Experimental Models of Type 2 Diabetic Nephropathy
Ru Yuan-yuan,Ye Shan-dong
DOI: https://doi.org/10.3760/cma.j.issn.0366-6999.20120675
IF: 6.133
2013-01-01
Chinese Medical Journal
Abstract:With the rapidly rising incidence of diabetes, one of its complications, diabetic nephropathy (DN) is becoming one of the most common causes of end-stage renal disease in many countries. Reliable animal models of DN are a valuable tool for identifying the molecular mechanisms responsible for this disease and preclinical development of new therapeutic strategies. Various types of animal models have been used in the past; the induced methods and characteristics of these commonly used animal models were summarized in this review. DN is one of the most common and serious complications in type 2 diabetes, which is now becoming one of the major causes of end-stage renal disease in many countries. There is an urgent need to find additional effective therapeutic strategies, and this will require a greater understanding of the etiology, pathogenesis, and pathophysiological mechanisms responsible for the development and progression of DN. Many animals have been used recently, such as mouse, dog, pig, cat and so on. As a high degree of homology between mouse and human genomes has been confirmed, mouse models have been most commonly used, which are not only an essential and effective experimental tool for investigating the above mechanisms in the development of DN, but also promote preclinical development of new therapeutic strategies. To further progress and assist researches, a description of current protocols and key issues for developing type 2 DN experimental models has been detailed in the present review. MOUSE STRAINS A large number of mouse models of diabetes have been identified and their kidney disease characterized to various degrees. Genetic background and gender have a profound effect on the evolution of type 2 diabetes mellitus and DN. Generally, the islet β-cells in male rodents are more susceptible to the effects of streptozotocin (STZ) and tend to develop greater hyperglycemia. However, none of these models developed renal lesions that fully resemble advanced human diabetic renal injury. It has been suggested that the DBA/2 mouse is most susceptible to diabetic glomerulopathy that is most similar to human DN and the B6 mouse seems to be the most resistant. Many models of type 2 diabetes have been used, such as the Goto-Kakisaki Wistar rats, OLETF mouse, Zucker diabetic fatty rats, KKAy mouse, New Zealand obese mouse, hyperinsulinemic ob/ob mouse, and various strains of obese hyperinsulinemic db/db mouse. Diabetes was initially recognized in mice from the C57BLKS/J strain. The C57BLKS/J mouse shares 84% of its alleles with the common C57BL/6 strain and 16% with the DBA/2J strain and was initially maintained by Dr. Kaliss (KS). The C57BL/6J mouse, genomic host of ob/ob mutation, was reported to be insulin resistant with a weak first-phase insulin release. Diabetes in the db/db mice is a recessive, autosomal single gene mutation on chromosome 4, with complete penetrance. The db gene encodes for a G-to-T point mutation of the leptin receptor, leading to abnormal splicing and defective signaling of the adipocyte-derived hormone leptin, and the db/db mouse in the C57 black Kaliss background was designated from Jackson Labs updated nomenclature using the term C57BLKS/JLepr. Hyperglycemia and hyperinsulinemia are demonstrated in 1-month-old strains. Diabetes in the ob/ob mice is autosomal recessive, which could cause obesity, hyperglycemia and hyperinsulinemia, and the severity of the symptoms depends on the genetic background. It was demonstrated that ob/ob mouse in the C57BLKS/J background develops β-cell atrophy and severe hyperglycemia, whereas ob/ob mice in the C57BL/6J background develop hyperplasia of the pancreatic ducts and only mild hyperglycemia. The ob/ob mouse differs from the db/db mouse in that it is deficient in the production of leptin but allows intact leptin signaling, which causes hepatic lipogenesis and gluconeogenesis. The renal disease in ob/ob mice primarily demonstrates diffuse and nodular lipohyaline changes in glomeruli. The KK strain was characterized by hyperphagia, moderate obesity, and polyuria. These mice were later crossed with the C57BL/6 strain and inbred under the name Toronoto-KK. The KKAy mouse was obtained by transferring the Ay gene to the KK mice, whose obesity and hyperglycemia are more obvious. The advantage of using these mice lies in the early and rapid development of kidney damage. GOTO-KAKISAKI WISTAR RATS These rats were non-obese and mildly hyperinsulinemic, and their diabetes was stable, inheritable, and neither deteriorates nor ameliorates with age. The pathophysiology of these mice is characterized by impaired glucose-stimulated insulin secretion, reduced beta cells, excessive glycogen, and insulin resistance in muscle and adipose tissue. NEW ZEALAND OBESE (NZO) STRAIN This strain was developed by selective breeding for heavy weight, of a mixed colony of mice, until obesity with mild hyperglycemia and hyperinsulinemia became established. NAGOYA-SHIBATA-YASUDA (NSY) MICE NSY mice are selected from JCL CR mice with impaired glucose tolerance. Impaired islet β-cell insulin secretion and insulin resistance may be the pathophysiological mechanisms of diabetes in NSY mice, which are similar to human type 2 diabetes. ZUCKER FA/FA RATS Zucker fa/fa rats represent a typical model of hyperinsulinemia with obesity. The fa gene is a spontaneous autosomal recessive mutation on rat chromosome 5. These mice exhibit early and persistent hyperphagia, long-lasting hyperinsulinemia, and obesity without ketosis, which is similar to non-insulin-dependent diabetes mellitus. KNOCKOUT MODELS OF TYPE 2 DIABETES Peripheral insulin resistance and impaired insulin secretion by pancreatic β-cells are the two key pathophysiological mechanisms of type 2 diabetes. Various degrees of insulin resistance can be created in mice, depending on the specific protein and on its localization or importance in the insulin signalling cascade.1 As type 2 diabetes is a polygenic disease which involves defects at multiple levels in the insulin signaling cascade, knockout animal models have been developed. When homologous recombination is applied to the insulin receptor (IR) gene itself, 10% of adult heterozygous knockout mice, which exhibit a 50% reduction in protein content, develop diabetes. Mice homozygous for the IR knockout gene die within the first week of life from diabetic ketoacidosis. The major substrates of the IR kinase in most cells are IRS-1 and IRS-2. Mutations in IRS-1 have been described in 10%-20% of type 2 diabetic patients in several populations but disruption of IRS-1 in mice does not lead to diabetes. It was reported that IRS-1 deficiency leads to insulin resistance, and increasing insulin secretion with normal glucose tolerance because of compensatory β-cell hyperplasia, whereas GK deficient mice showed mild impaired glucose tolerance, with the resulting double knockout mice GK/IRS-1 developing both insulin resistance and impaired glucose tolerance. Monogenic knockout models of IR+/- and IRS-1+/- have no obvious clinical symptoms of type 2 diabetes. Mice double-heterozygous for the IR and IRS-1 targeted genes (IR/IRS-1+/-) exhibit a synergistic impairment of insulin action and develop severe insulin resistance with β-cell hyperplasia, whereas double knockout mice IR+/-/IRS-1+/- can develop obvious insulin resistance and clinical symptoms of type 2 diabetes. It was reported that IR and IRS-1 level in mouse liver and muscle, auto-phosphorylation of IR by insulin-mediated, tyrosine phosphorylation of IRS-1 and IRS-2, and the p85 subunit combined with IRS-1 were all decreased. IRS-2-/- mice develop overt diabetes in early life, because of severe insulin resistance in the liver and a lack of compensatory hyperplasia from pancreatic β-cells; however, β-cell insulin secretion by single β-cell is normal or even elevated in the mice. Mice with eNOS deficiency, OVE26 mice, and the BTBR ob/ob mouse currently appear to be the best murine models of advanced disease.2 INDUCED MODELS BY INJECTING CHEMICAL AGENTS INTRAVENOUSLY OR INTRAPERITONEALLY Several chemicals are selectively toxic to the pancreatic beta cells. STZ and alloxan are two important chemical agents used for developing animal models of diabetes and DN. STZ is an analogue of N-acetylglucosamine (GlcNAc) which is readily transported into pancreatic β-cells by GLUT-2, then selectively inhibits the activity of O-GlcNAcase and causes β-cell toxicity, resulting in insulin deficiency and hyperglycemia.3 In addition, islet β-cells can be further damaged by STZ-induced nitric oxide (NO) synthesis and the autoimmune process activated. It is effective in different, species-specific doses, ranging from 25 to 200 mg/kg in rats, dogs, mice and pigs. Alloxan, which is quickly uptaken by pancreatic β-cells, can affect cell membrane permeability and intracellular ATP production, and consequently inhibits glucose-mediated insulin secretion. Another, alloxan can block insulin secretion by inducing oxygen free radical production, which destroys the β-cells structure and then leads to β-cells injury and necrosis. It is reported that alloxan also causes direct liver and kidney toxic damage. In recent studies, alloxan-induced DM animal model has been rarely used, because disease remission can spontaneously occur in these mice. STZ and alloxan can been administered by intraperitoneal or intravenous injections. Intravenous injections are more commonly used because they are relatively easy to administer in rats, give more consistent results, and require expectedly lower STZ dosages than intraperitoneal injections. An important difference between STZ and alloxan is that only multiple doses of STZ, in repeated small doses insufficient for immediate diabetogenesis, produces in susceptible species a delayed. It was demonstrated that STZ-induced diabetes can be achieved in mice either with multiple low-dose injections of STZ, or a single high dosage of STZ (≥200 mg/kg), or a 2-dose regimen of STZ (100-125 mg/kg) given on consecutive days. If a single high dosage of STZ is given, pancreatic β-cells can be damaged severely, and animals should be monitored for severe hyperglycemia (blood glucose >33 mmol/L). Multiple low-dose injections of STZ-induced type 2 diabetes mellitus models were administered with T-lymphocyte-mediated destruction of the β-cells. Owing to reduced β-cell insulin secretion and increased insulin resistance, hyperglycemia is often associated with fat metabolism disorders, and these features can be well interpreted by this model. There were still some defects in STZ-induced models, including high mortality of animals, severe metabolic disorders which had to be treated with exogenous insulin, moderate kidney tissue lesions, especially no glomerular sclerosis changes compared with human DN. It is noteworthy that STZ and alloxan had direct renal toxicity, which made it difficult to distinguish hyperglycemiainduced renal functional and structural abnormalities from the drug-related non-specific cytotoxicity effect secondary to high doses of STZ or alloxan. It was reported that the kidney needs at least 3 weeks to recover from acute renal toxicity of STZ, following which the role of STZ in STZ-induced DN can be examined. HIGH-ENERGY DIET-INDUCED MODEL OF TYPE 2 DN The basic pathophysiological mechanisms of type 2 diabetes mellitus are reduced insulin secretion and insulin resistance. It is known that a low-dose STZ can partly destroy islet β-cell function, and high-energy diet can cause hyperglycemia and hyperlipidemia and peripheral tissue insensitivity to insulin. Thus, on the basis of low-dose injections of STZ, high-sugar and/or high fat diet can develop a new type 2 diabetic animal model, which is closer to human type 2 diabetes mellitus and has been increasingly recognized by many researchers. Chen et al3 induced a type 2 diabetic model in male Sprague-Dawley rats by a combination of STZ injection and high-fat diet (combination group). They found that levels of glucose, triglyceride, and total cholesterol in the serum of these rats were significantly higher than those in age-matched control group rats and solely high-fat diet group rats, accompanied by a reduction in body weight. In addition, a higher level of serum insulin and higher HOMA-IR were also observed in combination group rats. However, the long duration of modeling time and high cost of the model could not be ignored. It was reported that B6-db/db mice on standard chow that contains only 4% fat (wt/wt) exhibit only transient hyperglycemia, whereas those on as little as 8% fat maintain hyperglycemia as well as obesity. UNINEPHRECTOMIZED RAT MODEL OF STZ-INDUCED DN Models of STZ-induced DN have also been reported in different rat strains following uninephrectomy, which is thought to accelerate the progression of renal injury. Uninephrectomy results in enlargement of the remaining kidney, which is further increased by the development of diabetes. Kang et al4 demonstrated that glomerular capillary pressure, which promotes diabetic glomerular injury, was increased following uninephrectomy in SHR rats. In addition, male Wistar rats were made diabetic by a single intravenous injection of STZ (65 mg/kg) after uninephrectomy, and their blood glucose was then maintained between 16 and 22 mmol/L (300-400 mg/dl) for the next 8 months with insulin treatment. Uninephrectomy can significantly reduce the modeling time of DN and involves low cost. However, as it is difficult to dissect the relative contributions of STZ-induced hyperglycemia and uninephrectomyinduced changes in glomerular hemodynamics in the development of renal injury, interpretation of this model is complex. BIOCHEMICAL ASSESSMENT AND PATHOLOGICAL CHANGES OF DIABETIC RENAL INJURY In humans, DN is characterized clinically by the development of microalbuminuria, which progresses to macroalbuminuria and a decline in renal function. These clinical features are also seen in animal models of DN, although the level of albuminuria and the loss of renal function are much less severe. Albuminuria, whose degree and extent in mouse models of DN can be determined by an immunoassay or dipstick method. Generally, quantification of proteinuria by dipstick in mice is probably inaccurate, while immunoassay provides a reliable assessment of proteinuria as a result of glomerular injury. Urine albumin excretion (UAER) is considered to be one of the most sensitive markers of renal injury. Akita mice (Ins2) and low-dose STZ models of DN develop only 30-80 μg albumin excretion per day. In the most robust diabetic mouse model (C57BLKS db/db mice) 24-hour protein excretion is only about 300 μg/24 hours.5 UAER, that is detected by enzyme-linked immunosorbent assay (ELISA), normally requires rodents to be maintained in metabolic cages for 24 hours to collect urine, and this technique is time-consuming. On the contrary, the urine albumin/creatinine ratio (UACR) is widely used, when urine for the measurement is collected over a brief duration. In normal mice and most models of diabetes, UACR is usually 0.1-0.2 μg/μg. However, the DBA/2 mouse model does not fully reproduce human conditions, particularly with respect to less progression of albuminuria. On the contrary, DBA/2J and KK/HlJ mice developed significantly more albuminuria than C57BL/6J, MRL/MpJ, and A/J mice. Severe glomerular mesangial expansion, nodular glomerulosclerosis, and arteriolar hyalinosis were observed in diabetic DBA/2J and KK/HlJ mice. Glomerular filtration rate (GFR) is most commonly used as an indicator of renal function. This involves obtaining creatinine measurements in the serum or plasma over a 24-hour urine collection period. The picric-acid-Jaffe method, which was traditionally carried out in rodent models of renal disease, has been found to overestimate the rodent serum or plasma creatinine values, because of interference from hemoglobin and possibly other factors. Recently, HPLC or a creatininase assay was used as a reliable technique for analyzing creatinine clearance in rodent models of diabetic kidney disease. The major histological findings in human DN are glomerular basement membrane thickening, mesangial expansion and sclerosis with or without the development of nodular mesangial sclerosis (i.e. Kimmelstiel-Wilson nodules), tubulointerstitial fibrosis and arteriolar hyalinosis. These features, with the exception of Kimmelstiel-Wilson nodules, have also been detected in rodent models of STZ-induced DN, although their severity in rodents is usually milder. CONCLUSION All type 2 DN animal models can not fully resemble human diabetic renal injury; they especially lack typical renal interstitial fibrosis and do not demonstrate advanced renal failure. Nephropathy in humans is normally a late complication of diabetes, occurring in susceptible patients after 10-20 years of diabetes. Consequently mouse diabetic models need to be maintained for extended periods after the onset of hyperglycemia, usually 3-8 months, in order to detect substantial renal injury, and glomerular and tubulointerstitial lesions. Because of issues of cost and convenience, most studies of DN in mice have focused on the earlier harbingers of DN, including the development of albuminuria and histopathologic changes, and have not explicitly used renal insufficiency as an end point. So, the major limitations currently seen in many mouse models are lack of observed hypertension, an inability to measure a decline in renal function and the absence of advanced pathological lesions. Consequently, further researches including genetic manipulation may be required to engineer more appropriate rodent models of DN. For different research purposes, different types of mouse should be carefully chosen at present.