Cardiology Plus

: 2021  |  Volume : 6  |  Issue : 2  |  Page : 109--120

Dapagliflozin Regulates Cardiac Metabolic Remodeling Partially via Uncoupling Protein 3 in a Nondiabetic Thoracic Aortic Constriction-Induced Mouse Model

Li-Wen Bao, Rong-Chen Liu, Fang-Ying Yan, Hui-Hua Fan, Guo-Qian Huang, Xiu-Fang Gao, Kun Xie, Yong Li, Hai-Ming Shi 
 Department of Cardiovascular Disease, Huashan Hospital, Fudan University, Shanghai, China

Correspondence Address:
Hai-Ming Shi
Department of Cardiovascular Disease, Huashan Hospital, Fudan University, Shanghai
Yong Li
Department of Cardiovascular Disease, Huashan Hospital, Fudan University, Shanghai


Objectives: We aimed to examine the protective effects of dapagliflozin (dapa) on thoracic aortic constriction (TAC)-induced heart failure in a nondiabetic mouse model. More specifically, we determined the effects of dapa on uncoupling protein 3 (UCP3) activation and subsequent metabolic remodeling. Methods: Sixty C57BL/6J mice were divided into six groups for TAC surgery and received different doses of dapa via gavage for 4 weeks. Echocardiography was performed to evaluate cardiac structure and function. Histological and molecular markers of cardiac remodeling and metabolic changes were assessed through staining assays, RT-PCR, western blot, and ELISA. HL-1 cells were used to explore the role of UCP3 in metabolic remodeling through transfection with UCP3 siRNA. Results: Mice that received TAC exhibited elevated heart weight/body weight ratios (HW/BW), left ventricular (LV) hypertrophy, impaired LV ejection fraction, and increased rates of fibrosis and apoptosis, unlike mice that received sham operation. Treatment with dapa after TAC restored HW/BW, improved LV parameters, and reduced fibrosis and apoptosis. dapa changed the expression levels of enzymes involved in glucose and fatty acid (FA) metabolism, such as pyruvate dehydrogenase lipoamide kinase isozyme 4, glucose transporter 4, carnitine palmitoyltransferase-1α, carnitine O-acetyltransferase, carnitine O-octanoyltransferase, acyl-CoA thioesterase 1, acyl-CoA thioesterase 2, peroxisome proliferator-activated receptors α and β, proliferator-activated receptor-gamma coactivator-1α, and UCP3, relative to the levels in mice in the TAC group. UCP3 siRNA reduced the expression of AMP-activated protein kinase and of factors involved in FA oxidation in vitro. Conclusions: Dapa exhibits cardioprotective effects in the model and augments expression of UCP3, which may be involved in metabolic remodeling in the failing heart.

How to cite this article:
Bao LW, Liu RC, Yan FY, Fan HH, Huang GQ, Gao XF, Xie K, Li Y, Shi HM. Dapagliflozin Regulates Cardiac Metabolic Remodeling Partially via Uncoupling Protein 3 in a Nondiabetic Thoracic Aortic Constriction-Induced Mouse Model.Cardiol Plus 2021;6:109-120

How to cite this URL:
Bao LW, Liu RC, Yan FY, Fan HH, Huang GQ, Gao XF, Xie K, Li Y, Shi HM. Dapagliflozin Regulates Cardiac Metabolic Remodeling Partially via Uncoupling Protein 3 in a Nondiabetic Thoracic Aortic Constriction-Induced Mouse Model. Cardiol Plus [serial online] 2021 [cited 2021 Sep 26 ];6:109-120
Available from:

Full Text


The global incidence of heart failure (HF) has been steadily increasing owing to the prevalence of arteriosclerotic cardiovascular disease (ASCVD).[1] According to the US National Health Interview Survey, the mortality rate associated with HF has exceeded the mortality rate associated with ASCVD.[2] Treatment with sodium-glucose co-transporter 2 inhibitors (SGLT2is), such as empagliflozin,[3] canagliflozin,[4] and dapagliflozin[5] (dapa), has been reported to reduce hospitalization for HF and cardiovascular death in cases of diabetes. Similar effects were also observed in the DAPA-HF trial and the EMPEROR-Reduced Trial.[6] Thus, SGLT2i confers cardioprotection in HF via mechanisms that inhibit the renin-angiotensin-aldosterone system (RAAS) and the sympathetic nervous system (SNS).

SGLT2 inhibition has been shown to benefit HF cases by enhancing Na+-H + exchange, inhibiting RAAS, and promoting diuresis;[7] however, these mechanisms do not fully explain the mechanism underlying the benefits of SGLT2i on HF. Improvement in several metabolic disorders, i.e., significantly decreased blood pressure, body weight, and uric acid levels, have been proposed in previous trials. Metabolic remodeling is an important mechanism primarily modulated by glucose and fatty acid (FA) metabolism that, in addition to abnormal activation of RAAS and SNS, leads to HF progression. Thus, we hypothesized that SGLT2 inhibition results in cardioprotection via the regulation of metabolic remodeling.[8]

Uncoupling protein 3 (UCP3) is a member of the UCP family and is expressed in the mitochondria of cardiac and skeletal muscles. Protons generated from oxidative phosphorylation leak across the inner mitochondrial membrane and return to the mitochondrial matrix independent of the ATP synthase, meaning that mitochondrial oxidative phosphorylation is not completely coupled. Given its role in energy dissipation, UCP regulates proton leakage and generates heat instead of ATP. Evidence suggests that UCP3 exerts cardioprotective effects and participates in the maintenance of mitochondrial biogenesis through the reduction of mitochondrial reactive oxygen species (ROS) and the regulation FA oxidation (FAO).

Herein, we established a model of HF via the thoracic aortic constriction (TAC) technique in 60 nondiabetic 8-week-old C57BL/6J male mice to determine metabolic processes that involve UCP3. Three different doses of dapa were administered to the mice to determine the benefits and adverse effects of SGLT2 inhibition in this model. UCP3 siRNA was utilized to verify whether UCP3 takes part in the regulation of metabolic remodeling and in the enhancement of FAO in cardiomyocytes with angiotensin II (AngII)-induced hypertrophy.

 Materials and Methods

Experimental design and animal grouping

Animal experiments were approved by the Animal Ethical Committee of Shanghai Medical School, Fudan University. All the experimental protocols were performed in accordance with the ARRIVE guidelines. The protocols of the current study were approved by the Department of Animal Laboratory of Fudan University (Approval No. 202001005Z).

A total of 60 nondiabetic 8-week-old C57BL/6J male mice were obtained from Shanghai Biomodel Organism Science and Technology Development Co. (SCXK (hu) 2017-0010; Shanghai, China). The mice were randomly divided into six groups: Sham group treated with saline (Sham; n = 10), 1.0 mg/kg/day dapa (Sham + dapa; n = 10), TAC + saline (n = 10), TAC + 1.0 mg/kg/d dapa (TAC + dapa1; n = 10), TAC + 2.0 mg/kg/d dapa (TAC + dapa2; n = 10), and TAC + 3.0 mg/kg/d dapa (TAC + dapa3; n = 10). dapa was obtained from Melone Pharmaceutical Co. (Dalian, China).

After 4 weeks of intragastric dapa or saline administration, the mice were anesthetized, blood samples were drawn from the posterior orbital venous plexus, and the hearts were rapidly excised.

Thoracic aortic constriction operation

Mice were anesthetized with a mixture of 100 mg/kg ketamine and 15 mg/kg xylazine. The neck and upper ventral chest of mice were shaved while mice were in the supine position. A horizontal incision (5 mm in length) was made at the level of the suprasternal notch to allow direct access to the transverse aorta without entering the pleural space. With the aid of a dissecting microscope, aortic constriction was performed by ligating the aorta between the right innominate artery and the left carotid artery with a 27-gauge needle using 5-0 silk suture. The needle was then quickly removed, leaving the constriction in place, and the skin was closed using wound clips.[9]


Mice were anesthetized with 1%–2% isoflurane in oxygen for echocardiography right before TAC operation and after the 4-week dapa/saline intragastric administration. Vivid E95 echo machine (GE Healthcare, Chicago, IL, USA) equipped with an 8–18 MHz phased array linear transducer was employed to examine left ventricular (LV) structure and function under the M-mode and two-dimensional echo.

Hematoxylin and eosin staining

Paraffin-embedded sections were used for hematoxylin and eosin staining. The tissue specimens were dewaxed with xylene for 10 min and hydrated in serial concentrations of ethanol (100%, 95%, 90%, and 85%) for 1 min each. The sections were washed with distilled water for 2 min and stained with hematoxylin for 4 min, differentiated in a hydrochloric acid-ethanol mixture for 5 min, and then stained with eosin for 2 min. Myocardial tissue morphology was then analyzed using ImageJ (NIH, Bethesda, MD, USA).

Masson's trichrome staining

Paraffin-embedded sections were washed with distilled water, and the nuclei were stained with Wiegert's hematoxylin for 10 min. Next, the stained sections were washed and treated with Masson's fuchsin acid solution for 10 min. Sections were bathed in 2% acetic acid. Subsequently, the sections were differentiated with 1% aqueous solution of phosphomolybdic acid for 5 min and then dried directly with aniline blue solution for 5 min without washing. The tissue sections were dipped in 95% alcohol, in anhydrous alcohol, and then in xylene. Then, the tissues sections were sealed using neutral gum after immersion in 0.2% acetic acid solution. The intensity of collagen staining was analyzed using ImageJ.

Terminal deoxynucleotidyl transferase dUTP nick-end labeling staining

Terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay was performed to detect apoptosis. Paraffin-embedded sections were dewaxed using xylene and hydrated in stepwise serial concentrations of ethanol and then washed twice with phosphate-buffered saline (PBS) for 5 min each wash. To remove proteins, the section was treated with 20 pg/ml protease K at 37°C for 15 min and washed three times with distilled water (2 min per wash). Subsequently, the section was transferred to a freshly prepared 3% hydrogen peroxide solution for 15 min at room temperature and washed twice with PBS (5 min per wash). Filter paper was used to remove excess liquid around the section, and 50 μl of the labeling solution was dropped onto the section. The section was placed in a wet box for reaction for 1 h at 37°C and then washed three times with PBS (3 min per wash). Afterward, peroxidase was added into the section, and the section was placed in the wet box for 30 min at 37°C and washed with PBS for 15 min. Next, 3,3'-diaminobenzidine was added to the section. Then, the section was stained with hematoxylin, counterstained blue with weak ammonia, and sealed with neutral gum. The sections were viewed under the microscope at high magnification. The nuclei of apoptotic cells appeared brown, and nonapoptotic nuclei appeared blue. Ten slices of the visual field were randomly selected from each section, and the number of apoptotic cells out of 100 cells was counted in each field. Average counts of TUNEL-positive cells per section were determined.

Quantitative real-time PCR

Total RNA was extracted from heart LV tissues using TRIzol (Life Technologies, Carlsbad, CA, USA). The RNA was reverse-transcribed to cDNA using PrimeScript RT Master Mix (RR036, Takara Bio, Shiga, Japan). Relative mRNA levels of genes that encode pyruvate dehydrogenase lipoamide kinase isozyme 4 (PDK4), glucose transporter 4 (GLUT4), glucose transporter 1 (GLUT1), carnitine palmitoyltransferase-1α (CPT-1α), acyl-CoA thioesterase 1 (ACOT1), acyl-CoA thioesterase 2 (ACOT2), carnitine O-acetyltransferase (CRAT), carnitine O-octanoyltransferase (CROT), and peroxisome proliferator-activated receptors α (PPARα) and β (PPARβ) were determined using the 2−ΔΔCt method. TB Green RR420 was used as the reference gene [Table 1].{Table 1}

Protein analysis by western blotting

Equal amounts of protein extract (15–25 μg) were separated through sodium-dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to a polyvinylidene fluoride membrane. After blocking with 5% bovine serum albumin, the membrane was probed at 4°C overnight with the appropriate dilutions of primary antibodies against the following proteins: β-actin, alpha smooth muscle actin (α-SMA); UCP-3; proliferator-activated receptor-gamma coactivator-1α (PGC-1α), 5' AMP-activated protein kinase (AMPK), caspase 3, B-cell lymphoma 2 (Bcl2), and Bcl-2-associated X protein (Bax) [Table 2]. After washing in Tris-buffered saline with 0.1% Tween 20, membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit or anti-mouse secondary antibodies for 60 min at room temperature. The blots were detected using the Tanon Imaging System (Shanghai, China) in accordance with the manufacturer's instructions. The densities of the protein blots were determined using ImageJ and normalized to the density of the β-actin bands.{Table 2}

Free fatty acid, glucagon, and glucose quantification

Circulating free fatty acids (FFAs) and glucagon were quantified using corresponding ELISA kits as per the manufacturer's instructions: Glucagon ELISA Kit (Linyi Azelasi, Shandong, China (EK-2873)); FFA ELISA Kit (Linyi Azelasi, Shandong, China (EK-2925)). After equilibration at room temperature for 20 min, the strips were taken out from the aluminum foil sealing bags. First, the standard and sample wells were prepared: 50 μl of different concentrations of the standard was transferred into the standard wells; 50 μl of sample was transferred into the sample wells; and no solution was added into the blank well. The HRP-conjugated antibody (100 μl per well) was added to the blank, standard, and sample wells. Afterward, the plates were sealed and incubated in 37°C for 60 min. The solutions were discarded from the wells, and the plates were patted dry on absorbent paper. Then, the following washing steps were repeated 5 times: Each well was filled with 350 μl washing solution, which was discarded after 1 min, and the plates were patted dry on absorbent paper. Next, 50 μl substrate was added into each well and incubated for 15 min at 37°C. Finally, 50 μl termination solution was added to the wells, and the optical density of each well at 450 nm was determined within 15 min. The dry chemistry method was used to measure glucose levels in the plasma.

Uncoupling protein 3 siRNA transfection

Mouse cardiomyocytes HL-1 (obtained from Cell Bank of the Chinese Academy of Sciences [Shanghai, China]) were inoculated into a 6-well plate (4 × 105 cells/well) and cultured overnight. Then, they were washed with PBS twice and stored in 500 μl basic culture medium for later use. UCP3 siRNA (80 pmol; Ucp3-mus-466: Sense (5'-3') CCAUUCGAAUUGGCCUCUATT; Antisense (5'-3') UAGAGGCCAAUUCGAAUGGTT) was mixed with 120 μl serum-free medium solution; then, 12 μl HilyMax transfection reagent (Dojindo, Kumamoto, Japan) was added to the mixture. After 15 min at room temperature, the static solution was added to the prepared mouse cardiomyocytes, and the solutions were mixed gently. The cells were incubated at 37°C for 6 h. The culture medium was replaced with fresh complete medium and drugs (10 μM dapagliflozin), and the cultures were incubated at 37°C for 48 h [Table 3]. To establish the cardiomyocyte remodeling model, we used the AngII-induced (1 μmol AngII) hypertrophy model.[10] The cells were centrifuged for subsequent RT-PCR and western blot analyses.{Table 3}

Statistical analysis

Baseline data are presented as means ± standard deviation for symmetric continuous variables and as median (25% quantile, 75% quantile) for skewed continuous variables. Student's t-test or log-rank test was used to compare symmetric or skewed data among groups. GraphPad Prism (GraphPad, San Diego, CA, USA) was used for statistical analysis and figure construction. A P < 0.05 was considered statistically significant.


Mouse survival

Sixty mice were randomized to undergo TAC or sham operation. Six mice (2 mice each from the TAC, TAC + dapa2, and TAC + dapa3 groups) died during or right after the surgical procedure, and two mice died during the echocardiography-anesthesia procedure (1 each from the dapa and the TAC + dapa1 groups). No mortality was recorded in the other stages of the study. In total, 52 mice (survival rate was 86.7%) survived till the end of the experiments.

Dapagliflozin attenuated left ventricle hypertrophy and restored left ventricle ejection fraction

Echocardiography at week 0 revealed similar LV parameters among all the examined mice. After the 4-week treatment, mice subjected to TAC showed LV hypertrophy with thickened interventricular septum and LV posterior wall, enlarged end diastolic LV internal diameter, and reduced LV ejection fraction (LVEF). However, intragastric administration of dapa attenuated LV hypertrophy and improved LVEF. Moreover, the benefits of dapa were more pronounced in the TAC + dapa2 group than in the rest of the treatment groups [Table 4] and [Figure 1]A, [Figure 1]B, [Figure 1]C, [Figure 1]D, [Figure 1]E, [Figure 1]F.{Table 4}{Figure 1}

The heart weight/body weight ratio (HW/BW) was higher in the TAC group and lower in the dapa treatment groups than in the sham group. In addition, mice in the TAC + dapa2 group had the lowest HW/BW. Although the change in BW was not linear, the TAC + dapa2 group showed the least weight gain among all the groups [Figure 1]G and [Figure 1]H.

Cardiomyocyte morphology was examined based on cell surface features via HE staining. Notably, cardiomyocyte and myocardial fiber arrangement were normal in the sham group. Hypertrophic cardiomyocytes were observed in mice subjected to TAC, and dapa treatment relieved cardiomyocyte hypertrophy [Figure 2].{Figure 2}

Dapagliflozin alleviates fibrosis and apoptosis

Masson's trichrome and TUNEL staining assays were performed to estimate the extent of fibrosis and apoptosis in the ventricle tissue sections. The mice in the sham group showed no significant fibrosis and apoptosis. In contrast, mice subjected to TAC displayed pronounced fibrosis and apoptosis. These characteristics of pathological remodeling were significantly abolished by dapa treatment, with more apparent effects in the TAC + dapa2 group. A similar trend was observed in the western blot for α-SMA, Bcl2, and Bax expression, where in the dapa treatment groups had lower α-SMA levels, which demonstrated less fibrosis and higher Bcl2/Bax ratios, which indicated less apoptosis, than those in the TAC group [Figure 3] and [Figure 4].{Figure 3}{Figure 4}

Changes in free fatty acid, glucagon, and glucose levels

Changes in serum FFAs and glucagon in the six groups were assessed using ELISA. Dapa treatment tended to reduce FFA levels, relative to the sham and the TAC group, and the FFA levels in the TAC + dapa2 group were significantly lower than those in the TAC group. A similar trend was observed for glucagon levels. Mice that received dapa treatment also had lower glucose levels than the mice in the sham and TAC groups [Figure 5].{Figure 5}

Dapagliflozin alters the activity and expression of enzymes involved in glucose and fatty acid metabolism in vitro and in vivo

The expression levels of several key enzymes involved in glucose and FA metabolism were estimated through RT-PCR and western blotting. The level of PDK4, which modulates pyruvate dehydrogenase (PDH) activity during glucose oxidation, was higher in the dapa groups than in the TAC group, suggesting suppressed glucose oxidation. Furthermore, the levels of GLUT4, which regulates glucose transport into cardiomyocytes, were lower in the TAC group than in the sham group and did not significantly improve after dapa treatment [Figure 6]A, [Figure 6]B, [Figure 6]C. The levels of CPT-1α, an enzyme that transports long-chain acyl-CoA moieties into the mitochondria for further FAO, were significantly lower in the TAC group than in the sham group. CPT-1α levels were slightly elevated in the TAC + dapa2 group compared to the TAC group. However, CRAT rather than CROT, which is involved in FAO reactions of medium-chain acyl-CoA moieties in the mitochondria, was also elevated in the dapa treatment groups [Figure 6]D, [Figure 6]F. The levels of ACOT1 and ACOT2, which catalyze the hydrolysis of long-chain acyl-CoA in the cytoplasm and the mitochondria, respectively, showed trends similar to that of CPT-1α. These findings indicate upregulation of FAO after dapa treatment [Figure 6]G and [Figure 6]H. AMPK, PGC-1α, and PPARs mediate metabolic adaptations in response to changes in FAO. Here, we found that PPARβ [Figure 7]A, PGC-1α [Figure 7]D, and AMPK [Figure 7]E had higher expression levels in the dapa treatment groups than in the TAC group. In addition, AMPK levels were higher in the TAC + dapa1 group than in the TAC group [Figure 7]D and [Figure 7]E. On the contrary, PPARα was downregulated in the dapa-treated mice [Figure 7]B. Moreover, UCP3 expression was significantly lower in the TAC group than in the dapa treatment groups [Figure 7]C.{Figure 6}{Figure 7}

In vitro, RT-PCR showed that AMPK expression was lower in the groups treated with UCP3 siRNA group than in the groups without UCP3 siRNA exposure [Figure 8]A. The levels of mRNA that encode the FAO-related protein, especially PPARβ, were substantially lower in the UCP3 siRNA group than in those without UCP3 siRNA [Figure 8]B. The expression of ACOT1 and CPT-1α was not affected by UCP3 siRNA. Furthermore, caspase 3 levels were lower in the groups treated with UCP3 siRNA than in those without UCP3 siRNA exposure [Figure 8].{Figure 8}


Several studies indicate that SGLT2i improve HF prognosis in patients with or without diabetes. Using nondiabetic C57BL/6 male mice with TAC-induced HF, we found that dapa treatment improves LVEF, relieves LV hypertrophy, and reduces fibrosis and apoptosis, which are consistent with results from previous studies.[11],[12] This confirms that SGLT2 inhibition improves cardiac function and alleviates fibrosis and apoptosis in the failing heart.

As the powerhouse of cardiomyocytes, the mitochondria synthesize ATP and fine-tune adaptation to changes in myocardial energetics under specific cellular contexts.[13] Mitochondrial ATP production depends on oxidative phosphorylation of nicotinamide adenine dinucleotide and flavin adenine dinucleotide produced from glucose oxidation and FAO. FAO accounts for 60%–90% of the ATP produced in the myocardium, while glucose oxidation provides 10%–40% of the ATP produced in a healthy adult. HF is accompanied with a shift in cardiac energy substrate from FA to glucose, which causes insufficiencies in cardiac substrates and compromises mitochondrial homeostasis.[14] UCP3-knockout mice exhibited larger infarct size with decreased cardiac function and increased apoptosis than wild-type mice in an ischemic reperfusion injury model.[15] Similarly, knocking out UCP3 has been reported to abolish the cardioprotective effects of H2O2 preconditioning,[16] and upregulation of UCP3 has been shown to suppress mitochondrial ROS production and to alleviate apoptosis.[17],[18] Our preliminary sequencing analysis indicates that SGLT2 inhibition augments UCP3 expression in cardiomyocytes; thus, we speculate that dapa amends metabolic remodeling, favoring FAO in the TAC-induced HF model.

Several studies have confirmed that SGLT2i can activate AMPK and improve mitochondrial function. Here, we explored FAO-related metabolic changes in mice subjected to TAC and then treated with dapa. Our results suggest that dapa treatment inhibits glucose oxidation as indicated by the increase in PDK4 activity. Moreover, the TAC mice treated with dapa had higher levels of FAO, especially higher medium-chain FAO, as demonstrated by increased CRAT, ACOT1, and ACOT2 expression and decreased levels of plasma FFA in vivo. In addition, significant upregulation of UCP3, which assists in ATP synthesis-independent proton leakage and in the optimization of FAO, was observed in dapa-treated mice. Thus, we speculate that the dapa-induced improvement in cardiac function among HF patients reflects the shift from glucose to FA as energy substrate. It is likely that dapagliflozin promoted optimal conditions for mitochondrial biogenesis and increased medium-chain FAO partly through UCP3 upregulation.

As previously mentioned, in the oxidative phosphorylation pathway, mitochondria transmit electrons, gradually releasing energy that drives ADP phosphorylation to ATP. This transforms chemical energy from FA and glucose substrates into mechanical power, providing energy for cardiac contractile function. Glucose, a major fuel in the heart, consumes far less oxygen than FA does in terms of ATP generation. A shift in cardiac energy substrate from FA to glucose has been demonstrated in HF.[19] Glucose oxidation in cardiomyocytes is an adaptive mechanism in the early stages of HF, when oxygen is insufficient. Long-term glucose utilization will eventually lead to the depletion of cardiac substrates. This reduces the activity of the mitochondrial oxidative respiratory chain and compromises mitochondrial homeostasis. Subsequently, electron leakage triggers the generation of reactive oxygen species, which induce cardiac apoptosis.

Our results show that dapa treatment enhanced the expression of PDK4, CRAT, ACOT1, and ACOT2, indicating a reciprocal shift in energy utilization from glucose to FA. Thus, we infer that SGLT2 inhibition augments FAO and then increases ATP production through UCP3 regulation. In diabetes, excessive accumulation of long-chain-acyl-CoAs derived from elevating FAO leads to lipotoxicity and, consequently, to cardiac injury. In our model, elevation in ACOT1 and ACOT2 expression may result in cardioprotection accompanied by enhanced FAO. ACOT1 is a cytosolic enzyme that is highly specific to medium-chain monounsaturated acyl-CoAs and is predicted to be a potent inhibitor of lipotoxicity.[20] Meanwhile, ACOT2 is specific to medium-chain fatty acyl-CoAs in the mitochondrial matrix, suggesting that it regulates mitochondrial FAO.[21],[22] Data have demonstrated that ACOT2 overexpression increases hepatic utilization of FA via an FA export-activation-uptake cycle, thereby optimizing the rate of FAO.[21]

UCP3 is important for regulating redox reactions and elevating FAO in mitochondria. Treatment with dapa increased UCP3 levels. It is therefore likely that UCP3 upregulation was responsible for reduced apoptosis in the dapa-treated TAC mice. However, UCP3 was not responsible for apoptosis alleviation in our model.

Studies have confirmed that UCP3 cannot uncouple protons until UCP3 is activated. A previous study has shown that AMPK and PPARα[23] activated UCP3 by promoting FAO. The expression of PPARα was not elevated in this study, raising the possibility that other molecular mechanisms may be involved in the activation of UCP3. Lima et al.[24] demonstrated that PGC-1α/PPARβ regulates the transcription of UCP3. Our findings show that the transcript levels of the gene that encodes PGC-1α were lower in the TAC group than in the sham group and were significantly increased in the TAC + dapa2 group. Dapa treatment also increased the PPARβ transcript levels, suggesting that PGC-1α and PPARβ were responsible for the upregulation of UCP3 and mediated the effects of dapa on HF progression.


dapa attenuates LV hypertrophy, fibrosis, and apoptosis, and improves LV function in a TAC-induced hypertrophy and HF mouse model. Dapagliflozin may augment medium-chain fatty acid catabolism partially through the UCP3 pathway.


Expression levels of several enzymes involved in fatty acid and glucose oxidation changed in response to dapa treatment in our HF mouse model. This was accompanied by attenuated fibrosis and hypertrophy, leading to improved LV systolic function. However, these results only demonstrate an association between the fatty acid catabolism pathway and improved cardiac function in vitro and in vivo. The association between metabolic modulation and the cardiac benefits of SGLT2 inhibition should be investigated further.

Financial support and sponsorship

This research was supported by the Research Starting Foundation of Huashan Hospital (2020QD009).

Conflicts of interest

Yong Li is Editorial Board members of Cardiology Plus. The article was subject to the journal's standard procedures.


1GBD 2016 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017;390:1211-59. doi: 10.1016/S0140-6736(17)32154-2.
2Cheng YJ, Imperatore G, Geiss LS, Saydah SH, Albright AL, Ali MK, et al. Trends and disparities in cardiovascular mortality among U.S. Adults with and without self-reported diabetes, 1988-2015. Diabetes Care 2018;41:2306-15. doi: 10.2337/dci18-0057.
3Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 2015;373:2117-28. doi: 10.1056/NEJMoa1504720.
4Neal B, Perkovic V, Mahaffey KW, de Zeeuw D, Fulcher G, Erondu N, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017;377:644-57. doi: 10.1056/NEJMoa1611925.
5Wiviott SD, Raz I, Bonaca MP, Mosenzon O, Kato ET, Cahn A, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med 2019;380:347-57. doi: 10.1056/NEJMoa1812389.
6Packer M, Anker SD, Butler J, Filippatos G, Pocock SJ, Carson P, et al. Cardiovascular and renal outcomes with empagliflozin in heart failure. N Engl J Med 2020;383:1413-24. doi: 10.1056/NEJMoa2022190.
7Verma S, McMurray JJ. SGLT2 inhibitors and mechanisms of cardiovascular benefit: A state-of-the-art review. Diabetologia 2018;61:2108-17. doi: 10.1007/s00125-018-4670-7.
8van Bilsen M, Smeets PJ, Gilde AJ, van der Vusse GJ. Metabolic remodelling of the failing heart: The cardiac burn-out syndrome? Cardiovasc Res 2004;61:218-26. doi: 10.1016/j.cardiores. 2003.11.014.
9Liu S, Lam A, Wazir A, Cheema AN. SGLT2 inhibitors and the changing landscape for treatment of diabetes. Ther Clin Risk Manag 2019;15:861-7. doi: 10.2147/TCRM.S207515.
10Chen Y, Yu S, Zhang N, Li Y, Chen S, Chang Y, et al. Atorvastatin prevents Angiotensin II induced myocardial hypertrophy in vitro via CCAAT/enhancer-binding protein β. Biochem Biophys Res Commun 2017;486:423-30. doi: 10.1016/j.bbrc.2017.03.057.
11Shi L, Zhu D, Wang S, Jiang A, Li F. Dapagliflozin attenuates cardiac remodeling in mice model of cardiac pressure overload. Am J Hypertens 2019;32:452-9. doi: 10.1093/ajh/hpz016.
12Lee HC, Shiou YL, Jhuo SJ, Chang CY, Liu PL, Jhuang WJ, et al. The sodium-glucose co-transporter 2 inhibitor empagliflozin attenuates cardiac fibrosis and improves ventricular hemodynamics in hypertensive heart failure rats. Cardiovasc Diabetol 2019;18:45. doi: 10.1186/s12933-019-0849-6.
13Lindquist C, Bjørndal B, Bakke HG, Slettom G, Karoliussen M, Rustan AC, et al. A mitochondria-targeted fatty acid analogue influences hepatic glucose metabolism and reduces the plasma insulin/glucose ratio in male Wistar rats. PLoS One 2019;14:e0222558. doi: 10.1371/journal.pone.0222558.
14Chen L, Song J, Hu S. Metabolic remodeling of substrate utilization during heart failure progression. Heart Fail Rev 2019;24:143-54. doi: 10.1007/s10741-018-9713-0.
15Ozcan C, Palmeri M, Horvath TL, Russell KS, Russell RR 3rd. Role of uncoupling protein 3 in ischemia-reperfusion injury, arrhythmias, and preconditioning. Am J Physiol Heart Circ Physiol 2013;304:H1192-200. doi: 10.1152/ajpheart.00592.2012.
16Chen Y, Liu J, Zheng Y, Wang J, Wang Z, Gu S, et al. Uncoupling protein 3 mediates HWang Z, Gu S, rfusion injury, arrhythmias, and preconditioning. the plasma insulin/glucose ratio in male Wi105:192-202. doi: 10.1093/cvr/cvu256.
17Nabben M, Shabalina IG, Moonen-Kornips E, van Beurden D, Cannon B, Schrauwen P, et al. Uncoupled respiration, ROS production, acute lipotoxicity and oxidative damage in isolated skeletal muscle mitochondria from UCP3-ablated mice. Biochim Biophys Acta 2011;1807:1095-105. doi: 10.1016/j.bbabio. 2011.04.003.
18Nabben M, Hoeks J, Bried. 003.ted skeletal muscle mitochondria from UCP3 et al. The effect of UCP3 overexpression on mitochondrial ROS production in skeletal muscle of young versus aged mice. FEBS Lett 2008;582:4147-52. doi: 10.1016/j.febslet. 2008.11.016.
19Bertero E, Maack C. Metabolic remodelling in heart failure. Nat Rev Cardiol 2018;15:457-70. doi: 10.1038/s41569-018-0044-6.
20Xia C, Dong R, Chen C, Wang H, Wang DW. Cardiomyocyte specific expression of Acyl-coA thioesterase 1 attenuates sepsis induced cardiac dysfunction and mortality. Biochem Biophys Res Commun 2015;468:533-40. doi: 10.1016/j.bbrc.2015.10.078.
21Moffat C, Bhatia L, Nguyen T, Lynch P, Wang M, Wang D, et al. Acyl-CoA thioesterase-2 facilitates mitochondrial fatty acid oxidation in the liver. J Lipid Res 2014;55:2458-70. doi: 10.1194/jlr.M046961.
22Bekeova C, Anderson-Pullinger L, Boye K, Boos F, Sharpadskaya Y, Herrmann JM, et al. Multiple mitochondrial thioesterases have distinct tissue and substrate specificity and CoA regulation, suggesting unique functional roles. J Biol Chem 2019;294:19034-47. doi: 10.1074/jbc.RA119.010901.
23Oka S, Alcendor R, Zhai P, Park JY, Shao D, Cho J, et al. PPARα-Sirt1 complex mediates cardiac hypertrophy and failure through suppression of the ERR transcriptional pathway. Cell Metab 2011;14:598-611. doi: 10.1016/j.cmet.2011.10.001.
24Lima TI, Guimarães D, Sponton CH, Bajgelman MC, Palameta S, Toscaro JM, et al. Essential role of the PGC-1α/PPARβ axis in Ucp3 gene induction. J Physiol 2019;597:4277-91. doi: 10.1113/JP278006.