Cardiomyocyte stromal interaction molecule 1 is a key regulator of Ca2+ -dependent kinase and phosphatase activity in the mouse heart

Abstract

Stromal interaction molecule 1 (STIM1) is a major regulator of store-operated calcium entry in non-excitable cells. Recent studies have suggested that STIM1 plays a role in pathological hypertrophy; however, the physiological role of STIM1 in the heart is not well understood. We have shown that mice with a cardiomyocyte deletion of STIM1 (^cr STIM1^-/- ) develop ER stress, mitochondrial, and metabolic abnormalities, and dilated cardiomyopathy. However, the specific signaling pathways and kinases regulated by STIM1 are largely unknown. Therefore, we used a discovery-based kinomics approach to identify kinases differentially regulated by STIM1. Twelve-week male control and ^cr STIM1^-/- mice were injected with saline or phenylephrine (PE, 15 mg/kg, s.c, 15 min), and hearts obtained for analysis of the Serine/threonine kinome. Primary analysis was performed using BioNavigator 6.0 (PamGene), using scoring from the Kinexus PhosphoNET database and GeneGo network modeling, and confirmed using standard immunoblotting. Kinomics revealed significantly lower PKG and protein kinase C (PKC) signaling in the hearts of the ^cr STIM1^-/- in comparison to control hearts, confirmed by immunoblotting for the calcium-dependent PKC isoform PKCα and its downstream target MARCKS. Similar reductions in ^cr STIM1^-/- hearts were found for the kinases: MEK1/2, AMPK, and PDPK1, and in the activity of the Ca²⁺ -dependent phosphatase, calcineurin. Electrocardiogram analysis also revealed that ^cr STIM1^-/- mice have significantly lower HR and prolonged QT interval. In conclusion, we have shown several calcium-dependent kinases and phosphatases are regulated by STIM1 in the adult mouse heart. This has important implications in understanding how STIM1 contributes to the regulation of cardiac physiology and pathophysiology.

Keywords: calcium-dependent; cardiomyocytes; kinases; protein kinase C (PKC); protein kinase G (PKG); store-operated calcium entry (SOCE); stromal interaction molecule 1 (STIM1).

FIGURE 1

Serine/threonine kinome of control and ^crSTIM1^−/− mice with saline treatment. Whole‐cell lysates of 2 μg from hearts extracted from control and ^crSTIM1^−/− mice subjected to subcutaneous in vivo saline treatment were loaded onto a serine/threonine kinome chip. Kinetic phosphorylation was visualized using BioNavigator 6.0 (PamGene) with additional analysis utilizing the PhosphoNET database and MetaCore. (a) Kinetic phosphorylation (x‐axis per cell) over time (y‐axis) of altered kinases between control and ^crSTIM1^−/− in response to saline treatment for selected peptides is displayed with control hearts (blue) overlaid with KO hearts (green); (b) Bar graph of kinases identified as reduced in ^crSTIM1^−/− versus control using the UpKin PamApp (BioNavigator) based on kinase scores from PhosphoNET. Kinases are scored on the left y‐axis by mean kinase statistic (k‐statistic) and on the right y‐axis by mean specificity score; (c) MetaCore network mapping of KO altered kinases (PKG isoforms) at baseline to an overlaid STIM1 node. A Dark circle inset to the upper right of a protein indicates decreased signal in KO. Arrowheads denote literature annotated directions of interactions with the type of interaction indicated by arrow color (green‐positive; red‐negative). Legend is located on the right‐hand side of the figure panel. STIM1, stromal interaction molecule 1

FIGURE 2

Serine/threonine kinome of control and ^crSTIM1^−/− mice after 15 min phenylephrine (PE) treatment. Whole‐cell lysates of 2 μg from hearts extracted from control and ^crSTIM1^−/− mice subjected to subcutaneous in vivo PE treatment (15 mg/kg) were loaded onto a serine/threonine kinome chip. Kinetic phosphorylation was visualized using BioNavigator 6.0 (PamGene) with additional analysis utilizing the PhosphoNET database and MetaCore. (a) Top 75 differentially regulated (upregulated and downregulated) peptide targets identified in control and ^crSTIM1^−/− hearts treated with saline and PE were arranged into a heatmap using MetaboAnalyst software; (b) Kinetic phosphorylation (x‐axis per cell) over time (y‐axis) of altered kinases between control and ^crSTIM1^−/− in response to PE. Green arrows highlight the four selected peptides that were differentially altered by PE in KO hearts; (c) Bar graph of altered kinases in control hearts treated with PE upstream of array peptides; (d) Bar graph of altered kinases in ^crSTIM1^−/− hearts treated with PE upstream of array peptides. Kinase prediction is performed using the UpKin PamApp (BioNavigator) based on kinase scores from PhosphoNET. Kinases are scored on the left y‐axis by mean kinase statistic (k‐statistic) and on the right y‐axis by mean specificity score; (e) Network mapping of KO altered kinases in response to PE treatment (PKC isoforms) to an overlaid STIM1 node. Network mapping of identified upstream kinases and their relation to STIM1. Dark circle inset to the upper right of a protein indicates decreased signal in KO. Arrowheads denote literature annotated directions of interactions with the type of interaction indicated by arrow color (green‐positive; red‐negative). Legend is located on the right‐hand side of the figure panel. STIM1, stromal interaction molecule 1

FIGURE 3

Decreased protein kinase C (PKC) isoforms in ^crSTIM1^−/− hearts. Immunoblots and densitometric analysis of PKCα (a) and MARCKS (b) phosphorylation in hearts isolated from control and ^crSTIM1^−/− mice treated subcutaneously with either 15 min saline or phenylephrine. Protein expression is normalized to total protein and/or GAPDH. STIM1, stromal interaction molecule 1. *p < 0.05, **p < 0.01, versus age‐matched control (ANOVA/Tukey's post hoc), n = 3 per genotype. STIM1, stromal interaction molecule 1

FIGURE 4

Decreased metabolic and growth kinases in ^crSTIM1^−/− hearts. Immunoblots and densitometric analysis of AMPK (a), PDPK1 (b), ERK1/2 (c), and MEK1/2 (d), phosphorylation in hearts isolated from control and ^crSTIM1^−/− mice treated subcutaneously with 15 min saline or phenylephrine. Protein expression is normalized to total protein and/or GAPDH. STIM1, stromal interaction molecule 1. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, versus age‐matched control (ANOVA/ Tukey's post hoc), n = 3 per genotype

FIGURE 5

Reduced calcineurin activity in ^crSTIM1^−/− hearts. (a) Immunoblots and densitometric analysis of calcineurin expression in hearts isolated from control and ^crSTIM1^−/− mice treated with 15 min saline or phenylephrine (PE). Protein expression is normalized to GAPDH. p > 0.05, versus age‐matched control (ANOVA, Tukey's post hoc), n = 3 per genotype. Calcineurin and phosphatase activity was assayed in ^crSTIM1^−/− and control heart lysates under saline and PE treatments showing the following parameters: (b) calcineurin activity (PP2B); (c) total phosphate released; (d) non‐calcineurin phosphatase activity (PP1, PP2A, and PP2C); and (e) PP2C phosphatase activity. STIM1, stromal interaction molecule 1. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, versus age‐matched control (ANOVA/Tukey's post hoc), n = 6 per genotype

FIGURE 6

Electrocardiogram parameters in ^crSTIM1^−/− hearts in response to saline and phenylephrine (PE) treatment. Analysis of electrocardiogram (ECG) parameters in control and ^crSTIM1^−/− hearts treated with 15 min saline (baseline) or PE. (a) Heart rate; (b) QT interval; (c) P wave duration; (d) PR interval; and (e) QRS interval. STIM1, stromal interaction molecule 1. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, versus age‐matched control (ANOVA/ Tukey's post hoc), n = 3–6 per genotype