Reconstitution of the Cytoplasmic Interaction between Phospholamban and Ca -ATPase of Cardiac Sarcoplasmic Reticulum

Phospholamban (PLN) reversibly inhibits the Ca -ATPase of cardiac sarcoplasmic reticulum (SERCA2a) through a direct protein-protein interaction, playing a pivotal role in the regulation of intracellular Ca in heart muscle cells. The interaction between PLN and SERCA2a occurs at multiple sites within the cytoplasmic and membrane domains. Here, we have reconstituted the cytoplasmic protein-protein interaction using bacterially expressed fusion proteins of the cytoplasmic domain of PLN and the long cytoplasmic loop of SERCA2a. We have developed two methods to evaluate the binding of the fusion proteins, one with glutathione-Sepharose beads and the other with a 96-well plate. Essentially the same results were obtained by the two methods. The affinity of the binding (KD) was 0.70 M. The association was inhibited by cAMP-dependent phosphorylation of the PLN fusion protein and by usage of anti-PLN monoclonal antibody. It was also diminished by substitution at the phosphorylation site of PLN of Ser to Asp. These results suggest that PLN can bind SERCA2a in the absence of the membrane domains and that the modifications of the cytoplasmic domain of PLN that activate SERCA2a parallel the disruption of the association between the two fusion proteins. It has been shown that the removal of PLN inhibition of SERCA2a rescues cardiac function and morphology in the mouse dilated cardiomyopathy model. Our assay system can be applied to the screening of novel inotropic agents that remove the inhibition of SERCA2a by PLN, improving the relaxation as well as the contractility of the failing heart. The Ca -ATPase of cardiac sarcoplasmic reticulum (SERCA2a) plays a pivotal role in the contraction and relaxation of heart muscle. It pumps Ca from the cytoplasm into the lumen of the sarcoplasmic reticulum (SR), resulting in muscle relaxation. In turn, the release of Ca from SR induces muscle contraction (Fleischer and Inui, 1989). The activity of SERCA2a is regulated by another SR membrane protein, phospholamban (PLN) (Tada and Katz, 1982). PLN in the dephosphorylated state inhibits SERCA2a by lowering its apparent affinity for Ca (Inui et al., 1986). Phosphorylation of PLN by cAMP-dependent protein kinase reduces the inhibition, resulting in facilitation of the Ca -pumping activity of SERCA2a, which leads to the acceleration of relaxation and the following increase in contractility of heart muscle. This mechanism constitutes a principal intracellular signaling pathway involved in the positive inotropic effects of -adrenergic stimulation. Thus, the SERCA2a-PLN system plays a key role in the regulation of cardiac contractility and relaxation (Kadambi and Kranias, 1997; Simmerman and Jones, 1998; Tada et al., 1988). PLN is a 52-amino acid integral membrane protein consisting of three domains: the N-terminal cytoplasmic helical domain (domain Ia), the less structured cytoplasmic region (domain Ib), and the C-terminal membrane helix (domain II) (Fujii et al., 1987). Chemical cross-linking between purified PLN and SERCA2a molecules has provided evidence that PLN directly binds to SERCA2a (James et al., 1989). Studies employing site-specific mutagenesis followed by transient expression in human embryonic kidney 293 cells have revealed that two, or possibly three, interaction sites in the cytoplasmic domains and the membrane helices of both proteins are involved in the inhibitory action of PLN on SERCA2a (Toyofuku et al., 1994a,b; Kimura et al., 1997, 1998; Asahi et al., 1999, 2001). Based on these observations, we have proposed that PLN and SERCA2a interact via a four (or possibly six)-base circuit through which long range inhibitory interactions are propagated among a series of cytoplasmic and intramembrane interaction sites (MacLennan et al., This work was supported in part by grants from the Ministry of Education, Science, Sports and Culture of Japan (to Y.K. and to M.I.). ABBREVIATIONS: SERCA, Ca -ATPase of sarco(endo)plasmic reticulum; SR, sarcoplasmic reticulum; PLN, phospholamban; PCR, polymerase chain reaction; PDZ, PSD-95/Dlg/ZO-1; BSA, bovine serum albumin; PAGE, polyacrylamide gel electrophoresis; PSD-95, postsynaptic density protein 95; GST, glutathione S-transferase. 0026-895X/02/6103-667–673$3.00 MOLECULAR PHARMACOLOGY Vol. 61, No. 3 Copyright © 2002 The American Society for Pharmacology and Experimental Therapeutics 1325/968358 Mol Pharmacol 61:667–673, 2002 Printed in U.S.A. 667 at A PE T Jornals on O cber 9, 2017 m oharm .aspeurnals.org D ow nladed from 1998). Disruption of one of these interactions probably removes the inhibitory effect of PLN on SERCA2a. Studies using gene targeting techniques have shown that excessive inhibition of SERCA2a by mutated PLN impairs cardiac contractility and relaxation (Zhai et al., 2000; Zvaritch et al., 2000), providing a model of heart failure and cardiomyopathy. On the other hand, PLN-knockout mice exhibit increased cardiac performance without catecholamine stimulation, although the enhancement of cardiac function by catecholamines is lost (Luo et al., 1994). Moreover, it has been reported that PLN ablation rescues the impaired cardiac function and morphological changes in mouse dilated cardiomyopathy generated by knockout of a cytoskeletal protein (Minamisawa et al., 1999). Thus, removal of PLN inhibition of SERCA2a could be a good therapeutic option for treatment of heart failure and cardiomyopathy. In this study, we have developed methods to reconstitute the cytoplasmic interaction between PLN and SERCA2a using fusion proteins of the cytoplasmic domains of the two proteins. We have found that the cytoplasmic domain of PLN physically associates with the cytoplasmic loop of SERCA2 in the absence of the membrane domains and that the modifications that remove the inhibition of the Ca -pumping activity of SERCA2a by PLN disrupt the physical association of the two fusion proteins. Our assay method can be applied to the screening of novel inotropic agents that improve the relaxation as well as the contractility of heart muscle by removing the inhibition of SERCA2a by PLN. Materials and Methods Construction of PLN-PDZ-Myc-His6 and GSTSERCA2LCL Fusion Proteins. The DNA fragment coding from Met to Gln of human phospholamban was obtained from a human heart cDNA library by polymerase chain reaction (PCR). An NcoI site was introduced using the forward PCR primer 5 -AGCCATGGAGAAAGTCCAATAC-3 , and an XbaI site was introduced using the reverse PCR primer 5 -CCTCTAGACGTTGACGTGCTTGTTGAGGCA-3 . The amplified DNA fragments were subcloned into the SmaI site of pBluescript (pBS/PLN). The construction was confirmed by dideoxynucleotide sequencing. The insert DNA fragment was excised by restriction digestion, and ligated into the NcoI and XbaI sites of pTrcHis2B (Invitrogen, Groningen, The Netherlands). This construct was named pTrc/PLN. The cDNA fragment coding for the three PDZ domains of rat PSD-95 (13–1176) was amplified using PCR. An SpeI site was introduced using the forward primer 5 -GCACTAGTCTGTATAGTGACAACCAAG-3 , and an XbaI site was introduced using the reverse primer 5 -CTTCTAGATTATACTGAGCGATGATCG-3 . The amplified DNA fragment was digested with SpeI and XbaI, and subcloned into the XbaI site of pTrc/PLN 26. This construct was named pTrc/PLN-PDZ. The cDNA fragment coding for the PDZ domains was also digested with XbaI, and subcloned into the XbaI site and blunted SacI site of pTrcHis2B; this construct was named pTrc/PDZ and was used as control. The fusion proteins (PLN-PDZ-Myc-His6 and PDZ-Myc-His6) were expressed in Escherichia coli DH5 and purified by Ni-NTA column chromatography (QIAGEN, Hilden, Germany). The large cytoplasmic loop of SERCA2a (SERCA2LCL) was prepared from rabbit heart mRNA by reverse transcription PCR. The DNA fragment coding for the long cytoplasmic loop of SERCA2a (994–2182) was amplified using the forward primer 5 -GCCATTGTTCGAAGCCTCC-3 and the reverse primer 5 -TTAAGCCACCGCAGTGCCAGA-3 by touchdown PCR (Don et al., 1991) in which the annealing temperature was shifted from 60° to 50°C by 1°C every 3 cycles. The amplified DNA fragment was subcloned into the SmaI site of pGEX-3X (pGEX/SERCA2LCL). The DNA fragments corresponding to the plus and minus strands encoding Ile-His of SERCA2a were synthesized on a DNA synthesizer, annealed, and subcloned into the SmaI site of pGEX-3X (pGEX/SERCA2LCL 405). The constructions were confirmed by dideoxynucleotide sequencing. The fusion proteins (GST-SERCA2LCL and GSTSERCA2LCL) were expressed in Escherichia coli DH5 and purified by glutathione-Sepharose column chromatography. Protein concentrations were determined by the method of Bradford (1976). Oligonucleotide-Directed Mutagenesis. Mutations of Ser to Asp (S16D) and Thr to Asp (T17D) were introduced into the PLN sequence inserted into pBluescript (pBS/PLN) by the method of Kunkel (1985). After confirming the sequences, the mutant PLN DNA fragments were excised by NcoI and XbaI digestion, and then subcloned into NcoI and SpeI sites of pTrc/PDZ. Binding Assay Using Glutathione-Sepharose Beads. PLN fusion protein was incubated with GST-SERCA2LCL or GST in 300 l of binding/wash buffer (20 mM HEPES-NaOH, pH 7.5, 100 mM KCl, 2 mM -mercaptoethanol, and 0.1% Triton X-100) containing 0.5% BSA at room temperature for 2 h with agitation. Twenty microliters of glutathione-Sepharose beads prewashed three times with binding/wash buffer was added and incubated for 1 h. PLN fusion protein bound to glutathione-Sepharose beads was sedimented, washed five times with binding/wash buffer, and then eluted with the use of Laemmli’s SDS sample buffer (Laemmli, 1970). The proteins were separated by SDS-PAGE and transferred onto a nitrocellulose sheet. Immunoblotting was performed with anti-PSD-95 polyclonal antibodies (Yamada et al., 1999) as a primary antibody and peroxidase-conjugated anti-rabbit IgG goat antibodies as a secondary antibody. Labeled bands were visualized by enhanced chemiluminescence. For the quantitation of bound PLN-PDZ-Myc-His6 to GST-SERCA2LCL, various amounts of PLN-PDZ-Myc-His6 were incubated with 1 g of GST-SERCA2LCL or 0.5 g of GST in 50 l of binding/wash buffer at room temperature for 2 h, and then glutathione-Sepharose beads were added. Elution, SDS-PAGE, and immunoblotting were done as described above. The amounts of PLN fusion proteins bound to GST or GST-SERCA2LCL were densitometrically determined using an ATTO Lane and Spot Analyzer (ATTO, Tokyo, Japan). Binding Assay Using a 96-Well Plate. Two micrograms of GSTSERCA2LCL was immobilized in each well of a 96-well plate in 100 l of plate buffer containing 20 mM HEPES-NaOH, pH 7.5, 100 mM KCl, and 2 mM -mercaptoethanol at 37°C for 1 h followed by blocking with 0.1% BSA in plate buffer at 37°C for 1 h. One microgram of GST with 1 g of BSA was immobilized in a well as a control. The plate was washed twice with plate buffer and then twice with plate buffer containing 0.05% Tween 20. One microgram of PLN fusion protein in 100 l of plate buffer containing 0.05% Tween 20 and 0.5% BSA was added to each well and incubated for 3 h at 25°C. In some experiments, anti-PLN monoclonal antibody A1 (1–25 g/ ml) (Suzuki and Wang, 1986) was included. The same concentrations of mouse IgG were used as a control. After washing five times with plate buffer containing 0.05% Tween 20, bound PLN fusion protein was cross-linked by 4% paraformaldehyde in phosphate-buffered saline for 20 min at room temperature, washed twice with Trisbuffered saline/Tween 20 (20 mM Tris-HCl, pH 7.4, 150 mM NaCl and 0.05% Tween 20), and then incubated with 0.5% skim milk in Tris-buffered saline/Tween 20 for 30 min at room temperature. AntiPSD-95 polyclonal antibodies were used as a primary antibody and peroxidase-conjugated anti-rabbit IgG goat antibodies as a secondary antibody to detect the fusion proteins. Thirty minutes after addition of TMB One solution (Promega, Madison, WI), the absorbance at 655 nm was measured using a model 550 microplate reader (Bio-Rad, Hercules, CA). The absorbance of PLN fusion protein bound to GST was subtracted from the absorbance of that bound to