateau of the response were then averaged for each experimental point.
The synthetic PAR-2-activating peptide agonist SLIGRL-NH2, >95% pure by HPLC and mass spectral criteria, was prepared by the peptide synthesis facility at the University of Calgary (peplab@ucalgary.ca). Stock solutions of SLIGRL-NH2 were prepared in 25 mmol HEPES, pH 7.40, and concentrations were verified by quantitative amino acid analysis. Stock solutions of 17-ODYA (Sigma) were prepared in ethanol. Freshly prepared stock solution of ANG II, nitro-L-arginine methyl ester (L-NAME), TEA, charybdotoxin, and apamin (all obtained from Sigma) were prepared with sterile water. Inhibition of the rat Na+-K+-ATPase isoform requires high concentrations of ouabain. For these experiments, 3 mmol/l ouabain (Sigma) was prepared by direct addition to the perfusate. All reagents were added to the perfusate, which emptied through the renal vein into the tissue bath. Accordingly, all agents reach both adluminal and abluminal vessel surfaces. Experiments employing apamin, charybdotoxin, and SLIGRL-NH2 required the use of a recirculating perfusion system to conserve reagents. Antibiotics (penicillin/streptomycin, Invitrogen, Carlsbad, CA) were added to the recirculating perfusate in these experiments. Kidneys were perfused using a single-pass system and switched to the recirculating system when these agents were employed. Finally, in addition to causing smooth muscle membrane depolarization by inhibition of the electrogenic Na+-K+-ATPase, ouabain can evoke the release of neurotransmitters by a similar action on nerves. Therefore, in all experiments using ouabain, 10 μmol/l phentolamine and 10 μmol/l propranolol (Sigma) were added to the perfusate.
All values are presented as means ± SE. Differences between treatment groups were evaluated by one-way ANOVA and Bonferroni's t-test, when multiple comparisons were evaluated. Differences exhibiting P values <0.05 were considered to be statistically significant.
RESULTS
Effects of PAR-2 agonists SLIGRL-NH2 and trypsin on afferent arteriole during ANG II induced vasoconstriction. To assess the characteristics of vasorelaxation elicited by PAR-2 activation, SLIGRL-NH2 (10 μmol/l) was administered to control kidneys and to kidneys that had been pretreated with 10 μmol/l ibuprofen and 100 μmol/l L-NAME. Afferent arteriolar tone was established by preconstricting the afferent arteriole with 0.1 nmol/l ANG II. In the controls, SLIGRL-NH2 elicited a biphasic response, characterized by an initial transient component and followed by a reduced but sustained vasodilation. As shown in the tracing depicted in Fig. 1A, ANG II reduced afferent arteriolar diameter from 20 to 8 μm, and SLIGRL-NH2 fully restored diameter at the initial peak response (20 μm). The diameter spontaneously returned to 16 μm over the next several minutes. As shown in Fig. 1B, SLIGRL-NH2 elicited only a transient response after arteriolar pretreatment with 100 μmol/l L-NAME plus 10 μmol/l ibuprofen. Thus in this tracing, ANG II reduced afferent arteriolar diameter from 16 to 4 μm, and SLIGRL-NH2 caused a transient increase in diameter to 13 μm, which returned to 4 μm within 35 min. Figure 2, A and B, illustrates tracings from similar experiments assessing responses to trypsin (2 nmol/l). In Fig. 2A, ANG II reduced afferent arteriolar diameter from 17 to 6.5 μm. Trypsin caused a rapid vasodilation to 16 μm, which persisted for 5 min. Proteolytic activation of the PAR-2 receptor, as would occur with trypsin, but not SLIGRL-NH2, is associated with rapid receptor internalization (see Ref. 36), perhaps contributing to the transient nature of the response to this agonist. In the presence of COX/NOS inhibition (Fig. 2B), the response to trypsin was much more transient. In this tracing, ANG II reduced the diameter from 20 to 5 μm, and trypsin caused a transient increase to 14 μm, which rapidly returned to 5 μm.
The persistence of a transient vasodilatory response to PAR-2 activation in the presence of NOS and COX blockade is consistent with previous reports. We had previously shown that this NO-independent component is blocked by KCl-induced depolarization (41), and others have attributed such responses to an EDHF (15, 29, 30). We have observed NO-independent afferent arteriolar responses to both acetylcholine and bradykinin that exhibit similar temporal characteristics (43, 44). In the case of acetylcholine, this EDHF-like response was fully abolished by combined treatment with charybdotoxin plus apamin (43), whereas with bradykinin, charybdotoxin plus apamin and the addition of either TEA or 17-ODYA was required to fully abolish the response (44). However, as shown in Fig. 3, A and B, these treatments did not eliminate the NO-independent response to SLIGRL-NH2. These studies were conducted in the presence of COX and NOS blockade (100 μmol/l L-NAME plus 10 μmol/l ibuprofen). Kidneys were treated with 10 nmol/l charybdotoxin, 1 μmol/l apamin, and either 1 mmol/l TEA (Fig. 3A) or TEA plus 50 μmol/l 17-ODYA (Fig. 3B). Afferent arteriolar tone was increased by the administration of 0.1 nmol/l ANG II. As shown, neither of these treatments was capable of eliminating the response to SLIGRL-NH2, although each treatment attenuated the response. In concert with our previous observations (43, 44), these findings suggest that the NO-independent response to SLIGRL-NH2 involves a separate or an additional mechanism that is not seen with either acetylcholine or bradykinin.
To determine whether these unusual characteristics of the NO-independent response to SLIGRL-NH2 were dependent on the condition that basal afferent arteriolar tone was established by ANG II, we performed additional experiments using either pressure or barium to establish basal tone. In each case, SLIGRL-NH2 evoked similar responses. For these studies, kidneys were pretreated with a "cocktail" containing 100 μmol/l L-NAME, 10 μmol/l ibuprofen, 10 nmol/l charybdotoxin, 1 μmol/l apamin, and 1 mmol/l TEA. Figure 4 illustrates the response obtained when elevated renal arterial pressure (RAP) was used to establish tone. The administration of the cocktail described above reduced diameters from a control of 18.7 ± 1.5 to 16.4 ± 1.4 μm. In this setting, elevating RAP from 80 to 160 mmHg reduced diameters to 7.8 ± 1.2 μm. The administration of 10 μmol/l SLIGRL-NH2 evoked a transient increase in diameter to 13.6 ± 1.2 μm (P = 0.001, n = 4), which spontaneously abated as diameters returned to 7.4 ± 1.2 μm. Similar results were obtained when barium was used to establish basal tone. In this series, diameters were 17.5 ± 1.4 μm in the control state and 16.5 ± 1.2 μm after the exposure to the cocktail of blockers. Barium (100 μmol/l) constricted the arterioles to 9.0 ± 2.1 μm, and the administration of 10 μmol/l SLIGRL-NH2 elicited a transient increase in diameter to 12.9 ± 2.2 μm (P = 0.015, n = 4). Diameters spontaneously returned to 8.3 ± 1.8 μm in the continued presence of the PAR-2 agonist. Thus, under conditions in which basal afferent arteriolar tone was established by either ANG II, elevated pressure, or barium-induced depolarization (see Ref. 4), SLIGRL-NH2 elicited a residual vasodilation in the combined presence of L-NAME, ibuprofen, charybdotoxin, apamin, and TEA.
Role of ouabain-sensitive pump in NO-independent response to SLIGRL-NH2. The literature suggests that, in addition to mechanisms sensitive to the administration of K channel blockers or inhibition of cytochrome P-450, some EDHF-like responses are blocked by ouabain (reviewed in Refs. 11 and 28), suggesting an involvement of the electrogenic Na+-K+-ATPase. We therefore examined whether ouabain would affect the NO-independent responses induced by PAR-2 activation. We had previously shown that ouabain elicits afferent arteriolar vasoconstriction (4), so in these experiments ouabain (3 mmol/l) was administered to elicit afferent arteriolar tone and the effects of SLIGRL-NH2 were evaluated. In this setting, SLIGRL-NH2 evoked a biphasic vasodilation in controls and a transient response in the presence of COX/NOS blockade (Fig. 5, A and B, left). The magnitude of the initial response was not altered by 100 μmol/l L-NAME plus10 μmol/l ibuprofen. As shown in Fig. 5A, right, ouabain reduced afferent arteriolar diameters from 18.8 ± 0.8 to 4.9 ± 0.7 μm in controls, and SLIGRL-NH2 returned diameters to 18.0 ± 1.1 μm (n = 6). In a separate series of kidneys pretreated with L-NAME and ibuprofen (Fig. 5B, right), ouabain reduced afferent arteriolar diameter from 18.1 ± 0.8 to 5.5 ± 0.7 μm, and SLIGRL-NH2 increased diameters to 16.1 ± 0.3 μm (n = 5). The corresponding vasodilation values expressed as a percentage were 94 ± 4% in controls and 84 ± 5% after L-NAME (P = 0.14).
We next examined the effects of K channel blocking agents on the NO-independent response elicited by SLIGRL-NH2 in the presence of ouabain. As depicted in Fig. 6A, the combination of charybdotoxin plus apamin did not prevent the response to SLIGRL-NH2. In these studies, ouabain reduced afferent arteriolar diameters from 17.7 ± 0.8 to 6.6 ± 1.3 μm (P = 0.0002), and SLIGRL-NH2 returned diameters to 12.3 ± 1.6 μm (P = 0.025 vs. control, n = 5). Similarly, the administration of TEA alone was not sufficient to block this response. Thus, as shown in Fig. 6B, in TEA-treated kidneys ouabain reduced afferent arteriolar diameters from 20.0 ± 1.0 to 6.9 ± 1.0 μm (P = 0.0004), and SLIGRL-NH2 returned the diameters to 14.2 ± 1.3 μm (P = 0.001, n = 6). Thus neither charybdotoxin plus apamin nor TEA was capable of preventing the NO-independent response to SLIGRL-NH2 when added alone, in this setting. However, when added together in the presence of ouabain, the combination of TEA, charybdotoxin, and apamin completely abolished this response. As depicted in Fig. 7A, in the presence of these blockers, ouabain reduced afferent arteriolar diameters from 18.7 ± 0.8 to 5.2 ± 0.7 μm, and the subsequent administration of SLIGRL-NH2 had no effect (6.0 ± 0.7 μm, P = 0.40, n = 7).
We had previously shown that the EDHF-like response of the afferent arteriole to bradykinin could be prevented by the combination of charybdotoxin and apamin plus either TEA or the cytochrome P-450 inhibitor 17-ODYA (50 μmol/l) (44). However, as shown in Fig. 7, 17-ODYA did not mimic the inhibition by TEA of the response to SLIGRL-NH2. In the presence of ibuprofen and L-NAME, charybdotoxin, apamin, and 17-ODYA, ouabain reduced afferent arteriolar diameters from 18.1 ± 0.4 to 4.7 ± 0.3 μm. In this setting, SLIGRL-NH2 increased the diameter to 11.1 ± 1.3 μm (P = 0.0001 vs. ouabain, Fig. 6B), suggesting that cytochrome P-450 products, such as an epoxyeicosatrienoic acid (EET), do not appear to mediate the TEA-sensitive component of the NO-independent response to SLIGRL-NH2.
The above observations indicated that a combination of ouabain, charybdotoxin, apamin, and TEA fully blocked the response of the afferent arteriole to SLIGRL-NH2 in the presence of NOS blockade. We next examined the effects of this treatment regime on the response when NOS was not inhibited. The results of these experiments are presented in Fig. 8. In kidneys that had been pretreated with the combination of TEA, apamin plus charybdotoxin, ouabain reduced afferent arteriolar diameters from 19.8 ± 0.3 to 8.7 ± 1.0 μm. In this setting, SLIGRL-NH2 returned the diameters to 14.0 ± 1.2 μm (P = 0.036 vs. ouabain alone, n = 4). Thus this treatment did not prevent the NO-dependent response to the PAR-2 agonist.
Figure 9 summarizes the studies described above examining the effects of various treatments on the SLIGRL-NH2-induced vasodilation in afferent arterioles constricted in the presence of ouabain. To facilitate comparisons, the data are expressed as the percentage dilation of ouabain-induced vasoconstriction. As illustrated, the only combination of treatments that fully abolished the response to SLIGRL-NH2 was L-NAME, ouabain, charybdotoxin, apamin, and TEA (7 ± 3%, P = 0.40 vs. ouabain alone). All other combinations attenuated the vasodilation [TEA, charybdotoxin, apamin, without L-NAME (52 ± 9%); L-NAME plus either TEA (57 ± 11%), charybdotoxin and apamin (53 ± 6%), or charybdotoxin, apamin, and 17-ODYA (51 ± 11%)] but did not fully abolish the response.
DISCUSSION
The present study demonstrates the complex nature of the mechanisms mediating PAR-2-induced afferent arteriolar vasodilation. A component of this response appeared to be endothelium dependent in that it was prevented by L-NAME, suggesting an involvement of endothelium-derived NO. A second component was prevented by combined treatment with apamin plus charybdotoxin, a manipulation that has been shown to block endothelium-dependent hyperpolarization in other vessel types (see Ref. 28 for a review). The remaining components were sensitive to TEA and ouabain and could reflect either additional endothelium-dependent mechanisms or direct actions of PAR-2 activation on the underlying smooth muscle.
Within the circulatory system, PAR-2 is expressed on the endothelium and on vascular smooth muscle myocytes (7). In a number of vascular preparations, PAR-2-induced vasodilation is eliminated by removal of the endothelium (21, 29, 30, 33, 39). Unfortunately, we cannot remove the endothelium in our preparation without affecting vascular reactivity. L-NAME prevented the sustained phase of the afferent arteriolar response to PAR-2 stimulation, and the NO-independent component that remained was transient in character, was not blocked by cyclooxygenase, but was abolished by depolarizing concentrations of extracellular K+. These characteristics are similar to responses attributed to EDHF in other preparations (21, 29, 33, 41). There is limited information on the nature of the EDHF involved in the responses of other vessel types to PAR-2 agonists. Nakayama et al. (33) found that in the porcine coronary artery, the EDHF-like response to trypsin was fully abolished by the combined treatment with charybdotoxin plus apamin. Similarly, Kawabata et al. (21) found the combination of charybdotoxin plus apamin to abolish the NO-independent dilator effects of SLIGRL-NH2 on gastric mucosal blood flow in the rat. McGuire et al. (29) also found this treatment to abolish the EDHF-like response to SLIGRL-NH2 in the mouse mesenteric arteriole. However, these authors also found the combination of ouabain and barium (30 μmol/l) attenuated this response. In the latter study, TEA, iberiotoxin, and cytochrome P-450 inhibition all had no effect. In contrast, McLean et al. (30) found that the NO-independent component of the response of the isolated perfused rat heart to SLIGRL or trypsin could be attenuated by TEA, charybdotoxin plus apamin, eicosatetraynoic acid, nordihydroguaiaretic acid, baicalein, capsaicin, and capsazepine, suggesting a complex and redundant signaling pathway involving K+ channels, lipoxygenase-derived ecosanoids, and vanilloid receptors. Thus while some common aspects of the PAR-2-induced EDHF-like responses are apparent, for example, the charybdotoxin/apamin-sensitive component, it appears that responses observed in different vascular beds display distinct characteristics.
We have previously characterized the nature of the EDHF-like responses of the afferent arteriole to acetylcholine (43) and bradykinin (44), and these findings along with the present results are summarized in Fig. 10. Each study employed the in vitro perfused hydronephrotic rat kidney model under identical experimental conditions, in which basal tone was established with ANG II. In Fig. 10, the open bars depict the peak dilations elicited under control conditions, whereas the filled bars depict peak responses after inhibition of NOS and COX. Each vasodilator elicited a transient response under the latter conditions and, in each case, the peak dilations were similar to the peak responses seen in controls. Moreover, these transient dilations were fully blocked by elevated extracellular K+ (16, 41, 44), consistent with the possible involvement of an EDHF. A common feature of all three agents was that a component of the dilation was blocked by treatment with apamin plus charybdotoxin (Figs. 9 and 10). We and others have suggested that this apamin/charybdotoxin-sensitive mechanism involves the activation of small (SKCa) and intermediate (IKCa) Ca-activated K channels that are located on the endothelium and that evoke smooth muscle hyperpolarization via myoendothelial gap junctions (e.g., Ref. 9; also discussed in Refs. 11 and 28). However, as further shown in Fig. 10, there are marked differences in other aspects of the NO-independent responses to these three agents. The response to acetylcholine was fully abolished by treatment with apamin plus charybdotoxin, whereas this treatment alone did not prevent the EDHF-like response to bradykinin. Rather, the bradykinin response involves one component that is blocked by charybdotoxin plus apamin and a second that is prevented by either TEA (1 mmol/l) or with 17-ODYA (further discussed in Ref. 44). The NO-independent response to SLIGRL-NH2 differed from that of either acetylcholine or bradykinin, in that it persisted in the combined presence of charybdodoxin, apamin, TEA, and 17-ODYA, indicating an additional mechanism that is not evoked by either of these agents.
As shown in Fig. 9, this additional mechanism was prevented by ouabain. Ouabain, which blocks the electrogenic Na+-K+ -ATPase, has been shown to inhibit responses attributed to EDHF (reviewed in Refs. 11 and 28), including those evoked by PAR-2 (e.g., Ref. 29). In the present study, we found that while ouabain fully prevented the NO-independent response to PAR-2 stimulation in the combined presence of TEA, apamin, and charybdotoxin (Figs. 7 and 9), this treatment did not prevent the NO-dependent response seen in the absence of L-NAME (Fig. 8), suggesting a specificity of action rather than a general suppression of PAR-2 activation. We cannot ascertain from our studies whether the ouabain-sensitive component of the PAR-2-induced afferent arteriolar dilation is endothelial dependent. However, mechanisms linking activation of the overlying endothelium to a stimulation of smooth muscle Na+-K+-ATPase have been suggested.
Edwards et al. (10) suggested that K+ efflux from the overlying endothelium in response to the activation of SKCa and IKCa causes an elevation of extracellular K+ near the sarcolemma of the underlying smooth muscle myocytes and that this elevation in K+ elicits hyperpolarization and vasodilation by stimulating the electrogenic Na+-K+-ATPase. While this hypothesis might seem an attractive means of explaining the involvement of the ouabain-sensitive component of actions of PAR-2 activation, there are a number of discrepancies that suggest this is not the case. First, while elevated extracellular K+ does indeed elicit afferent arteriolar vasodilation, we have previously shown that this response is not prevented by ouabain but rather is fully abolished by barium (4), suggesting that the dilation is mediated primarily by alterations in the inwardly rectifying K channel (KIR). Thus the properties of the response elicited by SLIGRL-NH2 are quite distinct from that induced by elevations in extracellular K+. Second, we observed that although the responses elicited by acetylcholine, bradykinin, and SLIGRL-NH2 were all sensitive to charybdotoxin and apamin, indicating an involvement of SKCa and IKCa, only the response to SLIGRL-NH2 was sensitive to ouabain. If K+ efflux via SKCa and IKCa stimulates the electrogenic Na+-K+-ATPase, why would not all three of these agents exhibit a component that is ouabain sensitive Finally, it should be noted that the proposed role of K+ as an EDHF remains controversial, and many observations underlying this premise are not confirmed by others (reviewed in Ref. 28), even by laboratories using the same vascular preparation (e.g., Ref. 1). A study by Pratt et al. (37) suggests an alternate mechanism whereby a vasodilator agent acting on the endothelium might cause a stimulation of the electrogenic Na+-K+-ATPase. These investigators found that ouabain inhibited both bradykinin- and 14,15 EET-induced dilation of the bovine coronary artery and suggested that endothelium-derived EETs act via a ouabain-sensitive mechanism in this vessel. However, as shown in Fig. 3B, we found that 17-ODYA, which would inhibit EET formation, did not prevent the ouabain-sensitive component of the afferent arteriole to SLIGRL-NH2.
It is equally possible that the ouabain-sensitive component of the PAR-2-induced response is not endothelium dependent but rather involves a direct action mediated by PAR-2 receptors located on the afferent arteriolar myocyte. For example, studies assessing the effects of adenosine on endothelium-denuded aorta indicate a direct vasorelaxant effect that is ouabain sensitive (14). Similarly, the relaxant effects of isoproterenol on isolated detrusor smooth muscle cells (32) and the effects of bradykinin on cultured tracheal smooth muscle myocytes (8) have both been suggested to involve a stimulation of smooth muscle Na+-K+-ATPase via mechanisms that clearly could not involve the endothelium. Future studies assessing the expression of PAR-2 receptors on afferent arteriolar myocytes and studies examining the direct effects of PAR-2 activation on these cells would be of interest in this regard.
We found TEA was required to completely abolish the response to PAR-2 (Fig. 9). This TEA-sensitive component could also involve either a direct action or an endothelium-dependent mechanism. We found TEA to similarly inhibit a component of the afferent arteriolar response to bradykinin, and this action was mimicked by 17-ODYA (Fig. 10). Several studies have implicated EETs in the EDHF-like response to bradykinin (12, 13, 19), and the vasodilator actions of 11,12 EET on the afferent arteriole are blocked by 1 mmol/l TEA (20, 45). Thus a model in which bradykinin stimulates the release of an EET, whose formation can be prevented by 17-ODYA and whose smooth muscle vasodilator actions can be blocked by TEA, could explain the bradykinin response (see Ref. 44 for further discussion). In the present study, we found that TEA also blocks a component of the NO-independent response to SLIGRL-NH2. However, unlike our observations with bradykinin, 17-ODYA had no effect on the PAR-2 response (Fig. 7). Accordingly, TEA must be affecting another signaling pathway. In this regard, it should be noted that, at the concentration used in our studies (1 mmol/l), TEA selectively blocks large-conductance Ca-activated K channels (BKCa) (23) and that a variety of vasodilatory agents, other than EETs, have been demonstrated to act by altering the activity of BKCa (reviewed in Ref. 40). Thus the TEA-sensitive component may involve a mechanism whereby activation of PAR-2 receptors, located either on the myocyte or on the endothelium, is linked to a stimulation of smooth muscle BKCa channels.
The role of NO in the response of the afferent arteriole to PAR-2 activation merits further discussion. As seen with other vessels, it appears that NO-independent mechanisms contribute predominantly to the initial phasic component of the response. The sustained component is fully blocked by L-NAME, reflecting an obligate role of sustained NO formation (Fig. 1). However, as shown in Figs. 9 and 10, when exposed to the combined presence of ouabain, TEA, and charybdotoxin plus apamin, SLIGRL-NH2 elicited a transient NO-dependent afferent arteriolar vasodilation. We observed similar transient NO-dependent responses to both acetylcholine (43) and bradykinin (44) when the EDHF component of each agent was blocked (see Fig. 10). Why does one not see a sustained NO-dependent response in this setting This likely reflects the obligate role of K+ channel activation in Ca2+ signaling of the endothelial cell. SKCa and IKCa are thought to play essential roles in the sustained entry of Ca2+ in the activated endothelial cell (see Ref. 34 for a review). Accordingly, blockade of these K+ channels would be anticipated to affect the sustained, Ca2+-dependent activation of eNOS but would have less effect on the initial response, which depends on the release of intracellular Ca2+ stores (34).
Finally, what could be the physiological or pathophysiological significance of the PAR-2-induced vascular responses of the afferent arteriole Currently, little is known regarding the role of PARs in the kidney, though PAR-2 is abundantly expressed in this organ (3, 2). PAR-1 and PAR-2 exert apposing effects on renal blood flow and glomerular filtration rate in the isolated, perfused rat kidney (15) and thus have the capability of bidirectional control of renal hemodynamics. In other organs, evidence implicates an important role of PARs in inflammatory responses and in responses to tissue injury (reviewed in Refs. 5, 36, and 42). The endogeneous activators of PAR-2 are proteinases, such as trypsin and mast cell tryptase. Increased numbers of renal mast cells are associated with glomerulonephritis, diabetic nephropathy, and renal graft rejection (17, 22, 38), and urinary proteinase activity is reported to be elevated in patients with acute and chronic renal failure (24). Does the present study provide any insights into the potential role of PARs in such settings One very curious aspect of our findings relates to the remarkable redundancy of the mechanisms mediating the actions of PAR-2 on the afferent arteriole. Why would such multiple and potentially overlapping vasodilatory mechanisms have evolved A similar complexity of PAR-2-induced dilatory responses has been reported for the coronary circulation (30). An important observation in this regard is the finding by McLean et al. (30) that vasodilator responses of the perfused heart to acetylcholine are impaired after ischemia whereas PAR-2 responses are preserved, suggesting the potential importance of PAR-2 in the conditions associated with ischemia-reperfusion. Did the redundancy in vasodilatory mechanisms seen in the afferent arteriole evolve to preserve vascular responsiveness of this vessel in pathophysiological settings These are all interesting questions for future investigations.
In conclusion, the present study demonstrates that redundant mechanisms contribute to the afferent arteriolar vasodilatory response to PAR-2 activation. NO plays a prominent role in the sustained response, whereas the initial response involves an NO-independent component, mediated by mechanisms involving K+ channels that are sensitive to blockade by charybdotoxin plus apamin (e.g., IKCa and SKCa) and by TEA (e.g., BKCa) and by a mechanism that is dependent on the electrogenic Na+-K+-ATPase. This remarkable redundancy may indicate an important role in settings associated with altered vascular reactivity, similar to that suggested to exist in the coronary circulation (30).
GRANTS
These studies were supported by grants from the Canadian Institutes for Health Research (R. Loutzenhiser and M. Hollenberg), the Heart and Stroke Foundation of Alberta and Nunavut (R. Loutzenhiser and M. Hollenberg), the Kidney Foundation of Canada (M. Hollenberg), and by a University-Industry grant in conjunction with Servier International (M. Hollenberg). R. Loutznehiser is a Alberta Heritage Foundation for Medical Research Scientist.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Andersson DA, Zygmunt PM, Movadeh P, Andersson TLG, and Hgesttt ED. Effects of inhibitors of small- and intermediate-conductance calcium-activated potassium channels, inwardly-rectifying potassium channels and Na+/K+ ATPase on EDHF relaxations in the rat hepatic artery. Br J Pharmacol 129: 14901496, 2000.
Bertog M, Letz B, Kong W, Steinhoff M, Higgins MA, Bielfeld-Ackermann A, Fromter E, Bunnett NW, and Korbmacher C. Basolateral proteinase-activated receptor (PAR-2) induces chloride secretion in M-1 mouse renal cortical collecting duct cells. J Physiol 521: 317, 1999.
Bohm SK, Kong W, Bromme D, Smeekens SP, Anderson DC, Connolly A, Kahn M, Nelken NA, Coughlin SR, Payani DG, and Bunnett NW. Molecular cloning, expression and potential functions of the human proteinase-activated receptor-2. Biochem J 314: 10091016, 1996.
Chilton L and Loutzenhiser R. Functional evidence of an inward rectifier potassium current in the renal afferent arteriole. Circ Res 88: 152158, 2001.
Cocks TM and Moffatt JD. Protease-activated receptors: sentries for inflammation Trends Pharmacol Sci 21: 103108, 2000.
Coughlin SR. Thrombin signalling and protease-activated receptors. Nature 407: 258264, 2000.
Damiano BP, D'Andrea MR, de Garavilla L, Cheung WM, and Andrade-Gordon P. Increased expression of protease activated receptor-2 (PAR-2) in balloon-injured rat carotid artery. Thromb Haemost 81: 808814, 1999.
Dodson AM and Rhoden KJ. Bradykinin increases Na+-K+ pump activity in cultured guinea-pig tracheal smooth muscle cells. Br J Pharmacol 133: 13391345, 2001.
Doughty JM, Plane F, and Langton PD. Charybdotoxin and apamin block EDHF in rat mesenteric artery if selectively applied to the endothelium. Am J Physiol Heart Circ Physiol 276: H1107H1112, 1999.
Edwards G, Dora KA, Gardener MJ, Garland CJ, and Weston AH. K+ is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 396: 269272, 1998.
Feletou M and Vanhoutte PM. Endothelium-dependent hyperpolarization of vascular smooth muscle cells. Acta Pharmacol Sin 21: 118, 2000.
Fulton D, Mahboubi K, McGiff JC, and Quilley J. Cytochrome p450-dependent effects of bradykinin in the rat heart. Br J Pharmacol 114: 99102, 1995.
Gauthier KM, Deeter C, Krishna UM, Reddy YK, Bondlela M, Falck JR, and Campbell WB. 14,15-Epoxyeicosa-5(Z)-enoic acid: a selective epoxyeicosatrienoic acid antagonist that inhibits endothelium-dependent hyperpolarization and relaxation in coronary arteries. Circ Res 90: 10281036, 2002.
Grbovic L and Radenkovic M. Analysis of adenosine vascular effect in isolated rat aorta: possible role of Na+/K+-ATPase. Pharmacol Toxicol 92: 265271, 2003.
Gui Y, Loutzenhiser R, and Hollenberg M. Bidirectional regulation of renal hemodynamics by proteinase-activated receptors 1 and 2. Am J Physiol Renal Physiol 285: F95F104, 2003.
Hayashi K, Loutzenhiser R, Epstein M, Suzuki H, and Saruta T. Multiple factors contribute to acetylcholine-induced vasodilation during myogenic, norepinephrine- and KCl-induced afferent arteriolar vasoconstriction: studies in the isolated perfused hydronephrotic kidney. Circ Res 75: 821828, 1994.
Hiromura K, Kurosawa M, Yano S, and Naruse T. Tubulointerstitial mast cell infiltration in glomerulonephritis. Am J Kidney Dis 32: 593599, 1998.
Hollenberg MD and Compton SJ. International union of pharmacology. XXVIII. Proteinase-activated receptors. Pharmacol Rev 54: 203217, 2002.
Imig JD, Falck JR, Wei S, and Capdevila JH. Epoxygenase metabolites contribute to nitric oxide-independent afferent arteriolar vasodilation in response to bradykinin. J Vasc Res 38: 247255, 2001.
Imig JD, Navar LG, Roman RJ, Reddy KK, and Falck JR. Actions of epoxygenase metabolites on the preglomerular vasculature. J Am Soc Nephrol 7: 23642370, 1996.
Kawabata A, Kubo S, Nakaya Y, Ishiki T, Kuroda R, Sekiguchi F, Kawao N, and Nishikawa H. Distinct roles for protease-activated receptors 1 and 2 in vasomotor modulation in rat superior mesenteric artery. Cardiovasc Res 61: 683692, 2004.
Lajoie G, Nadasdy T, Laszik Z, Blick KE, and Silva FG. Mast cells in acute cellular rejection of human renal allografts. Mod Pathol 9: 11181125, 1996.
Langton PD, Nelson MT, Huang Y, and Standen NB. Block of calcium-activated potassium channels in mammalian arterial myocytes by tetraethylammonium ions. Am J Physiol Heart Circ Physiol 260: H927H934, 1991.
Lee DY, Park SK, Yorgin PD, Cohen P, Oh Y, and Rosenfeld RG. Alterations in insulin-like growth factor-binding proteins (IGFBPs) and IGBBP-3 protease activity in serum and urine from acute and chronic renal failure. J Clin Endocrinol Metab 79: 13761382, 1994.
Loutzenhiser R. In situ studies of renal arteriolar function using the in vitro perfused hydronephrotic rat kidney. Int Rev Exp Pathol 36: 145160, 1996.
Loutzenhiser R, Horton C, and Epstein M. Flow-induced errors in estimating perfusion pressure of the isolated rat kidney. Kidney Int 22: 693696, 1982.
Macfarlane SR, Seatter MJ, Kanke T, Hunter GD, and Plevin R. Proteinase-activated receptors. Pharmacol Rev 53: 245282, 2001.
McGuire JJ, Ding H, and Triggle CR. Endothelium-derived relaxing factors: a focus on endothelium-derived hyperpolarizing factor(s). Can J Physiol Pharmacol 79: 443470, 2001.
McGuire JJ, Hollenberg MD, Andrade-Gordon P, and Triggle CR. Multiple mechanisms of vascular smooth muscle relaxation by the activation of proteinase-activated receptor 2 in mouse mesenteric arterioles. Br J Pharmacol 135: 155169, 2002.
McLean PG, Aston D, Sarkar D, and Ahluwalia A. Protease-activated receptor-2 activation causes EDHF-like coronary vasodilation: selective preservation in ischemia/reperfusion injury: involvement of lipoxygenase products, VR1 receptors, and C-fibers. Circ Res 90: 465472, 2002.
Milia AF, Salis MB, Stacca T, Pinna A, Madeddu P, Trevisani M, Geppetti P, and Emanueli C. Protease-activated receptor-2 stimulates angiogenesis and accelerates hemodynamic recovery in a mouse model of hindlimb ischemia. Circ Res 91: 346352, 2002.
Nakahira Y, Hashitani H, Fukuta H, Sasaki S, Kohri K, and Suzuki H. Effects of isoproterenol on spontaneous excitations in detrusor smooth muscle cells of the guinea pig. J Urol 166: 335340, 2001.
Nakayama T, Hirano K, Nishimura J, Takahashi S, and Kanaide H. Mechanism of trypsin-induced endothelium-dependent vasorelaxation in the porcine coronary artery. Br J Pharmacol 134: 815826, 2001.
Nilius B and Droogmans G. Ion channels and their functional role in vascular endothelium. Physiol Rev 81: 14151459, 2001.
Nystedt S, Emilsson K, Wahlestedt C, and Sundelin J. molecular cloning of a potential proteinase-activated receptor. Proc Natl Acad Sci USA 91: 92089212, 1994.
Ossovskaya VS and Bunnett NW. Protease-activated receptors: contribution to physiology and disease. Physiol Rev 84: 579621, 2004.
Pratt PF, Li P, Hillard CJ, Kurian J, and Campbell WB. Endothelium-independent, ouabain-sensitive relaxation of bovine coronary arteries by EETs. Am J Physiol Heart Circ Physiol 280: H1113H1121, 2001.
Roberts IS and Brenchley PE. Mast cells: the forgotten cells of renal fibrosis. J Clin Pathol 53: 858862, 2000.
Roy SS, Saifeddine M, Loutzenhiser R, Triggle CR, and Hollenberg MD. Dual endothelium-dependent vascular activities of proteinase-activated receptor-2-activating peptides: evidence for receptor heterogeneity. Br J Pharmacol 123: 14341440, 1998.
Schubert R and Nelson MT. Protein kinases: tuners of the BKCa channel in smooth muscle. Trends Pharmacol Sci 22:505512, 2001.
Trottier G, Hollenberg M, Wang X, Gui Y, Loutzenhiser K, and Loutzenhiser R. PAR-2 elicits afferent arteriolar vasodilation by NO-dependent and NO-independent actions. Am J Physiol Renal Physiol 282: F891F897, 2002.
Vergnolle N, Wallace JL, Bunnett NW, and Hollenberg MD. Protease-activated receptors in inflammation, neuronal signaling and pain. Trends Pharmacol Sci 22: 146152, 2001.
Wang X and Loutzenhiser R. Determinants of the renal microvascular response to ACh, afferent and efferent arteriolar action of EDHF. Am J Physiol Renal Physiol 282: F124F132, 2002.
Wang X, Trottier G, and Loutzenhiser R. Determinants of the renal afferent arteriolar actions of bradykinin: evidence for multiple pathways mediating the vasodilator responses attributed to EDHF. Am J Physiol Renal Physiol 285: F540F549, 2003.
Zou AP, Fleming JT, Faalck JR, Jacobs ER, Gebremedhin D, Harder DR, and Roman RJ. Stereospecific effects of epoxyeicosatrienoic acids on renal vascular tone and K+-channel activity. Am J Physiol Renal Fluid Electrolyte Physiol 270: F822F832, 1996.