AKAP complex regulates Ca2+ re-uptake into heart sarcoplasmic reticulum

Abstract

The beta-adrenergic receptor/cyclic AMP/protein kinase A (PKA) signalling pathway regulates heart rate and contractility. Here, we identified a supramolecular complex consisting of the sarcoplasmic reticulum Ca(2+)-ATPase (SERCA2), its negative regulator phospholamban (PLN), the A-kinase anchoring protein AKAP18delta and PKA. We show that AKAP18delta acts as a scaffold that coordinates PKA phosphorylation of PLN and the adrenergic effect on Ca(2+) re-uptake. Inhibition of the compartmentalization of this cAMP signalling complex by specific molecular disruptors interferes with the phosphorylation of PLN. This prevents the subsequent release of PLN from SERCA2, thereby affecting the Ca(2+) re-uptake into the sarcoplasmic reticulum induced by adrenergic stimuli.

Figure 1

AKAP18δ is present in the heart sarcoplasmic reticulum. (A) Fractions of rat heart sarcoplasmic reticulum (SR) were subjected to a solid-phase binding assay using ³²P-labelled RIIα (RII overlay) as a probe in the absence (upper panel) or presence (middle panel) of the Ht31 anchoring disruptor peptide (500 nM). The same fractions were analysed by immunoblot for the presence of SR proteins ryanodine receptor 2 (RYR2), Ins(1,4,5)P3 receptor II (IP3RII) and calsequestrin, a major Ca²⁺-binding protein of SR (lower panels). Calsequestrin was routinely used in the following as an SR marker and indicator for the quality of SR enrichment. Fraction numbers refer to a discontinuous sucrose density gradient fractionation. (B) Detection of AKAP18δ in rat heart SR fractions by immunoblotting (IB). Pep: AKAP18δ antibody was preincubated with the peptide used for immunization as specificity control. NCX was used as a sarcolemmal marker. (C) Levels of immunoreactive PKA regulatory (RIα, RIIα and RIIβ) and catalytic (C) subunits in SR fractions. (D) Rat heart homogenate was subjected to immunoprecipitation (IP) with AKAP18δ antibody or pre-immune IgG. Total extract and immunoprecipitates were analysed by using RII overlay. Recombinant AKAP18δ protein was used as a positive control. AKAP, A-kinase anchoring protein; NCX, Na⁺/Ca²⁺ exchanger; PKA, protein kinase A.

Figure 2

AKAP18δ and PKA colocalize with SERCA2 and PLN in heart tissue. (A) Rat heart tissue sections were immunostained for AKAP18δ (red) in combination with α-actinin (green), PKA-RIIα (green), SERCA2 (green) and PLN (green), and for SERCA2 (red) in combination with α-actinin (green) and PKA-RIIα (green). The relative fluorescence intensities along an axis perpendicular to the orientation of the sarcomeres are shown (right panels). Scale bar, 20 μm. (B) Immunogold staining was carried out using secondary antibodies labelled with gold particles of different sizes to allow dual staining. Co-staining of AKAP18δ (15 nm) and PLN (10 nm), AKAP18δ (18 nm) and SERCA2 (12 nm), and PLN (15 nm) and SERCA2 (10 nm). Scale bars, 1 μm. The magnified views show representative areas where the indicated proteins colocalize (arrowheads). AKAP, A-kinase anchoring protein; PKA, protein kinase A; PLN, phospholamban; SERCA2, sarcoplasmic reticulum Ca²⁺-ATPase.

Figure 3

AKAP18δ interacts with PLN in the heart sarcoplasmic reticulum. (A) Pooled rat heart sarcoplasmic reticulum (SR) fractions were subjected to affinity chromatography on Rp-8-AHA-cAMP-agarose in the absence (lane 1) or presence (lane 2) of excess cAMP. Eluates were analysed by immunoblot (IB) for the presence of PKA-RIIα, PKA-C, AKAP18δ and SERCA2 (Rp-8-AHA-cAMP is a PKA antagonist that does not dissociate PKA-C from holoenzyme). (B) Pooled rat heart SR fractions were subjected to immunoprecipitation (IP) with AKAP18δ or control rabbit IgG antibodies. Lysates and immunoprecipitates were analysed by immunoblot for the presence of PLN pentamer (upper band) and monomer (lower band; lysate not boiled; note the relative abundance of monomer). (C) HEK293 cells were transfected with expression vectors encoding AKAP18δ and PLN fused to GFP. Lysates were subjected to immunoprecipitation with anti-GFP (left panel) or anti-AKAP18δ (right panel), and lysates and precipitates were analysed for the presence of AKAP18δ and GFP–PLN by immunoblot. GFP-transfected cells were used as negative controls. (D) PLN residues important for AKAP18δ binding were identified by overlaying an array of immobilized PLN 20-mer peptides (2-amino-acid offset) with GST–AKAP18δ (top panel). PLN peptides with or without phosphorylated Ser 16 (pS) were subjected to GST–AKAP18δ overlay experiments (bottom panel). GST alone and GST–AKAP18δ preincubated with the AKAP18δ–PLN disruptor peptide were used as negative controls. Underscoring indicates residues relevant for AKAP18δ–PLN binding. AKAP, A-kinase anchoring protein; GFP, green fluorescent protein; GST, glutathione-S-transferase; PKA, protein kinase A; PLN, phospholamban; SERCA2, sarcoplasmic reticulum Ca²⁺-ATPase.

Figure 4

Disruption of the AKAP18δ–PLN complex influences PLN-Ser 16 phosphorylation and Ca²⁺ re-uptake in the sarcoplasmic reticulum. (A) Immunofluorescent labelling of AKAP18δ (red) and α-actinin (green) in rat neonatal cardiac myocytes. Scale bar, 20 μm. (B) Disruption of the AKAP18δ–PLN interaction with peptide PLN-Arg 11. Adult cardiac myocytes were attached to laminin-coated glass coverslips and incubated with the AKAP18–PLN disruptor peptide PLN-Arg 11 or the corresponding control peptide pSer 16-PLN. AKAP18δ was detected by immunofluorescence microscopy. Scale bar, 20 μm. (C) Rat neonatal cardiac myocytes were treated with or without Arg 9-PLN; peptide (50 μM, 30 min) before stimulation with isoproterenol (iso; 0.1 μM, 5 min) as indicated and analysed for immunoreactive pSer 16-PLN (IB; immunoblots; top panel). Phosphorylated PLN peptide (Arg 9-pSer 16-PLN; right lane) was used as a negative control. Dotted lines indicate lanes excised/combined from a single gel. The histogram shows levels of phosphorylated Ser 16-PLN quantified by densiometry relative to calsequestrin levels (bottom panel). Bars represent the mean±s.e.m. from 3–6 independent experiments (^*P<0.005, Student's t-test; NS, not significant). (D) Rat neonatal cardiac myocytes were transfected with the D1ER sensor. Responses to a 10 mM caffeine pulse (1 s, arrow) in control cells (red curves) or cells pretreated with PLN-Arg 11 peptide (25 μM, 40 min, green curves) with (open symbols) or without (filled symbols) treatment with norepinephrine (NE; 10 μM, 20 min) as indicated were recorded. Time constant averages (τ, mean±s.e.m.) were calculated (right). For each sample, more than 20 independent cells were examined (^*P<0.025, by Student's t-test and one-way ANOVA for paired and independent samples, respectively). AKAP, A-kinase anchoring protein; PLN, phospholamban; SERCA2, sarcoplasmic reticulum Ca²⁺-ATPase; SR, sarcoplasmic reticulum.

Figure 5

Knockdown of AKAP18δ affects Ca²⁺ re-uptake in the sarcoplasmic reticulum. Kinetics of Ca²⁺ release and re-uptake in the sarcoplasmic reticulum of depolarized cardiac myocytes transfected with the D1ER sensor alone (red curves), together with control siRNA (blue curves) or AKAP18δ siRNA (green curves) in the presence (open symbols) or absence (filled symbols) of 10 μM norepinephrine (NE; the arrow indicates the time of NE addition). SR Ca²⁺ was depleted by 50 μM BHQ, the cells were washed and extracellular Ca²⁺ was added. siRNA was Cy3-labelled and transfected cells were thus identified. For clarity, only every second data point is shown. Note: timescale differs at breakpoint. The time constant averages (τ, mean±s.e.m.) were calculated excluding outlier values outside mean±2 s.d. (right). For each sample, 10–17 independent cells were analysed (^*P<0.025, by Student's t-test and one-way ANOVA for paired and independent samples, respectively; NS, not significant). Immunoblot (IB, right): siRNA efficacy tested in easily transfectable HaCaT cells expressing GFP–AKAP18δ. Tx. ctr.: transfection control, an unrelated Flag-tagged construct (β-arrestin) was co-transfected and detected by anti-Flag to control for transfection and loading. AKAP, A-kinase anchoring protein; BHQ, 2,5-di-tert-butylhydroquinone; GFP, green fluorescent protein; siRNA, short interfering RNA; SR, sarcoplasmic reticulum.