[ Erik J. Behringer (left) earned his Ph.D. in Pharmacology. His dissertation research focused on calcium homeostasis in sympathetic neurons in the context of cerebral blood flow regulation throughout development and ageing. He joined the laboratory of Steven S. Segal for postdoctoral training, where he has developed novel paradigms for studying electrical signalling along the endothelium of resistance arteries. He is now combining his training in calcium imaging and electrical signaling to resolve mechanisms underlying endothelial dysfunction accompanying ageing. Steven S. Segal (right) began his career in research by studying the metabolic demands of exercise in human subjects. His doctoral studies focused on changes in muscle physiology and biochemistry during regeneration following transplantation, which led to his postdoctoral training in microcirculation. The long-term research interests of the Segal laboratory centre on resolving mechanisms of how blood flow control is governed within microvascular networks, particularly in light of the contractile activity of skeletal muscle fibres. The hypothesis developed in the present review reflects the creativity fostered in our laboratory complemented by the achievements of a talented young investigator. Our collective efforts focus on understanding how cell-to-cell signalling is modulated within microvessels that control tissue blood flow. Resolving these interactions provides new insight into how exercise and ageing influence human performance.]
Abstract Blood flow control requires coordinated contraction and relaxation of smooth muscle cells (SMCs) along and among the arterioles and feed arteries that comprise vascular resistance networks. Whereas smooth muscle contraction of resistance vessels is enhanced by noradrenaline release along perivascular sympathetic nerves, the endothelium is integral to coordinating smooth muscle relaxation. Beyond producing nitric oxide in response to agonists and shear stress, endothelial cells (ECs) provide an effective conduit for conducting hyperpolarization along vessel branches and into surrounding SMCs through myoendothelial coupling. In turn, bidirectional signalling from SMCs into ECs enables the endothelium to moderate adrenergic vasoconstriction in response to sympathetic nerve activity. This review focuses on the endothelium as the cellular pathway that coordinates spreading vasodilatation. We discuss the nature and regulation of cell-to-cell coupling through gap junctions, bidirectional signalling between ECs and SMCs, and how oxidative stress during ageing may influence respective signalling pathways. Our recent findings illustrate the role of small (SKCa) and intermediate (IKCa) Ca2+ activated K+ channels as modulators of electrical conduction along the endothelium. Gaps in current understanding indicate the need to determine mechanisms that regulate intracellular Ca2+ homeostasis and ion channel activation in the resistance vasculature with advancing age.
cGMP-dependent protein kinase I
endothelial nitric oxide synthase
intermediate conductance Ca2+ activated K+ channels
voltage-activated K+ channels
reactive oxygen species
small conductance Ca2+ activated K+ channels
sarcoplasmic/endoplasmic reticulum calcium ATPase
smooth muscle cell
sympathetic nerve activity
transient receptor potential
Blood flow regulation during exercise is dominated by the energetic demands of skeletal muscle. In turn the control of muscle blood flow is dictated by the functional organization of the resistance vasculature (Folkow et al. 1971). With smooth muscle cells (SMCs) as the effector, blood flow is regulated by vasodilatation and vasoconstriction through the interaction of SMCs with endothelial cells (ECs) and sympathetic nerves. Complementary stimuli include shear stress and transmural pressure in conjunction with vasoactive stimuli in the bloodstream and released by muscle fibres (Segal, 2005). Sympathetic nerve activity (SNA) is defined by the frequency of action potentials that trigger the release of noradrenaline from perivascular nerves to activate α-adrenoreceptors on SMCs. With advancing age, an increase in SNA may enhance smooth muscle contraction in tandem with an impaired ability of the endothelium to promote smooth muscle relaxation (Gates et al. 2009; Jackson et al. 2010; Seals et al. 2011; Muller-Delp et al. 2012). This review centres on the role of the endothelium as the principal cellular pathway by which the signal for vasodilatation is initiated and spreads along the vessel wall. We consider the cellular mechanisms that underlie the initiation, cell-to-cell conduction (i.e. spread) and modulation of hyperpolarization along arterioles and feed arteries that comprise the resistance networks controlling oxygen delivery to exercising skeletal muscle. In turn, we discuss the nature and regulation of signal conduction along the endothelium, bidirectional information transfer between ECs and SMCs, and the role of ECs in providing negative feedback to SMCs during activation of α-adrenergic receptors. In light of newly identified roles of small (SKCa) and intermediate (IKCa) Ca2+ activated K+ channels as modulators of electrical conduction, we consider how oxidative stress during ageing may influence key components of respective signalling pathways and propose a mechanism by which enhanced SNA can inhibit spreading vasodilatation.
Signalling pathways underlying endothelium-dependent vasodilatation
The coordination of vasodilatation in resistance networks is mediated in large part through electrical signals (e.g. hyperpolarization) conducted along the endothelium and into surrounding SMCs via cell-to-cell coupling through gap junction channels. We focus on those events that influence membrane potential (Vm) in respective cell layers (Fig. 1) with an emphasis on the role of K+ channels as their activation produces the hyperpolarization that results in vasodilatation.
Endothelial cells of resistance vessels express SKCa and IKCa abundantly in their plasma membranes, and thus the regulation of these ion channels is integral to the initiation of hyperpolarization (Ledoux et al. 2006; Grgic et al. 2009). Acetylcholine (ACh) is used widely as an endothelium-dependent vasodilator in light of its actions being highly reproducible and well-defined. Binding of ACh to muscarinic (M3) receptors of the endothelial plasma membrane stimulates phospholipase C to liberate inositol trisphosphate (IP3), which opens IP3 receptors on the endoplasmic reticulum, thereby releasing Ca2+ from internal stores. The rise in intracellular Ca2+ concentration ([Ca2+]i) activates SKCa/IKCa (Busse et al. 2002; Garland et al. 2011), resulting in the efflux of K+ to initiate hyperpolarization. This change in Vm spreads into surrounding SMCs via heterocellular (i.e. myoendothelial) gap junctions (Emerson & Segal, 2000a; Busse et al. 2002; Garland et al. 2011). In turn, hyperpolarization of SMCs inactivates voltage-gated (e.g. L-type) Ca2+ channels, reducing Ca2+ entry to promote relaxation and vasodilatation (Nelson & Quayle, 1995) (Fig. 1).
Complementary signalling events promoting hyperpolarization and relaxation of SMCs result from the activation of endothelial nitric oxide synthase (eNOS) as a consequence of the rise in EC [Ca2+]i. As nitric oxide (NO) diffuses from the endothelium to surrounding SMCs, it stimulates soluble guanylate cyclase to generate cyclic guanosine-3,5-monophosphate (cGMP). The resulting activation of cGMP-dependent protein kinase (cGKI) exerts multiple actions to promote SMC relaxation including phosphorylation of voltage-gated K+ channels (Kv) to increase their open probability (Irvine et al. 2003) (Fig. 1). Though beyond the focus of the present discussion it should be recognized that, independent of changes in Vm, the activity of myosin light chain phosphatase leads to dephosphorylation of the regulatory light chain of myosin II to promote SMC relaxation (Cole & Welsh, 2011). Additional actions of cGKI promoting SMC relaxation include inhibition of Ca2+ entry through the plasma membrane along with the release of Ca2+ from the sarcoplasmic reticulum (Feil et al. 2003).
The role of ACh as a mediator of spreading vasodilatation in vivo remains controversial (e.g. ‘spillover’ from the activation of motor end plates). Nevertheless, complementary mechanisms for initiating hyperpolarization in response to exercise involve the release of K+ from contracting skeletal muscle fibres. Thus elevation of extracellular [K+] (e.g. from 5 to 10 mm) can activate inward rectifying K+ channels and the Na+/K+-ATPase in plasma membranes. A major hurdle that has yet to be overcome is measuring Vm of ECs or SMCs in resistance vessels of exercising muscle. This obstacle is attributable to the pronounced tissue movement associated with contraction–relaxation of skeletal muscle fibres. Indeed, the key signalling events underlying the initiation (and spread) of hyperpolarization in response to ACh have been defined using preparations designed to minimize tissue movement and thereby enable measurements of Vm with microelectrodes. Thus key evidence concerning the role of the endothelium as the cellular pathway for spreading vasodilatation is based upon interventions in which selective disruption of ECs inhibits the spread of vasodilatation from arterioles into feed arteries (i.e. ascending vasodilatation) in response to muscle contraction as well as that initiated by ACh, thereby restricting the increase in muscle blood flow (Segal & Jacobs, 2001).
Spreading and regulating the signals for vasodilatation
Once hyperpolarization is initiated, it spreads rapidly from cell to cell along the endothelium through homocellular gap junctions (Fig. 2A). As shown by immunolabelling for connexin protein subunits and electron microscopy, gap junctions are highly expressed at borders of neighbouring ECs (Looft-Wilson et al. 2004; Wolfle et al. 2007) and discretely at myoendothelial junctions between ECs and SMCs (Sandow et al. 2006; Tran et al. 2012) (Fig. 1). Being oriented longitudinally along the axes of arterioles and resistance arteries, each EC can make contact with ∼20 adjacent SMCs circumscribed around the intima (Haas & Duling, 1997). These structural and functional characteristics of the endothelium underscore its role as the predominant cellular pathway for conducting hyperpolarization and, thereby, spreading the signal for vasodilatation along resistance vessels (Emerson & Segal, 2000b; Looft-Wilson et al. 2004; Wolfle et al. 2007).
Despite the relative paucity of myoendothelial projections, these microdomains serve as robust sites for signalling between respective cell layers to coordinate vasomotor control (Heberlein et al. 2009; Kerr et al. 2012). As confirmed with dual simultaneous intracellular recording from ECs and SMCs in pressurized resistance arteries, hyperpolarization spreading along the intima results in corresponding changes in the Vm of SMCs with ensuing vasodilatation (Emerson & Segal, 2000a). In turn, myoendothelial electrical coupling underlies the ability of hyperpolarization spreading along the intima to hyperpolarize and relax consecutive SMCs of the media. Myoendothelial communication through gap junctions is bidirectional and signals originating in SMCs can spread into ECs. For example, activation of α1-adrenoreceptors triggers a rise in SMC [Ca2+]i (and/or of IP3), which can diffuse through myoendothelial gap junctions and provide negative feedback to attenuate SMC contraction. Original studies implicated activation of eNOS with NO attenuating vasoconstriction (Dora et al. 1997) whereas more recent findings point to the activation of IKCa in endothelial projections (Tran et al. 2012). Nevertheless, these studies of myoendothelial signalling have consistently relied on pharmacological activation of α1-adrenoreceptors with phenylephrine. In light of the ability of SNA to inhibit spreading vasodilatation (Haug & Segal, 2005), we develop the following relationships to propose an integrated hypothesis of intercellular signalling underlying sympathetic regulation of muscle blood flow during exercise (Figs 1 and 2).
Investigations of spreading vasodilatation have centred on the regulation and expression of gap junctions (Figueroa & Duling, 2009), which can thereby determine cell-to-cell signalling. However, recent experiments suggest an alternative mechanism for modulating electrical signalling along the vessel wall. Whereas SKCa/IKCa activation has been viewed in the context of initiating hyperpolarization, opening ion channels in the plasma membrane increases the loss of electrical charge from the cell through the plasma membrane. With robust expression of SKCa/IKCa in the endothelium, activation of these ion channels increases the loss of electrical current from each EC, thereby diminishing the amount of signal available to spread into neighbouring ECs. Thus, graded activation of SKCa/IKCa progressively impairs electrical conduction along the endothelium (Behringer & Segal, 2012). At the same time, cell-to-cell coupling through gap junctions remains intact as confirmed by the maintenance of dye transfer from the injected cell into surrounding cells. In light of these findings we advance the following hypothesis: SNA impairs spreading vasodilatation by enhancing current leak along the endothelium. This effect can be explained by SKCa/IKCa activation along the endothelium in response to Ca2+ influx through myoendothelial gap junctions during enhanced sympathetic activation of α1-adrenoreceptors on SMCs (Fig. 2B).
Endothelial dysfunction with ageing
The impairment of vasodilatation with advancing age has been attributed to ‘endothelial dysfunction’, a disorder characterized by decreased NO production and/or availability with attenuated SMC relaxation in response to ACh (Taddei et al. 1995; Gates et al. 2009; Seals et al. 2011). Endothelial dysfunction is a marker of advanced cardiovascular disease (Schachinger et al. 2000; Bugiardini et al. 2004) and associated with increased oxidative stress (Heitzer et al. 2001). Older humans (>60 years) exhibit impaired endothelium-dependent vasodilatation during exercise (Proctor & Parker, 2006; Kirby et al. 2009), to the heating of skin (Kenney et al. 1997) and in response to ACh (James et al. 2006; Kirby et al. 2009). Complementary findings in older (>20 months) rats and mice support the use of animal models when investigating mechanisms underlying vascular dysfunction with ageing (Bearden et al. 2004; Csiszar et al. 2007; Jackson et al. 2010; Muller-Delp et al. 2012). Endothelial dysfunction may also be due to an impaired capacity to activate SKCa/IKCa (Feletou, 2009; Grgic et al. 2009). However, resolution of respective effects may be confounded by potential cross-talk between activation of SKCa/IKCa and the production of NO (or other autacoids, e.g. prostacyclin). For example, hyperpolarization arising from SKCa/IKCa activation enhances the electrical gradient for Ca2+ influx into ECs through transient receptor potential (TRP) channels, which may thereby promote the activation of eNOS (Sheng et al. 2009; Stankevicius et al. 2011).
A rise in [Ca2+]i is essential for the activation of SKCa/IKCa and eNOS (Busse et al. 2002; Ledoux et al. 2006). However, there is a lack of information with respect to endothelial [Ca2+]i homeostasis with ageing, particularly in light of ion channels and membrane transporters. We suggest that alterations in SKCa/IKCa activation or NO bioavailability may reflect corresponding changes in Ca2+ handling by the endoplasmic reticulum. Such an effect can be explained by an impaired ability of the sarcoplasmic/endoplasmic reticulum calcium ATPase (SERCA) to refill internal stores, particularly during sustained or repetitive Ca2+ release through IP3 receptors. In support of this scenario, studies from SMCs (Wray & Burdyga, 2010) and neurons (Buchholz et al. 2007) of old rats suggest that elevations in [Ca2+]i with ageing may reflect impaired SERCA function and/or mitochondrial Ca2+ uptake. There may also be a shift in the contribution of Ca2+ influx versus intracellular release in the regulation of [Ca2+]i. For example, hyperpolarization-induced Ca2+ entry through TRP channels (Stankevicius et al. 2011) may play a greater role in activating eNOS with impaired Ca2+ release from internal stores. Future studies are required to test such possibilities.
In light of events originating in SMCs being able to influence EC function, the presence of myoendothelial coupling may confound the interpretation of how ageing affects endothelium-dependent vasodilatation. For example, enhanced activation of α1-adrenoreceptors with ageing may serve a compensatory role in providing negative feedback to SMCs (e.g. via enhanced activation of eNOS and SKCa/IKCa in ECs). At the same time, the spread of vasodilatation along arterioles (Bearden et al. 2004) and from arterioles into feed arteries is impaired (Jackson et al. 2010), such that the resistance of proximal vessels can restrict muscle blood flow at rest and during exercise even when dilatation of intramuscular arterioles is maintained. Different ‘set points’ of signalling between SMCs and ECs and along the endothelium at respective levels of the resistance network should be considered in light of how ageing may affect blood flow control to skeletal muscle. Compounding this problem is the inability to functionally distinguish the effects of myoendothelial coupling without destroying one cell layer or the other (Emerson & Segal, 2000b; Segal & Jacobs, 2001). For example, commonly used gap junction blockers derived from glycyrrhetinic acid (e.g. carbenoxolone) are not selective for homocellular (endothelial) versus heterocellular (myoendothelial) gap junctions. Further, these agents can prevent endothelial hyperpolarization via SKCa/IKCa activation (Behringer et al. 2012), preempting their use in resolving potential changes in the role of gap junctions with ageing. As with Ca2+ signalling, the effect of ageing on signalling through gap junctions is ripe for future study.
Impact of ageing and oxidative stress on endothelial signalling
The reactive oxygen species (ROS) signalling pathway and the oxidative stress theory of ageing continue to gain prominence in explaining endothelial dysfunction (Heitzer et al. 2001; Bachschmid et al. 2012; Muller-Delp et al. 2012). The ROS pathway begins with superoxide (O2•−) production from a variety of intracellular sources (mitochondria, NADPH oxidases, uncoupled eNOS, and xanthine oxidase) which inactivates NO to form the reactive intermediate peroxynitrite (ONOO•−) (Bachschmid et al. 2012; Muller-Delp et al. 2012). Superoxide can also be rapidly transformed to the stable intermediate hydrogen peroxide (H2O2) by superoxide dismutase (Murphy, 2009; Bachschmid et al. 2012), which is converted into hydroxyl radicals (OH•). Through oxidizing thiol groups of disulfide bonds within and between protein subunits (Murphy, 2009), H2O2 can alter ion channel and transporter function including those discussed above (Fig. 3). Nevertheless, despite evidence of protecting endothelial function and restoring muscle blood flow with ROS scavenging (e.g. with ascorbic acid) in older humans (Kirby et al. 2009) and mice (Fleenor et al. 2012), the impact of oxidative stress on local and spreading vasodilatation remains unclear.
In contrast to the view that ROS are detrimental, investigators have also suggested that the ROS pathway may compensate for the decrease in endothelial function and reduced NO availability with age (Bachschmid et al. 2012; Muller-Delp et al. 2012). For example, H2O2 can activate SKCa/IKCa (Sesti et al. 2010) and thereby initiate EC hyperpolarization (Widlansky & Gutterman, 2011). Remarkably, H2O2 can also increase NO production (Cai et al. 2003). However, when in excess of the buffering capacity of catalases and peroxidases expressed within the endothelium, H2O2 can readily convert to cytotoxic OH• (Muller-Delp et al. 2012). In light of such ‘double-edged’ signalling of ROS, increased H2O2 levels with advanced age (Csiszar et al. 2007) may compensate for the decrease in NO bioavailability (Muller-Delp et al. 2012) while being detrimental to endothelial function if unregulated. Determining how local and integrated signalling along the endothelium is affected by the dual nature of the ROS pathway continues to advance current understanding into the aetiology of cardiovascular disease.
Summary and conclusions
The endothelium is integral to the initiation of hyperpolarization and spreading vasodilatation throughout resistance networks of skeletal muscle. In turn, these roles are influenced by surrounding SMCs through myoendothelial coupling and, thereby, the activation of α-adrenoreceptors through SNA. Whereas endothelial dysfunction with ageing has increasingly been attributed to oxidative stress, the associated influence of SMCs and perivascular nerves on endothelial function should not be dismissed. Adaptive signalling events during ageing can work towards maintaining the ability of the endothelium to promote smooth muscle relaxation along arterioles and feed arteries even while enhanced SNA impairs spreading vasodilatation and limits muscle blood flow. Greater insight into the regulation of [Ca2+]i homeostasis and ROS in respective cellular elements of the vessel wall is likely to provide therapeutic strategies for maintaining muscle blood flow, physical performance and the quality of life during ageing.
Drs Matthew Socha and Erika Westcott contributed helpful discussions. Research performed in the authors’ laboratory is supported by the United States Public Health Service, National Institutes of Health grants R37-HL041026 and R01-HL086483 to SSS and F32-HL110701 to E.J.B.