New Model of Experimental Diabetic Cardiomyopathy Using Combination of Multiple Doses of Anomer-Equilibrated Streptozotocin and High-Fat Diet: Sex Matters Free

More than half a billion individuals worldwide are affected by diabetes, with an expected increase of 46% to 783 million by 2045 (1). Cardiovascular complications are the leading causes of morbidity and mortality in diabetes (2). Diabetic cardiomyopathy (dbCM) is a distinct disease that leads to abnormal myocardial structure and function, in the absence of other cardiac risk factors (3). The development of dbCM consists of three stages. Initially, patients are asymptomatic but have increased cardiac fibrosis and stiffness. The second stage is defined by left ventricular (LV) hypertrophy (LVH), cardiac remodeling, and diastolic dysfunction, leading clinically to heart failure (HF) with preserved ejection fraction (EF). In the advanced stage, systolic dysfunction accompanies diastolic dysfunction, and patients develop HF with reduced EF (4). In clinical application, LV diastolic dysfunction is the most frequent echocardiographic finding in asymptomatic patients with diabetes with normal LVEF (5). A new definition of dbCM has been proposed as diastolic dysfunction in the presence of altered myocardial metabolism, but in the absence of other known causes of cardiomyopathy and/or hypertension (6).

Although diabetes is more prevalent in men, cardiovascular risk is higher in women with diabetes, leading to a more rapid development of dbCM and progression to HF. Premenopausal women have a lower risk of cardiovascular disease compared with men of the same age, but this advantage is diminished in postmenopausal women with diabetes because of the lost estrogen availability and protection from cardiovascular risk (7). Few studies in humans have reviewed the mechanisms behind the sex differences in the progression of dbCM. Furthermore, murine studies have seldom included female animals, creating preclinical bias, leading to a gap in available robust data for the better understanding of the effect of biological sex differences in dbCM (8), thus impeding translation to a heterogeneous clinical population (9).

Multiple experimental murine models have been used to investigate the pathogenesis and treatment of diabetes because of their resemblance to human disease (6). The most established methodology for induction of diabetes is the combination of high-fat diet (HFD) and streptozotocin (STZ) injections. Despite the demographics of patients with diabetes, the use of STZ combined with HFD (STZ + HFD) occurs predominantly in male preclinical studies; there is a dearth of published studies in female mice using this protocol (10).

STZ is an alkylating antineoplastic agent that is specifically toxic to insulin-secreting pancreatic β-cells. It enters the pancreatic β-cells through GLUT2 because of its similarity to glucose and leads to alkylation of DNA and consequently cell death (11). Multiple low doses of STZ cause the steady autoimmune destruction of β-cells, rather than the rapid destruction caused by a single high-dose STZ injection (12,13). Diabetic Complications Consortium guidelines suggest that STZ must be administrated within 15 min postdissolution in citrate buffer solution, because it degrades rapidly after this time point (14). However, it has been shown that dissolved STZ can remain stable for days at room temperature (15,16). STZ has two anomers, α-anomer and β-anomer, the first of which is more toxic. When STZ is dissolved in citrate buffer, the concentration of the α-anomer is up to 20-fold greater than that of the β-anomer, and it takes 60–90 min to reach anomer equilibrium (14,15). Therefore, STZ injections can be administered within 15 min of dissolution (fresh STZ) or 90 min after dissolution (anomer-equilibrated [EQ] STZ). However, with the use of freshly injected STZ, a diabetes mouse model has failed to achieve stable and consistent hyperglycemia or a robust diabetic pathophysiology (14). In contrast, mice injected with EQ STZ had a higher level of survival and persistent hyperglycemia with minimal weight loss compared with mice receiving freshly injected STZ (17).

Furthermore, EQ STZ has been shown to lead to development of severe fasted hyperglycemia, significant reduction in fasting insulin levels, and early onset of diabetic pathophysiology (14). However, whether EQ STZ–based diabetes induction protocols affect cardiac function and lead to development of dbCM remains unknown.

The primary aim of our study was to establish a novel robust model of inducible murine diabetes using the combination of EQ STZ injections with HFD feeding protocol. Additionally, we aimed to investigate the effect of biological sex on the course of development of dbCM. Here, we show that an 8-week protocol consisting of five doses of EQ STZ (40 mg/kg) combined with transient HFD feeding (36 days) leads to a stable early-onset preclinical model of diabetes that resembles human type 2 diabetes (T2D) pathology, including fed and fasting hyperglycemia, impaired glucose tolerance, development of dbCM, and severely perturbed systemic metabolome. Moreover, we show that biological sex significantly affects the severity of the dbCM cardiac phenotype. Unlike male mice, which develop diastolic dysfunction, female mice develop diastolic and systolic dysfunction as well as extensive cardiac metabolic perturbation.

This investigation conformed to the U.K. Home Office Guidance on the Operation of the Animals (Scientific Procedures) Act 1986. C57BL/6J mice (N = 109; aged 7 weeks) were purchased from Charles River Laboratories (London, U.K.). Animals were housed in individually ventilated cages and maintained under controlled temperature (21 ± 2°C) on a 12:12-h light-to-dark cycle with access to chow diet (PicoLab 5058; cat. no. I-DIET-5R58-9KG-BG; IPS Product Supplies, London, U.K.) and water ad libitum. Enrichment was provided in the form of tunnels and chew sticks. Body weight and nonfasting blood glucose were recorded weekly.

STZ (cat. no. 18883-66-4; Sigma-Aldrich, Gillingham, U.K.) contained 75% of the α-anomer as specified by the manufacturer. STZ (40 mg/kg) was dissolved in sterile 0.1 mol/L sodium citrate buffer (pH 4.5). STZ–sodium citrate buffer solutions were vortexed for 30 s and i.p. injected within 15 min (fresh STZ) or vortexed for 30 s every 10 min for 90 min before i.p. injection (90-min EQ STZ). Control diet used in the study (cat. no. TD.210879; Envigo, Indianapolis, IN) contained 10.5% fat, 69.1% carbohydrate, and 20.5% protein (percentage of energy in kcal). HFD (cat. no. TD.200185; Envigo) consisted of 60.3% fat, 21.4% carbohydrate, and 18.3% protein (percentage of energy in kcal). The dietary switch between HFD and standard chow was based on the previously described protocol for T2D induction in mice (18). HFD administration before STZ injections causes pancreatic inflammation, making mice more susceptible to STZ toxicity at low doses (18). Dietary switch from HFD to control chow also allows for induction of nonobesogenic diabetes and enables easier dietary incorporation of pharmacological agents for drug intervention studies (19). To help delineate the roles of dietary composition and STZ toxicity in the generation of our newly proposed model, we also examined the effect of HFD feeding alone as well as that of STZ injections (both fresh and equilibrated preparations) in the absence of dietary modification.

Eight-week-old C57BL/6J mice (N = 109) were separated into six experimental groups (summarized in Fig. 1):

1. EQ STZ and HFD (EQ + HFD): HFD days 1–36; control diet days 37–60 (n = 10 male; n = 9 female).

2. Freshly injected STZ and HFD (fresh + HFD): HFD days 1–36; control diet days 37–60 (n = 9 male; n = 8 female).

3. HFD feeding group (HFD): HFD days 1–60 (n = 8 male; n = 8 female).

4. EQ STZ injections (EQ STZ): control diet days 1–60 (n = 8 male; n = 8 female).

5. Freshly prepared STZ injections (fresh STZ): control diet days 1–60 (n = 8 male; n = 8 female).

6. Control group: control diet days 1–60 (n = 12 male; n = 13 female).

All STZ groups received i.p. injections on 5 consecutive days (days 22–26), and the control group received vehicle citrate buffer i.p. injections (Fig. 1). Nonfasting glucose was recorded weekly (weeks 0–7); i.p. glucose tolerance testing (IPGTT) was conducted on day 56 (week 8) and echocardiography on day 57 (week 8). IPGTT was performed after 12-h fast (8:00 p.m. to 8:00 a.m.). Fasting glucose measurements (0 min) were recorded before i.p. administration of glucose load (2 g/kg in 0.1 mL distilled water). Tail vein blood samples were used for glucose measurement (VivaChek glucometer; St Asaph, U.K.) at 15, 30, 60, 90, and 120 min postinjection. On day 60, all animals were humanely sacrificed by pentobarbital i.p. injection (140 mg/kg).

Cardiac function was assessed at the end of the protocols (week 8) by echocardiography (M-mode and Doppler) (20). Analysis was completed across three heartbeats, in three cineloops per animal.

Heart, liver, skeletal muscle and kidneys were rapidly collected and snap frozen in liquid N₂ using Wollenberger tongs (21). Heart and liver weights were recorded. Blood samples were collected from the thoracic cavity, centrifuged (3,000 rpm, 4°C), and analyzed by the Medical Research Council Mouse Biochemistry Laboratory (Addenbrookes Hospital, Cambridge, U.K.).

Nontargeted metabolomic profiling of the cardiac tissue was performed by liquid chromatography–mass spectroscopy (22). Snap-frozen and pulverized tissue (liver, skeletal muscle, and kidney) was analyzed using high-resolution ¹H nuclear magnetic resonance (NMR) spectroscopy (21).

Pancreata were fixed in 10% neutral formalin. Paraffin-embedded sections were hematoxylin-eosin stained by the Barts Cancer Institute histology facility. Tissue section images were acquired using a Nanozoomer S210 slide scanner, and the mean pancreatic islet area was measured using Image J software (version 1.54) (23).

Data are presented as mean ± SEM. Normality of data distribution was examined using the Shapiro-Wilk normality test. Comparison between two groups was performed using the Student t test (Gaussian data distribution); two- and one-way ANOVAs with Tukey correction were used for multiple comparisons where applicable. Power analysis was used to determine sample sizes. Based on a comparison between groups by ANOVA, a sample size of eight mice per group was shown to have 90% power to detect differences between fasting glucose means with a significance level of 0.05 in both sexes. Statistical analysis was performed using GraphPad Prism (version 10.2.2) software. Differences were considered significant when P < 0.05.

All data generated in this study are provided in the Supplementary Material.

Male mice on both STZ + HFD induction protocols experienced transient weight gain during 4 weeks of HFD. At the study end point (week 8), both groups had gained 8% in body weight versus week 0 (EQ + HFD P < 0.01; fresh + HFD P < 0.01) (Fig. 2A). Female mice on both induction protocols gained weight steadily, with 14% (EQ + HFD P < 0.0001) and 8% (fresh + HFD P < 0.001) weight increases at week 8 versus week 0 (Fig. 2B). In both sexes, weight gain was comparable to that in the age- and sex-matched control group (Fig. 2A and B). In both sexes, STZ + HFD experimental protocols successfully induced both nonfasting (Fig. 2C and E) and fasting hyperglycemia (Fig. 2G and I). However, male mice had an earlier onset of impaired glucose homeostasis (1 week post-STZ injections) (Fig. 2C), with more severe end point hyperglycemia (twofold vs. control; P < 0.01) (Fig. 2C). Female mice on the fresh + HFD protocol developed hyperglycemia 2 weeks post-STZ injections (40% increase vs. control; P < 0.0001), and those in the EQ + HFD group did so at the end of the protocol (40% increase in glucose vs. control; P < 0.001) (Fig. 2E). IPGTT response was impaired in both sexes in both diabetes induction protocols, with glucose concentration reaching maximum at 30 min and failing to return to baseline after 120 min (Fig. 2G–J). However, at the IPGTT end point, female groups had less severe hyperglycemia (EQ + HFD 1.5-fold and fresh + HFD twofold increase in glucose vs. control; P < 0.05) (Fig. 2I) than males (EQ + HFD 2.4-fold and fresh + HFD threefold increase in glucose vs. control; P < 0.001) (Fig. 2G).

The 8-week HFD feeding protocol led to an increase in body weight (30% male; 10% female) (Supplementary Fig. 1A and B) but did not cause hyperglycemia (neither fasting nor fed) (Supplementary Fig. 1C–G and I). At 120 min of IPGTT, HFD females returned to normoglycemia, with males showing 50% increase in glucose versus controls (Supplementary Fig. 1G–J).

Irrespective of sex or method of drug preparation, STZ injections in the absence of dietary modification did not negatively affect body weight by the end of the protocol (Supplementary Fig. 2A and B). Nonfasting plasma glucose was increased in males at week 8 (Supplementary Fig. 2C), whereas females remained normoglycemic (Supplementary Fig. 2E and F). IPGTT response was impaired in both sexes (Supplementary Fig. 2G–J).

The circulating metabolic profile showed that all experimental groups were normoinsulinemic (Supplementary Table 1). Male mice from all experimental protocols developed hypertriglyceridemia, and the male fresh + HFD group exhibited increased circulating creatine kinase activity, suggestive of muscle damage (Supplementary Table 1). Markers of liver dysfunction ALT and AST were elevated in male EQ + HFD, EQ STZ, and fresh + HFD groups (Supplementary Table 1), with EQ + HFD and fresh + HFD protocols also causing alterations in liver morphology, with significantly increased liver weight (Supplementary Table 2). No evidence of altered liver morphology or function was observed in female mice (Supplementary Tables 1 and 2). There was no evidence of anemia as the confounding pathologic factor in any of the groups examined (Supplementary Table 2).

In terms of pancreas morphology, in both sexes control and HFD groups were characterized by well-defined islets of Langerhans, surrounded by the exocrine portion of the pancreas, whereas STZ-treated groups were characterized by visually disrupted islet architecture (Fig. 3A and B). In all experimental protocols (HFD, STZ + HFD, and STZ injections), regardless of sex, the mean pancreatic islet area was increased compared with in controls (Fig. 3C and D).

Diabetes induction protocols led to the development of cardiac phenotypes. However, the extent of cardiac remodeling differed between experimental protocols and sexes (Supplementary Table 3). In male mice, the EQ + HFD protocol led to the development of a typical dbCM phenotype characterized by preserved EF and fractional shortening (FS) (Fig. 4A and B) accompanied by the development of diastolic dysfunction (increased mitral E-wave velocity [E]/peak early diastolic annular velocity [E′], reduced E′/peak late diastolic annular velocity [A′], and increased mitral valve [MV] deceleration time) (Fig. 4C–E). However, the fresh + HFD protocol did not have an effect on cardiac function (Fig. 4). LVH was not observed in the STZ + HFD groups (Supplementary Table 2).

In contrast, female mice on the EQ + HFD and fresh + HFD protocols developed distinct cardiac dysfunction. Both protocols led to systolic dysfunction, with reduction in EF and increased LV myocardial performance index in the EQ + HFD group (Fig. 5A–C). Furthermore, female mice developed diastolic dysfunction, as shown by the decrease in E′/A′ (Fig. 5D). However, the EQ + HFD induction protocol caused more extensive functional changes. Female mice on this protocol developed an increase in isovolumic relaxation time (Fig. 5E) as well as altered cardiac chamber geometry, with an increase in LV internal diameter in diastole (Fig. 5F).

The HFD feeding protocol led to the development of diastolic dysfunction in both sexes (Supplementary Fig. 1K and N), without an impact on systolic function (Supplementary Fig. 1L, M, O, and P). Increased heart weight was noted in the male HFD group (Supplementary Table 2). In both sexes, STZ administration alone, irrespective of the method of preparation, caused impairment of systolic (reduction of EF and FS) and diastolic function (MV deceleration and E/E′) (Supplementary Fig. 2K–P).

Diabetes induction led to cardiac metabolic perturbations. In the STZ + HFD groups, there were fewer cardiac metabolic changes in male mice (six metabolites changed vs. control) (Fig. 6) than in female mice (21 metabolites changed vs. control) (Fig. 7). In male hearts, both protocols led to a depletion of intermediates of the citric acid cycle (Fig. 6A) and remodeled amino acid metabolism (Fig. 6D). In female mice, there were alterations in citric acid cycle intermediates (Fig. 7A) and redox and energetic intermediates, including decreased total adenine nucleotide (TAN) pool (AMP + ADP + ATP) (Fig. 7B), decreased glycolytic intermediates (Fig. 7C), and altered amino acid levels (Fig. 7D).

The HFD feeding protocol affected concentrations of 19 metabolites in male mouse hearts (Supplementary Fig. 3) and 23 metabolites in the female group (Supplementary Fig. 4). HFD feeding in both males and females caused alterations in the concentrations of citric acid cycle, redox and energetic, glucose, and amino acid metabolism constituents (Supplementary Figs. 3 and 4).

Administration of the STZ drug alone, irrespective of the mode of drug preparation or sex, resulted in extensive cardiac metabolic remodeling, affecting citric acid cycle, redox and energetic, glucose, and amino acid metabolism constituents. Twenty-two metabolites were altered in males (Supplementary Fig. 5) and 24 metabolites in females (Supplementary Fig. 6).

¹H NMR spectroscopy metabolomic profiling of peripheral tissues revealed a systemic metabolic phenotype of the diabetes induction protocols in both sexes (Supplementary Figs. 7–12). The skeletal muscle metabolomic profile of male mice was not affected by STZ + HFD diabetes induction protocols (Supplementary Fig. 7A and B). In contrast, the female skeletal muscle metabolomic profile was altered on both diabetes protocols (Supplementary Fig. 8A and B). The kidney metabolomic profile of EQ + HFD male mice was characterized by increased concentrations of glucose, tyrosine, and AMP and reduced carnitine availability (Supplementary Fig. 7C). The fresh + HFD protocol caused fewer changes in the kidney metabolome (Supplementary Fig. 7D). In the kidneys of female mice, the fresh + HFD protocol caused extensive metabolic alterations (Supplementary Fig. 8D). In terms of liver metabolism, the EQ + HFD protocol changed the metabolome in males (Supplementary Fig. 7E). Female mice showed a similar degree of changes in liver metabolism on both diabetes induction protocols (Supplementary Fig. 8E and F).

Both male and female mice fed on HFD had alterations in the muscle metabolome to similar extents (Supplementary Figs. 9A and 10A). Kidney and liver metabolism were more extensively altered in female mice fed on HFD versus males, with mainly energetic constituents depleted (Supplementary Figs. 9B and C and 10B and C).

STZ injection protocols depleted skeletal muscle energy storage (glycogen and phosphocreatine) in both sexes (Supplementary Figs. 11A and B and 12A and B). In terms of kidney metabolism, both sexes had increased glycolytic intermediates and reduction in carnitine availability (Supplementary Figs. 11C and D and 12C and D). Liver metabolomic changes were more extensive in EQ STZ protocol males compared with fresh STZ groups regardless of sex (Supplementary Figs. 11E and F and 12E and F).

Our study shows that a protocol combining EQ STZ injections with HFD feeding produces a robust and reproducible murine model of diabetes characterized by hyperglycemia, impaired glucose tolerance, perturbed systemic metabolome, and development of dbCM with impaired cardiac function and metabolism (summarized in Fig. 8). Thus, our model mostly recapitulates features of T2D.

Crucially, our study highlights that biological sex is an important determinant of the severity of the diabetic phenotype, because we identified substantial cardiometabolic differences between male and female mice (summarized in Fig. 8). Female mice developed more severe cardiac remodeling, with impairment of both systolic and diastolic function accompanied by severely perturbed cardiac metabolome. Therefore, the EQ + HFD protocol creates a preclinical diabetes model consistent with human data showing that women with diabetes have an earlier onset of dbCM and deterioration to HF (7,8). This diabetes model also has the added advantages of being easy to induce, relatively short, and cost effective.

Although all STZ injection–containing protocols led to impaired glucose tolerance, the effect was less severe in female mice, indicating a different biological response to experimental diabetes induction. These differences could be attributed to sex-specific endocrine effects (24). Estrogen receptors ERα and ERβ are expressed in endothelial cells, vascular smooth muscle cells, and cardiomyocytes of both sexes, and they exert a vascular protective effect (25,26). In females, increased estrogen levels reduce cardiovascular risk factors and affect lipid metabolism as well as glucose and insulin handling (27). In our study, hyperlipidemia, alterations in liver morphology, and increase in circulating liver dysfunction markers were observed only in male mice; therefore, the protective effect of estrogen could account for their absence in females.

The EQ + HFD protocol in male mice led to diastolic dysfunction with preservation of EF, representative of human dbCM (28–30). In contrast, in female mice, both EQ + HFD and fresh + HFD protocols led to diastolic and systolic dysfunction. Premenopausal women tend to lose the hormone-related protection from cardiovascular disease in the presence of diabetes (7). Several clinical studies, including the Framingham Heart study, have shown that women with diabetes had a higher risk of HF than men (31). A recent meta-analysis also found that women with diabetes have a 40% greater risk of incident coronary artery disease compared with men with diabetes (32). Female db/db (type 2 diabetic) mice show an earlier onset of LV remodeling and cardiac fibrosis than age-matched males, despite comparable measurements in body weight, plasma triglycerides, and insulin levels (33). Therefore, although female mice have less severe glucose homeostatic impairment, they are more susceptible to changes in cardiac function than male mice, mirroring human data.

Despite its prevalence in human dbCM, LVH was not detected in any of our diabetes models. In general, murine models of dbCM do not always recapitulate all the aspects of human diabetes, including the development of cardiomyocyte hypertrophy (34). Obesity is thought to be one of the potential drivers of diabetes-induced LVH (35). Given that obesity was absent from our diabetes model, this could be the reason why LVH was not observed.

STZ administration alone without a dietary modification protocol led to less severe hyperglycemia; male mice did not exhibit fasting hyperglycemia, and female mice remained normolgycemic in the fed state. Consumption of HFD by mice on diabetes protocols was shown to cause pancreatic inflammation that enhances the sensitivity of pancreatic islets to STZ toxicity even at low doses, leading to faster development of hyperglycemia (36). However, the effect of STZ in diabetes models is not limited to pancreatic toxicity. STZ administration alone caused the development of both systolic and diastolic dysfunction as well as cardiac metabolic remodeling in both sexes, suggestive of a cardiotoxic effect of the drug. STZ was shown to exert a direct cardiomyocyte effect, causing atrophy and generating oxidative stress, thus impairing cardiomyocyte contractility (37,38). However, given the STZ injections caused comparable cardiac dysfunction in both sexes, the more severe cardiac phenotype observed with STZ + HFD protocols in females was not simply due to their enhanced susceptibility to the effects of the drug but to the enhanced susceptibility to diabetes.

Perturbations in cardiac substrate metabolism are at the center of dbCM pathology. This has led to a recent redefinition of dbCM: diastolic dysfunction in the presence of altered myocardial metabolism (6). Although the cardiac metabolomic analysis identified an effect of different diabetes induction protocols, it primarily accentuated the sex-specific differences. Female mice in our study showed extensive perturbation of cardiac metabolism. Despite the systemic milieu of hyperglycemia, reductions in glycolytic intermediates glucose-6-phosphate and phosphoenolpyruvate are indicative of impaired myocardial glucose use. Furthermore, there was depletion of citric acid cycle intermediates, reduced availability of CoA, perturbed amino acid metabolism, and alteration of NAD/NADH redox balance with excess NADH accumulation. Notably, the TAN pool in female hearts in all experimental models was significantly reduced, indicating a reduction in cardiac energy reserve and ATP turnover. Collectively, these changes are suggestive of energy imbalance, which is known to trigger mechano-energetic uncoupling, a central element in early contraction-relaxation disorders (34). Therefore, altered substrate metabolism in female diabetic hearts could have been the driver of both the systolic and diastolic dysfunction observed. Although we identified sex-specific and induction model–driven changes in the cardiac metabolomic profile, we did not examine the degree of cardiac inflammation, oxidative stress, or gene/protein expression of diabetic cardiomyopathy mediators.

Systemic metabolism was affected in experimental models, with different patterns between protocols and biological sexes. Elevation of glucose in the kidneys of male and female diabetes models confirms the presence of glucose excess in the kidney, subsequent augmented tubular reabsorption, and increased flux of glucose into the blood, resulting in exacerbation of hyperglycemia (39). Alterations of circulating carnitine in liver, kidney, and muscle tissue of male and female diabetic animals could suggest an increased use of carnitine for β-oxidation of free fatty acids (FFAs) and augment the shift of metabolism toward FFAs in the diabetic state (40). Alterations in energetics in the peripheral tissues are consistent with mitochondrial dysfunction and impaired energy homeostasis in diabetes (41). Overall, female diabetic mice had changes suggestive of diabetic nephropathy and myopathy, whereas males had signs of liver dysfunction, in line with previous findings (42).

Eight weeks of HFD feeding led to an obesogenic insulin-resistant phenotype in both sexes, although without evidence of overt diabetes. The use of STZ in combination with HFD led to a gradually developed stable diabetes model characterized by glucose intolerance, fed and fasting hyperglycemia, cardiac and liver dysfunction, and extensive metabolic remodeling across both sexes (Fig. 8). Therefore, it has characteristics of T2D without obesity as a comorbidity. EQ STZ in conjunction with HFD led to more evident signs of cardiac dysfunction (diastolic dysfunction in males and accompanied systolic dysfunction in females) compared with freshly injected STZ, consequently representing a more effective model of experimental dbCM. Our study highlights that biological sex is an important determinant of the severity of the diabetic phenotype, because we identified substantial cardiometabolic differences between male and female mice. This accentuates the importance of inclusion of both biological sexes in preclinical studies of metabolic disorders such as diabetes.

Acknowledgments. The authors thank Dr. Nasima Kanwal and the School of Physical and Chemical Sciences, Queen Mary University of London, for access to the NMR facility and technical assistance. This work forms part of the research themes contributing to the translational research portfolio of the Precision Cardiovascular Genomic Medicine Theme of the Barts Cardiovascular Biomedical Research Centre which is funded by the National Institute for Health Research (NIHR).

Funding. L.K. has received the Health Care Aid Fellowship/William Harvey Research Limited Clinical Fellowship and Leventis Foundation Scholarship Cyprus. K.B. has received Barts and the London Charity grant support (MGU0401) from the Metabolic Flux Analysis Facility of the Barts School of Medicine and Dentistry. S.M.H. was funded by Diabetes UK grant 19/0006057. D.A. has received grants from the Wellcome Trust (221604/Z/20/Z) and Barts Charity (G-002145).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. L.K., F.C., M.Y., S.H.L., V.M., and D.A. researched data and reviewed and edited the manuscript. L.K., M.Y., and D.A. wrote, reviewed, contributed to discussion of, and edited the manuscript. K.B. and S.M.H. reviewed and edited the manuscript. D.A. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the British Society for Cardiovascular Research 50th Anniversary Meeting, Oxford, U.K., 4–5 September 2023; British Cardiovascular Society Annual Conference, Manchester, U.K., 3–5 June 2024; and 38th Meeting of the European Section of the International Society for Heart Research, Toulouse, France, 11–14 June 2024.