Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Heterogeneity in the responsiveness of blood vessels is to a large extent related to the variability in the types and densities of pharmacological receptors and ion transport mechanisms of smooth muscle (Mulvany and Aalkjær, 1990). In addition, there are a number of reports indicating that endothelial secretions may vary, depending on the size and location of the artery (Hwa et al., 1994; Clark and Fuchs, 1997). The vascular endothelium can release contracting factors (Miller and Vanhoutte, 1985) and relaxing factors such as nitric oxide (NO), prostacyclin, and endothelium-derived hyperpolarizing factor (EDHF; Moncada et al., 1977; Furchgott and Zawadzki, 1980; Rubanyi and Vanhoutte, 1987). The release of EDHF from the endothelium is triggered by agonists such as acetylcholine (ACh), bradykinin (BK), histamine (Hist), or substance P (SP), and has been demonstrated in various arteries from different species (for a review, see Mombouli and Vanhoutte, 1997). The identity of EDHF has not yet been established, but its action is believed to occur via the activation of K⁺ channels in vascular smooth muscle cells. However, the types of K⁺ channels stimulated and the endothelial secretion responsible for this stimulation seem to vary according to the preparation investigated (for reviews, see Mombouli and Vanhoutte, 1997; Vanhoutte, 1998).

To further complicate matters, NO causes hyperpolarization in rabbit carotid artery (Cohen et al., 1997) and activates large conductance Ca²⁺-activated K⁺ channels (BK_Ca) in arterial smooth muscle cells (Bolotina et al., 1994; Weidelt et al., 1997) and ATP-activated K⁺ channels (K_ATP; Murphy and Brayden, 1995; Weidelt et al., 1997) in vascular smooth muscle. In the mesenteric vascular bed, including the main mesenteric artery and its smaller branches, the action of EDHF is not inhibited by K_ATP blockers (McPherson and Angus, 1991). The inhibitory influence of charybdotoxin (ChTX) suggests rather the involvement of BK_Ca (Hwa et al., 1994). Some studies performed in bovine, porcine, and rat coronary arteries suggest that EDHF may be a metabolite of arachidonic acid, derived from cytochrome P-450-dependent monooxygenase (Hecker et al., 1994). More recently, EDHF as been described as K⁺ in small resistance arteries of rat (Edwards et al., 1998). These observations emphasize the complexity of vascular endothelial secretions as well as the need to further explore the contribution of EDHF when both the tissue and the agonist are varied. Another unresolved question is whether the same EDHF released from different arteries targets different K⁺ channels depending on the location of the smooth muscle or whether there are multiple EDHFs, each targeting a specific K⁺ channel.

Comparative studies on resistance arteries have revealed that cerebral arteries differ pharmacologically and electrophysiologically from arteries in other areas. For example, cerebral arteries are more sensitive than peripheral arteries to dihydropyridine Ca²⁺ antagonists and agonists (Asano et al., 1993) and feature prominent coupling between electrical and mechanical events (Harder and Waters, 1984). Studies examining heterogeneity in isolated vessels have mainly been performed in large systemic arteries. To date, no study has compared endothelial function in cerebral and peripheral resistance arteries.

Thus, experiments were designed to compare the endothelial responses of cerebral resistance and small mesenteric arteries to a range of vasoactive agents. The relative contribution of EDHF to the total endothelium-mediated relaxation as a function of the mechanism whereby the stimulus enhances the cytoplasmic calcium concentration is explored. A unique set of targets for EDHF released from the endothelium of small mesenteric arteries is identified based on the experiments performed by using different K⁺ channel inhibitors. To recreate relevant physiological conditions, arteries were constricted by pressure rather than agonists. For this reason, it was important to initially compare the tone induced by pressure in both types of resistance vessels before studying the effects of different endothelial agonists.

The present study provides evidence for a role of EDHF in activating small conductance Ca²⁺-activated K⁺ channels (SK_Ca), BK_Ca, but not K_ATP.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Vessel Isolation and Cannulation. Male Sprague-Dawley rats (200-300 g) were anesthetized with i.p. injections of sodium pentobarbital (Somnotol; 30 mg/kg; MCT Pharmaceutical, Cambridge, Ontario, Canada) and heparin (Hepalean; 500 U/kg; Organon-Teknika, Toronto, Canada) and then sacrificed by decapitation. The brain or small intestine with attached mesentery was excised and transferred to a dissection dish filled with physiological salt solution (PSS) at 4°C. Distal middle cerebral or small mesenteric (third- or fourth-generation) arteries were dissected from surrounding connective tissues and transferred to the experimental chamber of an arteriograph filled with oxygenated PSS at 37°C.

Myogenic Tone. Microvessels with a diameter below 125 µm (70-120 µm) from cerebral or mesenteric beds were studied to identify region-specific differences in myogenic profile. After the development of myogenic tone during the equilibrium period (60 min), the relation between pressure and vessel diameter was studied. Intravascular pressure was decreased to 10 mm Hg and then raised in 10-mm-Hg steps from 10 to 120 mm Hg, while measuring corresponding changes in vessel diameter. At each step, diameter was monitored for 5 to 10 min until steady state was achieved. The protocol was repeated, and the results were averaged.

Endothelial Heterogeneity. To identify heterogeneity in receptor- and nonreceptor-mediated vasoactivity between cerebral and mesenteric vascular beds, responses to ACh (1 µM), BK (0.1 µM), SP (0.01 µM), Hist (3 µM), calcium ionophore (A23187; 0.1 µM), cyclopiazonic acid (CPA; 20 µM), an inhibitor of endoplasmic reticulum (ER) Ca²⁺-ATPase, and sodium nitroprusside (SNP; 10 µM), an exogenous NO donor, were investigated. A pilot study determined that these concentrations yielded a maximum response in both cerebral and small mesenteric arteries. Vasoactive compounds were added to the circulating buffer at the final concentrations reported above, and changes in luminal diameter were measured. The choice and the order of vasoactive agent exposure to the vessels were randomized so that treatment of the vessel with a particular agent would not influence the subsequent response to another.

Signaling in Endothelial Heterogeneity. To test the role of NO in the vasodilation induced by various agents, N-nitro-L-arginine methyl ester (L-NAME; 200 µM), a competitive inhibitor of constitutive and inducible nitric oxide synthase (NOS) isoforms, was added to the superfusing buffer and allowed to circulate for 20 min until a new steady-state diameter was reached. This was followed by reassessment of the vasodilation due to the agents used. Otherwise, in the two preparations (cerebral and mesenteric arteries), the same vessels were challenged by two vasoactive agents in the absence and presence of L-NAME. Several vessels were used as controls (by omitting the addition of L-NAME in an otherwise identical protocol) to verify that the diameter responses to a vasoactive agent (ACh, BK, SP, Hist, A23187, CPA, SNP) did not decrease with repeated exposure or with time. The effects of NO scavenging were also assessed in vessels challenged with ACh or CPA using oxyhemoglobin (Hb; 10 µM) at the concentration previously used in small mesenteric artery of the rat (Hwa et al., 1994).

8

3

8

3

Dilation Resistant to Indo plus L-NAME. The possible involvement of K⁺ channels in the vasodilation induced by ACh and CPA in myogenically active vessels was assessed in normal PSS containing L-NAME and Indo. In these experiments, K⁺ channel blockers were used at the concentrations known to selectively inhibit specific K⁺ channels; tetraethylammonium (TEA; 3-5 mM), charybdotoxin (ChTX; 100-150 nM), and iberiotoxin (IbTX; 30 nM) inhibited BK_Ca, apamin (0.3 µM) inhibited SK_Ca, glibenclamide (10 µM) inhibited K_ATP, 4-aminopyridine (4-AP; 100 µM) inhibited voltage-gated K⁺ channels, and BaCl₂ (30 µM) inhibited inward rectifier K⁺ channels.

9

8

Expression of Results and Statistical Analysis. Myogenic tone at each pressure was expressed as percent decrease in diameter from the "passive" diameter or

(1)

where D is the arterial diameter in Ca²⁺-free buffer or PSS.

(2)

(3)

Drugs and Solutions. The ionic composition of the PSS consisted of 119 mM NaCl, 4.7 mM KCl, 1.18 mM KH₂PO₄, 24 mM NaHCO₃, 1.17 mM MgSO₄ · 7H₂O, 1.6 mM CaCl₂, 5.5 mM glucose, and 0.026 mM EDTA. ACh, A23187, 4-AP, barium chloride, BK, ChTX, clotrimazole, 3-[(3-cholamidopropyl)dimethylammonio]propanesulfonate, DHO, glibenclamide, Hb, Hist, Indo, L-NAME, PE, TEA, 17-ODYA, and SKF 525a were purchased from Sigma (Ontario, Canada). Apamin, CPA, IbTX, OOPC, and SNP were purchased from Calbiochem (San Diego, CA). Quinacrine was purchased from Research Biochemicals, Inc. (Natick, MA). Stock solutions were diluted in deionized water (NANOpure). CPA, Indo, and glibenclamide were prepared in dimethyl sulfoxide. Clotrimazole, 17-ODYA, and SKF 525a were dissolved in absolute ethanol. Reduced Hb was prepared by treatment of Hb solution with sodium dithionite according to Martin et al. (1985). The effects of ethanol and other solvents were tested, and none of the vehicle solutions altered the pressure-diameter relation or the vascular responses to norepinephrine and ACh.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Myogenic Tone of Middle Cerebral and Small Mesenteric Arteries. Cerebral arteries of all sizes developed graded myogenic constrictions over the physiological pressure range. The steady-state response of distal middle cerebral arteries (range, 70-120 µm; mean, 96.1 ± 18.2 µm; n = 7) to stepwise intraluminal pressure under zero-flow conditions is shown in Fig. 1A. A maximum constriction (25.64 ± 1%) was obtained at 60 mm Hg. Beyond this point, further increases in intravascular pressure (up to 120 mm Hg) did not significantly change the diameter. Thus, maximal myogenic responsiveness was identified at an intravascular pressure of approximately 60 mm Hg in cerebral arteries of the rat. No statistical difference was observed in the tone of middle cerebral arteries at 60 mm Hg in the absence (26.8 ± 4.92%) and presence (25.6 ± 1%) of the endothelium.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Myogenic tone (A) and wall tension (B) in rat distal middle cerebral and small mesenteric resistance-sized arteries. The points are mean ± S.E.M. (vertical bars) of 14 experiments. *P < .05, **P < .01, ***P < .001, significantly different compared with distal middle cerebral arteries.

Vasomotion of Middle Cerebral and Small Mesenteric Arteries. Figure 2A shows the responses of cerebral arteries to several vasoactive agents. Only BK, SP, and SNP induced vasodilation; notably, ACh failed to induce significant dilation. Control experiments using a distal middle cerebral artery (approximately 100 µm) and proximal middle cerebral artery (approximately 300 µm) showed that the former was able to dilate modestly (0-5%) and the latter to a greater degree to ACh (1, 10 µM). However, in the same arteries, BK (0.1 µM) or SP induced somewhat greater dilation (20-50%). Hist, A23187, and CPA all caused vasoconstriction. In the presence of L-NAME, the dilations induced by BK and SP were abolished, whereas ACh induced a contraction. Hist, A23187, and CPA all caused greater vasoconstriction in the presence of L-NAME, whereas the SNP-induced dilation was not significantly different from control.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Comparative effects of agonists and vasoactive agents in rat (A) distal middle cerebral and (B) small mesenteric arteries with myogenic tone. Histograms show the responses induced by ACh (1 µM), BK (0.1 µM), SP (0.01 µM), Hist (3 µM), A23187 (0.1 µM), CPA (20 µM), and SNP (10 µM) in the absence (open columns) and in the presence of 200 µM L-NAME (filled columns). The points are mean ± S.E.M. (vertical bars) of seven experiments. **P < .01, ***P < .001, significantly different from the percentage of the dilation obtained in the absence of L-NAME.

Vasodilatation Resistant to NOS Inhibitor in Mesenteric Arteries. Because the L-NAME-resistant components of ACh and CPA vasodilation could be due to endothelial production of prostacyclin, Indo was used to inhibit COX (Figs. 3 and 4). L-NAME alone (P < .01) and Indo alone (P < .05) partially inhibited the dilation induced by ACh (Fig. 5A). The combination of L-NAME plus Indo is not greater than L-NAME alone or Indo alone (Table 1, Figs. 3 and 5A). In contrast, Indo alone, L-NAME alone, Hb alone, or the combination of L-NAME plus Hb did not affect the dilation produced by CPA (Fig. 6A). Only the combination of Indo plus L-NAME partially affected CPA-induced dilation of small mesenteric arteries (Table 1 and Fig. 6A).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Mechanisms of vasodilation independent of NO/COX pathway induced by ACh in rat small mesenteric arteries pressurized at 80 mm Hg. Representative diameter traces of five to seven experiments. ACh concentration-response curves (107 to 105 M) in small mesenteric arteries of the rat with myogenic tone at 80 mm Hg in the absence (A) and in the presence of either 200 µM L-NAME plus 10 µM Indo (B) or L-NAME plus Indo plus 0.3 µM apamin (C).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Mechanisms of vasodilation independent of the NOS/COX pathways induced by CPA in rat small mesenteric arteries pressurized at 80 mm Hg. Representative diameter traces of five to seven experiments. CPA concentration-response curves (107 to 3.105 M) in rat small mesenteric arteries with myogenic tone at 80 mm Hg in the presence of 200 µM L-NAME plus 10 µM Indo (A), L-NAME plus Indo plus 0.3 µM apamin (B), and L-NAME plus Indo plus apamin plus 30 nM IbTX (C).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Mechanisms of ACh-induced vasodilation resistant to L-NAME and Indo in myogenically active small mesenteric arteries of the rat. A, concentration-response curves to ACh in small mesenteric arteries of the rat pressurized at 80 mm Hg in the absence () and in the presence of 200 µM L-NAME (), 10 µM Indo (), and L-NAME plus Indo (). B, concentration-response curves to ACh in small mesenteric arteries of the rat pressurized at 80 mm Hg in the presence of L-NAME plus Indo (), L-NAME plus Indo plus 30 nM IbTX (), and L-NAME plus Indo plus 0.3 µM apamin (). The points are mean ± S.E.M. (vertical bars) of five to seven experiments. **P < .01, ***P < .001, significantly different compared with the percentage of the dilation induced by ACh in the absence of any drugs. ##P < .01, ###P < .001, significantly different compared with the percentage of the dilation induced by ACh in the presence of L-NAME plus Indo.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Mechanisms of CPA-induced vasodilation resistant to L-NAME and Indo in rat small mesenteric arteries with myogenic tone at 80 mm Hg. A, concentration-response curves to CPA in small mesenteric arteries of the rat pressurized at 80 mm Hg in the absence () and in the presence of 200 µM L-NAME (), 10 µM Indo (), and L-NAME plus Indo (). B, concentration-response curves to CPA in small mesenteric arteries of the rat pressurized at 80 mm Hg in the presence of L-NAME plus Indo (), L-NAME plus Indo plus 30 nM IbTX (), L-NAME plus Indo plus 0.3 µM apamin (), and L-NAME plus Indo plus IbTX plus 0.3 µM apamin (). The points are mean ± S.E.M. (vertical bars) of five to seven experiments. *P < .05, ***P < .001, significantly different compared with the percentage of the dilation induced by CPA in the absence of any drugs. ##P < .01, ###P < .001, significantly different compared with the percentage of the dilation induced by CPA in the presence of L-NAME plus Indo. tttP < .001, significantly different compared with the percentage of the dilation induced by CPA in the presence of L-NAME plus Indo plus IbTX.

View this table: [in this window] [in a new window]   TABLE 2 Effects of K+-free PSS (0 mM KPSS), 40 mM KPSS, and the removal of endothelium (E) on the L-NAME plus Indo-resistant effects of 20 µM CPA and 1 µM ACh in small mesenteric resistance arteries of the rat pressurized at 80 mm Hg

EDHF Signaling. Blockade of the BK_Ca channel with IbTX did not abolish the L-NAME plus Indo-resistant vasodilation to ACh or CPA (Table 3, Figs. 5B and 6B). Similar results were obtained using other K⁺-channel blockers (TEA, ChTX, BaCl₂, 4-AP, glibenclamide; Table 3). Only apamin was effective in reducing the L-NAME plus Indo-resistant vasodilation induced by ACh (Figs. 3C and 5B) or CPA (Figs. 4B and 6B); the combination of apamin and IbTX resulted in near-complete inhibition of vasodilation (Figs. 4C and 6B).

8

7

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Cerebral arteries of the rat display a higher level of myogenic reactivity compared with mesenteric arteries of comparable size. Johnson (1980) proposed that the stimulus for myogenic tone is wall tension rather than hydrostatic pressure. According to this hypothesis, myogenic constriction decreases wall tension and thus reduces the stimulus for further constriction. As shown in Fig. 1, the greater myogenic constriction in cerebral arteries was able to maintain a lower wall tension than was observed in the mesenteric arteries. The mechanism underlying this difference may be related to a higher state of activation of L-type voltage-dependent Ca²⁺ channels in cerebral arteries compared with peripheral arteries (Asano et al., 1993). Furthermore, in the cerebral circulation, both large and small arteries are important resistance vessels (Faraci, 1991). The marked myogenic tone coupled with the parallel arrangement of cerebral vessels would promote regional regulation of flow with minimal overall changes in blood volume. In contrast in the mesenteric vascular bed, resistance is mainly determined by the small arteries, which are arranged in series with the conduit arteries.

In myogenically active cerebral arteries, BK and SP produced endothelium-dependent dilation, which was completely inhibited by L-NAME, indicating the release of NO. This finding is consistent with previous studies performed in different types of cerebral arteries (Faraci and Brian, 1994; Zimmermann et al., 1997). In basilar artery of rat, Hist and A23187 induced endothelium-dependent dilation (Faraci and Brian, 1994), which was not the case in the present study. These results might indicate that there are regional differences in endothelial function within the same vascular bed. Alternatively, the observed discrepancy may be related to differences in smooth muscle sensitivity to the vasoactive agents.

For comparison with middle cerebral arteries, we chose small mesenteric arteries of similar size, which may also contribute to peripheral resistance. However, reports on their endothelial function are sparse and somewhat controversial (Chen and Cheung, 1997; Weidelt et al., 1997). Our results show that all the vasoactive agents tested produced endothelium-dependent dilations of small mesenteric arteries. As in cerebral arteries, SP-induced vasodilation appeared to be mediated by NO. However, dilation in response to CPA was due to NO-independent pathways, whereas responses to ACh, BK, Hist, and A23187 were due to activation of both NO-dependent and -independent pathways.

The existence of an NOS/COX inhibitor-resistant vasodilation in small mesenteric resistance arteries suggests an alternative pathway, as reviewed elsewhere (Moubouli and Vanhoutte, 1997). Although direct electrophysiological measurements have not been recorded, the following data strongly support the release of EDHF in the ACh- and CPA- induced vasodilation.

First, CPA- and ACh-induced vasodilation remaining in the presence of L-NAME plus Indo is abolished by high K⁺ solution. This finding indicates that hyperpolarization is likely responsible for these CPA- and ACh-induced dilations. It has been shown previously that exposure to high K⁺ concentrations does not affect NO-mediated relaxations in rat small mesenteric arteries (Fukao et al., 1995; Chen and Cheung, 1997) and does not impair the increase in tissue cGMP content (Kuhberger et al., 1994). A good correlation between ACh-induced hyperpolarization and relaxation (measured simultaneously) exists in the rat mesenteric vascular bed (McPherson and Angus, 1991). The present results indicate that L-NAME plus Indo-resistant ACh- and CPA-induced dilation of small mesenteric arteries do not involve K_ATP, voltage-gated K⁺ channels, and K_ir. These results are in contrast to those obtained in different vascular preparations (Cowan et al., 1993; Plane and Garland, 1993).

Second, relaxation induced by ACh, which was independent of the NO/COX pathway, was unaffected by IbTX but was abolished by apamin. This result suggests the participation of SK_Ca in ACh-induced dilation. This is in accordance with previous studies conducted in larger mesenteric arteries (Murphy and Brayden, 1995; Chen and Cheung, 1997). However, apamin had no effect on ACh-induced EDHF production in guinea pig coronary arteries (Eckmann et al., 1992).

A previous investigation in rat mesenteric preparations reported that CPA and thapsigargin, both inhibitors of Ca²⁺ uptake by endothelial ER, are able to produce EDHF by increasing intracellular calcium concentration (Fukao et al., 1995). However, the nature of the K⁺ channels activated by EDHF is not known. Our results suggest a major contribution by SK_Ca in the CPA-induced dilation that is resistant to NOS and COX inhibition. Complete abolition of the dilation of arteries could be achieved only when the two types of K_Ca were blocked with apamin and IbTX. The current study provides evidence that BK_Ca and SK_Ca participate in the hyperpolarization and dilation responses induced by ACh and CPA on myogenically active small mesenteric resistance arteries of the rat. However, we cannot completely explain the difference effects of IbTX on dilations produced by ACh and CPA.

It appears that ACh is more similar to CPA than SP in terms of the relative quantities of NO and EDHF that may be generated. Moreover, the dynamics of intracellular calcium concentration ([Ca²⁺]_i) changes during relaxation induced by ACh and CPA may be quite similar (Wang et al., 1995; Rahimian et al., 1998); both vasodilators relaxed mesenteric arteries to the same extent in the presence of L-NAME plus Indo (Figs. 5B and 6B). Thus, it is likely that the difference observed between ACh and CPA in terms of the inhibitory effects of IbTX may be due to the release of more than one EDHF, and their cumulative effects are slightly different in the case of ACh. Different candidates for EDHF have been suggested (Mombouli and Vanhoutte, 1997; Edwards et al., 1998; Vanhoutte, 1998). Therefore, the release of different substances by these vasodilators (ACh and CPA) leads to activation of different types (or subtypes) of K⁺ channels; alternatively, differential mechanisms of EDHF-mediated K⁺ channel activation may explain the results.

Recent evidence suggests that EDHF could be a cytochrome P-450-derived arachidonic acid metabolite (Hecker et al., 1994).We used the suicide-substrate inhibitor 17-ODYA, two mechanistically different inhibitors of a large number of cytochrome P-450-dependent systems (SKF 525a, clotrimazole), and phospholipase A₂ inhibitors on the CPA-induced endothelium-dependent vasodilation. Our data strongly suggest that in the small mesenteric artery, EDHF is unlikely to be an epoxyeicosatrienoic acid. That is in accordance with similar results observed in guinea pig carotid arteries (Corriu et al., 1996). This finding contrasts, however, with observations made in other test systems (Hecker et al., 1994). The apparent contradictions in the literature may be due to liberation of different EDHFs as well as the different distributions of K⁺ channel types in various smooth muscle preparations.

It is known that an increase in [Ca²⁺]_i in the endothelial cells is an important stimulus not only for the formation of NO but also for the activation of the NO-independent pathways (Busse et al., 1993; Fukao et al., 1993). This seems to be true also in myogenically active small mesenteric arteries, because the agonists used in this study are able to activate NO-dependent and -independent pathways and are known to increase [Ca²⁺]_i in endothelial cells (Dinerman et al., 1993). The increase in [Ca²⁺]_i induced by agonists is due to inositol trisphosphate-induced release from the ER as well as by extracellular Ca²⁺ influx. A23187 increases Ca²⁺ permeability of both cell and ER membranes, whereas CPA depletes intracellular Ca²⁺ stores by selectively inhibiting sarcoplasmic/ER Ca²⁺-dependent ATPase in a variety of tissues (Seidler et al., 1989; Lagaud et al., 1999). The striking difference between CPA, which selectively releases EDHF, and SP, which releases only NO (no EDHF), strongly suggests that the mode of Ca²⁺ elevation determines the nature of the secretory product. Hence, we postulate that CPA, by slowly releasing Ca²⁺ from the ER through inhibition of Ca²⁺ reuptake by sarcoplasmic/ER Ca²⁺-dependent ATPase, stimulates synthesis of EDHF. On the other hand, SP may increase [Ca²⁺]_i mainly by activating Ca²⁺ influx and perhaps Ca²⁺ release from the superficial ER (Sharma and Davis, 1995), thus stimulating endothelial NOS localized near the cell membrane (Pollock et al., 1993). On the other hand, we found that ACh was closer to CPA than SP in terms of the relative quantities of NO and EDHF. We have shown previously that CPA and ACh are able to raise [Ca²⁺]_i in endothelial cells by releasing Ca²⁺ from intracellular stores and activating Ca²⁺ influx through receptor-operated and store-operated channels (Wang et al., 1995; Rahimian et al., 1998). Thus, although we intend to further investigate specific [Ca²⁺]_i patterns, it is possible that those due to ACh and CPA may be quite similar. Clearly, further investigation providing more precise information on the exact mechanisms are necessary to resolve this issue.

In summary, we have shown that cerebral arteries of the rat display a higher level of pressure-dependent active tone compared with mesenteric arteries, suggesting that the myogenic responsiveness depends in part on the vascular bed from which the vessel is isolated. We also showed that the mechanisms of endothelium-dependent agonist responses differ between the distal middle cerebral and small mesenteric arteries of the rat. In addition, we suggest that the pathway by which agonists raise [Ca²⁺]_i may determine the nature of the endothelial secretory product. Heterogeneity of vascular reactivity is important for the plasticity of physiological responses. Adequate perfusion of individual vascular beds depends on heterogeneity in the responsiveness of vessels of different size, and this report supports the concept that this is at least partly due to variability in endothelial signaling.