Given the widespread prevalence of diabetes mellitus and its associated cardiovascular complications globally, there is an urgent need to uncover the underlying molecular mechanisms and develop preventative and therapeutic strategies for this condition20. The pathogenesis of DbCM, one of the cardiovascular complications for patients with T2DM, remains unclear, with no effective approaches to prevent and treat21. The absence of reliable preclinical models that accurately mimic the pathophysiological processes of the human heart highlights the significance of developing a stable and effective model specifically designed to investigate DbCM mechanisms.
The protocol we present outlines a well-established and stable method for generating a DbCM mouse model, which serves as a robust tool for studying T2DM-associated myocardial dysfunction. HFD feeding to induce insulin resistance is generally regarded as an effective approach to model many complications associated with human diabetes22. In this method, HFD is combined with low-dose STZ injections, an agent that induces pancreatic islet damage. This method can better simulate the process of insulin resistance and relative insulin insufficiency.
Generally, preclinical models of DbCM fall into two categories: genetic and diet-pharmacologically induced models. The ob/ob mouse23,24, the db/db mouse25,26, and the Zucker diabetic fatty rats27 are the three most widely used genetic models28,29. These three rodent models develop increased body weight, hyperglycemia, impaired glucose handling, and insulin resistance due to the intervention of leptin signaling or its receptor. However, leptin has cardioprotective properties30, which can potentially complicate the interpretation of results from these models since alterations in leptin signaling may directly affect cardiac function. Additionally, mutations in the leptin receptor are rarely the cause of T2DM in humans31. Moreover, genetic models are also costly and present important differences from the human condition32. Compared with rats, mice are more readily available, cost-effective, and widely used in animal experiments. Some studies use pure diet-indued methods to establish the T2DM model, such as HFD and/or sugar (high sucrose or high fructose), which may cause obesity, insulin resistance, and diabetes33,34,35,36. However, the severity of hyperglycemia and insulin resistance is less pronounced compared to genetic models like the ob/ob or db/db mice37.
Combining diet-induced with diabetogenic agent STZ is considered more suitable for medical research than diet-based models alone. Although the general procedures of the HFD/STZ DbCM model have been described in several studies, we have made several improvements. Although T2DM predominates among diabetes cases in humans, creating an animal model for it presents greater challenges compared to T1DM. The pivotal factor lies in the dosage and frequency of STZ injections, which significantly impact model establishment. Excessive STZ doses result in a model closely resembling T1DM and lead to higher mortality rates.
Previous studies indicated that administering multiple low-dose injections of STZ at a dosage of 50 mg/kg better models T2DM31,38,39. However, these studies often have relatively short HFD feeding durations, ranging from 3 weeks with varying fat content. A study by Ahlke Heydemann40 showed that prolonged HFD feeding results in excess lipid storage, leading to lipid toxicity, obesity, decreased metabolic flexibility, oxidative stress, and chronic inflammation-conditions similar to those observed in DbCM heart tissue. In this study, we extended the HFD feeding period to 24 weeks and used a lower STZ concentration (30 mg/kg) to mimic DbCM pathology. C57BL/6J mice, which carry a gene predisposing them to T2DM, were used as the model animal41,42. Our observations revealed that after 12 weeks, the mice in the HFD/STZ group began to lose weight. It is important to note that weight loss in diabetes can be a complex issue. While insulin insufficiency and hyperglycemia can lead to weight loss, other factors such as decreased appetite, dietary changes, and increased physical activity may also play a role. Additionally, not all diabetic patients experience weight loss; some may even experience weight gain, particularly if they have insulin resistance and consume a high-calorie diet.
This method has some limitations. First, it takes a relatively long time to develop DbCM, 24 weeks in total. Additionally, STZ may exhibit toxicity toward organs and tissues beyond pancreatic islets, making the HFD/STZ model less accurate in replicating human disease. Furthermore, citrate buffer, used for STZ, has a pH of 4.5. Administering acidic solutions intraperitoneally can harm animals by causing red blood cell hemolysis, as well as pain and discomfort43, which may limit the model's applicability. Finally, the cardiac manifestations in T2DM patients are varied and often accompanied by hypertension, coronary artery disease, and other cardiovascular diseases, complicating clinical management. In this model, cardiac manifestations include LV dilation and hypertrophy, as well as systolic and diastolic dysfunction, but these do not fully represent the spectrum of DbCM seen in clinical practice. DbCM also includes features such as electrophysiologic alterations44 or atrial changes45, whereas the focus here is primarily on LV structure and pump function.
Consistent with previous studies, the method described here of combining HFD feeding with STZ injections aims to induce diabetes more rapidly by partially destroying pancreatic tissue and inducing insulin resistance to model DbCM. Enlarged cardiomyocytes with disorganized arrangement, myocardial fiber rupture, dissolution, reduced EF, FS, etc., indicate myocardial dysfunction and reduced cardiac function. In summary, we have presented an approach for establishing DbCM in a cost-effective and convenient manner. Utilizing noninvasive echocardiographic techniques and pathological staining, we assessed the efficacy of this method.