INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

LOCAL CALCIUM TRANSIENTS or "Ca²⁺ sparks" have recently been measured in smooth muscle cells from cerebral arteries (28). Ca²⁺ sparks in arterial myocytes as well as in cardiac myocytes (6) appear to arise from the opening of one or the coordinated opening of several tightly clustered ryanodine-sensitive Ca²⁺ release (RyR) channels in the sarcoplasmic reticulum (SR). In arterial myocytes, Ca²⁺ sparks activate 10-100 nearby sarcolemmal Ca²⁺-sensitive K⁺ (K_Ca) channels to cause an outward K⁺ current (17, 28), previously referred to as a "spontaneous transient outward current" (STOC) (4). Ca²⁺ sparks, themselves, contribute little to overall cellular Ca²⁺ in pressurized cerebral arteries, since their activity is asynchronous (~1/s in a cerebral artery myocyte) (28). Ca²⁺ spark-activated K_Ca channels can continuously regulate, as a negative feedback element, the membrane potential of smooth muscle cells in intact pressurized cerebral arteries (7). Thus either blocking Ca²⁺ sparks or decreasing their frequency of occurrence (5) would lead to membrane potential depolarization and vasoconstriction through a decrease in K_Ca channel activity (Ref. 28; see also Refs. 7 and 31). Therefore, these results predict that increasing Ca²⁺ spark frequency could lead to vasodilation through activation of K_Ca channels.

The cyclic nucleotides, cAMP and cGMP, mediate vasodilation of important endogenous and therapeutic agents (e.g., -adrenergic agents, adenosine, calcitonin gene-related peptide, nitric oxide, atrial natriuretic factor, synthetic nitrovasodilators). Diverse mechanisms have been proposed to explain the vasodilator effect of cyclic nucleotides, including the stimulation of Ca²⁺ uptake by the SR (24), the direct activation of K_Ca channels (39), activation of ATP-sensitive K⁺ channels (18, 19, 36, 48), activation of voltage-dependent K⁺ channels (1), and changes in the Ca²⁺ sensitivity of the contractile process (11, 15, 41). Relaxation of cerebral and coronary arteries to agents that elevate cAMP or cGMP are inhibited by blockers of K_Ca channels (29, 32, 33, 35, 42, 47). In fact, in a number of other types of smooth muscle (airway, mesenteric, uterine, pulmonary), blockers of K_Ca channels (iberiotoxin or charybdotoxin) reduce significantly the relaxing effects of agents that elevate cAMP and cGMP (2, 3, 8, 16, 23). K_Ca channels can be activated by cAMP-dependent protein kinase (PKA) and cGMP-dependent protein kinase (PKG) (39, 44). Recent evidence suggests that K_Ca channels in pressurized arteries with tone are primarily activated by local Ca²⁺ transients, i.e., Ca²⁺ sparks, and not directly by average intracellular Ca²⁺ (28). We therefore explored the possibility that cAMP and cGMP can increase K_Ca channel activity by affecting Ca²⁺ sparks. Specifically, our results provide direct evidence that cAMP and cGMP increase the frequency of local Ca²⁺ release events (Ca²⁺ sparks) from the SR, which activate K_Ca channels. Furthermore, we provide evidence that cAMP can act, in part, to dilate pressurized arteries through the Ca²⁺ spark  K_Ca channel pathway.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Cell isolation. Single smooth muscle cells were enzymatically isolated from cerebral (basilar) and coronary (septal) arteries of rats. The cell isolation procedure was slightly modified from that previously described for cerebral (37) and coronary (38) arteries. Briefly, an artery was dissected out and placed in ice-cold Ca²⁺-free isolation solution of the following composition (in mM): 60 NaCl, 85 sodium glutamate, 5.6 KCl, 2 MgCl₂, 10 glucose, and 10 HEPES (pH 7.4). After 10 min, the artery was transferred to the first of two enzyme solutions, with the same ionic composition as the Ca²⁺-free isolation solution described above, containing 1 mg/ml albumin, 0.5 mg/ml papain (Worthington, Freehold, NJ), and 1 mg/ml dithioerythritol, and was digested for 25 min at 37°C. The tissue was then transferred into isolation solution containing 0.1 mM CaCl₂ and 1 mg/ml of a collagenase mixture of 30% type H + 70% type F (Sigma) and was digested for another 10 min at 37°C. The tissue was then washed twice in Ca²⁺-free fresh isolation solution without enzymes for 10 min. Single smooth muscle cells were obtained from the digested artery by gentle trituration with a polished wide-bore pipette. After trituration, cells were stored in the same solution at 4°C, to be used the same day. Cells were left to stick to the coverslip in the experimental chamber for 15-20 min.

Ca²⁺ spark measurements. The procedure for the measurement of sparks is described in Ref. 28. Briefly, the cells were loaded with the Ca²⁺-sensitive indicator fluo 3 by a 20-min incubation in 5 µM fluo 3-AM and 2.5 µg/ml pluronic acid (Molecular Probes, Eugene, OR) followed by a 20-min wash. All measurements were made within 30 min of the end of the wash. The cells were scanned with a Bio-Rad 1000 laser scanning confocal microscope, which was housed in the University of Vermont Cell Imaging Facility. The resolution was 0.4 µm × 0.4 µm (x and y) × 0.8 µm (z) (28). Images were acquired using the line scan mode of the confocal microscope; this mode repeatedly scans a single line through a cell. A scan duration of 6 ms was used. Cells were positioned so that the line would traverse the long axis of the cell to detect sparks occurring in as much of the cell volume as possible. The average length of the cell through which the scan line passed was 36 µm. Scan lines are displayed vertically, and each line is added to the right of the preceding line to form the line scan image. In these images, time passes in the horizontal direction running from left to right, and position along the scan line is given by the vertical displacement. To minimize the possibility of laser damage affecting the Ca²⁺ handling of the cells, control and test measurements were made from different cells, with each cell being scanned for the same duration, 18 s total (6 consecutive line scan images of 512 lines, at 6 ms/line, were recorded from each cell along a single line). Sparks were analyzed using custom-written (IDL, Research Systems, Boulder, CO) analysis programs. To test the significance in the changes of Ca²⁺ spark frequency, a one-way ANOVA on ranks (Student-Newman-Keuls) method was used.

Conventional Ca²⁺ imaging. Isolated smooth muscle cells and intact arteries were loaded with the Ca²⁺ indicator dye fura 2. Cells were incubated with 0.25 µM fura 2-AM for 15 min. Cells were then washed and allowed to sit for 20 min before measurements were made. Arteries were loaded by incubation with 2 µM fura 2-AM for 45 min. Arteries were then cannulated and continuously superfused with physiological salt solution at 37°C. Ca²⁺ was measured ratiometrically (340:380 nm) using the IMAGE-1/FL quantitative fluorescence measurement program (Universal Imaging, West Chester, PA).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Forskolin, SNP, and nicorandil increase Ca²⁺ spark frequency. To test the hypothesis that increases in cAMP and cGMP can increase the frequency of Ca²⁺ sparks, the effects of forskolin (an activator of adenylyl cyclase that elevates cAMP), SNP, and nicorandil (nitric oxide donors, which elevate cGMP) on Ca²⁺ sparks were examined in single smooth muscle cells isolated from rat cerebral and coronary arteries. Forskolin significantly increased (P < 0.05) the Ca²⁺ spark frequency 2.3-fold (Fig. 1, A and B), as measured with a laser scanning confocal microscope and the Ca²⁺ fluorescence indicator fluo 3. This increase was prevented by an inhibitor of PKA, H-89 (1 µM), which binds to the active center of the catalytic subunit and suppresses the kinase activity (Fig. 1, A and B). The nitrovasodilators, SNP and nicorandil, also increased Ca²⁺ spark frequency 2.2- and 2.6-fold, respectively (Fig. 1C; P < 0.05). Forskolin, H-89, or SNP did not cause statistically significant changes in Ca²⁺ spark amplitude or rate of decay. Nicorandil did, however, cause a small but significant increase in Ca²⁺ spark amplitude (8% increase). If, indeed, Ca²⁺ sparks can activate sarcolemmal K_Ca channels, then forskolin, SNP, and nicorandil should increase the frequency of STOCs in these preparations, as had been suggested to occur in other preparations (9, 22).

View larger version (53K): [in this window] [in a new window]   Fig. 1.   Forskolin, sodium nitroprusside, and nicorandil increase Ca2+ spark frequency in smooth muscle cells from cerebral and coronary arteries. A: forskolin (10 µM) increases Ca2+ spark frequency, and an inhibitor of protein kinase A (PKA), H-89 (1 µM), prevents the effects of forskolin on single smooth muscle cells from rat cerebral arteries. First column shows a series of 6 consecutive line scan images each of 3 s duration from a control cerebral artery myocyte in which sparks were observed in scans 1 and 2. Second column shows 6 line scan images from a different cerebral artery myocyte bathed in forskolin for 20 min, in which sparks were observed in all scans. Third column shows 6 line scans from a 3rd cerebral artery myocyte bathed for 20 min in 10 µM forskolin and the PKA inhibitor H-89, in which sparks were observed in scans 1, 2, 3, and 5. In each of the line scan images, position across scan line is given by vertical displacement, and time progresses from left to right. Inset (left of the line scan images): orientation of scan line through a smooth muscle cell. B: bar graph summarizing number of sparks per cerebral artery myocyte during 18 s of scanning in control (n = 40 cells), forskolin (n = 40 cells), and forskolin + H-89 cells (n = 40 cells). Spark frequency was significantly (* P < 0.05) increased by forskolin. This effect was blocked by presence of H-89. C: bar graph of effect of nitrovasodilators sodium nitroprusside (SNP; 10 µM) and nicorandil (100 µM) on Ca2+ spark frequency in coronary artery myocytes. Ca2+ spark frequency (n = 164 cells) was significantly increased (* P < 0.05) with both SNP (n = 132 cells) and nicorandil (n = 95 cells). Error bars are SE.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Forskolin increases Ca2+-sensitive K+ (KCa) channel currents in single smooth muscle cells from cerebral arteries. A: forskolin (10 µM) increases spontaneous transient outward current (STOC) frequency in a single smooth muscle cell isolated from a cerebral artery. PKA inhibitor H-89 (1 µM) reversed effects of forskolin. Whole cell currents were recorded with perforated-patch technique at holding potential (Vh) of 40 mV. Also shown on expanded time scale is average current signal of 20 STOCs averaged under control, forskolin-treated, and forskolin + H-89 conditions. B: expanded portion of experiment in A before and after forskolin. C: time course of STOC frequency from experiment in A, plotted in 20-s bins. Mean STOC amplitude and frequencies in this experiment were 25.4 pA and 2.08 Hz (control), 36.5 pA and 4.8 Hz (forskolin), and 28.8 pA and 1.97 Hz (H-89), respectively.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Adenosine, forskolin, and dibutyryl-cAMP (db-cAMP) increase STOC frequency (A) and amplitude (B), and H-89 reversed the effects of forskolin. Changes in STOC frequency and amplitude relative to pretreatment control levels with application of adenosine (10 µM, n = 7), forskolin (10 µM, n = 9), dibutyryl-cAMP (500 µM, n = 5), and forskolin (10 µM) in the presence of H-89 (1 µM, n = 9) are illustrated. Error bars are means ± SE. Statistical significance was tested with a paired Student's t-test (* P < 0.05, ** P < 0.01).

Forskolin increases the NP_o of K_Ca channels in the absence of Ca²⁺ sparks. The forskolin-induced increase in Ca²⁺ spark frequency would explain the observed elevation of STOC frequency. However, the effect of forskolin on STOC amplitude cannot be explained by its action on Ca²⁺ sparks alone, because the amplitude of the spark was unaffected. A possible explanation for the 1.3-fold increase in STOC amplitude to forskolin is through an elevation of open state probability (P_o) of K_Ca channels caused by PKA phosphorylation of the channel, as has been demonstrated in excised patches by others (23). An elevated P_o of K_Ca channels would lead to higher P_o of the K_Ca channels during a spark and hence increase the STOC amplitude, if it were not already maximal at that voltage. To test the effects of forskolin on K_Ca channels, single-channel currents through K_Ca channels were recorded from isolated arterial myocytes in the whole cell mode, using the perforated-patch technique. Single K_Ca channels were identified by their characteristic large single-channel conductance, voltage dependence, and sensitivity to block by tetraethylamonium and iberiotoxin (7, 23, 28, 31).

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Forskolin increases open probability (NPo) of single KCa channels, in absence of Ca2+ sparks, in a single smooth muscle cell isolated from a cerebral artery. Single-channel currents were recorded with the whole cell, perforated-patch technique from a single isolated smooth muscle cell at 0 mV. STOCs were indirectly inhibited by blocking sarcoplasmic reticulum (SR) Ca2+ uptake with thapsigargin (100 nM). Shown is a series of channel openings from a single cell recorded before (NPo of this series = 0.043) and during (NPo of this series = 0.054) application of 10 µM forskolin at 0 mV.

Role of Ca²⁺ sparks and K_Ca channels in forskolin-induced dilations. We tested the hypothesis that part of the forskolin dilation of pressurized cerebral arteries is through increasing Ca²⁺ spark frequency, activating K_Ca channels. Consistent with this hypothesis, an inhibitor of Ca²⁺ sparks, ryanodine, inhibited 79.8 ± 8.1% of the forskolin-induced dilation of pressurized cerebral arteries with tone (Fig. 5D). Furthermore, ryanodine constricted pressurized cerebral arteries to a similar diameter in the absence (to 81 ± 7 µm) (28) or presence of forskolin (100 nM; to 85 ± 20 µm, n = 5), consistent with forskolin acting on a ryanodine-sensitive process. The blocker of K_Ca channels, iberiotoxin, also inhibited 82.3 ± 7.0% of the forskolin-induced dilation (Fig. 5D) (as shown previously, see Ref. 42). The effects of ryanodine and iberiotoxin were not additive (Fig. 5, B and D; see also Ref. 28), consistent with the idea that Ca²⁺ sparks act to stimulate K_Ca channels (28). Part of the dilation to forskolin was insensitive to iberiotoxin or ryanodine, particularly at higher forskolin concentrations, suggesting that forskolin also acts through other mechanisms (1, 11, 15, 18, 19, 36, 41, 48). In this preparation, however, forskolin (10 µM or less) was ineffective on pressurized arteries that were constricted with high (60 mM) K⁺ (Fig. 5, A and D), consistent with forskolin acting in large part through activation of K⁺ channels (1, 31). Ryanodine and iberiotoxin were ineffective on pressurized arteries previously dilated with the Ca²⁺ channel blocker diltiazem (1 µM; Fig. 5C), a condition that would lower Ca²⁺ spark frequency and hence K_Ca channel activity (7, 20, 28, 31) through reducing average intracellular Ca²⁺ and SR Ca²⁺ content. These results suggest that forskolin dilates pressurized cerebral arteries through activation of K⁺ channels, with part of the dilation due to activation of the Ca²⁺ sparkK_Ca channel pathway.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Forskolin dilations of pressurized (60 mmHg) cerebral arteries were reversed by ryanodine, but diltiazem dilations were not. Arteries were partially constricted (~35%) by pressure (60 mmHg), as indicated by horizontal dotted line. A: forskolin did not dilate pressurized cerebral arteries constricted by 60 mM K+. Transient dilation to K+ was due to activation of inward rectifier K+ channel present in these arteries (31). B: ryanodine (10 µM) constricted pressurized arteries dilated by forskolin. KCa channel blocker iberiotoxin (IbTx; 100 nM) was without effect in presence of ryanodine. C: ryanodine and iberiotoxin were without effect on pressurized arteries dilated by Ca2+ channel blocker, diltiazem (1 µM). Ryanodine constricted arteries when diltiazem was removed. D: dilations to 30 nM forskolin were inhibited by presence of either ryanodine (10 µM, n = 5) or iberiotoxin (100 nM, n = 4). Ryanodine and iberiotoxin together (n = 5) were no more effective than either agent alone. Forskolin is ineffective in the presence of 60 mM external K+ (n = 4).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study proposes a new mechanism of action for vasodilators that work through cAMP and cGMP (Fig. 7). Our data suggest that cAMP and cGMP can increase K_Ca channel activity in two ways: 1) by increasing Ca²⁺ spark frequency and 2) by directly increasing the open probability of the K_Ca channel. Together these actions lead to increased frequency and amplitude of STOCs, which, when summed across the vessel wall, should cause membrane potential hyperpolarization and, ultimately, relaxation of the artery.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Forskolin decreases and ryanodine increases arterial wall Ca2+ in an intact pressurized (60 mmHg) cerebral artery (A). Ca2+ signal is from an intact artery loaded with fura 2. Elevating pressure from 10 to 60 mmHg increased Ca2+ from ~100 to 200 nM and constricted the artery by ~35%. Rapid application of caffeine (10 mM) to the bath produced a transient increase in cytosolic Ca2+ (arrows). B: ratio images of cerebral artery loaded with fura 2 (same artery as in A) pressurized to 10 mmHg (No tone; a), to 60 mmHg physiological salt solution (PSS tone; b), at 60 mmHg during a caffeine pulse (PSS/Caf pulse; c), at 60 mmHg dilated with forskolin (FSK tone; d), at 60 mmHg with forskolin present during a caffeine pulse (FSK/Caf pulse; e), and at 60 mmHg in presence of forskolin after ryanodine-induced constriction (FSK + RYA tone; f). Diameters of the artery are indicated below images. C: forskolin increases and diltiazem and ryanodine decrease caffeine-induced Ca2+ transients. Peak change in Ca2+ to caffeine is shown for control (PSS) and in presence of 100 nM forskolin, 3 µM diltiazem, and 10 µM ryanodine. * Caffeine transient was significantly increased (P < 0.05) in presence of forskolin and decreased in presence of diltiazem (which reduced arterial wall Ca2+ to 126 ± 4 nM, n = 8) or ryanodine (P > 0.05).

View larger version (65K): [in this window] [in a new window]   Fig. 7.   Proposed mechanisms of action of forskolin and nitrovasodilators on Ca2+ sparks and KCa channels in arterial smooth muscle cells. Adenylyl cyclase (AC) produces cAMP when activated by externally applied forskolin or adenosine, which in turn stimulates PKA. Activation of PKA has a direct effect on the KCa channel to increase its open probability, as has been previously demonstrated in inside-out excised patches. However, the predominant effect of PKA is on the SR to increase Ca2+ spark frequency. Possible targets of PKA and PKG are the KCa channel, RyR receptors, and Ca2+-ATPase (phospholamban). Combined effects of increased spark activity and direct effects on the KCa channel lead to significantly increased activity of the KCa channel. Resulting hyperpolarization closes voltage-dependent Ca2+ channels, leading to reduced global intracellular Ca2+ and vasodilation. GC, guanylate cyclase; G, Gs protein.

Comparison of Ca²⁺ spark and STOC frequency. The mean Ca²⁺ spark frequency (0.14 Hz) was ~15% of the mean STOC frequency (0.96 Hz). We estimate that we could detect Ca²⁺ sparks in no more than 23% of cell volume, based on the average spread of a Ca²⁺ spark (4.8 µm) and maximum width of a cell (10 µm) (28) (assuming a circular cross section of a cell). This detection volume would be less when a scan line is close to the side of the cell and when the length of the scan line is less than the length of the cell, which is the case when the cells are not straight. These results are consistent with the idea that most Ca²⁺ sparks cause STOCs (17).

Effects on Ca²⁺ sparks. Forskolin, SNP, and nicorandil increased Ca²⁺ spark frequency. Forskolin, through cAMP/PKA, could act directly on the RyR channel (43, 46) and/or indirectly through an elevation of SR Ca²⁺ load, which has been shown to increase the open state probability of RyR channels in skeletal and cardiac muscle (12, 26, 45, 46). cAMP/PKA and cGMP/PKG could increase SR Ca²⁺ load by enhancing SR Ca²⁺-ATPase activity by phosphorylating phospholamban (25). The observed increase in caffeine-induced Ca²⁺ transients in the presence of forskolin (Fig. 6) is consistent with cAMP/PKA increasing SR Ca²⁺ load. However, forskolin did not cause a detectable increase in Ca²⁺ spark amplitude, which could be indicative of SR Ca²⁺ load. Nicorandil did increase Ca²⁺ spark amplitude by ~10%. Our data do not distinguish between a direct effect on ryanodine receptors and an indirect effect of SR Ca²⁺ load on Ca²⁺ spark frequency. cAMP (PKA) and cGMP (PKG) could conceivably act on both the Ca²⁺-ATPase and the RyR channel to affect SR function and STOC frequency (see Fig. 7) (49).

Contribution of STOCs to overall K_Ca channel activity. K_Ca channels do not appear to contribute significantly to regulation of smooth muscle membrane potential in intact pressurized cerebral arteries when Ca²⁺ sparks are blocked (28). This suggestion is based on the observation that blockers of K_Ca channels (iberiotoxin) were without effect on pressurized cerebral arteries when inhibitors of Ca²⁺ sparks (ryanodine, thapsigargin, or cyclopiazonic acid) were present. Furthermore, iberiotoxin was without effect in the presence of ryanodine and forskolin (Fig. 5B), suggesting that, even in the presence of forskolin, K_Ca channels do not contribute significantly to the regulation of arterial diameter without Ca²⁺ sparks.

Proposed mechanism for regulation of arterial diameter by cAMP/PKA or by cGMP/PKG via Ca²⁺ sparks and K_Ca channels. Elevation of intravascular pressure causes a graded membrane potential depolarization and vasoconstriction of small myogenic arteries (Fig. 7) (7, 20, 27, 31). The membrane potential depolarization increases the steady-state open probability of voltage-dependent Ca²⁺ channels (30, 40), which elevates Ca²⁺ entry and global intracellular Ca²⁺ (21, 27). Ca²⁺ channel inhibitors (e.g., nimodipine, diltiazem) prevent pressure-induced constrictions of myogenic cerebral arteries (7, 20, 30). Intracellular Ca²⁺ could increase Ca²⁺ spark frequency through cytoplasmic and SR luminal activation of RyR channels. Ca²⁺ sparks, by themselves, would have little direct effect on intracellular Ca²⁺, even at the highest observed rate (4/s) (28). Ca²⁺ sparks would, however, have a profound effect on the activity of K_Ca channels, which would oppose the pressure-induced membrane potential depolarization. In support of this proposed mechanism, blockers of K_Ca channels (e.g., iberiotoxin) and Ca²⁺ sparks (ryanodine, cyclopiazonic acid, thapsigargin) depolarize and constrict pressurized (60 mmHg) myogenic cerebral arteries by ~8 mV and 30% (7, 20, 28). Lowering intravascular pressure (7) or blocking Ca²⁺ channels (7, 20, 28), which would decrease intracellular Ca²⁺ and thus K_Ca channel activity, abolished the effects of iberiotoxin and ryanodine on membrane potential and arterial diameter (also see Fig. 6B with diltiazem). Furthermore, iberiotoxin was without effect on membrane potential and diameter in intact pressurized arteries, when Ca²⁺ sparks were blocked by ryanodine or thapsigargin (28) (see also Fig. 6A), suggesting that global intracellular Ca²⁺ does not cause sufficient activation of K_Ca channels to modulate membrane potential (see above).

1

1