Carbon Monoxide Dilates Cerebral Arterioles by Enhancing the Coupling of Ca2+ Sparks to Ca2+-Activated K+ Channels
Carbon monoxide (CO) is generated endogenously by the enzyme heme oxygenase. Although CO is a known vasodilator, cellular signaling mechanisms are poorly understood and are a source of controversy. The goal of the present study was to investigate mechanisms of CO dilation in porcine cerebral arterioles. Data indicate that exogenous or endogenously produced CO is a potent activator of large-conductance Ca2+-activated K+ (KCa) channels and Ca2+ spark–induced transient KCa currents in arteriole smooth muscle cells. In contrast, CO is a relatively poor activator of Ca2+ sparks. To understand the apparent discrepancy between potent effects on transient KCa currents and weak effects on Ca2+ sparks, regulation of the coupling relationship between these events by CO was investigated. CO increased the percentage of Ca2+ sparks that activated a transient KCa current (ie, the coupling ratio) from ≈62% in the control condition to 100% and elevated the slope of the amplitude correlation between these events ≈2.6-fold, indicating that Ca2+ sparks induced larger amplitude transient KCa currents in the presence of CO. This signaling pathway for CO is physiologically relevant because ryanodine, a ryanodine-sensitive Ca2+ release channel blocker that inhibits Ca2+ sparks, abolished CO dilation of pial arterioles in vivo. Thus, CO dilates cerebral arterioles by priming KCa channels for activation by Ca2+ sparks. This study presents a novel dilatory signaling pathway for CO in the cerebral circulation and appears to be the first presents of a vasodilator that acts by increasing the effective coupling of Ca2+ sparks to KCa channels.
Carbon monoxide (CO) is an important paracrine and autocrine relaxing factor in the cerebral and systemic circulation (see review1). However, mechanisms leading to the regulation of arterial diameter by CO are poorly understood and controversial.
CO is produced physiologically via the metabolism of heme by heme oxygenase.1 Heme is found in virtually all cell types, and arterial smooth muscle and endothelial cells contain both heme oxygenase-1 and heme oxygenase-2.2,3⇓ Authentic CO or heme oxygenase substrates dilate vascular preparations from anatomically diverse locations.1 Thus, CO may constitute a novel and relatively unexplored paracrine and autocrine gaseous vasodilator. In tail,4 gracilis muscle,5 renal,6 and cerebral arteries,7 CO-induced dilations are inhibited by blockers of the large-conductance Ca2+-activated K+ (KCa) channel. CO has also been reported to activate KCa channels in cultured arterial smooth muscle cells via chemical modification of an external histidine residue.8 These findings support an important role for KCa channels in CO dilations.
In arterial smooth muscle cells, localized intracellular Ca2+ transients, termed Ca2+ sparks, activate several sarcolemma KCa channels to induce a “spontaneous transient outward current.”9–11⇓⇓ Ca2+ sparks occur due to the opening of several ryanodine-sensitive Ca2+ release (RyR) channels located on the sarcoplasmic reticulum and elevate the intracellular Ca2+ concentration in the immediate vicinity of the release site by ≈10 μmol/L, but do not significantly contribute to global Ca2+ levels.10,12⇓ Ca2+ spark or KCa channel blockers depolarize arteries at physiological levels of pressure, leading to the activation of voltage-dependent Ca2+ channels, an elevation in the intracellular Ca2+ concentration, and constriction.11,13,14⇓⇓ Genetic ablation of the associated β1 subunit reduces KCa channel Ca2+ sensitivity, attenuates Ca2+ spark coupling, and leads to an elevation in arterial tone and systemic blood pressure.15,16⇓ Conversely, activators of cAMP-dependent or cGMP-dependent protein kinases dilate cerebral arteries, in part, by activating Ca2+ sparks and KCa channels.17 To date, it is not clear whether vasodilators can act by elevating the coupling of Ca2+ sparks to KCa channels, but this mechanism seems to be a reasonable possibility. Furthermore, it is unknown whether Ca2+ sparks are important for the vasodilatory actions of CO, but because KCa channels appear to be required, this mechanism also deserves investigation.
The goal of the present study was to investigate potential mechanisms of CO-induced dilation using cerebral arterioles and arteriole smooth muscle cells of newborn pigs. Authentic CO was applied exogenously, or CO was elevated endogenously in smooth muscle cells with heme-l-lysinate, a heme oxygenase substrate. Exogenous or endogenous CO was a potent activator of single KCa channels. Although CO effectively elevated transient KCa current frequency and amplitude, CO only slightly increased Ca2+ spark frequency. To investigate this apparent discrepancy between potent activation of transient KCa currents and relatively poor activation of Ca2+ sparks, the effect of CO on the coupling of Ca2+ sparks to KCa channels was investigated. CO increased the effective coupling of Ca2+ sparks to KCa channels from ≈62% to 100% and elevated the slope of the amplitude correlation between Ca2+ sparks and transient KCa currents ≈2.6-fold. This dilatory signaling mechanism is physiologically relevant because ryanodine, an RyR blocker that inhibits Ca2+ sparks, abolished CO dilation in vivo.
Materials and Methods
Anesthetized newborn pigs (1 to 3 days old, 1 to 2.5 kg) were maintained on α-chloralose (50 mg/kg IV), and closed cranial windows were implanted over the left parietal cortices as previously described.7,18⇓ For removal of pial arterioles (50 to 200 μm), the brain was removed. Isolated arterioles were dissected from the brain, cleaned of basolateral connective tissue, and maintained in ice-cold HEPES-buffered PSS containing (in mmol/L) NaCl 134, KCl 6, CaCl2 2, MgCl2 1, HEPES 10, and glucose 10 (pH 7.4, NaOH). Smooth muscle cells were enzymatically dissociated from cerebral arterioles as previously described.13
K+ currents were measured in isolated smooth muscle cells by use of the perforated-patch configuration of the patch-clamp technique as previously described.22 The bathing solution was HEPES-buffered PSS (composition described above). The pipette solution contained (in mmol/L) potassium aspartate 110, KCl 30, NaCl 10, MgCl2 1, HEPES 10, and EGTA 0.05 (pH 7.2, KOH). In experiments in which the activities of single KCa channel currents were measured, Ca2+ sparks (and thus, transient KCa currents) were abolished with thapsigargin, a sarcoplasmic reticulum Ca2+-ATPase inhibitor that depletes the sarcoplasmic reticulum of Ca2+.11,13⇓ In each patch under each condition, KCa channel activity (NPo) was calculated from at least 60 seconds of continuous gap-free data by using Fetchan 6 (Axon Instruments). NPo was calculated from the following equation: NPo=Σ(t1+t2. . .ti), where ti is the relative open time for each channel level. In the present study, the average number of events used for single NPo analysis under each condition was as follows (mean±SEM): control, 896±169 (n=28 cells); CO, 3514±728 (n=7); heme-l-lysinate, 4980±11 (n=5); FeCl3, 522±266 (n=5); bilirubin, 582±214 (n=5); and NADP, 586±192 (n=6). Transient KCa current analysis was performed offline as previously described.22 A transient KCa current was defined as the simultaneous opening of ≥3 KCa channels (threshold, 6 pA at −40 mV). Simultaneous openings of ≥3 KCa channels were not observed in the presence of thapsigargin (100 nmol/L, n=4 cells).
Confocal Ca2+ Imaging
Arteriole segments or isolated smooth muscle cells were loaded with the fluorescent Ca2+ indicator fluo 4-AM (10 μmol/L) as previously described.22 Arteriole segments were imaged in an extracellular solution containing 30 mmol/L K+, which depolarizes smooth muscle cells from ≈−60 to −40 mV and elevates global intracellular Ca2+ concentration and Ca2+ spark frequency (see similar procedures13,19⇓). The 30 mmol/L K+ bath solution contained (in mmol/L) NaCl 110, KCl 30, HEPES 10, CaCl2 2, MgCl2 1, and glucose 10 (pH 7.4, NaOH). Smooth muscle cells were imaged by using a Noran Oz laser scanning confocal microscope as previously described.22 Fluorescent images of smooth muscle cells in arteriole segments were recorded every 16.7 ms (60 images/s). For isolated smooth muscle cells, images were acquired every 8.3 ms (120 images/s). In experiments in which confocal Ca2+ imaging was performed in combination with patch-clamp electrophysiology, current and fluorescence measurements were synchronized by using a light-emitting diode positioned above the recording chamber that was triggered during acquisition. Ca2+ sparks were detected in smooth muscle cells by using custom analysis software that was written with the use of IDL 5.0.2 (Research Systems Inc), kindly provided by Drs M.T. Nelson and A.D. Bonev (University of Vermont, Burlington). Detection of Ca2+ sparks was performed by dividing an area 1.54 μm (7 pixels)×1.54 μm (7 pixels) (ie, 2.37 μm2) in each image (F) by a baseline (F0), which was determined by averaging 10 images without Ca2+ spark activity. Ca2+ spark amplitude was calculated as F/F0. A Ca2+ spark was defined as a localized increase in F/F0 that was >1.2. Calculations of changes in local (within the 2.37-μm2 analysis area) or global Ca2+ concentration were performed by using methods previously described.19
In Vivo Arterial Diameter Measurement
Cranial windows were implanted as previously described.7,20⇓ The space under the window was filled with artificial cerebrospinal fluid, which was equilibrated with 6% CO2/6% O2/88% N2, producing pH 7.33 to 7.40, Pco2 42 to 46 mm Hg, and Po2 43 to 50 mm Hg. Pial arterioles were observed with a dissection microscope, and diameter was measured with a calibrated video micrometer attached to a monitor. Artificial cerebrospinal fluid was exchanged topically to introduce treatments to the periarachnoid space.
Values are expressed as mean±SEM. Differences among multiple groups were evaluated by ANOVA with the Tukey test. For comparison between two groups, paired and unpaired Student t tests were used as appropriate. First-order polynomial linear fits were used to calculate statistical correlation for Ca2+ spark and evoked transient KCa current amplitudes. A value of P<0.05 was considered significant and is illustrated as an asterisk where appropriate.
An expanded Materials and Methods section is available online at http://www.circresaha.org in the data supplement.
CO Activates KCa Channels
CO is generated endogenously from heme by the action of heme oxygenase.1 To investigate whether exogenously applied CO or CO produced endogenously by heme oxygenase activates KCa channels in newborn cerebral arteriole smooth muscle cells, NPo was measured in isolated arteriole smooth muscle cells by using the perforated-patch configuration of the patch-clamp technique.
Openings of large-conductance KCa channels were identified on the basis of the characteristic single-channel conductance (116±8 pS between −20 and 20 mV, n=6) and block by tetraethylammonium+ (1 mmol/L) (NPo, control 0.025±0.016, tetraethylammonium+ 0.004±0.002; n=3). At 0 mV, authentic CO (100 nmol/L) increased mean NPo 5.1-fold (Figure 1). Similarly, the heme oxygenase substrate, heme-l-lysinate (100 nmol/L), increased mean NPo 6.9-fold (Figure 1). After administration, CO rapidly increased NPo to a peak level, whereas heme-l-lysinate increased NPo gradually to reach a plateau after ≈3 to 4 minutes, consistent with endogenous generation of an intracellular KCa channel activator via the enzymatic action of heme oxygenase. Additional byproducts of heme degradation that are produced in equimolar concentration, namely, FeCl3 (100 nmol/L), bilirubin (100 nmol/L), and NADP (100 nmol/L), did not alter NPo, suggesting that CO was the KCa channel activator generated endogenously (Figure 1). These data suggest that exogenous CO or CO produced endogenously by heme oxygenase is a potent activator of KCa channels in cerebral arteriole smooth muscle cells.
CO Elevates Transient KCa Current Frequency and Amplitude
In smooth muscle cells, Ca2+ sparks activate several KCa channels, resulting in a transient KCa current.10 Inhibition of Ca2+ sparks or KCa channels leads to membrane depolarization, activation of voltage-dependent Ca2+ channels, elevated Ca2+ influx, and constriction.10 Conceivably, CO could dilate newborn cerebral arterioles via the activation of transient KCa currents. To examine the mechanisms of CO dilation, the regulation of transient KCa currents by exogenous CO or CO produced endogenously from heme-l-lysinate was investigated in isolated voltage-clamped (−40 mV) cerebral arteriole smooth muscle cells.
CO (100 nmol/L) increased mean transient KCa current frequency 2.4-fold and mean amplitude 1.6-fold (Figure 2). Heme-l-lysinate (100 nmol/L) increased mean transient KCa current frequency 2.9-fold and mean amplitude 1.9-fold (Figure 2). Similar to its effects on KCa channels, CO induced a rapid activation of transient KCa currents, whereas heme-l-lysinate gradually increased transient KCa current frequency and amplitude, consistent with endogenous generation of CO via heme oxygenase (Figure 2). These data suggest that exogenous CO or endogenous CO produced by cellular heme oxygenase activates transient KCa currents in cerebral arteriole smooth muscle cells.
Heme-l-Lysinate Activates Ca2+ Sparks
CO could increase transient KCa currents via the activation of Ca2+ sparks. To investigate this hypothesis, rapid changes in intracellular Ca2+ concentration were measured in the smooth muscle cells of cerebral arteriole segments by using a laser scanning confocal microscope and the fluorescent Ca2+ indicator fluo 4.
Heme-l-lysinate slightly increased mean Ca2+ spark frequency in the smooth muscle cells of cerebral arteriole segments from 0.7±0.08 to 0.99±0.11 Hz, or ≈1.4-fold (Figure 3). However, heme-l-lysinate did not significantly alter mean Ca2+ spark amplitude (for F/F0, control 1.44±0.01 [n=188 sparks], heme-l-lysinate 1.46±0.01 [n=244 sparks]). Heme-l-lysinate reduced global F/F0 in the same smooth muscle cells to 83±4% of control, suggesting that intracellular Ca2+ concentration decreased. Ryanodine (10 μmol/L), an RyR channel blocker, abolished Ca2+ sparks in smooth muscle cells (data not shown), suggesting that Ca2+ sparks occurred due to the opening of RyR channels, consistent with previous reports.10 These data suggest that CO elevates Ca2+ spark frequency and decreases global Ca2+ in arteriole smooth muscle cells.
CO Augments the Coupling Ratio and Amplitude Correlation Between Ca2+ Sparks and Transient KCa Currents
Heme-l-lysinate increased Ca2+ spark frequency only ≈1.4-fold in smooth muscle cells (Figure 3), which was considerably less than the ≈2- to 3-fold increase in transient KCa current frequency induced by heme-l-lysinate or exogenous CO (Figure 2). Furthermore, CO or heme-l-lysinate increased transient KCa current amplitude (1.6- to 1.9-fold), but heme-l-lysinate did not elevate Ca2+ spark amplitude (F/F0). To investigate the apparent discrepancies between the effective activation of transient KCa currents and weak activation of Ca2+ sparks, regulation of the coupling relationship between Ca2+ sparks and transient KCa currents by CO was investigated in voltage-clamped (−40 mV) cerebral arteriole smooth muscle cells.
In the control condition, only ≈62% of Ca2+ sparks evoked a transient KCa current (Figures 4 and 5⇓). The mean amplitude of uncoupled Ca2+ sparks (F/F0 1.68±0.11) was significantly smaller than the mean amplitude of coupled sparks (F/F0 2.0±0.07, P<0.05). Although the amplitude relationship between a coupled Ca2+ spark and the evoked transient KCa current was correlated, the slope was low (Figure 4). In the same cells, heme-l-lysinate (100 nmol/L) increased the percentage of Ca2+ sparks that evoked a transient KCa current (ie, the coupling ratio) to 100±0%. Mean Ca2+ spark amplitude was not altered by heme-l-lysinate (F/F0, control 1.88±0.06 [n=52], heme-l-lysinate 1.80±0.05 [n=73]), suggesting that the increase in coupling was not due to a change in Ca2+ spark properties. Heme-l-lysinate also did not significantly alter global F/F0 (105±6% of control, P<0.05). In the presence of heme-l-lysinate, the amplitude of a Ca2+ spark and that of the evoked transient KCa current were similarly correlated. However, heme-l-lysinate increased the slope of the amplitude correlation ≈2.6-fold, indicating that Ca2+ sparks of similar amplitude evoke significantly larger transient KCa currents after endogenous generation of CO. This finding explains the CO-induced increase in transient KCa current amplitude in the absence of a change in Ca2+ spark amplitude. These data suggest that heme-l-lysinate increases transient KCa current frequency and amplitude in newborn cerebral arteriole smooth muscle cells primarily by elevating the coupling ratio and amplitude relationship between Ca2+ sparks and transient KCa currents. The increase in effective coupling of Ca2+ sparks (>1.6-fold) contributes significantly to the increase in transient KCa current frequency (2- to 3-fold), particularly when in combination with an increase in spark frequency.
Ryanodine, an RyR Channel Blocker, Abolishes CO Dilation of Cerebral Arterioles In Vivo
Dilations to CO are prevented by KCa channel blockers.4–7⇓⇓⇓ To investigate the importance of Ca2+ sparks for CO dilation, arteriole diameter was measured in vivo by using cranial windows implanted in anesthetized piglets. Dilations to exogenous CO were measured before and after the administration of ryanodine, an RyR channel blocker that blocked Ca2+ sparks (also see Jaggar et al10). Dilations to sodium nitroprusside, an NO donor, or isoproterenol, an adrenergic agonist, were also studied in the same arterioles before and after ryanodine administration.
CO (1 nmol/L), isoproterenol (1 μmol/L), or sodium nitroprusside (10 μmol/L) significantly dilated pial arterioles (Figure 6). Ryanodine (10 μmol/L) completely abolished dilations of the same arterioles to CO but only slightly attenuated dilations induced by isoproterenol or sodium nitroprusside. These data suggest that Ca2+ sparks are required for CO to induce arterial dilation in the newborn pig cerebral circulation and that isoproterenol and sodium nitroprusside cause dilation primarily via other mechanisms.
We show that CO dilates cerebral arterioles primarily by augmenting the effective coupling of Ca2+ sparks to KCa channels in smooth muscle cells. This leads to an increase in transient KCa current frequency and amplitude. A CO-induced increase in transient KCa current frequency and amplitude will lead to membrane hyperpolarization, a decrease in voltage-dependent Ca2+ channel activity, a reduction in intracellular Ca2+ concentration, and arteriole dilation. This pathway is physiologically relevant and dependent on Ca2+ sparks, because ryanodine, an RyR channel blocker, abolishes CO dilation of cerebral arterioles in vivo. This is the first time that a vasodilator has been reported to act by increasing the effective coupling between Ca2+ sparks and transient KCa currents. This finding is particularly relevant when the lack of knowledge regarding the vasodilatory actions of CO is considered.
CO Augments KCa Channel Activation by Ca2+ Sparks in Cerebral Arteriole Smooth Muscle Cells
Exogenous CO or heme-l-lysinate activated KCa channels, transient KCa currents, and Ca2+ sparks in cerebral arteriole smooth muscle cells. Heme oxygenase-2 is constitutively expressed in arterial smooth muscle cells,2 and isolated arteries generate CO under basal conditions and in response to heme oxygenase substrate,18 suggesting that heme-l-lysinate activation of KCa channels is via CO. CO activation of KCa channels and transient KCa currents peaked almost immediately, whereas heme-l-lysinate activation took ≈3 to 4 minutes to reach a peak level, consistent with the generation of an endogenous secondary mediator. Furthermore, in control experiments, other products of heme degradation (Fe2+, bilirubin, and NADP) had no effect on KCa channels. At the same concentration, heme-l-lysinate was a more effective activator of KCa channels and transient KCa currents than was authentic CO, suggesting that heme oxygenase (and thus, CO production) may be localized near KCa channels.
Heme-l-lysinate decreased global Ca2+ concentration in smooth muscle cells of intact arteries to ≈83% of control, or from ≈19521 to 150 nmol/L, which would lead to dilation. The calculated peak elevation in local Ca2+ concentration caused by a Ca2+ spark would also decrease from 358 nmol/L in control to 265 nmol/L with heme-l-lysinate. This calculation suggests that that the micromolar Ca2+ concentration to which KCa channels located in the vicinity of the release site are exposed12 should also be reduced. Membrane hyperpolarization in the intact arteriole would also decrease KCa channel Ca2+ sensitivity and the driving force for K+, which would also reduce the impact of Ca2+ sparks. However, heme-l-lysinate increased the slope of the amplitude relationship between Ca2+ sparks and transient KCa currents ≈2.6-fold. Thus, the decrease in spark amplitude, KCa channel Ca2+ sensitivity, and driving force for K+ that would occur in the intact arteriole because of membrane hyperpolarization would be compensated by an increase in the sensitivity of KCa channels to Ca2+ sparks.
CO or heme-l-lysinate increased transient KCa current frequency 2- to 3-fold in voltage-clamped cells, whereas heme-l-lysinate increased Ca2+ spark frequency only ≈1.4-fold in intact arteries. Because Ca2+ sparks are regulated by intracellular Ca2+ concentration,10 the concomitant decrease in global Ca2+ concentration induced by heme-l-lysinate would be expected to attenuate the elevation in Ca2+ spark frequency. However, in voltage-clamped cells in which global Ca2+ did not change, heme-l-lysinate would increase transient KCa current frequency only 1.5-fold if the coupling ratio of 60% was maintained, ie, (0.49 Hz×0.6)/0.2 Hz. Therefore, the majority of the heme-l-lysinate–induced elevation in transient KCa current frequency (2.9-fold) occurs because of an increase in the effective coupling of Ca2+ sparks to KCa channels rather than because of an increase in spark frequency.
Ryanodine, an RyR channel blocker, abolished CO-induced pial arteriole dilation, suggesting that Ca2+ sparks are critical for vasoregulatory actions of CO. CO increased NPo ≈5- to 6-fold in the absence of Ca2+ sparks, but clearly, this was insufficient to induce dilation. Interestingly, in control conditions, ryanodine induced only a small arterial constriction, in contrast to effects in pressurized adult rat cerebral arteries.11 The lack of effect of ryanodine is not surprising because transient KCa current frequency and amplitude (and thus, NPo) were low in newborn piglet cerebral arteriole smooth muscle cells in the absence of CO. Because CO elevates the coupling ratio and amplitude relationship between Ca2+ sparks and KCa channels, and this ultimately leads to dilation, these data support the physiological role of Ca2+ sparks and KCa channels in negative-feedback regulation of arterial diameter with the use of an in vivo preparation.
Essentially 100% of Ca2+ sparks evoke a transient KCa current in cerebral artery smooth muscle cells of adult rats22,23⇓ and mice.15 Similar to the porcine preparation used in the present study, a significant proportion of Ca2+ sparks does not evoke a transient KCa current in human cerebral artery (28%)24, esophageal (27%),25 or Bufo marinus stomach (21%)26 smooth muscle cells. In those tissues, the slope of the amplitude relationship between a Ca2+ spark and the evoked KCa transient is also low, similar to the slope obtained in the absence of CO in the present study. Thus, CO may be an effective relaxing factor in other smooth muscle preparations by increasing spark–KCa channel coupling. Conceivably, membrane depolarization, which also increases KCa channel Ca2+ sensitivity, may be an additional mechanism to increase the coupling of Ca2+ sparks to KCa channels. If this is the case, membrane depolarization may somewhat reduce the dilator efficacy of CO.
Signal Transduction Pathways for CO in Arteriole Smooth Muscle Cells
The intracellular signaling pathways affected by CO are unclear. CO may activate KCa channels via direct effects,8 via activation of soluble guanylyl cyclase,27 via an associated membrane heme protein that acts as an oxygen sensor,28 or via inhibition of formation of 20-hydroxyeicosatetraenoic acid, a KCa channel blocker.29 Conceivably, CO could activate Ca2+ sparks via similar signaling mechanisms, although this would require further investigation. A major issue concerning signaling mechanisms for CO is whether the gas activates soluble guanylyl cyclase, which would increase cGMP and activate cGMP-dependent protein kinase. CO can activate soluble guanylyl cyclase but is only 1% as effective as NO. A saturated solution of CO (1 mmol/L) increases the activity of human soluble guanylyl cyclase only 4-fold.30 In the present study, the highest concentration of CO used was 10 000 times lower than this concentration, suggesting that guanylyl cyclase activation would be extremely small. For comparison, sodium nitroprusside at micromolar concentrations elevates soluble guanylyl cyclase activity ≈300-fold, with an EC50 of 2 μmol/L.30 Further support for an insignificant role for guanylyl cyclase comes from previous studies demonstrating that CO or heme-l-lysinate does not increase cGMP in cranial window perfusate at concentrations that dilate cerebral arterioles.7 In contrast, in the same study, sodium nitroprusside induced a similar dilation and significantly elevated cGMP.
There are significant differences in the cellular signaling mechanisms for CO when they are compared with the actions of vasodilators that elevate cyclic nucleotides. Forskolin, which elevates cAMP, or sodium nitroprusside, which elevates cGMP, increases KCa current primarily by elevating Ca2+ spark frequency, combined with a small activation of KCa channels.17 In contrast, CO appears to increase KCa current primarily via effects on KCa channels, with a small effect on Ca2+ sparks. When compared with agents that increase cyclic nucleotides, CO was a far more effective KCa channel activator (5- to 6-fold [present study] versus 1.3-fold17) and a weak activator of Ca2+ sparks (1.4-fold [present study] versus 2- to 3-fold17). Functional evidence also supports the hypothesis that CO acts via a mechanism distinct from dilators that elevate cyclic nucleotides. In piglet arterioles, ryanodine abolished dilations to CO but only slightly reduced dilations to isoproterenol, an agonist that elevates cAMP, or sodium nitroprusside, an NO donor. Cellular targets for cyclic nucleotides may also be slightly different in smooth muscle cells of the newborn vasculature, particularly because the contribution of NO to arterial diameter regulation appears to be far less in the newborn compared with the adult vasculature.31
KCa channels can be composed of a pore-forming α subunit and an associated β subunit that increases intracellular Ca2+ sensitivity.32 Genetic ablation of the associated β1 subunit of the KCa channel significantly reduces the coupling of Ca2+ sparks to KCa channels, leading to membrane depolarization, enhanced pressure-induced constriction of cerebral arteries, and increased blood pressure.15,16⇓ Mechanisms that lead to the uncoupling of Ca2+ sparks from KCa channels in wild-type smooth muscle cells are unclear but may also involve a reduced Ca2+ sensitivity of KCa channels within the vicinity of the Ca2+ release site. Because some, but not all, Ca2+ sparks induce a transient KCa current, a heterogeneous population of KCa channels may exist in the sarcolemma: those that are composed of only an α subunit and those that colocalize with a β subunit. CO has been proposed to increase KCa channel Ca2+ sensitivity in cultured arterial smooth muscle cells.8 Our data suggest that CO increases Ca2+ spark coupling by shifting the Ca2+ sensitivity of the KCa channel to within the micromolar range of intracellular Ca2+ concentrations produced by a Ca2+ spark.12 Conceivably, CO may act via the KCa channel β subunit to enhance this coupling relationship.
In summary, we demonstrate that exogenous or endogenously produced CO dilates cerebral arterioles by activating KCa channels primarily by increasing the coupling ratio and amplitude relationship between Ca2+ sparks and KCa channels. Although CO was a potent and effective activator of KCa channels, CO did not dilate arterioles in the absence of Ca2+ sparks. Therefore, CO appears to act by priming KCa channels for activation by Ca2+ sparks, and this ultimately leads to arteriole dilation via membrane hyperpolarization.
This study was supported by NIH grants NS-67061 (to Dr Jaggar), HL-042851 (to Dr Leffler), and HL-034059 (to Drs Leffler and Jaggar) and grants from the American Heart Association National Center and Southeast Affiliate (to Dr Jaggar). X. Cheng is a Predoctoral Fellow of the American Heart Association. We thank Liliya Balabanova for technical help with preliminary experiments and Alex Fedinec for in vivo measurements of arterial diameter.
Original received July 10, 2002; revision received August 30, 2002; accepted August 30, 2002.
- ↵Christodoulides N, Durante W, Kroll MH, Schafer AI. Vascular smooth muscle cell heme oxygenases generate guanylyl cyclase–stimulatory carbon monoxide. Circulation. 1995; 91: 2306–2309.
- ↵Zakhary R, Gaine SP, Dinerman JL, Ruat M, Flavahan NA, Snyder SH. Heme oxygenase 2: endothelial and neuronal localization and role in endothelium-dependent relaxation. Proc Natl Acad Sci U S A. 1996; 93: 795–798.
- ↵Zhang F, Kaide J, Wei Y, Jiang H, Yu C, Balazy M, Abraham NG, Wang W, Nasjletti A. Carbon monoxide produced by isolated arterioles attenuates pressure-induced vasoconstriction. Am J Physiol. 2001; 281: H350–H358.
- ↵Jaggar JH, Porter VA, Lederer WJ, Nelson MT. Calcium sparks in smooth muscle. Am J Physiol. 2000; 278: C235–C256.
- ↵Nelson MT, Cheng H, Rubart M, Santana LF, Bonev AD, Knot HJ, Lederer WJ. Relaxation of arterial smooth muscle by calcium sparks. Science. 1995; 270: 633–637.
- ↵Perez GJ, Bonev AD, Nelson MT. Micromolar Ca2+ from sparks activates Ca2+-sensitive K+ channels in rat cerebral artery smooth muscle. Am J Physiol. 2001; 281: C1769–C1775.
- ↵Jaggar JH. Intravascular pressure regulates local and global Ca2+ signaling in cerebral artery smooth muscle cells. Am J Physiol. 2001; 281: C439–C448.
- ↵Robinson JS, Fedinec AL, Leffler CW. Role of carbon monoxide in glutamate receptor-induced dilation of newborn pig pial arterioles. Am J Physiol. 2002; 282: H2371–H2376.
- ↵Leffler CW, Nasjletti A, Johnson RA, Fedinec AL, Leffler CW, Nasjletti A, Yu C, Johnson RA, Fedinec AL, Walker N. Contributions of prostacyclin and nitric oxide to carbon monoxide-induced cerebrovascular dilation in piglets. Am J Physiol. 2001; 280: H1490–H1495.
- ↵Gollasch M, Wellman GC, Knot HJ, Jaggar JH, Damon DH, Bonev AD, Nelson MT. Ontogeny of local sarcoplasmic reticulum Ca2+ signals in cerebral arteries: Ca2+ sparks as elementary physiological events. Circ Res. 1998; 83: 1104–1114.
- ↵Cheranov S, Jaggar JH. Sarcoplasmic reticulum calcium load regulates rat arterial smooth muscle calcium sparks and transient KCa currents. J Physiol. In press.
- ↵Perez GJ, Bonev AD, Patlak JB, Nelson MT. Functional coupling of ryanodine receptors to KCa channels in smooth muscle cells from rat cerebral arteries. J Gen Physiol. 1999; 113: 229–238.
- ↵Wellman GC, Nathan DJ, Saundry CM, Perez G, Bonev AD, Penar PL, Tranmer BI, Nelson MT. Ca2+ sparks and their function in human cerebral arteries. Stroke. 2002; 33: 802–808.
- ↵ZhuGe R, Fogarty KE, Tuft RA, Lifshitz LM, Sayar K, Walsh JV Jr. Dynamics of signaling between Ca2+ sparks and Ca2+-activated K+ channels studied with a novel image-based method for direct intracellular measurement of ryanodine receptor Ca2+ current. J Gen Physiol. 2000; 116: 845–864.
- ↵Riesco-Fagundo AM, Perez-Garcia MT, Gonzalez C, Lopez-Lopez JR. O2 modulates large-conductance Ca2+-dependent K+ channels of rat chemoreceptor cells by a membrane-restricted and CO-sensitive mechanism. Circ Res. 2001; 89: 430–436.
- ↵Coceani F. Carbon monoxide in vasoregulation: the promise and the challenge. Circ Res. 2000; 86: 1184–1186.
- ↵Lee YC, Martin E, Murad F. Human recombinant soluble guanylyl cyclase: expression, purification, and regulation. Proc Natl Acad Sci U S A. 2000; 97: 10763–10768.
- ↵Cox DH, Aldrich RW. Role of the β1 subunit in large-conductance Ca2+-activated K+ channel gating energetics. Mechanisms of enhanced Ca2+ sensitivity. J Gen Physiol. 2000; 116: 411–432.