Phospholemman regulates cardiac Na+/Ca2+ exchanger by interacting with the exchanger's proximal linker domain

Abstract

Phospholemman (PLM) belongs to the FXYD family of small ion transport regulators. When phosphorylated at Ser(68), PLM inhibits cardiac Na(+)/Ca(2+) exchanger (NCX1). We previously demonstrated that the cytoplasmic tail of PLM interacts with the proximal intracellular loop (residues 218-358), but not the transmembrane (residues 1-217 and 765-938) or Ca(2+)-binding (residues 371-508) domains, of NCX1. In this study, we used intact Na(+)/Ca(2+) exchanger with various deletions in the intracellular loop to map the interaction sites with PLM. We first demonstrated by Western blotting and confocal immunofluorescence microscopy that wild-type (WT) NCX1 and its deletion mutants were expressed in transfected HEK-293 cells. Cotransfection with PLM and NCX1 (or its deletion mutants) in HEK-293 cells did not decrease expression of NCX1 (or its deletion mutants). Coexpression of PLM with WT NCX1 inhibited NCX1 current (I(NaCa)). Deletion of residues 240-679, 265-373, 250-300, or 300-373 from WT NCX1 resulted in loss of inhibition of I(NaCa) by PLM. Inhibition of I(NaCa) by PLM was preserved when residues 229-237, 270-300, 328-330, or 330-373 were deleted from the intracellular loop of NCX1. These results suggest that PLM mediated inhibition of I(NaCa) by interacting with two distinct regions (residues 238-270 and 300-328) of NCX1. Indeed, I(NaCa) measured in mutants lacking residues 238-270, 300-328, or 238-270 + 300-328 was not affected by PLM. Glutathione S-transferase pull-down assays confirmed that PLM bound to fragments corresponding to residues 218-371, 218-320, 218-270, 238-371, and 300-373, but not to fragments encompassing residues 250-300 and 371-508 of NCX1, indicating that residues 218-270 and 300-373 physically associated with PLM. Finally, acute regulation of I(NaCa) by PLM phosphorylation observed with WT NCX1 was absent in 250-300 deletion mutant but preserved in 229-237 deletion mutant. We conclude that PLM mediates its inhibition of NCX1 by interacting with residues 238-270 and 300-328.

Fig. 1.

Heterologous expression of phospholemman (PLM) and Na⁺/Ca²⁺ exchanger (NCX1) loop deletion mutants in HEK-293 cells. Equal amounts of total plasmid DNA were transiently transfected in HEK-293 cells with wild-type (WT) NCX1 or its loop deletion mutants, with or without PLM. Cell lysates from indicated cell cultures were subjected to immunoblot analysis 48 h after transfection using monoclonal anti-NCX1 (R3F1), polyclonal anti-NCX1 [π11-13 (Pi11-13)], monoclonal anti-NCX1 (6H2), polyclonal anti-PLM (C2), and polyclonal anti-phospho-PLM (CP68) antibodies. Under nonreducing conditions (10 mM N-ethylmaleimide) used for R3F1 and π11-13, WT NCX1 is detected as a 160-kDa band. Under reducing conditions (5% β-mercaptoethanol) used for 6H2, WT NCX1 is detected as a 120-kDa band. R3F1 signal intensities of NCX1 and its deletion mutants, with or without PLM coexpression, are summarized in Table 1.CP68 signal intensities of PLM phosphorylated at Ser⁶⁸ in cells cotransfected with PLM and NCX1 or NCX1 deletion mutants are summarized in Table 2.

Fig. 2.

Localization of NCX1 deletion mutants by confocal microscopy in transfected HEK-293 cells. Cells were transiently transfected with control plasmid vector pAdTrack-CMV expressing green fluorescent protein (GFP; A), WT rat NCX1 (B), rΔ300–373 (C), rΔ265–373 (D), dΔ328–330 (E), or dΔ330–373 (F) loop deletion mutants. After 48 h, cells were fixed, permeabilized, labeled with anti-NCX1 monoclonal antibody R3F1, and visualized with Alexa Fluor 546-labeled goat anti-mouse IgG using a Zeiss LSM confocal microscope and ×60 oil objective. GFP expressed under control of a second CMV promoter is localized to the cytosol (A). NCX1 and its loop deletion mutants (red) are largely confined to the plasma membrane. R3F1 signals in images of cells expressing rΔ300–373, rΔ265–373, or dΔ328–330 mutants are weaker than raw signals of R3F1 in cells expressing WT NCX1 or dΔ330–373 mutant, and adjustment of image contrast and intensity was necessary so that the red rim is more clearly visible.

Fig. 3.

Phosphorylation of exogenous PLM in transfected HEK-293 cells. A: cells were transfected with control plasmid vector pAdTrack-CMV, WT PLM, WT rat NCX1, or PLM + NCX1. At 48 h after transfection, cell lysates were subjected to immunoblot analysis using monoclonal antibodies against the catalytic subunit of PKA (PKA_C) and R3F1 against NCX1. Note similar expression of NCX1 whether cells were transfected with NCX1 alone or NCX1 + PLM. B: HEK-293 cells were transfected with WT PLM. After 48 h, they were treated with PBS (control), DMSO, forskolin (10 μM), PMA (0.5 μM), or PMA + forskolin for 10 min at 37°C. Cell lysates were subjected to immunoblot analysis using CP68 antibody to specifically detect PLM phosphorylated at Ser⁶⁸ and C2 antibody to detect predominantly unphosphorylated PLM.

Fig. 4.

Effects of PLM on NCX1 current (I_NaCa) in HEK-293 cells expressing NCX1 or its loop deletion mutants. I_NaCa was measured in HEK-293 cells transfected with pAdTrack-CMV plasmid expressing WT rat NCX1 or various NCX1 loop deletion mutants, with PLM (⧫; n = 8, 5, 5, 8, 11, 6, 6, 8, and 7 for WT rat NCX1, dΔ328–330, dΔ330–373, rΔ300–373, rΔ250–300, rΔ265–373, dΔ240–679, dΔ229–237, and rΔ270–300 deletion mutants, respectively) or without PLM (▴; n = 8, 12, 10, 7, 9, 9, 8, 12, and 4 for WT rat NCX1, dΔ328–330, dΔ330–373, rΔ300–373, rΔ250–300, rΔ265–373, dΔ240–679, dΔ229–237, and rΔ270–300 deletion mutants, respectively). Top left: background currents in HEK-293 cells transfected with control plasmid vector pAdTrack-CMV expressing GFP (•; n = 5). Symbols represent means ± SE; error bars that fall within boundaries of symbols are not shown.

Fig. 5.

Deduction of functional interaction sites between PLM and NCX1. Top: schematic representation of intracellular loop of NCX1 (residues 218–764) containing exchange inhibitory peptide (XIP) region (purple box; residues 219–238), Ca²⁺-binding domain (CBD) 1 (orange box; residues 371–508), CBD 2 (brown box; residues 501–650), and putative interaction site for endogenous XIP (gray striped box; residues 562–679). There is significant overlap between CBD2 and interaction site for endogenous XIP. Two putative interaction sites with PLM are shown as white boxes (residues 238–270 and 300–328). Bottom: various NCX1 loop deletion mutants shown as preserved regions (thick blue line) and deleted segments (dotted black line). Ability of PLM to inhibit I_NaCa when coexpressed with WT NCX1 or its deletion mutants is shown at right. Alignment of deletion mutants indicates that regions encompassing residues 238–270 and 300–328 (white boxes) must be present in the intracellular loop for PLM to exert its modulatory effects on NCX1.

Fig. 6.

Inhibition of I_NaCa by PLM requires regions encompassing residues 238–270 and 300–328 in the intracellular loop of NCX1. I_NaCa was measured in HEK-293 cells expressing NCX1 loop deletion mutants, with PLM (•; n = 9, 4, and 4 for double-deletion rΔ238–270 + Δ300–328, rΔ238–270, and rΔ300–328 mutants, respectively) and without PLM (□; n = 12, 7, and 5 for double-deletion rΔ238–270 + Δ300–328, rΔ238–270, and rΔ300–328 mutants, respectively). Symbols represent means ± SE; error bars that fall within boundaries of symbols are not shown.

Fig. 7.

Glutathione S-transferase (GST) pull-down demonstrates physical association of PLM with residues 218–270 and 300–373, but not residues 250–300, of NCX1. Purified GST or GST-NCX1 fusion proteins linked to GSH-Sepharose beads were incubated with His-tagged PLM (1 μg), and GST pull-down assay was performed. Top: GST and GST-NCX1 fusion proteins detected by Coomassie blue staining. MW, molecular weight marker. Bottom: His-tagged PLM detected by C2 antibody. Experiment was performed 4 times using new sets of GST fusion proteins prepared from 4 independent bacterial cultures, and similar results were obtained.

Fig. 8.

Effects of forskolin on I_NaCa in HEK-293 cells coexpressing PLM and NCX1 or its deletion mutants. HEK-293 cells were transfected with PLM + WT rat NCX1 (n = 5), PLM + dΔ229–237 (n = 5), or PLM + rΔ250–300 (n = 4) loop deletion mutants. I_NaCa was measured before (□) and 5 min after (•) addition of forskolin (1 μM). For comparison, I_NaCa in cells expressing WT rat NCX1 alone (top, ▾; n = 4) is also shown. Symbols represent means ± SE; error bars that fall within boundaries of symbols are not shown. To decrease magnitude of I_NaCa inhibition by PLM (Fig. 4) before forskolin exposure, pAdTrack-CMV vector expressing PLM used in transfection was decreased from 1 to 0.5 μg. Bottom: Western blots demonstrate ∼58% of PLM (detected with C2 antibody) in cells transfected with 0.5 μg of PLM, 1 μg of NCX1, and 1.5 μg of empty vector compared with cells transfected with 1 μg of PLM, 1 μg of NCX1, and 1 μg of empty vector. There were no differences in NCX1 expression between the 2 groups of transfected cells.