Angiotensin II activates Ca_V1.2 Ca²⁺ channels through β‐arrestin2 and casein kinase 2 in mouse immature cardiomyocytes

Abstract

Key points

• Angiotensin II (AngII) is crucial in cardiovascular regulation in perinatal mammalians.

• Here we show that AngII increases twitch Ca²⁺ transients of mouse immature but not mature cardiomyocytes by robustly activating Ca_V1.2 L‐type Ca²⁺ channels through a novel signalling pathway involving angiotensin type 1 (AT₁) receptors, β‐arrestin2 and casein kinase 2.

• A β‐arrestin‐biased AT₁ receptor agonist, TRV027, was as effective as AngII in activating L‐type Ca²⁺ channels.

• Our results help understand the molecular mechanism by which AngII regulates the perinatal circulation and also suggest that β‐arrestin‐biased AT₁ receptor agonists may be valuable therapeutics for paediatric heart failure.

Abstract

Angiotensin II (AngII), the main effector peptide of the renin–angiotensin system, plays important roles in cardiovascular regulation in the perinatal period. Despite the well‐known stimulatory effect of AngII on vascular contraction, little is known about regulation of contraction of the immature heart by AngII. Here we found that AngII significantly increased the peak amplitude of twitch Ca²⁺ transients by robustly activating L‐type Ca_V1.2 Ca²⁺ (Ca_V1.2) channels in mouse immature but not mature cardiomyocytes. This response to AngII was mediated by AT₁ receptors and β‐arrestin2. A β‐arrestin‐biased AT₁ receptor agonist was as effective as AngII in activating Ca_V1.2 channels. Src‐family tyrosine kinases (SFKs) and casein kinase 2α’β (CK2α’β) were sequentially activated when AngII activated Ca_V1.2 channels. A cyclin‐dependent kinase inhibitor, p27^Kip1 (p27), inhibited CK2α’β, and AngII removed this inhibitory effect through phosphorylating tyrosine 88 of p27 via SFKs in cardiomyocytes. In a human embryonic kidney cell line, tsA201 cells, overexpression of CK2α’β but not c‐Src directly activated recombinant Ca_V1.2 channels composed of C‐terminally truncated α_1C, the distal C‐terminus of α_1C, β_2C and α₂δ₁ subunits, by phosphorylating threonine 1704 located at the interface between the proximal and the distal C‐terminus of Ca_V1.2α_1C subunits. Co‐immunoprecipitation revealed that Ca_V1.2 channels, CK2α’β and p27 formed a macromolecular complex. Therefore, stimulation of AT₁ receptors by AngII activates Ca_V1.2 channels through β‐arrestin2 and CK2α’β, thereby probably exerting a positive inotropic effect in the immature heart. Our results also indicated that β‐arrestin‐biased AT₁ receptor agonists may be used as valuable therapeutics for paediatric heart failure in the future.

Key points

• Angiotensin II (AngII) is crucial in cardiovascular regulation in perinatal mammalians.

• Here we show that AngII increases twitch Ca²⁺ transients of mouse immature but not mature cardiomyocytes by robustly activating Ca_V1.2 L‐type Ca²⁺ channels through a novel signalling pathway involving angiotensin type 1 (AT₁) receptors, β‐arrestin2 and casein kinase 2.

• A β‐arrestin‐biased AT₁ receptor agonist, TRV027, was as effective as AngII in activating L‐type Ca²⁺ channels.

• Our results help understand the molecular mechanism by which AngII regulates the perinatal circulation and also suggest that β‐arrestin‐biased AT₁ receptor agonists may be valuable therapeutics for paediatric heart failure.

Abbreviations

ACM

atrial cardiomyocyte

AKAP

A‐kinase anchoring protein

AngII

angiotensin II

AT₁

angiotensin type 1 receptor

AT₂

angiotensin type 2 receptor

AVCM

adult ventricular cardiomyocyte

Ca_V1.2 channel

L‐type Ca_V1.2 Ca²⁺ channel

Ca_V1.2α_1CΔ1821

Ca_V1.2α_1c subunit devoid of its distal C‐terminus_1822–2171

CDK

cyclin‐dependent kinase

CK2

casein kinase 2

CSK

C‐terminal Src kinase

DCRD

the distal C‐terminus regulatory domain

DCT

the distal C‐terminus_1822–2171

DMEM

Dulbecco's modified Eagle's medium

E

embryonic day

E_rev

reversal potential

FL

full‐length

GFP

green fluorescent protein

HA

hemagglutinin

IB

immunoblotting

IP

immunoprecipitation

NVCM

neonatal ventricular cardiomyocyte

PCRD

the proximal C‐terminal regulatory domain

PKA

protein kinase A

PKC

protein kinase C

P

postnatal day

p27

p27^Kip1

RAS

renin–angiotensin system

RFP

red fluorescent protein

SFK

Src‐family tyrosine kinase

shRNA

small hairpin RNA

Introduction

The renin–angiotensin system (RAS) is an important regulator of the cardiovascular system and controls sodium balance, body fluid volumes and arterial pressure (Hall, 2003; Paul et al. 2006; Karnik et al. 2015). Classically, RAS was viewed as an endocrinological system (Hall, 2003). The enzyme renin secreted from juxtaglomerular cells in the kidney digests angiotensinogen produced by the liver to a decapeptide, angiotensin I, in the plasma. Angiotensin I is then converted to an octapeptide, angiotensin II (AngII), by the angiotensin‐converting enzyme in the pulmonary circulation. AngII is a major player in this system and activates two specific G‐protein‐coupled receptors, angiotensin type 1 (AT₁) and angiotensin type 2 (AT₂) receptors (Hunyady & Catt, 2006; Karnik et al. 2015). AT₁ receptors mediate most of the known effects of AngII, whereas the functional role of AT₂ receptors is still under extensive investigation. AT₁ receptors activate various heterotrimeric G proteins (G_q/11, G_12/13 and G_i) and β‐arrestins. Although β‐arrestins were originally discovered to desensitize activated G‐protein‐coupled receptors, it is now well established that β‐arrestins mediate G‐protein‐independent signals on their own right (Shukla et al. 2011). In addition to the plasma RAS, many tissues, including the heart, possess many components of RAS and produce AngII in situ to regulate their own functions (local RAS) (Paul et al. 2006; Bader, 2010). RAS also plays significant roles in cardiovascular diseases such as hypertension, atherosclerosis, and cardiac hypertrophy and failure as evidenced by the fact that inhibitors of the angiotensin‐converting enzyme and AT₁ receptors significantly improve the outcome of these diseases (Jong et al. 2003; Pfeffer et al. 2003).

In fetal and postnatal development, the plasma and local RAS are extremely important (Heymann et al. 1981; Paul et al. 2006). In general, all components of RAS begin to be expressed early in fetal development, and their amounts culminate at the perinatal period. Indeed, the plasma renin concentration and activity are higher in newborns than adults in various animals, including humans (Wallace et al. 1980; Varga et al. 1981). At the time of birth, neonates experience major haemodynamic changes such as rapid increase in pulmonary blood flow, removal of placental circulation, closure of several sites of shunting within the heart and great vessels, and changes in cardiac output and its distribution (Hooper et al. 2015). Left ventricular output increases as much as ∼3 times at birth, and the systemic blood pressure increases immediately after birth and continues to rise for several weeks, concomitant with an increase in systemic vascular resistance. RAS together with the sympathetic nervous system is thought to play a critical role in these changes (Heymann et al. 1981). Furthermore, infants born vaginally have higher circulating concentrations of AngII than those delivered by caesarean section (Lumbers & Reid, 1977), and a minor degree of haemorrhage results in a marked increase in plasma AngII concentration (Broughton Pipkin et al. 1974), indicating that RAS is critical for newborns’ response to haemodynamic insults. Indeed, mice deficient of angiotensinogen or AT₁ receptors exhibit ∼50% perinatal lethargy (Kim et al. 1995; Oliverio et al. 1998).

Although it is well established that AngII causes vascular contraction, little is known about the effect of AngII on contraction of the immature heart. Herein, we have highlighted that AngII markedly increases twitch Ca²⁺ transients of immature but not mature cardiomyocytes by robustly increasing L‐type Ca_V1.2 Ca²⁺ (Ca_V1.2) channel currents through a novel pathway composed of AT₁ receptors, β‐arrestin2 and casein kinase 2 (CK2). Thus, β‐arrestin‐biased AT₁ receptor agonists, which have not only a positive inotropic effect but also an anti‐apoptotic effect on cardiomyocytes, may be used as valuable therapeutics for paediatric heart failure in the future (Kim et al. 2012).

Methods

Ethical approval

All animals used in this study received humane care in compliance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health. All experimental procedures were carried out in accordance with the Guidelines for Animal Experimentation of Shinshu University and approved by the Committee for Animal Experimentation (Approval numbers 250062 and 280035). All mice were given free access to water and standard diet throughout the study, and maintained in a temperature‐ (21–26°C) and humidity‐ (50–60%) controlled room with a 12 h light/dark cycle. All experimental mice were deeply anaesthetized with 0.3 mg kg^–1 medetomidine (Domitor, Nippon Zenyaku Kogyo Co., Fukushima, Japan), 4.0 mg kg^–1 midazolam (Midazolam Sandoz, Novartis, Tokyo, Japan) and 5.0 mg kg^–1 butorphanol (Vetorphale, Meiji Seika Pharma Co., Tokyo, Japan), i.p., or isoflurane (Escain, Mylan Seiyaku, Osaka, Japan) inhalation (mouse pups). We conform to the principles and standards for reporting animal experiments in The Journal of Physiology and Experimental Physiology (Grundy, 2015). All C57BL/6 mice were purchased from Japan SLC Inc.

Isolation of cardiomyocytes

Cardiomyocytes from male C57BL/6 mice at postnatal days (P) 10, P28–34 and P56–62 were enzymatically isolated as previously described (Horiuchi‐Hirose et al. 2011). Briefly, the hearts were removed from anaesthetized mice, mounted on the Langendorff apparatus with modified Tyrode solution containing (mm) 136.5 NaCl (Wako Pure Chemical Industries, Osaka, Japan), 5.4 KCl (Wako), 1.8 CaCl₂ (Wako), 0.53 MgCl₂ (Wako), 5.5 Hepes (Dojindo, Kumamoto, Japan) and 5.5 glucose (Wako) (pH 7.4 with NaOH), and digested with Ca²⁺‐free Tyrode solution containing 0.80 mg ml^–1 collagenase (Nitta Gelatin Inc., Osaka, Japan), 0.06 mg ml^–1 protease (Sigma‐Aldrich, Tokyo, Japan), 1.20 mg ml^–1 hyaluronidase (Sigma‐Aldrich), 0.03 mg ml^–1 DNase I (Roche, Tokyo, Japan) and 0.50 mg ml^–1 BSA (Wako) at 37°C for 2 min. Isolated cardiomyocytes were suspended in Ca²⁺‐free Tyrode solution containing 1 mg ml^–1 BSA at room temperature, and the Ca²⁺ concentration was gradually increased to 1.8 mm. After that, cardiomyocytes were treated with vehicle (deionized water) or AngII (Sigma‐Aldrich).

Cell cultures and preparations

Anaesthetized neonatal (P0 or P1) C57BL/6 mice were decapitated and hearts were then removed aseptically. Primary cultures of cardiomyocytes were prepared from the hearts by using the Neonatal Cardiomyocyte Isolation System (Worthington Biochemical Corp., Lakewood, NJ, USA). To enrich cardiomyocytes, the collected cells were plated onto plastic culture dishes in low‐glucose Dulbecco's modified Eagle's medium (DMEM) plus GlutaMAX (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum, 100 units ml^–1 penicillin (Thermo Fisher Scientific) and 100 μg ml^–1 streptomycin (Thermo Fisher Scientific) at 37°C and 5% CO₂ for 60 min and then plated onto gelatin‐ (Difco, Becton, Dickinson Company, Franklin Lakes, NJ) and fibronectin‐ (Sigma‐Aldrich) coated glass bottom dishes or coverslips. After culture in serum‐free medium for 24 h, the cardiomyocytes were treated with vehicle (deionized water) or AngII. We confirmed that these cardiomyocytes soon after isolation and those cultured for a few days exhibited almost the same degree of response of L‐type Ca²⁺ channels to AngII.

HL‐1 cardiomyocytes were kindly provided by Dr William C. Claycomb (Louisiana State University Health Science Centre, New Orleans, LA, USA) and were cultured in Claycomb Medium (Sigma‐Aldrich) supplemented with 10% fetal bovine serum (Sigma‐Aldrich), 100 units ml^–1 penicillin, 100 μg ml^–1 streptomycin, 2 mm l‐glutamine (Sigma‐Aldrich) and 100 μm noradrenaline (Sigma‐Aldrich) at 37°C and 5% CO₂ on gelatin/fibronectin‐coated flasks (Claycomb et al. 1998). HL‐1 cardiomyocytes spontaneously beating were passaged when they reached 100% confluency. For electrophysiological experiments, HL‐1 cardiomyocytes were plated onto gelatin/fibronectin‐coated coverslips at low density. After culture in noradrenaline‐ and serum‐free Claycomb medium for 24 h, HL‐1 cardiomyocytes were treated with vehicle (deionized water) or AngII.

For introduction of small hairpin (sh) RNAs with red fluorescent protein (RFP) or cDNA of C‐terminal Src kinase (CSK) with green fluorescent protein (GFP) in HL‐1 cardiomyocytes, lentiviral stocks were produced in HEK293T cells (American Type Culture Collection, Manassas, VA, USA) by cotransfection of the packaging plasmids pMD.G (VSVG envelope), pMDLg/pRRE (gag‐pol plasmid) and pRSV‐Rev (rev plasmid) with corresponding lentiviral constructs using polyethylenimine (Polysciences, Inc., Warrington, PA, USA) and concentrated by polybrene (Sigma‐Aldrich) (Kojima et al. 2014). Whole‐cell patch‐clamp recordings were performed with RFP‐ or GFP‐positive cells.

TsA201 human embryonic kidney cells (European Collection of Authenticated Cell, Salisbury, UK) were cultured in low‐glucose DMEM plus GlutaMAX, 10% fetal bovine serum, 100 units ml^–1 penicillin and 100 μg ml^–1 streptomycin at 37°C and 5% CO₂. TsA201 cells were grown to ∼70% confluency in plastic dishes and transiently transfected with an equimolar ratio of cDNA encoding Ca_V1.2 α_1c, β_2c and α₂δ₁ subunits (1.0, 0.7, and 0.8 μg of plasmid DNA per 35 mm dish, respectively) and other proteins by the calcium phosphate method. For electrophysiological experiments, a 10‐fold lower ratio of cDNA encoding enhanced GFP (EGFP) was added to each transfection mixture as an indicator of successful transfection, and 30 h after transfection, cells were plated onto collagen‐ (Nitta gelatin) coated coverslips at low density. Experiments were performed with 48–56 h post‐transfection cells.

Plasmid construction

The cDNA encoding rabbit α_1C was generously given by Dr William A. Catterall (University of Washington, Seattle, WA, USA). Ca_V1.2α_1CΔ1821 [Ca_V1.2α_1c subunit devoid of its distal C‐terminus_1822–2171 (DCT)] was generated by PCR using antisense primers containing a stop codon and SalI restriction site. All primer sequences used in this study are listed in Table 1. The PCR products were subcloned into a blunt‐ended pBlueScript SK‐ vector and again subcloned into the EcoRV/SalI site of the Ca_V1.2α_1C subunit. For immunoblotting and immunoprecipitation assays, hemagglutinin (HA)‐tagged Ca_V1.2α_1C and HA‐Ca_V1.2α_1CΔ1821 were used. An HA‐epitope was inserted into the extracellular loop between S5 and S6 in domain II as previously described (Nakada et al. 2012). To generate HA‐Ca_V1.2α_1CDCT and 3xFLAG‐tagged Ca_V1.2α_1CDCT, the cDNA encoding amino acids 1822–2171 of Ca_V1.2α_1C was amplified by PCR and then subcloned into pCMV‐HA (Clontech Laboratories, Inc., Mountain View, CA, USA) or p3xFLAG‐CMV‐10 (Sigma‐Aldrich).

Table 1.

List of oligonucleotides used in this study

Construct	Backbone	Direction	Sequence (5’→3’)	Note
CaV1.2α1CΔ1821	pcDNA3	S	AATCCAGCTGAACATACCCAATGCT	EcoRV/SalI*1
		AS	AGGTCGACCTACACAGTGCTGACTGTGCTG
CaV1.2α1CΔ1821T1704A	pcDNA3	S	TCTCCGGAGACCTGGCAGCTGAGGAAGAG	Mutagenesis*2
		AS	CTCTTCCTCAGCTGCCAGGTCTCCGGAGA
FLAG‐CaV1.2α1CDCT	p3xFLAG‐CMV‐10	S	CGAATTCCGTGGAGGGCCACGGGTCCCCC	EcoRV/BamHI*1
		AS	GGATCCCGCTACAGGCTGCTGACGCCGGC
HA‐CaV1.2α1CDCT	pCMV‐HA	S	CAGTCGACTGTGGAGGGCCACGGGTCCCCC	SalI/NotI*1
		AS	GGCGGCCGCCTACAGGCTGCTGACGCCGGC
CaVβ2c	pcDNA3.1	S	CACTGAGAACCTCCGATCGT
		AS	GCTTGTACTAGGCCAATTTCTGT
CaVα2δ1	pcDNA3.1	S	TGATCTTCGATCGCGAAGATGG
		AS	AGGGCATGGAATTAAGTTGCAGA
Myc‐c‐Src	pCMV‐Myc	S	TAGCCCGGGCGGATCCAGCCAGGACCATGGGCAG	In fusion*3
		AS	GCTTGATATCGAATTCTAGGTTCTCCCCGGGCTG
CK2α	pcDNA3.1	S	AAGTTCAGCTTTGTCTGTCAAC
		AS	CTGCTCAGGCATCAGAAGG
CK2α′	pcDNA3.1	S	CGTCCCTCCGGCAGCCGAA
		AS	ACGTGATCTCGCTACACGCGTTAAGAC
CK2β	pcDNA3.1	S	TTTTGCGCTGTAGTGGTCTC
		AS	CCATAGACTTCCTGAAAGGGTG
FLAG‐CK2α′	p3xFLAG‐CMV‐10	S	CGGTACCGCCCGGCCCGGCCGCGGGCAGTC	KpnI/BamHI*1
		AS	TGGATCCACAGACCGTCGATTCCCCAGGCT
FLAG‐CK2β	p3xFLAG‐CMV‐10	S	CGGTACCGAGTAGCTCTGAGGAGGTGTCCT	KpnI/BamHI*1
		AS	GGGATCCCAGAGGGAGAGGTGGGTGGGCAA
Myc‐p27 WT	pCMV‐Myc	S	CAGAATTCTGTCAAACGTGAGAGTGTCTAA	EcoRI/XhoI*1
		AS	AACTCGAGTTACGTCTGGCGTCGAAGGCCG
Myc‐p27 Y88F	pCMV‐Myc	S	TTGCCCGAGTTCTTCTACAGGCCCCCG	Mutagenesis*2
		AS	CGGGGGCCTGTAGAAGAACTCGGGCAA
Myc‐p27 Y88E	pCMV‐Myc	S	CTTGCCCGAGTTCGAGTACAGGCCCCCGC	Mutagenesis*2
		AS	GCGGGGGCCTGTACTCGAACTCGGGCAAG
sh‐Gq#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGGCCCTCTTGTAGTTTCTTTATGTTAATATTCATAGCATAAAGAAGCTACAAGAGGGCTTTTTTG	shRNA*4
		AS	CGAACAAAAAAGCCCTCTTGTAGCTTCTTTATGCTATGAATATTAACATAAAGAAACTACAAGAGGGCC
sh‐Gq#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCCTTCCTATCTGTCTACATAAGTTAATATTCATAGCTTGTGTAGGCAGATAGGAAGGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACCTTCCTATCTGCCTACACAAGCTATGAATATTAACTTATGTAGACAGATAGGAAGGC
sh‐G11#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	GCACCGGGCTCAAGATCCTCTATAAGTAGTTAATATTCATAGCTACTTGTAGAGGATCTTGAGCTTTTT	shRNA*4
		AS	CGAACAAAAAAGCTCAAGATCCTCTACAAGTAGCTATGAATATTAACTACTTATAGAGGATCTTGAGCC
sh‐G11#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCCTCTACAAGTATGAGTAGAAGTTAATATTCATAGCTTCTGCTCATACTTGTAGAGGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACCTCTACAAGTATGAGCAGAAGCTATGAATATTAACTTCTACTCATACTTGTAGAGGC
sh‐β‐arrestin1#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCCTTGAGGTATCATTGGATAAGTTAATATTCATAGCTTATCCAGTGATGCCTCAAGGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACCTTGAGGCATCACTGGATAAGCTATGAATATTAACTTATCCAATGATACCTCAAGGC
sh‐β‐arrestin1#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCCAATGATGACGATATTGTATGTTAATATTCATAGCATACAATGTCGTCATCATTGGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACCAATGATGACGACATTGTATGCTATGAATATTAACATACAATATCGTCATCATTGGC
sh‐β‐arrestin2#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCGGCTTATCATTAGAAAGGTAGTTAATATTCATAGCTACCTTTCTGATGATAAGCCGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACGGCTTATCATCAGAAAGGTAGCTATGAATATTAACTACCTTTCTAATGATAAGCCGC
sh‐β‐arrestin2#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGGACTATTTGAAGGACTGGAAAGTTAATATTCATAGCTTTCCGGTCCTTCAAGTAGTCTTTTTTG	shRNA*4
		AS	CGAACAAAAAAGACTACTTGAAGGACCGGAAAGCTATGAATATTAACTTTCCAGTCCTTCAAATAGTCC
sh‐CK2α#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGGCCAATATGATGTTAGGGATTGTTAATATTCATAGCAATCCCTGACATCATATTGGCTTTTTTG	shRNA*4
		AS	CGAACAAAAAAGCCAATATGATGTCAGGGATTGCTATGAATATTAACAATCCCTAACATCATATTGGCC
sh‐CK2α#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCCGAGTTGCTTCTTGATATTTGTTAATATTCATAGCAAATATCGAGAAGCAACTCGGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACCGAGTTGCTTCTCGATATTTGCTATGAATATTAACAAATATCAAGAAGCAACTCGGC
sh‐CK2α′#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCGTGGTGGAACAAATATTATTGTTAATATTCATAGCAATGATATTTGTTCCACCACGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACGTGGTGGAACAAATATCATTGCTATGAATATTAACAATAATATTTGTTCCACCACGC
sh‐CK2α′#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGAGACCTAGATCCACATTTCAAGTTAATATTCATAGCTTGAAGTGTGGATCTAGGTCTTTTTTTG	shRNA*4
		AS	CGAACAAAAAAAGACCTAGATCCACACTTCAAGCTATGAATATTAACTTGAAATGTGGATCTAGGTCTC
sh‐CK2β#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGTGGTTTCCCTTACATGTTCTTGTTAATATTCATAGCAAGAGCATGTGAGGGAAACCATTTTTTG	shRNA*4
		AS	CGAACAAAAAATGGTTTCCCTCACATGCTCTTGCTATGAATATTAACAAGAACATGTAAGGGAAACCAC
sh‐CK2β#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCACATGCTCTTTATGGTGTATGTTAATATTCATAGCATGCACCATGAAGAGCATGTGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACACATGCTCTTCATGGTGCATGCTATGAATATTAACATACACCATAAAGAGCATGTGC
sh‐p27#1	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCGCAAGTGGAGTTTCGGCTTTGTTAATATTCATAGCAAAGTCGAAATTCCACTTGCGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACGCAAGTGGAATTTCGACTTTGCTATGAATATTAACAAAGCCGAAACTCCACTTGCGC
sh‐p27#2	pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro	S	ACCGGCCGGTATTTGGTGGACTAAATGTTAATATTCATAGCATTTGGTCCACCAAATGCCGGTTTTTTG	shRNA*4
		AS	CGAACAAAAAACCGGCATTTGGTGGACCAAATGCTATGAATATTAACATTTAGTCCACCAAATACCGGC
CSK (lentivirus)	CSII‐EF‐MCS‐IRES‐hrGFP	S	GCAGAAGATGTCGGCAATACAG
		AS	GCCGGCTGGGGCTCAGTTCAA

^{* 1}PCR products were subcloned into the expression vectors after digestion with the restriction enzymes indicated. ^{* 2}Primers used for the quickchange site‐directed mutagenesis system. ^{* 3}PCR products were subcloned into the expression vectors by an in‐fusion cloning system. ^{* 4}Sense and antisense oligonucleotides were annealed and subcloned into the shRNA lentivector linearized with BbsI.

The cDNAs of Ca_Vβ₂c and Ca_Vα₂δ₁ subunits were isolated with RT‐PCR from a rat heart cDNA library. The cDNAs of CK2α, CK2α’, CK2β, p27^Kip1 (p27), c‐Src and CSK were isolated with RT‐PCR from a mouse heart cDNA library. These cDNAs were subcloned into pcDNA3.1, pCMV‐HA, pCMV‐Myc (Clontech) or p3xFLAG‐CMV‐10. Single amino acid substitution mutants [Ca_V1.2α_1CΔ1821T1704A (Thr¹⁷⁰⁴ of α_1C was substituted with alanine), Myc‐p27Y88F (Tyr⁸⁸ of p27 was substituted with phenylalanine, Y88F), and Myc‐p27Y88E (Tyr⁸⁸ of p27 was substituted with glutamic acid)] were generated with the QuickChange Site‐Directed Mutagenesis Kit (Stratagene, Agilent Technologies, Tokyo, Japan) according to the manufacturer's instructions.

The cDNAs of CSK were isolated with RT‐PCR from a mouse heart cDNA library and then subcloned into lentiviral transfer plasmid CSII‐EF‐MCS‐IRES‐hrGFP [kindly provided by Dr Hiroyuki Miyoshi (RIKEN BRC, Tsukuba, Japan)]. To prepare the shRNA expression vectors for p27, G_q, G₁₁, β‐arrestin1, β‐arrestin2, CK2α, CK2α’ and CK2β, the double stranded DNA oligo encoding a sense–loop–antisense sequence to the targeted genes was inserted into lentiviral shRNA expression vector pRSI9‐U6‐(sh)‐UbiC‐TagRFP‐2A‐Puro (Addgene, Cambridge, MA, USA). All DNA oligo sequences used in this study are listed in Table 1.

Antibodies

Antibodies against the following proteins were used: Ca_V1.2α_1c (1:2000, ACC‐003, Alomone Labs, Jerusalem, Israel), α‐tubulin (1:3000, T5168, Sigma‐Aldrich), G_q/11 (1:500, sc‐392, Santa Cruz Biotechnology, Dallas, TX, USA), β‐arrestin1 (1:1000, 610550, BD Transduction Laboratories, Franklin Lakes, NJ, USA), β‐arrestin2 (1:2000, ab167047, Abcam, Cambridge, UK), CSK (1:1000, H00001445‐M01, Abnova, Taipei, Taiwan), CK2β (1:1000, MAB0817, Abnova), HA (1:10 000, M180‐3, MBL, Nagoya, Japan), Myc (1:10 000, M192‐3, MBL), control mouse IgG2b (M077‐3, MBL), FLAG (1:5000, F1804, Sigma‐Aldrich), p27 (1:5000, 610241, BD Transduction), p27‐pTyr⁸⁸ (1:2000, AP20721b, Abgent, San Diego, CA, USA), mouse or rabbit IgG (1:30 000, 715‐036‐150 or 111‐035‐006, Jackson ImmunoResearch Laboratories, West Grove, PA, USA), and mouse IgG TrueBlot Ultra (1:2000, 18‐8817‐31, Rockland, Limerick, PA, USA). Antibodies against the CK2α and CK2α’ subunits were kindly provided by Dr David W. Litchfield (University of Western Ontario, Ontario, Canada).

Ca²⁺ imaging

The imaging of twitch Ca²⁺ transients was performed as previously described (Nakada et al. 2012). Briefly, cardiomyocytes in glass bottom dishes were incubated with 6 μm Fluo 4/AM (Dojindo) plus 0.01% Cremophore EL (Sigma‐Aldrich) and 0.02% BSA in serum‐free DMEM for 45 min at 37°C and 5% CO₂ followed by de‐esterification. Cardiomyocytes were superfused with modified Tyrode solution at room temperature and paced with 1 ms pulses of 50 V at 0.3 Hz across the 20 mm incubation chamber. Fluorescence images were acquired with an LSM 7 LIVE laser‐scanning microscope with a 20×/0.8 plan apochromatic objective (Carl Zeiss, Jena, Germany). Each image was taken with 128 pixels × 128 pixels every 2.8 ms. The time course of Ca²⁺ transients was assessed from the fluorescence change in individual cardiomyocytes selected by the ROI tool. An increment in fluorescence intensity from baseline was measured and normalized to baseline intensity (ΔF/F ₀).

Electrophysiology

Ionic and gating currents of Ca_V1.2 channels were recorded in the whole‐cell configuration of the patch‐clamp technique at 35–36°C with a patch‐clamp amplifier (Axopatch 200B, Molecular Devices, Sunnyvale, CA) and sampled at 10 kHz after low‐pass filtering at 5 kHz (Horiuchi‐Hirose et al. 2011). Patch pipettes (2–3 MΩ) were fabricated from borosilicate glass capillaries (Kimax‐51, Kimble Glass, Vineland, NJ, USA) and coated with Sylgard 184 (Dow Corning Toray Co., Tokyo, Japan). Series resistance was always kept less than 7 MΩ and routinely compensated for with the amplifier by ∼75%. Ca_V1.2 channel currents of cardiomyocytes were measured with the intracellular solution containing (mm): 20 TEA‐Cl (Tokyo Chemical Industry, Tokyo, Japan), 100 CsCl, 10 EGTA (Dojindo), 10 Hepes, 1 MgCl₂, 5 MgATP and 0.5 MgGTP [pH 7.2 with CsOH (Sigma‐Aldrich)]. The extracellular bath solution contained (mm): 150 TEA‐Cl, 5 CsCl, 5 BaCl₂ (Wako), 5 Hepes, 1.2 MgCl₂ and 5.5 glucose (pH 7.4 with CsOH). The membrane potential was stepped from the holding potential –80 mV to –45 mV for 1000 ms (to inactivate voltage‐gated Na⁺ channels and T‐type Ca²⁺ channels) and then for 300 ms to potentials between –40 and +70 mV with a 10 mV increment every 5 s. Coupling efficiency for recombinant Ca_V1.2 channels in tsA201 cells was assessed with the intracellular solution containing (mm): 135 CsCl, 10 EGTA, 1 MgCl₂, 4 MgATP (pH 7.3 with CsOH). The extracellular bath solution contained (mm): 150 Tris (Wako), 10 BaCl₂, 1 MgCl₂ and 10 glucose (pH 7.4 with HCl) (Hulme et al. 2006). Gating charge movement was measured by applying a series of test pulses from the holding potential of –80 mV to potentials between +60 mV and +80 mV with a 2 mV increment every 5 s and integrating the ON gating charge for 2 ms from the beginning at the reversal potential of Ca_V1.2 channels. Ba²⁺ tail currents were recorded after repolarization to –50 mV from each test pulse.

Immunoblotting and immunoprecipitation

Cells were lysed with ice‐cold lysis buffer composed of 10 mm Tris pH 7.5, 150 mm NaCl, 5 mm EDTA (Dojindo), 1% Triton‐X (Sigma‐Aldrich) and 10% glycerol (Wako) containing a protease inhibitor cocktail (Nacalai tesque, Kyoto, Japan). For analysis of Tyr⁸⁸ phosphorylation of p27, the lysis buffer with the phosphatase inhibitor cocktail (Nacalai tesque) and without EDTA was used. After sonication for 5 s and gentle rocking at 4°C for 30 min, lysates were centrifuged at 10 000 g at 4°C for 30 min in order to remove insoluble materials. In experiments shown in Fig. 7 C, lysates were ultracentrifuged at 100 000 g at 4°C for 1 h to further remove insoluble materials. The protein concentrations of samples were measured by a BCA protein assay kit (Thermo Fisher Scientific). For immunoprecipitation, lysates (250 μg) were incubated with antibodies (2 μg) covalently linked to Protein A‐Sepharose or Protein G‐Sepharose (17‐6002‐35, GE Healthcare Japan, Tokyo, Japan) at 4°C for 2 h, and the immunoprecipitates were then extensively washed with lysis buffer.

Figure 7. AngII removes the inhibitory effect of p27 on CK2α’β through SFKs and activates Ca_V1.2 channels.

A, effect of sh‐p27#2 on peak Ca_V1.2 channel current density at 0 mV in HL‐1 cardiomyocytes in the presence of bosutinib (2 μm), quinalizarin (3 μm) or AngII (3 μm, 2 h). ^* P < 0.01 versus basal of mock. B and C, p27 bound to CK2α’β with (C) or without (B) Ca_V1.2Δ1821 channels in tsA201 cells. B, Myc‐p27, FLAG‐CK2α’ and FLAG‐CK2β were expressed in tsA201 cells. C, HA‐Ca_V1.2Δ1821 channels, Myc‐p27, FLAG‐CK2α’ and FLAG‐CK2β were expressed in tsA201 cells. D, AngII (3 μm, 2 h) did not alter the expression level of p27; n = 3 independent blots. ^* P < 0.05 versus AngII 0 h group. E, time‐dependent effect of AngII (3 μm) on Tyr⁸⁸ phosphorylation of p27 (p27‐pY88) in HL‐1 cardiomyocytes; n = 3 independent blots. ^* P < 0.01 versus AngII 0 min group. F, effect of bosutinib on AngII‐ (3 μm, 2 h) dependent Tyr⁸⁸ phosphorylation in HL‐1 cardiomyocytes. HL‐1 cardiomyocytes were treated with vehicle or AngII (3 μm) for 2 h in the presence or absence of bosutinib (2 μm) before extraction; n = 4 independent blots. ^* P < 0.01 versus control in the absence of bosutinib. G, coupling efficiency of Ca_V1.2Δ1821 + DCT channels in tsA201 cells overexpressing indicated proteins. ^* P < 0.01 versus Ca_V1.2Δ1821 + DCT with Myc. The number of observed cells is indicated in the graph (A and G). H and I, effect of p27Y88E on interaction between CK2α’β and p27 (H) and Ca_V1.2Δ1821 channels (I) in tsA201 cells. H, Myc‐p27 (WT or Y88E), FLAG‐CK2α’ and FLAG‐CK2β were expressed in tsA201 cells. I, HA‐Ca_V1.2Δ1821 channels, Myc‐p27 (WT or Y88E), FLAG‐CK2α’ and FLAG‐CK2β were expressed in tsA201 cells. IB and IP were performed with indicated antibodies and data are representative of three independent blots (A–F, H and I). In each panel contrast and brightness were adjusted to the whole image to make the features of interest clearer.

For immunoblotting, samples (10–50 μg per lane) were separated on 7.5% or 10% SDS‐PAGE and electroblotted onto polyvinylidene difluoride membranes. Non‐specific binding was blocked with Blocking One (Nacalai tesque) or Blocking One‐P (Nacalai tesque) for 30 min at room temperature. Membranes were incubated with primary antibodies overnight at 4°C. After being washed with TBST, membranes were reacted with HRP‐conjugated secondary antibodies for 1 h at room temperature. The bound secondary antibody was visualized with Immobilon Western (Sigma‐Aldrich) according to the manufacturer's instructions. Signal intensities of bands were quantified with the Gel analysis program of the ImageJ software (NIH, Bethesda, MD, USA).

Statistics

All data are expressed as means ± SEM. Statistical significance was evaluated using the unpaired Student's t test. For multiple comparisons of data, ANOVA followed by Dunnett's test was used unless otherwise indicated. P < 0.05 was considered statistically significant. All statistical analyses were performed using the SPSS version 23 software (SPSS Inc., Armonk, NY, USA).

Results

AngII increases twitch Ca²⁺ transients in neonatal but not adult ventricular cardiomyocytes

It is of debate whether AngII augments cardiac contractility (Allen et al. 1988; Ikenouchi et al. 1994; Petroff et al. 2000; Aiello & Cingolani, 2001; Bkaily et al. 2005; Gassanov et al. 2006). We hypothesized that this is partly because AngII differentially regulates cardiac contractility at distinct developmental stages. Thus, we first compared the effects of AngII on twitch Ca²⁺ transients in mouse neonatal and adult ventricular cardiomyocytes (NVCMs and AVCMs, respectively). In NVCMs, 2 h treatment of AngII (3 μm) significantly increased the peak of Ca²⁺ transients (Fig. 1 A). However, AngII had no significant effect on Ca²⁺ transients in AVCMs (Fig. 1 A). These results suggested that AngII may exert a positive inotropic effect in immature cardiomyocytes.

Figure 1. AngII increases twitch Ca²⁺ transients and L‐type Ca²⁺ channel currents in immature cardiomyocytes.

A, effect of AngII on Ca²⁺ transients in Fluo‐4‐loaded NVCMs [postnatal day (P)0, left] and AVCMs (P56–62, right). Each left‐hand panel, representative Ca²⁺ transients of control cardiomyocytes (black lines) and cardiomyocytes treated with AngII (3 μm, 2 h) (red lines). Each right‐hand panel, the summary of peak ΔF/F ₀. The number of observed cells is indicated in the graph (NVCMs from 15 mouse pups and AVCMs from 3 mice). ^* P < 0.01 versus each control. B, effect of AngII (3 μm, 2 h) on peak current density–voltage relationships of L‐type Ca²⁺ channels in VCMs (P0 from 12 mouse pups, P10 from 3 mice, P28–34 from 3 mice, and P56–62 from 3 mice). The number of observed cells is indicated in the graph. ^* P < 0.05 versus each control. C, age‐dependent effect of AngII on peak L‐type Ca²⁺ channel currents at 0 mV shown in B. D, effect of AngII (3 μm, 2 h) on peak current density–voltage relationship of L‐type Ca²⁺ channels in HL‐1 cardiomyocytes, neonatal (P0 from 15 mouse pups) and adult (P56–62 from 4 mice) atrial cardiomyocytes (ACMs). The number of observed cells is indicated in the graph. ^* P < 0.05 versus each control. E, concentration–response curve of AngII‐induced increase in peak L‐type Ca²⁺ channel currents at 0 mV in HL‐1 cardiomyocytes (n = 6–21 cells in each concentration). The curve was fit by the Hill equation.

Age‐dependent effect of AngII on cardiac L‐type Ca²⁺ channels

L‐type Ca²⁺ channel currents are crucial for the Ca²⁺‐induced Ca²⁺ release from the cardiac sarcoplasmic reticulum. Thus, we next examined the effects of AngII on L‐type Ca²⁺ channel currents with Ba²⁺ as a charge carrier. Two hours of treatment of AngII (3 μm) almost doubled the amplitude of peak L‐type Ca²⁺ channel currents at 0 mV in immature cardiomyocytes such as NVCMs (P0) (Fig. 1 B), an atrial myocyte cell line, HL‐1 (Fig. 1 D), and mouse neonatal atrial cardiomyocytes (Fig. 1 D). This response to AngII, however, gradually decreased in an age‐dependent manner and was eventually abolished in adult ventricular and atrial cardiomyocytes (P56–62) (Fig. 1 B–D). AngII increased the amplitude of peak L‐type Ca²⁺ channel currents in HL‐1 cardiomyocytes in a concentration‐dependent manner with EC₅₀ and Hill coefficient of 360 nm and 0.90, respectively (Fig. 1 E).

Mechanism employed by AngII to increase L‐type Ca²⁺ channel activity

To clarify the mechanisms by which AngII augments L‐type Ca²⁺ channel currents, we measured, in NVCMs, the gating charge upon depolarization to the reversal potential of L‐type Ca²⁺ channels (E _rev) and the amplitude of peak tail currents after repolarization from E _rev to –50 mV (Fig. 2 A). AngII did not increase gating charges but increased the amplitude of tail currents in a time‐dependent manner within 2 h (Fig. 2 B and C). As a result, AngII increased the ratio of the amplitude of tail currents to gating charges (Fig. 2 D), indicating that AngII increased the efficiency of the coupling between the voltage‐sensing domain and activation gate of L‐type Ca²⁺ channels. This response continued as long as 18 h. Consistent with this, immunoblotting revealed that AngII did not change the expression level of Ca_V1.2α_1c subunits for 18 h in HL‐1 cardiomyocytes (Fig. 2 E). These results indicated that AngII activates L‐type Ca²⁺ channels by increasing the availability and/or open probability but not the number of single L‐type Ca²⁺ channels.

Figure 2. Mechanism underlying the activation of L‐type Ca²⁺ channels by AngII in NVCMs.

A, representative traces of gating charges and Ba²⁺ tail currents of L‐type Ca²⁺ channels in NVCMs (P0 or P1) in the presence or absence of AngII (3 μm, 2 h). Ordinate indicates the coupling efficiency (nA pC^–1). Inset, voltage protocol. B–D, time‐dependent effect of AngII (3 μm) on gating charge density (B), peak Ba²⁺ tail current density (C) and the coupling efficiency (D) of L‐type Ca²⁺ channels in NVCMs (P0 or P1); n = 6–9 cells from 10–15 mouse pups in each time point. ^* P < 0.05 versus each control (0 h). E, AngII did not alter the expression level of Ca_V1.2α_1c subunits in HL‐1 cardiomyocytes. HL‐1 cardiomyocytes were treated with AngII (3 μm) for the indicated duration. Whole‐cell lysates (α_1c, 50 μg per lane; tubulin, 10 μg per lane) from these HL‐1 cardiomyocytes were immunoblotted with antibodies against Ca_V1.2α_1c and tubulin. Tubulin was used as an internal control; n = 4 independent blots. ^* P < 0.05 versus control (0 h). In each panel contrast and brightness were adjusted to the whole image to make the features of interest clearer.

AngII activates L‐type Ca²⁺ channels through AT₁ receptors and β‐arrestin2

We henceforth analysed the signal transduction system mediating the effect of AngII on L‐type Ca²⁺ channels in immature cardiomyocytes. We found that a selective AT₁ receptor antagonist, candesartan (10 μm, AdooQ BioScience, Irvine, CA, USA), but not a selective AT₂ receptor antagonist, PD123319 (3 μm, Wako), completely abolished the AngII‐induced increase in peak L‐type Ca²⁺ channel current densities in HL‐1 cardiomyocytes (Fig. 3 A), indicating the involvement of AT₁ receptors in this response. Furthermore, knockdown of β‐arrestin2 but not G_q/11 or β‐arrestin1 significantly suppressed the effect of AngII (Fig. 3 B). Consistent with this, a β‐arrestin‐biased AT₁ receptor agonist, TRV027 (customarily produced by Peptide Institute, Inc., Osaka, Japan), activated L‐type Ca²⁺ channels as efficaciously as AngII in NVCMs (Fig. 3 C). These results indicated that AngII activates L‐type Ca²⁺ channels through AT₁ receptors and β‐arrestin2 in immature cardiomyocytes.

Figure 3. AngII activates L‐type Ca²⁺ channels through AT₁ receptors and  β‐arrestin2 in immature cardiomyocytes.

A, effect of candesartan (10 μm) or PD123319 (3 μm) on peak L‐type Ca²⁺ channel current density at 0 mV in HL‐1 cardiomyocytes in the presence or absence of AngII (3 μm, 2 h). HL‐1 cardiomyocytes were pretreated with antagonists for 30 min before treatment of AngII. The number of observed cells is indicated in the graph. ^* P < 0.05 versus control. B, left‐hand panel, representatives of three independent immunoblots of shRNA‐mediated knockdown (KD) of G_q/11, β‐arrestin1 and β‐arrestin2 in HL‐1 cardiomyocytes. Whole‐cell lysates (10–20 μg per lane) from these HL‐1 cardiomyocytes were immunoblotted with indicated antibodies. In each panel contrast and brightness were adjusted to the whole image to make the features of interest clearer. Right‐hand panel, effect of shRNA (sh)‐G_q#1/11#1, sh‐β‐arrestin1#2 or sh‐β‐arrestin2#2 on peak L‐type Ca²⁺ channel current density at 0 mV in HL‐1 cardiomyocytes in the presence or absence of AngII (3 μm, 2 h). The number of observed cells is indicated in the graph. ^* P < 0.01 versus basal of mock. C, effect of TRV027 (3 μm, 2 h) on the coupling efficiency of L‐type Ca²⁺ channels in NVCMs. ^* P < 0.01 versus basal. The number of observed cells is indicated in the graph (NVCMs from 14 mouse pups).

SFKs and CK2α’β are required for the effect of AngII on L‐type Ca²⁺ channels

The results described in Fig. 2 indicated that AngII post‐translationally modifies L‐type Ca²⁺ channels in immature cardiomyocytes. Thus, to elucidate possible involvement of protein kinases in the AngII‐induced response (Hunyady & Catt, 2006; Karnik et al. 2015), we examined, in NVCMs, the effects of an array of pharmacological inhibitors of kinases that are known to be activated by AngII and/or to regulate L‐type Ca²⁺ channels (Fig. 4 A). The effect of AngII was not affected by pretreatment of an epidermal growth factor receptor inhibitor, erlotinib (0.1 μm, Cayman Chemical Company, Ann Arbor, MI, USA); a protein kinase A (PKA) inhibitor, H‐89 (1 μm, Sigma‐Aldrich); a protein kinase B inhibitor, afuresertib (1 μm, ChemScene, Monmouth Junction, NJ, USA); a protein kinase C (PKC) inhibitor, Gö6983 (0.5 μm, Wako); a protein kinase D inhibitor, CID755673 (3 μm, Sigma‐Aldrich); a calcium/calmodulin‐dependent protein kinase II inhibitor, KN‐93 (1 μm, Sigma‐Aldrich); or an extracellular signal‐regulated kinase inhibitor, SCH772984 (1 μm, AdooQ). However, pretreatment of an Src‐family tyrosine kinase (SFK) and Abl tyrosine kinase inhibitor, bosutinib (2 μm, Sigma‐Aldrich), or a CK2 inhibitor, quinalizarin (3 μm, Sigma‐Aldrich), completely inhibited the effect of AngII.

Figure 4. Requirement for SFKs and CK2α’β for the AngII‐mediated activation of L‐type Ca²⁺ channels in immature cardiomyocytes.

A, effects of pharmacological inhibitors on AngII‐mediated activation of L‐type Ca²⁺ channels in NVCMs. Ordinate indicates the coupling efficiency (nA pC^–1) of L‐type Ca²⁺ channels. NVCMs were treated with vehicle, erlotinib (0.1 μm), H‐89 (1 μm), afuresertib (1 μm), Gö6983 (0.5 μm), CID755673 (3 μm), KN‐93 (1 μm), SCH772984 (1 μm), bosutinib (2 μm) or quinalizarin (3 μm) for 30 min before treatment of AngII (3 μm, 2 h). Dashed black line indicates the mean value in the basal condition. Twenty‐eight mouse pups were used. ^* P < 0.05 versus basal. B, upper panel, a representative immunoblot of three independent experiments overexpressing CSK in HL‐1 cardiomyocytes. Lower panel, peak L‐type Ca²⁺ channel current density at 0 mV in HL‐1 cardiomyocytes overexpressing CSK in the presence or absence of AngII (3 μm, 2 h). C, left‐hand panel, representatives of three independent experiments of shRNA‐mediated knockdown (KD) of CK2α, CK2α’ and CK2β in HL‐1 cardiomyocytes. Right‐hand panel, effect of sh‐CK2α#2, sh‐CK2α’#2 or sh‐CK2β#1 on peak L‐type Ca²⁺ channel current density at 0 mV in HL‐1 cardiomyocytes in the presence or absence of AngII (3 μm, 2 h). ^* P < 0.05 versus each basal. The number of observed cells is indicated in the graph (A–C). Whole‐cell lysates (10 μg per lane) from HL‐1 cardiomyocytes were immunoblotted with indicated antibodies (B and C). In each panel contrast and brightness were adjusted to the whole image to make the features of interest clearer.

We also found that overexpression of CSK, an endogenous inhibitor of SFKs (Okada, 2012), inhibited AngII‐induced activation of L‐type Ca²⁺ channels in HL‐1 cardiomyocytes (Fig. 4 B).

CK2 is a heterotetramer consisting of two catalytic subunits (α and/or α’) and two regulatory β subunits (Litchfield, 2003). Knockdown of CK2α’ or CK2β, but not CK2α, significantly suppressed the effect of AngII on L‐type Ca²⁺ channels in HL‐1 cardiomyocytes (Fig. 4 C), indicating a requirement of CK2α’β in this response. Immunoblotting revealed that the expression level of CK2α’ gradually decreased during maturation of the heart (Fig. 5 A and B), and it was 7 and 4.5 times higher in NVCMs (P0) and HL‐1 cardiomyocytes than in AVCMs (P56), respectively (Fig. 5 C and D).

Figure 5. Expression levels of CK2α, α’ and β during the development of mouse ventricular cardiomyocytes.

A and B, mouse VCMs [embryonic days (E) 14.5 from 3 pregnant mice, E19.5 from 3 pregnant mice, P0 from 5 pups in each sample, P10 from 3 mice, P28 from 3 mice and P56 from 3 mice] were extracted. These whole‐cell lysates (10 μg per lane) were immunoblotted with indicated antibodies. Data were normalized to each P0 group; n = 3 independent blots. ^* P < 0.05 versus each P0 group. C and D, expression levels of CK2α, α’ and β in NVCMs (P0) and HL‐1 cardiomyocytes and AVCMs (P56). Whole‐cell lysates (10 μg per lane) were immunoblotted with indicated antibodies. Data were normalized to tubulin; n = 3 independent blots. ^† P < 0.05 versus each NVCM (P0) group, ^# P < 0.05 versus HL‐1 cardiomyocytes group. Representative immunoblot (A and C) and summary of the data (B and D) are shown. In each panel contrast and brightness were adjusted to the whole image to make the features of interest clearer. Statistical significance was determined by ANOVA followed by Dunnett's test (B) and Bonferroni correction (D).

AngII sequentially activates SFKs and CK2α’β

The C‐terminus of cardiac Ca_V1.2α_1c subunit is proteolytically processed in vivo (De Jongh et al. 1996), and the cleaved distal C‐terminus non‐covalently rebinds through the distal C‐terminus regulatory domain (DCRD) to their proximal C‐terminal regulatory domain (PCRD), thereby autoinhibiting Ca_V1.2 channel activity (Fig. 6 A) (Gao et al. 2001; Hulme et al. 2006). To determine the order in which SFKs and CK2 mediate the effect of AngII to Ca_V1.2 channels, we examined whether these kinases could directly reverse the autoinhibition of recombinant Ca_V1.2 channels. We cotransfected cDNA encoding c‐Src or CK2 subunits with those encoding Ca_V1.2Δ1821, DCT, β_2C and α₂δ₁ subunits in tsA201 cells. No increase in the coupling efficiency for Ca_V1.2Δ1821 + DCT channels was observed with coexpression of c‐Src (Fig. 6 A). Overexpression of CK2α’, but not CK2α, CK2β or CK2αβ, slightly but significantly increased the coupling efficiency of Ca_V1.2Δ1821 + DCT channels (Fig. 6 A). Overexpression of CK2α’β significantly increased the coupling efficiency of Ca_V1.2Δ1821 + DCT channels to a level similar to that of Ca_V1.2Δ1821 channels (Fig. 6 A). Moreover, it did not further increase the coupling efficiency of Ca_V1.2Δ1821 channels. Threonine 1704 (Thr¹⁷⁰⁴) in the PCRD was reported to be the substrate of CK2 (Fuller et al. 2010). Indeed, CK2α’β failed to activate Ca_V1.2 channels composed of Ca_V1.2Δ1821T1704A, DCT, β_2C and α₂δ₁ subunits (Fig. 6 B), further suggesting that CK2α’β directly activates Ca_V1.2 channels.

Figure 6. Activation of autoinhibited Ca_V1.2Δ1821 + DCT channels by CK2α’β but not c‐Src.

A, coupling efficiency for Ca_V1.2Δ1821 and Ca_V1.2Δ1821 + DCT channels in tsA201 cells overexpressing Myc‐c‐Src, CK2α, CK2α’, CK2β, CK2αβ and CK2α’β as indicated. Inset, representative traces of gating currents and Ba²⁺ tail currents of recombinant Ca_V1.2 channels and voltage protocol. ^* P < 0.01 versus Ca_V1.2Δ1821 + DCT channels. B, requirement of Thr¹⁷⁰⁴ phosphorylation in Ca_V1.2 channels for CK2α’β‐mediated activation of Ca_V1.2Δ1821 + DCT channels. Coupling efficiency for Ca_V1.2Δ1821T1704A and Ca_V1.2Δ1821T1704A + DCT channels in tsA201 cells with or without overexpression of CK2α’β.  ^* P < 0.01 versus Ca_V1.2Δ1821T1704A + DCT channels. The number of observed cells is indicated in the graph (A and B). C and D, CK2α’β bound to Ca_V1.2Δ1821 through CK2β subunits. HA‐Ca_V1.2 channels, FLAG‐CK2α’ and FLAG‐CK2β were expressed in tsA201 cells as indicated. Antibody‐tagged protein complexes were co‐immunoprecipitated (IP) with non‐immune IgG or antibody against HA from lysates and analysed by immunoblotting (IB) with antibodies against HA and FLAG as indicated. Data are representative from three independent experiments. C, co‐immunoprecipitation of HA‐Ca_V1.2 (FL, Δ1821 or DCT) channels with FLAG‐CK2α’ + FLAG‐CK2β. D, HA‐Ca_V1.2Δ1821 channels and either FLAG‐CK2α’ (left) or FLAG‐CK2β (right) were expressed in tsA201 cells. Co‐immunoprecipitation of HA‐Ca_V1.2Δ1821 channels with FLAG‐CK2α’ (left) or FLAG‐CK2β (right). In each panel contrast and brightness were adjusted to the whole image to make the features of interest clearer.

To test how CK2α’β interacts with Ca_V1.2 channels, we coexpressed cDNA encoding HA‐Ca_V1.2 [full‐length (FL), Δ1821, or DCT], β_2C and α₂δ₁ subunits with FLAG‐CK2α’ and FLAG‐CK2β in tsA201 cells. FLAG‐CK2α’ and FLAG‐CK2β both co‐immunoprecipitated with HA‐FLCa_V1.2 and HA‐Ca_V1.2Δ1821 channels but not HA‐DCT (Fig 6 C). Furthermore, when either FLAG‐CK2α’ or FLAG‐CK2β was individually coexpressed with HA‐Ca_V1.2Δ1821, FLAG‐CK2β but not FLAG‐CK2α’ co‐immunoprecipitated with HA‐Ca_V1.2Δ1821 channels (Fig. 6 D). Thus, it is likely that CK2α’β binds to Ca_V1.2 channels devoid of DCT through CK2β.

SFKs removes the inhibition of CK2α’β through p27

Although the effect of CK2α’β on Ca_V1.2 channels was tightly regulated by AngII in cardiomyocytes (Fig. 4), CK2α’β activated Ca_V1.2 channels constitutively in tsA201 cells (Fig. 6 A). Therefore, there may be a means of inhibiting CK2α’β in cardiomyocytes, and this inhibitor might be released by AngII. It was reported that a cyclin‐dependent kinase (CDK) inhibitor, p27, inhibits CK2α’ in cardiomyocytes (Hauck et al. 2008). We found that knockdown of p27 indeed constitutively activated L‐type Ca²⁺ channels in HL‐1 cardiomyocytes (Fig. 7 A). This activity was almost completely inhibited by quinalizarin (3 μm) but not bosutinib (2 μm), consistent with direct activation of Ca_V1.2 channels by CK2α’β but not SFKs (Fig. 7 A). Knockdown of p27 occluded the effect of AngII on L‐type Ca²⁺ channel activity (Fig. 7 A). Furthermore, Myc‐p27 co‐immunoprecipitated with both FLAG‐CK2α’ and FLAG‐CK2β in tsA201 cells (Fig. 7 B). When HA‐Ca_V1.2Δ1821 was coexpressed with Myc‐p27, FLAG‐CK2α’ and FLAG‐CK2β, HA‐Ca_V1.2Δ1821 channels co‐immunoprecipitated with not only FLAG‐CK2α’β but also Myc‐p27 (Fig. 7 C). Thus, it is likely that Ca_V1.2 channels, CK2α’β and p27 form a macromolecular complex. AngII did not alter the expression level of p27 for 2 h in HL‐1 cardiomyocytes (Fig. 7 D).

We hypothesized that it is SFKs that remove the inhibitory effect of p27 on CK2α’β. A major phosphorylation site of p27 by SFKs is tyrosine 88 (Tyr⁸⁸), which plays an important role in regulating inhibition of CDK by p27 (Jakel et al. 2012). AngII significantly increased Tyr⁸⁸ phosphorylation of p27 in HL‐1 cardiomyocytes in a time‐dependent manner between 5 and 120 min (Fig. 7 E). Pretreatment of bosutinib (2 μm) almost completely inhibited Tyr⁸⁸ phosphorylation of p27 by AngII in HL‐1 cardiomyocytes (Fig. 7 F). In tsA201 cells, CK2α’β‐mediated increase in the activity of Ca_V1.2 channels was completely inhibited by coexpression of wild‐type (WT) p27 or a phosphorylation‐resistant mutant of p27Y88F (Fig. 7 G). In contrast, a phosphomimetic mutant p27Y88E was unable to inhibit CK2α’β (Fig. 7 G). Furthermore, coexpression of c‐Src fully reversed the inhibitory effect of WT p27 but not p27Y88F on CK2α’β (Fig. 7 G). As assessed with immunoprecipitation, p27Y88E could bind to CK2α’β as efficiently as WT p27 in the presence (Fig. 7 H) or absence of Ca_V1.2Δ1821 channels (Fig. 7 I). These results suggested that AngII does not physically dissociate p27 from CK2α’β but impairs the ability of p27 to inhibit CK2α’β through Tyr⁸⁸ phosphorylation by SFKs, thus leading to activation of Ca_V1.2 channels.

Discussion

We found that AngII augmented twitch Ca²⁺ transients by activating L‐type Ca²⁺ channels in immature but not mature cardiomyocytes (Fig. 8). AngII caused this effect through AT₁ receptors and β‐arrestin2. β‐Arrestin2 activated SFKs, which removed the inhibitory effect of p27 on CK2α’β. Then, CK2α’β phosphorylated Thr¹⁷⁰⁴ in the PCRD of Ca_V1.2α_1c subunits and reversed the autoinhibition of Ca_V1.2 channels by the DCT (Gao et al. 2001; Hulme et al. 2006), thereby activating Ca_V1.2 channels. It was likely that Ca_V1.2 channels, CK2α’β and p27 formed a macromolecular complex. These results suggest that AngII may exert a positive inotropic effect through this novel signalling pathway in immature cardiomyocytes although we cannot exclude the possibility that other changes in Ca²⁺ signalling may also contribute to an increase in Ca²⁺ transients in response to AngII.

Figure 8. A signaling pathway revealed by this study.

Schematic illustration of the AngII‐mediated activation of Ca_V1.2 channels in immature cardiomyocytes.

It has been of debate whether AngII causes a positive inotropic effect on the heart (Lefroy et al. 1996; Watanabe & Endoh, 1998; Petroff et al. 2000; Palomeque et al. 2006; Rajagopal et al. 2006; Liang et al. 2010; Acharya et al. 2011) and/or activates L‐type Ca²⁺ channels (Allen et al. 1988; Ikenouchi et al. 1994; Petroff et al. 2000; Aiello & Cingolani, 2001; Bkaily et al. 2005; Gassanov et al. 2006). This controversy might have arisen at least in part because of the different ages of animals used in experiments. We found that the expression level of CK2α’ subunits age‐dependently decreased in the postnatal period (Fig. 5). The expression level of p27 in cardiomyocytes also increases with age (Torella et al. 2004). These facts may explain the age‐dependent inotropic effect of AngII. In a study using sheep perinatal fetuses (Acharya et al. 2011), administration of AngII significantly increased +dP/dt _max of the left ventricle by ∼66%, consistent with our present observation. Lefkowitz and his colleagues reported that AT₁ receptors have a positive inotropic effect in adult mouse cardiomyocytes through cooperative action of G_q/11/PKC and β‐arrestin2 (Rajagopal et al. 2006). Thus, the molecular mechanisms underlying inotropic effects of AngII on mature cardiomyocytes may be different from that on immature cardiomyocytes.

AngII increased L‐type Ca²⁺ channel activity in a concentration‐dependent manner with an EC₅₀ of 360 nm, an ∼100 times higher value than that for AngII‐induced contraction of mouse abdominal aorta (4.6 nm) (Zhou et al. 2003). However, when rat NVCMs in culture were incubated with prorenin and angiotensinogen, cell proliferation equivalent to that induced by 100 nm AngII was observed, although the concentration of AngII in the medium remained < 1 nm (Saris et al. 2002). Thus, the local RAS exposes cardiomyocytes to higher concentrations of AngII than those in the plasma. Therefore, the signalling system identified in this study may be driven by the local RAS in the heart.

The effect of AngII on L‐type Ca²⁺ channels was attenuated by inhibition of SFKs or CK2 (Fig. 4). Both SFKs and CK2 are included in the interactome of β‐arrestins (Xiao et al. 2007). However, we considered that SFKs work upstream to CK2α’β in this signalling system from the following three pieces of evidence. First, in tsA201 cells, overexpression of CK2α’β but not c‐Src significantly activated recombinant Ca_V1.2 channels (Fig. 6 A). Second, the effect of AngII was almost completely inhibited by the alanine substitution of Thr¹⁷⁰⁴, a substrate of CK2 (Fuller et al. 2010) (Fig. 6 B). Finally, in HL‐1 cardiomyocytes, enhancement of L‐type Ca²⁺ channel activity by knockdown of p27 was inhibited by quinalizarin but not bosutinib (Fig. 7 A). However, because SFKs are also reported to directly phosphorylate and augment L‐type Ca²⁺ channel activity in native cells such as neurons, smooth muscle cells and cardiomyocytes (Hu et al. 1998; Bence‐Hanulec et al. 2000; Dubuis et al. 2006; Gui et al. 2006), AngII might upregulate L‐type Ca²⁺ channels not only through CK2α’β but also through SFKs.

Hauck et al. (2008) reported that AngII activated CK2α’ and thereby phosphorylated Ser⁸³ and Thr¹⁸⁷ of p27 in rat neonatal cardiomyocytes. These phosphorylations then impaired the ability of p27 to inhibit CK2α’. This is collectively suggestive of the existence of a positive feedback loop between p27 and CK2α’ that is regulated by CK2α’ itself. However, they did not identify what activated CK2α’ first. We propose that it is SFKs that first phosphorylate Tyr⁸⁸ of p27 and impair its ability to inhibit CK2α’. This would then be followed by positive feedback between p27 and CK2α’. Indeed, SFKs significantly phosphorylated Tyr⁸⁸ of p27 as early as 5 min after stimulation with AngII (Fig. 7 E). In addition, the phosphorylation‐resistant p27 mutant (Y88F) inhibited the effect of constitutively active CK2α’ on Ca_V1.2 channels, both in the presence and in the absence of c‐Src (Fig. 7 G). It was likely that physical dissociation of p27 from CK2α’ is not necessary for the disinhibition of CK2α’ (Fig. 7 H and I). Hauck et al. (2008) found that p27 knock‐out mice had significantly higher contractility of the left ventricle, especially in adulthood. We speculated that this was because CK2α’ activated Ca_V1.2 channels more strongly in the knock‐out compared with WT mice in adulthood when p27 expression level was high.

In the heart, the C‐terminus of most Ca_V1.2α_1c subunits is truncated (Gao et al. 1997) and reassociated with the PCRD, and autoinhibits Ca_V1.2 channels (Gao et al. 2001; Hulme et al. 2006). PKA is bound to Ca_V1.2α_1c subunits through A‐kinase anchoring proteins (AKAPs) associated with the distal C‐terminus of the Ca_V1.2α_1c subunit (Marshall et al. 2011). It phosphorylates Ser¹⁷⁰⁰ in the PCRD, and thereby attenuates the autoinhibitory effect of the distal C‐terminus and activates Ca_V1.2 channels (Fuller et al. 2010). We found that CK2α’ was bound to Ca_V1.2α_1c subunits devoid of DCT through CK2β (Fig. 6 C and D), phosphorylated Thr¹⁷⁰⁴ (Fig. 6 B), and thereby activated Ca_V1.2 channels (Fig. 6 A). Thus, it can be posited that what CK2β is for CK2α’ is analogous to what AKAPs are for PKA. Because both Ser¹⁷⁰⁰ and Thr¹⁷⁰⁴ are located in close proximity at the interface between the DCRD and the PCRD (Fuller et al. 2010), CK2α’ may activate Ca_V1.2 channels in a manner similar to PKA. Indeed, CK2α’β like PKA increased the coupling efficiency between the voltage‐sensing domain and the activation gate of Ca_V1.2 channels (Fig. 2) (Fuller et al. 2010).

β‐Adrenergic stimulation increases cardiac L‐type Ca²⁺ channel currents by ∼100 to 200% within several minutes (McDonald et al. 1994). This response is transient and desensitized quickly. AngII was as efficacious as catecholamines in activating L‐type Ca²⁺ channel currents (Fig. 1), but its effect was slower and more sustaining (Fig. 2). This may be because catecholamines activate PKA through G proteins whereas AngII activates CK2α’ through β‐arrestin2 (Ahn et al. 2004; Shenoy et al. 2006). In the perinatal period of mammals, catecholamines secreted from the adrenal medulla and circulating and local AngII play a crucial role in pups’ response to circulatory stresses (Heymann et al. 1981). These two systems may be complementary in regulating cardiac function in this period by means of their different time course of actions. The action of AngII on cardiac L‐type Ca²⁺ channels continues as long as > 18 h and seems to be important for subacute maintenance of the circulatory system in the perinatal period. However, we do not know the exact molecular mechanisms underlying the slow response of L‐type Ca²⁺ channels to AngII. Because SFKs submaximally phosphorylated p27 at 5 min after AT₁ receptor stimulation (Fig. 7 E), the rate‐limiting step in this signalling pathway may exist at the step of disinhibition of CK2 or phosphorylation of Ca_V1.2 by CK2. In this context, it is of note that CK2 usually requires negative charges near its target serine/threonine for effective phosphorylation (Litchfield, 2003). Thus, a negative charge added by coincident phosphorylation of Ser¹⁷⁰⁰ by PKA could enable CK2α’β to phosphorylate Thr¹⁷⁰⁴ more promptly.

We found that a β‐arrestin‐biased AT₁ receptor agonist TRV027 was as effective as AngII in activating L‐type Ca²⁺ channels (Fig. 3 C). This agent activates β‐arrestins but inhibits G_q/11 proteins coupled to AT₁ receptors and thus has an ideal profile for therapeutics against heart failure (Violin et al. 2010). However, this drug failed to meet either the primary or the secondary endpoints in the phase 2b BLAST‐AHF trial that enrolled adult patients (Felker et al. 2015; Greenberg, 2016). In this regard, TRV027 decreased mean atrial pressure and pulmonary capillary wedge pressure but only modestly increased cardiac output in a tachypacing‐induced heart failure model of adult canines (Boerrigter et al. 2012). We speculate that this was because of the age‐dependent effect of AT₁ receptors on cardiac L‐type Ca²⁺ channels. Thus, if we can extrapolate the present observation to humans, we consider that TRV027 might be more suitable for paediatric than for adult heart failure. In the field of children's health care, the little progress that has been made over the past decades for treating adult heart failure has been sufficiently translated into treatments regarding paediatric heart failure (Hsu & Pearson, 2009). Thus, β‐arrestin‐biased AT₁ receptor agonists, which have not only a positive inotropic effect but an anti‐apoptotic effect on cardiomyocytes, may be used as valuable therapeutics for paediatric heart failure in the future (Kim et al. 2012).

Additional information

Competing interest

The authors declare that they have no competing interests.

Author contributions

All studies were performed in the Department of Molecular Pharmacology at Shinshu University School of Medicine, Matusmoto, Japan. Conception and design of the work: T.K., T.N. and M.Y. Collection, assembly, analysis and interpretation of data: T.K., T.N., K.K., T.T. and M.Y. Drafting and revising the manuscript: T.K., T.N., K.K., T.T. and M.Y. All authors approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

This work was supported by JSPS KAKENHI Grant Numbers 24790210, 26860145, and 16K08546 to T.K.