INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

RECENT STUDIES (6, 7, 36) have indicated that nitric oxide (NO) plays an important role in the regulation of cerebral blood flow and mediates the cerebral vascular responses to a wide variety of stimuli, including acetylcholine (ACh), substance P, bradykinin, and ₂-adrenergic receptor agonists. Inhibitors of NO synthase constrict cerebral arteries in vitro and decrease cerebral blood flow in vivo (10, 14, 27). There is also evidence (10, 11) that impairment in NO-induced vasodilation plays an important role in the pathogenesis of cerebral vasospasm.

Despite the importance of NO in the control of cerebral vascular tone, its mechanism of action particularly in the middle cerebral artery (MCA) of the rat remains to be fully defined. The results of previous in vivo studies indicating that 1H-[1,2,4]oxadiazole[4,3-a] quinoxalin-1-one (ODQ), an inhibitor of guanylyl cyclase (GC), blocks 80% of the vasodilator response to ACh and NO in pial arteries of the mouse, rat, and rabbit (4, 9, 28) and in the basilar artery of the rat (1, 29) in vivo support a primary role for cGMP in mediating the response to NO. The potential mechanisms by which cGMP contributes to the vasodilator response to NO in vascular smooth muscle (VSM) cells have been reviewed by Lincoln et al. (16). They include activation of Ca²⁺-sensitive K⁺ (K_Ca) channels (15) and a reduction in intracellular Ca²⁺ concentration ([Ca²⁺]_i) either by stimulation of Ca²⁺ reuptake into intracellular Ca²⁺ stores (37), blockade of Ca²⁺ influx (23), and/or by inhibition of D-myo-inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃]-mediated Ca²⁺ release (26). In addition, cGMP has also been shown to desensitize the contractile mechanism to elevations of [Ca²⁺]_i in VSM (25). Many of these mechanisms have been examined in the basilar artery of rabbits (24), dogs (31), sheep (18), guinea pigs (35), and even rats (23). However, few of these studies have examined multiple mechanisms in the same vessel and none have been done in the MCA of the rat. This latter point is potentially significant, because we (1, 33) and other researchers (19, 20) have noted that the mechanisms underlying the vasodilator response to NO appear to differ in the basilar artery and MCA of the rat. The vasodilator response to NO is completely cGMP dependent in the basilar artery of the rat (1, 29) and guinea pig (22) and is not affected by K_Ca channel blockers (27). However, ~50% of the response to NO in MCA of rats is cGMP independent and is secondary to blockade of the formation of 20-hydroxyeicosatetraenoic acid (20-HETE) and activation of K_Ca channels (1, 33). Furthermore, in patch-clamp studies, blockade of GC with ODQ has no effect on the ability of NO to activate K_Ca channels in MCA of rats (33). These findings suggest that the vasodilator response to NO is mediated by different pathways in the MCA and basilar arteries. Given the many mechanisms that may impact on the vasodilator response to NO, and the fact that different pathways seem to underlie the vasodilator response to NO in basilar arteries and MCA of rats, it was therefore critical to determine the mechanisms that contribute to the cGMP-dependent vasodilator response to NO in the MCA of the rat. Therefore, the present study examined the following: 1) the contribution of K⁺ channels to the vasodilator responses to NO and cGMP in MCA of the rat, 2) the effects of NO and cGMP on intracellular Ca²⁺ levels in VSM cells isolated from the MCA, and 3) the effects of NO and cGMP on the vasoconstrictor response to increases in [Ca²⁺]_i in MCA permeabilized with -toxin and ionomycin.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Experiments were performed on 10- to 12-wk-old male Sprague-Dawley rats purchased from Harlan Sprague Dawley (Indianapolis, IN). The rats were housed in an Animal Care Facility at the Medical College of Wisconsin that is approved by the American Association for the Accreditation of Laboratory Animal Care. All protocols involving animals received approval by the Animal Care Committee of the Medical College of Wisconsin.

Isolated vessel studies. Rats were anesthetized with pentobarbital sodium (50 mg/kg ip), the brain was removed, and small branches of the MCA (inner diameter <100 µm) were microdissected. The MCA branches were then mounted on glass micropipettes under a microscope in a chamber filled with physiological saline solution (PSS) containing (in mM) 119 NaCl, 4.7 KCl, 1.17 MgSO₄, 1.6 CaCl₂, 12 NaHCO₃, 1.18 NaH₂PO₄, 0.03 EDTA, 10 glucose, and 10 HEPES (pH 7.4). The bath was bubbled with a 95% O₂-5% CO₂ gas mixture and maintained at 37°C. The vessels were secured to the pipettes with 10-0 silk suture, and intraluminal pressure was maintained at 80 mmHg during the experiment. Vascular inner diameters were measured with a stereomicroscope, a charge-coupled device television camera (model KP-130AU, Hitachi), and video measurement system (model VIA-100, Boeckeler Instrument; Tucson, AZ).

Role of K+ channels in the vascular responses to NO and cGMP. The contribution of K⁺ channels to the vasodilator response to NO and cGMP in rat MCA was assessed by comparing the response to different concentrations of diethylaminodiazen-1-ium-1,2-dioate (DEA-NONOate; 10⁹ to 10⁵ M) and 8-bromo-cGMP (8-BrcGMP; 10⁸ to 10⁴ M) in vessels preconstricted with serotonin (5-HT; 10⁷ M) or a depolarizing concentration of KCl (80 mM). The role that activation of K_Ca channels plays in the vasodilator response was assessed by comparing the response to DEA-NONOate or 8-BrcGMP before and after addition of the K_Ca channel blocker iberiotoxin (IbTX) (10⁷ M) or a depolarizing concentration of K⁺ to the bath. The dose of IbTX used in the present study is based on previous patch-clamp data (41) and functional data (43), indicating that this concentration of IbTX completely blocks K_Ca channels in VSM cells and isolated perfused arterioles.

Effects of NO and cGMP on Ca²+ sensitivity in MCA. These experiments were performed using MCA mounted in a vessel myograph that were treated with -toxin (10 µg/ml) and ionomycin (10⁵ M) to allow for equilibration of the [Ca²⁺]_i and extracellular [Ca²⁺]. To eliminate the possibility that changes in bath [Ca²⁺] might alter vascular tone by releasing endothelial factors, the endothelium was first removed by perfusion of the vessels with PSS containing an antibody raised against the von Willebrand Factor (1:1,000 dilution) and guinea pig complement (2%) for 10 min, followed by a 20-min wash with normal PSS (12). Endothelial function was tested in each experiment by measuring the vasodilator response to addition of ACh (10³ M) to the bath after preconstriction of the vessel with 5-HT (10⁷ M). After being tested for functional removal of the endothelium, the vessels were bathed with a low-Ca²⁺ (10⁸ M) cytoplasmic substitution solution (CSS) containing (in mM) 100 potassium propionate, 4 MgCl₂, 20 piperazine-N,N'-bis(2-ethanesulfonic acid), 4 Na₂ATP, 10 creatinine phosphate, 0.1 mg/ml creatinine phosphokinase, and 2 EGTA, pH 7.1. Staphylococcal -toxin (10 µg/ml) and ionomycin (10⁵ M) were added to the bath to permeabilize the vessel to Ca²⁺. After 30 min, the vessels were washed with fresh low-Ca²⁺ CSS. The degree of permeabilization was tested by comparison of the maximal vasoconstrictor response to addition of Ca²⁺ (10⁵ M) to the bath versus the vasoconstrictor response to a depolarizing concentration of KCl (80 mM) before the vessels were treated with ionomycin and -toxin.

7

5

6

4

5

Study on [Ca²+]_i in VSM cells. VSM cells were isolated from rat MCA by incubating the vessels in a low-Ca²⁺ Tyrode solution containing (in mM) 145 NaCl, 1 MgCl₂, 4 KCl, 0.05 CaCl₂, 10 glucose, and 10 HEPES (pH 7.4) with 1.5 mg/ml papain (14 U/mg), and 1 mg/ml dithiothreitol for 15 min at 37°C. The vessels were then spun down and transferred to a low-Ca²⁺ Tyrode solution composed of (in mg/ml) 0.5 elastase (90 U/ml), 1 soybean trypsin inhibitor (10,000 U/ml), and 2 collagenase (196 U/ml) for 20 min at 37°C. The supernatant was collected, and the cells were pelleted by centrifugation at 500 g for 1 min, resuspended in fresh low-Ca²⁺ Tyrode solution, and stored at 4°C.

Effect of NO and cGMP on [Ca²+]_i in VSM cells permeabilized with -toxin and ionomycin. These experiments examined whether the inhibitory actions of cGMP and NO on the vasoconstrictor response to Ca²⁺ in permeabilized MCA were associated with changes in [Ca²⁺]_i. During the control period, the cells were bathed with a CSS solution containing a Ca²⁺ concentration of 10⁶ M, and the [Ca²⁺]_i response to DEA-NONOate (10⁶ M) or 8-BrcGMP (10⁴ M) was studied. The cells were then permeabilized with ionomycin (10⁵ M) and -toxin (10 µg/ml), and the experiment was repeated.

Effect of NO and cGMP on [Ca²⁺]_i in VSM cells stimulated with 5-HT. These experiments examined whether cGMP and NO alter [Ca²⁺]_i in VSM cells isolated from the MCA of the rat after stimulation with 5-HT. [Ca²⁺]_i were measured under control condition and after 5-HT (10⁵ M) was added to the bath. 5-HT produced a transient increase in [Ca²⁺]_i, followed by a steady-state increase. After [Ca²⁺]_i reached a stable plateau value, DEA-NONOate (10⁶ M) or 8-BrcGMP (10⁴ M) was added to the bath and changes in [Ca²⁺]_i were monitored for an additional 10 min.

Role of K_Ca channels in the change in [Ca²+]_i induced by NO in 5-HT-stimulated VSM cells. Because DEA-NONOate decreased [Ca²⁺]_i in VSM cells stimulated with 5-HT, the contribution of activation of K_Ca channels to the fall in [Ca²⁺]_i produced by NO was examined. In this experiment, IbTX (10⁷ M), a selective K_Ca channel blocker, was added to the bath before administration of 5-HT and DEA-NONOate.

Drugs and chemicals. All chemicals were of analytic grade. Collagenase type II was purchased from Worthington (Freehold, NJ). DEA-NONOate and 8-BrcGMP were purchased from Calbiochem (La Jolla, CA), ODQ was obtained from Alexis (San Diego, CA), ionomycin was from Biomol (Plymouth Meeting, PA), and fura 2-AM and pluronic acid were purchased from Molecular Probes (Eugene, OR). All other chemicals used in this study were purchased from Sigma (St. Louis, MO).

Statistics. Values are presented as means ± SE. The significance of differences in mean values between and within groups in the isolated vessel studies was examined using analysis of variance for repeated measurements, followed by the Duncan's multiple-range test. A paired t-test was used to examine the significance of differences in [Ca²⁺]_i in the studies using isolated VSM cells. P < 0.05 was considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Role of K+ channels in the vasodilator response to NO and cGMP. Under control conditions, DEA-NONOate (10⁹ to 10⁵ M) dose dependently dilated MCA preconstricted with 5-HT (10⁷ M) to 79 ± 4% of control (Fig. 1). Preconstricting the vessels with a depolarizing concentration of KCl (80 mM) or addition of the selective K_Ca channel blocker IbTX (10⁷ M) to the bath significantly impaired the vasodilator response to the NO donor by 50% (Fig. 1A). In contrast, blocking K⁺ channel activity with 80 mM KCl or IbTX had little effect on the vasodilator response to 8-BrcGMP (10⁴ M) (Fig. 1B).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Role of K+ channels in vasodilator response to nitric oxide (NO) and 8-bromo-cGMP (8-BrcGMP) in the middle cerebral artery (MCA) of rats. A: cumulative concentration-response curves depicting the effect of diethylaminodiazen-1-ium-1,2-dioate (DEA-NONOate) on the inner diameter of rat MCA preconstricted with serotonin (107 M) before and after blocking Ca2+-sensitive K+ (KCa) channels with iberiotoxin (IbTX) (107 M) or by adding a depolarizing concentration of KCl (80 mM) to the bath. Control inner diameter of these vessels averaged 108 ± 3 µm. B: cumulative concentration-response curves depicting the effect of 8-BrcGMP on the inner diameter of rat MCA before and after blocking KCa channel with IbTX (107 M) or by adding a depolarizing concentration of KCl (80 mM) to the bath. Control inner diameter of these vessels averaged 110 ± 4 µm. Numbers in parentheses indicate the number of vessels studied. * P < 0.05, significant difference compared with the corresponding control value.

Effect of NO and cGMP on the vasoconstrictor response to Ca²+ in MCA permeabilized with -toxin and ionomycin. Baseline diameter of the MCA used in these studies averaged 110 ± 4 µm. The diameter of these vessels fell to 51 ± 3 µm after addition of 5-HT to the bath and rose to 108 ± 5 µm after the administration of ACh. Removal of the endothelium markedly impaired the vasodilator response to ACh. Vessel diameters averaged 109 ± 4, 50 ± 1, and 52 ± 4 µm, respectively, before and after the addition of 5-HT and ACh to the bath. The control vasoconstrictor responses to elevations in bath [Ca²⁺] in deendothelialized MCA subsequently treated with DEA-NONOate or 8-Br-cGMP were similar. Therefore, the results from these two groups were pooled and presented together in Fig. 2. Under control conditions, an elevation in bath [Ca²⁺], from 10⁸ to 10⁵ M, reduced the diameter of the MCA by 76 ± 5%. DEA-NONOate (10⁶ M) had no significant effect on the baseline diameter of permeabilized MCA measured at a bath [Ca²⁺] of 10⁸ M (136 ± 17 vs. 135 ± 17 µm before and after DEA-NONOate, respectively). However, it did reduce the vasoconstrictor response to an elevation in bath [Ca²⁺] to 10⁵ M by 55%. Similarly, 8-BrcGMP (10⁴ M) had no significant effect on the baseline diameter of permeabilized MCA, but it also reduced the vasoconstrictor response to the elevation in bath [Ca²⁺] to 10⁵ M by 60%. The inhibitory effect of DEA-NONOate on the vasoconstrictor response to changes in bath [Ca²⁺] was completely blocked by addition of the GC inhibitor ODQ (10⁵ M) to the bath.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   The effects of NO and cGMP on the vasoconstrictor response of rat MCA that were permeabilized with -toxin and ionomycin to elevations in bath Ca2+ concentration ([Ca2+]). Paired experiments were performed under control conditions in high-K+ bath solution before and after addition of DEA-NONOate (106 M), 8-BrcGMP (104 M), or DEA-NONOate (106 M) plus 1H-[1,2,4]oxadiazole[4,3-a]quinoxalin-1-one (ODQ; 105 M) to the bath. Results are expressed as %reduction in the diameter of the permeabilized vessels measured at a bath [Ca2+] of 108 M. Control inner diameter of these vessels averaged 107 ± 4 µm. * P < 0.05, significant difference compared with the corresponding control values. #P < 0.05, significant difference compared with the corresponding values seen in vessels treated with DEA-NONOate. Numbers in parentheses indicate the number of vessels studied.

Effect of NO and cGMP on [Ca²+]_i in VSM cells permeabilized with -toxin and ionomycin. The results of these experiments are presented in Figs. 3 and 4. Under control conditions (Fig. 3A), DEA-NONOate (10⁶ M) had no significant effect on [Ca²⁺]_i in VSM cells isolated from MCA of rats. Mean [Ca²⁺]_i averaged 98 ± 11 nM before and 88 ± 7 nM 2 min after the addition of DEA-NONOate to the bath. [Ca²⁺]_i increased rapidly from 88 ± 11 nM to 1,580 ± 211 nM after permeabilizing the cells with ionomycin (10⁵ M) and -toxin (10 µg/ml) in a CSS solution with a 10⁵ M free [Ca²⁺] (Fig. 3B). DEA-NONOate (10⁶ M) had no significant effect on [Ca²⁺]_i under these experimental conditions. Similarly, 8-BrcGMP (10⁴ M) had no significant effect on [Ca²⁺]_i in the VSM cells before (72 ± 7 vs. 67 ± 7 nM; Fig. 4A) or after (1,047 ± 98 vs. 1,043 ± 99 nM; Fig. 4B) the cells were permeabilized with ionomycin and -toxin. To exclude the possibility that the lack of effects of DEA-NONOate and 8- BrcGMP on [Ca²⁺]_i was due to an inability of our system to detect decreases in [Ca²⁺]_i, the cells were also treated with EDTA (10 mM). As shown in Fig. 4B, the addition of EDTA to the bath markedly reduced [Ca²⁺]_i in the permeabilized VSM cells to values close to 0 nM.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Effect of NO on the intracellular [Ca2+] ([Ca2+]i) in VSM cells isolated from the MCA of rats. Top: representative Ca2+ response of a vascular smooth muscle (VSM) cell exposed to DEA-NONOate (106 M) under control conditions (A) and after the cell was permeabilized with -toxin and ionomycin (B). Bottom: summary of the mean effects of DEA-NONOate on [Ca2+]i in control VSM cells and VSM cells permeabilized with -toxin and ionomycin. Con, control; NO, 2-min after addition of DEA-NONOate to the bath.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of cGMP on the [Ca2+]i in VSM cells isolated from the MCA of rats. Top: representative Ca2+ response of a VSM cell exposed to 8-BrcGMP (104 M) under control conditions (A) and after the cell was permeabilized with -toxin and ionomycin (B). Bottom: summary of the mean effects of 8-BrcGMP on [Ca2+]i in control VSM cells and VSM cells permeabilized with -toxin and ionomycin. cGMP, 2 min after addition of 8-BrcGMP to the bath. To exclude the possibility that the lack of effects of DEA-NONOate and cGMP on [Ca2+]i was due to an inability of our system to detect decreases in [Ca2+]i, the response to addition of EDTA (10 mM) to the bath was studied at the end of each experiment.

Effect of NO and cGMP on [Ca²+]_i in VSM cells stimulated with 5-HT. The results of these experiments are presented in Figs. 5 and 6. Stimulation of the VSM cells with 5-HT produced a transit increase in [Ca²⁺]_i from 90 ± 7 to 211 ± 27 nM, followed by a sustained plateau phase, in which [Ca²⁺]_i remained ~15% above control (Figs. 5 and 6). Addition of 8- BrcGMP to the bath during the plateau phase had no significant effect on [Ca²⁺]_i in VSM cells isolated from the MCA that were stimulated with 5-HT (Fig. 5). In contrast, the NO donor DEA-NONOate reduced [Ca²⁺]_i by 16 ± 1% (Fig. 6). Pretreatment of the cells with IbTX potentiated the steady-state rise in [Ca²⁺]_i produced by 5-HT from 11 ± 2 to 20 ± 3% above control (Fig. 6). Furthermore, blockade of K_Ca channels also significantly reduced the fall in [Ca²⁺]_i from 16 ± 1 to 11 ± 1% after administration of DEA-NONOate to the bath (Fig. 6).

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Effect of cGMP on [Ca2+]i in VSM cells stimulated with serotonin (5-HT). A: representative Ca2+ response to 8-BrcGMP (104 M) in a VSM cell stimulated with 5-HT. B: summary of the mean effects of 8-BrcGMP on [Ca2+]i in all cells studied. * P < 0.05 compared with baseline value.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Effect of NO on [Ca2+]i in VSM cells stimulated with 5-HT. A: representative Ca2+ response to DEA-NONOate (106 M) in a control VSM cell stimulated with 5-HT. B: representative Ca2+ response to DEA-NONOate (106 M) treated with iberiotoxin IbTX (107 M) and stimulated with 5-HT. Bottom: summary of the mean effects of DEA-NONOate on [Ca2+]i in control and IbTX-treated VSM cells. Control, cells stimulated with 5-HT only; IbTX, cells treated with IbTX and 5-HT. * P < 0.05 compared with baseline value; #P < 0.05 compared with plateau phase in same group. Numbers in parentheses indicate the number of cells studied.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Previous studies (5, 17, 39) have indicated that the vasodilator response to NO is mediated by cGMP-dependent and -independent pathways (3, 8). In renal arterioles (32) and MCA (17, 19, 33) of the dog and rat, activation of K_Ca channels has been previously reported to contribute ~50% to the vasodilator response to NO. The increase in K_Ca channel activity appears to be mediated by inhibition of the formation of 20-HETE, rather than to an elevation in cGMP levels, because preventing the NO-induced fall in 20-HETE levels by adding it to the bath abolishes the ability of a NO donor to activate K_Ca channels and attenuates the vasodilator response in rat renal arterioles and MCA (32, 33). In addition, blockade of GC with ODQ has no effect on the activation of K⁺ channels produced by NO in VSM cells isolated from rat renal arterioles and MCA (32, 33). Nevertheless, ODQ still reduces the vasodilator response to NO by 50% in rat renal and cerebral arteries (1, 17, 33). However, the vasodilator response to NO in basilar arteries of the rat (1, 29) and guinea pig (22) can be completely blocked by inhibition of GC. These observations suggest that the pathways underlying the vasodilator response to NO are different in the basilar artery and MCA and that cGMP contributes to the vasodilator response to NO in rat MCA via a mechanism that is independent of activation of K_Ca channels.

In the present study, we first confirmed that blockade of K⁺ channels with a depolarizing concentration of KCl (80 mM) or K_Ca channels with IbTX decreased the vasodilator response to NO by 50% in rat MCA under the present experimental conditions (Fig. 1). This observation fits with previous results, indicating that activation of K⁺ channels secondary to inhibition of the formation of 20-HETE contributes ~50% to the vasodilator response to NO in the MCA of the rat (1, 33). These results are also consistent with the results of other studies demonstrating that activation of K_Ca channels contributes to the vasodilator response to NO in other vascular beds, including cerebral arteries studied in vitro (2, 3, 17, 21, 24, 32, 33). However, it is important to recognize that there is not unequivocal support for a role for K_Ca channels in mediating the response to NO. Indeed, studies in rat MCA (33), basilar arteries (29), pulmonary arteries (42), and bovine coronary arteries (15) have indicated that NO activates 4-aminopyridine-sensitive K⁺ channels. Blockade of K_Ca channels with IbTX had little effect on the vasodilator response to NO and ACh in pial arteries of rabbits (34) or the basilar artery of the rat (27) studied in vivo. The vasodilator responses to sodium nitroprusside, 8-BrcGMP and ACh in these studies were attenuated by the blockade of voltage-sensitive K⁺ (K_v) channels with 4-aminopyridine. These studies indicate that in some vessels, activation of K_v channels contributes to the vasodilator response to NO, whereas in others, it involves activation of K_Ca channels. The reason for the differences in these results remains to be determined but may reflect species differences or differences in the types or density of K⁺ channels expressed in VSM cells in different vessels.

In contrast to what was observed after administration of the NO donor in the present study, the vasodilator response to 8-BrcGMP in MCA was not affected by blockade of K_Ca channels with IbTX (10⁷ M) or by the addition of a depolarizing concentration of K⁺ to the bath. These observations are consistent with previous patch-clamp data (33), indicating that blockade of GC with ODQ does not alter the NO-induced activation of K⁺ channels in VSM cells isolated from rat MCA. Thus cGMP must contribute to the vasodilator response to NO in the MCA of the rat by some other mechanism. To evaluate this hypothesis further, we compared the effects of DEA-NONOate and 8-BrcGMP on the vasoconstrictor response to elevations in bath Ca²⁺ in MCA permeabilized with ionomycin and -toxin. We found that the NO donor DEA-NONOate (10⁶ M) produced a rightward shift of the concentration-constriction curve for Ca²⁺ (Fig. 2). In addition, the inhibitory effect of DEA-NONOate on Ca²⁺-induced vasoconstriction was not associated with a change in [Ca²⁺]_i in VSM cells isolated from the MCA of rats that were permeabilized with ionomycin and -toxin (Fig. 3). These findings suggest that the effect of NO to reduce the vasoconstrictor response to Ca²⁺ in permeabilized rat MCA is not mediated by a reduction in [Ca²⁺]_i, but rather is due to a decrease in the contractile response to an elevation in [Ca²⁺]_i. The effects of NO on the vasoconstrictor response to Ca²⁺ were completely blocked by ODQ (Fig. 2), indicating that this effect is cGMP dependent. We also found that 8-BrcGMP mimicked the response to NO in permeabilized rat MCA and decreased the vasoconstrictor response to elevations in [Ca²⁺]_i (Fig. 2). 8-BrcGMP also failed to change [Ca²⁺]_i in MCA VSM cells permeabilized with ionomycin and -toxin (Fig. 4). Moreover, 8-BrcGMP had no effect on [Ca²⁺]_i in VSM cells isolated from the MCA when studied under control condition or after the cells were stimulated with 5-HT (Fig. 5). These results indicate that cGMP-induced attenuation of the vasoconstrictor response to elevated [Ca²⁺]_i in permeabilized MCA is not mediated by a decrease in [Ca²⁺]_i, but rather by a decrease in the contractile response to an elevation in [Ca²⁺]_i. In contrast to the results obtained with cGMP, we found that addition of the NO donor significantly reduced [Ca²⁺]_i in VSM cells stimulated with 5-HT (Fig. 6) and the decrease in [Ca²⁺]_i was attenuated, but not eliminated, by blockade of K_Ca channels with IbTX (Fig. 6). These findings are consistent with the view that NO-induced activation of K_Ca channels contributes to the fall in [Ca²⁺]_i by hyperpolarizing the cell membrane and reducing Ca²⁺ influx through voltage-gated Ca²⁺ channels (23). They also indicate that other mechanisms contribute to the decrease in [Ca²⁺]_i produced by NO in these vessels, because blockade of K_Ca channels did not completely block the NO-induced fall in [Ca²⁺]_i. These mechanisms include activation of other type of K⁺ channels (29, 33, 34) and/or stimulation of Ca²⁺ reuptake into intracellular stores as reported by Twort and van Breemen (37).

The mechanism by which cGMP decreases the vasoconstrictor response to elevations in [Ca²⁺]_i in the rat MCA remains to be determined. Contraction of VSM is dependent on the phosphorylation and dephosphorylation of the myosin light chain protein (30). Phosphorylation of the myosin light chain is mediated by Ca²⁺-calmodulin activated myosin light-chain kinase, and dephosphorylation depends on myosin light chain phosphatase. Thus a decrease of the activity of Ca²⁺-calmodulin activated myosin light-chain kinase and/or an increase in the activity of myosin light chain phosphatase could contribute to the decrease in the contractile response to an elevation of [Ca²⁺]_i. Indeed, several investigators (13, 25, 38, 40) have reported that cGMP activates protein kinases that phosphorylate and modify the activity of both of these enzymes. Thus it is likely that both mechanisms contribute to the cGMP-induced fall in the vasoconstrictor response to elevations in [Ca²⁺]_i in rat MCA.

In summary, the present results indicate that the vasodilator response to NO in the MCA of the rat is mediated by activation of K_Ca channels (cGMP independent) and cGMP-dependent pathways and the cGMP-dependent pathway is mediated by an effect of cGMP to decrease the contractile response to elevations in [Ca²⁺]_i, rather than by activation of K⁺ channels or a fall in [Ca²⁺]_i.