Another-regulin regulates cardiomyocyte calcium handling via integration of neuroendocrine signaling with SERCA2a activity

Abstract

Calcium (Ca²⁺) dysregulation is a hallmark feature of cardiovascular disease. Intracellular Ca²⁺ regulation is essential for proper heart function and is controlled by the sarco/endoplasmic reticulum Ca²⁺ ATPase (SERCA2a). Another-regulin (ALN) is a newly discovered cardiomyocyte-expressed SERCA2a inhibitor, suggesting cardiomyocyte Ca²⁺-handling is more complex than previously appreciated. To study the role of ALN in cardiomyocytes, we generated ALN null mice (knockout, KO) and found that cardiomyocytes from these animals displayed enhanced Ca²⁺ cycling and contractility compared to wildtype (WT) mice, indicating enhanced SERCA2a activity. In vitro and in vivo studies show that ALN is post-translationally modified via phosphorylation on Serine 19 (S19), suggesting this contributes to its ability to regulate SERCA2a. Immunoprecipitation and FRET analysis of ALN-WT, phospho-deficient ALN (S19A), or phosphomimetic ALN (S19D) revealed that S19 phosphorylation alters the SERCA2a-ALN interaction, leading to relief of its inhibitory effects. Adeno-associated virus mediated delivery of ALN-WT or phospho-mutant ALN-S19A/D in ALN KO mice showed that cardiomyocyte-specific expression of phospho-deficient ALN-S19A resulted in increased SERCA2a inhibition characterized by reduced rates of cytoplasmic Ca²⁺ clearance compared to ALN-WT and ALN-S19D expressing cells, further supporting a role for this phosphorylation event in controlling SERCA2a-regulation by ALN. Levels of ALN phosphorylation were markedly increased in cardiomyocytes in response to Gα_q agonists (angiotensin II, endothelin-1, phenylephrine) and Gα_q-mediated phosphorylation of ALN translated to increased Ca²⁺ cycling in cardiomyocytes from WT but not ALN KO mice. Collectively, these results indicate that ALN uniquely regulates Ca²⁺ handling in cardiomyocytes via integration of neuroendocrine signaling with SERCA2a activity.

Keywords: Another-regulin; Calcium; Microprotein; Neuroendocrine signaling; SERCA2a.

Fig. 1.. ALN KO mice have normal cardiac structure and function.

(A) To generate ALN KO mice, CRISPR gRNAs were used to target exons 1 and 2 of the mouse ALN gene resulting in a 220 bp deletion and 1 bp insertion. (B) ALN protein expression analyzed in heart homogenates from WT and ALN KO mice by Western blotting. (C) Representative images of histological sections stained with H&E from 12-month-old mice with the indicated genotypes. Scale bar is 1 mm. (D) Heart weight to body weight (HW/BW) ratios of 12-month-old WT and ALN KO mice. Data are expressed as mean ± SD and statistical comparison between groups was performed using an unpaired student’s t-test. (E-I) Serial echocardiography (ECHO) was performed to measure left ventricular (LV) function and structure in mice with the indicated genotypes. (E) Representative para-sternal short-axis view (left) and M-mode tracings (right) from 12-month-old mice. M-mode images were used to quantify ejection fraction (F), fractional shortening (G), and LV posterior wall thickness during diastole (H) and systole (I). Data are expressed as mean ± SD and number of mice analyzed are indicated within each graph. Statistical analysis was performed by 2-way ANOVA with Bonferroni post-hoc analysis. Abbreviations: gRNA, guide RNA; bp, base pair; WT, wildtype; KO, knockout; ECHO, echocardiography; EF, ejection fraction; FS, fractional shortening; LVPW;s, left ventricular posterior wall; systole; LVPW;d, left ventricular posterior wall; diastole.

Fig. 2.. ALN KO cardiomyocytes display an intermediate phenotype in Ca²⁺ handling compared to WT and PLN KO cardiomyocytes.

(A) Representative fractional shortening tracings measured by sarcomere length during pacing-induced cardiomyocyte contraction in cells from 8-week-old mice with the indicated genotypes. (B) Representative Ca²⁺ transient traces recorded in Fura-2 loaded cardiomyocytes from mice with the indicated genotypes. (C, D) Ca²⁺ transients were used to quantify mean peak transient amplitude (C) and cytosolic Ca²⁺ decay rates (D) in cardiomyocytes from mice with the indicated genotypes. Transient decay rates (Tau) were measured by fitting a single exponential to the decay phase of the Ca²⁺ transient. (E) Quantification of cardiomyocyte fractional shortening measured by sarcomere length for the indicated genotypes. Data points in panels A-E represent individual cardiomyocytes from n = 6 mice per genotype. Statistical analysis was performed by One-way ANOVA with Bonferroni post-hoc testing. Abbreviations: WT, wildtype; PLN, phospholamban; ALN/PLN dKO, ALN/PLN double-knockout.

Fig. 3.. The microprotein ALN shares sequence homology with PLN and is phosphorylated on Serine 19.

(A). Sequence alignment of ALN and PLN highlighting their conserved transmembrane domain (blue). Identical residues within the PLN/ALN SERCA2a binding domain are underlined. Validated PLN phosphorylation and candidate phosphorylation sites of ALN are indicated in yellow (PLN) and green (ALN). The ALN phosphorylation site at S19 is indicated with an asterisk. (B) AlphaFold prediction of ALN and PLN structures with phosphorylation sites indicated. AlphaFold model confidence is depicted with the following colors: dark blue, very high; light blue, high; yellow, low. (C) Western blot analysis of protein lysates from AD293 cells expressing HA-ALN-WT separated by Phos-tag (top) or conventional (bottom) SDS-PAGE. Phos-tag SDS-PAGE separates HA-ALN into a phosphorylated species (pALN, top) and a non-phosphorylated species (ALN, bottom). (D) Western blot analysis of protein lysates from AD293 cells transfected with WT HA-ALN or versions of ALN with candidate phosphorylation mutated to alanine separated by Phos-tag SDS-PAGE. The loss of the high molecular weight species of ALN in the ALN-S19A lane indicates Serine 19 is the ALN phosphorylation site. (E) Western blot analysis of protein lysates from AD293 cells expressing HA-ALN-WT, HA-ALN-S19A (mouse) or HA-ALN-S21A (human) separated by Phos-tag (top) or conventional (bottom) SDS-PAGE. Phos-tag SDS-PAGE separates HA-ALN into a phosphorylated species (pALN, top) and a non-phosphorylated species (ALN, bottom). Immunoblots shown in D and E are representative of 3 independent experiments. Abbreviations: HA, hemagglutinin; pALN, phosphorylated ALN; pPLN, phosphorylated PLN; WB, Western blot.

Fig. 4.. ALN S19 phosphorylation regulates its interaction with SERCA2a.

(A) Co-immunoprecipitation experiments performed on protein extracts from AD293 cells expressing WT or phospho-mutant versions of HA-tagged ALN or PLN with Myc-tagged SERCA2a. Pulldowns were performed with an antibody for Myc (SERCA2a). Immunoblots are representative of 3 independent experiments. (B) Average FRET from Cer-SERCA2a to YFP-ALN variants. (C) The dependence of FRET on expression level (WT, gray; S19A, red; S19D, green). FRET increased with protein concentration toward a maximum (FRET_max). (D) FRET_max values for SERCA2a-ALN (WT, S19A, S19D) obtained from hyperbolic fitting of data (as in panel C). (E) Fluorescent lifetime decays for mCyRFP-SERCA2a donor alone (Cntrl) or in the presence of acceptor mMaroon1-ALN (WT, S19A, S19D). In the absence of ALN, mCyRFP-SERCA2a was characterized by a long lifetime. Decays were obtained from cells. (F) Comparison of amplitude-weighted average fluorescence lifetimes. (G) Quantification of a short lifetime component τ₂ obtained from fitting of the decay data with a multiexponential model. (H) Quantification of short lifetime component τ₃. S19D-expressing cells consistently showed larger values of τ₂ and τ₃, consistent with a longer donor-acceptor separation distance for the phosphomimetic mutant. Abbreviations: HA, hemagglutinin; FRET, fluorescence resonance energy transfer; mCer, monomeric cerulean; YFP, yellow fluorescent protein; mCyRFP, monomeric cyan-excitable red fluorescent protein; mMaroon, monomeric maroon.

Fig. 5.. ALN S19 phosphorylation enhances SERCA2a activity.

(A) Schematic depicting MyoAAV constructs for cardiomyocyte-specific expression of ALN WT and phosphomutants or MyoAAV-Empty (control). (B) Immunofluorescence (IF) images of heart sections from 8-week-old ALN KO mice infected with the indicated MyoAAVs. Tissue sections were immunostained with antibodies for HA (green) and PLN (red). The merged channel also include WGA (purple) and a nuclear DAPI stain (blue). Scale bars are 100 100 μm. For additional images, see Fig. S4A. (C) HA-ALN protein expression was analyzed by Western blotting of protein extracts from cardiomyocytes isolated from ALN KO mice infected with the indicated MyoAAVs. (D-F) Mean peak Ca²⁺ transient amplitude (D), cytosolic Ca²⁺ decay rates (E), and cardiomyocyte fractional shortening (F) quantified from pacing induced Ca²⁺ transients and sarcomere shortening measurements from Fura-2 loaded cardiomyocytes isolated from 8-week-old ALN KO mice infected with the indicated MyoAAVs. Data points represent individual cardiomyocytes from n = 5 mice. One-way ANOVA with Bonferroni post-hoc testing was used for statistical analysis. (G) Co-immunoprecipitation experiments performed on protein extracts from isolated cardiomyocytes from ALN KO mice infected with the indicated MyoAAVs. Pulldowns were performed with an antibody for HA to IP HA-ALN proteins. Western blotting was performed on eluted proteins from HA IPs (top) and total protein extracts (input, bottom) with the indicated antibodies. Abbreviations: AAV, adeno associated virus; IP, intraperitoneal; P7, postnatal day 7; DAPI, 4′,6-diamidino-2-phenylindole; HA IP, HA immunoprecipitation.

Fig. 6.. Gα_q/PKC activation increases pALN S19 levels in isolated cardiomyocytes.

(A) Isolated cardiomyocytes from 8-week-old WT mice were stimulated with 1nM Iso or 1μM AngII and phosphorylated ALN and PLN levels were analyzed via Western blotting with antibodies for phospho-ALN (pALN-S19) and phospho-PLN (pPLN-S16). Total ALN and PLN (tALN, tPLN) protein levels were also analyzed. Quantification is shown on the right. (B) Schematic of the Gα_q pathway with relevant agonists, drugs, and signaling molecules indicated. (C) Adult cardiomyocytes from WT mice were stimulated with Gα_q agonists (AngII, 1μM; ET-1, 10nM; PE, 10μM) with or without pre-treatment with a pan-PKC inhibitor (Go6983, 2μM) and protein extracts were analyzed and quantified by Western blotting using antibodies for pALN S19 and total ALN. A Gα_s agonist (Iso, 1nM) was used as a negative control. (D) Gα_q protein expression analyzed in isolated cardiomyocytes from WT and cardiomyocyte-specific Gα_q KO mice by Western blotting. (E) Western blot analysis and quantification of pALN S19 levels in isolated cardiomyocytes from cardiomyocyte-specific Gα_q KO mice stimulated with Gα_q agonists (AngII, 1μM; ET-1, 10nM; PE, 10μM) or a PKC activator (PMA, 1μM) compared to unstimulated WT cardiomyocytes. (F) Western blot analysis and quantification of pALN S19 levels in isolated cardiomyocytes from cardiomyocyte-specific Gα_q overexpression mice (GNAQ^Q209L) stimulated with Gα_q agonists (AngII, 1μM; ET-1, 10nM; PE, 10μM) and a PKC activator (PMA, 1μM) compared to unstimulated WT cardiomyocytes. For all Western blots shown in this Figure, images shown are representative of n=5 (A) or n=3 (C-F) independent experiments. Abbreviations: Iso, isoproterenol; AngII, angiotensin II; ET-1, endothelin-1; PE, phenylephrine; PMA, Phorbol 12-myristate 13-acetate (PKC activator); PLC, phospholipase C; DAG, diacyl glycerol; PKC, protein kinase C; Gα_q KO, Gna11^−/−;Gnaq^fl/fl;αMHC-Cre; GNAQ^Q209L, Rosa26-floxed stop-GNAQ^Q209L+/−;αMHC-Cre.

Fig. 7.. Phosphorylation of ALN via Gα_q stimulation increases cardiomyocyte Ca²⁺ transient and contractility properties.

(A) Schematic of experimental design and methods used for quantification. (B-E) Pacing-induced Ca²⁺ transients and cardiomyocyte contractility analyzed in Fura-2 loaded cardiomyocytes isolated from 8-week-old mice with the indicated genotypes. Cytosolic Ca²⁺ decay rates (B, D, Tau) and fractional shortening measured by sarcomere length (C, E) were quantified before and after treatment with Gα_s or Gα_q agonists. Each datapoint represents a single cardiomyocyte at baseline and after agonist stimulation. 400 nM Iso was used as the Gα_s agonist in panels B and C, and 7.5μM AngII was used as the Gα_q agonist in panels D and E. Data points represent individual cardiomyocytes from n = 6 mice. 2-way ANOVA with Bonferroni post-hoc testing was used for statistical analysis. Abbreviations: WT, wildtype; ALN/PLN dKO, ALN/PLN double-knockout.

Fig. 8.. Working model.

Neuroendocrine stimuli activate Gα_q-coupled receptors on the cell membrane which initiates the signaling cascade resulting in PKC phosphorylation of ALN, its dissociation from SERCA2a, and increased Ca²⁺ cycling in cardiomyocytes (Left). In contrast, β-adrenergic agonists activate Gα_s-coupled receptors which elevate cAMP and induce PKA to phosphorylate PLN, resulting in SERCA2a disinhibition and increased Ca²⁺ cycling (Right). Abbreviations: PLC, phospholipase C; DAG, diacyl glycerol; PKC, protein kinase C; AC, adenylate cyclase; cAMP, cyclic adenosine monophosphate; PKA, protein kinase A.