Gene and Drug-Mediated SERCA2a Activation Restores Cardiac Function and Metabolic Balance in Diabetic Mice

Abstract

Diabetic cardiomyopathy is a major complication of diabetes mellitus, characterised by impaired calcium homeostasis, insulin resistance, and metabolic dysregulation. Notably, the expression and activity of sarco/endoplasmic reticulum Ca²⁺-ATPase 2a (SERCA2a) are diminished in diabetic heart failure, contributing to both systolic and diastolic dysfunction. Although enhancing SERCA2a function has shown beneficial cardiac effects, the underlying metabolic mechanisms remain unclear. The objective of this study was to investigate the effect of pharmacological activation of SERCA2a in the diabetic heart. This study aimed to evaluate the impact of pharmacological activation of SERCA2a on cardiac function and metabolism in a diabetic context. Leptin-deficient (Ob/ob) diabetic mice were treated using two strategies: a novel SERCA2a allosteric agonist (CDN1163) and cardiac-specific overexpression via AAV9-mediated gene delivery. Both approaches attenuated the expression of genes involved in lipid synthesis and fatty acid oxidation, thereby improving lipid homeostasis. Enhanced SERCA2a activity promoted mitochondrial biogenesis, increased mitochondrial DNA content, improved oxidative phosphorylation, and elevated ATPase activity in diabetic hearts. In vitro, SERCA2a restoration in high glucose and H₂O₂-treated H9C2 myocytes normalised mitochondrial membrane potential and improved mitochondrial function. Additionally, SERCA2a activation mitigated lipotoxicity and cell injury through upregulation of antioxidant enzymes and suppression of ROS by inhibiting NADPH oxidase activity. Crucially, these molecular improvements translated into enhanced diastolic function in diabetic mice following both CDN1163 and AAV9-SERCA2a treatments. Collectively, our findings suggest SERCA2a activation may offer a promising therapy for diabetic cardiomyopathy by improving cardiac energetics, lipid metabolism, and mitochondrial function.

Graphical Abstract

1. Introduction

Diabetes and obesity are known to elicit changes in the heart muscle, leading to a condition termed diabetic cardiomyopathy [1, 2]. This cardiac alteration is characterised by disruptions in intercellular calcium homeostasis, myocardial insulin resistance, and cardiometabolic disturbances, all of which are closely associated with mitochondrial dysfunction and shifts in energy substrate utilization [2–10]. The heart, being the most energy-demanding organ in the body, predominantly relies on glucose as its primary substrate for generating ATP. However, in animal models with diabetes and obesity, such as Ob/ob mice, db/db mice, and Zucker diabetic fatty rats, there is a notable preference for fatty acids over glucose as a source for ATP production [4–6, 10–14]. This metabolic shift towards increased fatty acid oxidation results in reduced cardiac efficiency, where more oxygen is required to produce the same amount of ATP compared to glucose metabolism [10, 11]. Moreover, the pathophysiology of diabetic cardiomyopathy involves intricate interactions that lead to structural changes in the heart, including myocardial hypertrophy, increased collagen formation, fibrosis, and alterations in coronary vessels, ultimately culminating in cardiac remodeling and dysfunction [13, 15–17]. Elevated levels of lipid peroxidation and the accumulation of toxic lipids within cardiac cells can contribute to heart dysfunction by inducing oxidative stress, mitochondrial dysfunction, and potentially leading to apoptosis triggered by endoplasmic reticulum stress [18–20]. Furthermore, dysregulation of myocardial energy metabolism, particularly in the context of altered substrate utilization, has been implicated in the progression of diabetic cardiomyopathy [10, 11, 21]. Understanding the molecular mechanisms underlying altered substrate utilization and mitochondrial dysfunction is crucial for developing targeted therapies for diabetic cardiomyopathy.

The activity or expression of sarco/endoplasmic reticulum Ca²⁺-ATPase 2a (SERCA2a) plays a pivotal role in cardiac function, particularly in heart failure and diabetic cardiomyopathy. Studies have shown that the levels of SERCA2a are reduced in hearts subjected to various conditions, including high glucose perfusion, high-fat/high-sugar diets, and in obesity-related heart dysfunction [22–26]. Increasing SERCA2a activity or expression reverses calcium impairment and contractile dysfunction [22, 26–31]. We and others have previously shown that pharmacological activation or overexpression of SERCA influences lipid metabolic pathways [32–34] and cardiac metabolism [22, 35–37].

CDN1163, a quinoline derivative, directly enhances SERCA activity by displacing inhibitors, promoting ATP-dependent calcium transport [38]. In experimental models, CDN1163 improves calcium homeostasis, reduces oxidative stress-related cell damage, and supports glucose and lipid metabolism [32, 34]. Additionally, it mitigates age-related muscle atrophy by restoring SERCA function and lowering mitochondrial ROS production [39]. These findings highlight the potential of CDN1163 as a therapeutic agent for conditions associated with impaired SERCA function.

In this study, we have employed two distinct parallel strategies to enhance either the activity or the amount of SERCA2a in the diabetic hearts: pharmacological activation via the synthetic allosteric agonist, CDN1163, and cardiac-specific adeno-associated virus 9 (AAV9)-mediated gene transfer. Through these methodologies, the effects of SERCA2a activation on lipid homeostasis, lipotoxicity, cellular damage, and mitochondrial function in the context of heart dysfunction associated with obesity were investigated. We found that restoring proper calcium handling through SERCA2a modulation improved lipid homeostasis, mitigated lipotoxicity and cellular damage, and enhanced mitochondrial capacity and preserved cardiac function in the diabetic heart following CDN1163 treatment. By understanding how SERCA2a influences cardiac function and metabolism, we aim to provide new therapeutic interventions that target SERCA2a to improve outcomes in individuals with cardiac pathologies associated with metabolic disorders

2. Materials and Methods

2.1. Animals

Male Ob/ob mice (10–12 weeks old), which are leptin-deficient obese mice, and lean C57BL/6J mice were purchased from Jackson Laboratory. The obese mice were divided into four groups: two groups were injected intraperitoneally with either a vehicle (10% DMSO + 10% Tween 80 in 0.9% NaCl) or CDN1163 (50 mg/kg/day) for five consecutive days, while the other two groups were injected with AAV9.LacZ or AAV9.SERCA2a (3×10^12 viral particles) intravenously. After the respective treatments, the mice were fasted overnight and sacrificed either 7 weeks after CDN1163 administration or 14 weeks after AAV9 injection. The study was conducted following the guidelines approved by the Mount Sinai Institutional Animal Care and Use Committee, adhering to the principles of laboratory animal care.

2.2. Cell Culture

H9C2 myocytes were cultured in high glucose media (33 mmol/L glucose) to mimic diabetic conditions or 5 mmol/L glucose as a control. Cells were incubated with CDN1163 (2 μmol/L in DMSO) for 24 hours. Then, cells were treated with 400 μmol/L H₂O₂ for 2 hours to determine the mitochondrial membrane potential, bioenergetics, and cell death.

2.3. Metabolic parameters

Sixteen-hour fasting blood glucose levels were measured in whole blood drawn from the tail vein using the OneTouch Ultra 2 Meter (LifeScan, Inc.). To determine plasma levels of insulin, triglycerides, cholesterol, free fatty acids, and malondialdehyde (MDA), blood was collected from the retro-orbital plexus, and the parameters were assayed using commercially available kits. Cardiac lipids were extracted using the chloroform-methanol method[40], and MDA was lysed in RIPA buffer. Both were quantified using the corresponding assay kits.

2.4. Real-Time Quantitative PCR

RNA was extracted from heart tissues using TRIzol (Invitrogen), and cDNA synthesis was carried out with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Real-time PCR was performed using iTaq Fast SYBR Green Supermix with ROX (Bio-Rad) on a 7500 Real-Time PCR System (Applied Biosystems). Gene expression was normalized to 18S rRNA. Data from qPCR were analysed using the ΔΔCT method. Primer sequences utilised in this study are listed in the online Supplement Table S2.

2.5. Western blotting

Heart tissues were homogenized in RIPA buffer containing protease and phosphatase inhibitors (Roche). Fifty micrograms of protein were applied to SDS-PAGE and transferred onto a nitrocellulose membrane. The antibodies used were SERCA2a (produced in our lab), PGC1α (Cell Signaling Technology), CV-ATP5A, CIII-UQCRC2, CIV-MTCO1, CII-SDHB, and CI-NDUFB8 (MitoSciences) p22phox and nitrotyrosine-γ (Santa Crus), p47phox and gp91phos(abcam). Protein expression was normalized to GAPDH (Cell Signaling Technology).

2.6. Mitochondrial DNA copy number

Mitochondrial (mtDNA) and nuclear (nDNA) DNA contents were quantified by real-time quantitative PCR using specific primers for mtDNA genes (mtCytB, mtCOX1, mtND1) and the nuclear H19 gene. The mtDNA copy number was determined by calculating the difference in threshold cycle numbers between mtDNA and nDNA. Primer sequences are provided in the online Supplement Table S2.

2.7. Ca²⁺-ATPase activity

Ca²⁺-dependent ATPase activity of SERCA2a was evaluated using a colorimetric ATPase assay kit (Innova Biosciences). Heart homogenates (50 μg) were pre-incubated with the ionophore A23187 and EGTA, each at a final concentration of 1 μg/mL, for 5 minutes to prevent calcium accumulation inside the vesicles that could inhibit Ca²⁺-ATPase activity. The activity rates were measured at 650 nm and normalized to the total protein content, as determined by the Micro BCA protein assay kit (Pierce).

2.8. ATP content

Total cellular ATP levels were assayed using the ATP determination kit (Molecular Probes). The luminescence was measured at 560nM.

2.9. Mitochondrial membrane potential and Cell death ELISA

H9C2 myocytes were cultured in high glucose media (33 mmol/L glucose) to mimic diabetic conditions, with 5 mmol/L glucose as a normal control. Forty-eight hours later, cells were incubated with CDN1163 (2 μmol/L in DMSO) for 24 hours. Cells were treated with 400 μmol/L H₂O₂ for 2 hours to determine mitochondrial membrane potential and cell death. For mitochondrial membrane potential assessment (Abcam; ab112134), the fluorescent intensity of JC-10 was measured with excitation/emission wavelengths of 490/525 nm and 540/590 nm. Cell death (Roche; 11 544 675 001) was determined by the formation of cytoplasmic histone-associated DNA fragments in the form of mononucleosomes after apoptosis. The total amount of apoptosis was calculated by dividing the absorbance of the sample (405 nm) by the absorbance of controls without treatment (490 nm).

2.10. Mitochondrial respiration

The Seahorse XF24 Analyser (Agilent Technologies) quantifies oxygen consumption rates (OCR) in a 24-well format by detecting fluctuations in oxygen levels through a fluorescence-based biosensor. H9C2 myoblasts were plated at a density of 10,000 cells per well in 250 μL of Ham’s F-10 medium containing 5% fetal bovine serum and cultured overnight at 37 °C in a humidified incubator with 5% CO₂. The following day, cells were serum-starved using Ham’s F-10 medium with either low (5.5 mmol/L) or high (30 mmol/L) glucose for 6 hours. After starvation, cells were treated overnight with CDN1163 at concentrations of 2 or 10 μmol/L. Subsequently, cells were exposed to 400 μmol/L hydrogen peroxide (H₂O₂) for 4 hours to induce oxidative stress. All treatments, including controls, contained 0.1% DMSO as the vehicle. Prior to the assay, the culture medium was replaced with 500 μL of Seahorse XF assay medium (catalog #102365–100, Agilent Technologies), and cells were incubated for 45 minutes at 37 °C in a CO₂-free environment to allow for equilibration. The plate was then transferred to the XF24 Analyser for real-time measurement. To assess mitochondrial function, oligomycin (2 μg/mL) was added to inhibit ATP synthase and differentiate between ATP-linked respiration and proton leak. Maximal respiration was evaluated using the mitochondrial uncoupler FCCP (1 μmol/L), while rotenone (1 μmol/L) was used to inhibit electron transfer from complex I to ubiquinone[41]. OCR parameters were calculated using the XF Cell Mito Stress Test Report Generator and analysed in Wave Desktop Software (Agilent Technologies, Santa Clara, CA) and values were normalised to protein content per well. Calculations followed manufacturer-validated algorithms and established protocols (Agilent Technologies)[42], Non-mitochondrial Oxygen Consumption = Minimum rate measurement after Rotenone/antimycin A injection. Basal Respiration= (Last rate measurement before first injection) – (Non-Mitochondrial Respiration Rate). Maximal Respiration= (Maximum rate measurement after FCCP injection) – (Non-Mitochondrial Respiration). H+ (Proton) Leak= (Minimum rate measurement after Oligomycin injection) – (Non-Mitochondrial Respiration). ATP Production= (Last rate measurement before Oligomycin injection) – (Minimum rate measurement after Oligomycin injection). Spare Respiratory Capacity= (Maximal Respiration) – (Basal Respiration)[42]. Wave software applies baseline correction, normalization, and kinetic modelling to ensure reproducibility of analyses.

2.11. Echocardiography

Echocardiography was conducted at the beginning of the study and again after 7 weeks of treatment. All procedures were carried out under sedation using an intraperitoneal injection of ketamine (80–100 mg/kg). The sedation protocol was carefully adjusted to (1) administer the lowest effective dose required to minimize movement and avoid imaging artifacts, and (2) keep the heart rate close to 550 beats per minute. Short-axis parasternal 2D images of the left ventricle (LV) at the mid-papillary level, along with long-axis parasternal views, were obtained using a Vivid7 ultrasound machine equipped with a 14 MHz linear array transducer (i13L probe, GE). Image analysis was performed offline by an investigator blinded to both treatment and time point, using Echopac PC software (GE Vingmed Ultrasound, Horten, Norway). Each measurement was averaged over six cardiac cycles. M-mode echocardiographic measurements of the LV cavity and wall thickness were taken at the mid-papillary level, guided by 2D imaging from the short-axis view. End-diastolic LV internal diameter, anterior wall thickness, and posterior wall thickness were recorded. LV mass was estimated using the cubic formula. Additionally, systolic LV diameter was measured, and both fractional shortening and ejection fraction were calculated. [43].

To assess LV diastolic function, mitral inflow was measured using pulsed-wave Doppler from the apical four-chamber view. To closely approximate physiological conditions while evaluating LV function with ultrasound, we only lightly sedated the animals to maintain higher heart rates, despite acknowledging the challenges associated with noninvasively assessment of LV diastolic function under light anaesthesia [44].

2.12. Statistics

Data are expressed as means ± SEM. The significance of the differences in mean values was evaluated by using Student’s t test or ANOVA(N=5). Values of p < 0.05 were considered to be statistically significant.

3. Results

3.1. SERCA2 reduced cardiac contractile dysfunction in the obese mice.

To investigate the effect of SERCA2 activation with CDN1163 in Ob/ob mice with cardiac dysfunction, left ventricular (LV) function and dimensions were measured by echocardiography at baseline before treatment and 7 weeks post-treatment. Compared to control lean mice, untreated Ob/ob mice displayed depressed cardiac function, increased LV wall thickness, and increased LV diameter (Table 1A). However, the CDN1163 treatment resulted in the attenuation of ventricular dysfunction, as evidenced by improvements in systolic dysfunction and contractility assessed by fractional shortening and ejection fraction (Table 1A). Additionally, CDN1163 attenuated decompensated LV diameters in Ob/ob mice, as assessed by decreases in LV end systolic dimensions (LVESD) and diastolic dimensions (LVEDD) (Table 1A). We further assessed diastolic function using Doppler echocardiography. Interestingly, CDN1163 attenuated diabetes-induced diastolic decline by restoring the mean E wave - the early diastolic peak flow velocity - and the E wave deceleration time DT), important indicators of LV diastolic function (Table 1A). Collectively, these findings document amelioration in global diabetic cardiac function with CDN1163 treatment.

Table 1A.

Echocardiography of Ob/ob mice at baseline and end of follow-up treatment with CDN1163.

	Control (n=6)	Before CDN (n=14)	After CDN (n=14)	P (before vs. after)
Weight (g)	32.6±1.6	57.3±3.8	57.9±3.3	0.47
Heart rate (bpm)	541.6±47.9	549.1±27.4	550.1±23.5	0.91
Systolic function
IVS (mm)	0.86±0.04	1.02±0.09	0.98±0.08	0.18
LVDD (mm)	3.32±0.25	2.97±0.18	2.84±0.17	0.06
PW (mm)	0.86±0.08	0.96±0.10	0.93±0.06	0.2
LVSD (mm)	1.87±0.27	1.54±0.26	1.32±0.22	0.002
LV Mass (mg)	97.2±17.7	100.2±14.1	89.2±12.9	0.03
Relative Wall Thickness	0.52±0.04	0.65±0.09	0.66±0.06	0.7
LVEF (%)	90.6±6.5	84.4±5.5	88.9±4.0	0.0006
LVFS (%)	55.8±6.5	48.4±7.9	53.6±6.5	0.01
Diastolic function
E (cm/s)	79.8±5.6	66.8±10.4	80.3±12.5	0.02
DT (ms)	29.4±3.8	31.7±4.2	28.8±3.8	0.04

For validation of SERCA targeting and for comparative purposes, we also used a gene transfer strategy to overexpress SERCA2a specifically in the heart tissue using adeno-associated virus serotype 9 (AAV9.SERCA2a). The AAV9.SERCA2a gene transfer would allow us to demonstrate whether SERCA small molecules evoke similar dynamics to SERCA2a overexpression on cardiac function and if indeed their bioactivity is mediated through alteration in SERCA2a function, and also to determine the long-term effects of SERCA2a restoration in insulin-resistant mice. Ob/ob mice were infected with a control vector (AAV9-Empty) or AAV9.SERCA2a. As noted earlier, Ob/ob mice exhibited significantly reduced contractility in comparison to control lean animals (Table 1B). Fourteen weeks after gene transfer, the AAV9.SERCA2a group displayed a substantial increase in cardiac contractility compared to the control AAV9-Empty (FS%, 44.57±2.06 vs 59.558±0.73, p=0.0001). Altogether, the data suggest that CDN1163 SERCA2a activation produces similar functional improvement as SERCA2a gene overexpression, thus validating SERCA2a enzyme as a target for gene therapy or pharmacological intervention in diabetes and obesity.

Table 1B.

Echocardiographic parameters following AAV9-SERCA2 overexpression in Ob/ob mice.

	Lean	Ob+Empty	Ob+SERCA2a	P (Ob+Ept Vs. Ob+S2a)
Weight (g)	30.8±1.14	59.1.2±3.38	59.3± 2.39	0.9884
Heart weight(mg)	136.0±12.04	145.7±1.45	143.0±10.65	0.15
HW/BW ratio(mg/g)	5.00 ± 0.33	2.47±0.10	2.55±0.15	0.7
LVFS (%)	60.57±1.54	44.57±2.06	59.558±0.73	0.0001
LVSD (mm)	1.0275±0.08	0.9575±0.03	1.088±0.02	0.0009
Heart rate (bpm)	633±25.4	532.08±24.88	580.4±9.89	0.0001

IVS, interventricular septum; LVDD, left ventricular diastolic diameter; PW, posterior wall; LVSD, left ventricular systolic diameter; LV, left ventricular; LVEF, left ventricular ejection fraction; LVFS, left ventricular fractional shortening; E, mitral inflow peak velocity; DT, E wave deceleration time. Data expressed as mean ± standard deviation. n=

3.2. SERCA2 significantly lowered plasma glucose and triglyceride levels in the Ob/ob mice.

Next, we examined the effects of SERCA2a activation on glucose homeostasis in diabetic mice. Ob/ob were treated with either the small-molecule SERCA2 activator CDN1163 (Ob + CDN) or AAV9.SERCA2a (Ob + S2a). As summarized in Supplementary Table S1, AAV9-mediated SERCA2a overexpression (Ob + S2a) produced a dramatic decrease in fasting glucose versus AAV9.Empty controls (Ob + Ept). CDN1163 treatment also reduced blood glucose (270±29.99 Ob+Vhc Vs 237.33±19.25 mg/dl Ob+CDN), mirroring the improvements we previously reported with CDN1163 treatment [34]. In addition, Ob + S2a mice showed significant drops in circulating insulin, plasma triglyceride, and plasma cholesterol levels (Supplementary Table S1), indicating that restoring SERCA2a function in heart tissue can drive systemic metabolic benefits.

3.3. SERCA2 normalized lipid homeostasis in Ob/ob mouse hearts.

Obesity-associated dysregulation of lipid metabolism- marked by elevated circulating free fatty acids and triglycerides, constitutes an early, potent risk factor for metabolic syndrome, type 2 diabetes, and related cardiac dysfunction. The drops in plasma triglyceride and cholesterol levels following SERCA2a activation prompted us to evaluate the cardiac expression of key genes involved in cardiac lipid metabolism in Ob/ob mice. Both CDN treatment (Fig.1A, left panel) and AAV9.SERCA2a overexpression (Fig.1A, right panel) significantly decreased diabetes-induced expression of lipid oxidation genes, such as fatty acid translocase, carnitine palmitoyltransferase-1β, -2, malonyl-Coenzyme A decarboxylase, and peroxisome proliferator-activated receptor-α.

Figure 1. SERCA2 overexpression and CDN1163 improve lipid homeostasis in Ob/ob mouse hearts.

Quantitative RT-PCR analysis of genes involved in cardiac lipid oxidation (A: FAT, fatty acid translocase; CPT1β, carnitine palmitoyltransferase-1β; CPT2, carnitine palmitoyltransferase-2; MCD, malonyl-CoA decarboxylase; PPARα, peroxisome proliferator-activated receptor-α) and de novo lipogenesis (B: SCD1, stearoyl-CoA desaturase-1; DGAT2, diacylglycerol acyltransferase-2; FAS, fatty acid synthase; ACC1, acetyl-CoA carboxylase-1; SREBP1c, sterol regulatory element-binding protein-1c). Groups are divided as Ln+Vhc (Lean+Vehicle), Ob+Vhc (Obese+Vehicle) and Ob+CDN (Obese+CDN1163). Ln+Ept (Lean + Empty), Ob+Ept (Obese+Empty) and Ob+S2a (Obese+ AAV9.SERCA2a). Data are expressed as means ± SEM. *P<0.05, **P<0.01, ***P <0.001 and ****P <0.0001. P-values are derived from two-way ANOVA.

Similarly, CDN treatment (Fig.1B, left panel) and AAV9.SERCA2a overexpression (Fig.1B, right panel) repressed diabetes-induced expression upregulation of genes involved in de novo lipogenesis, namely stearoyl-CoA desaturase-1, diacylglycerol acyltransferase-2, fatty acid synthase, acetyl-CoA carboxylase-1, and sterol regulatory element binding protein-1c to control lean mice levels. These remarkable metabolic properties displayed by SERCA2 overexpression and CDN1163 demonstrate their ability to favourably modulate cardiac lipid homeostasis under obese diabetic conditions

3.4. CDN1163 promoted Mitochondrial Respiratory function.

To further gain deeper insights into the mechanisms underlying CDN1163’s metabolic effects, we assessed mitochondrial respiratory function and energy metabolism in H9C2 cardiomyoblasts, which were selected because Seahorse XF assays require intact adherent cells for accurate oxygen consumption measurements, a condition technically unfeasible in whole heart tissue preparations. Moreover, H9C2 cells share metabolic and mitochondrial features with cardiomyocytes and provide a reproducible model to study bioenergetic responses under controlled high-glucose and oxidative stress conditions. As anticipated, mitochondrial oxidative phosphorylation, measured by the oxygen consumption rate (OCR), was lower in the high glucose group compared to the control. Remarkably high glucose exposure resulted in reductions in basal respiration (Figure 2B), maximal respiration (Figure 2C), ATP production (Figure 2D), proton leak levels (Figure 2G), spare respiratory capacity (Figure 2F), and non-mitochondrial oxygen consumption (Figure 2E) in H9C2 cells. CDN1163 treatment notably countered these effects, significantly increasing basal respiration (Figure 2B), maximal respiration (Figure 2C), and proton leak levels (Figure 2G), which were diminished by high glucose. Additionally, CDN1163 substantially enhanced ATP production (Figure 2D). The improvements in respiration induced by CDN1163 were also evident in the spare respiratory capacity of mitochondria, where it increased spare oxidative capacity and mitigated the high glucose-induced reduction in spare respiratory capacity (Figure 2F). These findings suggest that CDN1163’s potential to protect against high glucose-induced mitochondrial dysfunction in H9C2 cells is likely linked to its ability to enhance ATP production.

Figure 2: CDN1163 improves mitochondrial respiration.

Effect of CDN1163 on the oxygen consumption rate of H9C2 cells. (A) Mitochondrial respiration, OCR of H9C2 treated with CDN1163 in the presence of High glucose, (B) Basal respiration, (C) Maximal respiration, (D)ATP production, (E) Non-mitochondrial oxygen consumption, (F) Spare respiratory capacity, (G) Proton leak. Data are expressed as means ±SEM *P<0.05, **P< 0.01, ***P <0.001 and ****P <0.0001. P-values are derived from two-way ANOVA.

3.5. SERCA2 enhanced mitochondrial Capacity in Ob/ob mouse hearts and H9C2 myocytes.

Since CDN1163 enhanced ATP production (Figure 2D), we determined SERCA2a ATPase activity in microsomes isolated from heart tissues from the indicated groups in figure 3A. Diabetes-depressed ATPase activity was reversed by CDN1163 treatment (Ob+CDN; Fig. 3A, left panel) or AAV9-SERCA2a (Ob+S2a; Fig.3A, right panel) compared to vehicle (Ob+Veh) or AAV9-Empty (Ob+Ept), respectively. To further evaluate the impact of CDN1163 treatment on mitochondrial efficiency in cardiac cells, we analysed the relative levels of mitochondrial-encoded proteins involved in the OXPHOS pathway by Western blotting. Protein levels of ATP synthase, H⁺ transporting, mitochondrial F1 complex, alpha subunit (Atp5A) of complex V, cytochrome c oxidase I (MTCO1) of complex IV, ubiquinol cytochrome c reductase core protein 2 (UQCRC2) of complex III, succinate dehydrogenase [ubiquinone] iron sulfur subunit (SDHB) of complex II, and NADH dehydrogenase (ubiquinone) 1 beta subcomplex subunit 8 (NDUFB8), a representative of complex I were increased in mitochondria from CDN1163-treated Ob/ob mice compared to Ob/ob vehicle-treated or expressing AAV9-SERCA2a (Fig.3B). Consistent with Seahorse assay results, these data support a role for SERCA2a in reversing obesity/diabetes-induced cardiac mitochondrial defect.

Figure 3: Upregulation of ATPase activity and mitochondrial-related protein by SERCA2a overexpression and CDN1163 (N=5).

(A) ATPase activity, (B) Protein expression levels of oxidative phosphorylation (OXPHOS) complexes, mitochondrial biogenesis markers, and SERCA2a(N=5). Data are expressed as means ±SEM *P<0.05, **P< 0.01, ***P <0.001 and ****P <0.0001. P-values are derived from two-way ANOVA.

Furthermore, Mitochondrial biogenesis was decreased in obese mice, as indicated by the downregulation of PGC1α (Fig. 3B). However, SERCA2 overexpression and CDN1163 treatment upregulated the protein expression of PGC1α (Fig.3B). SERCA2a’s improvement of mitochondrial biogenesis was further indicated by the enhanced expression of estrogen-related receptor-α (ERRa), nuclear respiratory factor-1 (NRF1), and mitochondrial transcription factor A (TFam) (Fig. 4A), along with elevated mitochondrial DNA content, shown by increased levels of cytochrome c oxidase subunit-1 (mtCOX1), cytochrome-b (mtCytB), and NADH-ubiquinone oxidoreductase subunit-1 (mtND1) (Fig. 4B). These findings support the notion that fixing calcium cycling via SERCA2a not only promotes the PGC1α axis but triggers new mitochondrial capacity, providing a mechanistic bridge between calcium homeostasis and mitochondrial health in obese hearts.

Figure 4. SERCA2 enhanced mitochondrial capacity and function in Ob/ob mouse hearts and H9C2 myocytes.

Quantitative RT-PCR analysis of genes involved(N=5) in (A) mitochondrial DNA content and (B) mitochondrial biosynthesis. (C) Mitochondrial membrane potential (ΔΨm) assessed using JC-10 staining. (D) Cell death measured by ELISA-based detection of cytoplasmic histone-associated DNA fragments. Data are expressed as means ±SEM *P<0.05, **P< 0.01, ***P <0.001 and ****P <0.0001. P-values are derived from two-way ANOVA.

To assess mitochondrial membrane potential (Δψm) as an important parameter of mitochondrial function and an indicator of cytotoxicity, H9C2 myocytes cultured in the presence of the fluorescent probe JC-10 were employed. Mitochondrial membrane potentials were significantly depolarized in high glucose and/or superoxide (H₂O₂)-treated H9C2 myocytes (Fig. 4C). However, CDN1163 treatment restored Δψm in the high glucose and H₂O₂-treated cells (Fig. 4C). Additionally, CDN1163 protected against cell death induced by high glucose and H₂O₂ oxidative stress in H9C2 myocytes (Fig. 4D), suggesting that restoring SERCA2a activity protects against diabetic and/or oxidative stress-induced cell injury.

3.6. SERCA2a diminished lipid peroxidation and enhanced antioxidant potential in the heart of obese mice.

Although oxidative phosphorylation is indispensable for ATP production, exaggerated respiratory activity inevitably generates more reactive oxygen species (ROS), which can cause cellular injury. We therefore examined how SERCA2a impacts mitochondrial antioxidant defences. SERCA2a overexpression (Fig. 5A, right panel) and CDN1163 treatment (Fig. 5A, left panel) led to the upregulation of antioxidant enzymes glutathione peroxidase-1 (GPX1), peroxiredoxin-3 (PRDX3), superoxide dismutase-2 (SOD2) (Fig. 5A), and the downregulation of ROS-generating NADPH oxidases, p22phox, p47phox, gp91phox and nitrotyrosine-γ (Fig. 5B). Moreover, in line with the enhancement in antioxidant enzyme activity, lipid peroxidation was significantly lower in the hearts of obese mice treated with CDN1163, compared to vehicle-treated controls, or received AAV9.SERCA2a gene transfer. This was evident from the significantly decreased formation of malondialdehyde (MDA), a marker of lipid peroxidation (Fig. 5C), potentially suggesting that SERCA2a plays a key role in neutralizing excess ROS and might attenuate ROS-induced apoptosis in the heart of obese mice.

Figure 5. SERCA2a diminished lipotoxicity, upregulated the expression of genes involved in antioxidant enzymes and attenuated ROS-induced cell damage in the heart of obese mice.

(A)Gene expression of Antioxidant enzymes (B), Protein expression of ROS-generating NADPH oxidase complex (C), Heart Malondialdehyde. Data are expressed as means ±SEM *P<0.05, **P< 0.01, ***P <0.001 and ****P <0.0001. P-values are derived from two-way ANOVA

4. Discussion

Our study demonstrates that SERCA2a exerts a profound influence on cardiac metabolism and function in obesity-induced heart dysfunction. Specifically, we provide the first evidence that pharmacological activation of SERCA2a with CDN1163, and parallel AAV9.SERCA2a gene transfer improves both cardiac performance and metabolic homeostasis in Ob/ob mice and H9c2 myocytes. By dissecting three interrelated themes - Metabolic Remodelling, Mitochondrial Rescue, and Oxidative Stress - we integrate our data with emerging literature to underscore the multifaceted cardioprotective actions of SERCA2a.

Influence on Contractile and Metabolic Remodelling

Growing evidence indicates that diabetes directly impairs left ventricular (LV) function, independent of coronary artery disease or hypertension [7, 9, 10, 25]. This contractile decline is coupled with a metabolic shift: reduced glucose utilization and increased fatty acid oxidation [1, 8, 13]. Depressed cardiac function present in Ob/ob mice was significantly improved following pharmacological activation of SERCA2a with CDN1163, a finding that was also exhibited with AAV9.SERCA2a gene transfer. The attenuation of ventricular dysfunction observed in the study aligns with prior interventions, such as gene therapy and caloric restriction, that reverse ventricular dysfunction, hypertrophy, inflammation and oxidative stress in obese and diabetic rodents [26, 28, 29, 45].

Lipotoxic cardiomyopathy, driven by excess fatty acids and ceramides, contributes to heart failure in obese individuals [46]. Our data show that SERCA2a activation via CDN1163 or AAV9.SERCA2a enhances SERCA2a Ca²⁺-dependent ATPase activity (which is consistent with improved Ca²⁺ transport function, although direct Ca²⁺ uptake was not measured), lowers blood glucose and insulin, and rebalances cardiac lipid metabolism in Ob/ob mice. By normalizing the expression of lipid-oxidation and lipogenic genes to lean-mouse levels, these interventions reduced plasma triglycerides and myocardial steatosis, thereby reestablishing balanced substrate utilization. Hyperlipidemia shifts cardiac energy use toward long-chain fatty acids and away from glucose oxidation. Normally, non-diabetic hearts switch to glucose under stress, but high intracellular non-esterified fatty acids in diabetic (and, to a lesser extent, obese) hearts inhibit glycolysis and glucose oxidation [47, 48], increasing their ischemic vulnerability. This metabolic inflexibility also correlates with insulin resistance and reduced mitochondrial oxidative capacity in Ob/ob mice[48], as well as impaired cardiac performance in obese humans [49]. Restoring SERCA2a function with CDN1163 helps reverse these maladaptive changes and shifts the heart away from damaging pathways like excessive fatty acid oxidation and lipogenesis and toward more efficient glucose metabolism, thereby preserving heart function under obesity-induced stress. These results extend prior evidence that SERCA2a’s control of calcium handling, contractility, and transcription is vital for cardiac health, and that its loss predisposes to heart failure [50, 51]. Additionally, our results suggest that CDN1163 may mitigate dyslipidemia by suppressing SREBP1, a key transcription factor governing lipogenic gene expression [52]. This suppression appears to reduce cardiac lipotoxicity, downregulate lipogenic genes, and potentially decrease the rate of de novo fatty acid and triglyceride synthesis. These findings are consistent with prior reports showing that short-term SERCA2b gene transfer [53] or the attenuation of ER stress via chemical or molecular chaperones [34, 54, 55] in Ob/ob mice reduces SREBP1c activity and lipogenesis, leading to marked improvements in hepatic lipid accumulation and insulin sensitivity. Therefore, it is conceivable to speculate that modulating SERCA2a cardiac activity may improve not only heart metabolism but also potentially systemic metabolic dysregulation as well. We previously demonstrated that CDN1163 may confer protective metabolic benefits through SERCA2b activation of AMPK, leading to improvements in glucose homeostasis, ER stress, and mitochondrial dysfunction in liver tissue [34]. This is a key pathway in reversing insulin resistance and metabolic inflexibility[56]. Other studies have reported that CDN1163-induced activation of SERCA improves energy metabolism in human myotubes, which may be beneficial for conditions associated with skeletal muscle diseases and type 2 diabetes mellitus [32, 34]. Overall, these studies collectively support the notion that targeting SERCA2a to enhance its enzymatic activity and downstream metabolic effects can have beneficial outcomes in metabolic disorders like diabetes, thereby establishing SERCA2a as a critical regulator of cardiac energetics and metabolic balance.

It is important to note that previous clinical attempts to restore SERCA2a expression using AAV1 vectors in advanced heart failure (CUPID-2 trial)[57] did not achieve clinical efficacy, despite encouraging preclinical and pilot data. This outcome highlights the need to revisit vector design, delivery efficiency, disease stage, and patient heterogeneity. In the present study, our rationale for using AAV9-SERCA2a is distinct from CUPID-2 in several ways. Vector choice: AAV9, which has superior myocardial tropism compared with AAV1, may provide more efficient gene delivery. Disease context: Our study focuses on diabetic cardiomyopathy, which is driven by metabolic and calcium-handling defects, rather than advanced end-stage heart failure with extensive remodelling, fibrosis, and comorbidities. Intervening at this earlier disease stage may provide a therapeutic window where restoring SERCA2a is more effective. Complementary approach: We used AAV9-SERCA2a in parallel with pharmacological activation (CDN1163), demonstrating convergent benefits on bioenergetics, lipid metabolism, and oxidative stress. This supports the concept that SERCA2a modulation remains a viable therapeutic target when combined with improved vectors or delivery methods.

It is also reasonable to consider the concern that SERCA2a up-regulation in diabetic models (or patients) might promote RyR2 leakiness, potentially worsening cardiac function[58]. It is important to notice that SERCA2a manipulation in the Liu et al. [58] study was performed in a calsequestrin (CASQ2)-deficient mouse model, and therefore the arrhythmogenic phenotype observed may reflect the combined influence of CASQ2 absence, elevated calreticulin and RyR2 expression [59–62], rather than SERCA2a overexpression alone. This context warrants careful consideration when interpreting their findings. Furthermore, targeting SERCA2a with CDN1163 demonstrates a high degree of disease-specific selectivity. In healthy cells, normal SR/ER Ca²⁺ levels suppress SERCA2 activity, effectively overriding any allosteric modulation. Under physiological conditions, SERCA2 activity is already in excess, and cytoplasmic Ca²⁺ concentrations are maintained at approximately 10 nM or lower. At these low levels, SERCA2 exhibits minimal activity, and allosteric modulation has negligible effect. In contrast, pathological states such as diabetes are characterized by elevated resting cytoplasmic Ca²⁺ (pCa ≤ 7), where SERCA2 activity becomes functionally relevant and responsive to allosteric activation. For example, we and others showed that SERCA2b activation in euglycemic lean mice does not induce hypoglycemia, whereas in diabetic ob/ob mice, SERCA2b activation significantly lowers blood glucose levels [34, 53]. These findings underscore the context-dependent efficacy of SERCA2 modulation and help explain why increased SERCA activity—whether through genetic mutations, pharmacological agents, or overexpression—has not been associated with pathological outcomes. Additionally, our interventions using CDN1163 and AAV9-SERCA2a led to reductions in oxidative stress markers (e.g., decreased NOX subunits and MDA; increased GPX1, PRDX3, and SOD2) and improved diastolic function as assessed by echocardiography. Given that RyR2 leak is exacerbated by oxidative modifications, these molecular changes are expected to stabilize RyR2 and reduce leakiness, rather than promote it.

Mitochondrial Rescue

To further explore the metabolic benefits of CDN1163 SERCA2a activation, we investigated its effects on mitochondrial respiratory function and energy metabolism. Hyperglycemia and diabetes reduce mitochondrial respiration, lower ATP synthesis, and cause membrane potential loss. SERCA2 deficiency exacerbates these defects by depleting ER Ca²⁺ and overloading mitochondria with Ca²⁺, leading to Complex III downregulation and ATP decline [63], potentially leaving cardiomyocytes energy-starved, prone to oxidative damage, apoptosis, and maladaptive remodelling. We show that treatment with CDN1163 significantly increased OCR levels and ATP production, and preserved mitochondrial membrane potential under stress conditions, indicating an enhancement in mitochondrial function and protection against high glucose/diabetes-induced mitochondrial deficiency. By enhancing SERCA2a ATPase activity in Ob/ob mice, CDN1163 may relieve ER stress and improve mitochondrial-ER communication, thereby supporting mitochondrial function. Because mitochondria depend on Ca²⁺ for the TCA cycle and ATP synthesis, normalizing SR Ca²⁺ signalling interrupts a self-amplifying cycle of ER stress and mitochondrial dysfunction. This rescue of ER mitochondrial crosstalk augments mitochondrial biogenesis, mtDNA content, and OXPHOS capacity, as shown by the upregulation of NRF1, ERRα, TFAM, and PGC1α, increased mtDNA levels, and enhanced respiratory function. CDN1163-induced PGC1α elevation, in association with ERRα and NRF1, drives mitochondrial gene expression and amplifies TFAM and other electron-transport-chain subunits, promoting mtDNA stability, transcription, and replication. These findings are consistent with previous reports highlighting the importance of mitochondrial protection in preventing cellular damage in pancreatic MIN6 cells and in renal proximal tubules, where it rescued mitochondrial homeostasis disrupted by SERCA2 oxidation [33, 64]. Beyond the heart, CDN1163 also enhanced endothelial function and skeletal muscle oxygen consumption in diabetic mice, indicating systemic metabolic benefits of SERCA2 activation[65]. Therefore, proper function of SERCA2a via CDN1163 activation ensures that mitochondria receive rhythmic Ca²⁺ pulses to influence ATP synthesis, while preventing the chronic Ca²⁺ overload that triggers bioenergetic collapse and cell death.

Oxidative Stress Attenuation

Obesity and diabetes enhance cardiac oxidative stress via NADPH oxidase overactivation and mitochondrial electron-transport uncoupling, fostering lipid peroxidation and apoptosis [66–68]. In the failing heart, oxidative stress is primarily attributed to the functional uncoupling of the respiratory chain due to complex I inactivation[66]. Modulation of SERCA2a activity appears to play a significant role in mitigating oxidative stress, improving antioxidant defences, and potentially preventing cell death in obese and diabetic hearts. We demonstrate that SERCA2a activation via CDN1163 or gene transfer significantly lowers malondialdehyde levels, indicators of lipid peroxidation, restores antioxidant-enzyme expression and suppresses ROS-generating NADPH-oxidase subunits in obese mouse hearts. This SERCA2a-evoked improvement in defence capacity may potentially lead to a reduction in ROS-induced apoptosis. In vitro, CDN1163 protected H9C2 myocytes from high glucose- and H₂O₂-induced apoptosis, underscoring its capacity to inhibit ROS-driven cell death. These observations parallel our previous report, where CDN1163 SERCA2b activation attenuated ER stress and reduced oxidative stress in the liver of Ob/ob mice [34]. Altogether, restoration of SERCA2a activity - via administration of small-molecule activator CDN1163 or targeted gene therapy - improves mitochondrial efficiency, by maximizing ATP production per unit of oxygen consumed, attenuates calcium overload and limits reactive oxygen species generation, thereby reducing oxidative stress and safeguarding cardiomyocyte viability, resulting in improved myocardial contractility and cardiac output and performance.

5. Conclusion

Overall, the findings of this study provide strong evidence for the beneficial effects of SERCA2a on cardiac metabolic disorders associated with obesity and type 2 Diabetes. SERCA2a interventions resulted in the normalization of metabolic parameters, improvement in lipid homeostasis, enhancement of mitochondrial bioenergetics, attenuation of lipid peroxidation, upregulation of antioxidant enzymes, and repression of ROS-induced cell death. These findings position SERCA2 as a promising therapeutic target for managing cardiac disorders in obesity and type-2 diabetes. Future work should focus on strategies aimed at improving SERCA2a function, dissecting the precise molecular pathways by which SERCA2 overexpression and CDN1163 confer cardiometabolic protection and optimizing these strategies for clinical translation. The CDN1163 compound activates SERCA2 and is not isoform-specific, which is a limitation of this compound. Future efforts should focus on developing SERCA2a-selective activators or cardiac-targeted delivery strategies to minimize off-target effects in non-cardiac tissues.

Highlights.

• Dysregulation of SERCA2a impairs calcium handling and contributes to the pathogenesis of heart failure, muscle disorders, and diabetes.

• SERCA2a overexpression and pharmacological activation with CDN1163 improve ventricular function and attenuate cardiac dysfunction in Ob/ob mice.

• These interventions enhance glucose and lipid metabolism, improve mitochondrial function, and reduce oxidative stress.

• SERCA2a activation represents a promising therapeutic strategy for obesity-related cardiac dysfunction.

• Targeting SERCA2a may improve overall metabolic health and reduce complications associated with diabetes and obesity-related cardiovascular disease.

Abbreviations

SERCA2

Sarco/endoplasmic reticulum Ca²⁺-ATPase 2

SCD1

stearoyl-CoA desaturase-1

DGAT2

diacylglycerol acyltransferase-2

FAS

fatty acid synthase

ACC1

acetyl co-A carboxylase-1

SREBP1c

sterol regulatory element binding protein-1c

FAT

fatty acid translocase

CPT1β

CPT2 carnitine palmitoyltransferase-1β, −2

MCD

malonyl-Coenzyme A decarboxylase

PPARα

peroxisome proliferator-activated receptor-α

TG

triglyceride

FFA

free fatty acids

MDA

malondialdehyde

ERRα

estrogen-related receptor-α

NRF1

nuclear respiratory factor-1

TFAm

mitochondrial transcription factor A

mtCOX1

cytochrome c oxidase subunit-1

mtCytB

cytochrome-b

mtND1

NADH-ubiquinone oxidoreductase subunit-1

PGC1α

peroxisome proliferator activated receptor γ coactivator-1α

GPX1

glutathione peroxidase-1

PRDX3

peroxiredoxin-3

SOD2

superoxide dismutase-2

ROS

reactive oxygen species

ANOVA

Analysis of variance

EF or LVEF

Left ventricular ejection fraction

FS%

Fractional shortening percentage

Δψm

Mitochondrial membrane potential

OCR

Oxygen consumption rate

LVEDD / LVESD

Left ventricular end-diastolic/systolic diameter

IVS

Interventricular septum

PW

Posterior wall

AAV9

Adeno-associated virus serotype

Nitro-γ

Nitrotyrosine-γ

ATP5A, UQCRC2, MTCO1, SDHB, NDUFB8

Subunits of mitochondrial OXPHOS complexes

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.