A high-energy diet can be used to induce type 2 diabetes in animals. So far, many diabetic animal models have been established.
In 2000, Reed et al  proposed a method for inducing insulin resistance in rats by using a high-fat diet and a low-dose streptozotocin (STZ). Later, Srinivasan et al  further modified and improved this model by testing four STZ doses (25, 35, 45, and 55 mg/kg) and using insulin secretion stimulating agent (glipizide) and insulin-sensitive agent (pioglitazone) to verify the insulin resistance and insulin secretion. They suggested a high-fat feeding for two weeks and 35 mg/kg STZ to induce a type 2 diabetes model in rats. Consequently, this approach has become the most commonly used T2DM model worldwide. This method can induce diabetic phenotype symptoms in a relatively short test period. Yet, when using pigs, this approach may not be enough [10, 14].
Some research suggested that a similar effect could be achieved in pigs by feeding animals with high-fat feeding for three months as two weeks feeding in rats [10, 15]. To achieve significant hyperglycemia, in this study, 90 mg/kg STZ was used. This dosage was selected based on previous rat model studies [11, 13] and our previous research results in minipigs.
In this experiment, Bama minipigs were fed with high-fat/high-cholesterol high-fat diet for three months. The body weight of animals without STZ was significantly higher than that of the control group. Although the fasting blood glucose and fasting insulin values were not significantly different from the control group, in the IVGTT experiment, the area under the blood glucose curve after glucose administration was larger than that of the control group (the blood glucose values at multiple time points after glucose administration was significantly higher than that of the control group, and the recovery of blood glucose values was slowed). The blood lipid levels (TC, HDL-C, and LDL-C) were also increased. These results were consistent with that of Larson et al . All of this suggested a mild insulin resistance.
In animals treated with 90 mg/kg STZ, fasting blood glucose increased compared to the control group. During hyperglycemia, the glucose tolerance of animals was impaired, but insulin secretion capacity still existed. IVGTT test showed that the insulin secretion level did not significantly increase after glucose administration, showing insufficient insulin secretion. This indicated that the Bama minipig T2DM model with significant hyperglycemia symptoms could be successfully established by feeding a high-fat or high-cholesterol and high-fat diet for three months and then applying 90 mg/kg STZ. The hyperglycemia symptoms can last for at least six months.
Recently, Yanjun et al  successfully established a type 2 diabetes model in Bama minipigs by using a similar method. They fed pigs with a high-sucrose and high-fat diet for six months, followed by an IV injection of 60 mg/kg STZ. Hyperglycemia symptoms continued until the end of the experiment (four months). These data suggested that a model can be established using a lower concentration of STZ; yet, it takes a longer time to establish a model.
Toxic effect of STZ on islet β cells
STZ has a selective toxic effect on islet β cells and relatively low toxicity to other tissue. The chemical composition of STZ is methyl-1-nitrosyl-C2-D glucose. Once STZ enters the body, it binds a GLUT2 receptor expressed on the β cells, causing DNA strand breakage, after which, repair mechanism is activated, which reduces nicotine adenine diglycoside nucleic acid and ATP, and in turn, leads to cell death.
The effect of STZ is related to the expression of glucose transporter GLUT2. The difference in toxicity to STZ between different animal species may result from different expression levels of GLUT2 [18–20]. Dufrane et al  showed that compared with rats and macaques, pigs express less GLUT2 receptors. Therefore, in pigs, a large dose of STZ is required to destroy a sufficient number of islet β cells, thus causing diabetes. Dufrane et al also suggested that150 mg/kg STZ could be used to destroy 97% of islet β cells in Landrace pigs, resulting in persistent diabetes. This result is consistent with our previous research results using Bama minipigs.
In this study, we used 90 mg/kg STZ and found that the application of STZ before or after high-fat/high-cholesterol and high-fat diet feeding may lead to different effects. The pre-application of STZ caused substantial damage to islet β cells, which was further confirmed by pathological examination (the number of islet β cells in minipigs with pre-application of STZ was significantly lower than that in the control group, and the morphological structure of islets could not completely recover to normal at nine months after STZ injection). However, nine months after treatment, the concentrations of fasting blood glucose and insulin were not significantly different in animals with pre-application of STZ and control pigs. In contrast, after three months of high-fat/high-cholesterol and high-fat feeding and the same dose of STZ, the animals showed significant hyperglycemia. The reason for such a difference in sensitivity to STZ toxicity is presumed to be related to the expression level of GLUT2 in islet β cells.
Minipigs have a large number of islet β cell reserves . Under normal feeding conditions, only a part of islet β cells are active, and the secreted insulin is sufficient to meet the needs of body blood glucose regulation. When feeding animals with a high-fat diet, they need more insulin to metabolize excess energy intake. Therefore, more islet β cells are activated, which then increases insulin secretion. The active islet β cells express more GLUT2 on the surface to transport more glucose. However, higher expression of GLUT2 receptor leads to higher STZ uptake. Therefore, STZ doses of 60 mg/kg in Yanjun’s experiment  and 90 mg/kg in our study do not cause blood glucose changes in minipigs under normal feeding conditions, but can induce damage to a large number of islet β cell in minipigs fed with high-fat diet, thus causing animal insulin secretion to fail to meet the needs of body metabolism, and eventually turning into diabetes.
Changes in the number of islet β cells in T2DM
T2DM is characterized by insulin resistance and insufficient insulin secretion. However, it is unclear whether this insulin secretion is caused by a decrease in the number of islet β cells, a substantial defect in secretion capacity, or both. Studies have shown that most patients with T2DM, whether fat or thin, have an absolute decrease in the number of islet β cells [21, 22]. It is worth noting that these evaluations on the number of β cells are mainly based on autopsy studies, not prospective studies. Therefore, knowing the number of β cells before onset is of crucial importance. In this study, two experimental groups received 90 mg/kg STZ in advance (pre-treatment). According to previous research results, it is predicted that although this dose of STZ does not cause hyperglycemia, it tends to cause a significant reduction in the number of islet β cells. Pathological examination results further confirmed this prediction result. However, there was no significant difference in fasting blood glucose and fasting insulin concentration between these animals and the control group, whether they were fed with a high-fat diet or high-cholesterol and high-fat diet for nine months. The animals’ weight was increasing, showing a trend toward obesity. There was no abnormality in any of the liver function and renal function with detected indexes, which shows that although the number of islet β cells was significantly reduced, the susceptibility of Bama minipigs to T2DM did not increase after nine months of treatment with high-fat or high-cholesterol high-fat feed . Therefore, we speculate that, although there are individual differences in the number of islet β cells during growth and development, the absolute decrease in the number of islet β cells in patients with T2DM is the consequence of disease occurrence and development, which may be related to the difference in susceptibility of individual animal islet β cells to pathogenic factors, and is irrelevant to the number of pre-existing islet β cells.
The relationship between diabetes mellitus, high-fat diet, and atherosclerosis
Diabetic patients are 2–6 times more likely to develop atherosclerosis than non-diabetic patients . Coronary heart disease is the most common fatal factor for adult diabetic patients . Up to 80% of patients with type 2 diabetes die of atherosclerotic macrovascular disease. Since wild-type mice and rats do not easily develop atherosclerosis, and experiments in transgenic mice have indicated that diabetes has a weak effect on promoting atherosclerosis , pigs spontaneously develope atherosclerosis with increasing age and in which diabetes can aggravate atherosclerosis are of utmost importance [26, 27].
In this study, obvious atherosclerotic plaque lesions in the abdominal aorta and iliac artery and mild fatty streak lesions of coronary arteries were observed in animals fed with high-cholesterol and high-fat diet (STZ + HCFD group and HCFD + STZ group), which is basically consistent with our previous research results . The proportion of abdominal aortic plaque lesion to the intimal area was used as an indicator of the severity of atherosclerosis. The results showed that diabetes could aggravate atherosclerosis induced by high-cholesterol and a high-fat diet, which is consistent with Gerrity et al . Unfortunately, the coronary artery lesions of minipigs in this experiment were mild. After 90 days of high-cholesterol and high-fat feeding and 60 days of diabetes, no obvious lesions were detected by the coronary CT. Pathological examination showed that only a few animals had mild fatty streak lesions in coronary arteries, which is inconsistent with the research results in other pig models [26, 27].
It is believed that saturated fatty acids (SFAs) in food increase the level of serum cholesterol, thus increasing the risk of atherosclerosis. Although Siri-tarino et al  admitted that replacing saturated fat with unsaturated fat is beneficial in preventing ischemic heart disease compared with carbohydrates, they proposed no clear evidence for the relationship between SFA intake and cardiovascular diseases . Scholars investigating the effect of diet insist that SFA intake should be reduced to 6–7% of total calories in a diet with a balanced and reasonable nutritional structure . Since these studies all come from statistical analysis of clinical data, they are inevitably affected by patients' heredity, disease background, and other conditions.
The study of animal models can selectively control the experimental conditions and provide a reliable tool for further analysis of SFA's mechanism. Our data suggested no difference in the occurrence and severity of atherosclerosis compared with the normal diet by simply increasing the content of saturated fatty acids in the feed of Bama minipig. Moreover, besides increased high-density lipoprotein cholesterol (HDL-C), serum triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and free fatty acid (FFA) did not increase, which is consistent with previous studies . In addition, this study further confirmed that cholesterol has an important role in the occurrence and development of atherosclerosis. As a risk factor, cholesterol is significantly higher than diabetes.
About diabetic microangiopathy and complications
Diabetic microangiopathy is one of the most vital factors causing persistent complications; yet, its pathogenesis is not completely clear. In the present study, we observed the retina and kidney changes, which are the most common microvascular complications of diabetes mellitus.
Diabetic ophthalmopathy, which includes diabetic retinopathy (DR) and diabetic macular edema (DMO), is the most common microvascular complication of diabetes . It is also the main cause of blindness in working-age adults . Before clinical symptoms appear, DR has undergone some histological changes. These early morphological changes include capillary basement membrane thickening, pericyte loss or ghost remnants, acellular capillary structure, endothelial cell proliferation, and microaneurysm formation. Capillary basement membrane thickening is considered the most common and consistent feature in the early stage of DR, which is commonly used as an early marker of experimental DR. In this study, four animals in each group treated with STZ later experienced significant hyperglycemia that lasted six months. No obvious abnormalities were found under conventional tissue sections, HE staining, and optical microscope of miniature pig retina. Consequently, the thickness of the capillary basement membrane was measured using a transmission electron microscope. It was found that the retinal capillary basement membrane of diabetic minipigs was significantly thickened, which was basically consistent with the research results of Lee et al  and Hainsworth et al . Lee and colleagues induced type 1 diabetes in Yorkshire pigs with STZ (150 mg/kg). Five diabetic pigs were obtained in this experiment; retinal capillary basement membrane thickening was detected in three pigs, which first occurred at the 18th week of hyperglycemia, and was not detected in the other two pigs, with detection being carried out in the 18th and 26th weeks of diabetes respectively. Hainsworth et al induced type 1 diabetes of Yucatan miniature pigs with alloxan (150–200 mg/kg). After 20 weeks of diabetes, the retinal capillary basement membrane thickened (five). Interestingly, in their study, the diabetic pigs (six) fed with a high-fat diet were the same as the control group, and the capillary basement membrane did not thicken. In our study, diabetic animals were fed with a high-fat or high-cholesterol and high-fat diet. The detection time was 24 weeks after suffering from diabetes, and the retinal capillary basement membrane was definitely thickened.
Diabetic nephropathy is the main cause of end-stage renal disease worldwide, with 35% of diabetic patients progressing to severe chronic renal disease . Histopathological findings of diabetic nephropathy include glomerular capillary basement membrane thickening, mesangial dilatation, mesangial sclerosis, and vascular lesions, such as hyaline degeneration . In this study, diabetic minipigs experienced continuous and significant hyperglycemia for six months. Serum GRE and BUN showed no significant difference compared to the control group. Except for urine sugar index, there was no obvious abnormality in the urine routine test and micro-urine protein detection. In addition, the optical microscope examination showed that there was no obvious abnormality in renal tissue morphology. However, the glomerular capillary basement membrane thickened. These results are consistent with Khairoun et al  and Maile et al . Maile and colleagues induced diabetes in Yorkshire pigs with STZ (150 mg/kg) for 22 weeks. They observed an increase in glomerular mesangial coefficient and thickening of the capillary basement membrane. Khairoun et al induced diabetes in pigs (Landrace x Yorkshire, T-line) with STZ (140 mg/kg) fed with an atherosclerotic diet for 15 months. They observed thickening of the glomerular capillary basement membrane (only typical individuals were detected), and glomerular mesangium showed a trend of expansion. At this time, no abnormality was found in clinical indexes (serum CRE level, urine albumin/creatinine ratio).
Clinically, ocular microvessels are often used as indicators of diabetic microangiopathy. This study showed that in the Bama minipig T2DM model, the symptoms of diabetes lasted for six months, and the conventional clinical indicators of microvascular complications did not show abnormalities. However, the retina and kidney's capillary basement membrane was thickened, which could be used as an indicator for the pathogenesis research and treatment evaluation of early microvascular diseases.