PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of jphysiolLink to Publisher's site
 
J Physiol. 2017 September 15; 595(18): 6147–6164.
Published online 2017 August 14. doi:  10.1113/JP274754
PMCID: PMC5599484

Calcium/calmodulin‐dependent kinase 2 mediates Epac‐induced spontaneous transient outward currents in rat vascular smooth muscle

Abstract

Key points

  • The Ca2+ and redox‐sensing enzyme Ca2+/calmodulin‐dependent kinase 2 (CaMKII) is a crucial and well‐established signalling molecule in the heart and brain.
  • In vascular smooth muscle, which controls blood flow by contracting and relaxing in response to complex Ca2+ signals and oxidative stress, surprisingly little is known about the role of CaMKII.
  • The vasodilator‐induced second messenger cAMP can relax vascular smooth muscle via its effector, exchange protein directly activated by cAMP (Epac), by activating spontaneous transient outward currents (STOCs) that hyperpolarize the cell membrane and reduce voltage‐dependent Ca2+ influx. How Epac activates STOCs is unknown.
  • In the present study, we map the pathway by which Epac increases STOC activity in contractile vascular smooth muscle and show that a critical step is the activation of CaMKII.
  • To our knowledge, this is the first report of CaMKII activation triggering cellular activity known to induce vasorelaxation.

Abstract

Activation of the major cAMP effector, exchange protein directly activated by cAMP (Epac), induces vascular smooth muscle relaxation by increasing the activity of ryanodine (RyR)‐sensitive release channels on the peripheral sarcoplasmic reticulum. Resultant Ca2+ sparks activate plasma membrane Ca2+‐activated K+ (BKCa) channels, evoking spontaneous transient outward currents (STOCs) that hyperpolarize the cell and reduce voltage‐dependent Ca2+ entry. In the present study, we investigate the mechanism by which Epac increases STOC activity. We show that the selective Epac activator 8‐(4‐chloro‐phenylthio)‐2′‐O‐methyladenosine‐3′, 5‐cyclic monophosphate‐AM (8‐pCPT‐AM) induces autophosphorylation (activation) of calcium/calmodulin‐dependent kinase 2 (CaMKII) and also that inhibition of CaMKII abolishes 8‐pCPT‐AM‐induced increases in STOC activity. Epac‐induced CaMKII activation is probably initiated by inositol 1,4,5‐trisphosphate (IP3)‐mobilized Ca2+: 8‐pCPT‐AM fails to induce CaMKII activation following intracellular Ca2+ store depletion and inhibition of IP3 receptors blocks both 8‐pCPT‐AM‐mediated CaMKII phosphorylation and STOC activity. 8‐pCPT‐AM does not directly activate BKCa channels, but STOCs cannot be generated by 8‐pCPT‐AM in the presence of ryanodine. Furthermore, exposure to 8‐pCPT‐AM significantly slows the initial rate of [Ca2+]i rise induced by the RyR activator caffeine without significantly affecting the caffeine‐induced Ca2+ transient amplitude, a measure of Ca2+ store content. We conclude that Epac‐mediated STOC activity (i) occurs via activation of CaMKII and (ii) is driven by changes in the underlying behaviour of RyR channels. To our knowledge, this is the first report of CaMKII initiating cellular activity linked to vasorelaxation and suggests novel roles for this Ca2+ and redox‐sensing enzyme in the regulation of vascular tone and blood flow.

Keywords: Epac, CaMKII, vascular smooth muscle

Abbreviations

2‐APB
2‐aminoethoxydiphenyl borate
8‐pCPT‐AM
8‐(4‐chloro‐phenylthio)‐2′‐O‐methyladenosine‐3′, 5‐cyclic monophosphate‐AM
AIP
autocamtide‐2‐inhibitory peptide
BKCa
large‐conductance Ca2+‐activated potassium channel
CaMKII
Ca2+/calmodulin kinase II
Epac
exchange protein directly activated by cAMP
CaV1.2
voltage‐dependent L‐type Ca2+ channel
GEF
guanine nucleotide exchange factor
GST
glutathione S‐transferase
IP3
inositol 1,4,5‐trisphosphate
IP3R
inositol 1,4,5‐trisphosphate receptor
PKA
cAMP‐dependent protein kinase
PLC
phospholipase C
RalGDS‐RBD
Ral‐guanine nucleotide‐dissociation stimulator Rap‐binding domain
RMASMC
rat mesenteric arterial smooth muscle cell
RIPA
radioimmunoprecipitation assay
RyR
ryanodine receptors
SDS
sodium dodecyl sulfate
SERCA
sarcoplasmic reticulum Ca2+ ATPase
SR
sarcoplasmic reticulum
STOCs
spontaneous transient outward currents

Introduction

The opening of plasma membrane large‐conductance Ca2+‐activated potassium (BKCa) channels in vascular smooth muscle cells is triggered by localized Ca2+ release from the subjacent sarcoplasmic reticulum (SR) (Jaggar et al. 2000). The resultant outward K+ currents, spontaneous transient outward currents (STOCs), hyperpolarize the cell membrane and decrease Ca2+ entry via voltage‐dependent L‐type (Cav1.2) Ca2+ channels. Cav1.2 channels provide the main Ca2+ influx pathway in vascular smooth muscle and, consequently, membrane hyperpolarization lowers global intracellular Ca2+ and induces muscle relaxation (Moosmang et al. 2003). The coupling between localized SR Ca2+ release (termed Ca2+ sparks) and the generation of STOCs is thus a central mechanism that opposes arterial constriction induced by intravascular pressure or vasoactive transmitters. Indeed, inhibition of Ca2+ spark activity elicits contraction in most vascular beds, emphasizing the functional significance of this negative‐feedback pathway (Jaggar et al. 2000). Ca2+ sparks themselves originate from the opening of single or clustered groups of ryanodine‐sensitive Ca2+ release channels (RyRs) on the peripheral SR. RyRs respond to changes in both cytosolic and luminal SR Ca2+ levels (ZhuGe et al. 1999) and sparks can occur spontaneously or be triggered directly or indirectly by the influx of extracellular Ca2+ (Earley et al. 2005; Essin et al. 2007).

The ability of Ca2+ sparks to affect the membrane potential through the activation of cell‐surface BKCa channels means that changes of spark frequency and/or amplitude influences vascular tone. In arterial tissue and isolated vascular myocytes, nitric oxide and a range of β‐adrenergic agonists induce vasorelaxation by increasing both spark frequency and the activity of BKCa channels (Jaggar et al. 1998; Porter et al. 1998; Mauban et al. 2001; Pucovsky et al. 2002; Kim et al. 2006). There is also strong evidence that the coupling between spark and STOC activity regulates vascular resistance and blood pressure in vivo. In mice, disruption of the gene encoding the regulatory β‐subunit of BKCa channels leads to defective coupling between sparks and STOCs and is associated with elevated blood pressure and left ventricular hypertrophy (Brenner et al. 2000; Pluger et al. 2000). Hypertensive rat models have decreased expression of BKCa β‐subunits and BKCa channels with low Ca2+ sensitivity that fail to respond normally to Ca2+ sparks (Amberg & Santana, 2003). In humans, population‐based genetic studies have identified a single‐nucleotide gain‐of‐function substitution in the β1 subunit gene that results in BKCa channels with a heightened sensitivity to Ca2+. This mutation is associated with low prevalence of diastolic hypertension and has a strong protective effect against myocardial infarction and stroke (Fernandez‐Fernandez et al. 2004; Senti et al. 2005). Enhanced expression of the pore‐forming α subunits of BKCa channel is also widely reported in hypertensive animal models, suggesting that these channels may form part of a compensatory mechanism in response to abnormal vascular tone (Cox & Rusch, 2002).

Mechanisms by which vasorelaxants increase spark‐STOC activity have not been fully defined. β‐adrenergic agonists that stimulate this pathway bind to cell surface Gs‐coupled receptors that elevate intracellular levels of cAMP, indicating a possible role for this second messenger. Downstream of cAMP lies the major cAMP effector, exchange protein directly activated by cAMP (Epac). Epacs are guanine nucleotide exchange factors (GEFs) for the small Ras‐related G proteins Rap1 and Rap2 (de Rooij et al. 1998; Kawasaki et al. 1998) and are abundant in the vasculature where they modulate cytokine signalling, strengthen endothelial barrier function and participate in vascular remodelling (Roberts & Dart, 2014). Recent studies suggest additional roles for Epac as a potent vasodilator (Sukhanova et al. 2006; Roscioni et al. 2011; Zieba et al. 2011; Roberts et al. 2013; Stott et al. 2016). β‐adrenergic and prostacyclin‐induced elevation of cAMP changes the sensitivity of the contractile proteins through Rap‐initiated alteration of myosin light chain phosphorylation (Sukhanova et al. 2006; Roscioni et al. 2011; Zieba et al. 2011) and, using the Epac‐specific cAMP homologue 8‐(4‐chloro‐phenylthio)‐2′‐O‐methyladenosine‐3′,5‐cyclic monophosphate‐AM (8‐pCPT‐AM), we have recently shown that Epac increases Ca2+ spark and STOC activity to induce membrane hyperpolarization and relaxation of vascular smooth muscle (Roberts et al. 2013). Importantly, these effects persist in the presence of potent and selective cAMP‐dependent kinase (PKA) inhibitors and suggest an important vasorelaxant pathway that is distinct from the more traditional cAMP‐PKA axis.

The mechanism by which Epac activation increases STOC activity in smooth muscle cells is unknown. In cardiomyocytes, Epac activation increases Ca2+ release through Ca2+/calmodulin‐dependent kinase 2 (CaMKII)‐dependent phosphorylation of both RyRs and phospholamban, a small pentameric protein complex that controls the activity of the SR Ca2+ ATPase (SERCA) (Oestreich et al. 2009). Phosphorylation by CaMKII increases cardiac RyR2 activity by enhancing its sensitivity to Ca2+ (Wehrens et al. 2004), whereas CaMKII‐induced phosphorylation of phospholamban relieves its inhibition of SERCA, thereby increasing Ca2+ uptake into stores and thus store content and RyR activity. Although CaMKII is recognized as an important signalling molecule in the heart, surprisingly little is known about its role in controlling vascular tone. In the present study, we assess the possible involvement of CaMKII in Epac‐mediated increases in STOC activity in contractile vascular smooth muscle cells.

Methods

Ethical approval

Tissues were obtained from adult male Wistar rats (175–225 g; Charles River Laboratories, Wilmington, MA, USA) and from wild‐type and CaMKIIδ and CaMKIIγ knockout mice (C57B/6 background; provided by Dr Tim Curtis, Queens University, Belfast, UK). Wild‐type mice were littermates of the homozygous knockouts. All animals were killed by a rising concentration of CO2 followed by cervical dislocation. The care and killing of animals conformed with the requirements of the UK Animals (Scientific Procedures) Act 1986 and was approved by the University of Liverpool and Queens University, Belfast local ethics committees.

Chemicals

The chemicals employed in the present study were KN‐93, NS11021 (Tocris Bioscience, St Louis, MO, USA), 8‐(4‐chlorophenylthio)‐2′‐O‐methyladenosine‐3′,5′‐cyclic monophosphate‐AM (8‐pCPT‐2′‐O‐Me‐cAMP‐AM; Biolog Life Science Institute, Bremen, Germany), Fura‐2‐pentapotassium (Thermo Fisher Scientific Inc., Waltham, MA, USA). All other chemicals were purchased from Sigma‐Aldrich (St Louis, MO, USA).

Cell isolation

Rat mesenteric arterial smooth muscle cells (RMASMCs) were isolated by enzymatic digestion of first‐order branches of the rat superior mesenteric artery as described previously (Hayabuchi et al. 2001). Briefly, proximal first‐order branches of rat superior mesenteric arteries were dissected and placed in low Ca2+ buffer containing (mm): 134 NaCl, 6 KCl, 0.42 Na2HPO4, 0.44 NaH2PO4, 10 Hepes, 10 glucose, 1 MgCl2 and 0.1 CaCl2 (pH 7.4) at 35°C for 10 min. After incubation, branches were moved to the first stage digestion buffer consisting of low Ca2+ buffer containing papain (1.4 mg ml−1), 4‐dithioerythritol (0.9 mg ml−1) and bovine serum albumin 0.9 mg ml−1 for 8–10 min at 35°C. Branches were then washed three times in pre‐warmed low Ca2+ buffer and transferred to second stage digestion buffer consisting of low Ca2+ buffer containing hyaluronidase (0.9 mg ml−1), collagenase (1.4 mg ml−1) and bovine serum albumin (0.9 mg ml−1) at 35°C. The incubation period for second stage digestion varied from between 8 and 12 min. Branches were subsequently transferred to fresh low Ca2+ buffer and carefully washed before dispersion of cells by gentle trituration of the tissue through a heat‐polished glass pipette. RMASMCs were stored in low Ca2+ buffer on ice and used in experiments between 1 and 8 h after digestion.

Electrophysiology

STOCs were recorded from single freshly isolated RMASMCs using the whole‐cell configuration with an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA, USA) as described previously (Roberts et al. 2013). Currents were filtered at 5 kHz and digitized at 10 kHz using a Digidata 1320A interface (Molecular Devices). The pipette‐filling solution contained (mm): 140 KCl, 3 MgCl2, 0.2 EGTA, 10 Hepes and 3 Na2ATP (pH 7.2). Extracellular recording solutions contained (mm): 6 KCl, 134 NaCl, 1 MgCl2, 2 CaCl2 and 10 Hepes (pH 7.4). STOCs were recorded at a holding potential of −20 mV and defined as events deviating from the baseline by a factor of 4 standard deviations above baseline noise. Autoinhibitory‐2‐related peptide (1 μm) was placed in the intracellular solution and allowed to dialyse into the cell for 5 min before the start of recording. In experiments to assess direct effects of 8‐pCPT‐AM on BKCa channel activity, the pipette‐filling solution contained 300 nm free calcium as calculated by Maxchelator Ca‐Mg‐ATP‐EGTA Calculator, version 1.0, using constants from NIST database #46, version 8 (http://maxchelator.stanford.edu).

Calcium fluorometry: measurement of SR Ca2+ content

Transient Ca2+ release from the SR was triggered by pulse applications of the RyR activator caffeine in single freshly isolated voltage clamped and fura‐2‐dialysed RMASMCs as described previously (Kamishima & Quayle, 2002). Cells were suspended in low Ca2+ extracellular solution containing (in mm) 145 NaCl, 6 KCl, 1 MgCl2, 0.1 CaCl2, 10 Hepes and 10 glucose (pH 7.4) and placed in a perspex perfusion chamber. Background fluorescence counts for dual excitation wavelengths were taken after establishing a tight seal but before achieving the whole‐cell clamp configuration. On achieving the whole‐cell configuration, the cell was dialysed with pipette‐filling solution containing (in mm) 145 CsCl, 3 MgCl2, 3 Na2ATP, 10 Hepes and 0.05 fura‐2 pentapotassium (pH 7.2 with CsOH). The cell membrane potential was held at −60 mV, and normal Ca2+ extracellular solution, where CaCl2 was raised to 3 mm, was superfused. The cell was illuminated alternately with 340 and 380 nm light (bandpass 5 nm) using a PTI DeltaRAM Illuminator (Horiba Scientific, Burlington, ON, Canada) and the emission signal was detected at 510 nm (40 nm bandpass). Temporal changes in photon counts were measured, backgrounds subtracted, and the complete ratio obtained at 50 Hz. Fluorescence signals and cell membrane potentials were acquired using FelixGX (Horiba Scientific). RyRs were activated by rapid application of 5 mm caffeine using a U tube superfusion system (Evans & Kennedy, 1994). After the third or fourth caffeine application, the extracellular solution was switched to normal Ca2+ solution containing either vehicle control (DMSO) or 8‐pCPT‐AM (5 μm), thereafter caffeine transients were continuously triggered at regular intervals. To convert the fluorescence ratio to [Ca2+]i, in vitro calibration was carried out using a range of EGTA‐buffered intracellular solutions containing a desired free Ca2+ calculated with MAXchelator. Fluorescence ratios in the absence of added CaCl2 (R min) and in the presence of saturating Ca2+ (R max) were decreased by 15% to account for viscosity (Poenie et al. 1986), and the K d of fura‐2 was determined as 160 nm. To profile Ca2+ transients, three parameters, amplitude, duration and rate of cytosolic Ca2+ increase, were examined. Amplitude was measured as the [Ca2+]i difference between resting and peak [Ca2+]i. Duration was determined as the time that [Ca2+]i remained above the sum of resting [Ca2+]i and a half‐amplitude. To examine the early stage of the cytosolic Ca2+ rise, data points from baseline to [Ca2+]i of 200 nm were fit with a third‐order polynomial fit using Sigmaplot, version 13 (Systat Software Inc., Chicago, IL, USA) and the rate of Ca2+ increase was determined as a half‐time taken to reach a [Ca2+]i of 200 nm. For each parameter, the value of the third Ca2+ transient after switching to the test solution was normalized against that of the Ca2+ transient immediately before solution change.

SDS‐PAGE and immunoblotting

First‐ or second‐order branches of rat superior mesenteric arteries (five to eight branches) were homogenized in a pre‐cooled hand‐held homogenizer containing 100–500 μl of ice‐cold lysis buffer comprising (mm): 20 Tris‐HCl; 250 NaCl, 3 EDTA, 3 EGTA, 0.5% (v/v) Triton‐X 100 (pH 7.6), 1% (v/v) protease inhibitor cocktail (Sigma‐Aldrich) and 1X PhosSTOP Phosphatase Inhibitor Cocktail Tablets (Roche, Basel, Switzerland). Resultant lysates were placed on ice for 10 min then centrifuged at 16 100 g for 10 min at 4°C. Supernatants were removed and mixed 1:3 (v/v) with 4 × SDS sample buffer, before heating to 98°C for 10 min. Lysates for phospholamban immunoblots were prepared by lysis in either radioimmunoprecipitation assay (RIPA) buffer or 1 × SDS sample buffer to maximize phosphorylation preservation and increase protein solubilization. RIPA lysates were then mixed 1:3 with 4 × sample buffer as with Triton X‐100 based lysates. Samples were kept at −20°C until use. Proteins within the lysates were resolved by SDS‐PAGE on either 5%, 10% or 15% polyacrylamide‐Tris‐glycine gels (dependent on the protein of interest) and transferred electrophoretically onto nitrocellulose membranes (Hybond ECL; GE Healthcare, Little Chalfont, UK). When immunoblotting for phospholamban, proteins were transferred to 0.22 μm polyvinylidene fluoride membrane. Immunoblotting was performed as described previously (Sampson et al. 2003). Where indicated, membranes were stripped at room temperature for 1 h in 50 mm Tris‐HCl (pH 6.8), 1% SDS.

Antibodies

For immunoblotting, the primary antibodies used were: phospho‐CaMKII (T286: dilution 1:1000); phospholamban (dilution 1:500), Rap 1A/B (dilution 1:1000) were from Cell Signaling Technology (Beverly, MA, USA). CaMKII (pan: dilution 1:2500) was from Abcam (Cambridge, MA, USA). Smooth muscle α actin (dilution 1:10,000) was from Sigma‐Aldrich. The secondary antibodies used were anti‐mouse IgG (H+L) HRP conjugated polyclonal antibody and anti‐rabbit IgG (H+L) HRP conjugated polyclonal antibody (Stratech Scientific Ltd, Newmarket, UK).

Expression of GST‐RalGDS‐RBD and Rap1‐GTP pull‐down

GST‐RalGDS‐RBD fusion protein, consisting of glutathione S‐transferase (GST) fused to the Rap binding domain (RBD) of Ral guanine nucleotide dissociation stimulator (RalGDS), was expressed from the pGEX4T3‐GST‐RalGDS‐RBD plasmid, kindly donated by Professor Johannes Bos (University Medical Centre, Utrecht, The Netherlands), in the BL21 strain of Escherichia coli as described previously (Van Triest et al. 2001). Pull‐down of active (GTP‐bound) Rap was carried out as described previously (Roberts et al. 2013).

Total RNA extraction and reverse transcription

Total RNA was extracted and purified from first‐ or second‐order branches of rat superior mesenteric arteries using an RNeasy Mini Kit® (Qiagen, Valancia, CA, USA) in accordance with the manufacturer's instructions. Extracted RNA was treated with DNase I by incubating 8 μl of total RNA, 1 μl of 10 × DNase I buffer and 1 μl of DNase I (1 U μl−1; Invitrogen, Carlsbad, CA, USA) at room temperature for 15 min. Then, 1 μl of EDTA (25 mmol l−1) was added and the reaction heated at 65°C for 10 min. First‐strand cDNA was synthesized using SuperScript® III reverse transcriptase (Invitrogen) in accordance with the manufacturer's instructions. The purity and concentration of the resulting cDNA template was determined by measuring absorbance at 260 and 280 nm using a NanoDrop spectrophotometer (Thermo Fisher Scientific Inc.).

PCR

Touchdown PCR was carried out using HotStarTaq Master Mix Plus (Qiagen) in accordance with the manufacturer's instructions. Primers used to amplify specific CaMKII isoforms were: δ: Forward (5′‐GCATGTGGCGTCATCCTCTA‐3′) Reverse (5′‐TCCTTTACCCCATCCGGTT‐3′); γ: Forward (5′‐CATCCACCAGCATGACATCG‐3′) Reverse (5′‐CTTTCCTCAAGACCTCAGG‐3′). δ splice variants: Forward (5′‐CGGAAATTGAAGGGTGCCATC‐3′) Reverse (5′‐CCCTACCAGGTGTACGTGAG‐3′); γ splice variants: Forward (5′‐GAAGGGTGCCATCCTCACAA‐3′) Reverse (5′‐GTTACCAAGGGCTTCTGGCT‐3′). Following an initial 5 min denaturing step at 95°C, each reaction went through a touchdown PCR protocol for 15 cycles: 94°C for 1 min, annealing temperature (ranging from 65°C to 50°C, decreasing by 1°C each cycle) for 30 s, extension at 72°C for 1 min. This was followed by 25 cycles at the lowest annealing temperature. Final extension was at 72°C for 5 min. Products were analysed by running on a 3% agarose gel containing Midori Green (dilution 1:10,000; GC Biotech, Alphen aan den Rijn, The Netherlands) for ~1 h at 80 V. Bands were excised under ultraviolet light and products purified using a QIAquick Gel Extraction kit (Qiagen) in accordance with the manufacturer's instructions. Products were verified by sequencing (GATC Biotech, Konstanz, Germany).

Statistical analysis

Results are expressed as the mean ± SEM. Intergroup differences were analysed using repeated measures ANOVA followed by Newman–Keuls post hoc test, or Student's t test for simple comparisons; levels of significance were P < 0.05 (*) and P < 0.01 (**). For electrophysiological and calcium fluorometry recordings, n is the number of cells recorded from. Cells were from at least three different animals. For immunoblots, n is the number of experimental repeats. These repeats were on different days and with tissue taken from different animals.

Results

Selective Epac activation causes autophosphorylation of specific CaMKII isoforms in rat mesenteric artery

CaMKII is encoded by four closely related genes (α, β, γ and δ) with alternative splicing producing multiple variants (Soderling & Stull, 2001). The enzyme itself is a large complex made up of 12 subunits, with the predominant CaMKII holoenzymes found in vascular smooth muscle forming as a mixture of splice products from γ and δ genes (Singer, 2012). Reverse transcription‐PCR using cDNA derived from rat mesenteric artery as a template and primers designed to detect (i) CaMKII isoforms and (ii) γ and δ splice variants confirmed the presence of mRNA for at least two γ variants and two δ variants in this tissue (Fig. 1 A). Sequencing of these products revealed them to be γ predicted isoforms X18 and X16, and δ isoforms 2 and/or 3. Consistent with the expression of multiple CaMKII isoforms, immunoblots of arterial lysates using pan‐specific CaMKII antibodies produced three/four distinct immunoreactive bands (Fig. 1 Ba, lower). To assess the effect of Epac activation on CaMKII activity, we incubated arteries in the selective Epac activator 8‐pCPT‐2′‐O‐Me‐cAMP‐AM (8‐pCPT‐AM; 10 μm) or vehicle control (DMSO). As a positive control, we also incubated arteries in the Ca2+ ionophore A23187 (12.5 μm), which would be expected to globally raise intracellular Ca2+ and activate CaMKII. Immunoblot analysis of 8‐pCPT‐AM‐treated arterial lysates indicated phosphorylation of CaMKII at Thr286/7, an autophosphorylation site that indicates CaMKII activation, following 8‐pCPT‐AM exposure (n = 3) (Fig. 1 Ba, upper, Bb). Of the three/four immunoreactive bands detected by pan‐specific anti‐CaMKII, one band (second from bottom) strongly bound phospho‐CaMKII antibodies against Thr286/7 following 8‐pCPT‐AM treatment. Exposure to A23187 induced phosphorylation of all immunoreactive bands detected by pan‐specific anti‐CaMKII (Fig. 1 Ba, right).

Figure 1
Selective activation of Epac causes autophosphorylation of specific CaMKII isoforms in rat mesenteric artery

To identify the CaMKII isoform/s differentially phosphorylated by Epac activation, we used immunoblotting to analyse arterial tissue obtained from mice in which the gene encoding either CaMKIIδ or CaMKIIγ had been globally knocked out (Backs et al. 2009; Backs et al. 2010). Immunoblots of arterial lysates from wild‐type mice using CaMKII antibodies showed a similar pattern of three/four immunoreactive bands as observed in rat tissue (Fig. 1 Ca and Cb). In arterial lysates obtained from CaMKIIδ knockout animals, the lowest molecular weight band was absent. By contrast, in CaMKIIγ knockout lysates, the lower band remained but the higher molecular weight bands were missing (Fig. 1 Ca and Cb). This suggests that Epac activation predominantly induces phosphorylation of CaMKII γ variants in vascular tissue. Whole arterial lysates contain a combination of different cell types and we thus undertook electrophysiological experiments on single isolated mesenteric myocytes to directly assess the effects of Epac‐mediated CaMKII activation in vascular smooth muscle.

Inhibition of CaMKII attenuates the increase in STOC frequency mediated by Epac in RMASMCs

Selective activation of Epac with 8‐pCPT‐AM induces vasorelaxation by increasing both the frequency and amplitude of STOCs in rat mesenteric beds (Roberts et al. 2013). These 8‐pCPT‐AM‐mediated changes in STOC activity persist in the presence of the potent PKA inhibitor PKI‐(Myr‐14‐22)‐amide, indicating that they occur independently of PKA activation (Roberts et al. 2013). To assess whether 8‐pCPT‐induced activation of CaMKII is involved in the increase of STOC activity, we used the whole‐cell recording technique to record STOCs in mesenteric myocytes following activation of Epac in the presence and absence of pharmacological inhibitors of CaMKII.

In whole‐cell recordings from single, freshly isolated RMASMCs, application of the CaMKII inhibitor KN‐93 (500 nm) reversed the increase in STOC frequency (1.71 ± 0.22 to 0.61 ± 0.18 s−1; mean ± SEM) induced by 8‐pCPT‐AM (5 μm; P < 0.05; n = 5; paired t test) (Fig. 2). The increase in STOC amplitude was the result of an increase in the probability of larger events occurring in the presence of 8‐pCPT‐AM (Fig. 2 C). This was not mimicked by application of KN‐92 (500 nm), an inactive KN‐93 analogue (n = 3) (Fig. 2 Ab). Because KN‐93 has cellular effects other than inhibition of CaMKII (Pellicena & Schulman, 2014), we also undertook experiments using autocamtide‐2‐inhibitory peptide (AIP; 1 μm), a highly specific peptide inhibitor of CaMKII that corresponds to its autoinhibitory domain (Ishida et al. 1995). AIP was included in the pipette‐filling solution and allowed to dialyse into the cell for a minimum of 5 min on establishment of the whole‐cell configuration. In the presence of AIP, application of 8‐pCPT‐AM did not increase STOC frequency or amplitude in RMASMCs (Fig. 3 A; control response shown in Fig. 3 B). Indeed, there was a significant reduction in STOC frequency following 8‐pCPT‐AM addition from 1.08 ± 0.23 s−1 under basal conditions to 0.59 ± 0.12 s−1 in the presence of AIP (P < 0.05, n = 4; paired t test) (Fig. 3 C).

Figure 2
Epac‐mediated increases in STOC activity are reversed by KN‐93 inhibition of CaMKII in mesenteric smooth muscle cells
Figure 3
Autocamtide‐2‐inhibitory peptide blocks Epac‐mediated increases in STOC activity

Inhibition of inositol 1,4,5‐trisphosphate (IP3) receptors blocks Epac‐mediated CaMKII activation and increases in STOC activity

Because CaMKII is activated by the binding of Ca2+/calmodulin (Soderling & Stull, 2001), we next explored possible sources of Ca2+ for Epac‐mediated CaMKII activation. 8‐pCPT‐AM was unable to increase STOC activity in RMASMCs pre‐incubated in 2‐aminoethoxydiphenyl borate (2‐APB) (100 μm), an IP3 receptor (IP3R) inhibitor. 2‐APB application following STOC activation with 8‐pCPT‐AM caused an initial rapid increase in STOC frequency followed by a decline to levels significantly below those measured in 8‐pCPT‐AM alone (from 1.72 ± 0.27 s−1 to 0.83 ± 0.23 s−1; P < 0.05; n = 6; Student–Newman–Keuls) (Fig. 4 A and B). These data suggest that Epac‐induced CaMKII activation may be triggered by Ca2+ release from intracellular stores via IP3Rs. To directly link the effects of 2‐APB to CaMKII activity, we stimulated rat mesenteric arteries with 8‐pCPT‐AM (5 μm) in the presence and absence of 2‐APB (100 μm) before homogenizing the tissue and quantifying the changes in CaMKII phosphorylation by immunoblot analysis using phospho‐specific antibodies against the CaMKII Thr286/7 autophosphorylation site. In the presence of 2‐APB, 8‐pCPT‐AM was unable to induce phosphorylation of Thr286/7, indicating that Epac‐induced Ca2+ efflux through IP3Rs is a probable mechanism that initiates CaMKII activity (n = 3) (Fig. 4 C). Consistent with the idea that store Ca2+ is needed to activate CaMKII, pretreatment with thapsigargin (500 nm), which depletes intracellular Ca2+ stores by blocking Ca2+ uptake via SERCA, also abolished 8‐pCPT‐AM‐induced phosphorylation of CaMKII at Thr286/7 (n = 3) (Fig. 4 D).

Figure 4
Inhibition of IP3 receptors blocks Epac‐mediated CaMKII phosphorylation and increases in STOC activity

Upstream of IP3Rs, inhibition of the IP3‐generating enzyme phospholipase C (PLC) with U73122 (2 μm) significantly reduced the ability of 8‐pCPT‐AM to induce sustained activation of Rap1 (Fig. 5). Epac directly activates Rap1, which in turn can activate PLCε, an unusual PLC isoform that possesses an N‐terminal Ras GEF domain and two C‐terminal Ras binding domains (Kelley et al. 2001). Activated PLCε has two roles, further activation of Rap1 through GEF activity and phosphoinositol hydrolysis to produce IP3 (Oestreich et al. 2009). Active Rap levels were determined by incubating arterial lysates with fusion proteins comprising GST and the Rap‐binding domain of Ral‐guanine nucleotide‐dissociation stimulator (GST‐RalGDS‐RBD), which only bind the active GTP‐bound form of Rap1. Glutathione‐sepharose beads were then used to specifically pull‐down Rap1‐GTP, which was quantified by immunoblot analysis. The abolition of 8‐pCPT‐AM‐induced Rap activation by U73122 suggests that this PLC‐mediated feedback mechanism is essential for sustained Rap activity in vascular smooth muscle. Further investigation downstream of PLC using U73122 (i.e. its effect on Epac‐induced STOCs) was not undertaken as a result of the known side effect of U73122 on SR Ca2+ pumps (MacMillan & McCarron, 2010).

Figure 5
The PLC inhibitor U73122 blocks sustained Rap1 activation by 8‐pCPT‐AM

8‐pCPT‐AM does not directly activate BKCa channels

We have previously shown that selective activation of Epac with 8‐pCPT‐AM in the presence of PKA inhibition increases the frequency of Ca2+ sparks (Roberts et al. 2013). In neurones, however, CaMKII phosphorylation of BKCa channels has been shown to increase their open probability by shifting their window of voltage activation towards more hyperpolarized membrane potentials (van Welie & du Lac, 2011). We therefore investigated whether a component of the 8‐pCPT‐AM effect on STOC activity could originate from direct effects on BKCa channels. In whole‐cell recordings from isolated mesenteric myocytes held at −20 mV with ryanodine (15 μm) and Ca2+ (300 nm) included in the pipette‐filling solution, application of 8‐pCPT‐AM (5  μm) failed to increase BKCa channel activity (n = 6) (Fig. 6 A). Under these conditions, Ca2+ release via RyRs will be blocked but BKCa channels will still be available for activation. This is demonstrated by the subsequent addition of the selective BKCa channel activator NS11021 (10 μm), which induced a large outward current (Fig. 6 A). Consistent with the idea that Epac‐mediated changes in STOC activity rely on underlying changes in the activity of RyRs, application of ryanodine (15 μm) to the superfusing bath solution reversed the increases in STOC activity induced by the application of 8‐pCPT‐AM (5  μm; n = 4) (Fig. 6 B and Da). Similarly, in cells pre‐incubated in ryanodine (15 μm), subsequent application of 8‐pCPT‐AM does not increase STOC activity (n = 3) (Fig. 6 C and Db).

Figure 6
8‐pCPT‐AM does not directly activate BKCa channels but STOCs cannot be generated by 8‐pCPT‐AM in the presence of ryanodine

Epac activation decreases the rate of caffeine‐induced Ca2+ release from intracellular stores

Epac activation may increase RyR activity by enhancing RyR Ca2+ sensitivity, or by increasing the Ca2+ content of the RyR‐accessible Ca2+ store. We thus measured the effect of Epac activation on store content and the rate of Ca2+ clearance from the cytosol by inducing transient Ca2+ release from the SR by pulse applications of the RyR activator caffeine in voltage clamped and fura‐2‐dialysed myocytes (Fig. 7).

Figure 7
8‐pCPT‐AM slows the initial rate of [Ca2+]i rise induced by caffeine without affecting the Ca2+ transient amplitude

Repeated 100 ms application of caffeine (5 mm; indicated by arrow heads) at 60 s intervals triggered a train of Ca2+ release transients (Fig. 7 A). At the point indicated by the bar above the trace, the extracellular superfusing solution was switched to test solutions containing either vehicle control (DMSO) or 8‐pCPT‐AM (5 μm). In both cases, the amplitude of the caffeine‐induced Ca2+ transients decreased during the course of the experiment, suggesting that the SR Ca2+ is being slowly depleted. Epac activation with 8‐pCPT‐AM had no significant effect on Ca2+ transient amplitude during the course of the experiment (relative amplitude for DMSO‐treated cells 0.68 ± 0.07 and for 8‐pCPT‐AM treated cells 0.80 ± 0.03, n = 5, 4 respectively; P > 0.05; unpaired t test) (Fig. 7 C, right). There was however a significant shortening of the relative duration of the transient following Epac activation (DMSO treated cells 1.39 ± 0.15 and 8‐pCPT‐AM treated cells 0.98 ± 0.08, n = 5, 4 respectively; P < 0.05; unpaired t test) (Fig. 7 C, middle).

During the experiments, it was noted that the initial upward deflection of the Ca2+ transient appeared slower following 8‐pCPT‐AM application. Thus, the early period of [Ca2+]i increase in response to caffeine was examined more closely. The initial rate of Ca2+ rise (from baseline to [Ca2+]i of 200 nm) was almost indistinguishable before and after DMSO application, despite the later Ca2+ transients being smaller in amplitude (Fig. 7 B, left). The initial rate of Ca2+ rise was, however, considerably slower after application of 8‐pCPT‐AM (5 μm) (Fig. 7 B, right). The relative initial rate of Ca2+ rise was 1.17 ± 0.07 and 2.00 ± 0.41 for DMSO and 8‐pCPT‐AM treated cells respectively (n = 4, 5; P < 0.05, unpaired t test) (Fig. 7 C, left). These results suggest that Epac activation has no significant effect upon store content, although it may affect the kinetic behaviour of the RyR release channels.

Epac and subsequent CaMKII activation probably induces changes in RyR activity through phosphorylation of the release channel itself or associated regulatory proteins. We were unable to detect 8‐pCPT‐AM‐induced phosphorylation of RyR at canonical CaMKII phosphorylation sites. This may simply reflect the levels of these proteins within vascular smooth muscle cells, which is considerably less than in cardiomyocytes. Equally, we could not reliably detect phosphorylation of the SERCA pump regulator phospholamban at Thr17, another key CaMKII target in the heart. Interestingly, we did see 8‐pCPT‐AM‐mediated changes with respect to both phospholamban mobility within gels and susceptibility to detergent extraction from membranes, which may indicate protein modification following Epac activation (Fig. 8).

Figure 8
8‐pCPT‐AM exposure induces changes in phospholamban solubility and gel mobility

Discussion

The Ca2+‐sensing enzyme CaMKII is a well‐established and important modulator of cardiomyocyte contraction, although its role in the control of vascular tone is surprisingly poorly understood (Toussaint et al. 2016). In the heart, aside from its Ca2+ sensitivity, CaMKII responds to the cellular redox state through oxidation of methionine residues in its regulatory domain (Erickson et al. 2008). However, in differentiated vascular smooth muscle, which relies on complex patterns of Ca2+ signalling and functions to translate changes in oxidative stress into contractile behaviour, little is known about its role. We show that Epac‐mediated activation of membrane currents driving smooth muscle hyperpolarization occur via the activation of CaMKII. To our knowledge this is the first report of CaMKII activation being associated with the initiation of vasorelaxation.

Activation of Epac has previously been shown to induce smooth muscle relaxation by increasing the frequency of localized Ca2+ release from RyRs located on the peripheral sarcoplasmic reticulum (Roberts et al. 2013). These subsurface Ca2+ sparks activate BKCa channels in the plasma membrane, evoking STOCs that hyperpolarize the cell and reduce voltage‐dependent Ca2+ entry (Roberts et al. 2013). In the present study, we investigate the mechanism by which Epac increases STOC activity. We show that (i) the ability of Epac to increase STOC frequency and amplitude in rat mesenteric artery smooth muscle cells depends upon the activation of CaMKII; (ii) Epac activation preferentially induces autophosphorylation of specific CaMKII γ isoform/s; (iii) Epac‐induced CaMKII activation is probably initiated by IP3‐mobilized Ca2+; and (iv) Epac activation has little effect on intracellular Ca2+ store content but affects caffeine‐induced store release, possibly as a result of changes in the kinetic behaviour or synchronous opening of RyRs. We do not consider that the observed increase in STOC activity is attributable to Epac‐activated CaMKII directly phosphorylating and activating BKCa channels. We see no significant BKCa channel activation by 8‐pCPT‐AM under conditions where RyR channels are blocked, even though BKCa channels are still susceptible to activation by NS11021.

The main CaMKII holoenzymes found in vascular smooth muscle form from splice products from γ and δ genes (Singer, 2012). Consistent with this, we find messenger RNA encoding at least two CaMKII γ variants and two CaMKII δ variants in rat mesenteric artery. The expression of multiple CaMKII isoforms in this tissue is confirmed by immunoblots of arterial lysates using pan‐specific CaMKII antibodies that produce three/four immunoreactive bands, only one of which is significantly phosphorylated following Epac activation. By immunoblotting arterial tissue obtained from mice in which the gene encoding either CaMKIIδ or CaMKIIγ had been globally deleted (Backs et al. 2009, 2010), we were able to determine that Epac activation predominantly induces phosphorylation of CaMKII γ variants. This is consistent with the finding that CaMKIIγ expression dominates in differentiated smooth muscle (Kim et al. 2000; Marganski et al. 2005). CaMKIIδ, in contrast, is a relatively minor isoform in contractile muscle. It is up‐regulated in synthetic/proliferating smooth muscle and is required for injury‐induced neointima formation, largely through its regulation of the expression and activity of cell cycle activators (Li et al. 2011). We cannot entirely rule out the possibility that δ variants make up a minor component of the immunoreactive band that strongly binds phospho‐CaMKII antibodies following 8‐pCPT‐AM treatment. Faint anti‐CaMKII immunoreactive bands persist at this molecular weight in CaMKIIγ knockout lysates, and it is possible that δ and γ subunits exist together in the functional holoenzyme complex. Interestingly, the two major PCR products we detect using primers designed to amplify γ isoforms are predicted splice variants of the γ gene whose expression has never been reported. Thus, the CaMKII isoforms involved in this Epac‐activated pathway may be novel. We did not undertake any functional work with the δ or γ knockout tissue because CaMKIIδ and CaMKIIγ show redundancy and are able to functionally compensate for each other (Kreusser et al. 2014).

Because these experiments were conducted in whole arteries, we cannot fully distinguish whether the isoforms we detect are in smooth muscle cells, endothelial cells or innervating nerves. However, our electrophysiological experiments have all been conducted in single isolated smooth muscle cells and clearly show that inhibition of CaMKII activity by KN‐93 or AIP, a highly specific peptide inhibitor of CaMKII, blocks the ability of Epac to generate STOC activity.

Intuitively, as a Ca2+‐sensing enzyme, agents that elevate intracellular Ca2+ should activate CaMKII and, indeed, in large blood vessels, CaMKII is activated in response to contractile stimuli and this activation maintained throughout tonic contraction (Kim et al. 2000; Rokolya & Singer, 2000). In ferret aortae, inhibition of CaMKII with KN‐93 decreases the amplitude of contraction induced by high extracellular K+, although it has no significant effect on the size of contractions induced by the α adrenoceptor agonist phenylephrine (Kim et al. 2000). Similarly, in pig carotid artery, KN‐93 inhibits both the amplitude of contraction and the ability to maintain force in response to high extracellular K+, although it inhibits only force maintenance in response to histamine (Rokolya & Singer, 2000). It is suggested that the site of Ca2+ release/influx and/or calmodulin compartmentalization controls the spatial activation of CaMKII in response to different upstream stimuli. This would be in keeping with studies in cardiomyocytes that demonstrate distinct pools of CaMKIIδ linked to different upstream pathways (Mishra et al. 2011). This specificity of activation is controlled by the mobilization of different Ca2+ stores as opposed to the subcellular compartmentation of different subtypes of the enzyme.

By contrast to large vessels, in mesenteric arteries from transgenic mice expressing the inhibitor peptide CaMKIIN in smooth muscle, CaMKII inhibition has no effect on vasoconstriction in response to KCl, angiotensin II, or phenylephrine (Prasad et al., 2013). Instead, CaMKII supports intracellular Ca2+ homeostasis by phosphorylating and maintaining the activity of voltage‐dependent Ca2+ channels and phospholamban. The functionality of Epac‐induced CaMKII activation in mediating vasorelaxation will presumably rely on the selective activation of CaMKII holoenzymes residing sufficiently close to activating IP3‐sensitive release channels and downstream targets such as RyRs and their associated regulatory proteins. We would predict that these pools of CaMKII are distinctly accessible to Epac‐mediated upstream events and shielded from contractile stimuli.

Our data suggest that CaMKII activity in vascular smooth muscle is triggered by Epac‐induced Ca2+ efflux through IP3Rs. The immediate downstream target of Epac is the small‐Ras‐related G‐protein Rap1. We propose that, following activation, Rap interacts with the C‐terminal Ras binding domains of PLCε (Kelley et al. 2001). Subsequent activation of PLCε results in two separate events: phosphoinositol hydrolysis and the formation of IP3 and further activation of Rap through the N‐terminal Ras GEF domain of PLCε. Consistent with this model, the PLC inhibitor U73122 significantly reduced the ability of the selective Epac activator, 8‐pCPT‐AM, to induce sustained activation of Rap1. Indeed, the abolition of 8‐pCPT‐AM‐induced Rap activation by U73122 suggests that this PLC‐mediated feedback mechanism is essential for sustained Rap activation in vascular smooth muscle. We also find that, in the presence of the IP3 receptor inhibitor 2‐APB, 8‐pCPT‐AM is unable to induce phosphorylation of CaMKII at Thr286/7 or increase STOC frequency/amplitude, which would be consistent with IP3‐released Ca2+ being essential for activation of the CaMKII holoenzyme. Pretreatment with thapsigargin, which depletes intracellular Ca2+ stores by blocking Ca2+ uptake via SERCA, also abolished 8‐pCPT‐AM‐induced CaMKII phosphorylation. During the course of the experiments, we noted that 2‐APB application following STOC activation with 8‐pCPT‐AM had a bi‐phasic effect on STOC activity, causing an initial rapid increase in STOC frequency followed by a decline to levels significantly below those measured in 8‐pCPT‐AM alone. Further experiments showed that application of 2‐APB alone caused a transient increase in basal STOC frequency (data not shown), which may indicate a constant Ca2+ leak from a common Ca2+ store via IP3Rs which, when blocked, alters store content and RyR activity. We see some evidence of basal phosphorylation of CaMKII in immunoblots, which may be consistent with this idea.

Downstream of CaMKII activation, we investigated possible mechanisms by which CaMKII could affect STOC activity. STOCs are generated by the synchronized opening of groups of sarcolemmal BKCa channels in response to localized Ca2+ release from RyRs on the subjacent SR. Changes in STOC frequency or amplitude thus give a direct indication of changes in underlying behaviour of RyRs. To increase STOC activity, Epac may affect RyR behaviour directly via enhancement of Ca2+ sensitivity, or indirectly by increasing the content of the RyR‐accessible Ca2+ store. We thus measured the effect of Epac activation on store content and the rate of Ca2+ clearance from the cytosol by inducing transient Ca2+ release from the SR by pulse applications of the RyR activator caffeine in voltage clamped and fura‐2‐dialysed myocytes. Epac activation significantly slowed the initial global rise in [Ca2+]i in response to caffeine but had no effect on the relative transient amplitude. This suggests that depleted SR Ca2+ content is probably not the reason for the sluggish Ca2+ rise in 8‐pCPT‐AM treated cells. A possible explanation for these observations is that the rapid and synchronized opening of RyRs is compromised by Epac activation. CaMKII probably induces changes in RyR activity through phosphorylation of the release channel itself or associated regulatory proteins (Van Petegem, 2012). RyR1, 2 and 3 have been reported to be expressed in vascular smooth muscle and all contain potential phosphorylation sites for CaMKII. We were unable to detect 8‐pCPT‐AM‐induced phosphorylation of vascular RyRs using phospho‐specific antibodies against serine 2814, the CaMKII phosphorylation sites on RyR2 that enhances Ca2+ sensitivity and increases RyR open probability in cardiac muscle (Ai et al. 2005). This may simply reflect the low level of RyR proteins within vascular smooth muscle or phosphorylation at an alternative site. CaMKII may also directly or indirectly modify the behaviour of associated regulatory proteins such as FK‐506 binding protein, which functions to synchronize the activity of RyRs by coupling the opening and closure of neighbouring channels (Marx et al. 1998). Loss of the ability to simultaneously open in response to caffeine would account for our Epac‐mediated slowing of global [Ca2+]i rise. Although we see no clear changes in SR content following activation of Epac, we cannot rule out an increase in uptake that balances the increased store leak (spark activity). As with RyR, we could not reliably detect phosphorylation of the SERCA pump regulator phospholamban at Thr17, another key CaMKII target in the heart. Interestingly, we did see 8‐pCPT‐AM‐mediated changes in both phospholamban mobility within gels and susceptibility to detergent extraction from membranes, which may indicate protein modification following Epac activation (Bidlack & Shamoo, 1980).

In conclusion, the results of the present study indicate that Epac‐induced STOC activity in contractile vascular smooth muscle occurs via the activation of CaMKII and suggests that functionally distinct pools of CaMKII may be pivotal in mediating cyclic AMP‐induced vasodilatation.

Additional information

Competing interests

The authors declare that they have no competing interests.

Author contributions

ESAH, TK, JMQ and CD were responsible for the conception and design of experiments. ESAH was responsible for the electrophysiological, biochemical and molecular experiments. TK was responsible for the calcium fluorometry experiments. ESAH, TK and CD were responsible for analysis and interpretation of data. ESAH and CD were responsible for drafting the article. ESAH, TK, JMQ and CD were responsible for revising the article critically for important intellectual content and approving the final version to be published.

Funding

This work is supported by a British Heart Foundation project grant PG/14/55/30973 and BBSRC Doctoral Training Grant (DTP1‐NLD) to ESAH.

Acknowledgements

We thank Professor Johannes Bos (University Medical Centre, Utrecht, The Netherlands) for the pGEX4T3‐GST‐RalGDS‐RBD plasmid. We thank Dr Tim Curtis (Queen's University Belfast) and Professor Eric Olson (University of Texas Southwestern Medical Centre, USA) for tissue from CaMKII δ and γ knockout mice.

References

  • Ai X, Curran JW, Shannon TR, Bers DM & Pogwizd SM (2005). Ca2+/calmodulin‐dependent protein kinase modulates cardiac ryanodine receptor phosphorylation and sarcoplasmic reticulum Ca2+ leak in heart failure. Circ Res 97, 1314–1322. [PubMed]
  • Amberg GC & Santana LF (2003). Downregulation of the BK channel beta 1 subunit in genetic hypertension. Circ Res 93, 965–971. [PubMed]
  • Backs J, Backs T, Neef S, Kreusser MM, Lehmann LH, Patrick DM, Grueter CE, Qi XX, Richardson JA, Hill JA, Katus HA, Bassel‐Duby R, Maier LS & Olson EN (2009). The delta isoform of CaM kinase II is required for pathological cardiac hypertrophy and remodeling after pressure overload. Proc Natl Acad Sci USA 106, 2342–2347. [PubMed]
  • Backs J, Stein P, Backs T, Duncan FE, Grueter CE, McAnally J, Qi XX, Schultz RM & Olson EN (2010). The gamma isoform of CaM kinase II controls mouse egg activation by regulating cell cycle resumption. Proc Natl Acad Sci USA 107, 81–86. [PubMed]
  • Bidlack JM & Shamoo AE (1980). Adenosine 3′,5′‐monophosphate‐dependent phosphorylation of a 6000 and a 22000 dalton protein from cardiac sarcoplasmic reticulum. Biochim Biophys Acta 632, 310–325. [PubMed]
  • Brenner R, Perez GJ, Bonev AD, Eckman DM, Kosek JC, Wiler SW, Patterson AJ, Nelson MT & Aldrich RW (2000). Vasoregulation by the beta 1 subunit of the calcium‐activated potassium channel. Nature 407, 870–876. [PubMed]
  • Cox RH & Rusch NJ (2002). New expression profiles of voltage‐gated ion channels in arteries exposed to high blood pressure. Microcirculation 9, 243–257. [PubMed]
  • de Rooij J, Zwartkruis FJT, Verheijen MHG, Cool RH, Nijman SMB, Wittinghofer A & Bos JL (1998). Epac is a Rap1 guanine‐nucleotide‐exchange factor directly activated by cyclic AMP. Nature 396, 474–477. [PubMed]
  • Earley S, Heppner TJ, Nelson MT & Brayden JE (2005). TRPV4 forms a novel Ca2+ signaling complex with ryanodine receptors and BKCa channels. Circ Res 97, 1270–1279. [PubMed]
  • Erickson JR, Joiner MLA, Guan X, Kutschke W, Yang JY, Oddis CV, Bartlett RK, Lowe JS, O'Donnell SE, Aykin‐Burns N, Zimmerman MC, Zimmerman K, Ham AJL, Weiss RM, Spitz DR, Shea MA, Colbran RJ, Mohler PJ & Anderson ME (2008). A dynamic pathway for calcium‐independent activation of CaMKII by methionine oxidation. Cell 133, 462–474. [PubMed]
  • Essin K, Welling A, Hofmann F, Luft FC, Gollasch M & Moosmang S (2007). Indirect coupling between Ca(v)1.2 channels and ryanodine receptors to generate Ca2+ sparks in murine arterial smooth muscle cells. J Physiol 584, 205–219. [PubMed]
  • Evans RJ & Kennedy C (1994). Characterization of P2 purinoceptors in the smooth muscle of the rat tail artery – a comparison between contractile and electrophysiological responses. Br J Pharmacol 113, 853–860. [PubMed]
  • Fernandez‐Fernandez JM, Tomas M, Vazquez E, Orio P, Latorre R, Senti M, Marrugat J & Valverde MA (2004). Gain‐of‐function mutation in the KCNMB1 potassium channel subunit is associated with low prevalence of diastolic hypertension. J Clin Invest 113, 1032–1039. [PubMed]
  • Hayabuchi Y, Davies NW & Standen NB (2001). Angiotensin II inhibits rat arterial KATP channels by inhibiting steady‐state protein kinase A activity and activating protein kinase Ca. J Physiol 530, 193–205. [PubMed]
  • Ishida A, Kameshita I, Okuno S, Kitani T & Fujisawa H (1995). Novel highly specific and potent inhibitor of calmodulin‐dependent protein kinase II. Biochem Biophys Res Commun 212, 806–812. [PubMed]
  • Jaggar JH, Porter VA, Lederer WJ & Nelson MT (2000). Calcium sparks in smooth muscle. Am J Physiol Cell Physiol 278, C235–C256. [PubMed]
  • Jaggar JH, Wellman GC, Heppner TJ, Porter VA, Perez GJ, Gollasch M, Kleppisch T, Rubart M, Stevenson AS, Lederer WJ, Knot HJ, Bonev AD & Nelson MT (1998). Ca2+ channels, ryanodine receptors and Ca2+‐activated K+ channels: a functional unit for regulating arterial tone In 3rd Acta Physiologica Scandinavica International Symposium on Signalling Mechanisms in Vascular Smooth Muscle, pp. 577–587. Margretetorp, Sweden. [PubMed]
  • Kamishima T & Quayle JM (2002). Mitochondrial Ca2(+) uptake is important over low Ca2(+) (i) range in arterial smooth muscle. Am J Physiol Heart Circ Physiol 283, H2431–H2439. [PubMed]
  • Kawasaki H, Springett GM, Mochizuki N, Toki S, Nakaya M, Matsuda M, Housman DE & Graybiel AM (1998). A family of cAMP‐binding proteins that directly activate Rap1. Science 282, 2275–2279. [PubMed]
  • Kelley GG, Reks SE, Ondrako JM & Smrcka AV (2001). Phospholipase C epsilon: a novel Ras effector. EMBO J 20, 743–754. [PubMed]
  • Kim I, Je HD, Gallant C, Zhan Q, Van Riper D, Badwey JA, Singer HA & Morgan KG (2000). Ca2+‐calmodulin‐dependent protein kinase II‐dependent activation of contractility in ferret aorta. J Physiol 526, 367–374. [PubMed]
  • Kim M, Han IS, Koh SD & Perrino BA (2006). Roles of CaM kinase II and phospholamban in SNP‐induced relaxation of murine gastric fundus smooth muscles. Am J Physiol Cell Physiol 291, C337–C347. [PubMed]
  • Kreusser MM, Lehmann LH, Keranov S, Hoting MO, Oehl U, Kohlhaas M, Reil JC, Neumann K, Schneider MD, Hill JA, Dobrev D, Maack C, Maier LS, Grone HJ, Katus HA, Olson EN & Backs J (2014). Cardiac CaM kinase II genes delta and gamma contribute to adverse remodeling but redundantly inhibit calcineurin‐induced myocardial hypertrophy. Circulation 130, 1262–1273. [PMC free article] [PubMed]
  • Li WW, Li H, Sanders PN, Mohler PJ, Backs J, Olson EN, Anderson ME & Grumbach IM (2011). The multifunctional Ca2+/calmodulin‐dependent Kinase II delta (CaMKII delta) controls neointima formation after carotid ligation and vascular smooth muscle cell proliferation through cell cycle regulation by p21. J Biol Chem 286, 7990–7999. [PubMed]
  • MacMillan D & McCarron JG (2010). The phospholipase C inhibitor U‐73122 inhibits Ca2+ release from the intracellular sarcoplasmic reticulum Ca2+ store by inhibiting Ca2+ pumps in smooth muscle. Br J Pharmacol 160, 1295–1301. [PubMed]
  • Marganski WA, Gangopadhyay SS, Je HD, Gallant C & Morgan KG (2005). Targeting of a novel Ca2+/calmodulin‐dependent protein kinase II is essential for extracellular signal‐regulated kinase‐mediated signaling in differentiated smooth muscle cells. Circ Res 97, 541–549. [PubMed]
  • Marx SO, Ondrias K & Marks AR (1998). Coupled gating between individual skeletal muscle Ca2+ release channels (ryanodine receptors). Science 281, 818–821. [PubMed]
  • Mauban JRH, Lamont C, Balke CW & Wier WG (2001). Adrenergic stimulation of rat resistance arteries affects Ca2+ sparks, Ca2+ waves, and Ca2+ oscillations. Am J Physiol Heart Circ Physiol 280, H2399–H2405. [PubMed]
  • Mishra S, Gray CBB, Miyamoto S, Bers DM & Brown JH (2011). Location matters clarifying the concept of nuclear and cytosolic CaMKII subtypes. Circ Res 109, 1354–U1104. [PubMed]
  • Moosmang S, Schulla V, Welling A, Feil R, Feil S, Wegener JW, Hofmann F & Klugbauer N (2003). Dominant role of smooth muscle L‐type calcium channel Ca(v)1.2 for blood pressure regulation. EMBO J 22, 6027–6034. [PubMed]
  • Oestreich EA, Malik S, Goonasekera SA, Blaxall BC, Kelley GG, Dirksen RT & Smrcka AV (2009). Epac and phospholipase C epsilon regulate Ca2+ release in the heart by activation of protein kinase C epsilon and calcium‐calmodulin kinase II. J Biol Chem 284, 1514–1522. [PubMed]
  • Pellicena P & Schulman H (2014). CaMKII inhibitors: from research tools to therapeutic agents. Front Pharmacol 5, 21. [PMC free article] [PubMed]
  • Pluger S, Faulhaber J, Furstenau M, Lohn M, Waldschutz R, Gollasch M, Haller H, Luft FC, Ehmke H & Pongs O (2000). Mice with disrupted BK channel beta 1 subunit gene feature abnormal Ca2+ spark/STOC coupling and elevated blood pressure. Circ Res 87, E53–E60. [PubMed]
  • Prasad AM, Nuno DW, Koval OM, Ketsawatsomkron P, Li WW, Li H, Shen FY, Joiner MLA, Kutschke W, Weiss RM, Sigmund CD, Anderson ME, Lamping KG & Grumbach IM (2013). Differential Control of Calcium Homeostasis and Vascular Reactivity by Ca2+/Calmodulin‐Dependent Kinase II. Hypertension 62, 434–441. [PubMed]
  • Poenie M, Alderton J, Steinhardt R & Tsien R (1986). Calcium rises abruptly and briefly throughout the cell at the onset of anaphase. Science 233, 886–889. [PubMed]
  • Porter VA, Bonev AD, Knot HJ, Heppner TJ, Stevenson AS, Kleppisch T, Lederer WJ & Nelson MT (1998). Frequency modulation of Ca2+ sparks is involved in regulation of arterial diameter by cyclic nucleotides. Am J Physiol Cell Physiol 274, C1346–C1355. [PubMed]
  • Pucovsky V, Gordienko DV & Bolton TB (2002). Effect of nitric oxide donors and noradrenaline on Ca2+ release sites and global intracellular Ca2+ in myocytes from guinea‐pig small mesenteric arteries. J Physiol 539, 25–39. [PubMed]
  • Roberts OL & Dart C (2014). cAMP signalling in the vasculature: the role of Epac (exchange protein directly activated by cAMP). Biochem Soc Trans 42, 89–97. [PubMed]
  • Roberts OL, Kamishima T, Barrett‐Jolley R, Quayle JM & Dart C (2013). Exchange protein activated by cAMP (Epac) induces vascular relaxation by activating Ca2+‐sensitive K+ channels in rat mesenteric artery. J Physiol 591, 5107–5123. [PubMed]
  • Rokolya A & Singer HA (2000). Inhibition of CaM kinase II activation and force maintenance by KN‐93 in arterial smooth muscle. Am J Physiol Cell Physiol 278, C537–C545. [PubMed]
  • Roscioni SS, Maarsingh H, Elzinga CRS, Schuur J, Menzen M, Halayko AJ, Meurs H & Schmidt M (2011). Epac as a novel effector of airway smooth muscle relaxation. J Cell Mol Med 15, 1551–1563. [PubMed]
  • Sampson LJ, Leyland ML & Dart C (2003). Direct interaction between the actin‐binding protein filamin‐A and the inwardly rectifying potassium channel, Kir2.1. J Biol Chem 278, 41988–41997. [PubMed]
  • Senti M, Fernandez‐Fernandez JM, Tomas M, Vasquez E, Elosua R, Marrugat J & Valverde MA (2005). Protective effect of the KCNMB1 E65K genetic polymorphism against diastolic hypertension in aging women and its relevance to cardiovascular risk. Circ Res 97, 1360–1365. [PubMed]
  • Singer HA (2012). Ca2+/calmodulin‐dependent protein kinase II function in vascular remodelling. J Physiol 590, 1349–1356. [PubMed]
  • Soderling TR & Stull JT (2001). Structure and regulation of calcium/calmodulin‐dependent protein kinases. Chem Rev 101, 2341–2351. [PubMed]
  • Stott JB, Barrese V & Greenwood IA (2016). Kv7 channel activation underpins EPAC‐dependent relaxations of rat arteries. Arterioscler Thromb Vasc Biol 36, 2404–2411. [PubMed]
  • Sukhanova IF, Kozhevnikova LM, Popov EG, Podmareva ON & Avdonin PV (2006). Activators of Epac proteins induce relaxation of isolated rat aorta. Dokl Biol Sci 411, 441–444. [PubMed]
  • Toussaint F, Charbel C, Allen BG & Ledoux J (2016). Vascular CaMKII: heart and brain in your arteries. Am J Physiol Cell Physiol 311, C462–C478. [PubMed]
  • Van Petegem F (2012). Ryanodine receptors: structure and function. J Biol Chem 287, 31624–31632. [PubMed]
  • Van Triest M, De Rooij J & Bos JL (2001). Measurement of GTP‐bound Ras‐like GTPases by activation‐specific probes. Method Enzymol 333, 343–348. [PubMed]
  • van Welie I & du Lac S (2011). Bidirectional control of BK channel open probability by CAMKII and PKC in medial vestibular nucleus neurons. J Neurophysiol 105, 1651–1659. [PubMed]
  • Wehrens XHT, Lehnart SE, Reiken SR & Marks AR (2004). Ca2+/calmodulin‐dependent protein kinase II phosphorylation regulates the cardiac ryanodine receptor. Circ Res 94, E61–E70. [PubMed]
  • ZhuGe RH, Tuft RA, Fogarty KE, Bellve K, Fay FS & Walsh JV (1999). The influence of sarcoplasmic reticulum Ca2+ concentration on Ca2+ sparks and spontaneous transient outward currents in single smooth muscle cells. J Gen Physiol 113, 215–228. [PubMed]
  • Zieba BJ, Artamonov MV, Jin L, Momotani K, Ho R, Franke AS, Neppl RL, Stevenson AS, Khromov AS, Chrzanowska‐Wodnicka M & Somlyo AV (2011). The cAMP‐responsive Rap1 guanine nucleotide exchange factor, Epac, induces smooth muscle relaxation by down‐regulation of RhoA activity. J Biol Chem 286, 16681–16692. [PubMed]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society