Dox hydrochloride (D1515), phloridzin dihydrate (P3449), and βOHB (54920) were obtained from Sigma-Aldrich (St. Louis, MO, USA). EMPA (Boehringer Ingelheim, Ingelheim am Rhein, Germany) is a potent SGLT2 inhibitor. We made the EMPA diet by mixing EMPA in NCD (Harlan Teklad, Madison, WI, USA) at a concentration 300 mg/kg of diet.

Male C57BL/6J (B6J) mice were purchased from the Institute of Medical Science, University of Tokyo. The mice were housed at ambient temperature (22±1°C) in a specific pathogen-free facility under a 12 hours light-dark cycle. All diets and water were provided ad libitum. The serum βOHB levels of mice were measured using a βOHB Colorimetric Assay kit (Cayman Chemical, Ann Arbor, MI, USA) according to the manufacturer's protocol. All experiments were performed in accordance with relevant guidelines and regulations. Our animal protocol was designed to minimize pain and discomfort to the mice. For euthanasia, mice were continuously exposed to isoflurane during tissue sampling. All animal procedures were approved by the Institutional Animal Care and Use Committee of CHA University (IACUC No. 160033, 170047). Our study did not involve human data or tissue.

We constructed acute cardiomyopathy and chronic cardiomyopathy mouse models as previously described.11) Male B6J mice (8-weeks-old, weighting 22–24 g) were randomly divided into 4 groups. Control group: B6J mice (n=5) were fed an NCD with phosphate-buffered saline (PBS) injection. EMPA group: B6J mice (n=5) were fed the EMPA diet with PBS injection. Dox group: B6J mice (n=5) were fed an NCD with Dox injection. Dox+EMPA group: B6J mice (n=5) were fed the EMPA diet with Dox injection. Acute cardiomyopathy was induced in the B6J mice by a single intraperitoneal (IP) injection of Dox (15 mg/kg). Diets were changed at night, 12 hours before PBS/Dox injection. To analyze the phenotype of these mice, all mice were sacrificed 2 weeks later. Chronic cardiomyopathy was induced by administering Dox at 2.5 mg/kg every other day for 12 days (cumulative dose=15 mg/kg). All mice were sacrificed 6 weeks later.

Cardiac magnetic resonance imaging (MRI) was performed to assess the structural and functional changes in mice hearts by Dox-induced cardiotoxicity. Male B6J mice (10-weeks-old, n=5 per each group) were examined using a Mouse Agilent 9.4 Tesla MRI Scanner (Agilent Technologies, Santa Clara, CA, USA). Heart sizes were measured using Image J software and functional parameters were calculated as previously described.12)

H9C2 cells (rat cardiac myoblasts) were purchased from the Korean Cell Line Bank (21446) and maintained in high glucose Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 100 U penicillin/streptomycin. Cells were treated with various concentration of Dox (ranging from 0.1 to 10 µM) and tested after 12 hours or 24 hours. Dox and Phloridzin dihydrate were dissolved in dimethyl sulfoxide (DMSO). DMSO was used as a control vehicle.

The cytotoxic effects of Dox and protective effects of other drugs in vitro on H9C2 cells were measured as previously described.13) The viability of H9C2 cells were analyzed using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide based colorimetric assay using a Cell Proliferation Kit I (Sigma-Aldrich). Adenosine triphosphate (ATP) was measured using ATPLite Luminescence Assay System (PerkinElmer, Waltham, MA, USA).

Mitochondrial membrane potential (MPP) and reactive oxygen species (ROS) were measured as previously described.14) To measure the MPP and ROS, H9C2 cells were preincubated βOHB for 2 hours, and then with Dox for 12 hours in serum free media. For mitochondrial analysis, H9C2 cells were incubated for 30 minutes with 50 nM tetramethylrhodamine methyl ester (TMRM; Sigma-Aldrich). TMRM fluorescence images were obtained using confocal microscopy (Leica TCS SP5 II). Intracellular ROS were measured using 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) (Thermo Fisher Scientific, Waltham, MA, USA).

Total RNA was extracted using the TRIzol^® reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. For complementary DNA synthesis, total RNA (1µg) was reverse transcribed with the Qiagen Quantitect Reverse Transcription Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocols. Quantitative real-time polymerase chain reaction (PCR) was performed using a FastStart Essential DNA Green Master kit (Roche Life Sciences, Indianapolis, IN, USA). The primer sequences are provided in Supplementary Table 1.

Heart tissue lysates and whole-cell lysates were used for western blotting analysis. Proteins were separated by electrophoresis and transferred to membranes. The membranes were incubated with specific primary antibodies (1:1,000) recognizing β-actin (#4970), cleaved-caspase 3 (#9664), and C/EBP homologous protein (CHOP; #5554), purchased from Cell Signaling Technology (Danvers, MA, USA). The membranes were washed with Tris-buffered saline containing 0.05% Tween 20 and incubated with horseradish peroxidase-linked anti-rabbit immunoglobulin G (IgG) antibody (1:2,000; AP188p) or anti-mouse IgG antibody (1:2,000; AP160P) from Sigma-Aldrich. Signals were captured using a ChemiDoc MP system (Bio-Rad, Hercules, CA, USA). Densitometric analysis was performed using Image J software.

Mouse hearts were harvested, fixed in 4% paraformaldehyde in PBS, dehydrated, and embedded in paraffin. Then, 5-µm-thick tissue sections were deparaffinized and stained with hematoxylin and eosin (H&E). Fibrosis was detected with Masson's trichrome staining (Abcam, Cambridge, UK) according to the manufacturer's protocol.

Data are presented as mean±standard error of mean. Differences between groups were evaluated for statistical significance using Student's t-test or 1-way analysis of variance followed by Tukey's post hoc text. The p values <0.05 were considered significant.

To test our hypothesis that EMPA might improve cardiac function in HF, we induced cardiac toxicity in mice by Dox administration and then fed them with an EMPA diet. We divided these mice into 4 groups: control group, EMPA group, Dox group, and Dox+EMPA group. Two weeks after Dox/vehicle injection, the left ventricles (LVs) of the mice in these 4 groups were examined to analyze the effects of EMPA on mouse hearts. To evaluate whether EMPA protects against Dox-induced cardiomyopathy, functional analysis was performed by heart MRI. In the acute stage, Dox causes cardiac hypertrophy in mice.15) The Dox+EMPA group showed body weight reduction after 2 weeks treatment but this change was not significant (Supplementary Figure 1). After 2 weeks of Dox treatment, mouse hearts showed hypertrophic changes (Figure 1A and B). As expected, EMPA reduced these pathological changes. The Dox+EMPA group displayed significantly reduced LV mass and improved fractional shortening compared with the Dox group (Figure 1B and C, Supplementary Figure 2). Masson's trichrome staining revealed that EMPA reduced perivascular and interstitial fibrosis in mouse hearts treated with Dox (Figure 1D-F and Supplementary Figure 3). These results suggested that the SGLT2 inhibitor protects against Dox-induced cardiac toxicity in vivo.

We observed that an SGLT2 inhibitor has protective effects against Dox-induced cardiac injury in vivo. However, the exact mechanism underlying the cardio-protective effect of SGLT2 inhibitors is still unclear. The SGLTs are present in heart. However, the major SGLT is SGLT1, not SGLT2.16)

We checked the expression of Sglt genes in the mouse heart. As expected, Sglt1 is expressed in both the kidney and LV. However, Sglt2 gene expression was very low in the LV compared with its expression in the kidney (Figure 2A and B). To investigate the direct effect of SGLT in the heart, we treated H9C2 cells with phloridzin (PHL), a non-selective SGLT inhibitor. After 2 hours of pretreatment with PHL, H9C2 cells were incubated with Dox for 24 hours. As shown in Figure 2C and D, cell viability was significantly reduced after Dox treatment and PHL pretreatment did not show a protective effect on Dox-induced cytotoxicity. These results suggested that SGLT has no direct role in cardiomyocytes during Dox-induced cardiotoxicity.

We focused on βOHB, a ketone body that is increased in the blood after SGLT2 inhibition, even in non-diabetic conditions.9)17)18) First, we treated H9C2 with βOHB at various concentrations (100 μM, 1 mM, 10 mM, and 30 mM) according to previous studies.18)19) As shown in Figure 2C and E, βOHB pretreatment had significant protective effects against Dox-induced cardiotoxicity.

Increased free radical production is the major mechanism of Dox-induced cardiac injury.20) Therefore, we measured intracellular ROS generation in H9C2 cells using H₂DCFDA. H9C2 cells produced ROS after the addition of 5 μM Dox. However, when H9C2 cells were pretreated for 2 hours with βOHB, Dox-induced ROS production was significantly decreased (Figure 3A and B, Supplementary Figure 4). We then performed a western blotting assay to detect cell stress markers. We found that the levels of both cleaved caspase 3 and CHOP were increased after Dox treatment. In contrast, βOHB pretreatment suppressed the levels of these proteins (Figure 3C and D).

Dox treatment also induces mitochondrial dysfunction in cardiomyocytes.21) So, we measured the ATP levels after βOHB treatment. As expected, βOHB significantly increased intracellular ATP levels in H9C2 cells (Figure 3E). We then analyzed the MPP of Dox-treated H9C2 cells using TMRM fluorescent probe. Dox-treated H9C2 cells showed decreased uptake of TMRM. βOHB pretreatment restored the uptake of TMRM in H9C2 cells. These results suggested that βOHB protects cardiomyocytes from Dox-induced mitochondrial dysfunction (Figure 3F and G).

Collectively, these in vitro results suggested that the cardio-protective effects of SGLT2 inhibition depend on elevated serum βOHB. We measured serum βOHB levels of our 4 groups of mice after 2 weeks of treatment. As expected, the βOHB level was elevated in EMPA diet-fed groups (EMPA and Dox+EMPA) (Figure 4A). We then examined whether the EMPA diet reduced Dox-induced ROS generation in the mouse heart in vivo. We analyzed antioxidant gene expression in the LVs from the 4 groups of mice. In the qPCR analysis, the Dox treated LVs showed reduced anti-oxidative gene expression and the EMPA diet restored this suppression induced by Dox treatment (Figure 4B).

We then investigated whether βOHB could reduce Dox-induced cardiomyopathy in vivo. After a single IP injection of Dox (15 mg/kg), mice were injected with saline or βOHB (20 mmoL/kg), according to a previously described method.22) After 2 weeks of IP injection, mice were sacrificed, histological analysis showed preserved myofibril structure and reduced cardiac fibrosis in the LVs of mice treated with Dox+ βOHB compared with that in the LVs of mice treated with Dox (Figure 4C and Supplementary Figure 5).

In addition, we investigated whether EMPA could inhibit Dox-induced chronic structural changes, such as eccentric hypertrophy in the heart. We fed the EMPA diet to Dox-treated mice and sacrificed them 6 weeks after the last injection, as previously described.23) After 6 weeks of Dox injection, the hearts of Dox treated mice showed dilated cardiomyopathy (increased size and eccentric hypertrophy) (Figure 4D). The hearts of the EMPA diet-fed mice retained a similar size and structure to those of the control mice. Histological analysis also showed preserved myofibril structure and reduced cardiac fibrosis in the LVs of mice fed with Dox+EMPA compared with those of mice fed with Dox (Figure 4E). Collectively, these results suggested that elevation βOHB by SGLT2 inhibition attenuated the oxidative stress induced by Dox treatment in mice.