NEPHROLOGY 2007; 12, 261–266
doi:10.1111/j.1440-1797.2007.00796.x
Methods in Renal Research
Rodent models of streptozotocin-induced diabetic nephropathy GREG H TESCH1,2 and TERRI J ALLEN3 Departments of 1Nephrology and 2Medicine, Monash University, Monash Medical Centre, Clayton, and 3Einstein JDRF Centre for Diabetic Complications, Baker Heart Research Institute, Melbourne, Victoria, Australia SUMMARY: Streptozotocin-induced pancreatic injury is commonly used for creating rodent models of type 1 diabetes which develop renal injury with similarities to human diabetic nephropathy. This model can be established in genetically modified rodents for investigating the role of molecular mechanisms and genetic susceptibility in the development of diabetic nephropathy. In this report, the authors describe and compare the current protocols being used to establish models of diabetic nephropathy in rat and mouse strains using streptozotocin. The authors also list some of the histological criteria and biochemical measurements which are being used to validate these models. In addition, our review explains some of the key aspects involved in these models, including the impact of streptozotocin-dosage, uninephrectomy, hypertension and genetically modified strains, which can each affect the development of disease and the interpretation of findings. KEY WORDS:
diabetic nephropathy, mouse, rat, streptozotocin.
Diabetic nephropathy is clinically defined as the progressive development of renal insufficiency in the setting of hyperglycaemia. This disease is now the major single cause of end stage renal failure in many countries. Reliable animal models of diabetic renal injury are a valuable tool for identifying the molecular mechanisms responsible for this disease and for the preclinical development of new therapeutic strategies. Recently, a number of genetically modified (knockout and transgenic) mouse strains have been used to provide important insights into the roles of oxidative stress, advanced glycation end products, inflammation and profibrotic mechanisms in the development of diabetic nephropathy. Chemical agents, such as streptozotocin (STZ) and alloxan, that can selectively damage the insulin-producing b-cells in the pancreas resulting in hyperglycaemia, are important tools for developing animal models of diabetic complications. These reagents can be used to study diabetic tissue injury in most rodent strains, although the severity of injury is partly dependent on genetic background. Models that use STZ to induce type 1 diabetes, have been shown to develop modest elevations in albuminuria and serum creatinine and some of the histological lesions associated with diabetic nephropathy. Obtaining meaningful data from such Correspondence: Dr Greg Tesch, Department of Nephrology, Monash Medical Centre, 246 Clayton Road, Clayton, Vic. 3168, Australia. Email:
[email protected] Accepted for publication 26 February 2007. © 2007 The Authors Journal compilation © 2007 Asian Pacific Society of Nephrology
models is dependent on various factors, including: (i) a reliable method for establishing a consistent level of diabetes; (ii) being able to maintain a steady level of diabetes for the duration of the experiment; (iii) understanding the disease characteristics and progression of injury in the rodent strain being used; and (iv) the achievement of a pathological state which has clinical relevance. In order to assist researchers, this paper provides a description of current protocols and key issues for developing a rodent model of STZ-induced diabetic renal injury.
MATERIALS AND REAGENTS The following items are required to establish a rodent model of STZ-induced diabetes (Table 1).
METHODS Preparation and storage of reagents For each experiment, aliquots of STZ from the same batch are preweighed into plastic microfuge tubes which are then wrapped in aluminium foil (to protect against light sensitivity) and stored at -20°C with desiccant until use. Sodium citrate buffer (10 mmol/L, pH 4.5) is prepared by dissolving 147 mg of tri-sodium citrate in 49.5 mL of normal saline and adjusting the pH to 4.5 with approximately 0.5 mL of 1 mol/L citric acid. The citrate buffer should be used fresh or frozen in 1 mL aliquots and stored at -20°C. After thawing, each vial of frozen citrate buffer should be used immediately and unused contents discarded.
262
GH Tesch and TJ Allen
Table 1 Items required for establishing STZ-induced diabetes in rodents Chemical reagents Streptozotocin† Tri-sodium citrate† Citric acid† Normal saline (0.9%) Isophane insulin‡
Equipment
Consumables
Electronic weighing balance for streptozotocin (10.1 mg) Electronic weighing balance for mice (10.1 g) or rats (11 g) Electronic pH meter (10.1 units) Dispensing pipette (200–1000 mL) Blood glucometer and test strips§
Plastic microfuge tubes Aluminium foil 1 mL plastic pipette tips Disposable plastic syringe (1 mL) Needles for injection (27G or 29G)
†Available from Sigma-Aldrich, St Louis, MO, USA (website: http://www.sigma-aldrich.com). ‡Protophane, Novo Nordisk A/S, Bagsvaerd, Denmark. §Available from Abbott Laboratories, Bedford, MA, USA.
Effect of streptozotocin on pancreatic b-cells Streptozotocin is an analogue of N-acetylglucosamine (GlcNAc) which is readily transported into pancreatic b-cells by GLUT-2 and causes b-cell toxicity, resulting in insulin deficiency. STZ selectively inhibits the activity of b-cell O-GlcNAcase, which is responsible for removing O-GlcNAc from protein. This causes irreversible O-glycosylation of intracellular proteins and results in b-cell apoptosis.1
Low-dose mouse model of STZ-induced diabetic nephropathy Based on experimental studies performed over the past decade, the authors have formulated a reliable protocol for establishing diabetes in mice with multiple low-dose injections of STZ.2 Male mice aged 6–7 weeks (20–25 g body weight) are fasted for 6 h prior to injection. To induce diabetes, a microfuge tube containing preweighed STZ (ª10 mg) is mixed immediately before use with a predetermined volume of sodium citrate buffer to produce a final concentration of 7 mg/mL, and is dissolved with continuous pipetting for about 5 s. This solution is then injected intraperitoneally into each prestarved mouse at 55 mg/kg (7.86 mL/g) using a 29G insulin needle (total volume injected = 160– 200 mL). STZ degrades quickly in aqueous solutions and should be administered rapidly to obtain the best experimental results. Each tube with 10 mg of STZ will provide enough solution to inject six mice. Any remaining contents should be discarded according to the safety protocols of the researcher’s institute. To complete the induction of disease, this procedure must be repeated so that each mouse receives one STZ injection for five consecutive days. This protocol normally induces a suboptimal injury of pancreatic b-cells and progression of diabetes relies, in part, on a secondary autoimmune insulitis. After the completion of STZ injections, mice should be examined for the appearance of hypoglycaemia (blood glucose C57BL/6 > MRL/ MP > 129/SvEv > BALB/c); however, it is uncertain whether this order would apply generally to all mouse models using STZ. Using our above protocol, the authors usually find that >90% of STZ-treated C57BL/6 mice obtain sufficient diabetes to be used in animal model studies of diabetic nephropathy. The US-based Animal Models of Diabetes Complications Consortium (AMDCC, http://www.amdcc.org) is also proposing the adoption of a standard low-dose model for STZ-induced diabetic complications,
which include nephropathy. In their proposed model, which is still being finalized, mice (7–8 weeks of age) are starved for 4 h then briefly anaesthetized with isoflurane and injected intraperitoneally with 50 mg/kg of STZ for five consecutive days. Preliminary studies using this protocol indicate that approximately 50% of C57BL/6 mice will develop overt diabetes after 3 weeks (see http://www.amdcc.org) with non-fasting blood glucose levels 322 mmol/L (400 mg/dL). However, it is likely that some investigators will consider a 50% incidence rate of diabetes to be undesirable in terms of wastage of animals and resources. Therefore, it is uncertain whether this protocol will be widely used.
Moderate and high-dose mouse models of STZ-induced diabetic nephropathy Some studies examining diabetic nephropathy in mouse strains which are resistant to STZ-induced pancreatic injury have used either a single high dosage of STZ (3200 mg/kg) or a two-dose regimen of STZ (2 ¥ 100–125 mg/kg) given on consecutive days. Increasing the STZ dosage causes greater cytotoxicity and more rapid destruction of pancreatic b-cells, resulting in a higher incidence and severity of diabetes. However, at high doses, STZ has a non-specific cytotoxicity effect which has been shown to cause acute kidney damage in mice and rats.4,5 Consequently, models using high doses of STZ can develop a nephropathy which results from hyperglycaemia-induced injury superimposed on acute renal STZ-cytotoxicity, making it difficult to interpret any findings.4 The following protocol describes a two-dose procedure (2 ¥ 125 mg/kg per day STZ) for establishing diabetes in C57BL/6 mice with genetic deficiencies which facilitate mild resistant to STZ.6 Renal injury in this model does not appear to be associated with acute tubular cytotoxicity, based on the ability of insulin treatment to prevent renal pathology.6 An aliquot of STZ (10–15 mg/tube) is dissolved immediately before use with a predetermined volume of sodium citrate buffer to produce a final concentration of 15.6 mg/mL. This solution is then injected intraperitoneally into each mouse at 125 mg/kg (8 mL/g). The same procedure is repeated for each mouse at 24 h after the first injection. Using this procedure, approximately 90% of wild type C57BL/6 mice will develop overt diabetes within 2 weeks, with a lower incidence expected for more resistant genotypes.6 Because the pancreatic injury is more severe in this model, the diabetic mice will need to be monitored for severe hyperglycaemia (blood glucose >33 mmol/L, 600 mg/dL) and administered isophane insulin (see section Animal Welfare and Maintenance) to reduce blood glucose to a more tolerable range (16– 33 mmol/L, 300–600 mg/dL).
Rat models of STZ-induced diabetic nephropathy Models of STZ-induced diabetic nephropathy are commonly performed in Sprague-Dawley (SD), Wistar-Kyoto (WKY) or spontaneously © 2007 The Authors Journal compilation © 2007 Asian Pacific Society of Nephrology
263
Streptozotcin-induced diabetic nephropathy in rodents
hypertensive (SHR) rats. In these models, male rats at 8 weeks of age (200–250 g) are starved for 16 h and injected once into the tail vein with STZ (SD = 55 mg/kg, WKY = 60 mg/kg, SHR = 45 mg/kg) in sodium citrate buffer (1 mL/kg).7,8 STZ has also been administered intraperitoneally to rats, however, this is less common as intravenous injections are relatively easy to perform in rats and give more consistent results. In addition, the STZ dosage required to achieve diabetes via an intraperitoneal route is expectedly higher. Following the STZ injection, rats should be given drinking water supplemented with sucrose (15 g/L) for 48 h, to limit early mortality as stores of insulin are released from damaged pancreatic islets. At 1 week after STZ, rats should be assessed for hyperglycaemia and those with fasting blood glucose of over 15 mmol/L (280 mg/dL), which is usually around 90%, should be included in studies of diabetic nephropathy. To prevent subsequent development of ketonuria, diabetic rats should be given daily subcutaneous injections of long-acting insulin (2–4 U/rat, Protophane, Novo Nordisk Industries A/S, Bagsvaerd, Denmark) to maintain blood glucose levels in a desirable range (16–33 mmol/L, 300–600 mg/dL).9 Studies examining the effects of treatments on the development of diabetic nephropathy should not be started until at least 3 weeks after STZ when the kidneys have recovered from the acute mild nephrotoxic effects of STZ.5 Following induction of diabetes with STZ, the development of renal injury is accelerated and becomes more profound in SHR compared with normotensive rats (WKY).8 Vascular hypertension activates the renin-angiotensin system which alters renal haemodynamics, increases glomerular basement membrane thickness and promotes the development of inflammation and fibrosis in the setting of renal injury.10 A long-term study of STZ-induced diabetic nephropathy involving SHR has shown that the urine albumin excretion rate (UAER) is threefold higher in diabetic SHR (149 1 1 mg/24 h) at 32 weeks compared with control SHR (49 1 1 mg/24 h).9
Uninephrectomized rat model of streptozotocin-induced diabetic nephropathy Models of STZ-induced diabetic nephropathy have also been performed in different rat strains (SD, Wistar, SHR) 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. Uninephrectomy has been shown to increase glomerular capillary pressure in SHR rats which promotes diabetic glomerular injury.11 However, interpretation of this model is complex, as it is difficult to dissect the relative contributions of STZ-induced hyperglycaemia and uninephrectomyinduced changes in glomerular haemodynamics in the development of renal injury. In a study by Utimura et al.12 male Wistar rats (ª250 g) were uninephrectomized (right nephrectomy) during anaesthesia (50 mg/kg intraperitoneal sodium pentobarbital) and allowed to recover from surgery (3 weeks). They were then made diabetic by a single intravenous injection of STZ (65 mg/kg) and blood glucose assessed 2 days later. The blood glucose was then maintained between 16 and 22 mmol/L (300–400 mg/dL) for the next 8 months with insulin treatment. These uninephrectomized diabetic rats achieved a UAER of approximately 60 mg/24 h at 8 months which was nearly three times higher than non-diabetic control rats at the same age.
Biochemical assessment of diabetes Evaluation of rodent hyperglycemia is routinely performed by obtaining a drop of blood from the tail vein, placing it on a test strip, and measuring the glucose level with a standard patient glucometer. However, more accurate readings can be obtained by automated © 2007 The Authors Journal compilation © 2007 Asian Pacific Society of Nephrology
glucose-oxidase assays performed in biochemistry labs. To reduce the variations in blood glucose readings associated with feeding habits, blood glucose should be measured on animals after a standard fasting period at a designated time of day. This fasting time typically varies between 3 and 6 h among research groups performing mouse studies. However, fasting is not routinely performed before blood glucose measurements in rat models of STZ-induced diabetic nephropathy. For a more comprehensive measurement of average blood glucose levels, heparinized tail vein blood (ª25 mL) can be collected from rodents and assessed to determine the percentage of glycated haemaglobin. This assay is routinely performed by HPLC in hospital pathology labs. Because the blood cell turnover for rodents is approximately 30 days, a glycated haemaglobin reading provides an indirect assessment of the average blood glucose level over the previous month. Glycated haemaglobin levels greater than 7% have led to significant renal lesions in diabetic mouse kidneys.
Biochemical assessment of renal injury Urine albumin excretion is considered to be one of the most sensitive markers of renal injury. Normal mice have a UAER of approximately 10 mg/day. Studies of STZ-treated mice with a C57BL/6 background have detected a modest increase in UAER of 30–90 mg/day after 18–20 weeks with the highest levels being observed in hyperlipidaemic ApoE deficient mice.13 Measurements of UAER normally requires rodents to be maintained in metabolic cages for 24 h to collect urine. The urine volume is measured and aliquots stored frozen for subsequent measurement of albumin by enzyme-linked immunosorbent assay (ELISA). Previous studies have successfully used radioimmunoassay for assessing albuminuria;10 however, this technique is time-consuming. Reliable ELISA kits are now available from Bethyl Laboratories (Montgomery, TX, USA, http://www.bethyllabs.com) for measuring mouse and rat albumin and from Exocell (Philadelphia, PA, USA, http:// www.exocell.com) for mouse albumin. The albumin : creatinine ratio in urine can also be used to measure diabetic renal injury in rodents. This technique can be applied when metabolic cages are not available or if there is concern about the potential stress imposed on mice housed in metabolic cages. For these measurements, urine is collected by briefly allowing animals to urinate into a Petri dish. Urine creatinine can be assessed by a common colourimetric assay (picric-acid-Jaffe method), an enzymatic assay or an HPLC method.14 Renal function is most commonly assessed by calculating creatinine clearance as a measure of glomerular filtration rate (GFR). This involves obtaining creatinine measurements in serum or plasma and in a 24 h urine collection. This analysis has been traditionally performed in rodent models of renal disease using the picric-acid-Jaffe method. However, recent studies indicate that rodent serum or plasma creatinine values are overestimated using the Jaffe method, because of interference from haemaglobin and possibly other factors. In contrast, an enzymatic method (CREA plus, Roche Diagnostics, Mannheim, Germany) using creatininase has been shown to produce measurements of mouse plasma creatinine which correlate with HPLC values when samples show no visible haemolysis.14 Therefore, analysis of creatinine clearance in rodent models of diabetic kidney disease should be performed using reliable techniques such as HPLC or a creatininase assay. An alternative approach for determining GFR is to measure clearance of labelled inulin or diethylene triamine penta-acetic acid (DTPA). Inulin clearance measurements have been achieved in rats and mice by surgical intraperitoneal implantation of osmotic minipumps (Alza Corporation, Palo Alto, CA, USA) which are filled with FITC-conjugated inulin (Sigma, St Louis, MO, USA) that is released at a steady state.15 After implantation, urine from a 24 h collection and plasma are assessed for levels of FITC-inulin by fluorometry. The GFR based on clearance of inulin or creatinine is calculated
264
by the amount excreted in urine divided by the plasma concentration and is usually expressed as mL/min per g body weight in rodents. GFR measurements have also been determined in rodents by a single tail vein injection of 99mtechnetium-labelled DTPA (99mTc-DTPA).16 In this technique, GFR is calculated by measuring plasma radioactivity at a specified time after injection (43 min) which is then compared with a reference made at the time of injection.
Assessment of renal pathology Renal pathology in diabetic rodent kidneys can be routinely assessed on tissue sections stained with periodic acid Schiff ’s reagent and Harris haematoxylin. Kidney cross-sections 3–4 mm thick are fixed in 10% neutral buffered formalin for 2–3 h and then processed for paraffin embedding. In order to best preserve kidney morphology, some groups also perfuse the kidneys with fixative prior to removal and dissection, however, this procedure is not routinely used in literature. After processing, tissue sections 2–3 mm thick are attached to slides, dewaxed and stained with periodic acid Schiff’s reagent followed by haematoxylin according to standard textbook protocols. Microscope images of these sections can be used in the analysis of glomerular hypertrophy, glomerular and interstitial hypercellularity, tubular dilatation and atrophy, and interstitial volume.17 Additional pathological characterization can be performed by a number of different techniques. Electron microscopy is classically used to assess morphological changes including thickening of the glomerular basement membrane and mesangial expansion.10 Total collagen deposition, using text book histochemical stains such as Sirius Red or Masson Trichrome, can be used to evaluate fibrosis. Also, specific collagens, myofibroblast accumulation or inflammatory cells can be identified by immunostaining which is usually best performed on sections fixed in paraformaldehyde.17
Assessment of hypertension Although the progression of human diabetic nephropathy is strongly associated with hypertension, the blood pressure changes seen in STZinduced diabetic rodent models is usually mild unless the strain being used is spontaneously hypertensive. Hypertension is routinely measured by indirect tail-cuff plethysmography in non-anaesthetized rodents, and requires the averaging of repeated measurements at selected timepoints.18 This technique is particularly difficult in mice which need a lot of training to become familiar to the procedure without causing additional stress. More recently, radio telemetry has allowed continuous direct blood pressure monitoring in studies involving conscious rodents by inserting a radio-implant into an artery.9 Both of these methods have been used to evaluate the effects of antihypertensive treatments on the progression of STZ-induced diabetic nephropathy. The equipment used is relatively expensive and the procedures involved require a significant amount of training to be sufficiently skilled, however, the high sensitivity of these techniques can lead to results which provide important insight into therapeutic applications and disease mechanisms.
ADDITIONAL KEY ISSUES FOR EXPERIMENTAL DESIGN Animal maintenance and welfare When designing experiments in animal models of diabetic nephropathy, it is important to predetermine protocols for animal monitoring and criteria for intervention. This will help avoid animal wastage. Severely diabetic rodents can
GH Tesch and TJ Allen
suffer from weight loss, dehydration, cataracts, lethargy and diabetic coma. Diabetic animals should be visually monitored at regular intervals (2–3 times weekly) and assessed for health status using a checklist with specific scoring criteria (see example by David B. Morton at http://dels.nas.edu/ ilar_n/ilarjournal/41_2/Systematic.shtml). Rodents with suspected welfare problems should be examined more often, including measurements of food and water intake. Guidance for rodent monitoring, appropriate treatment or humane euthanasia can usually be obtained from journals (http:// www.lal.org.uk), animal welfare committees, veterinarians and trained animal technicians. Rodents with non-fasting blood glucose levels between 16 and 30 mmol/L can normally be maintained without intervention. Rodents with a non-fasting blood glucose above 35 mmol/L require insulin treatment to avoid weight loss and those below 4 mmol/L require administration of glucose or glucagon to avoid diabetic coma. Insulin treatment to lower blood glucose into a manageable range is best achieved by subcutaneous injection of a suboptimal dose of long-acting isophane insulin (e.g. Protophane). The insulin dose required will vary with species, strain and disease severity and should be determined by the researcher. Subcutaneous implants which continuously release insulin are less reliable and often result in episodes of hypoglycaemia and diabetic coma. Liquid nutrition supplements (e.g. Ensure, Abbott Laboratories) can help in preventing weight loss in severely diabetic animals when combined with insulin treatment.
Creating and validating a new model of STZ-induced diabetic nephropathy In order to create a reliable model of STZ-induced diabetic nephropathy, a number of preliminary findings need to be established with each rodent strain being used. Gender and genetic background will affect the susceptibility of rodents to STZ-induced pancreatic injury and to the development of diabetic renal injury. Male rodents are generally more susceptible to the effects of STZ and tend to develop greater hyperglycaemia. In addition, some strains of rodents are more hypertensive than others and will develop a more profound renal injury after the onset of diabetes. Recently, Qi et al.19 evaluated the development of STZ-induced diabetic nephropathy in various mouse strains with hyperglycaemia. This study showed that the level of hyperglycaemia alone was unable to account for the differences between strains in the severity of renal injury. When compared with the commonly used C57BL/6 strain, DBA/2J and KK/H1J mice were found to develop increased albuminuria and greater severity in renal morphological changes, including mesangial expansion, nodular glomerulosclerosis and arteriolar hyalinosis. Therefore, choosing the appropriate strain and gender of rodents should be considered carefully. When determining the effects of specific molecules in genetically modified strains (knockouts or transgenics), it is particularly important to make sure that the genetically modified rodents are only different to the wild type controls in the gene of interest. The appropriate dose of STZ © 2007 The Authors Journal compilation © 2007 Asian Pacific Society of Nephrology
265
Streptozotcin-induced diabetic nephropathy in rodents
required to induce a sustainable diabetes in >50% of rodents, without inducing direct renal injury, should be determined in both wild type and genetically modified strains. The incidence of diabetes obtained with the same STZ dose may vary between these strains. For example, the authors have previously found that a STZ dose inducing a >90% incidence of diabetes in wild type C57BL/6 mice produced only a 60% incidence of diabetes in MCP-1 deficient C57BL/6 mice.6 Therefore, the results of doseseeking studies in each strain should be considered together in selecting the single most appropriate dosage of STZ to be used in a major study which compares strains. A pilot study, with the selected dose of STZ, should then be performed in order to establish a time course of diabetic renal injury and choose appropriate endpoints. This information can then be used to design a major study and also determine appropriate points for potential therapeutic intervention. In humans, diabetic nephropathy is characterized clinically by the development of microalbuminuria, which progresses to macro-albuminuria and a decline in renal function. These clinical features are also seen in rodent models of STZ-induced diabetic nephropathy, although the level of albuminuria and the loss of renal function are much less severe. The major histological findings in human diabetic nephropathy are glomerular basement membrane thickening by electron microscopy in the absence of immune deposits, 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 STZinduced diabetic nephropathy, although their severity in rodents is usually milder. Based on present knowledge of human diabetic nephropathy, the AMDCC is currently proposing that a desirable rodent model of diabetic renal disease should include the following components: (i) a greater than 50% decline in GFR over the lifetime of the animal; (ii) a 3100-fold increase in the UAER compared with controls of the same strain, age and gender; and (iii) histopathology findings which include mesangial sclerosis (a 50% increase in mesangial volume), any degree of arteriolar hyalinosis, glomerular basement membrane thickening (a >25% increase compared with baseline by electron microscopy morphometry) and tubulo-interstitial fibrosis. Currently, there are no mouse or rat models which achieve the first two criteria as a result of diabetes, however, a number of studies have shown significant histopathological lesions which achieve or approach the histological criteria. Such models have already proved useful in our understanding of the mechanisms of diabetic renal disease and, often, the conclusions are supported by clinical and biopsy findings in human patients. Future developments in STZ-induced models of diabetic nephropathy, perhaps involving novel rodent strains, may provide the additional conditions necessary to achieve all the recommended criteria defined by the AMDCC. © 2007 The Authors Journal compilation © 2007 Asian Pacific Society of Nephrology
DISCUSSION Although the use of STZ is a robust method for inducing diabetes in rodents, the development of diabetic nephropathy in these animals shows limited resemblance to the human disease, presumably because of physiological, metabolic and hormonal differences. Consequently, extensive genetic manipulation may be required to engineer more appropriate rodent models of diabetic nephropathy. Genetic modified rodents have recently been used to create models of STZ-induced diabetic nephropathy with increased renal injury. These models are useful for testing novel therapies which target disease mechanisms. Mice which are genetically deficient in apolipoprotein-E (ApoE) have a reduced ability to clear plasma lipoproteins,13 which results in increased circulating levels of cholesterol and triglycerides. These ApoE–/– mice are more susceptible to vascular injury and, consequently, diabetic complications progress more rapidly in an ApoE deficient strain compared with equally diabetic wild type mice of the same background strain. Models of STZ-induced diabetic nephropathy in ApoE–/– mice have been used to examine the effects of PPAR-a and PPAR-g agonists and specific tyrosine kinase inhibitors as potential intervention treatments.20,21 Hypertensive transgenic (mREN-2)27 rats which have tissue overexpression of renin develop a more severe renal injury than either normotensive or SHR strains following induction of diabetes with STZ.21 Diabetic (mRen-2)27 rats have a greater than 50% decline in GFR with nodular and diffuse glomerulosclerosis reminiscent of diabetic nephropathy.22 This rat model has been used to examine the therapeutic benefits of antihypertensive agents and inhibitors of advanced glycation end products, specific kinases and transforming growth factor-b. However, a recent article suggests that long-term studies in this model may more closely resemble severe hypertensive nephrosclerosis than progressive diabetic nephropathy.23 In conclusion, the extensive use of STZ to create models of diabetic nephropathy indicates that this technique is an important and widely accepted tool for examining the mechanisms of diabetic renal injury and potential therapeutic interventions. In order to better compare and interpret findings obtained from different experiments performed around the world, it will be beneficial to obtain some general agreement on protocols for establishing and analysing models of STZ-induced diabetic nephropathy in specified rodent strains. It is hoped that information presented in this manuscript will help in developing such an agreement. ACKNOWLEDGEMENTS GHT is supported by a Career Development Award from the National Health and Medical Research Council of Australia, Kidney Health Australia and the Australian and New Zealand Society of Nephrology. TJA is a recipient of a Career Development Award/RD Wright Fellowship from the National Health and Medical Research Council of
266
Australia and Diabetes Australia. Animal model studies were supported by a Einstein Juvenile Diabetes Research Foundation Centre grant. REFERENCES 1. Lee TN, Alborn WE, Knierman MD, Konrad RJ. The diabetogenic antibiotic streptozotocin modifies the tryptic digest pattern for peptides of the enzyme O-GlcNAc-selective N-acetyl-beta-d-glucosaminidase that contain amino acid residues essential for enzymatic activity. Biochem. Pharmacol. 2006; 72: 710–18. 2. Candido R, Jandeleit-Dahm KA, Cao Z et al. Prevention of accelerated atherosclerosis by angiotensin-converting enzyme inhibition in diabetic apolipoprotein E-deficient mice. Circulation 2002; 106: 246–53. 3. Gurley SB, Clare SE, Snow KP, Hu A, Meyer TW, Coffman TM. Impact of genetic background on nephropathy in diabetic mice. Am. J. Physiol. Renal Physiol. 2006; 290: F214–22. 4. Tay YC, Wang Y, Kairaitis L, Rangan GK, Zhang C, Harris D. Can murine diabetic nephropathy be separated from superimposed acute renal failure? Kidney Int. 2005; 68: 391–8. 5. Kraynak AR, Storer RD, Jensen RD et al. Extent and persistence of streptozotocin-induced DNA damage and cell proliferation in rat kidney as determined by in vivo alkaline elution and BrdUrd labeling assays. Toxicol. Appl. Pharmacol. 1995; 135: 279–86. 6. Chow FY, Nikolic-Paterson DJ, Ozols E, Atkins RC, Rollins BJ, Tesch GH. Monocyte chemoattractant protein-1 promotes diabetic renal injury in streptozotocin-treated mice. Kidney Int. 2006; 69: 73–80. 7. Ma G, Allen TJ, Cooper ME, Cao Z. Calcium channel blockers, either amlodipine or mibefradil, ameliorate renal injury in experimental diabetes. Kidney Int. 2004; 66: 1090–98. 8. Cooper ME, Allen TJ, Macmillan P, Bach L, Jerums G, Doyle AE. Genetic hypertension accelerates nephropathy in the streptozotocin diabetic rat. Am. J. Hypertens. 1988; 1: 5–10. 9. Davis BJ, Johnston CI, Burrell LM et al. Renoprotective effects of vasopeptidase inhibition in an experimental model of diabetic nephropathy. Diabetologia 2003; 46: 961–71. 10. Allen TJ, Cao Z, Yousef S, Hulthen UL, Cooper ME. The role of angiotensin II and bradykinin in experimental diabetic nephropathy: Functional and structural studies. Diabetes 1997; 46: 1612–18.
GH Tesch and TJ Allen
11. Kang MJ, Ingram A, Ly H, Thai K, Scholey JW. Effects of diabetes and hypertension on glomerular transforming growth factor-beta receptor expression. Kidney Int. 2000; 58: 1677–85. 12. Utimura R, Fujihara CK, Mattar AL, Malheiros DM, Noronha IL, Zatz R. Mycophenolate mofetil prevents the development of glomerular injury in experimental diabetes. Kidney Int. 2003; 63: 209– 16. 13. Lassila M, Seah KK, Allen TJ et al. Accelerated nephropathy in diabetic apolipoprotein e-knockout mouse: Role of advanced glycation end products. J. Am. Soc. Nephrol. 2004; 15: 2125–38. 14. Keppler A, Gretz N, Schmidt R et al. Plasma creatinine determination in mice and rats: An enzymatic method compares favorably with a high-performance liquid chromatography assay. Kidney Int. 2007; 71: 74–8. 15. Qi Z, Whitt I, Mehta A et al. Serial determination of glomerular filtration rate in conscious mice using FITC-inulin clearance. Am. J. Physiol. Renal Physiol. 2004; 286: F590–96. 16. Allen TJ, Cooper ME, O’Brien RC, Bach LA, Jackson B, Jerums G. Glomerular filtration rate in streptozocin-induced diabetic rats. Role of exchangeable sodium, vasoactive hormones, and insulin therapy. Diabetes 1990; 39: 1182–90. 17. Chow FY, Nikolic-Paterson DJ, Atkins RC, Tesch GH. Macrophages in streptozotocin-induced diabetic nephropathy: Potential role in renal fibrosis. Nephrol. Dial. Transplant. 2004; 19: 2987–96. 18. Cooper ME, Allen TJ, Macmillan PA, Clarke BE, Jerums G, Doyle AE. Enalapril retards glomerular basement membrane thickening and albuminuria in the diabetic rat. Diabetologia 1989; 32: 326–8. 19. Qi Z, Fujita H, Jin J et al. Characterization of susceptibility of inbred mouse strains to diabetic nephropathy. Diabetes 2005; 54: 2628–37. 20. Calkin AC, Giunti S, Jandeleit-Dahm KA, Allen TJ, Cooper ME, Thomas MC. PPAR-alpha and -gamma agonists attenuate diabetic kidney disease in the apolipoprotein E knockout mouse. Nephrol. Dial. Transplant. 2006; 21: 2399–405. 21. Lassila M, Jandeleit-Dahm KA, Seah KK et al. Imatinib attenuates diabetic nephropathy in apolipoprotein E-knockout mice. J. Am. Soc. Nephrol. 2005; 16: 363–73. 22. Kelly DJ, Wilkinson-Berka JL, Allen TJ, Cooper ME, Skinner SL. A new model of diabetic nephropathy with progressive renal impairment in the transgenic (mRen-2)27 rat (TGR). Kidney Int. 1998; 54: 343–52. 23. Hartner A, Cordasic N, Klanke B, Wittmann M, Veelken R, Hilgers KF. Renal injury in streptozotocin-diabetic Ren2transgenic rats is mainly dependent on hypertension, not on diabetes. Am. J. Physiol. Renal Physiol. 2007; 292: F820–27.
© 2007 The Authors Journal compilation © 2007 Asian Pacific Society of Nephrology
