Large Conductance, Ca²⁺-Activated K⁺ Channels (BK_Ca) and Arteriolar Myogenic Signaling

Summary

Myogenic, or pressure-induced, vasoconstriction is critical for local blood flow autoregulation. Underlying this VSM response are events including membrane depolarization, Ca²⁺ entry and mobilization, and activation of contractile proteins. BK_Ca has been implicated in several of these steps including, 1) channel closure causing membrane depolarization, and 2) channel opening causing hyperpolarization to oppose excessive pressure-induced vasoconstriction. As multiple mechanisms regulate BK_Ca activity, (subunit composition, Em and Ca²⁺ levels, post-translational modification) tissue level diversity is predicted. Importantly, heterogeneity may contribute to tissue-specific differences in regulation of myogenic vasoconstriction, allowing local hemodynamics to be matched to metabolic requirements. Knowledge of such variability will be important to exploiting the BK_Ca channel as a therapeutic target and understanding systemic effects of its pharmacological manipulation.

K⁺ channels play key roles in the regulation of diameter of small arteries and arterioles. Principally this results from K⁺ channel - induced hyperpolarization of smooth muscle cell plasma membranes decreasing the opening of voltage-gated Ca²⁺ channels. While arterioles are known express a variety of K⁺ channels (Table 1; see also Jackson[1]) the dominant channels in regulation of vascular tone are the large conductance Ca²⁺ - activated K⁺ channel (BK_Ca) of the vascular smooth muscle (VSM) cells and the small (SK) - and intermediate (IK) – conductance, Ca²⁺ activated K⁺ channels which are particularly associated with endothelial cells. As a result of its large single channel conductance (approximately 150 pS under physiological conditions with respect to K⁺ gradient) and high protein expression levels, BK_Ca plays a significant role in determining membrane potential (Em). In addition, and of direct relevance to microvascular smooth muscle, BK_Ca is activated by both Ca²⁺ and voltage which favors it functioning as a dominant repolarizing current to act as a negative feedback control mechanism for contractile stimuli while being sensitive to a number of metabolic stimuli (pO2, ROS, phosphorylation, fatty acids and their metabolites).

Table 1.

K⁺ Channels Implicated in Arteriolar Vasomotor Function.

Channel	Distribution	Activation	Function
BKCa	Predominately smooth muscle	Ca2+, voltage, protein kinases, steroid hormones, lipids	Membrane hyperpolarization; smooth muscle relaxation; oppose myogenic tone
Kv	Smooth muscle	Voltage, protein kinases	Membrane hyperpolarization; Appear to play a negative feedback role in limiting myogenic reactivity. Inhibitors of Kv (4-aminopyridine, correolide) cause vasoconstriction of myogenically active small arteries.
Kir	Smooth muscle	Extracellular K+, voltage	Metabolic and endothelial – dependent dilation. Targeted disruption of Kir2.1 leads to impaired dilator responses to 15 mM K+ without altering myogenic tone or cAMP - mediated dilation.
Katp	(Endothelium) and smooth muscle	Availability of ATP/ADP, protein kinases	Protection against hypoxia/ischemia. Dilation to vasoactive substances including PGI2, adenosine, isoproterenol. Inhibited by vasoconstrictors.
K2p e.g. TREK-1 (TWIK-related K+channels)	Endothelium and smooth muscle	Mechanical, osmolality, polyunsaturated fatty acids	Operate at physiological levels of membrane potential. Basilar and mesenteric arteries from TREK-1−/− mice do not show altered myogenic responsiveness, suggested that under physiological conditions these channels modulate vasconstriction rather than determining levels of myogenic tone.
SK	Predominately EC	Ca2+ via constitutively bound calmodulin	Membrane hyperpolarization; Endothelial-dependent dilation
IK	Predominately EC, but also expressed in proliferating VSM cell	Ca2+ via Constitutively Bound calmodulin	Membrane hyperpolarization; Endothelial-dependent dilation

The role of VSM BK_Ca in the regulation of arteriolar tone, specifically myogenic tone, is the focus of this review, whereas the role of SK and IK in endothelial-dependent regulation of arteriolar diameter has been the subject of a number of recent publications [2,3]. Diversity in the molecular structure of BK_Ca, its spatial localization relative to other signaling elements and its ability to be regulated by a number of post-translational mechanisms (including phosphorylation, fatty acids, hormones and redox state) increases the likelihood that this channel may exhibit both tissue and functional heterogeneity. Further, BK_Ca may not only play a role in control of vascular tone under physiological conditions, but may contribute to the alterations in tone characteristic of pathophysiological states, such as hypertension, stroke, atherosclerosis, diabetes and complications of cardiovascular surgery [4–7].

Molecular and Electrophysiological Characteristics of BK_Ca

i. Subunit Composition

Based on both functional and structural evidence [8,9], it is now established that the holo-BK_Ca channel is comprised of four similar α subunits that co-assemble around a central axis to form a single K⁺-selective conduction pore. Structurally, the BK_Ca α subunit contains 7 putative transmembrane (TM)-spanning α helical segments (denoted S0–S6), with the ion conduction pathway and selectivity filter formed by the pore-forming loop structure connecting TM segments S5 and S6 (Figure 1). The amino terminus of each α subunit is located extracellularly; an unusual characteristic for ion channels. The selectivity filter itself is comprised of a signature “G-Y-G motif that is evolutionarily conserved amongst K⁺-selective ion channels. X-ray crystallographic structures of both prokaryotic (i.e. KcsA, KvAP, MthK) [10–13] and eukaryotic (i.e. Kv1.2)[14] K⁺ channels have demonstrated that the S5-pore loop-S6 tertiary structure is highly conserved, and itself may be considered a distinct module or domain. Moreover, it is now apparent that the S1–S4 segments present in BK_Ca and other voltage-gated K⁺ channels (i.e. Shaker and mammalian Kv homologues) function as a voltage sensor domain (VSD), and collectively represent a distinct structural module. In BK_Ca channels, 4–5 positively and negatively charged residues distributed within the S2-S4 helices appear to represent the major voltage-sensing residues in the VSD of these channels[15]. This situation differs from that of Shaker-type Kv channels, in which the charged residues contributing to voltage sensing (i.e. Arg residues) are largely contained within the S4 segment. BK_Ca channels also display weaker voltage sensitivity compared to Kv channels (i.e. ~2.5 versus 12–13 effective gating charges per holo-channel, respectively), most likely due to fewer positively charged residues in the BK_Ca VSD being sensitive to the transmembrane voltage. Electrophysiologically, this weaker voltage sensitivity of BK_Ca channels is reflected by the shallower slope of plotted conductance-voltage relations describing BK_Ca versus Kv-channel opening.

Figure 1.

Schematic diagram illustrating the subunit structure of BK_Ca (upper panel) and the activation of BK_Ca through SR-mediated Ca²⁺ sparks (lower panel). Upper Panel: in VSM cells BK_Ca exists as a tetramer of α subunits arranged around a central conducting pore. The α subunit has 7 membrane spanning domains and an intracellular C-terminal tail containing a number of regulatory sites (see text for detail). Co-assembled with each α subunit is a β1 subunit, which confers additional Ca²⁺ and voltage sensitivity on the channel. Lower Panel: close apposition of the SR and plasma membranes creates a restricted space in which Ca²⁺ can increase to levels required for activation of BK_Ca. Ca²⁺ increases occur transiently in the form of Ca²⁺ sparks due to periodic release from the SR via ryanodine-sensitive mechanisms.

In addition to the transmembrane segments the α subunit has an extensive cytoplasmic carboxy tail (S7–S10) containing a number of regulatory domains. Three distinct Ca²⁺/divalent metal ion binding sites (RCK 1 and 2) and the ‘Ca²⁺ bowl’) have been functionally identified by site-directed mutagenesis in this C-terminal region of the α subunit (Figure 1A). The RCK domains in holo-BK_Ca channels are thought to assemble into “gating rings” beneath the cytosolic mouth of the conduction pore. Direct Ca²⁺ binding to the high affinity sites within the RCK1 gating ring, in addition to the Ca²⁺ bowl site, is thought to trigger channel opening via an allosteric coupling mechanism. The α subunit may also exhibit Ca²⁺ sensing domains within the membrane spanning domains as Ca²⁺ sensitivity has been shown to persist following deletion of the cytoplasmic sequence [16]. Other studies however, using chimeric channels, have shown the C-terminus to be necessary for Ca²⁺ sensitivity of the α-subunit [17]. In addition to regulation by Ca²⁺, the carboxy tail contains phosphorylation sites and RCK1, in particular, has been implicated in the direct gating actions of protons, carbon monoxide and reactive oxygen species[18].

Co-associated with the main channel complex are auxiliary β subunits, which are presumed to be present in a 1:1 ratio with α subunits (Figure 1). This pattern of BK_Ca subunit co-expression and stoichiometry is thus similar to other members of the large family of voltage-gated K⁺ (Kv) channel complexes that are typically comprised of α and β subunits [19]. Four distinct genes have been identified for BK_Ca β subunits (KCNMβ1–4). Thus, four beta subunits (β1–4), have been described with β1 being the principal form found in vascular smooth muscle, with β2 (chromaffin cells, brain, lung), β3 (testis, pancreas, spleen) and β4 (neuronal, lung) generally considered to be restricted to other tissues [20–22]. Structurally, the β subunit exists as 2 transmembrane segments and an extracellular loop with both carboxy and amino termini being intracellular. Further, it is now evident that the BK_Ca α subunit’s S0 segment is critically important for interaction and functional regulation of channel gating by the auxiliary β subunit [23,24].

Functionally, BK_Ca β subunits are reported to influence the kinetics and calcium sensitivity of channel gating [25–29], the responses of BK_Ca activity to pharmacologic modulators [30–34], and the trafficking of channels to the plasma membrane [35–37]. Although differing subunit stoichiometries (i.e. α:β subunit ratio) may exist, a full complement of four β1 subunits are required for maximal effect [38]. Genetic deletion of the β1 subunit in mice leads to hypertension and increased contractility of vascular smooth muscle [39,40], whereas a gain-of-function mutation in the β1 subunit is associated with lower blood pressure in humans [41].

ii. Splice Variants

The BK_Ca α subunit is translated from a single gene [42] denoted KCNMA1 that contains 27 distinct exons [43], thus the presence of multiple exon boundaries in the pre-messenger RNA provides ample opportunity for structural modification of the final protein product through alternative splicing of the initial gene transcript. Up to 13 alternative splice sites have been described in mammalian BK_Ca α subunit mRNA. Such variants have been shown to contribute to differences in the regulatory properties of the channel due to variability in responses to steroids, cyclic nucleotides and availability of phosphorylation sites. One such splice variant of the α subunit involves the STRess axis regulated EXon (STREX) which changes the response to cAMP from stimulatory to inhibitory relative to the variant lacking this exon (termed ZERO). Importantly, the expression of different splice variants between tissues could also contribute to differences in voltage sensitivity of the channel [44] with STREX-1 showing a leftward shift in half-maximal voltage for activation (i.e. V_1/2 shifted 20 -30 mV more negative [45] to a level of Em consistent with that observed in cannulated arterioles [46,47]. To date, BK_Ca channels cloned from smooth muscle typically do not contain the STREX sequence, which is consistent with functional data demonstrating that activation of the smooth muscle cAMP/PKA signaling pathway enhances BK_Ca channel opening, leading to relaxation [48]. Comparisons of the predicted amino acid sequences of smooth muscle BK_Ca α subunit clones further reveal strong conservation in primary sequence throughout the protein, with the most diversity occurring at the extreme C-terminus. With regard to the β subunit, while considerable variation exists in primary structure between species (reference) there is little information regarding splice variants of the β1 subunit, the predominant beta subunit in vascular smooth muscle. In contrast splice variants of β3 have been reported [49].

iii. Regulation of BK_Ca by Ca²⁺ and Voltage

BK_Ca channels belong to the family of voltage-gated K⁺ channels as shown by both primary sequence analysis and their biophysical behavior [50–52], however, the presence of intrinsic binding sites for intracellular divalent metals (i.e. Ca²⁺ and Mg²⁺) affords an independent and synergistic mechanism of activation via stimulus-evoked elevations in cytosolic free Ca²⁺ concentration. Detailed biophysical characterization of BK_Ca channel gating by Aldrich and colleagues, along with Stefani and co-workers, have demonstrated that BK_Ca channel opening may be induced by voltage alone, in the absence of calcium, or solely by cytosolic free calcium, under conditions in which the voltage sensor domains remain inactive [52,53]. At resting levels of cytosolic Ca²⁺ (i.e. 100 nM), very positive voltages are required to achieve significant BK_Ca channel opening (i.e. half-maximal voltage of activation is >100 mV), due to the inherently weak voltage sensitivity of these channels. With increasing cytosolic free [Ca²⁺], however, this voltage-dependent activation is shifted left-ward along the voltage axis to more negative membrane potentials [50,54], so that more channel activity is observed within the physiologic range of membrane potentials. Calcium thus appears to enhance the voltage-dependent activation of BK_Ca channels, such that channel opening will be greater at any given membrane potential with increasing levels of cytosolic Ca²⁺. This mechanism is particularly important in smooth muscle cells of myogencially active resistance arteries, which do not typically use action potentials to produce graded changes in tone and display depolarized membrane potentials in the range of −40 to −30 mV. In this situation, stimulated rises in cytosolic free Ca²⁺ will be the major determinant of cellular BK_Ca activity, as channel open probability over this negative voltage range will only be ~0.1% (at an intracellular Ca²⁺ level of 100 nM). Thus, in vascular smooth muscle, BK_Ca channels appear to operate primarily as ligand-gated channels, with membrane potential in the physiologic range producing only a modest level of background activity upon which Ca²⁺ elevations evoke further increases. Functionally, the smooth muscle-associated BKβ1 subunit significantly increases the activity of the holo-channel, primarily by enhancing the apparent sensitivity of channel activation by Ca²⁺ [25,27,55,56]. Smooth muscle BK_Ca channels assembled from α + β1 subunits will thus have greater activity at a given [Ca²⁺] compared to channels containing only α subunit, as well as channels containing less than a 1:1 ratio of β1 to α subunits [38].

Spatial Considerations in BK_Ca Activity

The local environment in which BK_Ca resides in part determines its function within a given tissue. For example, efficient activation by locally produced Ca²⁺ transients, or sparks (via ryanodine-sensitive release channels) requires a close spatial arrangement between SR and plasma membranes (Figure 1B). In smooth muscle the distance between the plasma membrane and superficial SR has been reported to be less than 20 nm [57]. Facilitating such a spatial arrangement is the superficial SR found in smooth muscle and possibly plasma membrane invaginations or caveolae. The α-subunit of BK_Ca is known to contain two caveolin binding sites and BK_Ca has been demonstrated to associate with caveolin 1 in cultured endothelium [58] and myometrium [59]. In cerebral arteries, caveolae have been shown to be functionally important in coupling Ca²⁺ sparks to spontaneous transient outward currents (STOCs) (Lohn et al., 2000; Drab et al., 2001) presumably, which are presumably mediated through the opening of a population/cluster of BK_Ca channels [60]. Thus, genetic ablation of caveolin-1 was shown to lead to both a loss of caveolae and a decrease in STOC frequency [61] while cholesterol depletion with β-methyl cyclodextrin decreased spark frequency, amplitude and spatial dimensions by ablating caveolae and possibly disrupting the local coupling between plasmalemmal VGCC and RyR [62]. A consequence of this architecture is that subpopulations of BK_Ca may exist [63] to perform specific functions independently of the more global BK_Ca complement.

Additional localized regulation of BK_Ca activity may result from the colocalization of other signaling molecules (for example, c-Src; G-protein coupled receptors) and cytoskeletal elements (for example, actin) [64]. Similarly, targeting of BK_Ca to specific membrane domains has also been implicated in the functional coupling of this population of BK_Ca to other ion channels, including cation channels comprised of TRP proteins and voltage-gated Ca²⁺ channels [65,66].

iv. Post-Translational Modification

Regulation of BK_Ca conductance is also exerted via direct phosphorylation by a number of protein kinases (including PKA, PKG, PKC, Src) and BK_Ca is reported to have a large number of potential phosphorylation sites [67]. Importantly, a number of vasodilator and vasoconstrictor stimuli utilize phosphorylation-mediated mechanisms to regulate ion channel activities. In general, VSM cell BK_Ca is activated by PKA [48,68] and PKG [48,69] while being inhibited by PKC (Schubert et al., 1999) and the tyrosine kinase cSrc [70,71], although the latter effect is controversial [72,73]. It is apparent, however, that tissue specific differences exist in responses to various protein kinases, as pulmonary artery smooth muscle BK_Ca is activated by particular isoforms of PKC [74]. Cross-talk between cyclic nucleotide-dependent systems may also occur as cAMP-dependent activation of BK_Ca has been reported to occur via cGMP-dependent protein kinase G [75]. Similarly, the activation of BK_Ca in pulmonary artery smooth muscle by PKC is reported to occur via a cGMP-dependent mechanism [74]. Collectively, these kinase-based mechanisms allow for modulation of BK_Ca open probability by a number of vasoactive substances. Biophysically, vasodilators acting through cAMP or cGMP tend to increase channel open probability by shifting voltage-activation curves towards more negative membrane potentials [48]. In VSM of skeletal muscle arterioles, such pathways could provide a mechanism whereby a relatively quiescent channel could be activated during metabolic activity to oppose the normally high resting vascular resistance [47,76].

In addition to phosphorylation, BK_Ca is potentially regulated by a number of other post-translational modifications. For example, palmitoylation of a cysteine-rich domain within the STREX insert can potentially affect membrane targeting/anchoring, trafficking and signaling [77]. Interestingly, when the palmitoylated STREX variant is phosphosphorylated by PKA, the C-terminus of the α-subunit alters its interaction with the plasma membrane and channel activity is inhibited [77]. These observations highlight the complexity of BK_Ca regulation and the potential for interaction between such mechanisms.

BK_Ca is also modulated by a number of small molecules including steroids, fatty acids, PIP₂ and reactive oxygen species [18,34,78]. In the case of steroid hormones, modulation of BK_Ca activity has been shown to occur through genomic [79] and/or non-genomic [33,34] effects, with the latter being mediated via cGMP in the case of estrogen-induced relaxation of coronary arteries [80]. In cerebral arteries it is likely that the increase in cGMP occurs secondary to increased NO production [81] although NO has been suggested to exert direct effects on BK_Ca [82]. Estrogen - like molecules, including 17β-estradiol and tamoxifen exert a direct non-genomic effect to cause enhanced opening of BK_Ca through an action on the β1 subunit of BK_Ca [33,34]. Genomically-mediated effects include control of expression of spice variants. During pregnancy, the hormonal environment dictates alternate splicing patterns of the BK_Ca α subunit expression, with the STREX variant diminishing and PKA signaling switching from inhibitory to excitatory [79]. Similarly, splice variant expression (i.e. STREX) is regulated by stress hormones of the hypothalamic – adrenal axis [83].

BK_Ca and the Arteriolar Myogenic Response

Introduction to the Myogenic Response

The term ‘myogenic response’ refers to the ability of arterioles to respond to changes in intraluminal pressure. Specifically vasoconstriction follows an increase in pressure while vasodilation occurs in response to a pressure decrease. The physiological significance of myogenic responsiveness relates to its role in the local autoregulation of microvascular blood flow and provision of a level of basal vascular tone (resistance) upon which vasodilators and vasoconstrictors can act. A further role relates to the control of capillary pressure, affording protection against pressure-induced capillary leakage and the formation of edema. This mechanically-induced vasomotor response has been shown to be a function inherent to vascular smooth muscle and is independent of the endothelium or neural input. While the exact signaling mechanisms underlying myogenic vasoconstriction are not completely defined, a plausible pathway involves activation of non-selective cation and/or stretch-activated channels, membrane depolarization, opening of voltage-gated Ca²⁺ channels, Ca²⁺ influx and activation of the contractile proteins (Figure 2). These events are likely supported by a number of other mechanisms, including the rearrangement of cytoskletal elements, Ca²⁺ release from the sarcoplasmic reticulum and increased Ca²⁺ sensitivity of the contractile elements (Figure 2). As changes in Em and global intracellular Ca²⁺ appear central to myogenic constriction, considerable interest has been shown in the potential influence of dynamic BK_Ca activity. Further detail regarding myogenic signaling mechanisms can be found in recent reviews [84–87].

Figure 2.

Schematic diagram illustrating signaling mechanisms underlying arteriolar myogenic vasoconstriction. Emphasis is placed on ion channels, membrane depolarization, mobilization of Ca²⁺ and activation of the contractile proteins. In addition, negative feedback activation of BK_Ca by SR-mediated Ca²⁺ sparks is shown (from Hill et al.[124]).

Mechanisms Linking BK_Ca to Myogenic Vasoconstriction

i. Modulation of BK_Ca by stretch (direct and indirect)

Several studies have examined the direct stretch sensitivity of BK_Ca in smooth muscle. The potential relevance of such a mechanism to the myogenic response is that mechanical stretch, during an alteration in intraluminal pressure, would alter BK_Ca channel activity thereby impacting Em and Ca²⁺ entry via voltage-gated Ca²⁺ channels. Kirber et al (1992) and Dopico et al (1994) in studies (pulmonary and mesenteric arteries, respectively) at both the whole cell and excised patch levels reported that stretch exerts a direct stimulatory effect on BK_Ca channel opening. The data from excised membrane patches were used to exclude a role for SR-mediated Ca²⁺ release. These data would not, however, be consistent with mechanical activation of BK_Ca being involved in myogenic contraction, per se, as stretch would induce hyperpolarization leading to relaxation. In studies of coronary arteriolar smooth muscle cells, [88] showed that the stretch-induced activation of BK_Ca channels was largely overridden by the concurrent activation of an inward cation current. Thus, collectively, these studies support direct stretch activation of BK_Ca as acting as a brake on stretch-induced depolarization to limit the magnitude of depolarization and contraction.

A number of alternative mechanisms indirectly linking BK_Ca to a stretch stimulus have been proposed. An increase in intraluminal pressure has been shown to lead to the production of the arachidonic acid metabolite 20-HETE, which causes membrane depolarization by closure of BK_Ca [89]. Interestingly, arachidonic acid and other metabolites of this fatty acid have been shown to cause an increase in opening of BK_Ca [90]. A possible explanation for these differences may relate to specific actions such as 20-HETE-induced activation of PKC [91], as this kinase has been reported to inhibit BK_Ca. Alternatively, differences may relate to the physico-chemical properties of lipids which have been shown to dictate both inhibitory and stimulatory actions on BK_Ca [92].

A further mechanism which may link BK_Ca to the pressure/stretch-induced mechanical stimulus involves an action on cell surface integrins mediated through extracellular matrix proteins. Integrins have been implicated in both myogenic responsiveness and activation of BK_Ca. In particular, α₅β₁ integrin activation with fibronectin was shown to enhance BK_Ca opening in arteriolar smooth muscle cells [93] through a mechanism dependent upon cSrc [73]. Blockade of α₅β₁ integrin with specific antibodies, however, inhibits myogenic constriction in intact arterioles [94].

ii. Ca²⁺ entry

Direct activation of BK_Ca in arteriolar smooth muscle by stretch-induced Ca²⁺ entry is unlikely. Ca²⁺ levels in the vicinity of the channel are required to reach 5 – 10 µM to markedly affect channel opening, whereas smooth muscle global cytosolic Ca²⁺ concentrations remain in the range 100 – 300 nM in cannulated arterioles over a physiological range of intraluminal pressures. It is further unlikely that this level of Ca²⁺, in combination with the expected pressure-induced depolarization, would significantly alter BK_Ca opening.

Clustering of voltage-gated Ca²⁺ channels in a membrane domain in which BK_Ca channels are co-localized could provide a mechanism for focal increases in Ca²⁺, conceivably of significant magnitude to stimulate BK_Ca opening. Consistent with this possibility Ca²⁺ sparklets (resulting from Ca²⁺ entry through clustered VGCCs) have been identified in arterial smooth muscle cells [95,96], while other studies have suggested a functional co-localization of BK_Ca and VGCC in caveolae [97]. Data are not currently available, however, linking these events to pressure-induced activation of BK_Ca in intact arterioles.

Ca²⁺ entry may also indirectly facilitate BK_Ca activation through increased filling of SR stores and increasing the sensitivity of RyR to subsequent CICR. In cerebral artery smooth muscle cells, Ca²⁺ sparks and Ca²⁺ waves are increased in frequency by depolarization (30 mM KCl) and inhibited by blockade of voltage-gated Ca²⁺ channels [98]. Similarly, increasing intraluminal pressure from 10 to 60 mmHg in isolated cerebral arteries (a stimulus expected to cause both depolarization and Ca²⁺ entry via VGCC) caused an approximate 2.5 fold increase in the frequency of Ca²⁺ sparks and waves. Whether a similar situation applies in all vascular beds is uncertain. In cremaster muscle, elevating intraluminal pressure increases the frequency of Ca²⁺ waves, and this relationship persists in the presence of nifedipine, but is blocked by putative inhibitors of non-selective cation channels [99]. Conceivably this may point to some of the effects of increasing Ca²⁺ entry being independent of the precise entry mechanism.

iii. Ca²⁺ sparks and SR-mediated events

While myogenic constriction lowers pressure distal to the point of constriction there is a tendency for upstream pressure to increase. The result of this situation is that myogenic constriction can act as a positive feedback system with the potential for uncontrolled vasoconstriction. Under physiological conditions, this is not observed, however, as myogenic constriction exhibits upper and lower pressure limits while the increased pressure may also activate hyperpolarizing mechanisms that oppose the pressure-induced depolarization and contraction. A candidate hyperpolarizing pathway involves the large conductance, Ca²⁺-activated, K⁺ channel (BK_Ca).

The idea that BK_Ca could be activated as a repolarizing event thus opposing the action of vasoconstrictor stimuli evolved from the work of Benham and colleagues [60,100]. Nelson and colleagues [101,102] subsequently implicated BK_Ca as a negative feedback mechanism limiting pressure-induced or myogenic constriction. Activation of BK_Ca was hypothesized to occur in response to focal Ca²⁺ release (Ca²⁺ sparks) from the SR. Specifically, a spark was proposed to allow local Ca²⁺ concentrations below the plasma membrane to reach levels approaching 10 µM for brief periods. Such high local Ca²⁺ concentrations would be sufficient to activate a cluster of BK_Ca channels (giving rise to spontaneous transient outward currents) while not contributing markedly to global cytosolic Ca²⁺ levels and contraction. In fact Ca²⁺ sparks are viewed as activating hyperpolarizing current which leads to closure of voltage-gated Ca²⁺ channels and vasodilation.

The idea that Ca²⁺ sparks oppose myogenic constriction through opening of BK_Ca would appear to be at odds with the hypothesis that an increase in intraluminal pressure causes the production of the arachidonic acid metabolite, 20-HETE, resulting in closure of BK_Ca with subsequent depolarization and contraction [89,103]. Interestingly, evidence for both Ca²⁺ sparks and 20-HETE production has been provided in rat cerebral blood vessels [98,102,103]. The precise relationship between these apparently opposing pathways, however, remains unclear, although it is questionable that both mechanisms could be simultaneously present, even if arguments of spatio-temporal separation are proposed.

Heterogeneity in BK_Ca Function

i. Variation at the Level of the Channel

As outlined above, a number of characteristics of the BK_Ca channel are consistent with the possibility of vascular heterogeneity. Differences in subunit stoichiometry, isoform expression or the expression of splice variants could result in channels of varying Ca²⁺ sensitivity and responsiveness to post translational modifications including phosphorylation.

On the basis of differences in the relationships between intraluminal pressure, level of smooth muscle Em, and observed myogenic tone in small cerebral arteries and arterioles from skeletal muscle of the rat, it has been hypothesized that these two tissues may exhibit differing contributions of BK_Ca. At the tissue level, it was further suggested that these differences allowed arterioles in resting skeletal muscle to maintain a relatively high level of vascular resistance [47]. This idea was supported by the observation that cremaster muscle arterioles showed relatively small changes in Em and diameter in response to IBTX and that the extent of the constriction (i.e. effectiveness of IBTX) was apparently independent of the level of intraluminal pressure. Jackson and Blair [76] similarly reported a comparative lack of effect of IBTX on the in vivo resting diameters of cremaster muscle second and third order arterioles (diameter approximately 20 – 40 µm) from hamster while Frisbee et al[104] found that IBTX did not alter myogenic responsiveness in isolated gracilis muscle from normotensive Dahl rat. Similarly, both in vivo and in vitro studies of renal [105] and coronary arteries [106] have shown little effect of BK_Ca inhibitors (IBTX and penitrem A) on resting vascular resistance. Collectively, these observations appear to provide evidence against BK_Ca being recruited as a general negative feedback mechanism as pressure and hence myogenic tone increase.

To examine the apparent differences in BK_Ca between two different vascular beds, comparative electrophysiological studies were performed using patch clamp on freshly isolated VSM cells. Relative to VSM cells from cerebral arteries, VSM from skeletal muscle arterioles displayed a lower density of IBTX-sensitive K⁺ current while Kv-associated current appeared similar. This difference was particularly evident at high pipette Ca²⁺ levels (5 µM), consistent with the observed difference in whole-cell K⁺ current being mediated by differences in BK_Ca. Further supporting a decreased contribution of BK_Ca was the observation that STOCs were less frequent and of smaller amplitude in VSM cells from skeletal muscle arterioles compared to cells from cerebral arteries despite identical recording conditions (Figure 3). Pharmacological studies suggested a difference at the level of the β1 regulatory subunit, as skeletal muscle VSM showed diminished sensitivity to acute estrogen exposure compared to cerebral VSM while both vasculatures showed comparable increases in K⁺ current in response to NS-1619. Estrogen had previously been shown to exert an enhancing effect on BK_Ca opening via the β1 subunit, while NS-1619 acts through the α subunit. A relative decrease in expression of BK_Ca was shown both at the mRNA and protein levels in skeletal muscle VSM. Further to this, the ratio of β1:α subunit protein expression in cremaster VSM was shown by Western blotting to be decreased. The importance of this relates to the fact that a simple reduction in protein might not necessarily explain the differences in BK_Ca characteristics, due to its known large conductance while the differences in subunit composition are consistent with the observed heterogeneity involving Ca²⁺ and Em sensitivities. Interestingly, in vitro treatment of cerebral artery VSM with siRNA directed at the β1 subunit causes BK_Ca function to more closely resemble that of native skeletal muscle muscle VSM [107].

Figure 3.

Characteristics of STOCs in VSM cells enzymatically isolated from cremaster muscle arterioles and small cerebral arteries. The upper panels show example STOC tracings and illustrate that despite differences in apparent voltage dependency, ryanodine eliminates the outward currents in both cell preparations. Recordings were made at a pipette [Ca²⁺] of 100 nM. The lower panel shows that at any given holding potential STOCs are both less frequent and of smaller amplitude in cremaster VSM cells compared to those from cerebral arteries. Data are re-plotted from Yan et al and are shown as mean SEM for n = 6 – 8 cells.

To provide additional insight into differences in BK_Ca function in skeletal muscle and cerebral artery VSM cells STOCs were recorded at various holding potentials. STOCs have previously been suggested to arise from the opening of clusters of BK_Ca [108] and reflect the simultaneous opening of a variable number of individual channels [108–110]. At any holding potential cerebral VSM cells showed an increased frequency of STOCs and further the outward currents were greater in amplitude compared to those recorded in skeletal muscle VSM cells [111] (Figure 3). From the comparative data, it can be seen that skeletal muscle VSM cells require approximately 30 mV additional depolarization (compared to cerebral VSM cells) to generate a comparable STOC frequency. Although STOCs could be evoked in cerebral VSM cells at levels of Em approximating the physiological range, cells from skeletal muscle muscle required considerably more positive holding potentials (Figure 3).

ii. Variation in Activation of the Channel

As outlined above, Ca²⁺-mediated activation of BK_Ca is thought to occur as a result of focal Ca²⁺ release in the form of ‘sparks’ from the SR. Ca²⁺ sparks comparable to those in cerebral VSM cells do not, however, appear to be found in similarly prepared cells from skeletal muscle arterioles[111]. However, as STOCs are inhibited in skeletal muscle VSM cells by both IBTX and ryanodine, it is evident that, at least under the conditions of whole cell patch clamp, a BK_Ca activation mechanism involving SR Ca²⁺ release does exist (Figure 3). The question is, why are comparable Ca²⁺ spark events not identified in the two cell preparations? The difference does not appear to relate to methods used for isolating the cells nor differences in resting Em as determined using a perforated patch recording strategy under current clamp conditions. Differences therefore reflect variation in how the VSM cell preparations respond to the patch clamp conditions or the occurrence of Ca²⁺ release events in skeletal muscle VSM cells that were below the limits of detection for the approach used. Regardless, as both cell types were investigated using the same experimental strategy, this suggests differences in the exact cellular mechanism by which the SR is coupled to activation of BK_Ca.

The possibility that Ca²⁺ release events are yet to be detected in some VSM cell types, including those of skeletal muscle muscle arterioles, is highly likely. It is clear that there is considerable variation in spatiotemporal aspects of Ca²⁺ signaling [112]. Further, even in cerebral VSM cells, approximately 50% of BK_Ca current was not associated with typical Ca²⁺ sparks [113]. Interestingly, the currents not associated with sparks tended to be of smaller amplitude, perhaps more consistent with the situation in cremaster VSM, where sparks were not detected and STOCs were of smaller amplitude (at a given Em) compared to cerebral VSM cells yet the observed STOCs were inhibited by depletion of SR Ca²⁺ with ryanodine (Figure 3).

An alternate explanation is that BK_Ca may be activated by distinct mechanisms in differing vascular beds. Variation in phosphorylation–mediated mechanisms perhaps may result from tissue – specific expression of α subunit splice variant. Similarly, regional variation in the production of vasoactive factors that activate BK_Ca may contribute to apparent functional heterogeneity of the channel. In this regard Hercule et al [114] have demonstrated that the vasodilator 17,18-epoxyeicosatetrenoic acid stimulates BK_Ca by an action on the α subunit. Further, these authors showed that this action occurred despite the presence of low Ca²⁺ concentrations (100 nM) or genetic deletion of either RyR3 or BK_Ca β1, indicating that neither local nor global Ca²⁺ increases were required.

iii. Other Possible Factors Contributing to BK_Ca Heterogeneity Between Vascular Beds

In addition to differences in the contribution of the β1 subunit to BK_Ca function between vascular beds it is conceivable that heterogeneity arises from the existence of tissue-specific expression of splice variants. Such expression patterns may subsequently confer differences in response to vasoactive stimuli, which post-translationally regulate BK_Ca (for example, via phosphorylation) as well as determining cellular localization and trafficking. To date, however, relatively little data specific to arteriolar smooth muscle are available. Consistent with this possibility, Poulsen et al. [115] have recently shown C-terminal splice variants in rat cerebral blood vessels. Further, these authors suggest that while the β1 subunit is the predominant form in the vessel wall of cerebral arteries the β2 and β3 isoforms can also be detected. These results await confirmation in purified VSM cell preparations and extension to other vascular beds.

It is currently also unknown whether the same cellular architecture exists in all vascular smooth muscle cells or if variation affecting the exact localization of BK_Ca occurs between vascular beds. As the extent of peripheral SR has been reported to vary between blood vessels [116–118] it may be expected that the relationship between the SR, Ca²⁺ sparks and BK_Ca-mediated STOCs also show regional variation. Consistent with this, the coupling efficiency between Ca²⁺ sparks and STOCs has been reported to vary between smooth muscle tissues [119].

Similarly, differences in coupling between the SR and BK_Ca between vascular beds could also relate to variation in the expression, or cellular distribution, of ryanodine receptors. In regard to this, little is known as to whether there are quantitative differences in the various ryanodine receptor isoforms (RyR 1, 2 and 3) between cerebral and cremaster muscle VSM cells.

Although not discussed in detail in this review, variation in the role of BK_Ca in the vasculature may exist between species and as a result of differing genetic backgrounds within strains of the same species. In relation to the latter point, Stadnickna and colleagues have reported that differences in the expression and function of BK_Ca between normotensive Dahl salt-sensitive and Brown Norway rats underlies differences in the cardiovascular responsiveness to the anesthetic agent, propofol [120]. Specifically, differences in chromosome 13 resulted in an increased expression of BK_Ca (α-subunit) and enhanced hyperpolarization of mesenteric artery smooth muscle cells in the Dahl rats. Similarly, age-dependent changes in BK_Ca expression and function may contribute to differences in the physiological role of the channel within the cardiovascular system [121–123].

Conclusion

Apparent differences in the structure and function of BK_Ca in VSM cells of skeletal muscle compared to cerebral vasculature raise challenges to the classical view of Ca²⁺ sparks as the primary mechanism for channel activation and dilation in all arteries. The complex array of mechanisms that regulate BK_Ca in vascular smooth muscle appear likely to contribute to this regional heterogeneity in channel function and may allow the channel function to be adapted to local requirements. Current challenges important to understanding the physiological significance of such variation include identifying vascular bed-specific differences in channel subunit composition, BK_Ca regulation by post-translational mechanisms and its exact relationship to SR-mediated Ca²⁺ release. In regard to the involvement of BK_Ca in the regulation of pressure-induced membrane depolarization and contraction, such heterogeneity may be utilized to allow for variation in the regional control of vascular resistance to match local hemodynamics with metabolic requirements. Further, knowledge of the molecular and cellular basis for tissue heterogeneity will be important to both exploiting BK_Ca as a potential therapeutic target and understanding the systemic effects of pharmacologically manipulating this K⁺ channel.

Abbreviations

BK_Ca

Large conductance, Ca²⁺ - activated, K⁺ channel

CICR

Ca²⁺ - induced Ca²⁺ release

EC

Endothelial cell

Em

Membrane potential

IBTX

Iberiotoxin

IK

Intermediate conductance, Ca²⁺ - activated, K⁺ channel

IP3

Inositol trisphosphate

MLC

Myosin light chain

PIP₂

Phosphatidylinositol 4,5 – bisphosphate

PLC

Phospholipase C

RCK

Regulator of conductance for K+

ROS

Reactive oxygen species

RyR

Ryanodine receptor

SK

Small conductance, Ca²⁺ - activated, K⁺ channel

SR

Sarcoplasmic reticulum

STOC

Spontaneous transient outward current

STREX

STRess axis regulated EXon

VGCC

Voltage-gated Ca²⁺ channel

VSD

Voltage sensor domain

VSM

Vascular smooth muscle