Combination of genetic eNOS deficiency and high-fat diet induces reproducible cardiometabolic HFpEF

Abstract

Heart failure with preserved ejection fraction (HFpEF) is a complex, multi-organ cardiometabolic syndrome lacking universally effective therapies. We developed and validated a robust murine model leveraging complete, constitutive genetic endothelial nitric oxide synthase deficiency (eNOS KO) coupled with metabolic stress via a high-fat diet (HFD) in both male and female cohorts. eNOS KO mice subjected to HFD (10 weeks for males; 15 weeks for females) developed a severe and highly reproducible cardiometabolic HFpEF phenotype. Despite preserved left ventricular ejection fraction, both sexes exhibited profound diastolic dysfunction, characterized by elevated E/e’ ratios and left ventricular end-diastolic pressures (LVEDP), alongside severe exercise intolerance. Consistent with the systemic clinical syndrome, pathological remodeling extended to multi-organ damage, featuring prominent cardiac fibrosis, severe hepatic steatosis, and renal tubulointerstitial fibrosis. Finally, chronic administration of the dual GIP/GLP-1 receptor agonist tirzepatide largely reversed the hemodynamic, functional, and systemic fibrotic derangements. In conclusion, this novel eNOS KO mouse + HFD model provides a rapidly developing, highly penetrant, and chemically unconfounded platform that ensures robust HFpEF induction across both sexes. This model can be readily adopted for the investigation of mechanistic insights into HFpEF pathobiology and for the study of novel HFpEF therapeutics.

Graphical abstract

Highlights

• •ENOS knockout plus high-fat diet drives rapid, robust, multi-organ HFpEF in mice.
• •This genetic-metabolic model induces HFpEF in both male and female mice.
• •Tirzepatide reverses HFpEF hemodynamics and multi-organ fibrosis.

Non-standard abbreviations and acronyms

eNOS

endothelial nitric oxide synthase

HFD

high-fat diet

HFpEF

heart failure with preserved ejection fraction

LVEDP

left ventricular end-diastolic pressure

LVEF

left ventricular ejection fraction

1. Introduction

Heart failure with preserved ejection fraction (HFpEF) has emerged as the dominant form of heart failure, now accounting for more than 50% of diagnoses in the United States [1]. Driven by the convergence of aging, obesity, and hypertension, HFpEF carries a grim prognosis with a 5-year mortality rate exceeding 50%; and curative therapies remain elusive [2,3]. A critical and necessary component of therapeutic discovery is preclinical models that faithfully recapitulate the chronic, systemic nature of the human syndrome [4].

We validated a new murine cardiometabolic model leveraging constitutive, genetic endothelial NOS deficiency (eNOS KO). We hypothesized that eNOS KO mice subjected to metabolic stress (HFD) would develop a stable and robust HFpEF phenotype that results from the combination of severe hypertension and obesity. Here, we report that this genetic-metabolic model system recapitulates the key HFpEF criteria—preserved ejection fraction, diastolic dysfunction, and exercise intolerance—alongside systemic multi-organ fibrosis, and hepatic steatosis offering a novel and clinically relevant platform for translational research.

2. Methods

2.1. Animals and study design

Homozygous eNOS knockout mice (eNOS KO; B6.129P2-Nos3^tm1Unc/J, Stock No. 002684) on a C57BL/6J background, along with appropriate wild-type control mice, were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals were housed under standard laboratory conditions at a controlled temperature of 23-26°C and humidity of 30-70%, with a 12-h light/dark cycle and ad libitum access to water and standard chow unless otherwise specified. All animal procedures were conducted in accordance with the ARRIVE guidelines and approved by the Institutional Animal Care and Use Committee (IACUC) at Cedars-Sinai Medical Center.

To comprehensively evaluate the HFpEF phenotype and therapeutic interventions, mice were assigned to three distinct experimental protocols: Experiment 1 (Male validation): Two groups of male mice were utilized. The control group was fed standard chow, while the experimental eNOS KO group received a 60% kcal high-fat diet (HFD; D12492, Research Diets) for 10 weeks. Experiment 2 (Female validation): Two groups of female mice were evaluated to assess sex-specific responses. The control group received standard chow, while the eNOS KO group was fed the HFD for 15 weeks. Experiment 3 (Pharmacological intervention): Two groups of male eNOS KO mice were fed the HFD for 15 weeks. After 5 weeks of HFD feeding, mice were randomized to receive either vehicle (0.05% Tween-80 in PBS) or the dual GIP/GLP-1 receptor agonist tirzepatide (Cayman Chemical, Item No. 39748) administered subcutaneously at 30 nmol/kg twice weekly for the remaining 10 weeks. Body weights were recorded weekly for all animals across the three experiments.

2.2. Exercise capacity testing

Treadmill exercise capacity was evaluated at baseline and every 5 weeks utilizing an IITC Life Science 800 Series rodent-specific treadmill (Woodland Hills, CA) as described previously [[5], [6], [7]]. Animals were initially placed on the treadmill for a 15-min acclimation period. Following acclimation, mice underwent a 15-min warm-up protocol on a flat incline, beginning at a speed of 1 m/min and increasing by 1.1 m/min until a speed of 12 m/min was achieved at the 10-min mark. This speed was maintained for the remaining 5 min. Immediately following the warm-up, the graded exercise testing protocol was initiated at 12 m/min. The speed was incrementally increased by 2 m/min until reaching a maximum speed of 18 m/min, which was maintained until the mouse reached a state of physical exhaustion.

2.3. Transthoracic echocardiography

Cardiac structure and function were evaluated non-invasively at baseline and every 5 weeks using a Vevo 3100 high-resolution ultrasound imaging system (VisualSonics, Toronto, Canada) [[5], [6], [7]]. After appropriate hair removal, animals were induced with 3% isoflurane and maintained under 1–3% isoflurane anesthesia during the echocardiographic assessment. To ensure physiological accuracy, diastolic parameters were recorded at a target heart rate of >400 beats per minute (bpm), and systolic measures were recorded at >450 bpm. Left ventricular ejection fraction (LVEF) was quantified using M-mode imaging in the parasternal long-axis view. Early mitral inflow velocity (E), late atrial filling velocity (A), early diastolic tissue velocity (E’), and isovolumetric relaxation time (IVRT) were assessed utilizing the apical four-chamber view.

2.4. Systemic and left ventricular hemodynamic measurements

At the designated study endpoints, systemic and left ventricular pressures were measured as previously described [7,8]. Briefly, animals were anesthetized with 3% isoflurane until completely unresponsive to external stimuli. The right common carotid artery was surgically isolated. Following exposure of the artery, isoflurane was reduced to 1% to stabilize hemodynamics. A 1F high-fidelity solid-state pressure Millar catheter (SPR-1000, ADInstruments) was inserted into the carotid artery to record systemic arterial blood pressures. The catheter was subsequently advanced retrograde through the aortic valve into the left ventricular lumen to continuously measure left ventricular end-diastolic pressure (LVEDP) [7,8].

2.5. Quantitative real-time PCR

Total RNA was extracted from cardiac tissue utilizing TRIzol Reagent (Life Technologies Corporation, USA). First-strand cDNA was synthesized using the iScript Reverse Transcription Supermix (Bio-Rad, USA). Quantitative real-time PCR (RT-qPCR) was performed utilizing SYBR Green qPCR Master Mix (Selleck Chemicals, USA) on a CFX Duet Real-Time PCR System (Bio-Rad, USA). Relative mRNA expression levels were calculated using the 2^–ΔΔCt method, normalized to the housekeeping gene HPRT1. All primer sequences utilized in this study are detailed in Supplemental Table S1.

2.6. Histology and fibrosis quantification

Cardiac, renal, and hepatic tissues were harvested and fixed in a 10% zinc formalin buffered solution for 48 h. Tissues were subsequently transferred to a storage solution consisting of 0.01% sodium azide in PBS. Samples were dehydrated, embedded in paraffin, and sectioned at a thickness of 5 μm. Sections were stained with either hematoxylin and eosin (H&E) for morphological assessment and hepatic lipid visualization, or Picrosirius Red with Fast Green counterstain for the identification of collagen deposition [[5], [6], [7], [8], [9]]. Digital images were acquired and processed using QuPath software. Quantitative analysis of cardiac interstitial and perivascular fibrosis, renal tubulointerstitial fibrosis, and hepatic lipid content was performed using ImageJ software. Images were converted to 8-bit grayscale within an RGB stack, and color thresholds were standardized to accurately isolate and quantify the percentage area of fibrosis and lipid droplet area [[5], [6], [7], [8]].

2.7. 8-Isoprostane ELISA

Plasma levels of 8-isoprostane, a marker of oxidative stress, were quantified using a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Item No. 516351, Cayman Chemical, USA), strictly adhering to the manufacturer's instructions.

2.8. Cardiac nitrite measurements

Nitrite concentrations in myocardial tissue homogenates were determined using a specialized high-performance liquid chromatography (HPLC) system equipped with a NO-specific detector (ENO-30 Analyzer; Amuza Inc, San Diego, CA) [5,[9], [10], [11]]. Biological samples were subjected to methanol precipitation to effectively remove contaminating lipids and proteins. The resulting supernatants were injected into the analyzer for precise nitrite quantification. A standard curve was generated using varying concentrations of pure sodium nitrite solutions (0.039 to 5 μmol/L) to calculate the absolute nitrite concentration within the biological samples.

2.9. Statistical analysis

All data are presented as the mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism version 10 (San Diego, CA, USA). Differences between two independent groups were evaluated using Welch's test. Longitudinal data involving multiple groups and time points were analyzed using a two-way repeated measures analysis of variance (ANOVA), followed by Šídák's multiple comparisons post hoc test. A P-value of <0.05 was considered statistically significant.

3. Results

3.1. Validation of the genetic model and development of cardiometabolic HFpEF in male cohorts

We first validated the genetic model by assessing myocardial levels of nitrite (Fig. 1B), a well-established biomarker for NO bioavailability. Consistent with constitutive eNOS ablation, cardiac nitrite levels were negligible when compared to wild-type controls (0.02 ± 0.01 vs. 2.2 ± 0.59 nmol/g). We hypothesized that subjecting both male and female eNOS KO mice to metabolic stress via a HFD would drive the development of a stable, highly reproducible cardiometabolic HFpEF phenotype in a relatively short period of time. Indeed, we report that this genetic-metabolic paradigm successfully recapitulates key clinical criteria required for a true HFpEF diagnosis—specifically, a preserved ejection fraction, left ventricular (LV) diastolic dysfunction, severe exercise intolerance, multi-organ systemic fibrosis, and metabolic associated hepatic steatosis in both sexes. This comprehensive phenotype confirms that this genetic murine model of cardiometabolic HFpEF mimics in large part the key aspects of HFpEF observed in the clinic.

Fig. 1.

Cardiometabolic, functional, and structural characterization of the male eNOS KO + HFD model. (A) Schematic of the experimental timeline for male eNOS KO mice fed a high-fat diet (HFD) for 10 weeks compared to standard chow (normal diet, ND)-fed Wild-Type mice. (B) Cardiac nitrite levels (n = 5). (C) Body weight trajectory over the 10-week study period. (D) Invasive hemodynamic assessment of systolic and diastolic blood pressures. (E) Left ventricular (LV) ejection fraction (LVEF) derived from M-mode echocardiography, demonstrating preserved systolic function. (F) Calculated LV mass. (G) Interventricular septal thickness at end-diastole (IVSd). (H) LV posterior wall thickness at end-diastole (LVPWd). (I) LV internal diameter at end-diastole (LVIDd). (J) Echocardiographic tissue Doppler-derived ratio of early mitral inflow velocity to early diastolic tissue velocity (E/e’). (K) Isovolumetric relaxation time (IVRT). (L) Invasive LV end-diastolic pressure (LVEDP) assessment. (M) Assessment of exercise intolerance via graded treadmill testing, displaying total running distance and total work performed. Data are presented as mean ± SEM (n = 10 per group). Statistical significance was determined by Welch's test or two-way repeated measures analysis of variance (ANOVA) where appropriate.

Following the experimental timeline depicted in Fig. 1A, eNOS KO male mice fed an HFD for 10 weeks developed a severe and progressive cardiometabolic phenotype that closely mimics the key human comorbidities. When compared to standard chow-fed wild-type mice, the eNOS KO + HFD mice exhibited a significant increase in body weight (Fig. 1C) alongside profound elevations in both systolic (153 ± 2.2 vs. 104 ± 3.6 mmHg) and diastolic (109 ± 2.5 vs. 69 ± 2.9 mmHg) blood pressures (Fig. 1D). Crucially, despite the presence of chronic, sustained hypertension, left ventricular ejection fraction (LVEF) remained entirely preserved (>60%) (Fig. 1E). However, detailed echocardiographic analysis revealed underlying concentric hypertrophy. This structural remodeling was characterized by significantly increased left ventricular mass and wall thickness, coupled with reduced internal ventricular diameters (Fig. 1F–I). Thus, the model faithfully recapitulates the pathological myocardial structural changes frequently observed in HFpEF patients.

Functionally, these structural alterations translated into pronounced diastolic impairment. Echocardiographic assessment revealed a significantly increased E/e’ ratio (Fig. 1J) and isovolumetric relaxation time (IVRT, Fig. 1K), indicative of impaired LV diastolic relaxation and increased resting filling pressures. This non-invasive echocardiographic finding was directly corroborated by invasive hemodynamic assessments, which demonstrated a marked elevation in left ventricular end-diastolic pressure (LVEDP) in the eNOS KO + HFD mice (Fig. 1L and 33.8 ± 1.1 vs. 12.7 ± 2.3 mmHg). Clinically, these compounded cardiac perturbations manifest as severe exercise intolerance. Consistent with the hallmark symptoms of human HFpEF, graded treadmill testing demonstrated a highly significant reduction in both running distance and total work performed by eNOS KO + HFD mice compared to their control counterparts (Fig. 1M). The combination of elevated cardiac filling pressures and severe physical performance limitations unequivocally confirms that this model accurately replicates the core physiological deficits required for a valid HFpEF diagnosis.

3.2. Myocardial remodeling and mechanisms of oxidative and inflammatory stress

The profound functional impairments observed in the eNOS KO + HFD cohort were accompanied by pathological remodeling of the myocardium. Histological evaluation of cardiac tissue sections demonstrated a marked accumulation of extracellular matrix proteins. Specifically, quantitative image analysis revealed highly significant increases in both cardiac interstitial and perivascular fibrosis in the eNOS KO + HFD mice compared to wild-type mice fed a normal diet (Fig. 2A). This extensive histopathological remodeling was directly corroborated at the transcriptional level. Quantitative PCR analysis of myocardial tissue revealed a robust upregulation in the expression of key profibrotic regulatory genes, including collagen type I alpha 1 (Col1), collagen type III alpha 1 (Col3), and transforming growth factor-beta (Tgf-β) (Fig. 2B).

Fig. 2.

Cardiac fibrosis, oxidative stress, and inflammation. (A) Representative histological images and quantitative analysis of cardiac interstitial and perivascular fibrosis utilizing Picrosirius Red staining (scale bar = 20 μm). (B) Relative cardiac mRNA expression of key profibrotic genes, including collagen type I alpha 1 (Col1), collagen type III alpha 1 (Col3), and transforming growth factor-beta (Tgf-β), measured by RT-qPCR. (C-D) Markers of pro-inflammatory and oxidative stress, displaying relative cardiac mRNA expression of interleukin-6 (IL-6) and circulating plasma levels of 8-isoprostane quantified via ELISA. (E) Relative cardiac mRNA expression of inducible nitric oxide synthase (iNOS). Data are presented as mean ± SEM (n = 7-10 per group). Statistical significance was determined by Welch's test. HFD, high-fat diet; ND, normal diet.

Mechanistically, this fibrotic deposition and subsequent functional deterioration were driven by a persistent state of pro-inflammatory and oxidative stress within the cardiac microenvironment. Analysis of the cardiac tissue showed a significant upregulation in the expression of the pro-inflammatory cytokine interleukin-6 (IL-6, Fig. 2C). This local myocardial inflammation coincided systemically with profoundly elevated plasma levels of 8-isoprostane, an established biomarker of lipid peroxidation and oxidative stress (Fig. 2D).

Crucially, to evaluate the extent of nitrosative imbalance in this model, further molecular profiling was performed. Quantitative PCR confirmed significant upregulation of oxidative stress and inflammation in the absence of any significant increase in cardiac inducible NOS (iNOS) expression (Fig. 2E). Together, these data demonstrate that the left ventricular remodeling and resulting HFpEF phenotype are directly driven by the targeted endothelial NOS deficiency combined with metabolic overload, operating independently of inflammatory iNOS upregulation.

3.3. Systemic involvement and multi-organ pathology

Clinical HFpEF is increasingly recognized not merely as an isolated cardiac condition, but as a complex, systemic syndrome driven by severe metabolic dysfunction, fibrosis, lipid accumulation, and microvascular inflammation across multiple organs including but not limited to the liver, kidney, and skeletal muscle [[12], [13], [14]]. Importantly, mirroring this multi-organ pathology observed in the clinic, the pathological consequences of the eNOS KO mouse + HFD paradigm extended beyond the myocardium as multiple organs were impacted significantly.

Histological examination of extra-cardiac tissues revealed severe systemic end-organ damage. Specifically, the experimental cohort exhibited severe hepatic steatosis, characterized by widespread pathological lipid accumulation within the liver parenchyma (Fig. 3). Concurrently, analysis of renal sections demonstrated significant tubulointerstitial fibrosis, suggestive of progressive kidney injury resulting from chronic cardiometabolic stress (Fig. 3).

Fig. 3.

Hepatic and renal pathology in the eNOS KO + HFD mouse model.(A) Representative histological images (H&E stain) and quantitative image analysis of hepatic steatosis (% area), characterized by widespread lipid droplet accumulation within the liver parenchyma (scale bar = 20 μm). (B) Representative histological images and quantitative assessment of renal tubulointerstitial fibrosis (% collagen fraction) resulting from chronic cardiometabolic stress (scale bar = 20 μm). Data are presented as mean ± SEM (n = 10 per group). Statistical significance was determined by Welch's test. HFD, high-fat diet; ND, normal diet.

Together, these structural and histological data provide evidence that the eNOS KO + HFD mouse model successfully recapitulates the systemic inflammatory, fibrotic, and metabolic derangements that define the complex pathophysiology of human cardiometabolic HFpEF. Future studies are aimed at further characterization of hepatic, renal, vascular, and skeletal muscle pathology in the eNOS KO mouse + HFD in both sexes.

3.4. Overcoming female resistance: robust HFpEF induction in both sexes

Following the 15-week experimental timeline outlined in Fig. 4A, female eNOS KO + HFD mice developed the core multi-organ disease derangements required for a clinical HFpEF diagnosis. Closely mirroring the extent of pathology observed in the male cohort, female mice subjected to the genetic-metabolic intervention exhibited significantly increased body weight (Fig. 4B) alongside elevated systolic and diastolic blood pressures (Fig. 4C). Despite the induction of this robust hypertensive state, left ventricular ejection fraction (LVEF) remained entirely preserved (Fig. 4D). Furthermore, structural assessments confirmed the presence of cardiac hypertrophy (Supplemental Fig. S1A–C).

Fig. 4.

Cardiometabolic HFpEF phenotype in female eNOS KO mice. (A) Schematic of the experimental timeline for the female cohort eNOS KO mice fed an high-fat diet (HFD) for 15 weeks compared to Wild-Type controls fed a normal diet (ND). (B) Body weight during the 15-week experimental protocol. (C) Systolic and diastolic blood pressures (mmHg). (D) Preserved left ventricular ejection fraction (LVEF). (E) Elevated E/e’ ratio via tissue Doppler imaging. (F) Invasive LV pressures (mmHg) demonstrating elevated LVEDP in female eNOS KO mice fed a high fat diet. (G) Treadmill exercise capacity, highlighting significant reductions in both running distance and total work. (H) Plasma levels of 8-isoprostane indicating systemic oxidative stress. (I) Quantitative histological assessment confirming marked increases in both cardiac fibrosis (interstitial and perivascular) and metabolic hepatic steatosis (scale bar = 20 μm). Data are presented as mean ± SEM (n = 8 per group). Statistical significance was determined by Welch's test or two-way repeated measures analysis of variance (ANOVA) where appropriate.

Crucially, these morphological changes translated into profound functional and hemodynamic deficits. The female cohort displayed pathological cardiac filling pressures, characterized non-invasively by a high E/e’ ratio and increased IVRT (Fig. 4E, Supplemental Figure S1D) and invasively by markedly elevated LVEDP (Fig. 4F). Clinically, these severe hemodynamic perturbations resulted in a dramatic deterioration of exercise capacity. Both total running distance and physical work performed on the treadmill were severely reduced in the female eNOS KO + HFD mice compared to their control counterparts, with this physical limitation becoming especially prominent after 15 weeks of HFD feeding (Fig. 4G). Despite the delayed onset of HFpEF in female eNOS KO mice, the decline in exercise capacity (exercise distance and work) was very similar to that observed in the male eNOS KO mice.

Mechanistically, this functional decline was accompanied by systemic oxidative stress, as evidenced by significantly elevated plasma 8-isoprostane levels in the female eNOS KO cohort (Fig. 4H). This increases oxidative stress ultimately drove adverse multi-organ tissue remodeling. Histological evaluation confirmed that the female eNOS KO + HFD mice developed marked increases in both cardiac interstitial and perivascular fibrosis, coupled with extensive hepatic steatosis (Fig. 4I).

While the development of a stable, clinically relevant HFpEF phenotype in female eNOS KO mice evolved over at a slower rate (i.e., 15 weeks vs. 10 weeks to exhibit HFpEF), the ultimate degree of HFpEF severity was very similar between male and female mice.

Thus, this genetic model combined with a high-fat diet ensures robust, reproducible HFpEF induction in both sexes, providing an indispensable tool for the rigorous evaluation sex-specific differences in the mechanisms of HFpEF progression and therapeutic efficacy.

3.5. Pharmacological validation: tirzepatide reverses the cardiometabolic HFpEF phenotype

To determine the translational utility of this genetic-metabolic model, we evaluated its responsiveness to a clinically relevant pharmacological intervention that has previously been shown to be effective in cardiometabolic HFpEF in the clinic [[15], [16], [17], [18], [19]]. We selected tirzepatide, a dual GIP/GLP-1 receptor agonist that has demonstrated highly promising results in human HFpEF trials by broadly targeting underlying metabolic comorbidities such as obesity and diabetes [15]. Following the experimental timeline outlined in Fig. 5A, male eNOS KO mice were fed an HFD for 5 weeks to establish the initial cardiometabolic disease phenotype, followed by 10 weeks of systemic therapeutic intervention with tirzepatide.

Fig. 5.

Pharmacological Therapy: Tirzepatide reverses the core cardiometabolic HFpEF phenotype. (A) Experimental timeline for male eNOS KO mice fed an high-fat diet (HFD) for 5 weeks prior to randomization to either vehicle (0.05% Tween-80 in PBS) or the dual GIP/GLP-1 receptor agonist tirzepatide (30 nmol/kg twice weekly) for 10 subsequent weeks. (B) Significant reduction in body weight following tirzepatide treatment. (C) Attenuation of systolic and diastolic blood pressures. (D) Left ventricular ejection fraction (LVEF) remaining preserved throughout treatment. (E) Longitudinal reduction in the E/e’ ratio at 10 and 15 weeks following tirzepatide. (F) LVEDP (mmHg) measured with pressure catheter demonstrating marked reductions following tirzepatide treatment. (G) Restoration of exercise capacity in mice receiving tirzepatide as demonstrated by improved running distance (m) and total work (Kg x m) at 10 and 15 weeks. (H) Reduction of systemic oxidative stress (plasma 8-isoprostane). (I) Histological reversal of maladaptive tissue remodeling, showing significant decreases in cardiac interstitial/perivascular fibrosis and liver fat content (scale bar = 20 μm). Data are presented as mean ± SEM (n = 6 per group). Statistical significance was determined by Welch's test or two-way repeated measures analysis of variance (ANOVA) where appropriate.

Treatment with tirzepatide effectively reversed the core morphometric and hemodynamic derangements of the disease as early as 5 weeks following treatment and these marked benefits persisted throughout the 10 week follow up period (Fig. 5). Compared to the vehicle-treated control eNOS KO group, mice receiving tirzepatide exhibited a robust and significant reduction in total body weight (Fig. 5B). Concurrently, the pharmacological intervention profoundly ameliorated the hypertensive phenotype, significantly lowering both systolic and diastolic blood pressures (Fig. 5C). Throughout the treatment period, left ventricular ejection fraction (LVEF) remained completely preserved and unchanged (Fig. 5D).

Crucially, tirzepatide administration rescued LV diastolic function and exercise capacity. Echocardiographic assessment revealed that the E/e’ ratio, an index of left ventricular filling pressure, was significantly reduced at the 10-week study timepoint (i.e. 5 weeks following the initiation of treatment), and this improvement was sustained through the 15-week study endpoint (Fig. 5E) together with improved IVRT (Supplemental Fig. S2A). This non-invasive finding was directly confirmed by terminal invasive hemodynamics, which demonstrated a marked reduction in LVEDP following tirzepatide treatment (Fig. 5F). These robust hemodynamic improvements translated directly into the restoration of physical function; tirzepatide-treated mice demonstrated profound, significant improvements in both treadmill running distance and total work capacity at the 10- and 15-week timepoints compared to vehicle controls (Fig. 5G).

Finally, we assessed the impact of the drug on systemic oxidative stress and adverse multi-organ structural remodeling. Tirzepatide therapy effectively suppressed circulating plasma 8-isoprostane levels (Fig. 5H). At the tissue level, this systemic metabolic rescue was associated with a striking reversal of maladaptive pathology. Importantly, tirzepatide treatment significantly attenuated cardiac hypertrophy, as evidenced by a reduced heart weight-to-tibia length ratio and LV wall thickness (Supplemental Fig. S2B–D). Histological analysis revealed substantial decreases in both cardiac interstitial and perivascular fibrosis, alongside a marked reduction in hepatic lipid content (Fig. 5I). Together, these data demonstrate that the eNOS KO + HFD model is highly responsive to standard-of-care HFpEF pharmacotherapy, establishing it as a viable model system for future drug discovery and validation.

Although tirzepatide profoundly reversed the cardiometabolic HFpEF phenotype, the present study design does not distinguish weight-dependent from weight-independent mechanisms of cardiac benefit. Delineating the direct cardiovascular effects of dual GIP/GLP-1 agonism from those driven primarily by improved systemic metabolic status and weight loss remains an important area for future investigation.

4. Discussion

In the present study, we characterized and validated a robust, reproducible murine model of cardiometabolic HFpEF by leveraging complete genetic eNOS deficiency (eNOS KO) in combination with HFD. We demonstrated that this genetic-metabolic paradigm faithfully recapitulates the core components of human HFpEF, including preserved LV ejection fraction, severe LV diastolic dysfunction, profound exercise intolerance, and cardiac hypertrophy. Furthermore, our findings establish that this model accurately reflects the systemic, multi-organ complexity of the clinical syndrome, driving not only cardiac interstitial and perivascular fibrosis, but also severe hepatic steatosis and renal tubulointerstitial fibrosis. The use of gene-targeted mice that completely lack the eNOS enzyme and physiological levels of nitric oxide accelerates disease onset, drives a severe cardiometabolic phenotype, and effectively overcomes any barriers of female resistance, ensuring robust disease penetrance across both sexes. Moreover, the successful reversal of these systemic and functional derangements using the clinically relevant dual GIP/GLP-1 receptor agonist tirzepatide confirms the eNOS KO + HFD mouse model as a highly translatable preclinical model for therapeutic discovery in cardiometabolic HFpEF.

The strength of our eNOS KO + HFD mouse model lies in its ability to replicate the complex, multi-organ reality of clinical cardiometabolic HFpEF, where hypertension, obesity and systemic metabolic derangement are indispensable disease drivers. While eNOS deficiency induces hypertension, previous studies examining cardiac diastolic function in isolated eNOS-deficient mice have produced highly heterogeneous findings, indicating that the consequences of eNOS loss depend strongly on age, experimental conditions, and compensatory pathways. For instance, one study found that despite hypertension, eNOS KO mice exhibited normal baseline ventricular relaxation (τ)—likely due to compensatory mechanisms like increased atrial natriuretic peptide (ANP) expression [20]. Conversely, another study demonstrated clear baseline diastolic dysfunction (elevated LVEDP and prolonged τ) that was unexpectedly rescued by endothelin-1 overexpression [21]. The addition of cardiovascular stress further highlights this context dependence; indeed, eNOS KO mice were protected from chronic pressure overload (transverse aortic constriction), as the stressor induces pathological eNOS uncoupling and ROS generation in wild-type hearts [22]. Conversely, during acute stress like ischemia/reperfusion or cardiac arrest, eNOS appears protective; one study reported comparable baseline τ but greater functional impairment in eNOS KO mice post-resuscitation [23], while a more recent ischemia/reperfusion study noted pre-existing echocardiographic diastolic abnormalities that were not further exacerbated by the ischemic event [24]. This baseline variability is heavily influenced by aging, which progressively drives hypertrophy and fibrosis [25], differences between global versus endothelial- or red-blood-cell-specific eNOS deletion, and profound compensatory mechanisms, such as nNOS-derived H₂O₂/endothelium-dependent hyperpolarization that can preserve function when eNOS alone is deleted [26]. Collectively, these studies prove that genetic eNOS deficiency alone does not produce a uniform or comprehensive HFpEF phenotype. By superimposing a high-fat diet onto this genetic background, our model overwhelms these innate compensatory pathways. This addition incorporates the critical obesity-driven metabolic stress seen in human patients, shifting the paradigm from heterogeneous, isolated diastolic abnormalities to a highly reproducible, multi-organ cardiometabolic HFpEF syndrome.

5. Limitations and future directions

While this study establishes the eNOS KO + HFD mouse model as a highly robust and reproducible translational platform, several limitations and future directions warrant discussion and consideration. First, the use of a constitutive, lifelong genetic eNOS knockout may not perfectly mirror the progressive, acquired endothelial dysfunction typical of the human clinical scenario. However, this study serves as a critical proof-of-concept; future investigations will utilize adult-onset, inducible eNOS knockout models to perhaps more faithfully replicate the clinical disease scenario. Second, because this initial validation focused primarily on translational relevance, we did not include single-hit cohorts (i.e., wild-type mice on HFD alone or eNOS KO mice on standard chow). Future studies aiming to dissect the specific, isolated contributions of metabolic stress versus nitric oxide depletion (i.e. hypertension) will benefit from including these study groups. Third, although our model induces HFpEF in both sexes, female mice did require an additional 5 weeks of HFD exposure (15 weeks total) compared to males (10 weeks) to develop a comparably severe HFpEF phenotype. Nevertheless, a 15-week induction period remains highly efficient and is significantly shorter than the prolonged timelines required for traditional chronological aging models. The stable and highly robust eNOS KO + HFD mouse model can be readily crossed with other genetically modified lines to generate double knockouts, providing an invaluable tool for deeply interrogating specific molecular pathways and novel therapeutic targets in HFpEF.

CRediT authorship contribution statement

Mahmoud H. Elbatreek: Formal analysis, Methodology, Visualization, Writing – original draft. Vinayak Khodade: Writing – review & editing, Formal analysis. Darshan Amin: Methodology. Miguel Huerta: Methodology. Timothy D. Allerton: Validation, Writing – review & editing. David J. Polhemus: Investigation, Validation, Methodology, Writing – review & editing. Sanjiv J. Shah: Investigation, Writing – review & editing. Thomas M. Vondriska: Funding acquisition. Traci T. Goodchild: Project administration, Investigation, Supervision. David J. Lefer: Conceptualization, Visualization, Methodology, Funding acquisition, Investigation, Project administration, Supervision, Writing – review & editing, Writing – original draft.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Sources of funding

This work was supported by the National Institutes of Health Grants R01 HL181016 and R01 HL174996 to D.J.L., California Institute for Regenerative Medicine (CIRM) PDEV-19141 to D.J.L, P20GM135002 to T.D.A., and 5T32HL007227 to D.J.P.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: