ALPK2 prevents cardiac diastolic dysfunction in heart failure with preserved ejection fraction

Abstract

Protein phosphorylation, controlled by protein kinases, is central to regulating various pathophysiological processes, including cardiac systolic function. The dysregulation of protein kinase activity plays a significant role in the pathogenesis of cardiac systolic dysfunction. While cardiac contraction mechanisms are well documented, the mechanisms underlying cardiac diastole remain elusive. This gap persists owing to the historical focus on systolic dysfunction in heart failure research. Recently, heart failure with preserved ejection fraction (HFpEF), an age‐related disease characterized by cardiac diastolic dysfunction, has emerged as a major public health concern. However, its underlying mechanism remains unclear. In this study, we investigated cardiac protein kinases by analyzing the gene expression of 518 protein kinases in human tissues. We identified alpha‐kinase 2 (ALPK2) as a novel cardiac‐specific atypical kinase and generated tamoxifen‐inducible, cardiomyocyte‐specific Alpk2‐knockout mice and Alpk2‐overexpressing mice. Alpk2 deficiency did not affect cardiac systolic dysfunction in the myocardial infarction model or the pressure‐overload‐induced heart failure model. Notably, cardiomyocyte‐specific Alpk2 deficiency exacerbated cardiac diastolic dysfunction induced by aging and in the HFpEF model. Conversely, Alpk2 overexpression increased the phosphorylation of tropomyosin 1, a major regulator that binds myosin to actin, and mitigated cardiac stiffness in HFpEF. This study provides novel evidence that ALPK2 represents a potential therapeutic target for cardiac diastolic dysfunction in HFpEF and age‐related cardiac impairments.

Genetic manipulations of Alpk2 did not affect cardiac systolic dysfunction induced by myocardial infarction or pressure overload. Alpk2 deficiency exacerbated cardiac diastolic dysfunction in response to advanced aging and HFpEF, while Alpk2 overexpression protected against cardiac diastolic dysfunction induced by a high‐fat diet and L‐NAME treatment. Our findings can inform future research and drug development efforts aimed at targeting ALPK2 or related pathways for the treatment of HFpEF and other cardiac conditions.

Abbreviations

CaMK II

calcium/calmodulin‐dependent kinase II

cmc‐Alpk2 KO

tamoxifen‐inducible cardiomyocyte‐specific Alpk2 knockout

dPmax

peak rate of pressure rise

E

mitral inflow velocity

E′

annular velocity

EDPVR

end‐diastolic pressure‐volume relationship

Ees

end‐systolic elastance

HFD

high‐fat diet

HFpEF

heart failure with preserved ejection fraction

HFrEF

heart failure with reduced ejection fraction

KO

knockout

PKA

protein kinase A

PKG

protein kinase G

TAC

transverse aortic constriction

Tau

relaxation time constant

TPM

tropomyosin

UMAP

uniform manifold approximation and projection

WT

wild type

1. INTRODUCTION

Protein phosphorylation by protein kinases is integral to various cellular processes, with approximately 30% of the proteins encoded by the human genome containing covalently bound phosphate. ¹ Phosphorylation sites have been identified in human, murine, and rat tissues with many sites being tissue‐specific. ² , ³ , ⁴ Most proteins with tissue‐specific phosphorylation sites are expressed across multiple tissues, suggesting that many proteins are regulated by phosphorylation independently of protein expression and that tissue‐specific phosphorylation reflects differences in the complex tissue‐specific signaling. ³ To data, a total of 518 members of the human protein kinase family, comprising 478 conventional eukaryotic protein kinases and 40 atypical protein kinases, have been identified. ⁵ While most protein kinases contain a common eukaryotic protein kinase catalytic domain, the physiological and pathological roles of atypical kinases are not well investigated owing to their lack of sequence similarity to the eukaryotic protein kinase domain.

Despite the efficacy of multiple therapies for cardiovascular diseases, the prevalence of chronic heart failure continues to increase drastically in older adults. ⁶ , ⁷ Various clinical classification systems have been used for heart failure, with the most important distinctions being heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF). ⁸ HFrEF typically manifests in the end stages of various cardiovascular diseases, with common causes including myocardial infarction and pressure overload induced by hypertension and valvular heart disease. Inappropriate activation or inactivation of protein kinase A (PKA) and calcium/calmodulin‐dependent kinase (CaMK) II, which phosphorylate key molecules involved in the regulation of cardiac contraction, ⁹ results in contractile dysfunction, contributing to the pathogenesis of HFrEF. ¹⁰ , ¹¹ Therefore, protein kinase‐mediated phosphorylation plays a central role in mediating pathological cardiac ventricular contraction.

Despite extensive knowledge of cardiac contraction, the mechanism underlying cardiac diastole remains under‐investigated, as heart failure has traditionally been associated with the pathogenesis of cardiac systolic dysfunction in HFrEF. Epidemiological features of HFpEF include increasing prevalence with advancing age, female sex, and comorbidities contributing to myocardial stiffness such as metabolic disorders and hypertension. ¹² Notably, diastolic dysfunction is the fundamental component underlying HFpEF pathophysiology. ¹³ Patients with HFpEF exhibit abnormal left ventricular relaxation and increased left ventricular chamber stiffness. ¹⁴ Left ventricular chamber stiffness, caused by alterations in the myocardium and extracellular matrix, represents the key component underlying the pathophysiology of HFpEF. However, the mechanisms underlying this stiffness remain unclear.

Based on these observations, we hypothesized that cardiac‐specific atypical protein kinases mediate pathological pathways in heart failure through cardiac‐specific phosphorylation. In this study, we identified ALPK2 as a cardiac‐specific atypical kinase and subsequently investigated its role of ALPK2 in various murine models of heart failure, given its unknown function.

2. MATERIALS AND METHODS

2.1. Materials and chemicals

Antibodies were used against the following proteins: ALPK2 (H00115701‐M05; Abnova, Taipei, Taiwan), actin (sc‐47778; Santa Cruz Biotechnology, Dallas, TX, USA), Flag (#14793; Cell Signaling Technology, Danvers, MA, USA), GAPDH (#2118; Cell Signaling Technology), Myc (017‐21871; Wako Pure Chemicals, Osaka, Japan), TPM1 (ab109505; Abcam, Cambridge, UK), and Phospho TPM1 (600‐4‐1‐J52; Rockland, Philadelphia, PA, USA).

2.2. Western blotting

Cell and tissue samples were homogenized in radioimmunoprecipitation assay buffer (Wako, Osaka, Japan) with protease inhibitor and phosphatase inhibitor mixtures (Roche, Basel, Switzerland), and the protein concentration was measured using a bicinchoninic acid protein assay (Thermo Fisher Scientific, Waltham, MA, USA). Equal amounts of proteins were loaded onto an SDS‐PAGE gel (Bio‐Rad Laboratories, Hercules, CA, USA) and subsequently transferred onto a PVDF membrane. ECL or ECL plus western blotting detection kits (GE Healthcare, Chicago, IL, USA) were used to detect signals. Relative phosphorylation or protein levels were quantified using ImageJ (National Institutes of Health, Bethesda, MD, USA).

2.3. In situ hybridization assay

A DNA fragment corresponding to sequence positions 5120–5972 of mouse Alpk2 (GenBank accession number NM_001037294.1) was subcloned into the pGEMT‐Easy vector (Promega, Madison, WI, USA) and used for the generation of sense or antisense RNA probes. RNA probes were prepared with the DIG RNA labeling Mix (Roche Diagnostics, Basel, Switzerland). C57BL/6 mouse embryos at E18.5 were dissected, fixed with G‐Fix (Genostaff, Tokyo, Japan), embedded in paraffin on CT‐Pro20 (Genostaff) using G‐Nox (Genostaff) as a less toxic organic solvent for xylene, and sectioned at 8 μm. In situ hybridization was performed with the ISH Reagent Kit (Genostaff) following the manufacturer's instructions. Paraffin‐embedded sections were hybridized with probes at a concentration of 300 ng/mL in G‐Hybo‐L (Genostaff) for 16 h at 60°C. Overnight color reactions were performed with NBT/BCIO solution (Sigma‐Aldrich, St. Louis, MO, USA), and the sections were then washed with phosphate‐buffered saline. The sections were counterstained with Kernechtrot stain solution (Muto Pure Chemicals, Tokyo, Japan) and mounted with G‐Mount (Genostaff).

2.4. In vitro experiments

Adult mouse cardiomyocytes were isolated as previously described. ¹⁵ Male mice (8–10 weeks old) were anesthetized, and the chests were opened to fully expose the heart. The hearts were rapidly removed, cannulated from the aorta with a blunted 24‐gauge needle, and connected to a perfusion apparatus for retrograde perfusion. EDTA buffer and collagenase buffer (type II and IV collagenase: Worthington, Columbus, OH, USA; protease XIV: Sigma‐Aldrich) were perfused via the aortic cannula. The ventricles were cut and gently separated into small pieces with forceps and dissociated through gentle pipetting. Cells were plated in MEM (Thermo Fisher Scientific) containing 2.0 mM L‐glutamine, 10 mM 2,3‐butanedione monoxime, 10 μg/mL insulin, 5.5 μg/mL transferrin, 5.0 ng/mL selenium, and 0.1% bovine serum albumin (BSA) for 1 h. Subsequently, the medium was removed, and adherent cells were resuspended in a BSA‐free culture medium until use.

NIH 3T3 cells (JCRB0615, JCRB Cell Bank) were transfected with 0.2 nmol/L pRP‐CB‐Myc‐human TPM1 (VectorBuilder, Chicago, IL, USA) and 0.15 nmol/L pRP‐CBh‐3xFlag‐human ALPK2 catalytic domain (VectorBuilder) using Lipofectamine 2000 (Thermo Fisher Scientific) following the manufacturer's protocol.

2.5. Mass spectrometry

Protein extraction was performed using the Minute Detergent‐Free Protein Extraction Kit (Invent Biotechnologies Inc., Plymouth, MN, USA) according to the manufacturer's instructions. Phosphatase inhibitor and protease inhibitor cocktail tablets (Roche) were added to the buffer before use. The proteins were digested with trypsin for 16 h at 37°C after reduction and alkylation. The peptides were analyzed via liquid chromatography‐mass spectrometry (LC–MS) using an Orbitrap Fusion mass spectrometry setup (ThermoFisher Scientific) coupled to an UltiMate3000 RSLCnano LC system (Dionex Co., Amsterdam, The Netherlands), a nano HPLC capillary column (150 mm × 75 μm i.d.: Nikkyo Technos Co., Tokyo, Japan), and a nano electrospray ion source. Reversed‐phase chromatography was performed with a linear gradient (0 min, 5% B; 100 min, 40% B) of solvent A (2% acetonitrile with 0.1% formic acid) and solvent B (95% acetonitrile with 0.1% formic acid) at an estimated flow rate of 300 nL/min. A precursor ion scan was conducted using a 400–1600 mass‐to‐charge ratio (m/z) prior to MS/MS analysis. Tandem MS was performed by isolation at 0.8 Th with the quadrupole, HCD fragmentation with a normalized collision energy of 35%, and rapid scan MS analysis in the ion trap. Only precursors with charge states 2–6 were sampled for MS2. The dynamic exclusion duration was set to 15 s with a 10‐ppm tolerance. The instrument was operated in top‐speed mode with 3‐s cycles. Raw data were processed using Proteome Discoverer 1.4 (Thermo Fisher Scientific) in conjunction with the MASCOT search engine, version 2.6.0 (Matrix Science Inc., Boston, MA, USA) for protein identification. Peptides and proteins were identified against the human or mammal protein database in UniProt (release 2020_04), with a precursor mass tolerance of 10 ppm and a fragment ion mass tolerance of 0.8 Da. A fixed modification was set to carbamidomethylation of cysteine, and variable modifications were set to oxidation of methionine. Two missed cleavages by trypsin were allowed.

2.6. Protein kinase gene expression analysis

Human healthy adult tissue RNAs were procured from BioChain (Newark, CA, USA). Tissue gene expression was examined using nCounter® Kinase Ver2 gene expression panels (human) according to the manufacturer's protocol. ALPK2 expression in cardiac cells was analyzed using a publicly available dataset from the Heart Cell Atlas (https://www.heartcellatlas.org). ¹⁶ R (4.1.0) and Seurat (4.3.0) were used for downstream processing. Processed data from single‐cell/single‐nucleus RNA‐sequencing were imported using the anndata package (0.7.5.6). Doublets were identified using scDblFinder (1.6.0) and subsequently removed. ALPK2 gene expression was visualized using Seurat's feature plot function.

2.7. Alpk2 mouse

Alpk2‐floxed mice were developed using the CRISPR/Cas system (Figure S1A) at the Institute of Immunology Co., Ltd. (Tokyo, Japan). First, loxP was inserted downstream of exon 13 to produce an individual with loxP inserted downstream. Subsequently, genome editing was performed to insert loxP between exons 1 and 2 in the fertilized egg prepared using the obtained male individual with loxP inserted downstream. Individuals selected as floxed mouse candidates based on the polymerase chain reaction‐restriction fragment length polymorphism (PCR‐RFLP) results were subjected to Sanger sequencing by GENEWIZ (Azenta Life Sciences, Burlington, MA, USA) for validation and confirmation of the accurate insertion of loxP sites (Figure S1B). The generation of the α‐myosin heavy chain (αMHC)‐CreERT2 mice has been previously reported. ¹⁷ Tamoxifen‐inducible, cardiomyocyte‐specific Alpk2‐KO mice (cmc‐Alpk2 KO [cardiomyocyte‐specific Alpk2 KO]) were generated by intercrossing the αMHC–CreERT2 positive mice with Alpk2 ^flox/flox mice.

Conventional Alpk2 KO (Alpk2 ^KO/KO) mice were developed using the CRISPR/Cas system at the Institute of Immunology Co., Ltd. Guide RNA‐1 and guide RNA‐2 were knocked into the upstream intron of exon 2, and a similar short construct was inserted into downstream of exon 13 (Figure S1C). Guide RNA‐1 and ‐2 were knocked in C57BL6 mouse embryos via electroporation with Alpk2 gRNA and Cas9 protein. Alpk2 ^KO/KO mice were selected via PCR and sequencing analyses. Western blotting confirmed that Alpk2 was undetectable in the hearts of Alpk2 ^KO/KO mice (Figure S1D).

CAG‐Alpk2‐overexpressing mice were developed at the Institute of Immunology Co., Ltd. A CAG‐Alpk2 expression vector was constructed using murine Alpk2 cDNA prepared via gene synthesis (Figure S1E; Azenta Life Sciences). The CAG‐Alpk2 expression vector was microinjected into the embryos, and the embryos were transplanted into pseudo‐pregnant CD‐1 females (The Jackson Laboratory Japan, Inc.). All the obtained individuals were genotyped using tail DNA samples via PCR and genomic Southern blotting. Southern blotting was performed on PCR‐positive individuals. The CAG‐Alpk2 fragment was labeled with ³²P and used as a probe for detecting Alpk2 gene and identifying a 1.9 kb restriction fragment (Figure S1F).

2.8. Animal experiments

Myocardial infarction was induced in 8‐week‐old male cmc‐Alpk2 wildtype (WT) and KO mice. Briefly, after anesthesia using medetomidine, midazolam, and butorphanol at a dose of 0.3, 4.0, and 5.0 mg/kg, i.p., respectively, and as well as intubation, the left anterior descending coronary artery was permanently ligated with an 8–0 nylon suture. Mice were euthanized 7 days after surgery.

Eight‐week‐old male cmc‐Alpk2 WT and KO mice were subjected to transverse aortic constriction (TAC), as previously described. ¹⁷ Briefly, TAC was performed under anesthesia via intraperitoneal administration with intubation. The transverse aorta was constricted between the right innominate and left common carotid arteries using a blunted 24‐gauge needle with an 8–0 suture. Mice were euthanized 14 days after surgery.

The murine HFpEF model was induced by feeding mice with L‐NAME (0.5 g/L in drinking water; Sigma‐Aldrich, St. Louis, MO, USA) and a high‐fat diet (HFD) consisting of 60% kcal from fat (Oriental Yeast, Tokyo, Japan). ¹⁸ Blood pressure in conscious mice was measured noninvasively using the tail‐cuff system (Softron, Tokyo, Japan).

The animals were euthanized, and their organs were harvested. Frozen mouse heart samples were embedded in an optimal cutting temperature compound (Sakura Finetek, Tokyo, Japan).

Sections of the left ventricles were stained with Picrosirius red as standard protocol. The cardiac fibrosis area was assessed using a BZ‐X710 fluorescence microscope (Keyence, Osaka, Japan). Infarct size measurement was obtained by assessing the length of the affected area.

Cardiac systolic and diastolic function were assessed using transthoracic echocardiography and a Vevo 1100 imaging system (FUJIFILM VisualSonics, Toronto, Canada) equipped with an MS400 (18–38 MHz) phased‐array transducer. Fractional shortening (%) was measured from parasternal short‐axis M‐mode scans at the midventricular level using the papillary muscles in conscious mice. Diastolic dysfunction was assessed in anesthetized mice to maintain the heart rate at 410–460 beats/min. Apical four‐chamber views were obtained in anesthetized mice for diastolic function measurements using pulsed‐wave and tissue Doppler imaging at the level of the mitral valve.

Left ventricular pressure–volume measurements were performed using a Millar catheter (SPR 839, Millar Instruments Inc., Houston, TX, USA). A catheter was inserted into the left ventricular cavity via the left ventricular apex. Data were analyzed using ADInstruments LabChart 7 (ADInstruments Pty Ltd., Bella Vista, Australia).

2.9. Statistical analysis

Data are presented as mean ± standard error of mean (SEM) for all experiments. All statistical analyses were performed using GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, USA). Data were analyzed using the unpaired Student's t‐test to compare means between two experimental groups, one‐way analysis of variance (ANOVA) with Tukey's post‐hoc test to compare means among more than two groups, and two‐way ANOVA with Tukey's post‐hoc test or two‐way repeated‐measures ANOVA with Bonferroni post‐hoc testing when there were ≥2 independent variables. In the study, n refers to the number of independent experiments or mice per group. Significance was defined as p < .05. ns, not significant; *p < .05; **p < .01; ***p < .001; ****p < .0001.

2.10. Ethical approval

All procedures in this study were approved by the Ethics Review Board of the Nagoya University School of Medicine (M240112).

3. RESULTS

3.1. ALPK2 is a cardiac‐specific atypical kinase

To investigate the protein kinases expressed in the heart, the gene expression levels of the 40 atypical protein kinases (Figure 1A) and 478 conventional eukaryotic protein kinases (Figure S2A–I) were analyzed in human healthy tissues using the NanoString nCounter gene expression system. ¹⁹ Human hearts expressed various protein kinases; among them, ALPK2 was highly expressed in the heart, but not in other tissues. To further investigate ALPK2 expression in human cardiac cells, publicly available single‐cell RNA‐sequencing data from the Heart Cell Atlas (Source: https://www.heartcellatlas.org/), ¹⁶ which comprises data from heart tissue samples obtained from transplant donors without a history of cardiac disease or arrhythmia, were analyzed. The results revealed that ALPK2 was expressed in both ventricular and atrial cardiomyocytes, but not in other cardiac cell types (Figure 1B). Moreover, in situ hybridization using murine embryos confirmed that the Alpk2 mRNA was exclusively expressed in the heart (Figures 1C and S2J). Western blot analysis indicated that Alpk2 protein was strongly expressed in the murine heart and weakly in the skeletal muscle (Figure 1D).

FIGURE 1.

ALPK2 is a cardiac‐specific atypical kinase. (A) Gene expression analysis of atypical kinases in human tissues. (B) Feature plot of cardiac cells labeled according to ALPK2 transcription expression, visualized using Uniform Manifold Approximation and Projection (UMAP). (C) In situ hybridization analysis of Alpk2 expression in murine fetal hearts. (D) Western blot analysis of ALPK2 protein level in murine tissues (n = 3). (E) Protein extracts from isolated adult cardiomyocytes (cmc) of vesicle‐ or tamoxifen‐treated αMHC‐CreERT2‐positive Alpk2 ^wt/wt or Alpk2 ^flox/flox mice were blotted and probed with antibodies against Alpk2 and actin (loading control). (F) Left ventricular fractional shortening, as determined via echocardiography, before and 7 days after coronary artery ligation (n = 6). (G) Infarct area in the left ventricles at 7 days after coronary artery ligation, as determined via Picrosirius red staining (MI, myocardial infarction; (n = 6). (H, I) Echocardiographic analyses of left ventricular fractional shortening (H), and interventricular septal thickness at end‐diastole (I) were performed before and 1 and 2 weeks after transverse aortic constriction (TAC) surgery (n = 6). Data are presented as mean ± SEM and were analyzed using two‐way analysis of variance (ANOVA) followed by Tukey's post‐hoc test (F, G), or two‐way repeated‐measures ANOVA followed by Bonferroni post‐hoc test (H, I); ns, not significant; ****p < .0001.

To examine the functional significance of ALPK2 in the heart in vivo, mice with tamoxifen‐inducible cardiomyocyte‐specific Alpk2 deficiency (cmc‐Alpk2 KO) were generated by mating αMHC‐Cre‐ERT2 mice with Alpk2 ^flox/flox mice. The efficiency of tamoxifen‐inducible recombination was analyzed via western blotting, and the results confirmed that ALPK2 expression was undetectable in cardiomyocytes isolated from tamoxifen‐treated cmc‐Alpk2 KO mice (Figure 1E). The pathological role of ALPK2 in the heart was examined using murine HFrEF models induced by myocardial infarction or pressure overload. The left coronary artery in cmc‐Alpk2 WT and KO mice was ligated to induce murine models of myocardial infarction, which is a leading cause of morbidity and mortality globally. ²⁰ Picrosirius red staining and echocardiography were performed to assess cardiac functionality 7 days after left coronary artery ligation. The coronary artery ligation significantly decreased left ventricular fractional shortening and increased the infarct area (Figure 1F,G). No differences in cardiac systolic dysfunction induced by the coronary artery ligation were observed between cmc‐Alpk2 WT and KO mice.

To investigate the role of ALPK2 in other HFrEF models, a murine transverse aortic constriction (TAC) model was used to study pressure‐overload‐induced cardiac dysfunction in response to maladaptive cardiac hypertrophy. ¹⁷ No differences were observed in pressure‐overload‐induced cardiac systolic dysfunction between cmc‐Alpk2 WT and KO mice (Figure 1H). In both cmc‐Alpk2 WT and KO mice, TAC resulted in a significant increase in cardiac hypertrophy, as determined by interventricular septal thickness at end‐diastole (Figure 1I). These data indicate that ALPK2, a cardiac‐specific atypical kinase, does not mediate left ventricular systolic function in HFrEF models.

3.2. ALPK2 contributes to aging‐induced cardiac diastolic function

Aging is accompanied by a progressive decline in left ventricular diastolic function in both humans and mice. ²¹ To examine whether the protein expression level of ALPK2 changes with age, its expression was investigated in the hearts of 3‐, 8‐, 24‐, and 78‐week‐old mice (Figure 2A,B). Western blot analysis indicated that ALPK2 protein expression decreased with age. Next, cardiac function was investigated in 8‐ and 78‐week‐old male mice. Body weight and blood pressure equally increased with age (Figure 2C,D), whereas left ventricular fractional shortening equally decreased in cmc‐Alpk2 WT and KO mice (Figure 2E). The ratio of mitral E velocity to mitral annular E′ velocity (E/E′), a predictor of left ventricular end‐diastolic pressure, ²² was elevated in 78‐week‐old mice (Figure 2F). Notably, cardiomyocyte‐specific Alpk2 deficiency exacerbated cardiac diastolic dysfunction related to aging.

FIGURE 2.

Cardiomyocyte‐specific Alpk2 deficiency exacerbates cardiac diastolic dysfunction. (A, B) Endogenous ALPK2 expression in the murine heart was examined via immunoblotting in 3‐, 8‐, 24‐, and 78‐week‐old mice (n = 4). (C–F) Body weight (C), systolic blood pressure (D), fractional shortening (E), and the ratio of mitral E velocity to mitral annular E′ velocity (E/E′) (F) in 8‐ and aged 78‐week‐old cmc‐Alpk2 WT and KO male mice (F). (G) RNA‐sequencing analysis on hearts obtained from donor controls (n = 24), patients with HFrEF (n = 30), and patients with HFpEF (n = 41). (H) Eight‐week‐old cardiomyocyte‐specific Alpk2 ^wt/wt and Alpk2 ^flox/flox male and female mice exposed to a two‐hit pre‐clinical mouse model that resembles human HFpEF, induced by a 5‐week combination of high‐fat diet (HFD) and L‐NAME (constitutive nitric oxide synthase inhibitor). (I, J) Body weight and systolic blood pressure of mice from different experimental groups (n = 9). (K) Percentage of fractional shortening, examined via echocardiography (n = 9). (L) Ratio of left ventricular weight to tibia length. (n = 9). (M) Percentage of fibrosis area in Picrosirius red‐stained transversal sections (n = 9). (O) Representative pulsed‐wave Doppler (top) and tissue Doppler (bottom) tracings (yellow, E wave; red, E′ wave). (P) Ratio between the mitral E wave and the E′ wave (E/E′) (n = 9). Data are presented as the mean ± SEM and analyzed using one‐way ANOVA followed by Tukey's post‐hoc test (B, G) or two‐way ANOVA followed by Tukey's post‐hoc test (C–F, I–M, and P); ns, not significant; *p < .05; **p < .01; ***p < .001; ****p < .0001.

3.3. Cardiomyocyte‐specific Alpk2 deficiency exacerbates HFpEF

To examine ALPK2 expression in patients with heart failure, publicly available RNA‐sequencing data, ²³ which comprise data from heart tissue samples obtained from brain‐dead organ donors (controls, n = 24), HFrEF explanted hearts (n = 30) in patients undergoing transplantation, and endomyocardial biopsy HFpEF heart samples collected from patients with HFpEF (n = 41), were analyzed. The results revealed that ALPK2 expression decreased in patients with HFpEF but not in those with HFrEF (Figure 2G).

To investigate the role of ALPK2 in another model of cardiac diastolic dysfunction, 8‐week‐old male and female cmc‐Alpk2 WT and KO mice were exposed to a two‐hit pre‐clinical murine model that resembles human HFpEF. This model involves unhealthy food consumption, mimicked by a high‐fat diet (HFD), and hypertension induced by L‐NAME, which is a nitric oxide synthase inhibitor ¹⁸ (Figure 2H). The 5‐week exposure to the combination of HFD and L‐NAME equally increased body weight and systolic blood pressure in cmc‐Alpk2 WT and KO mice (Figure 2I,J). No differences were observed in left ventricular fractional shortening (Figure 2K), cardiac hypertrophy (Figure 2L), and cardiac fibrosis (Figure 2M) between cmc‐Alpk2 WT and KO mice. Cardiomyocyte‐specific Alpk2 deficiency significantly increased E/E′ induced by the combination of HFD and L‐NAME (Figure 2O,P).

Subsequently, ventricular pressure–volume loop analysis was performed using a conductance catheter to measure cardiac systolic function (end‐systolic elastance [Ees] and peak rate or pressure rise [dPmax]) and cardiac diastolic function (relaxation time constant [Tau] and the slope of the end‐diastolic pressure–volume relationship [EDPVR]) (Figure S3A,B). ²⁴ Cardiac Alpk2 deficiency enhanced the increase in Tau and EDPVR induced by the combination of HFD and L‐NAME, suggesting that Alpk2 deficiency exacerbates ventricular stiffness in HFpEF.

3.4. ALPK2‐mediated TPM1 phosphorylation in the heart

As ALPK2 is a novel protein kinase that displays little sequence similarity to conventional eukaryotic protein kinases, ²⁵ the substrates phosphorylated by ALPK2 have not been identified. ²⁶ Therefore, to examine ALPK2‐mediated phosphorylation in the heart, mass spectrometry analysis was performed using cardiomyocytes obtained from Alpk2 ^WT/WT or Alpk2 ^KO/KO mice (n = 3 mice/group). The results indicated that Alpk2 deficiency resulted in a decrease in the phosphorylation of 15 peptides in the heart, including tropomyosin 1 (TPM1) (Figure 3A). In the heart, TPM1 regulates actin‐myosin interactions by blocking the interaction of myosin heads with actin. ²⁷ Therefore, to examine whether TPM1 is an ALPK2 substrate, FLAG‐tagged human ALPK2 catalytic domain (FLAG‐tagged ALPK2‐CAT; 1800‐2169 aa) and myc‐tagged human TPM1 (myc‐tagged TPM1) were transfected in NIH3T3 cells (Figure 3B). FLAG‐tagged ALPK2‐CAT phosphorylated myc‐tagged TPM1 at Ser283 in a manner that depends on the expression level of FLAG‐tagged ALPK2‐CAT protein (Figure 3C,D). To assess endogenous TPM1 phosphorylation in murine tissues, immunoblotting analysis was performed (Figure 3E). Although TPM1 was expressed in both the heart and skeletal muscle, TPM1 was mainly phosphorylated in the heart and, to a lesser extent, in the skeletal muscle. In addition, the phosphorylation of TPM1 at Ser283 was examined in cardiomyocytes isolated from Alpk2 ^WT/WT or Alpk2 ^KO/KO mice (Figure 3F,G), revealing that Alpk2 deficiency significantly decreased endogenous TPM1 at Ser283.

FIGURE 3.

ALPK2 mediates TPM1 phosphorylation in cardiomyocytes. (A) Mass spectrometry analysis of cardiomyocytes obtained from Alpk2 ^WT/WT and Alpk2 ^KO/KO mice reveals that 15 phosphorylation sites decreased in the Alpk2‐deficient mice. (B) Structure of ALPK2. (C, D) Myc‐tagged human TPM1 (0.5 μg/well) and FLAG‐tagged human ALPK2 catalytic domain (FLAG‐tagged ALPK2‐CAT; 0, 0.2, or 0.5 μg/well) were transfected in NIH3T3 cells. Statistical evaluation of myc‐tagged TPM1 at Ser283 phosphorylated by FLAG‐tagged ALPK2‐CAT (n = 4). (E) Endogenous TPM1 phosphorylation at Ser283 in murine tissues was examined using immunoblotting analysis. The results are representative of three independent experiments. (F, G) Endogenous TPM1 phosphorylation in isolated murine cardiomyocytes from Alpk2 ^WT/WT and Alpk2 ^KO/KO mice was examined via immunoblotting (n = 3). (H) The ratio of mitral E velocity to mitral annular E′ velocity (E/E′) was examined via echocardiography before and after 2 and 5 weeks of exposure to the HFD and L‐NAME combination (n = 5). (I, J) Endogenous TPM1 phosphorylation in the heart was examined via immunoblotting before and after 2 and 5 weeks of exposure to the HFD and L‐NAME combination (n = 4). (K, L) Endogenous ALPK2 protein expression in the heart was examined using immunoblotting before and after 2 and 5 weeks of exposure to the HFD and L‐NAME combination (n = 4). (M, N) Endogenous TPM1 phosphorylation in the hearts of cmc‐Alpk2‐deficient mice was examined via immunoblotting before and at 2 weeks of exposure to both HFD and L‐NAME (n = 5). Phosphorylated TPM1 in the hearts of cmc‐Alpk2 WT mice unexposed to the combination of HFD and L‐NAME was used as a loading control. Data are represented as mean ± SEM and were analyzed using an unpaired Student's t‐test (G, N) or one‐way ANOVA followed by Tukey's post‐hoc test (D, H, J, and L); ns, not significant; *p < .05; **p < .01; ***p < .001; ****p < .0001.

Next, TPM1 phosphorylation and ALPK2 expression were assessed at 0, 2, and 5 weeks of exposure to both HFD and L‐NAME. Left ventricular diastolic dysfunction occurred after a 5‐week exposure to the combination (Figure 3H). TPM1 phosphorylation and ALPK2 expression in the heart increased after a 2‐week exposure to the combination of HFD and L‐NAME but decreased following 5 weeks of exposure (Figure 3I–L). Moreover, Alpk2 deficiency significantly decreased TPM1 phosphorylation following 2 weeks of exposure to the combination (Figure 3M,N). These data indicate that ALPK2 mediates TPM1 phosphorylation in the heart.

3.5. Alpk2 overexpression prevents cardiac diastolic dysfunction in HFpEF

To examine whether ALPK2‐meditated signaling would attenuate the progression of cardiac diastolic dysfunction induced by aging and the combination of HFD and L‐NAME, Alpk2‐full‐length (FL)‐overexpression mice were generated (Figure 4A,B). Notably, TPM1 phosphorylation at Ser283 in the heart increased with increasing levels of Alpk2 (Figure 4A,C). Therefore, cardiac function was further assessed in 8‐, 24‐, and 78‐week‐old Alpk2‐FL‐overexpression mice. Parameters such as body weight, blood pressure, and left ventricular wall thickness exhibited comparable age‐related increases in both control and Alpk2‐FL‐overexpression mice (Figures 4D,E and S4A). A decline in cardiac systolic function with age was also observed in both groups of mice (Figure 4F). Notably, Alpk2‐FL‐overexpression mice displayed a reduced progression of diastolic dysfunction with age (Figure 4G).

FIGURE 4.

Alpk2 overexpression prevents cardiac diastolic dysfunction. (A–C) Protein extracts from isolated cardiomyocytes of Alpk2 full‐length (FL)‐overexpressing mice were blotted and probed with antibodies against ALPK2 and GAPDH (n = 3). Endogenous TPM1 phosphorylation at Ser283 in isolated murine cardiomyocytes from Alpk2‐overexpressing mice examined via immunoblotting (n = 3). (D–G) Body weight (D), systolic blood pressure (E), percentage of fractional shortening (F), and the ratio of mitral E velocity to mitral annular E′ velocity (E/E′) (G) in 8‐, 24‐, and 78‐week‐old control and Alpk2 FL‐overexpressing mice (line 2) (n = 8). (H–K) Control and Alpk2 FL‐overexpressing mice (line 2) were exposed to a 5‐week combination of HFD and L‐NAME. (H) Ratio between the mitral E wave and E′ wave before and after 5 and 10 weeks of exposure to a combination of high‐fat diet (HFD) and L‐NAME (constitutive nitric oxide synthase inhibitor) (n = 8). (I and J) Cardiac diastolic function (relaxation time constant [Tau: I] and the slope of the end‐diastolic pressure–volume relationship [EDPVR: J]) examined via ventricular pressure–volume loop analysis (n = 6). (K) Ratio of lung weight to tibia length after 10 weeks of exposure to the HFD and L‐NAME combination (n = 6). (L) Schematic diagram showing ALPK2‐mediated cardiac diastolic function. Data are presented as mean ± SEM and were analyzed using one‐way ANOVA followed by Tukey's post‐hoc test (B, C), two‐way ANOVA followed by Tukey's post‐hoc test (I–K), or two‐way repeated‐measures ANOVA followed by Bonferroni post‐hoc test (D–H); ns, not significant; *p < .05; **p < .01; ****p < .0001.

Finally, Alpk2‐FL‐overexpression mice were exposed to a combination of HFD and L‐NAME. No differences were observed in left ventricular fractional shortening in Alpk2‐WT and overexpressing mice (Figure S4B). However, Alpk2‐FL‐overexpression mice developed milder diastolic dysfunction than control mice (Figure 4H). Ventricular pressure–volume loop analysis indicated that Alpk2 overexpression protected mice from further deterioration in cardiac diastolic dysfunction after a 10‐week exposure to both HFD and L‐NAME without affecting cardiac hypertrophy and fibrosis (Figures 4I,J and S4C,D). Moreover, Alpk2 overexpression significantly suppressed the HFD‐ and L‐NAME‐induced increase in lung weight, a highly reliable marker for cardiac dysfunction (Figure 4K). ²⁸ These data suggest that ALPK2 activation enhanced TPM1 phosphorylation in the heart and prevented cardiac diastolic dysfunction in aged and HFpEF murine models.

4. DISCUSSION

Over the past decade, novel therapeutic targets for cardiac diastolic dysfunction have been extensively investigated, as therapies effective against HFrEF have failed to improve outcomes. ¹³ In vitro, cardiomyocyte stiffness is acutely reduced by the activation of protein kinase G (PKG) or PKA through the phosphorylation of titin (TTN), the largest sarcomere protein in humans. ²⁹ However, TTN phosphorylation represents only one of the numerous phosphorylation events potentially contributing to diastolic dysfunction. For example, in a high‐salt diet‐induced Dahl salt‐sensitive rat model of left ventricular diastolic dysfunction, the phosphorylation levels of 529 sites in the heart are altered. ³⁰ However, the extent to which kinases other than PKG and PKA contribute to diastolic dysfunction remains poorly understood. In the present study, our gene expression analysis revealed that numerous kinases are expressed in the heart, including widely expressed kinases such as PRKACA (PKA), PRKG1 (PKG), and CAMK2 ¹⁰ as well as kinases with cardiac‐specific expression patterns such as ALPK2, OBSCN, TTN, and MYLK3. OBSCN and TTN are giant cytoskeletal molecules associated with hypertrophic cardiomyopathy and dilated cardiomyopathy. ³¹ , ³² In addition, Mylk3 deficiency induces cardiac ventricle dilation in zebrafish embryos and mice. ³³ In contrast to these conventional eukaryotic protein kinases with well‐studied cardiac function, little is known about the role of ALPK2 in mammalian heart disease.

The ALPK family consists of three mammalian atypical kinase isoforms: ALPK1, ALPK2, and ALPK3. ²⁶ Our gene expression analysis indicated that Alpk2 is exclusively expressed in the heart, whereas Alpk1 and Alpk3 are expressed in multiple tissues. ALPK1 depletion using siRNA treatment in human colorectal adenocarcinoma cells has been reported to decrease the phosphorylation of myosin I, which regulates the delivery of vesicles to the plasma membrane. ³⁴ Moreover, Alpk3 ^−/− mice manifest a rapidly progressive cardiomyopathy, resulting in premature death, ³⁵ while Alpk2 deficiency leads to severe cardiac defects in zebrafish. ³⁶ However, global Alpk2 KO mice with a 31‐bp deletion in exon 3 or 17‐bp deletion in exon 6 of Alpk2 exhibit normal cardiac systolic function and morphology up to 12 months of age. ³⁷ Our data demonstrate that cardiac‐specific Alpk2 deficiency does not affect physiological cardiac contraction and cardiac systolic dysfunction induced by myocardial infarction and pressure overload. However, we observed that Alpk2 overexpression protects the heart from cardiac diastolic dysfunction induced by advanced aging and a combination of HFD and L‐NAME treatment.

The epidemiology of HFpEF is unique owing to the potent association with the female sex and advanced age in the development of HFpEF. ³⁸ Despite the predominance of women in HFpEF cases, we found no evidence that Alpk2 contributes to sex differences in a murine HFpEF model. Aging results in an increase in the prevalence of left ventricular hypertrophy and a decline in diastolic function. ³⁹ Cardiac diastolic function decreases naturally with age in mice not subjected to HFD and L‐NAME diets. ⁴⁰ In our study, echocardiography analysis revealed no cardiac dysfunction in cardiomyocyte‐specific 8‐week‐old Alpk2‐deficient mice under normal conditions. However, at 78 weeks of age, Alpk2 deficiency resulted in the deterioration of cardiac diastolic dysfunction without affecting cardiac systolic dysfunction. These data suggest that Alpk2 affects the association of advanced age with cardiac diastolic dysfunction.

Myocardial stiffness is influenced by various factors including alterations at the level of cardiomyocytes, actin‐myosin interactions, and the composition of the extracellular matrix. ⁴¹ In the present study, histopathological analyses indicated that neither deficiency nor overexpression of Alpk2 affects cardiac fibrosis. In cardiomyocytes, relaxation is regulated by a complex process that involves intracellular calcium decline, thin filament deactivation, and cross‐bridge detachment. ⁴² Our mass spectrometry analysis on murine hearts detected ALPK2‐mediated phosphorylation of TPM1 at Ser283. In vertebrates, this multigene family comprises four genes (TPM1, 2, 3, and 4), ⁴³ and TPM1 plays a central role in the regulation of cross‐bridge detachment by blocking the interaction of myosin heads with actin in the heart. ²⁷ In vitro, recombinant FLAG‐tagged DAPK1 phosphorylates TPM at Ser283 in HEK293 cells, ⁴⁴ and CK2 holoenzymes increase the phosphorylation of purified TPM1. However, the kinase responsible for this phosphorylation in vivo remains unclear. ⁴⁵ Our findings indicated that cardiomyocyte‐specific Alpk2‐deficient mice exhibited decreased phosphorylation of TPM1 at Ser283 in the heart. Conversely, Alpk2 overexpression in mice resulted in a dose‐dependent increase in cardiac TPM1 phosphorylation, suggesting that ALPK2 mediates the phosphorylation of TPM1 at Ser283 in the heart (Figure 4L). Striated muscle contraction is regulated in a Ca²⁺‐dependent manner through the dynamics of the TPM1 polymer, a multi‐component complex helically wrapped around actin‐containing thin filaments. ⁴⁶ At low Ca²⁺ concentrations, TPM1 sterically blocks the interaction of myosin heads with actin, whereas at high Ca²⁺ concentrations, it shifts to expose myosin‐binding sites, leading to muscle contraction. However, further studies are required to elucidate the mechanisms by which TPM1 phosphorylation mediates cardiac diastolic function.

This study had some limitations. Our single‐cell analysis further demonstrated ALPK2 expression in both atrial and ventricular human cardiomyocytes. However, we primarily investigated the role of ALPK2 in the left ventricle, with the atrium remaining relatively unexplored. Although ALPK2 expression in the heart may change with age or in the context of HFpEF, we did not investigate the mechanism underlying ALPK2 activation. In addition, our mass spectrometry analysis revealed a significant decrease in the phosphorylation of 14 other peptides in Alpk2‐deficient mice. Among them, GLUT4 is a crucial sugar transporter protein in cardiomyocytes, playing a pivotal role in glucose homeostasis. ⁴⁷ Moreover, HK1 is responsible for converting glucose into glucose‐6‐phosphate, thereby maintaining the concentration gradient necessary for the efficient transport of glucose into the cell via glucose transporters. ⁴⁸ Therefore, ALPK2‐mediated phosphorylation of GLUT4 and HK1 may contribute to cardiometabolic abnormalities. RBPMS is an RNA‐binding protein, and its deficiency is associated with cardiovascular developmental defects. ⁴⁹ PLCB1 cleaves phosphatidylinositol 4,5‐bisphosphate to produce inositol 1,4,5‐trisphosphate and diacylglycerol in response to Gαq activation, thus contributing to cardiac hypertrophy. ⁵⁰ Therefore, future studies focusing on the cellular signaling pathways involved in ALPK2 in atrial and ventricular cardiomyocytes are warranted to elucidate the pathological role of ALPK2 in the heart.

In this study, we demonstrated that frenetic manipulations targeting the cardiac atypical kinase Alpk2 do not influence myocardial infarction‐ and pressure‐overload‐induced cardiac systolic dysfunction in vivo. However, Alpk2 deficiency exacerbates advanced aging‐ and HFpEF‐induced cardiac diastolic dysfunction. Conversely, Alpk2 overexpression protects mice from the cardiac diastolic dysfunction induced by exposure to a combination of HFD and L‐NAME with an increase in TPM1 phosphorylation. These results suggest that ALPK2 may provide unique therapeutic targets for HFpEF.

AUTHOR CONTRIBUTIONS

Tatsuya Yoshida performed most of the in vitro and in vivo experiments. Satoya Yoshida, Kohei Inukai, and Tomoki Hattori performed in vivo experiments and mouse generation. Katsuhiro Kato and Kentaro Taki performed in vitro experiments. Atsushi Enomoto, Nina Wettschureck, and Stefan Offermanns supervised the study, discussed data, and commented on the manuscript. Koji Ohashi, Takahiro Okumura, Noriyuki Ouchi, Haruya Kawase, and Toyoaki Murohara supervised the study and commented on the manuscript. Mikito Takefuji initiated the study, performed in vitro and in vivo experiments, analyzed and discussed data, and wrote the manuscript.

DISCLOSURES

The authors declare that no conflict of interest exists.

Supporting information

Data S1.

Yoshida T, Yoshida S, Inukai K, et al. ALPK2 prevents cardiac diastolic dysfunction in heart failure with preserved ejection fraction. The FASEB Journal. 2024;38:e70192. doi: 10.1096/fj.202402103R

DATA AVAILABILITY STATEMENT

All data are available in the main text or the Supporting Information.