2.2.2. Models of Type2 induced diabetes-associated atherosclerosis
The prevalence of Type2 diabetes driven by insulin resistance and obesity is on an exponential rise worldwide with 700 million individuals predicted to have Type 2 diabetes by 2045[41]. Metabolic features of Type2 diabetes include hyperglycemia, hyperinsulinemia, obesity and dyslipidemia encompassing an increase in LDL, vLDL and a decrease in HDL levels. Thus, it is pertinent to choose an animal model that at best encompasses all of these features in order to study the mechanisms and explore therapeutic approaches to limit macrovascular disease induced by Type2 diabetes. The majority of the animal models available to study Type2 diabetes-associated atherosclerosis are either genetically manipulated or diet induced.
Two of the most common genetic models are generated by crossbreeding genetic models of Type2 diabetes, such as the leptin (ob/ob) and leptin receptor (db/db) deficient mice, with atherosclerotic mice such as the ApoE-/- and LDLR-/- mice. Leptin, the satiety hormone that is produced by adipose cells and enterocytes of the small intestine, regulates appetite via receptors in the arcuate nucleus of the hypothalamus. Ob/ob mice and db/db mice carry mutations in the leptin gene and the leptin receptor-coding gene respectively, causing the induction of hyperphagia leading to obesity and Type 2 diabetes[42, 43]. The db/db;ApoE-/- mouse is well established and exhibits typical features of Type2 diabetes by 20 weeks, such as increased body weight, hyperglycemia, hyperinsulinemia, elevated cholesterol levels (triglycerides, LDL and vLDL) with a 3-4 fold increase in atherosclerotic plaque burden[42, 43]. From a more diabetic perspective, these mice display higher expression of RAGE, inflammatory adhesion molecules (such as VCAM-1) and matrix metalloproteinase (MMP-9) activity[43]. Pharmacological interventions, such as soluble RAGE and PPAR-agonists have proven to be effective in their pre-clinical evaluation with this model[43]. Similar atherosclerotic observations were noted in the ob/ob;ApoE-/- and ob/ob;LDLR-/-mice, however, in these models either hyperglycemia was not evident (ob/ob; ApoE-/- mice) or hyperlipidemia was too pronounced masking the diabetic effects (ob/ob;LDLR-/-mice). A modification of the ob/ob;ApoE-/- and ob/ob;LDLR-/- mice to more closely align with the human lipoprotein profile is to cross these mice to ApoB100/100 producing mice which do not make ApoB-48 in their liver[44, 45]. Cholesterol in mice is mostly carried in ApoB-48 containing vLDL particles and chylomicron remnants whereas humans do not produce ApoB-48 in the liver[44, 45]. Ob/ob;ApoE-/-;ApoB100/100 and ob/ob;LDLR-/-;ApoB100/100 mice exhibit a metabolic phenotype which includes obesity, insulin resistance as well as atherosclerosis and hypertension[44]. Due to their specificity in lipoprotein profiles, these models are useful for studying cholesterol-lowering therapies from a Type2 diabetic/metabolic syndrome perspective[44].
For research that requires less influence of lipid levels and consistent features of Type2 diabetes, there are other models with a more clinically relevant lipoprotein profile that can be considered. The LDLR-/- mouse that is able to synthesise only ApoB100/100 protein represents a model more closely aligned to the human lipoprotein profile. This model exhibits accelerated atherogenesis[45]. Furthermore, Heinonen et al crossbred the LDLR-/-;ApoB100/100 mouse with transgenic mice overexpressing insulin-like growth factor-II in pancreatic β cells which led to the development of Type2 diabetes without major hyperlipidemia[46]. These mice demonstrated aggravated atherosclerotic lesions with increased complications such as lesion calcification which coincided with an increase in the expression of a variety of genes involved in calcification (OPN, ALP-2 and BMP-2) and inflammation (MCP-1)[46]. Moreover, the complexity of the lesion and increased calcification was more prominent in aged mice, which is in line with clinical studies that demonstrate calcification in peripheral and coronary arteries, particularly in older patients with a longer duration of diabetes, and is considered a strong independent risk factor for cardiovascular events[46-49]. Another useful genetic model is the ApoE-/- mouse that has a homozygous or heterozygous deletion of the insulin receptor substrate (IRS1 and IRS2) gene. ApoE-/- mice that are IRS1+/-, IRS2+/- and IRS2-/- mice display a Type2 diabetes/metabolic syndrome phenotype that includes, insulin resistance, hyperglycemia, hyperinsulinemia, and impaired glucose tolerance as compared to IRS2+/+ mice, however lipid levels were only modestly altered[50-52]. This is consistent with clinical findings in Type2 diabetic patients that have reduced expression of IRS2 in their pancreatic β cells and a highly polymorphic IRS1 gene[53]. These mice have augmented atherosclerotic lesions and vascular inflammation that is attributed to impaired IRS2 signalling of the AKT and ERK pathways that lead to the upregulation of pro-atherogenic MCP-1[54]. Lastly, another note-worthy genetic model of T2D-induced atherosclerosis is the heterozygous glucokinase (GK) knockout mouse on the ApoE-/- background. GK is the rate-limiting enzyme of glucose-stimulated insulin secretion, thus is it not surprising that GK+/-/ApoE-/-mice on a western diet show significant glucose intolerance and impaired glucose-stimulated insulin secretion, a phenomenon that was consistent over time[55]. Despite plasma lipid levels being comparable to the ApoE-/- mouse, GK+/-/ApoE-/- mice have significantly accelerated and highly developed atherosclerotic lesions, strongly suggesting that lesions are driven by hyperglycemia and insulin resistance[55].
Diet-induced insulin resistance is extensively studied in models of Type2 diabetes associated atherosclerosis. In particular, a high-fat western diet (comprising cholesterol (0.2% total cholesterol) + total fat (21% by weight; 42% kcal from fat) + high saturated fatty acids (>60% of total fatty acids) + high sucrose (34% by weight)) in ApoE-/- or LDLR-/- mice increases lesions throughout the aortic tree and exhibits hyperglycemia, hyperinsulinemia and hypertriglyceridemia. Interestingly, in mice fed a diabetogenic diet where 35% of the energy is sourced from fats, atherosclerotic lesions were inconsistent with the LDLR-/- mice being more susceptible to diabetes[56, 57]. LDLR-/- mice on a diabetogenic diet had severe dyslipidemia and increased atherosclerotic lesions, albeit with modest increases in glucose and insulin levels. On the contrary, diabetogenic diet-fed ApoE-/- mice displayed no significant differences between in lipids, glucose and atherosclerotic lesion sizes as compared to chow-fed ApoE-/- mice[56, 57]. In comparing neointimal formation, ApoE-/- mice fed a western diet exhibited larger lesions, accompanied by higher glucose and insulin levels, as compared to mice on a diabetogenic diet that maintained euglycemia but developed insulin resistance[58]. Treatment of these mice with Rosiglitazone, a thiazolidinedione class of PPAR agonist, led to a marked reduction in macrophage content and neointimal formation[58].
Due to ongoing discrepancies in lesion size between diet-induced and genetically manipulated mouse models, research groups have continually strived to modify diet composition to more closely reflect the human Type2 diabetic cardio-metabolic risk profile. Indeed, a diabetogenic diet supplemented with 0.15% cholesterol in the LDLR-/- mice further enhances obesity, insulin resistance, systemic inflammation and atherosclerosis with increased macrophage accumulation[59, 60]. Additionally, Neuhofer et al found that by altering complex carbohydrates with a high-fat, high-sucrose diet and the inclusion of 0.15% cholesterol, they were able to further augment atherosclerotic lesion size and increase adipose tissue inflammation, hallmarks often seen in human obesity and insulin resistance, thus creating a more robust model of metabolic syndrome with diet-induced obesity[61]. Furthermore, supplementing rodent diets with high fructose to emulate consumption of sugary drinks in humans and the impact on atherosclerotic lesions has been investigated by several groups. ApoE-/- mice fed a high fructose diet exhibited a pro-atherogenic state with insulin resistance, independent of hypercholesterolemia[62]. This cohort of mice exhibited increased vascular inflammation (VCAM-1), overproduction of NADPH-oxidase induced ROS and significant outward vascular remodelling[62]. Supplementation of liquid fructose to western-diet fed LDLR-/- mice demonstrated similar results with the mice developing increased atheromatous plaque driven by increased monocyte/macrophage infiltration and local inflammation. Importantly, atherosclerotic plaque size strongly correlated with plasma lipid levels[63]. Another method that has been described to model the metabolic syndrome associated with diabetes is the incorporation of low-dose STZ along with high fat feeding. Although this model has not been validated in terms of atherosclerotic lesions, it has been shown to develop cardiac inflammation and dysfunction, which combines both early and late stages of the disease[64, 65].