Canagliflozin restores electrical properties and suppresses structural damage in isoproterenol-induced myocardial injury models Authors: Lisa Kitasato, Yusuke Murayama, Naruya Ishizue, Shuhei Kobayashi, Jun Oikawa, Hidehira Fukaya, Shinichi Niwano, Junya Ako Published: 2025 ABSTRACT Canagliflozin (Cana), a sodium-glucose cotransporter 2 (SGLT2) inhibitor with additional affinity for SGLT1, has demonstrated cardioprotective effects. However, its role in modulating ventricular electrophysiology and structural remodeling under stress conditions remains insufficiently characterized. We have previously reported electrical and structural remodeling in a rat model of isoproterenol (ISP)-induced myocardial injury. In this study, we investigated the protective effects of Cana and its underlying mechanism in an ISP-induced myocardial injury model using Sprague Dawley rat hearts and in neonatal rat ventricular cardiomyocytes (NRVCMs). Our data revealed that Cana significantly suppressed ISP-induced prolongation of electrophysiological parameters, reduced oxidative stress, preserved mitochondrial function, and elevated ketone body levels. In NRVCMs, Cana -rather than β-hydroxybutyrate (βOHB)- improved mitochondrial integrity, attenuated apoptosis, and downregulated SGLT1, sodium–calcium exchanger 1 (NCX1), and sodium–hydrogen exchanger 1 (NHE1) expression. These effects were at least partly abolished by phosphoinositide 3-kinase (PI3K)/ protein kinase B (Akt) inhibition. Collectively, these findings suggest that Cana modulates electrical remodeling and exerts cardioprotective effects by enhancing mitochondrial function and regulating calcium transport. These results underscore the therapeutic potential of Cana in cardiovascular diseases.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-29126-8. Introduction Canagliflozin (Cana) is a diabetes medication that improves hyperglycemia through inhibition of sodium-glucose cotransporter 2 (SGLT2), which is involved in glucose and sodium reabsorption in the renal proximal convoluted tubules. In addition to SGLT2 inhibition, Cana also blocks sodium-glucose cotransporter 1 (SGLT1), which is expressed in the small intestine, proximal tubule, and myocardium, albeit with lower affinity1. SGLT1 is crucial for glucose absorption in the intestines and kidneys, and its inhibition contributes to reduced postprandial glucose levels. This dual action of Cana may provide additional therapeutic benefits beyond those offered by SGLT2 inhibition alone. Recently, the cardioprotective effects of Cana have attracted increased attention. Large clinical trials have demonstrated that Cana reduces cardiovascular events in heart failure patients, regardless of ejection fraction and diabetes status2,3. Although various mechanisms have been proposed for the cardiovascular benefits of Cana, including improvement of infarct size4, cardiac metabolism5, and diastolic function6, no preclinical or clinical evidence is available regarding the potential effects and mechanisms of SGLT2 inhibitors on ventricular electrophysiological characteristics and structural changes in myocardial injury. Changes in ventricular electrophysiology refer to alterations in the electrical properties of cardiac cells, which are often observed in heart failure and contribute to abnormal electrical conduction and arrhythmias7. This process involves alterations in ion channel function, resulting in changes in action potential duration and refractoriness. These alterations exacerbate heart failure by impairing the ability of the heart to maintain a normal rhythm and efficient contraction. Structural remodeling, often occurring alongside electrical remodeling, includes fibrosis and myocyte loss, further disrupting heart function8. The isoproterenol (ISP)-induced myocardial injury model in rats9 is a reliable and reproducible model associated with arrhythmias, myocyte loss, and fibrosis, ultimately progressing to heart failure and myocardial injury10. We have previously reported on the electrical and structural remodeling in ISP-induced myocardial injury model rats, characterized by endocardium-dominant myocardial injury, prolongation of the ventricular action potential duration (VAPD) and ventricular effective refractory period (VERP), and a marked increase in oxidative stress. These pathological markers were improved by treatment with linagliptin, a dipeptidyl peptidase 4 (DPP-4) inhibitor, by reducing oxidative stress11. Additionally, we demonstrated that Cana mitigated atrial fibrillation and suppressed oxidative stress mediated by ketone bodies in a canine atrial fibrillation model12. However, the potential of Cana to influence ventricular remodeling in the context of myocardial injury remains unexplored. Building on these findings, we aimed to investigate the effects of Cana on ventricular electrophysiological and structural changes using an ISP-induced myocardial injury rat model in both in vivo and in vitro settings. Results Cana pretreatment reduced ISP-induced oxidative stress and elevated ketone body levels in rats Previous reports have shown that subcutaneous injection of ISP in rats induces myocardial injury within 1 week13,14, while other studies have reported that minimal histological changes, including myocyte necrosis and inflammatory responses, occur within the first 3 to 24 h after isoproterenol injection15. We therefore analyzed the effect of Cana in rats injected with ISP at 1 day (Acute phase) and 1 week (Sub-Acute phase) (Fig. 1). Fig. 1Schematic representation of the in vivo experimental protocol Sprague-Dawley rats were assigned to four groups: Sham group (n = 9): Rats were injected with saline (1 mL) on day 0 and received oral buffer water (1 mL/kg/day) via gastric tube starting 7 days prior to the saline injection (day − 7) and continuing until day 7 (total of 14 days). ISP group (n = 11): Rats received a subcutaneous ISP injection (10 mg/kg) on day 0 and were administered buffer water identically to the sham group. ISP + Cana group (n = 15): Rats received an ISP injection on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Cana group (n = 3): Rats were injected with saline (1 mL) on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Day 1 was defined as the Acute phase, during which d-ROMs and ketone body levels were measured. Day 7 was defined as the Sub-Acute phase, during which the same parameters were re-evaluated. Additionally, electrophysiological studies, Azan staining, and TUNEL assays were performed on day 7. * Sprague-Dawley rat model; † 0.5% hydroxypropyl methylcellulose (HPMC); Cana, canagliflozin; ISP, isoproterenol; EPS, electrophysiological studies. Schematic representation of the in vivo experimental protocol Sprague-Dawley rats were assigned to four groups: Sham group (n = 9): Rats were injected with saline (1 mL) on day 0 and received oral buffer water (1 mL/kg/day) via gastric tube starting 7 days prior to the saline injection (day − 7) and continuing until day 7 (total of 14 days). ISP group (n = 11): Rats received a subcutaneous ISP injection (10 mg/kg) on day 0 and were administered buffer water identically to the sham group. ISP + Cana group (n = 15): Rats received an ISP injection on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Cana group (n = 3): Rats were injected with saline (1 mL) on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Day 1 was defined as the Acute phase, during which d-ROMs and ketone body levels were measured. Day 7 was defined as the Sub-Acute phase, during which the same parameters were re-evaluated. Additionally, electrophysiological studies, Azan staining, and TUNEL assays were performed on day 7. * Sprague-Dawley rat model; † 0.5% hydroxypropyl methylcellulose (HPMC); Cana, canagliflozin; ISP, isoproterenol; EPS, electrophysiological studies. In the Acute phase, although the administration of ISP increased the heart/body weight ratio and decreased the serum blood sugar level compared with those in the Sham group, these metrics did not significantly differ between the ISP and ISP + Cana groups (Supplemental Table S1). The levels of derivatives of reactive-oxygen metabolites (d-ROMs), which are markers of oxidative stress, increased in response to ISP stimulation but significantly decreased in response to Cana pretreatment (Fig. 2A). A consistent effect of SGLT2 inhibitors in patients is an increase in plasma ketone body levels through the enhancement of hepatic ketogenesis, which contributes to improved energy metabolism16,17. Interestingly, in our rat model, the serum levels of total ketone bodies, acetoacetate, and β-hydroxybutyrate significantly increased in the ISP + Cana group compared with those in the group treated only with ISP (Fig. 2B-D). To clarify whether this elevation was due to myocardial injury or Cana treatment alone, we added a Cana-only group without ISP. Notably, compared with the control treatment, the Cana treatment alone did not significantly increase serum ketone levels, whereas compared with all the other groups, the ISP + Cana group exhibited marked increases in both β-hydroxybutyrate and acetoacetate levels. These findings support the hypothesis that the increase in ketone bodies is potentiated by myocardial injury and is not attributable to Cana alone. Fig. 2Cana Pretreatment Mitigated ISP-Induced Oxidative Stress and Enhanced Ketone Bodies in Rats during the Acute-Phase Serum levels of (A) d-ROMs. Data represent measurements from Sham (n = 6), ISP (n = 8), and ISP + Cana (n = 15) groups. Serum levels of (B) Total Ketone bodies, (C) Acetoacetic acid, and (D) β-hydroxybutyric acid levels in rats during the Acute phase following ISP stimulation. Data represent measurements from Sham (n = 9), Cana (n = 3), ISP (n = 6), and ISP + Cana (n = 8) groups. Bar graphs show mean +/-SE. *P < 0.05. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. Cana, canagliflozin; ISP, isoproterenol. Cana Pretreatment Mitigated ISP-Induced Oxidative Stress and Enhanced Ketone Bodies in Rats during the Acute-Phase Serum levels of (A) d-ROMs. Data represent measurements from Sham (n = 6), ISP (n = 8), and ISP + Cana (n = 15) groups. Serum levels of (B) Total Ketone bodies, (C) Acetoacetic acid, and (D) β-hydroxybutyric acid levels in rats during the Acute phase following ISP stimulation. Data represent measurements from Sham (n = 9), Cana (n = 3), ISP (n = 6), and ISP + Cana (n = 8) groups. Bar graphs show mean +/-SE. *P < 0.05. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. Cana, canagliflozin; ISP, isoproterenol. Cana improved electrophysiological changes and reduces the apoptosis induced by ISP in rats After 7 days of ISP injection (Sub-Acute phase), the heart weight/body weight ratio significantly decreased in the Cana-treated group compared to the ISP group (3.5 ± 0.4 vs. 3.0 ± 0.2, Supplemental Table S1), suggesting an improvement of cardiac hypertrophy18. The electrophysiological parameters are shown in Fig. 3. Representative electrograms: VERP, monophasic potential duration (MAPD) 20 and MAPD 90 with BCL140 and BCL 120 were markedly prolonged in the ISP group in comparison with the Sham group, which were considered as the result of ISP-induced myocardial injury. In contrast, this prolongation was relatively suppressed in the ISP + Cana group, with the parameters being shorter than those in the ISP group. Fig. 3Cana Improved ISP-induced Electrophysiological Alterations in Rats during the Sub-Acute phase A, B: VERP was measured after 8 basic stimuli with cycle lengths of 140 ms (A) and 120 ms (B). Data represent measurements from the Sham (n = 3), ISP (n = 11), and ISP + Cana (n = 7) groups. C–F: MAPs were recorded from the last beat of 8 consecutive beats with fixed cycle lengths of 140 ms (C, E) and 120 ms (D, F). The MAP duration (MAPD) was determined as the interval from the MAP onset to 20% (MAPD20; C, D) and 90% (MAPD90; E, F) repolarization. Data represent measurements from the Sham (n = 3), ISP (n = 5), and ISP + Cana (n = 7) groups. Bar graphs show mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. VERP, ventricular effective refractory period; MAP, monophasic action potential; Cana, canagliflozin; ISP, isoproterenol. Cana Improved ISP-induced Electrophysiological Alterations in Rats during the Sub-Acute phase A, B: VERP was measured after 8 basic stimuli with cycle lengths of 140 ms (A) and 120 ms (B). Data represent measurements from the Sham (n = 3), ISP (n = 11), and ISP + Cana (n = 7) groups. C–F: MAPs were recorded from the last beat of 8 consecutive beats with fixed cycle lengths of 140 ms (C, E) and 120 ms (D, F). The MAP duration (MAPD) was determined as the interval from the MAP onset to 20% (MAPD20; C, D) and 90% (MAPD90; E, F) repolarization. Data represent measurements from the Sham (n = 3), ISP (n = 5), and ISP + Cana (n = 7) groups. Bar graphs show mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. VERP, ventricular effective refractory period; MAP, monophasic action potential; Cana, canagliflozin; ISP, isoproterenol. Cardiac fibrosis is known to disrupt normal electrical pathways, create heterogeneous conduction, and increase the risk of arrhythmias, contributing to electrical remodeling19,20. However, in our study, although Azan-positive staining tended to increase in the ISP group, there was no significant difference in the fibrotic index compared to the ISP + Cana group (Fig. 4A, B). Consequently, we focused on myocardial apoptosis, as cardiomyocyte apoptosis can contribute to electrical remodeling by reducing the number of viable cells, promoting interstitial fibrosis, and altering ion channel expression, which impacts the myocardium’s electrophysiological properties21. The TUNEL staining demonstrated that Cana significantly attenuated ISP-induced cardiomyocyte apoptosis (ISP vs. ISP + Cana, p < 0.01), reducing the apoptosis index to levels that were not significantly different from the Sham group (p = 0.08), although a slight residual increase was observed (Fig. 4C, D). Fig. 4Cana Reduced ISP-induced Myocardial Apoptosis in Rats during the Sub-Acute Phase A. Representative Azan staining images obtained from heart tissue sections of ISP-injected rats. Scale bar = 100 μm. B. Quantification of the Azan-positive fibrotic area in the Sham (n = 2), ISP (n = 3), ISP + Cana (n = 3) groups. C. TUNEL assay showing apoptotic myocardium in ISP-injected rats. Double immunofluorescence staining was performed for sarcomeric actin (red), TUNEL-positive cells (green), and DAPI nuclear counterstaining (blue). Scale bar = 100 μm. D. Quantification of TUNEL-positive cells (n = 4 per group). Bar graphs show mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. H.E, hematoxylin-eosin; Cana, canagliflozin; ISP, isoproterenol. Cana Reduced ISP-induced Myocardial Apoptosis in Rats during the Sub-Acute Phase A. Representative Azan staining images obtained from heart tissue sections of ISP-injected rats. Scale bar = 100 μm. B. Quantification of the Azan-positive fibrotic area in the Sham (n = 2), ISP (n = 3), ISP + Cana (n = 3) groups. C. TUNEL assay showing apoptotic myocardium in ISP-injected rats. Double immunofluorescence staining was performed for sarcomeric actin (red), TUNEL-positive cells (green), and DAPI nuclear counterstaining (blue). Scale bar = 100 μm. D. Quantification of TUNEL-positive cells (n = 4 per group). Bar graphs show mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. H.E, hematoxylin-eosin; Cana, canagliflozin; ISP, isoproterenol. In the Sub-Acute phase, no significant difference in d-ROMs, serum total ketone bodies levels or ketone body components was detected (Supplemental Fig. S1). In summary, in our rat model, we demonstrated that Cana protects against ISP-induced electrical and physiological changes, including cardiac hypertrophy and apoptosis. These effects were accompanied by a reduction in oxidative stress and an elevation of ketone bodies in the Acute phase. Cana pretreatment improved ISP-induced cardiomyocyte injury To further investigate the underlying mechanism of the protective effect of Cana on cardiomyocyte damage, we stimulated neonatal rat ventricular cardiomyocytes (NRVCMs) with ISP. A TUNEL assay revealed that the apoptotic rate of NRVCMs was significantly lower after treatment with Cana than after treatment with ISP alone (Fig. 5A, B). Fig. 5Cana reduced apoptosis and increased viability in ISP-stimulated cardiomyocytes A. TUNEL assay showing apoptotic cardiomyocytes after 48 h of ISP stimulation. Double immunofluorescence staining was performed for sarcomeric actin (red), TUNEL-positive cells (green), and DAPI nuclear counterstaining (blue). Scale bar = 100 μm B. Quantification of TUNEL-positive cells (n = 4 per culture). C. Cardiomyocyte viability in 24 h after ISP stimulation with or without dose-dependent Cana pretreatment (n = 5 per culture). D. Cardiomyocyte viability after 24 h of ISP stimulation with or without pretreatment with Empa, Cana and/or βOHB. The effect of LY29002 on PI3K inhibition was also evaluated (n = 3 for Empa, n = 5 for other conditions, per culture). Bar graphs represent mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Dunnet post-hoc test. Cana, canagliflozin; ISP, isoproterenol; βOHB, β-hydroxybutyrate; PI3K, phosphatidylinositol 3-kinase. Cana reduced apoptosis and increased viability in ISP-stimulated cardiomyocytes A. TUNEL assay showing apoptotic cardiomyocytes after 48 h of ISP stimulation. Double immunofluorescence staining was performed for sarcomeric actin (red), TUNEL-positive cells (green), and DAPI nuclear counterstaining (blue). Scale bar = 100 μm B. Quantification of TUNEL-positive cells (n = 4 per culture). C. Cardiomyocyte viability in 24 h after ISP stimulation with or without dose-dependent Cana pretreatment (n = 5 per culture). D. Cardiomyocyte viability after 24 h of ISP stimulation with or without pretreatment with Empa, Cana and/or βOHB. The effect of LY29002 on PI3K inhibition was also evaluated (n = 3 for Empa, n = 5 for other conditions, per culture). Bar graphs represent mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Dunnet post-hoc test. Cana, canagliflozin; ISP, isoproterenol; βOHB, β-hydroxybutyrate; PI3K, phosphatidylinositol 3-kinase. When NRVCMs were exposed to ISP, cell viability decreased significantly compared to the control group. This cell damage was significantly and dose-dependently reduced when the cells were pretreated with Cana (Fig. 5C). Interestingly, β-hydroxybutyrate (βOHB) treatment also significantly mitigated ISP-induced cell damage. However, there was no significant difference between the βOHB + Cana 10 µM pretreatment and Cana 10 µM alone (Fig. 5D). Taken together, our results suggest that Cana, rather than βOHB, is primarily responsible for the protection of NRVCMs in our model. To determine whether this effect is specific to Cana or shared with other SGLT2 inhibitors, we also tested empagliflozin (Empa). Empa partially alleviated ISP-induced cell injury but the effect was less pronounced than that observed with Cana (Fig. 5D). However, given the limited sample size of Empa, these findings between the two drugs should be considering preliminary. The phosphoinositide 3-kinase/ protein kinase B (PI3K/AKT) pathway has imperative roles in the regulation of the size and survival of cardiomyocytes, angiogenic processes and inflammatory responses22. Notably, the protective effect of Cana on ISP-induced cardiomyocyte damage was abolished when LY29002, a PI3K inhibitor, was added (Fig. 5D). Cana pretreatment improved mitochondrial function in CMs Cardiomyocyte viability critically depends on mitochondrial function, as mitochondria regulate energy production, apoptosis, oxidative stress, and calcium homeostasis—key factors for cell survival23. Under ISP stimulation, Cana treatment significantly improved mitochondrial membrane potential in cardiomyocytes compared to untreated cells. Similarly, βOHB enhanced mitochondrial membrane potential; however, consistent with the cell viability data, no significant difference was observed between the βOHB + Cana pretreatment and Cana alone (Fig. 6A, B). Fig. 6Changes in mitochondrial function in cardiomyocytes following ISP stimulation A. MMP was assessed using MT-1 fluorescent dye after 12 h after ISP stimulation. Scale bar = 100 μm. B. Quantification of area of the MT-1-positive area (n = 100 cells per group). C. mtROS levels were evaluated by DCFH-DA staining 12 h after ISP stimulation. Scale bar = 100 μm. D. Quantification of the DCFH-DA-positive area (n = 100 cells per group). Bar graphs represent mean +/-SE. *P < 0.05 vs. control. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. MMP, mitochondrial membrane potential; mtROS, mitochondrial reactive oxygen species; Cana, canagliflozin; ISP, isoproterenol. Changes in mitochondrial function in cardiomyocytes following ISP stimulation A. MMP was assessed using MT-1 fluorescent dye after 12 h after ISP stimulation. Scale bar = 100 μm. B. Quantification of area of the MT-1-positive area (n = 100 cells per group). C. mtROS levels were evaluated by DCFH-DA staining 12 h after ISP stimulation. Scale bar = 100 μm. D. Quantification of the DCFH-DA-positive area (n = 100 cells per group). Bar graphs represent mean +/-SE. *P < 0.05 vs. control. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. MMP, mitochondrial membrane potential; mtROS, mitochondrial reactive oxygen species; Cana, canagliflozin; ISP, isoproterenol. Additionally, Cana treatment under ISP stimulation significantly reduced mitochondrial ROS levels in cardiomyocytes compared to cells without Cana treatment (Fig. 6C, D). Cana pretreatment improved CMs sarcomere extension and attenuated hypertrophy ISP induces hypertrophy in NRVCMs, leading to increases in cell size and structural changes24. Sarcomere organization and lengthening are key determinants of mechanical function, and it has been shown that sarcomere elongation25 can contribute to myocyte hypertrophy26. To measure cardiomyocyte sarcomere length, we created a two-dimensional graph of pixel intensity along a line drawn over the sarcomeres. Sarcomere length was determined by measuring the distance between consecutive peaks in pixel intensity on the graph (Fig. 7A). The stimulation with ISP elongated the sarcomere length but was shortened by Cana treatment (Fig. 7B). Fig. 7Effects of Cana on Sarcomere Length and Mitochondrial Ca²⁺ Uptake in ISP-Stimulated Cardiomyocytes A. Sarcomere length was evaluated by analyzing a two-dimensional pixel intensity graph along sarcomeric actin filaments. The distance between consecutive intensity peaks was measured, and the average of five peaks was calculated per cell. The cells were immunostained for sarcomeric actin (green) and nuclei (DAPI; blue). Scale bar = 25 μm. B. Quantification of sarcomere length (n = 5 cells per condition).C. Cell perimeter and D. Cell area of cardiomyocytes (ISP, n = 7; ISP + Cana, n = 9). E. Mitochondrial Ca²⁺ uptake was measured after 12 h of ISP stimulation using Rhod-2 AM staining (n = 5 cells per condition). Values are the mean +/-SE. *P < 0.05 vs. control. Significance was assessed using one-way ANOVA. Cana, canagliflozin; ISP, isoproterenol. Effects of Cana on Sarcomere Length and Mitochondrial Ca²⁺ Uptake in ISP-Stimulated Cardiomyocytes A. Sarcomere length was evaluated by analyzing a two-dimensional pixel intensity graph along sarcomeric actin filaments. The distance between consecutive intensity peaks was measured, and the average of five peaks was calculated per cell. The cells were immunostained for sarcomeric actin (green) and nuclei (DAPI; blue). Scale bar = 25 μm. B. Quantification of sarcomere length (n = 5 cells per condition).C. Cell perimeter and D. Cell area of cardiomyocytes (ISP, n = 7; ISP + Cana, n = 9). E. Mitochondrial Ca²⁺ uptake was measured after 12 h of ISP stimulation using Rhod-2 AM staining (n = 5 cells per condition). Values are the mean +/-SE. *P < 0.05 vs. control. Significance was assessed using one-way ANOVA. Cana, canagliflozin; ISP, isoproterenol. To further support the antihypertrophic effect of Cana, we quantitatively analyzed cardiomyocyte size using perimeter and area measurements. Both parameters were significantly increased in the ISP group but markedly reduced in the ISP + Cana group (Fig. 7C, D), supporting the conclusion that Cana attenuates ISP-induced cardiomyocyte hypertrophy. Calcium transient Mitochondria are involved in the regulation of cellular Ca2+ in cardiomyocytes27. To elucidate the effect of Cana on cellular calcium, NRVCMs were loaded cells with Rhod-2 AM. The mitochondrial Ca2+ uptake in cardiomyocytes, which was increased by ISP stimulation, decreased significantly in Cana pretreated cells after 30 min of ISP exposure (Fig. 7C). Expression of membrane channels involved in calcium transport in cardiomyocytes Changes in mitochondrial Ca²⁺ transients suggest that alterations in calcium handling at the plasma membrane may drive the observed changes in mitochondrial calcium levels28. We measured the protein expression of sodium–calcium exchanger 1 (NCX1), and sodium–hydrogen exchanger 1 (NHE1). As shown in Fig. 8A-C, while ISP increased the expression of NCX1 and NHE1, Cana tended to reduce their protein levels, suggesting that Cana reduced calcium entry primarily via NCX1 and NHE1. Fig. 8Western Blotting and Immunofluorescence Analysis of Cardiomyocytes A Representative Western blot images for NCX1 and NHE expression. B, C. Densitometric quantification of NCX1 and NHE1 expression normalized to GAPDH. Each condition was assessed in 3 independent experiments, yielding a total of 5 technical replicates (lanes) per group. Data are expressed as fold change relative to the corresponding control. D. Double immunofluorescence staining for sarcomeric actin (red), SGLT1 (green), and DAPI nuclear counterstaining (blue).E. Quantification of SGLT1 expression based on immunofluorescence intensity (n = 5 per condition). Graphs represent mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. NCX1, sodium–calcium exchanger 1; NHE1, sodium–hydrogen exchanger 1; SGLT1, sodium–glucose cotransporter 1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; Cana, canagliflozin; ISP, isoproterenol. Western Blotting and Immunofluorescence Analysis of Cardiomyocytes A Representative Western blot images for NCX1 and NHE expression. B, C. Densitometric quantification of NCX1 and NHE1 expression normalized to GAPDH. Each condition was assessed in 3 independent experiments, yielding a total of 5 technical replicates (lanes) per group. Data are expressed as fold change relative to the corresponding control. D. Double immunofluorescence staining for sarcomeric actin (red), SGLT1 (green), and DAPI nuclear counterstaining (blue).E. Quantification of SGLT1 expression based on immunofluorescence intensity (n = 5 per condition). Graphs represent mean +/-SE. *P < 0.05. **P < 0.01. Significance was assessed using one-way ANOVA, followed by the Tukey post-hoc test. NCX1, sodium–calcium exchanger 1; NHE1, sodium–hydrogen exchanger 1; SGLT1, sodium–glucose cotransporter 1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; Cana, canagliflozin; ISP, isoproterenol. ISP-induced SGLT1 expression is suppressed by Cana treatment Several studies documented that heart failure conditions increase left ventricular (LV) SGLT1 mRNA or protein expression as compared with non-failing controls29–31, whereas SGLT2 is not expressed in murine and human hearts to a relevant extent32. As Cana is an SGLT2 inhibitor with affinity for SGLT1, we next examined its expression in cardiomyocytes. The upregulation of SGLT1 in cardiomyocytes induced by ISP was suppressed by treatment with Cana (Fig. 8D, E). SGLT2 was not detected under any conditions. Discussion We demonstrated that Cana improves electrical and structural remodeling induced by ISP in nondiabetic rats. ISP stimulation prolonged the VERP and MAPD, effects that were significantly reduced by Cana pretreatment. These electrophysiological improvements were accompanied by reductions in d-ROMs, cardiomyocyte apoptosis, and markers of hypertrophy. The prolongation of VERP and MAPD in ISP-treated animals is consistent with previous reports linking calcium overload and oxidative stress to impaired repolarization via modulation of Ca²⁺ channels (both membrane and sarcoplasmic)33 and potassium channels (IKr and IKs)33,34, ultimately leading to slower repolarization. Although myocardial fibrosis can contribute to inhomogeneous repolarization11, we observed no significant fibrotic changes in our model, suggesting that cellular remodeling, such as hypertrophy and apoptosis, may underlie the observed electrophysiological changes. Cana significantly reduced ISP-induced cardiomyocyte apoptosis35, an effect that appears to be mediated by the PI3K/Akt pathway. This signaling cascade promotes cell survival36, stabilizes mitochondrial function, suppresses proapoptotic proteins37,38 and enhances antioxidant responses39. In our study, Cana preserved the mitochondrial membrane potential (MMP), reduced mitochondrial ROS (mtROS), and limited mitochondrial Ca²⁺ uptake in cardiomyocytes. These findings are supported by prior reports showing that Cana enhances the expression of endogenous antioxidants such as catalase (CAT) and superoxide dismutase (SOD) while downregulating the expression of proapoptotic mediators such as iNOS, Nox4, caspase-3, and Bax40. Cana also elevated serum ketone body levels in our rat model, which is consistent with the findings of previous studies showing that SGLT2 inhibitors enhance cardiac energy metabolism via enhanced ketogenesis41,42. Although βOHB treatment improved mitochondrial parameters in vitro, it did not exert additive effects when combined with Cana, suggesting that the cardioprotective effects of Cana are not solely dependent on ketone signaling. Several factors may explain the limited effect of βOHB in our model. First, the concentration and timing of βOHB exposure may have been suboptimal, as ketone-mediated signaling often requires sustained exposure or higher systemic levels to activate protective pathways. Second, βOHB acts indirectly, functioning as both an energy substrate and epigenetic modulator (e.g., HDAC inhibitor43, which may not elicit acute effects in isolated cardiomyocytes. Finally, Cana exerts broader pharmacological actions—including PI3K/Akt activation, mitochondrial preservation, and the suppression of NHE1 and NCX1—none of which are directly targeted by βOHB. Therefore, the observed benefit of Cana likely involves a combination of metabolic and nonmetabolic mechanisms beyond simple ketone elevation. The PI3K/Akt pathway plays important roles in regulating cardiomyocyte growth, survival, and resistance to stress-induced apoptosis22. PI3K is an intracellular kinase that, upon activation, generates PIP3 (phosphatidylinositol-3,4,5-trisphosphate), which serves as a second messenger to recruit and activate Akt by promoting its phosphorylation44. Activated Akt exerts its regulatory effects on cell survival by phosphorylating a series of downstream substrates that inhibit apoptosis and enhance cell viability45. In our study, Cana rescued ISP-induced cardiomyocyte apoptosis and restored cell viability in NRVCMs, effects that were abolished by the PI3K inhibitor LY294002, confirming that Cana’s protective mechanisms are PI3K/Akt-dependent. This is consistent with previous reports showing that Cana and other SGLT2 inhibitors, such as dapagliflozin, activate PI3K/Akt signaling to promote autophagy in diabetic atrial fibrillation models and reduce oxidative and inflammatory stress in ISP-induced injury40,46,47. We also observed sarcomere shortening in the Cana group, which may indicate the attenuation of hypertrophic remodeling. Increased sarcomere length has been associated with mechanical stress and hypertrophic signaling25,48–51. In line with these findings, our in vitro analysis using neonatal NRVCMs revealed that ISP stimulation increased the cell area and perimeter, indicating hypertrophic growth. Cana pretreatment significantly reduced both the cell area and perimeter, supporting its antihypertrophic effect at the cellular level. Prior studies have also shown that hypertrophic signaling involves upregulation of fetal genes and MAPK/PKC pathways26,51, which may be modulated by Cana. Cana modulates calcium-handling membrane proteins by downregulating NCX1 and NHE1, both of which are upregulated during myocardial stress and contribute to calcium overload52,53. This modulation likely contributes to the suppression of SGLT154, which is upregulated by calcium overload and oxidative stress via HIF-1α and NF-κB signaling55. Prior evidence from neuronal models supports the Cana’s ability to downregulate SGLT1 and protect against ischemic injury via the SGLT1/AMPK axis56. Although SGLT1 downregulation may contribute to the effects of Cana, whether this represents a direct therapeutic mechanism, or a secondary response remains unclear. Several studies have shown that SGLT2 inhibitors with low affinity to SGLT1, such as dapagliflozin, can directly suppress NHE157,58 or NCX activity59. In our study, empagliflozin partially mitigated ISP-induced cardiomyocyte injury, but the effect appeared weaker than that of Cana, although this observation is limited by the small sample size. Importantly, recent study demonstrated that empagliflozin inhibits the late sodium current (INa) via CaMKIII inhibition60. Taken together, while our findings add to the emerging evidence that SGLT2 inhibitors can modulate cardiac electrophysiology, the precise contribution of SGLT1 remains to be clarified. To further delineate the role of SGLT1 in this context, studies employing selective SGLT1 inhibitors, SGLT1 knockout models, or another dual inhibitor such as sotagliflozin61 would be informative. However, caution is warranted, as SGLT1 inhibition can lead to adverse gastrointestinal effects due to its essential role in intestinal glucose absorption62. Nevertheless, clinical studies with sotagliflozin have reported cardioprotective benefits, including reductions in heart failure events and mortality in patients with type 2 diabetes63, reinforcing the therapeutic relevance of dual SGLT1/SGLT2 inhibition. From a translational perspective, the ability of Cana to attenuate oxidative stress, improve mitochondrial function, suppress Ca²⁺ overload, and reduce electrical remodeling suggests its therapeutic potential in heart failure, even in the absence of diabetes. By normalizing the VERP and MAPD, Cana may reduce the risk of ventricular arrhythmias and atrial fibrillation. Activation of PI3K/Akt and downregulation of NCX1, NHE1, and SGLT1 further support its role in improving cardiomyocyte survival. Limitations of our study include the use of male rats only, the absence of functional knockdown of SGLT1, the lack of larger sample sizes and dose–response analyses with empagliflozin, and the lack of assessment of SR-specific calcium handling proteins such as SERCA2a and RyR2. Although we used a dose of canagliflozin (10 mg/kg/day) supported by prior rodent studies64–66, direct extrapolation to human therapeutic dosing is limited by species differences in pharmacokinetics and drug metabolism67. Thus, while our findings support a mechanistic role for canagliflozin in modulating myocardial remodeling, the clinical relevance of the precise dosing should be interpreted with caution. Additionally, while our in vitro data provide mechanistic insights, these findings should be validated in human cardiomyocytes. In conclusion, Cana attenuated ISP-induced electrical and structural remodeling by reducing oxidative stress, preserving mitochondrial function, and modulating calcium-handling membrane transporters. These effects were partly mediated via PI3K/Akt activation and SGLT1 suppression. Our preclinical observations provide novel insights into the possible mechanisms through which Cana preserve cardiac health. Materials and methods Rats All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85 − 23, revised 1996) and were approved by the Kitasato University ethics committee for animal experiments, which strictly adhered to the guidelines for animal experiments of the Kitasato University (approval number 2023 − 120). Eight-week-old male Sprague-Dawley rats (CLEA JAPAN, Inc, Tokyo, Japan) were kept in the departmental animal house under controlled conditions of temperature at 25 ± 2 °C, 60 ± 5% relative humidity, and standard light. We have determined the body weights of the rats as described in a previous study11. They were fed with standard food pellets and water adlib tum. A total of 38 rats were used in this study and assigned to four groups: Sham group (n = 9): Rats were injected with saline (1 mL) on day 0 and received oral buffer water (1 mL/kg/day) via gastric tube starting 7 days prior to the saline injection (day − 7) and continuing until day 7 (total of 14 days).ISP group (n = 11): Rats received a subcutaneous ISP injection (10 mg/kg) on day 0 and were administered buffer water identically to the sham group.ISP + Cana group (n = 15): Rats received an ISP injection on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7.Cana group (n = 3): Rats were injected with saline (1 mL) on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Sham group (n = 9): Rats were injected with saline (1 mL) on day 0 and received oral buffer water (1 mL/kg/day) via gastric tube starting 7 days prior to the saline injection (day − 7) and continuing until day 7 (total of 14 days). Sham group (n = 9): Rats were injected with saline (1 mL) on day 0 and received oral buffer water (1 mL/kg/day) via gastric tube starting 7 days prior to the saline injection (day − 7) and continuing until day 7 (total of 14 days). ISP group (n = 11): Rats received a subcutaneous ISP injection (10 mg/kg) on day 0 and were administered buffer water identically to the sham group. ISP group (n = 11): Rats received a subcutaneous ISP injection (10 mg/kg) on day 0 and were administered buffer water identically to the sham group. ISP + Cana group (n = 15): Rats received an ISP injection on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. ISP + Cana group (n = 15): Rats received an ISP injection on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Cana group (n = 3): Rats were injected with saline (1 mL) on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Cana group (n = 3): Rats were injected with saline (1 mL) on day 0 and were given Cana (10 mg/kg/day in 1 mL buffer water) via a gastric tube starting on day − 7 and continuing until day 7. Cana was provided by Mitsubishi Tanabe Pharma Corporation. Myocardial injury was induced by subcutaneous (s.c.) injection of ISP (10 mg/kg, I6054, Sigma Aldrich, St. Louis, USA) as previously reported11, and this day of injection was defined as day 0 in the protocol. The rats were sacrificed on either day 1 (Acute phase) or day 7 (Sub-Acute phase). Euthanasia was performed by administering an overdose of a combination anesthetic consisting of medetomidine, midazolam, and butorphanol. The experimental protocol is described in Fig. 1. The sample size (n) for each figure and experiment is detailed in the figure legends. Discrepancies in n-values, particularly in electrophysiological measurements (e.g., VERP and MAPD), are due to mortality in a subset of rats during or following the electrophysiological procedure. These deaths were unrelated to the treatment but occurred during invasive recording under anesthesia. As a result, some animals could not be included in downstream analyses. These numbers have been carefully updated and are now consistent across the figures, figure legends, and the Materials and Methods section. Histology and measurement of myocardial fibrosis The heart was sampled for histological evaluation in selected rats at the end of the protocol. Each heart was transversely sliced and fixed in 10% formalin and embedded in paraffin. Transverse sections of the sliced left ventricle were subjected to hematoxylin and eosin (H&E) staining for the morphological analysis and with Azan staining for the evaluation of tissue fibrosis. The fibrotic area in each slice was quantified by a digitized image analyzer (LuminaVision Co. Ltd., Tokyo, Japan). Measurements of heart rate and blood pressure Heart rate and systolic blood pressure were evaluated in awake mice using a tail-cuff system with a pneumatic pulse transducer (BP-98; Softron, Tokyo, Japan). Electrophysiological study (EPS) At the end of the protocol and prior to sacrifice, electrophysiological evaluation was performed under anesthesia using intraperitoneal sodium pentobarbital (10–50 mg/kg), followed by tracheotomy and controlled ventilation (Shinano, SN480-7, Tokyo, Japan). The heart was exposed by the median sternotomy, and the electrical stimulation and/or recordings were performed through a pair of plunged platinum needle electrodes (φ 0.1 mm) as previously described11,68. The heart rate was controlled not to exceed 400 beats/minute by additional intravenous anesthesia with sodium pentobarbital. The analog signals were converted into digital signals at a sampling frequency of 1,000 Hz (PowerLab/8SP; ADInstruments Pty Ltd, Bella Vista, NSW, Australia) and stored on a computer hard disk. The bandpass filter was set at 50–300 Hz for standard cardiac electrogram recording and at open − 300 Hz for recording the MAP. To evaluate the VERP, a single extrastimulus was delivered after 8 basic stimuli with a cycle lengths of 140 and 120 msec. The coupling interval of the extrastimulus was shortened in 2-msec steps until ventricular refractoriness was reached. To standardize the influence of the activation cycle length, MAP was evaluated using the last beat of 8 consecutive passed beats with fixed cycle lengths of 140 and 120 msec. The MAPD was determined as the interval between the onset of the MAP trace and the 20% (MAPD20) and 90% (MAPD90) repolarization times. Heart and body weights were evaluated in all rats, including the rats used in the electrophysiological study, but the following histological and biochemical evaluations were performed on the remaining mice. Clarification regarding anesthesia and pacing Although sodium pentobarbital has known negative chronotropic and inotropic effects, all electrophysiological measurements were performed under controlled conditions with ventricular pacing, minimizing variability in heart rate and rhythm. Importantly, ISP was administered 24 h prior to the EPS and not on the day of measurement, so its acute hemodynamic effects (e.g., tachycardia or enhanced inotropy) did not influence the electrophysiological recordings. The heart rate was uniformly controlled not to exceed 400 beats/min using titrated pentobarbital, and the key parameters such as the VERP and MAPD were assessed during the fixed-rate pacing protocols. These conditions ensured that group comparisons reflected intrinsic myocardial electrical properties rather than differences in autonomic tone or systemic drug effects. Evaluation of oxidative stress The serum levels of d-ROMs were measured using the Free Radical Analytical System 4 (FRAS4; H&D srl, Parma, Italy) during the Acute and Sub-Acute phases. The measurement of d-ROMS levels is defined as CARRATELLI UNITS (U.CARR) according to a previous study11. Evaluation of ketone bodies Rat serum samples were sent to SRL, Inc. (Tokyo, Japan) for the measurements of total keone bodies, acetoacetic acid, andβ-hydroxybutyric acid. Cell culture Primary NRVCMs were isolated from 0 to 1-day-old Wistar rats as previously described69. Briefly, hearts were quickly minced and digested using a Neonatal Heart Dissociation Kit (Miltenyi Biotec, Bergisch Glad Bach, Germany). Isolated cells were plated in laminin-coated culture dishes and 96-well culture plates. The density of cells was approximately 5.0 × 105 cells/mL. NRVCMs were collected and plated in the medium, consisted of DMEM, 5% fetal bovine serum (FBS), 10% horse serum (HS), and 1% penicillin/streptomycin for another 48 h. NRVCMs were maintained at 37℃ with 5% CO2 and 95% O2. NRVCMs were subjected to DMEM starvation and treatment by Cana (10 µM, unless indicated otherwise) and/or βOHB, 2 mM, Sigma Aldrich 166898, St. Louis, USA) for 12 h before stimulation with ISP (10 µM). Cells without Cana or ketone body pretreatment were used as controls. Cell viability assay NRVCM cell viability was measured by a 3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide (MTT) assay (Roche 11465007001, Basel, Switzerland) as previously described70. NRVCMs were plated on 96-well culture plates. MTT was added to each well and incubated for 4 h in a CO2 incubator, after which cells were lysed with acid isopropanol in 0.04 N HCL. The plates were then analyzed with a microplate reader at 590 nm. MTT assays were also performed with or without a phosphoinositide 3-kinase (PI3K) inhibitor (LY294002, Sigma Aldrich L9908, St. Louis, USA). Terminal Deoxynucleotidyl transferase dUTP Nick end labeling (TUNEL) assay Cell apoptosis was detected by staining cryosectioned heart tissue samples or NRVCMs with the cell death detection kit (Takara MK500, Shiga, Japan), as described before71. Following three washes with PBS, drug-treated cardiomyocytes were fixed with 4% paraformaldehyde. Cells were then permeabilized with 0.1% Triton X-100 and incubated with freshly prepared TUNEL reaction mixture for 1 h at 37 °C. After TUNEL labeling, cells were incubated with anti-α-actinin (Actn2, 1:400; Sigma A7811; St. Louis, MO, USA), followed by an Alexa Fluor–conjugated secondary antibody (1:500), and counterstained with DAPI to visualize the nuclei. Multichannel 16-bit images were acquired with a Zeiss LSM 710 microscope. The apoptotic index was defined as the percentage of TUNEL-positive cells/total nuclei. Mitochondrial membrane potential (MMP) assay The MMP was evaluated using an MT-1 MitoMP Detection Kit (Dojindo 343–09781, Kumamoto, Japan) following the manufacturer’s protocol. Treated cells were incubated with MT-1 working solution for 30 min at 37℃. After treatment with MT-1 working solution, the cells were fixed with 4% paraformaldehyde. The red fluorescence intensity derived from MT-1 was determine with a BZ-X710 fluorescence phase contrast microscope (Keyence, Osaka, Japan), and the average intensity value of the stained cells was calculated as by image J software (NIH, Bethesda, USA). Intracellular Ca2+ measurement The NRVCMs were loaded with a Ca2+ sensitive dye, the acetoxymethyl-ester of Rhod-2 (Rhod-2-AM; Dojindo R002, Kumamoto, Japan) using a Wash type Calcium Kit (Dojindo CS22, Kumamoto, Japan), which resulted in preferential electrical potential-driven uptake preferentially into the mitochondria, according to the supplier’s protocol. After loading, the cells were incubated for 1 h. The loading buffer was removed carefully, and the cells were washed in PBS. Afterward, the recording buffer (37 °C) was added to each well, fluorescence intensity changes were measured at Ex: 553 nm, Em: 576 nm, at 5-, 15-, and 30-minutes following ISP treatment. Mitochondrial ROS measurement Mitochondrial reactive oxygen species (ROS) were detected using a ROS assay DCFH-DA (Fujifilm-Wako 345–0998, Osaka, Japan) in NRVCMs induced by ISP for 12 h, following the manufacturer’s protocol. Briefly, 10µM DCFH-DA dye working solution was added to NRVCMs, which were then incubated for 30 min at 37 ℃. After 12 h of ISP stimulation, the cells were fixed by 4% PFA. The green fluorescence intensity derived from DCFH-DA was determined with a BZ-X710 fluorescence phase contrast microscope (Keyence, Osaka, Japan), and the average intensity value of the stained cells was calculated by image J software. Western blotting Proteins were extracted for Western blotting in RIPA lysis buffer (20 mM HEPES, 150 mM NaCl, 1% IGEPAL CA630 (NP-40), 0.1% SDS, 0.5% Sodium deoxycholate/pH 7.5). Equal amounts of proteins were separated on SDS–polyacrylamide gel electrophoresis and transferred to PVDF membranes. Proteins were detected with the primary antibodies against Na2+/Ca2+ exchanger, isoform 1 (NCX1; 1/1,000; Cell Signaling 79350; Danvers, USA) and anti-sodium/hydrogen exchanger 1 (NHE1; 1/500; Abcam 67314; Cambridge, UK) to analyse the activation of Ca channels or GAPDH (1/1,000; Cell Signaling 5174; Danvers, USA) as a control, after 48 h of ISP treatment, followed by HRP-conjugated secondary antibodies (1/10, 000; Cytiva, Tokyo, Japan).The protein quantification was obtained by densitometry analysis using Image Quant 800 (Cytiva, Tokyo, Japan) and normalized to the GAPDH level. Immunofluorescence Immunofluorescence was performed with a standard protocol69, using primary antibodies against Actn2 (1/400, Sigma A7811, St. Louis, USA), SGLT1 (1/100, Abcam 14686, Cambridge, UK) and SGLT2 (1/50, Abcam 37296, Cambridge, UK), Alexa Fluor conjugated secondary antibodies (1/500) and DAPI nuclear staining. Multichannel 16-bit images were acquired with a Zeiss LSM 710 microscope and a 63 ×/1.4 oil objective. The SGLT1 or SGLT2 index was defined as the percentage of SGLT1 or SGLT2 positive cells/total nuclei. Sarcomere length was also evaluated by measuring the distance between intensity peaks, corresponding to adjacent α-actinin-labelled filaments, using Image J. Statistical analysis Statistical analysis was performed with JMP 11.2 (SAS Institute, Cary, North Carolina, USA) statistical software package. Continuous variables are presented as the mean ± standard deviation and were compared using one-way ANOVA, Tukey or Dunnett post-hoc analysis. A P value of < 0.05 was considered to indicate statistical significance. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2