Physiological functions and pathological involvement of ion channel trafficking in the vasculature

Abstract

A variety of ion channels regulate membrane potential and calcium influx in arterial smooth muscle and endothelial cells to modify vascular functions, including contractility. The current (I) generated by population of ion channels is equally dependent upon their number (N), open probability (Po) and single channel current (i), such that I=N.P_O.i. A conventional view had been that ion channels traffic to the plasma membrane in a passive manner, resulting in a static surface population. It was also considered that channels assemble with auxiliary subunits prior to anterograde trafficking of the multimeric complex to the plasma membrane. Recent studies have demonstrated that physiological stimuli can regulate the surface abundance (N) of several different ion channels in arterial smooth muscle and endothelial cells to control arterial contractility. Physiological stimuli can also regulate the number of auxiliary subunits present in the plasma membrane to modify the biophysical properties, regulatory mechanisms and physiological functions of some ion channels. Furthermore, ion channel trafficking becomes dysfunctional in the vasculature during hypertension, which negatively impacts the regulation of contractility. The temporal kinetics of ion channel and auxiliary subunit trafficking can also vary depending on the signaling mechanisms and proteins involved. This review will summarize recent work that has uncovered the mechanisms, functions, and pathological modifications of ion channel trafficking in arterial smooth muscle and endothelial cells.

Graphical Abstract

Ion channel trafficking in smooth muscle and endothelial cells regulates arterial contractility. SMC, smooth muscle cell; EC, endothelial cell, IEL, internal elastic lamina. Created with BioRender.

INTRODUCTION

Resistance-size arteries and arterioles regulate regional organ blood flow and total peripheral resistance to establish systemic blood pressure. The contraction of smooth muscle cells within the vascular wall reduces luminal diameter, restricts regional blood flow and increases systemic pressure. In contrast, smooth muscle cell relaxation leads to vasodilation which increases flow and decreases blood pressure. A wide variety of stimuli, including intravascular pressure and vasoactive substances, directly regulate arterial smooth muscle cell contractility. Endothelial cells line the wall of all blood vessels and modulate smooth muscle cell contractility through the release of several diffusible substances and direct electrical coupling mediated via gap junctions.

Membrane potential is a key determinant of arterial contractility (Nelson et al., 1990). Depolarization activates voltage-dependent calcium (Ca²⁺) channels (Ca_V1.2) which are located on the plasma membrane of smooth muscle cells, leading to an increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) and vasoconstriction. In contrast, membrane hyperpolarization reduces [Ca²⁺]_i, resulting in vasodilation (Nelson et al., 1990). Arterial smooth muscle and endothelial cells both express a wide variety of ion channels that regulate membrane potential and plasma membrane Ca²⁺ influx. Several surface ion channels that are expressed in the vasculature, including Ca_V1.2 and large-conductance Ca²⁺-activated potassium (BK) channels, couple to auxiliary β and γ subunits (Hullin et al., 1992; Brenner et al., 2000; Bannister et al., 2009; Evanson et al., 2014). These subunits can modify the trafficking, biophysical properties and regulatory mechanisms of ion channels. Thus, vascular ion channels are important regulators of contractility, regional organ blood flow and systemic blood pressure.

The current (I) generated by population of ion channels is equally dependent upon their number (N), open probability (P_O), and single channel current (i), such that I=N.P_O.i. Physiological stimuli, including intravascular pressure, receptor ligands and other vasoactive substances regulate the open probability (P_O) of a variety of different surface ion channels in both smooth muscle and endothelial cells (Nelson et al., 1990; Quayle et al., 1997; Earley & Brayden, 2015; Foster & Coetzee, 2016; Dopico et al., 2018). A conventional view had been that the number of surface ion channels was fixed and not open to modification by vasoregulatory stimuli. Similarly, it was considered that ion channels assemble with their auxiliary subunits prior to anterograde surface-trafficking of the heteromeric complex to the plasma membrane. Recent studies have demonstrated that vasoactive stimuli can regulate the surface abundance of several different ion channels in both arterial smooth muscle and endothelial cells. It has also been shown that physiological stimuli can independently regulate the trafficking of ion channels and regulatory subunits to alter their biophysical properties and functions. Ion channel trafficking is now also recognized to become dysfunctional during vascular diseases, which negatively impacts arterial contractility. This review will summarize recent work that has identified the mechanisms, physiological functions and pathological modifications of ion channel trafficking in arterial smooth muscle and endothelial cells. The reader is referred to excellent earlier review articles which summarize the molecular and cellular mechanisms of protein trafficking in cells (Maxfield & McGraw, 2004; Grant & Donaldson, 2009; Stenmark, 2009; Hutagalung & Novick, 2011; Mizuno-Yamasaki et al., 2012).

ION CHANNEL TRAFFICKING IN ARTERIAL SMOOTH MUSCLE CELLS

BK channels

BK channels form from the homotetrameric assembly of seven transmembrane domain (S0-S6) pore-forming α subunits which are encoded by the KCNMA1 gene (Butler et al., 1993). The transmembrane segments S1-S4 confer channel voltage-sensitivity, whereas S5-S6 forms the pore (Tao et al., 2017). BK channels are voltage-and Ca²⁺-dependent and have a large single channel conductance (Torres et al., 2014). BK channel accessory subunits include β (β1–4) and γ (γ1–4) proteins (Yan & Aldrich, 2012). Vascular smooth muscle cells express β1 subunits, which increase BK channel apparent Ca²⁺-sensitivity and slow channel activation and deactivation (Torres et al., 2014). Vascular smooth muscle cells also express γ1 (also termed LRRC26), which increases BK channel voltage-sensitivity and apparent Ca²⁺ sensitivity to produce vasodilation (Evanson et al., 2014).

BK channel β1 subunits

Only approximately 10% of the total amount of β1 subunit protein was located in the plasma membrane of human and rat smooth muscle cells of resistance-size arteries (Leo et al., 2014; Leo et al., 2015; Leo et al., 2017; Zhai et al., 2017). This finding suggested that arterial smooth muscle cells contain an intracellular pool of β1 subunits that could be mobilized to increase the amount of surface protein. Intracellular β1 subunits were stored in recycling endosomes which were positive for Rab11A, a small Rab GTPAse, in arterial smooth muscle cells (Leo et al., 2014) (Fig. 1). Several vasodilatory stimuli rapidly (seconds) increased surface β1 protein, including carbachol, a muscarinic receptor agonist which stimulates endothelial cell nitric oxide (NO) production, iloprost, a PGI₂ analog, nitric oxide donors which stimulate cGMP-dependent protein kinase (PKG), and forskolin, an adenylyl cyclase activator which stimulates cAMP-dependent protein kinase (PKA) (Leo et al., 2014) (Fig. 1). This rapid increase in surface β1 could have occurred due to the stimulation of anterograde trafficking or inhibition of internalization. Supporting an anterograde trafficking mechanism, NO increased Rab11A activity in cerebral arteries (Zhai et al., 2017). Data obtained using pharmacological trafficking inhibitors, Rab11A knockdown and constitutively-active and dominant-negative Rab11A mutants indicated that Rab11A-dependent anterograde trafficking was the primary mechanism by which NO elevated surface β1 protein (Leo et al., 2014). In contrast, Rab11B was not involved in β1 subunit trafficking in arterial smooth muscle cells, suggesting that the recycling endosomes which deliver β1 subunits are regulated by distinct Rab subtypes (Leo et al., 2014). Inhibition of protein internalization also increased surface β1 subunit protein, indicating that β1 subunits constitutively recycle in arterial smooth muscle cells (Leo et al., 2014). Thus, vasodilator stimuli may not only stimulate anterograde trafficking of β1 subunits, but also inhibit their internalization to elevate the amount of surface protein (Leo et al., 2014). PKG and PKA may directly phosphorylate Rab11A or other regulatory protein(s) which then stimulate anterograde trafficking of recycling endosomes. NO stimulated β1 subunit trafficking more than did a constitutively active Rab11A mutant (Rab11AQ70L). This finding suggested that PKG activation increases surface β1 protein through a mechanism additional to Rab11A in arterial smooth muscle cells (Leo et al., 2014). An increase in plasma membrane cholesterol also rapidly increased surface β1 subunits through a trafficking-dependent mechanism, without altering surface BKα protein in cerebral arteries (Bukiya et al., 2021). This increase in surface β1 subunits switched the regulation of BK channels by cholesterol from inhibition to activation in arterial smooth muscle cells.

Figure 1.

Signaling mechanisms which regulate BK channel α and β1 subunit trafficking in arterial smooth muscle cells and functional significance. Ang II, angiotensin II; ET-1, endothelin 1; PKC, protein kinase C; PKG, protein kinase G; RE, recycling endosome; sGC, soluble guanylate cyclase. Created with BioRender.

Recombinant BK channels can contain between one and four β1 subunits, with the ratio of α:β1 subunits altering voltage- and Ca²⁺-dependence (Knaus et al., 1994; Wang et al., 2002). In smooth muscle cells, BK channels are activated by localized intracellular Ca²⁺ transients, termed “Ca²⁺ sparks”, which occur due to the opening of ryanodine-sensitive Ca²⁺ release (RyR) channels on the sarcoplasmic reticulum (Nelson et al., 1995; Jaggar et al., 2000). The presence of β1 subunits increases the sensitivity of BK channels to Ca²⁺ sparks (Brenner et al., 2000). PKA and PKG activation stimulated rapid surface trafficking of β1 subunits, which increased BK channel apparent Ca²⁺-sensitivity and activation by Ca²⁺ sparks in arterial smooth muscle cells (Leo et al., 2014). PKG-mediated vasodilation was attenuated by inhibition of β1 subunit trafficking and was dependent upon BK channel activation (Leo et al., 2014). These studies demonstrated that NO acting via PKG stimulated Rab11A-positive recycling endosomes to rapidly deliver β1 subunits to the plasma membrane (Fig. 1). These de novo β1 subunits associated with surface BK channels, leading to an increase in apparent Ca²⁺-sensitivity, enhanced coupling to Ca²⁺ sparks, membrane hyperpolarization and vasodilation (Fig. 1).

BKα subunits

In contrast to what was observed with β1 subunits, more than 95% of BKα subunit protein was located in the plasma membrane of smooth muscle cells from human and rat resistance-size arteries (Leo et al., 2014; Leo et al., 2015; Leo et al., 2017; Zhai et al., 2017). BKα was not located in close proximity to Rab11A and knockdown of Rab11A or Rab11B did not alter the surface amount of BKα, suggesting that recycling endosomes do not traffic these subunits (Leo et al., 2014; Leo et al., 2015). Vasodilators and trafficking inhibitors that rapidly regulated surface β1 protein did not alter the amount of surface BKα over the same time course in arterial smooth muscle cells (Leo et al., 2014). These findings provided evidence that BKα and β1 subunits are trafficked by different pathways (Fig. 1). Early endosomes are characterized by expression of the RabGTPase Rab4 (Hutagalung & Novick, 2011). BKα colocalized with Rab4 in cerebral artery smooth muscle cells and Rab4A knockdown decreased plasma membrane BKα protein in cerebral arteries, suggesting a mechanistic link (Leo et al., 2015). In contrast, Rab4B knockdown did not alter surface BKα, indicating subtype-specific regulation by Rab4 (Leo et al., 2015). Contrary to expectations, Rab4A knockdown did not result in the intracellular accumulation of BKα, indicating that channels which were not surface-trafficked by early endosomes were instead targeted for degradation (Leo et al., 2015). Consistent with this observation, angiotensin II (Ang II), a vasoconstrictor, stimulated protein kinase C (PKC)-dependent internalization and degradation of BKα protein in arterial smooth muscle cells (Leo et al., 2015) (Fig. 1). The reduction in surface BKα by Ang II occurred over a longer time course (hours) than had previously been described for rapid trafficking of β1 subunits by Rab11A-positive recycling endosomes (Leo et al., 2014; Leo et al., 2015). Ang II-induced BKα internalization and degradation reduced BK currents in arterial smooth muscle cells, elevated myogenic tone and reduced vasoregulation by BK channel activators and inhibitors (Leo et al., 2015) (Fig. 1). Ang II type-1 receptors (AT₁R) inhibit BK channels via a G protein-independent physical coupling mechanism in rat renal artery smooth muscle cells (Zhang et al., 2014). Conceivably, BK channel inhibition by Ang II may be mediated in the short term by a G-protein independent mechanism and in the long term by G-protein dependent internalization and degradation. Collectively these studies demonstrated that the surface abundance of BKα and β1 subunits is regulated by distinct vasoactive stimuli, Rab proteins and endosomes, and over different time scales in arterial smooth muscle cells.

BKα contains a 58 amino acid insert between RCK1 and RCK2 of the C-terminus termed the “STRess axis regulated EXon” (STREX) (Xie & McCobb, 1998). In heterologous expression systems, BKα containing the STREX variant exhibited higher cell surface expression than channels of the ZERO variant (Nourian et al., 2014). Either a very small proportion of BK channels or no BK channels contained the STREX insert in rat cerebral, cremaster and pulmonary artery smooth muscle cells, (Jaggar et al., 2005; Zhao et al., 2007; Nourian et al., 2014; Detweiler et al., 2016). Thus, STREX does not appear to modify the trafficking of BKα by Rab4A-dependent early endosomes in arterial smooth muscle cells.

Regulation by membrane potential

Physiological intravascular pressures stimulate arterial depolarization to membrane potentials between ~−60 and −35 mV (Brayden & Nelson, 1992). This depolarization stimulates Ca_V1.2 channels, leading to an increase in [Ca²⁺]_i between ~100 and 350 nM in arterial smooth muscle cells (Brayden & Nelson, 1992). Membrane depolarization within this physiological range stimulated rapid anterograde trafficking and plasma membrane insertion of intracellular β1 subunits in human and rat cerebral artery smooth muscle cells (Leo et al., 2017) (Fig. 1). Depolarization stimulated Ca_V1.2 channels, leading to Ca²⁺ influx which activated Rho kinase 1 and 2 (ROCK1, ROCK2) (Leo et al., 2017). In contrast, PKC, phosphoinositide 3-kinase (PI3K) or Ca²⁺-calmodulin-dependent protein kinase II (CaMKII) were not involved in depolarization-induced β1 subunit trafficking (Leo et al., 2017). Knockdown of either ROCK1 or ROCK2 inhibited depolarization-induced β1 trafficking, indicating a requirement for both isoforms. Why both ROCK isoforms were required was unclear, but potential mechanisms include that they may function in a series signaling cascade or form a heterodimeric protein (Hartmann et al., 2015). ROCK may have stimulated the trafficking of β1 subunit-containing recycling endosomes either by directly phosphorylating Rab11A or upstream kinases that control Rab11A activity. The Rab11A amino acid sequence contains five potential phosphorylation sites for ROCK, at Thr³², Thr⁴³, Thr⁷⁷, Thr⁹⁸ and Ser¹⁴⁹, although it was not determined if any of these were involved in this signaling pathway. These studies demonstrated that membrane depolarization and NO mobilized the same intracellular β1 subunit pool, but through distinct mechanisms. NO acted through PKG, whereas membrane depolarization functioned through Ca_V1.2 channels and ROCK (Leo et al., 2014; Leo et al., 2017).

Membrane depolarization increases the coupling of BK channels to Ca²⁺ sparks in arterial smooth muscle cells (Jaggar et al., 2000). β1 subunits that were surface-trafficked in response to membrane depolarization associated with plasma membrane BKα channels and increased their apparent Ca²⁺-sensitivity (Leo et al., 2017) (Fig. 1). This trafficking-dependent increase in BK channel Ca²⁺-sensitivity would enhance coupling to Ca²⁺ sparks, leading to vasodilation. Rab11A knockdown inhibited the depolarization-induced increase in surface β1 subunit protein and reduced BK channel apparent Ca²⁺ sensitivity (Leo et al., 2017). Knockdown of Rab11A also reduced diameter responses to BK channel activators and inhibitors in pressurized arteries. Inhibition of depolarization-induced β1 subunit trafficking reduced, but did not abolish, the coupling of BK channels to Ca²⁺ sparks (Leo et al., 2017). Such coupling likely remained because two populations of surface β1 subunits exist: those which maintain a small surface population of β1 subunits and those which are contained within Rab11A-positive recycling endosomes and are mobile (Leo et al., 2014). In summary, membrane depolarization stimulated Ca_V1.2 channels, leading to Ca²⁺ influx which activated ROCK1 and 2. ROCK signaling phosphorylated Rab11A, which activated Rab11A-positive recycling endosomes to deliver β1 subunits to the plasma membrane. The increase in surface β1 subunits elevated BK channel apparent Ca²⁺-sensitivity and activity to produce vasodilation.

Modulation by vasoactivate receptor ligands

Many vasoregulatory stimuli act via protein kinases to regulate BK channels in arterial smooth muscle cells (Barman, 1999; Schubert et al., 1999; Betts & Kozlowski, 2000; Taguchi et al., 2000; Barman et al., 2004; McNair et al., 2004; Rainbow et al., 2009). Endothelin-1 (ET-1), a vasoconstrictor, inhibited the surface trafficking of β1 subunits in smooth muscle cells of cerebral arteries (Zhai et al., 2017) (Fig. 1). ET-1 activates G_q/11 protein, which stimulates phospholipase C to produce diacylglycerol (DAG) and inositol-1, 4, 5-triphosphate (IP₃) from phosphatidylinositol-4, 5-bisphosphate (Ivey et al., 2008). ET-1 activated PKC and inhibited Rab11A, which blocked anterograde trafficking of β1 (Zhai et al., 2017). In contrast, ET-1 did not alter surface BKα protein, consistent with previous evidence that β1 and BKα traffic through distinct mechanisms (Leo et al., 2014; Leo et al., 2015; Zhai et al., 2017). Through this mechanism, ET-1 reduced the number of β1 subunits that associated with BKα channels. PKC may have directly phosphorylated Rab11A to reduce its activity or modulated upstream proteins that inhibit Rab11A. Sequence analysis revealed five putative phosphorylation sites for PKC, with S177 the highest probability residue. ET-1 stimulated Rab11A phosphorylation and this was attenuated in arteries expressing Rab11A S177A, a Rab11A mutant that is unable to be phosphorylated by PKC at serine 177 (Zhai et al., 2017). The expression of Rab11A S177A also prevented ET-1 from reducing surface β1 protein in cerebral arteries. Rab11A S177A did not alter the ability of NO to stimulate β1 surface trafficking, indicating that NO activated Rab11A through a mechanism independent of S177. These observations also suggested that PKC and PKG independently modulate the activity of Rab11A. Rab11A S177 expression alone did not increase surface β1 subunits, suggesting that PKC phosphorylation of Rab11A is not what maintains a small amount of surface β1 subunits in unstimulated smooth muscle cells (Zhai et al., 2017; Leo et al., 2018). Rather, an anterograde trafficking stimulus, such as that evoked by NO or membrane depolarization, was necessary to mobilize recycling endosomes to deliver β1 subunits to the surface.

Protein kinases can directly and indirectly regulate BK channel activity through a variety of signaling mechanisms in arterial smooth muscle cells. PKA and PKG activation directly increased BK channel activity, whereas PKC reduced activity, in excised membrane patches of arterial smooth muscle cells (Robertson et al., 1993; Bonev et al., 1997; White et al., 2000; Schubert & Nelson, 2001; Zhou et al., 2010). PKA, PKG, and PKC indirectly modulated BK channel activity by regulating the frequency of Ca²⁺ sparks (Jaggar et al., 2000). NO stimulated the surface-trafficking of β1 subunits which increased BK channel open probability in arterial smooth muscle cells and this effect was retained even after patch excision (Leo et al., 2014). Similarly, PKC decreased the amount of surface β1 subunits and the reduction in BK channel open probability was maintained in excised patches. ET-1 reduced BK channel activation by Ca²⁺ sparks and the subsequent vasoconstriction through Rab11A S177 phosphorylation (Zhai et al., 2017). Collectively, these studies indicated that ET-1 activates PKC which phosphorylates Rab11A S177, leading to a decrease in Rab11A activity and inhibition of β1 subunit surface trafficking. The ET-1-induced reduction in β1 subunit surface delivery inhibited BK channels and transient BK currents, leading to vasoconstriction. In contrast, NO activated Rab11A, which surface-trafficked β1 subunits through a Rab11A S177-independent mechanism. The differential mechanisms by which NO and ET-1 regulate Rab11A activity control BK channel activity and arterial contractility.

Pathological impact of hypertension

Arteries from hypertensive animals are depolarized and constricted, and endothelium-dependent vasodilation is attenuated in these vessels (Harder et al., 1983; Harder et al., 1985) (Luscher et al., 1990; Cordellini, 1999; Heitzer et al., 1999; Zhou et al., 2001; Yang et al., 2004; Quaschning et al., 2006; Jimenez et al., 2007; Choi et al., 2011). The total and surface amounts of both BKα and β1 proteins were similar in unstimulated cerebral arteries of normotensive Wistar-Kyoto (WKY) rats and hypertensive stroke-prone spontaneously hypertensive (SP-SHR) rats (Leo et al., 2018). BKα and β1 total proteins were also similar in cerebral arteries of a hypertensive model of Fawn Hooded rats, when compared with their normotensive genetic controls (Pabbidi et al., 2014). Another study found that although BKα mRNA was similar, BKα protein was higher in cerebral arteries of spontaneously hypertensive (SHR) rats when compared to WKY rats. (Liu et al., 1998). These slightly different findings may relate to the different approaches or animal models used.

NO and membrane depolarization stimulated surface-trafficking of β1 subunits in cerebral arteries of WKY rats, but failed to mobilize β1 subunits in cerebral arteries of SP-SHR rats (Leo et al., 2018) (Fig. 1). Total Rab11A protein was lower in SP-SHR than WKY rat arteries, but the overexpression of wild-type Rab11A did not reestablish β1 subunit trafficking in SP-SHR (Leo et al., 2018). These data were consistent with the pathophysiology being due to Rab11A inhibition rather than loss of Rab11A protein (Leo et al., 2018). In line with these findings, NO activated Rab11A in arteries of WKY rats but did not activate Rab11A in arteries of SP-SHR rats (Leo et al., 2018). Given previous evidence that PKC inhibits Rab11A, the authors measured PKC activity in arteries of these rat models (Zhai et al., 2017). The total amounts and activities of two PKC isoforms, PKCα and PKCβII, were both higher in arteries of SP-SHR than WKY rats (Leo et al., 2018). Pharmacological inhibition of PKC or the overexpression of Rab11A S177A restored β1 subunit trafficking by NO in SP-SHR rat arteries (Leo et al., 2018). Impaired β1 subunit trafficking prevented BK channel activation by NO in arterial smooth muscle cells of SP-SHR rats and this dysfunction was restored by the expression of PKC phosphorylation-deficient Rab11A S177A (Leo et al., 2018). PKC inhibition reestablished vasodilation to NO by reinstating the ability of NO to surface-traffic β1 subunits and increase their association with BKα channels (Leo et al., 2018). This study demonstrated that spontaneously active PKC phosphorylates Rab11A at serine 177, which inhibits surface trafficking of β1 subunits in arterial smooth muscle cells of SP-SHR rats (Leo et al., 2018). The mechanisms which upregulated PKC signaling in arterial smooth muscle cells of SP-SHRs were not determined, but several possibilities exist, including an increase in the expression, translocation and activity of PKC (Salamanca & Khalil, 2005).

Voltage-dependent potassium (K⁺, K_V) channels

K_V channel activation produces membrane hyperpolarization, which reduces the activity of Ca_V1.2 channels in arterial smooth muscle cells, leading to vasodilation (Nelson & Quayle, 1995; Korovkina & England, 2002).The K_V channel family contains 40 different members which are classified into 12 subtypes (K_V1–12) (Jiang et al., 2003; Gutman et al., 2005). Each K_V channel α subunit contains six transmembrane domains, an S4 segment which is the main voltage sensor, and a P-loop between S5 and S6 which forms the channel pore (Gutman et al., 2005). Depolarization stimulates the S4 segment to translocate to the extracellular surface of the membrane, resulting in an increase in channel P_O (Jiang et al., 2003; Ghosh et al., 2006). Several K_V family members are expressed in vascular smooth muscle cells, including K_V1, K_V2, K_V3, K_V4, K_V6, K_V7, K_V9 and K_V11 (Lu et al., 2002; Cox, 2005; Miguel-Velado et al., 2005; Yeung et al., 2007; Mackie et al., 2008; Moreno-Dominguez et al., 2009; Vandenberg et al., 2012; Kidd et al., 2015; Barrese et al., 2017). It is generally considered that subtypes of K_V1, K_V2 and K_V7 channels are the most functionally relevant in arterial smooth muscle cells. K_V channel subunits can assemble as homotetramers or heterotetramers, which creates a wide array of current phenotypes that differ with respect to their voltage-sensitivity, temporal kinetics, amplitude and pharmacology. Currents produced by both homomeric and heteromeric K_V channels have been described in arterial smooth muscle cells (Kerr et al., 2001; Plane et al., 2005; Zhong et al., 2010).

Kv1 channels

Approximately 48% of total K_V1.5 protein was located at the plasma membrane in mesenteric arteries (Kidd et al., 2015). Smooth muscle cells are the major cell type in these arteries and K_V channels do not appear to be expressed in endothelial cells. As such, this observation suggested that a significant pool of K_V1.5 channels was stored intracellularly and may be available for anterograde trafficking in arterial smooth muscle cells. It was found that K_V1.5 channels constantly recycle to and from the plasma membrane in arterial smooth muscle cells (Kidd et al., 2015) (Fig. 2). Importantly, intravascular pressure and membrane potential regulated this recycling process to modulate the surface abundance of K_V1.5 channel protein (Kidd et al., 2015). The time course by which these stimuli regulated surface K_V1.5 channels was between one and three hours (Kidd et al., 2015). At low pressure, internalized K_V1.5 channels were targeted for lysosomal and proteosomal degradation, which reduced surface channel abundance (Kidd et al., 2015). Physiological levels of intravascular pressure and membrane potential activated Ca_V1.2 channels, leading to Ca²⁺ influx which inhibited lysosomal and proteasomal degradation of internalized K_V1.5, thereby increasing surface channels (Kidd et al., 2015). Patch-clamp electrophysiology and myography experiments supported these observations and demonstrated that intravascular pressure controls K_V1.5 surface abundance to regulate K_V currents and arterial contractility (Kidd et al., 2015). Data suggested that Ca²⁺-sensing proteins with affinities for Ca²⁺ in the nanomolar range inhibit the degradation of K_v1.5 channels. To better understand mechanisms that may be involved, an extrapolation to experiments performed in other cell types is reasonable. Heat shock protein 70 (Hsp70), a chaperone that targets misfolded proteins for ubiquitylation and degradation, heat shock cognate protein 70 (Hsc70), and carboxyl-terminus heat shock cognate 70-interacting protein (CHIP), an E3 ubiquitin ligase, all regulated K_V1.5 degradation in mammalian cell lines (Hirota et al., 2008; Li et al., 2015). In HEK293 cells, H9c2 rat cardiac myoblasts, and HL-1 mouse atrial myocytes cells expressing K_V1.5, constitutively internalized K_V1.5 channels recycled to the plasma membrane through a Rab4- and Rab11-dependent pathway (McEwen et al., 2007; Zadeh et al., 2008). K_Vβ2 subunits may also be important for K_V1.5 channel trafficking in coronary artery smooth muscle cells, as these channels are less abundant in the plasma membrane of global K_Vβ2 knockout mice (Nystoriak et al., 2017). Future studies should investigate the possibilities that these or other signaling pathways regulate surface K_V1.5 channels in arterial smooth muscle cells. In summary, intravascular pressure and membrane potential regulate K_V1.5 currents by controlling both the number and the open probability of surface-localized channels in arterial smooth muscle cells. These mechanisms integrate to alter arterial contractility.

Figure 2.

The regulation of K_V1.5 channel trafficking in arterial smooth muscle cells. Ang II, angiotensin II; [Ca²⁺]_i, intracellular calcium concentration; PKC, protein kinase C; SR, sarcoplasmic reticulum. Created with BioRender.

Ang II binds to AT₁ receptors, which stimulate G_q/11 protein and phospholipase C in vascular smooth muscle cells, leading to contraction (Berk & Corson, 1997). Ang II activated PKC, which stimulated the degradation of internalized K_V1.5 channels in mesenteric artery smooth muscle cells (Kidd et al., 2017) (Fig. 2). This decreased K_V1.5 surface protein, which reduced K_V1.5 current density and K_V1.5 function in pressurized arteries (Kidd et al., 2017). Thus, membrane potential and Ang II regulate K_V1.5 channel surface abundance through distinct mechanisms: membrane depolarization acts via Ca_V1.2 channels and Ca²⁺ influx, and Ang II through PKC activation. Vascular smooth muscle cells express three different isoforms of PKC: PKCα and PKCβ, which are DAG- and Ca²⁺-dependent, and PKCε, which is stimulated by DAG (Newton, 1995). Ang II activates PKCε to reduce K_V currents in arterial smooth muscle cells (Rainbow et al., 2009). Similarly, Ang II-induced vasoconstriction was attenuated by PKCε inhibition in mesenteric artery rings (Rainbow et al., 2009). Thus, Ang II may stimulate Ca²⁺-independent PKCε to degrade K_V1.5 channels and induce vasoconstriction.

Serotonin stimulates the endocytosis of K_V1.5 channels in pulmonary artery smooth muscle cells (Cogolludo et al., 2006). Caveolin-1, 5-HT_2A receptors and K_V1.5 channels colocalized, suggesting that channel internalization may be regulated by a compartmentalized signaling pathway (Cogolludo et al., 2006). K_V1.2 channels contain a high affinity binding site for the scaffolding protein postsynaptic density-95 (PSD95), whereas K_V1.5 has a low affinity binding site (Joseph et al., 2011). Knockdown of PSD95 reduced K_V1 expression and K_V1 currents in cerebral artery smooth muscle cells, which attenuated vasoregulation by these channels (Joseph et al., 2011). The authors proposed that PSD95 may function to increase the number of functional K_V1 channels at the plasma membrane in arterial smooth muscle cells.

Larger K_V currents were recorded in pial arteriole smooth muscle cells of (Tg)Notch3^R169C mice, which is a genetic model of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) (Dabertrand et al., 2015). Pial arterioles from (Tg)Notch3^R169C mice also constricted more to K_V channel blockers than did controls (Dabertrand et al., 2015). Data suggested that an increase in the number of surface K_V1 channels in smooth muscle cells attenuated pressure-induced constriction in CADASIL mice (Dabertrand et al., 2015).

H₂O₂, an endothelium-derived relaxing factor, activated K_V1.5 channels to produce vasodilation in adipose arterioles from healthy human subjects (Rogers et al., 2006; Park et al., 2015; Nishijima et al., 2017). In contrast, H₂O₂-induced vasodilation occurred primarily through the activation of large-conductance BK channels in adipose arterioles from patients with coronary artery disease (CAD) (Miura et al., 2003; Phillips et al., 2007; Liu et al., 2011; Zhang et al., 2012). Less surface K_V1.5 protein was observed in smooth muscle cells of CAD arteries than non-CAD subjects when performing immunofluorescence (Nishijima et al., 2017). K_V currents were also smaller in smooth muscle cells of CAD patients (Nishijima et al., 2017). Consistent with these findings, vasodilation to H₂O₂ through K_V1.5 channels was smaller in CAD arteries (Nishijima et al., 2017). This study proposed that a reduction in K_V1.5 channel surface expression in arterial smooth muscle cells attenuates vasodilation to H₂O₂ in CAD patients.

An increase in extracellular oxyhemoglobin stimulated K_V1.5 channel endocytosis, which reduced K_V currents in cerebral artery smooth muscle cells following subarachnoid hemorrhage (Ishiguro et al., 2006). The oxyhemoglobin-induced suppression of K_V currents in smooth muscle cells produced vasoconstriction that was attenuated by inhibitors of tyrosine kinase (Ishiguro et al., 2006). This study demonstrated that oxyhemoglobin activated tyrosine kinases, which stimulated K_V1.5 channel endocytosis in arterial smooth muscle cells, leading to vasoconstriction (Ishiguro et al., 2006).

K_V2 channels

K_V2 channels contributed to K_V currents in arterial smooth muscle cells, with message for both K_V2.1 and K_V2.2 detected (Cox, 2005; Amberg & Santana, 2006; Kidd et al., 2015; Cox & Fromme, 2016a, b). Less than 1% of K_V2.1 channels were conductive in arterial smooth muscle cells (O’Dwyer et al., 2020). In male arterial myocytes, K_V2.1 channels regulated membrane potential, whereas in females K_V2.1 expression was higher and channels controlled the clustering of Ca_V1.2 channels (O’Dwyer et al., 2020). This structural function of K_V2.1 channels increased Ca_V1.2 channel-dependent Ca²⁺ influx (O’Dwyer et al., 2020). Thus, K_V2.1 channels performed both electrical and structural functions in arterial smooth muscle cells that differed between males and females.

Approximately 80% of K_V2.1 was plasma membrane-localized in rat mesenteric arteries (Kidd et al., 2015). In contrast to observations with K_V1.5 channels, intraluminal pressure, membrane potential or Ang II did not alter the surface abundance of K_V2.1 channels in mesenteric arteries (Kidd et al., 2015; Kidd et al., 2017). These observations illustrated that distinct processes regulate the surface abundance of K_V1.5 and K_V2.1 channels in arterial smooth muscle cells. Dissimilar amino acid sequences of these K_V channels, or their regulation by different auxiliary subunits or signal transduction pathways may explain their isoform-selective mechanisms of surface trafficking.

Hypertension caused by systemic administration of Ang II stimulated Ca_V1.2 channels, leading to the activation of calcineurin (CaN), a Ca²⁺/calmodulin-dependent protein phosphatase in cerebral artery smooth muscle cells (Amberg et al., 2004). Activated CaN stimulated nuclear accumulation of NFATc3, a transcription factor, which decreased K_V2.1 channel expression and reduced currents (Amberg et al., 2004). K_V2.1 transcript and protein were also lower in cerebral and mesenteric artery smooth muscle cells of mice fed a high-fat diet (Nieves-Cintron et al., 2015). In this model of type-2 diabetes, AKAP150 anchored CaN to dephosphorylate and activate the transcription factor NFATc3 (Nieves-Cintron et al., 2015). The nuclear translocation of activated NFATc3 reduced K_V2.1 channel transcription, reducing its surface expression in smooth muscle cells. K_V2 currents were smaller and K_V2 function was lost in pressurized arteries of diabetic mice, which increased myogenic tone (Nieves-Cintron et al., 2015). Taken together, these studies suggest that short-term exposure to Ang II does not alter K_V2.1 channel trafficking, but a long-term elevation in Ang II inhibits K_V2.1 transcription which reduces surface channel number (Nieves-Cintron et al., 2015; Kidd et al., 2017).

K_V7 channels

Ang II reduced K_V7.4 channel protein by decreasing its interaction with HSP90, a molecular chaperone in mesenteric artery smooth muscle cells (Barrese et al., 2018). This resulted in K_V7.4 ubiquitination and proteasomal degradation (Barrese et al., 2018). Through this mechanism Ang II reduced K_V7 currents and reduced vasorelaxation to ML213, a K_V7 agonist (Barrese et al., 2018).

The microtubule depolymerizing agents colchicine, nocodazole and ciliobrevin D, a dynein inhibitor, augmented relaxation to K_V7 channel activators in rat mesenteric and renal arteries (Lindman et al., 2018; van der Horst et al., 2021). Colchicine and nocodazole also increased vasodilation to isoprenaline and this was reversed by XE991, a K_V7 channel inhibitor (Lindman et al., 2018). Immunofluorescence suggested that dynein and K_V7.4 channels colocalized in arterial smooth muscle cells and colchicine and ciliobrevin increased K_V7.4 channel surface protein (Lindman et al., 2018; van der Horst et al., 2021). Cholesterol depletion with methyl-β-cyclodextrin reduced vasorelaxation to K_V7 channel activators (van der Horst et al., 2021). The authors proposed that dynein traffics K_V7.4 channels via a mechanism that was caveolae-dependent in arterial smooth muscle cells. The time course of changes in surface K_V7 channels was not determined. It was also not clear whether these effects were specific to K_V7.4 channels or a general mechanism that also regulates the surface abundance of other K_V channel subtypes and other plasma membrane proteins.

L-type voltage-gated Ca²⁺ channels

Voltage-gated Ca²⁺ channels are complexes of multiple subunits, including the pore-forming α_1C (Ca_V1.2) and auxiliary β and α₂δ subunits. Ca_V1.2 is encoded by the CACNA1C gene, which can undergo alternative splicing to produce several variants with unique biophysical properties (Soldatov, 1994; Tang et al., 2004; Cheng et al., 2007; Cheng et al., 2009). Rat arterial smooth muscle cells primarily express Ca_V1.2 containing a novel N-terminus encoded by exon 1c (Ca_V1.2e1c), with a minority of protein encoded by the exon 1b variant (Cheng et al., 2007). Ca_V1.2 channels expressed in arterial smooth muscle cells are sensitive to several inhibitors, including dihydropyridines, benzothiazepines and phenylalkyamines (Harder, 1984; Hofmann et al., 1994; Knot & Nelson, 1998).

Ca_V1.2 channels

Ca_V1.2 subunits are primarily plasma membrane-localized (>95 %) in smooth muscle cells of cerebral arteries (Bannister et al., 2009; Bannister et al., 2011; Bannister et al., 2012; Bannister et al., 2013). Rab25, a small GTPase associated with apical recycling endosomes, regulates the surface abundance of Ca_V1.2 channels in cerebral artery smooth muscle cells (Bannister et al., 2016) (Fig. 3). Rab25 knockdown using RNA interference reduced both plasma membrane and total Ca_V1.2 protein (Bannister et al., 2016). In the absence of Rab25-mediated anterograde trafficking, Ca_V1.2 protein was targeted for degradation through both proteasomal and lysosomal pathways (Bannister et al., 2016). The reduction in surface Ca_V1.2 protein decreased Ca_V1.2 current density, which decreased both depolarization and pressure-induced vasoconstriction (Bannister et al., 2016). These findings indicated that Rab25 was essential for surface trafficking of Ca_V1.2 channels and suggested that Rab25 may be a viable vasodilatory target.

Figure 3.

Ca_V1.2 channel surface expression is regulated by α₂δ−1 and impacted during hypertension in arterial smooth muscle cells. [Ca²⁺]_i, intracellular calcium concentration; SR, sarcoplasmic reticulum. Created with BioRender.

Regulation by α₂δ−1 subunits

The CACNA2D1 gene can undergo alternative splicing, which yields four different α₂δ subunits (α₂δ−1–4) (Cole et al., 2005; Davies et al., 2007). Post-translational cleavage of α₂δ protein is followed by re-association of a glycosylated extracellular α₂ with a membrane-spanning δ subunit to form a functional protein (Cole et al., 2005; Andrade et al., 2007; Davies et al., 2007). Arterial smooth muscle cells only express α₂δ−1, with more than 95% of total protein located at the surface (Bannister et al., 2009). α₂δ−1 trafficked Ca_V1.2 channels to the plasma membrane in cerebral artery smooth muscle cells (Bannister et al., 2009). This function was Ca_V1.2 splice variant-dependent and more effective for Ca_V1.2e1c than Ca_V1.2e1b (Bannister et al., 2011). Knockdown of α₂δ−1 subunits reduced surface Ca_V1.2 channels and inhibited both pressure-and depolarization-induced vasoconstriction (Bannister et al., 2009) (Fig. 3). Pregabalin, an α₂δ−1/2 ligand that is used to treat neuropathic pain, reduced plasma membrane α₂δ−1 and Ca_v1.2 subunits in arterial smooth muscle cells (Bannister et al., 2009) (Fig. 3). Pregabalin caused vasodilation through two mechanisms: primarily by inhibiting the anterograde trafficking of Ca_V1.2 protein, but also by acting as a weak pore blocker of surface-resident Ca_V1.2 channels (Bannister et al., 2009).

Ca_V1.2 currents are larger in arterial smooth muscle cells of several different hypertension models and during high fat diet (Wilde et al., 1994; Lozinskaya & Cox, 1997; Simard et al., 1998; Wilde et al., 2000; Hirenallur et al., 2008). In genetic hypertension, the transcription and translation of both Ca_V1.2 and α₂δ−1 were both upregulated in cerebral artery smooth muscle cells (Bannister et al., 2012) (Fig. 3). Other α₂δ subunit isoforms did not emerge during hypertension (Bannister et al., 2012). Plasma membrane-localized α₂δ−1 protein increased more than did Ca_V1.2α protein, which caused a ratiometric shift in Ca_V1.2 channel composition (Bannister et al., 2012). This increased Ca_V1.2 current density and produced a larger non-inactivating current, which elevated Ca²⁺ influx at resting membrane potentials (Bannister et al., 2012). Pregabalin reduced surface Ca_V1.2 and α₂δ−1 subunits more in arteries of hypertensive rats than in normotensive rats (Bannister et al., 2012). This unequal reduction in Ca_V1.2 and α₂δ−1 subunits normalized surface amounts of these proteins in arteries of hypertensive rats (Bannister et al., 2012). Pregabalin also normalized the amplitude of the non-inactivating Ca_V1.2 current, suggesting that the increase in α₂δ−1 surface protein was a major factor responsible for current modification during hypertension (Bannister et al., 2012). Pregabalin was a more effective vasodilator of cerebral arteries from hypertensive rats than normotensive rats and attenuated the pathological vasoconstriction (Bannister et al., 2012). These studies suggested that smooth muscle cell α₂δ−1 could represent a novel target for antihypertensive therapy. Clinical pregabalin does not appear to modify systemic blood pressure in normotensive humans at doses used to treat neuropathic pain, fibromyalgia, and epileptic seizures (Gajraj, 2007). It is possible that clinical doses of pregabalin that are used to treat these diseases may be insufficient to induce vasodilation in vivo, at least in normotensive subjects. Gabapentin, a lower affinity pregabalin analog, enters cells through system-L neutral amino acid transporters (Hendrich et al., 2008). Arterial myocytes may not uptake pregabalin as effectively as neurons. Other mechanisms that regulate blood pressure, including baroreceptors or the renin-angiotensin system, may also compensate for pregabalin-induced systemic vasodilation, leading to no net change in blood pressure. Similarly, pregabalin may be a more effective vasodilator in hypertensive than normotensive subjects. Regardless of the mechanisms involved, smooth muscle cell α₂δ−1 may represent a novel target for antihypertensive therapy.

Modulation by β3 subunits

Four different genes encode Ca_V β subunits (β1–β4), with each subject to splice variation (Buraei & Yang, 2013). β subunits can regulate several properties of Ca_V1.2 channels, including surface expression, voltage-sensitivity and interaction with other proteins (Buraei & Yang, 2013). The binding of β3 subunits to Ca_V1.2 channels weakens the endoplasmic retention signal, leading to an increase in surface trafficking (Bichet et al., 2000). In aortic smooth muscle cells of global β3^−/− mice, total and surface Ca_V1.2α protein was lower, Ca_V1.2 current density was smaller and Ca_V1.2 current inactivation occurred more slowly (Murakami et al., 2003). Ang II infusion increased Ca_V1.2 protein in mesenteric arteries of wild-type mice and elevated blood pressure (Kharade et al., 2013). These effects were attenuated in global β3^−/− mice, suggesting that β3 subunits contributed to this response and that targeting β3 subunits can attenuate these pathological modifications (Kharade et al., 2013).

Collectively, these studies have identified unique mechanisms by which the trafficking and surface expression of Ca_V1.2 channels are regulated in arterial smooth muscle cells and unearthed novel proteins that could be targeted to modulate contractility.

Transient Receptor Potential (TRP) channels

The TRP family of non-selective cation channels contains 28 distinct proteins in mammals (Earley & Brayden, 2015). TRP channels are divided into six subfamilies, designated as canonical (TRPC), vanilloid (TRPV), melastatin (TRPM), ankyrin (TRPA), polycystin (TRPP), and mucolipin (TRPML). All TRP channels contain six transmembrane domains with both the C- and N-termini located intracellularly. Arterial smooth muscle cells express TRP channels from several different families (Earley & Brayden, 2015).

TRPC3 channels

Only ~30% of TRPC3 channel total protein was present in the plasma membrane of rat cerebral artery smooth muscle cells, indicating a significant intracellular pool of this protein (Adebiyi et al., 2010). In contrast, virtually all TRPC3 protein was surface-resident in rat mesenteric artery smooth muscle cells (Adebiyi et al., 2012). These data suggested that the surface to intracellular distribution of TRPC3 channels may differ depending on the anatomical origin of the vasculature. The surface abundance of TRPC3 channels was the same in mesenteric arteries of normotensive WKY rats and young prehypertensive SHR rats (Adebiyi et al., 2012). In contrast, total and surface TRPC3 proteins were both more than 2-fold higher in mesenteric artery smooth muscle cells of hypertensive SHR rats than in age-matched normotensive WKY rats (Adebiyi et al., 2012). It was concluded that an increase in TRPC3 channel expression occurred during hypertension and this de novo TRPC3 protein was trafficked to the plasma membrane (Adebiyi et al., 2012). The increase in surface TRPC3 protein led to a greater ET-1 and IP₃-induced vasoconstriction in hypertensive rats (Adebiyi et al., 2012).

TRPM4 channels

Intravascular pressure and PKC stimulate TRPM4 channels in arterial smooth muscle cells, leading to sodium (Na⁺) influx, membrane depolarization and vasoconstriction (Crnich et al., 2010). Biotinylation indicated that ~64% of TRPM4 protein located to the plasma membrane of cerebral arteries (Crnich et al., 2010). Intravascular pressure activated PKCδ, which stimulated surface-trafficking of TRPM4 channels in smooth muscle cells, leading to depolarization and constriction (Crnich et al., 2010) (Fig. 4).

Figure 4.

TRP channel trafficking in arterial smooth muscle cells. [Ca²⁺]_i, intracellular calcium concentration; PKCδ, protein kinase C δ; SR, sarcoplasmic reticulum, SUMO, small ubiquitin-like modifier 1. Created with BioRender.

PKD2 (TRPP1, polycystin 2) channels

PKD2 channels are expressed in arterial smooth muscle cells (Griffin et al., 1997; Sharif-Naeini et al., 2009; Narayanan et al., 2013). PKD2 channel activation generated non-selective cation currents in cerebral, mesenteric and hindlimb artery smooth muscle cells that induced membrane depolarization and vasoconstriction (Narayanan et al., 2013; Bulley et al., 2018; Hasan et al., 2019). PKD2 channels were primarily plasma membrane-localized (~85%) in cerebral and hindlimb arteries (Narayanan et al., 2013; Bulley et al., 2018). Cell swelling activated PKD2 currents in arterial smooth muscle cells but did not alter their surface-to-intracellular distribution, indicating that this mechanical stimulus activated channels located in the plasma membrane (Narayanan et al., 2013). PKD2 channels undergo post-translational modification by SUMO1 (small ubiquitin-like modifier 1) protein in arterial smooth muscle cells of resistance-size hindlimb arteries (Hasan et al., 2019) (Fig. 4). In arteries at physiological intravascular pressure, PKD2 existed in equal proportions as either non-sumoylated (PKD2) or triple SUMO1-modifed (SUMO-PKD2) proteins. SUMO can attach to other proteins through either a canonical consensus motif (Ψ-K-x-D/E), a SUMO interaction motif, or a non-consensus domain (Hendriks & Vertegaal, 2016). PKD2 contains two SUMO consensus motifs at K686 and K864, three SIM domains located between amino acids 229–233, 490–494 and 511–514, and a non-consensus motif at K759. Sumoylation of PKD2 altered its surface abundance and physiological functions in arterial smooth muscle cells. SUMO-PKD2 constantly recycled between the plasma membrane and an intracellular compartment, whereas PKD2 was surface-resident (Hasan et al., 2019) (Fig. 4). Voltage-dependent Ca²⁺ influx stimulated the return of internalized SUMO-PKD2 channels to the plasma membrane (Hasan et al., 2019). Mechanisms that reduced [Ca²⁺]_i, such as a decrease in intravascular pressure, membrane hyperpolarization or Ca_V1.2 channel blockers shuttled internalized SUMO-PKD2 into a lysosomal degradation pathway. This prevented the return of SUMO-PKD2 to the plasma membrane and decreased the number of surface channels. Through this signaling mechanism, intravascular pressure regulated the surface abundance of SUMO-PKD2 channels to control the amplitude of Na⁺ currents (I_Na) in smooth muscle cells. It was also found that intravascular pressure activated SUMO-PKD2, not PKD2, channels, as desumoylation lead to the loss of I_Na activation in arterial smooth muscle cells and vasodilation (Hasan et al., 2019) (Fig. 4).

Total and surface PKD2 protein was more abundant in mesenteric and hindlimb arteries of Ang II-induced hypertensive mice than in normotensive controls (Bulley et al., 2018). In contrast, the surface-to-intracellular distribution of PKD2 channels was unchanged (Bulley et al., 2018). These data indicated that PKD2 protein increased during hypertension, presumably due to transcriptional upregulation. The de novo PKD2 protein was then trafficked to the plasma membrane, which increased surface channel number (Fig. 4). Tamoxifen-inducible genetic knockout of PKD2 in arterial smooth muscle cells reduced both vasoconstriction and systemic blood pressure in Ang II-induced hypertensive mice (Bulley et al., 2018). This study supported the concept that a transcriptional increase in the abundance of surface PKD2 channels in arterial smooth muscle cells contributed to vasoconstriction during hypertension (Bulley et al., 2018).

TRPV4 channels

Less than half of the total amount of TRPV4 channel protein was located in the plasma membrane of mesenteric and hindlimb arteries, indicating that a significant intracellular pool exists in the vascular wall (Peixoto-Neves et al., 2023). Short term (minutes) exposure to acetylcholine (ACh) or intravascular flow did not alter the surface-to-intracellular distribution of TRPV4 channels in these arteries, nor did brefeldin, an anterograde trafficking inhibitor (Peixoto-Neves et al., 2023). It is possible that physiological stimuli can regulate the surface abundance of TRPV4 channels in endothelial cells and smooth muscle cells over a longer time course than was investigated in this study. Similarly, TRPV4 channels are expressed in both arterial smooth muscle and endothelial cells and the lack of change in arterial surface protein may be the result of differential regulation in each cell type (Watanabe et al., 2002; Earley et al., 2005; Marrelli et al., 2007).

ION CHANNEL TRAFFICKING IN ENDOTHELIAL CELLS

Small-conductance Ca²⁺-activated K⁺ (SK) and intermediate-conductance Ca²⁺-activated K⁺ (IK) channels

Three SK (SK1–3) and one IK (SK4) channel are encoded by four different KCNN genes. SK and IK channels contain six transmembrane domains, intracellular C- and N- termini and a selectivity filter located between S5 and S6 in the pore region. SK and IK channels assemble as homotetramers, although heterotetrameric channels have also been described (Tuteja et al., 2010). SK and IK channels are highly homologous, with principal variation occurring in the amino acid sequences of their C- and N- termini (Kohler et al., 1996; Ishii et al., 1997). Calmodulin associates with both SK and IK channels and acts as a Ca²⁺ sensing β subunit which endows Ca²⁺-dependent activation (Xia et al., 1998; Fanger et al., 1999). Endothelial cells express SK3 and IK channels (Burnham et al., 2002; Brahler et al., 2009). SK3 channels exhibited uniform distribution in the plasma membrane, whereas IK channels localized to myoendothelial projections in endothelial cells (Sandow et al., 2006; Ledoux et al., 2008; Ottolini et al., 2020).

Approximately 21% of total SK3 channel protein and 25% of total IK channel protein located to the plasma membrane in endothelial cells of mesenteric arteries (Peixoto-Neves et al., 2023). ACh and intravascular flow each stimulated the rapid (seconds) anterograde trafficking of an intracellular pool of SK3 channels, which increased surface SK3 protein (Peixoto-Neves et al., 2023) (Fig. 5). It was not clear whether ACh and flow mobilize the same pool of SK3 channels or different pools, although both possibilities may exist. ACh activated TRPV4 channels, leading to Ca²⁺ influx which stimulated Rab11A to traffic SK3 channels (Peixoto-Neves et al., 2023). These data suggested that intracellular SK3 channels are stored within Rab11A-positive recycling endosomes. Consistent with this Ca²⁺-dependent signaling mechanism, a high pipette Ca²⁺ concentration activated an SK current that was blocked by trafficking inhibitors in aortic endothelial cells (Lin et al., 2012). The rapidity of trafficking suggested that Rab11A-positive recycling endosomes containing SK3 may be situated nearby TRPV4 channels and be directly impacted by Ca²⁺ influx. Alternatively, Ca²⁺-dependent proteins located nearby TRPV4 channels may constitute part of a transduction pathway which activates Rab11A. The stimulation of SK3 channel anterograde trafficking increased SK3 currents in endothelial cells and produced vasodilation (Peixoto-Neves et al., 2023) (Fig. 5).

Figure 5.

SK3 channel trafficking in endothelial cells and functional significance. RE, recycling endosome. Created with BioRender.

In contrast to SK3 channels, ACh or flow did not alter IK channel surface protein over the same time course in endothelial cells (Peixoto-Neves et al., 2023). Explanations for this difference include that mechanisms of IK channel trafficking may be slower than for SK3 channels. IK channels may not be stored in Rab11A-positive recycling endosomes or may be within Rab11A-positive recycling endosomes that are distant from surface TRPV4 channels.

Mechanisms that regulate the properties of surface ion channel clusters, including the influence of de novo trafficked proteins are poorly understood. ACh-induced anterograde trafficking of SK3 channels increased the size of surface SK3 clusters which colocalized with TRPV4 clusters (Peixoto-Neves et al., 2023). In contrast, ACh did not alter the size of SK3 clusters that were not co-localized with TRPC4 clusters (Peixoto-Neves et al., 2023) (Fig. 5). Surprisingly, ACh did not alter SK3 cluster density, suggesting that actively trafficked SK3 channels were targeted to specific surface locations nearby TRPV4 channels (Peixoto-Neves et al., 2023). Trafficked SK3 channels may be targeted to existing SK3 clusters or delivered randomly to the plasma membrane where they then diffuse and join with existing SK3 clusters. Experimental and in silico data have suggested that ion channels are randomly inserted into the plasma membrane, where they subsequently form clusters (Sato et al., 2019). What caused surface-trafficked SK3 channels to form larger clusters was uncertain but may have involved preferential interaction with other SK3 subunits or another protein in endothelial cells (Peixoto-Neves et al., 2023). The mechanisms by which de novo SK3 channels were targeted to TRPV4 clusters was also unclear. Possibilities include that Rab11A-positive recycling endosomes containing SK3 channels locate in close spatial proximity to TRPV4 clusters which activate their surface trafficking. SK3 channels are then delivered to the region of plasma membrane which contains the same TRPV4 channels (Fig. 5). This targeting is beneficial because Ca²⁺ entering through TRPV4 channels would be more effective at activating nearby SK3 channels than distant SK3 channels and produce larger currents. Consistent with these results, ACh also did not alter the size or density of TRPV4 surface clusters or their proximity to SK3 clusters in endothelial cells (Peixoto-Neves et al., 2023). It is possible that stimuli other than ACh or flow may alter the surface abundance of TRPV4 channels or that such changes may occur over a longer time course than was measured in this study.

Mechanisms that regulate basal levels of SK and IK channels in the plasma membrane of endothelial cells is poorly understood. The recycling of recombinant SK3 channels was dependent upon RME-1 and Rab35 in HMEC-1, a human endothelial cell line (Gao et al., 2010). In contrast, recombinant IK channels underwent endocytosis and degradation (Gao et al., 2010). These data indicated that basal turnover of SK3 and IK channels occurs through different mechanisms.

In summary, the surface abundance of SK3 channels is dynamic, targeted, and rapidly increased by vasodilator stimuli in endothelial cells to elicit vasorelaxation. The unique trafficking mechanisms for SK and IK channels may explain their different plasma membrane localization in endothelial cells.

CONCLUSIONS AND PERSPECTIVE

It is now recognized that physiological vasoactive stimuli regulate ion channel trafficking in both arterial smooth muscle and endothelial cells. Through this mechanism, stimuli alter the surface abundance of ion channels to modulate cellular current, membrane potential and arterial contractility. Ion channels which undergo surface regulation in arterial smooth muscle cells include BKα and β1, K_V1.5, K_V 7.4, Ca_V1.2, TRPC3, TRPM4 and PKD2 (Bannister et al., 2009; Crnich et al., 2010; Adebiyi et al., 2012; Leo et al., 2014; Kidd et al., 2015; Leo et al., 2015; Hasan et al., 2019; van der Horst et al., 2021). In endothelial cells, SK3 channels are also highly mobile (Peixoto-Neves et al., 2023). Importantly, a single physiological stimulus does not impact all ion channels in these cell types but targets specific channels. For instance, NO stimulates rapid surface-trafficking of β1 subunits, but does not alter the amount of pore-forming BKα channels in arterial smooth muscle cells (Leo et al., 2014). Membrane depolarization increases surface K_V1.5 channels but does not alter surface K_V2.1 channels in arterial smooth muscle cells (Kidd et al., 2015). ACh rapidly increases the number of surface SK3 channels, but does not alter surface IK channels in endothelial cells (Peixoto-Neves et al., 2023). Individual physiological stimuli activate signal transduction pathways which promote the trafficking of specific ion channels. Ion channels are also stored within vesicles whose trafficking is controlled by specific Rab proteins. Studies have shown that trafficking mechanisms which rapidly insert de novo channels and regulatory subunits into the plasma membrane in response to a stimulus are different to those which regulate the steady-state surface population of the same proteins under resting conditions (Leo et al., 2014; Peixoto-Neves et al., 2023). Thus, the surface population of a specific ion channel may also be regulated by more than one trafficking mechanism in the same cell type.

Stimuli may activate local or global intracellular signals to control the surface trafficking of vesicles which contain specific ion channel types. Ion channels may be delivered to precise locations in the plasma membrane, which will affect their clustering and coupling to other surface proteins, including other ion channels. For example, surface-trafficked SK3 channels increase the size of SK3 clusters which locate in close spatial proximity to TRPV4 clusters in endothelial cells (Peixoto-Neves et al., 2023). What produces this location specificity is unclear, but it may occur because trafficking vesicles which deliver SK3 channels are located nearby TRPV4 channels which activate their trafficking. The same TRPV4 clusters then receive the trafficked SK3 channels through a local signaling mechanism.

Plasma membrane microdomains, such as caveolae and lipid rafts, are also likely to be important for determining the surface abundance and clustering of individual ion channels. In smooth muscle cells, peripheral coupling sites exist where the plasma membrane and sarcoplasmic reticulum membranes locate within ~20 nm nanometer apposition (Devine et al., 1972). At these sites, spatially restricted signaling can occur between plasma membrane and sarcoplasmic reticulum proteins. For example, the opening of sarcoplasmic reticulum RyR channels produces local Ca²⁺ transients termed sparks which activate nearby plasma membrane BK channels at peripheral coupling sites (Jaggar et al., 2000; Pritchard et al., 2019; Saeki et al., 2019; Krishnan et al., 2022). Ca²⁺ released from sarcoplasmic reticulum IP₃ receptors also activates plasma membrane TRPM4 channels at peripheral coupling sites (Krishnan et al., 2022). Conceivably, these ion channels may be targeted to the plasma membrane at peripheral coupling sites. Evidence supporting this concept includes that β1 subunits are essential for coupling BK channels to Ca²⁺ sparks and that Rab11A-dependent recycling endosomes deliver β1 subunits to the plasma membrane to enhance the coupling of Ca²⁺ sparks to BK channels in arterial smooth muscle cells (Brenner et al., 2000; Zhai et al., 2017). Thus, recycling endosomes may deliver β1 subunits to the plasma membrane at peripheral coupling sites to control BK channel activity and function in smooth muscle cells.

Physiological stimuli regulate both the amounts of surface ion channels and their regulatory subunits and their rates of change in smooth muscle and endothelial cells to elicit functional responses. Rapid anterograde trafficking of BK channel β1 subunits in smooth muscle cells and SK3 channels in endothelial cells quickly produces vasodilation (Leo et al., 2014; Peixoto-Neves et al., 2023). In contrast, the trafficking mechanisms which control surface K_V1.5 and PKD2 channels in arterial smooth muscle cells are slower (Kidd et al., 2015; Hasan et al., 2019). These slower trafficking-dependent mechanisms will produce long-term alterations in contractility, particularly if surface channels are internalized and targeted for degradation. Some physiological stimuli can simultaneously alter the surface abundance of several different ion channel subunits in endothelial and smooth muscle cells. For instance, ACh rapidly increases the surface amounts of both SK3 channels in endothelial cells and BK channel β1 subunits in arterial smooth muscle cells (Leo et al., 2014; Peixoto-Neves et al., 2023). The combination of these mechanisms produces a vasodilation that is larger than if trafficking of one channel type occurred in only one cell type.

Recent work which has studied ion channel trafficking in the vasculature has primarily focused on its functional significance and pathological involvement in smooth muscle cells. In contrast, far less is known about ion channel trafficking in endothelial cells. Endothelial cells express a wide variety of ion channels, including several K⁺, Na⁺, Cl⁻ and non-selective cation channels (von Beckerath et al., 1996; Earley & Brayden, 2015; MacKay et al., 2020; Mata-Daboin et al., 2023). The types of ion channels present in the plasma membrane of arterial and capillary endothelial cells also varies (Ledoux et al., 2006; Longden et al., 2017). Investigating the mechanisms which regulate ion channel trafficking in endothelial cells, its regulation of arterial contractility and alterations that occur in vascular diseases, including hypertension and stroke should be a focus of future research.

A wide variety of diseases, including hypertension, diabetes, obesity, and stroke impact vascular function, including contractility. Dysfunctional signaling mechanisms may impact ion channel trafficking in arterial smooth muscle and endothelial cells and contribute to cardiovascular diseases in humans. For instance, surface trafficking of BK channel β1 subunits was attenuated in arterial smooth muscle cells during hypertension and produced vasoconstriction (Leo et al., 2018). Amino acid mutations in ion channels and their auxiliary subunits can also cause or contribute to human diseases. Using BK channels as an example, single nucleotide polymorphisms in BKα and β1 genes contribute to human cardiovascular diseases, including hypertension (Kohler, 2010). More than 140 SNPs have been reported in or nearby the β1 subunit gene. An E65K polymorphism in β1 produces a “gain-of-function” mutant which increases BK channel apparent Ca²⁺-sensitivity and lowers diastolic hypertension (Fernandez-Fernandez et al., 2004). It is possible that mutations in β1 subunits affect trafficking and impact the physiological regulation of arterial contractility. Mutations that alter transcription and translation will also change the amount of an ion channel protein that is available for trafficking to the surface. Along these lines, the apparent Ca²⁺-sensitivity of BK channels was lower in arterial smooth muscle cells of β1 subunit knockout mice and this lead to vasoconstriction and systemic hypertension (Brenner et al., 2000). Ion channel mutations may also affect surface-localization, internalization and degradation and modify coupling to other proteins and surface abundance. The dysfunctional surface expression of an ion channel may immediately produce a disease phenotype or initially be compensated by other signaling mechanisms. Vascular disease may only occur when other pathological states affect this compensation; a scenario analogous to a two-hit mechanism. Although recent studies have begun to uncover the functional significance and pathological involvement of ion channel trafficking in the vasculature, much is unknown. Future studies should aim to investigate these many possibilities.

Biographies

Dieniffer Peixoto-Neves, Ph.D. is an Instructor in the Department of Physiology at the University of Tennessee Health Science Center in Memphis, Tennessee, USA. Dr. Peixoto-Neves obtained her Ph.D. from the State University of Ceará in Fortaleza, Brazil. Dr. Peixoto-Neves has published papers investigating physiological functions of ion channels in the vasculature. Currently, she is studying the functional significance of ion channel trafficking in endothelial cells.

Jonathan Jaggar, Ph.D. is the Maury Bronstein Endowed Professor of Physiology in the same Department as Dr. Peixoto-Neves. Dr. Jaggar obtained his Ph.D. from the University of Sheffield in the United Kingdom. His laboratory studies signaling mechanisms, trafficking, physiological functions and pathological alterations in a wide variety of different vascular ion channels. Research in his laboratory uses molecular, biochemical, cellular, imaging, and functional techniques and genetic animal models.