Animal models have been extensively used in diabetes research, including pharmacological rodent models of diabetes such as streptozotocin (STZ) and alloxan, drugs that when administred to rats or mice induce diabetes mellitus (DM) as soon as 24–48 h post-injection. These substances are selectively toxic to pancreas β-cells, induce insulin deficiency and hyperglycemia and therefore represents a model of type-1 diabetes [6]. On the other hand, selective inbreeding has produced several strains of animal that are considered reasonable models of type 2 diabetes and related phenotypes such as obesity and insulin resistance (Fig. 19.1 and Table 19.1). Apart from their use in studying the pathogenesis of the disease and its complications, all new treatments for diabetes, including islet cell transplantation and preventive strategies, are initially investigated in animal models. In recent years, a large number of new genetic animal models for the study of diabetes, including knockin or generalized and tissue-specific knockout mice have been described. Rodent models of type-2 diabetes include the Zucker fatty rat, as well as db/db and ob/ob mice, all of which display dysfunctional or absent leptin homeostasis and therefore develop insulin resistance at different timepoints (Table 19.1).

These rodent models share many features of human diabetic cardiomyopathy. For example, rodent models of obesity, insulin resistance and type 2 diabetes present left ventricle (LV) hypertrophy, diastolic dysfunction, increased cardiac fatty acid uptake and utilization, decreased cardiac efficiency, impaired mitochondrial energetics, increased myocardial lipid storage and impaired Ca²⁺ handling [7]. Moreover, depending on the model, in vivo studies have revealed susceptibility to systolic and diastolic dysfunction assessed by echocardiography and hemodynamics [8] as well as propensity to ischemia/reperfusion injury following ligation/unligation of the left anterior descending coronary artery (LAD), which occurs in conjunction with structural and functional changes to the LV [9].

However, there are several limitations when comparing these models to human diabetic cardiomyopathy, as spontaneous ischemia and atherosclerotic disease are not prominent in rodents [10]. The latter becomes simultaneously a valuable aspect as the effects of obesity, insulin resistance and diabetes on the heart can be studied independently of coronary artery disease [11]. Importantly, rodent models present sudden and uncontrolled hyperglycemia or insulin resistance, whilst in the clinical setting patients with diabetes are increasingly well controlled. Moreover, because DM develops at varying stages in these models, it is important to keep in mind that studies performed in animals before the onset of diabetes may reflect changes that are secondary to the underlying obesity and insulin resistance, and studies performed after the onset of diabetes may reflect the added effects of hyperglycemia of different durations.

Obesity is associated with premature death by increasing the risk of many chronic diseases, including type 2 DM, CV disease and certain cancers. Thus, not surprisingly, many obesity animal models are also diabetic and held major CV abnormalities (Fig. 19.2). As previously mentioned, many available models of obesity derive from selective crossings between rats comprising one out of the two most significant mutations in leptin receptor (i.e. fa and cp). Such is the case of Zucker fatty rats, the fa (fatty) gene “family”, whose obesity is not as marked as in SHROB or JCR:La-cp, the cp (corpulent) gene “family”. Figure 19.1 display several rat models of obesity derived from selective crossings among and between these two “families”. Many obese animal models, such as Zucker fatty rat, SHROB or even SHHF, are also models of hypertension or MS and thus were mentioned in previous sections. Another example is the Otsuka Long-Evans Tokushima Fatty (OLETF) rat obtained by selecting spontaneously diabetic rats exhibiting polyuria, polydipsia and obesity from an out-bred colony of Long-Evans rats. These rats exhibit obesity, hyperglycemia (at the adult age, 18 weeks), hyperinsulinemia, chronic DM, which is first non-insulin dependent but that over time progresses to insulin dependent. The control strain, LETO appears to share some of diabetogenic genes with OLETF rats [44].

In the human setting, the development of obesity and metabolic syndrome is mostly linked with increased caloric intake and lack of physical activity, in addition to genetic predisposition. Thus it would be relevant to study the pathogenesis of CV disease resulting from abnormal food intake. Many models of obesity are induced by the consumption of modified diets, such as high-fat or high-carbohydrate diets, therefore representing potentially more relevant pathophysiological models of the human disease. Such is the case of C57BL/6 mouse fed with a high-fat diet described previously as an animal model or vascular dysfunction and atherosclerosis. Still regarding mice, ob/ob mice [45], with leptin deficiency, and db/db mice [46], presenting a mutation in leptin receptor, constitute the most used models of obesity, which also display hyperglycemia and hyperlipidemia without hypertension and thus represent a good animal model of metabolic syndrome.

Apart from these models, other genetic modifications gave rise to a whole range of transgenic including knockout and knockin mice, which are briefly described throughout Table 19.1, wherein the most usual DM animal models are also described.

Aortic constriction (banding) is a well-established surgical technique to induce LV chronic pressure overload and hypertrophy. Initially, aortic banding imposes almost no restriction to aortic flow but as the animal grows, the severity of the constriction gradually increases, resulting in cardiac hypertrophy. Banding in several aortic segments (ascending (AAC), transverse (TAC) or supra/infra-renal abdominal aorta) has been used to mechanically reproduce the cardiac consequences of aortic stenosis, coarctation of the aorta and systemic hypertension [47]. The timeline of the disease progression depends on the selected specie, age, gender, anatomic locations or degree of the constriction (Fig. 19.3).

Changes in nitric oxide (NO) pathway are believed to play an important role in the pressure-overloaded heart and pathological cardiac remodelling (REM). In fact, reports showed that the phosphodiesterase-5 inhibitor sildenafil reduces LV HYP and dilatation in a mouse TAC model [48]. Additionally, the exogenous administration of the NOS cofactor BH4, has been shown to reduce LV HYP, fibrosis, and cardiac dysfunction in mice with pre-established pressure-overload hypertrophy. In this setting, BH4 re-coupled endothelial NOS, with subsequent reduction of NOS-dependent oxidative stress and reversal of maladaptive remodelling [49].

Among the major advantages that these banding models share compared to other hypertensive or HF models, is the ability to manipulate the degree of pressure overload by changing the constriction severity [50]. Concerning the thoracic aorta constrictions the main advantage is the similarities to human HF progression, especially to aortic stenosis patients. Accordingly, it is characterized by initial compensatory phase, with concentric LV hypertrophy followed by an enlargement of cardiac chambers associated with a further deterioration of LV function [51]. Another advantage is the extensive information regarding TAC model: it was first described in 1994 by Rockman [52] and it has been extensively used since then, especially in mice either by traditional thoracotomy approach or by minimally invasive aortic banding through a small incision in the proximal sternum [15]. Finally, this method permits the quantification of the pressure gradient across the aortic constriction and the stratification of LV hypertrophy [13]. However, the TAC model has several drawbacks such as the prolonged duration of the protocols (Fig. 19.3), inter-individual variability in the response to pressure overload, high proportion of debanding due to internalization of the constriction knot [15]. These two last disadvantages require the use of large experimental groups and the use of accessory methods for visualization of the constriction integrity and progression of disease, such as echocardiography. Both AAC and TAC models have a common disadvantage resultant from the complex surgical method and equipment necessary for open-chest microsurgery. The lack of such an extended learning curve is the major advantage of abdominal constriction model, alongside with the lower mortality rate associated with banding (10 %). Activation of RAAS might however limit the use of abdominal aorta constriction in some studies [53]. Moreover, decrease of LV relaxation rates makes such models valuable for the evaluation of diastolic dysfunction, which is an important factor in the progression of LV failure [54]. In addition, the stimulus for HF is gradual in onset as is the progression from compensated HYP to HF in humans, thus making it clinically more relevant.

Arteriovenous shunts or fistulas have been used to induce volume overload, HYP and later DCM and HF in rodents. Femoral artery to femoral vein or aorta to cava fistulas lead to HF, but present a mortality rate above 25 % in all studies [55]. The more recent aortocaval shunt is a relatively simpler and a faster alternative to induce HF with good survival rates and no need to perform thoracotomy [56]. Compared to the rat, reports on the cardiac consequences of arteriovenous shunts in mice are still scarce. The progression of myocardial changes is represented in Fig. 19.3. Notably conflicting observations have been reported probably due to differences in the techniques utilized to induce volume overload, duration and size of the shunt or even the stage of cardiac hypertrophy. This procedure has the advantage of being fast and usually well tolerated, despite the limitation of requiring a laparotomy. Nonetheless, shunt closure has been reported in 7 % of the cases, which means that it is necessary to confirm the patency of the shunt at the end of investigation.

Another procedure used to induce volume overload in rats is aortic valve regurgitation, in which an aortic valve cusp is punctured [57].

Doxorubicin (DOX) is an anthracycline widely used in cytostatic treatments. One of the major long-term consequences of DOX therapy is the development of dose-dependent cardiomyopathy and ultimately congestive HF either in humans as in various animal species. Therefore, understanding the pathogenesis of cardiotoxic cardiomyopathy is essential to the development of new measures to prevent cardiotoxicity associated with antineoplastic therapies [15].

DOX administration once a week for 6 weeks or on alternate days for 2 weeks has been shown to induce cardiomyopathy and HF in rodent, while in rabbits the first abnormalities are observed after 3 months’ injections twice weekly. Interestingly, a single intravenous administration of DOX has been shown to induce significant LV dysfunction in mice after 5 days, while its intracoronary injection allows a smaller dose of DOX to induce HF without systemic toxicity [15].

DOX-induced cardiomyopathy is characterized by ventricular wall thinning and dilatation (reduced mass-to-volume ratio), and depressed systolic and diastolic function accompanied by fluid retention and by neurohumoral activation. At the cardiac muscle level, DOX promotes intrinsic contractile dysfunction and reduced contractile reserve. Furthermore, DOX impairs vascular, as well as cardiac endothelial function and induces inflammatory reactions in the heart, leading to thrombosis in the atria and myocarditis [15].

Multiple pathways of antracycline-induced cardiac cellular injury have been proposed, such as the release of cardiotoxic substances which subsequently accumulate in cardiomyocytes, the generation of free radical, lipid peroxidation and suppression of DNA, RNA and protein synthesis. Other studies suggest that cardiotoxicity pathways include abnormalities in Ca²⁺ handling; induction of mitochondrial DNA lesions; degradation of myofilamental and cytoskeletal proteins, including titin and dystrophin; interference with various pro-survival kinases [59] and changes in adrenergic and adenylate cyclase function. All these mechanisms may contribute to cardiac cell damage, ultimately leading to myocyte death, either by necrosis or by apoptosis [15].

DOX model has a short time course of induction of HF and also the advantages of being technically simple, reproducible, non-invasive, and economical. Additionally, it can be used in several animal species to promote either chronic or acute HF [60]. The main limitations of this model are related to the variable degree of ventricular dysfunction and the high incidence of arrhythmias that contributes to the high mortality rate. Importantly, it is known that doxorubicin cardiotoxicity is dose-dependent and may appear either during and within 2–3 days of its administration or later within 30 days of administration of its last dose. In humans, it may occur even after 6–10 years after its administration [61]. Anthracycline administration also has undesirable bone marrow, gastrointestinal and renal toxicity that can however be avoided by the intracoronary injection of the drug [60