Sage Journals: Discover world-class research

Abstract

Male obese Zucker Diabetic Fatty (ZDF) rats develop type 2 diabetes around eight weeks of age, and are widely used as a model for human diabetes and its complications. The objective of the study was to test whether the complications manifested in the kidney and nerves of ZDF rats really correspond to human diabetic complications in their being related to the hyperglycaemic state. Four groups of ZDF rats were used. One lean (Fa/?) and one obese (fa/fa) untreated group served as non-diabetic and diabetic controls. In two further groups of obese (fa/fa) rats, diabetes was prevented by pioglitazone or delayed by food restriction. All rats were monitored up to 35 weeks of age with respect to their blood glucose, HbA1c and insulin levels, their kidney function (urinary glucose excretion, renal glucose filtration, glomerular filtration rate, albumin/creatinine ratio), and their nerve function (tactile and thermal sensory threshold and nerve conduction velocity). Pioglitazone prevented the development of diabetes, while food restriction delayed its onset for 8–10 weeks. Accordingly, kidney function parameters were similar to lean non-diabetic rats in pioglitazone-treated rats and significantly improved in food-restricted rats compared with obese controls. Kidney histology paralleled the functional results. By contrast, nerve functional evaluations did not mirror the differing blood glucose levels. We conclude that the ZDF rat is a good model for diabetic nephropathy, while alterations in nerve functions were not diabetes-related.

Keywords

Animal model disease model rodents metabolism refinement

Microvascular late complications of diabetes mellitus represent an important unmet medical need. Diabetic nephropathy is the most common cause of end-stage renal disease in developed countries.¹ It starts with microalbuminuria and progresses to macroalbuminuria, decreased glomerular filtration rate (GFR) and finally renal failure.² Up to 50% of diabetic patients develop peripheral polyneuropathy.³ They suffer from pain, paraesthesia, hypersensitivity or sensory loss and numbness in the distal extremities.⁴ Tight glycaemic control reduces the risk of late complications,^5–8 although it is not able to eliminate it. Symptomatic treatment provides improvement of quality of life, but cannot prevent progression of disease. Treatments targeting the pathophysiological alterations underlying microvascular complications are needed, but so far, no approach has proven fully satisfactory.

For the development of new therapeutic options for the microvascular complications, animal models that mirror the pathogenesis in humans are essential. Male Zucker Diabetic Fatty (ZDF) rats with fa/fa-genotype develop obesity, insulin resistance and non-insulin dependent diabetes within the first months of life due to a leptin receptor defect and a genetically reduced insulin promoter activity.⁹ Female ZDF rats with fa/fa-genotype similarly become obese and insulin-resistant, but do not progress to hyperglycaemia, except when fed a high-fat diet. Male ZDF rats are widely used as animal models for human type 2 diabetes, and in a number of studies of diabetic nephropathy^10–14 and neuropathy.^15–17 However, as the primary genetic alteration in ZDF rats is a leptin abnormality, it cannot be taken for granted that pathological changes observed in this model are in all cases consequences of diabetes. The aetiological relationship between hyperglycaemia and microvascular pathologies remains to be verified.

Treatment of prediabetic ZDF rats with a PPARγ agonist normalizes blood glucose through an insulin-sensitizing effect and preserves beta cell function.^18,19 As obesity of fa/fa-ZDF rats is caused by hyperphagia, food restriction can counteract or at least delay diabetes development.²⁰ In both cases, incidence, progression and severity of microvascular complications should be reduced, if their pathogenesis corresponds to the human disease.

The aim of the current study was to evaluate whether nephropathy and neuropathy found in male diabetic ZDF rats are linked to the hyperglycaemic state of the animals and are therefore real diabetic late complications, or rather related to other characteristics of the fa/fa-genotype. For this purpose, male obese ZDF rats were treated with either pioglitazone or food restriction over several months and metabolic parameters as well as manifestations of nephropathy and neuropathy were examined.

Materials and methods

Animals

The study was performed under the terms of the German Animal Protection Law and in accordance with the Sanofi-Aventis Company Charter on the Humane Care and Use of Laboratory Animals. Male lean (genotype Fa/?) and obese (genotype fa/fa) ZDF rats (ZDF-Lepr ^fa /Crl) were purchased from Charles River Laboratories (Brussels, Belgium) and allowed to acclimatize for at least one week after arrival. They were maintained two per cage on a constant 12 h light/dark cycle under controlled temperature and humidity conditions. Water was freely accessible. They were fed Purina 5008 diet (PMI Nutrition International, Richmond, IN, USA) ad libitum, with the exception of the food-restricted group, which got 70% of the amount of food consumed by the obese control rats the day before.

Experimental design

At the age of seven weeks, rats were allocated to one of four groups (n = 8 each): lean untreated control rats, obese untreated control rats, obese rats treated with pioglitazone (Actos^®, Takeda, Osaka, Japan) and obese rats treated with food restriction to 70% of the consumption of obese control rats. All animals received daily oral gavage of either vehicle (5% solutol, 0.5% hydroxyethyl cellulose) or dissolved pioglitazone (10 mg/kg). Treatment was continued for 28 weeks until the end of the study. Rats were then aged 35 weeks. Body weight and food consumption were recorded twice a week.

Blood parameters

Tail tip capillary blood was collected weekly for determination of glucose and HbA1c. Every four weeks, blood was sampled from the retrobulbar venous plexus under light isoflurane anaesthesia for analysis of serum insulin and creatinine. All metabolic parameters were determined under fed conditions.

Urinary parameters

At four-week intervals, rats were placed individually in metabolic cages for 24 h and urine was collected. Urine volume was determined and urinary glucose, albumin and creatinine concentrations were analysed. Urinary glucose excretion (UGE), renal glucose filtration (RGF), urinary albumin/creatinine ratio (ACR) and GFR were calculated.

Analytical procedures of blood and urinary parameters

Standard procedures were used to determine blood glucose, HbA1c, serum creatinine, urinary glucose, urinary albumin and urinary creatinine.²¹ Serum insulin concentrations were assayed with an enzyme-linked immunosorbent assay obtained from Mercodia AB (Uppsala, Sweden).

Sensory thresholds and nerve conduction

Every four to six weeks, tactile and thermal sensory thresholds were assessed. Tactile withdrawal thresholds were measured with a Dynamic Plantar Aesthesiometer (Ugo Basile, Comerio, Italy). Rats were placed in a cage with wire mesh floor and allowed to acclimatize for several minutes. The testing device was placed under the plantar face of the hind paw and applied an incremental force (measured in grams) via a metal rod (diameter 0.5 mm), until the animal withdrew the paw. A cut-off value of 50 g was set to prevent tissue damage. Testing was repeated on the contralateral paw and the mean withdrawal threshold was calculated for each animal. Thermal withdrawal thresholds were measured using the tail flick test (Ugo Basile). Rats were gently restrained on the testing apparatus and the tail tip was placed over a radiant heat source (157 mW/cm²). Time from the onset of heat stimulus to tail withdrawal was recorded. Cut-off point was set to 16 s to prevent tissue damage.

Nerve conduction parameters of the tail nerve were measured every two to four weeks. Rats were lightly anaesthetized with isoflurane. Needle electrodes were inserted subcutaneously into the tail, with the recording electrode at the tail base and the stimulating electrode placed 5 cm distally. Supramaximal stimulation pulses of 0.2 ms duration were applied, and amplitude, duration and latency of the resulting whole nerve action potential were recorded. Maximum nerve conduction velocity was calculated from the latency between stimulus onset and beginning of the action potential and the distance between stimulating and recording electrode.

Kidney histology

At 35 weeks of age, rats were euthanized; kidneys were taken out, weighed and fixed in 10% buffered formalin. Tissues were processed by routine procedures. The specimens were stained with haematoxylin-eosin and periodic acid Schiff. A longitudinal section of one kidney per animal was examined. Absolute incidences of renal lesions were recorded, and the extent of lesions was assessed semi-quantitatively by an examiner blinded to the treatment of the animal. Damage to individual glomeruli and distribution of the damaged glomeruli were graded in a combined scheme of glomerulopathy, which reflects the experience in a semi-quantitative scoring system. The grading score describes severity (A) and quantity (B) in each slide according to the following definitions:
(A) Severity: The glomerulopathy observed in ZDF rats consists of the following findings (in sequence of increasing severity): increase in glomerular size, widening of glomerular capillaries (microaneurysma), broadening of mesangial structures, occasionally accompanied by an increase in cellular population, segmental synechia with Bowman's capsule, segmental or nodular thickening of the mesangium, partial or complete sclerosis.

(B) Quantity: Grade 1 (minimal): single glomeruli affected; grade 2 (mild): few glomeruli affected; grade 3 (moderate): some glomeruli affected; grade 4 (marked): many glomeruli affected; grade 5 (severe): majority of glomeruli affected.
A similar semi-quantitative scheme was applied to assess tubulointerstitial damage.

Statistical analysis

Data are expressed as means ± SEM. Statistical significance was determined only for endpoint values of each parameter by one-way analysis of variance with Bonferroni's multiple comparison post-test using the GraphPad Prism^® 5 software. P values <0.05 were considered statistically significant.

Results

During the treatment period of 28 weeks (age of the rats: 7–35 weeks), there were spontaneous deaths in the diabetic obese control group (group II: 3 rats in treatment weeks 9, 15 and 28, respectively) and in the pioglitazone-treated obese group (group III: 2 rats in study weeks 15 and 28, respectively). All pioglitazone-treated rats reaching the regular end of the study (n = 6) demonstrated histologically verified cardiomyopathy (which was not the subject of this manuscript). In the lean control group (group I), unfortunately, we lost two rats during anaesthesia for the measurement of nerve conduction velocity during the last two measurements; in this group six rats finished the treatment period of 28 weeks although kidneys from only five rats were histologically evaluated (Table 1). There were no losses in the food-restricted group (group IV).
Table 1
Kidney histology. Incidence and average grading of glomerulopathy, tubulopathy and Armani-Ebstein cells in kidneys of lean and obese untreated ZDF rats and obese ZDF rats treated with pioglitazone or food restriction

Lean control (n = 5)
Obese control (n = 5)
Pioglitazone (n = 6)
Food restriction (n = 8)

Incidence (%) Score Incidence (%) Score Incidence (%) Score Incidence (%) Score

Glomerulopathy 0 0 100 3.2 0 0 100 1.4

Tubulopathy 0 0 100 3.2 50 1.7 100 1.5

Armani-Ebstein cells 0 0 100 1.6 0 0 88 1.3

Metabolic parameters

Body weight of obese control rats was higher than that of lean rats at the beginning of the study, but then started to decline at around 16 weeks of age. At 25 weeks of age, lean rats reached the same body weight as obese rats and finally grew slightly heavier. Food restriction did not influence body weight of obese rats. Pioglitazone resulted in an enormous boost in body weight, which came to be more than 1000 g (Figure 1a). The three groups of obese ZDF rats showed hyperinsulinaemia at the start of the study. Serum insulin of obese untreated rats dropped to the level of control rats during the trial. Rats on food restriction showed prolonged hyperinsulinaemia and delayed decline. Pioglitazone preserved increased insulin secretion throughout the study (Figure 1b). Lean control rats had stable blood glucose values of around 6 mmol/L and HbA1c values of around 4%. Blood glucose of all three groups of obese rats was already elevated at the beginning of the study and rose to 28 mmol/L in obese control rats. Food restriction as well as pioglitazone treatment initially resulted in normoglycaemia, which was maintained during the study period in the pioglitazone group. In the food-restriction group, blood glucose started mounting again after four weeks of treatment and reached 20 mmol/L after eight weeks. Final blood glucose values were 22 mmol/L in this group (Figure 1c). HbA1c values started at 5% in obese control rats and ascended to around 9%. HbA1c of food-restricted obese rats stayed constant at 5% until 14 weeks of age and then increased to reach a level slightly below that of obese controls between 20 and 24 weeks of age. Pioglitazone treatment normalized HbA1c levels to that of lean control rats (Figure 1d).

Figure 1
Body weight (a), serum insulin levels (b), blood glucose (c) and HbA1c values (d) of lean (white circles) and obese untreated ZDF rats (black circles) and obese ZDF rats treated with pioglitazone (white squares) or food restriction (black squares) (n = 5–8, mean ± SEM, fed conditions). P < 0.05 versus lean control, ^† P < 0.05 versus obese control. The error bars (SEM) in (a) as well as for lean control in (b), lean control and pioglitazone-treated groups in (c) and (d) are smaller than the size of the symbols

Kidney function and histology

Lean control rats and rats treated with pioglitazone never showed UGE during the trial. Obese control rats already had high UGE at 12 weeks, while UGE of food-restricted rats augmented from week 12 to week 20, but did not reach the level of untreated obese rats (Figure 2a). RGF showed a similar time course, with high levels in untreated obese rats from week 12 onwards and an increase from near normal levels to clearly elevated ones in food-restricted obese rats. Pioglitazone treatment held RGF on normal levels (Figure 2b). Urinary albumin excretion rose massively and progressively in obese control rats and to a much smaller degree in obese rats on food restriction. Pioglitazone prevented urinary albumin excretion (Figure 2c). Urinary ACR showed progressive increase in obese control rats and – to a lesser extent – in obese rats with food restriction, while it was constantly low in lean controls and pioglitazone-treated rats (Figure 2d). GFR was augmented in obese controls and in the food-restriction group compared with controls, and slightly diminished in the pioglitazone group (Figure 2e). Serum creatinine was slightly reduced in obese control and pioglitazone-treated rats compared with lean controls and even more reduced in obese food-restricted rats (Figure 2f).

Figure 2
Urinary glucose excretion (UGE) (a), renal glucose filtration (RGF) (b), urinary albumin (c), urinary albumin/creatinine ratio (ACR) (d), glomerular filtration rate (GFR) (e) and serum creatinine (f) of lean (white circles) and obese untreated Zucker Diabetic Fatty (ZDF) rats (black circles) and obese ZDF rats treated with pioglitazone (white squares) or food restriction (black squares) (n = 5–8, mean ± SEM, fed conditions). P < 0.05 versus lean control, ^† P < 0.05 versus obese control

Absolute kidney weight at autopsy was significantly elevated in the obese control versus lean control group. Both treatment groups had significantly lower kidney weight compared with obese controls, but kidney weight of the pioglitazone group was still significantly higher than that of lean controls (Figure 3a). Relative kidney weight related to body weight was also significantly raised in the obese control versus lean control group. Relative kidney weights of food-restricted rats were between those of lean and obese control rats. Pioglitazone-treated rats had lower relative kidney weights than lean controls (Figure 3b).

Figure 3
Absolute kidney weight (a) and relative kidney weight per body weight (b) of lean and obese untreated Zucker Diabetic Fatty (ZDF) rats and obese ZDF rats treated with pioglitazone or food restriction (n = 5–8, mean ± SEM). P < 0.05 versus lean control, ^† P < 0.05 versus obese control. KW: kidney weight; BW: body weight

None of the lean control rats showed kidney glomerulopathy or tubulopathy, while 100% of obese untreated rats did, with an average scoring of 3.2 for both pathologies. Food-restricted rats also had 100% incidence for glomerulopathy and tubulopathy, but with milder appearance (average scores 1.4 and 1.5). Pioglitazone-treated rats did not show glomerulopathy and only 50% incidence of mild tubulopathy (score 1.7) (Table 1 and Figure 4). Armani-Ebstein cells as a correlate of glucosuria were present in all diabetic control rats (score 1.6) and in 88% of diabetic food-restricted rats (score 1.3), but not in lean controls and in pioglitazone-treated rats (Table 1).

Figure 4
Kidney glomeruli (first row) and tubuli (second row) of lean and obese untreated Zucker Diabetic Fatty (ZDF) rats and obese ZDF rats treated with pioglitazone or food restriction. Representative histological slices of results are summarized in Table 1

Nerve function

Tactile withdrawal thresholds measured with the Dynamic Plantar Aesthesiometer were not different between lean and obese control rats and obese rats with food restriction. Only pioglitazone-treated rats showed slightly lower withdrawal thresholds in the beginning (Figure 5a). Thermal withdrawal thresholds measured with the tail flick test were lower in the three groups of obese rats, treated and untreated, compared with lean control rats (Figure 5b). Maximum nerve conduction velocities of the tail nerve were not different among the four groups (Figure 5c). The amplitude of the nerve action potential was significantly higher in untreated obese rats compared with the three other groups (Figure 5d), while the duration of the action potential did not differ between groups (data not shown).

Figure 5
Paw withdrawal thresholds measured with the Dynamic Plantar Aesthesiometer (a), tail withdrawal latency in the tail flick test (b), nerve conduction velocity (c) and amplitude of the nerve action potential (d) at the tail nerve of lean (white circles) and obese untreated ZDF rats (black circles) and obese ZDF rats treated with pioglitazone (white squares) or food restriction (black squares) (n = 5–8, mean ± SEM). P < 0.05 versus lean control, ^† P < 0.05 versus obese control. NCV: nerve conduction velocity; AP: action potential

Discussion

Metabolic data of the present study confirm that treatment of prediabetic male ZDF rats with the PPARγ agonist pioglitazone completely prevents development from insulin resistance to overt diabetes due to its insulin-sensitizing effect.^18,19 Pioglitazone-treated animals remain hyperinsulinaemic but normoglycaemic. Food restriction to 70% of the amount consumed by ad libitum fed male obese control rats delayed diabetes development for approximately six weeks, but did not prevent it. This is explicable by the fact that the food amount consumed was still higher than that of lean control rats, in contrast to the study by Ohneda et al.,²⁰ who pair-fed obese ZDF rats to lean controls. Nevertheless, blood glucose and HbA1c values of food-restricted rats were slightly but significantly lower than those of untreated obese rats until the end of the study period. When comparing blood glucose values of untreated and of food-restricted obese rats, it must be taken into account that the food-restricted rats consumed their food in less than 24 h. This means they had a short fasting period at the end of the night, before new food was given, while for control rats, food was available all the time. As blood collection for glucose determination took place in the morning, values of the food-restricted group were slightly lower than what they would have been if the rats had had the possibility to eat all night. This explains why the difference in HbA1c between the two groups was smaller (only about 0.5%) than would have been expected from the difference in glucose levels.

Taken together, metabolic data allow the three groups of obese male ZDF rats to be regarded as a non-diabetic group of fa/fa-genotype (the pioglitazone group), an untreated diabetic group, and a diabetic group with shorter diabetes duration and less severe hyperglycaemia (the food-restriction group). Incidence, progression and severity of nephropathy and neuropathy should then be reduced in the two treatment groups compared with the untreated diabetic group, if the two pathologies were diabetes-related in the male ZDF model.

RGF and UGE augment in parallel with blood glucose levels during diabetes progression, and this was observed in untreated and in food-restricted obese ZDF rats. Lean control rats had constantly low RGF and no UGE, and so had pioglitazone-treated obese rats. The first clinical sign of nephropathy is progressive albuminuria with a rising urinary ACR. Both parameters showed major increase in obese control rats and a milder increase in the food-restriction group, but no change compared with healthy controls in pioglitazone-treated rats. If nephropathy progresses to renal failure, GFR decreases and serum creatinine increases. This did not occur in diabetic ZDF rats during the study period. On the contrary, GFR was higher and serum creatinine lower than in healthy controls. Glomerular hyperfiltration is commonly observed in early diabetic nephropathy^22,23 and is probably linked to poor glycaemic control.²⁴ Maybe osmotic effects of high RGF are involved in this phenomenon. Lower serum creatinine levels of diabetic rats might be explained by increased urinary creatinine loss due to the large urine volume excreted. In addition, obese ZDF rats have less muscle mass than lean ZDF rats, and serum creatinine levels are proportionally influenced by muscle mass. The lack of decrease in GFR and increase in serum creatinine indicate that ZDF rats display only mild nephropathy without progression to renal failure up to 35 weeks of age.

Kidney enlargement is another characteristic of nephropathy²⁵ observed in the untreated diabetic ZDF group. Absolute kidney weight was not increased in the food-restriction group, but correction to body weight showed an increase of the relative kidney weight in this group as well, although it was less pronounced than in the untreated group. Pioglitazone-treated rats had slightly elevated absolute kidney weights, but as their body weight was massively augmented, relative kidney weight was significantly lower than that of healthy controls. Correction to lean body mass would have been more appropriate for comparison, but this parameter has not been determined in the present study.

Histological evaluation of the kidneys revealed no sign of nephropathy in 35-week-old healthy control rats. Untreated diabetic ZDF rats showed pathological changes with an incidence of 100% in glomeruli and tubuli similar to those seen in human diabetic nephropathy. Changes were moderate to marked. In the food-restriction group, the same changes were observed with the same incidence, but only graded as minimal to small. In the non-diabetic group of pioglitazone treatment, glomerulopathy was fully prevented and only slight tubulopathy could be seen in 50% of the animals.

Taken together, these data clearly show that male diabetic ZDF rats develop clinical and histological signs of progressive nephropathy and that this is related to the hyperglycaemic state of the animals, as nephropathy development can be prevented by diabetes prevention and at least slowed down by a delayed onset of diabetes and reductions of blood glucose and HbA1c. Therefore, nephropathy seen in male diabetic ZDF rats can be regarded as a model of human diabetic nephropathy, which is also influenced by the quality of glycaemic control.

For neuropathy, results are far less conclusive. Measurements of sensory thresholds are widely variable between animals and between time points, clearly showing the importance of repeated measurements at intervals of several days when assessing sensory parameters. Only reproducible differences between groups can be considered reliable.

Tactile withdrawal thresholds showed no difference between diabetic obese and non-diabetic lean rats at all. By contrast, pioglitazone-treated non-diabetic obese rats seemed to show transient hypersensitivity from 13 to 21 weeks of age. It remains unclear if this was a random difference, or if it had any (patho-)physiological background. So far, only antinociceptive effects of pioglitazone have been described.^26,27

Although endpoint measurements of thermal withdrawal thresholds were significantly lower for obese diabetic rats compared with lean non-diabetic rats, this difference only became pronounced at the very last measurement for the untreated diabetic group. Food-restricted diabetic rats were most sensitive at most time points, and non-diabetic pioglitazone-treated rats showed thresholds varying between those of untreated diabetic and food-restricted diabetic rats except for the last time point. Thus, the level of hyperglycaemia did not influence tail flick thresholds. If the observed decrease in thermal sensitivity in obese ZDF rats is a sign of neuropathy, this neuropathy is not related to diabetes.

Our results regarding sensory thresholds are partly in contradiction to those of Brussee et al.,¹⁷ Li et al.²⁸ and Piercy et al.,²⁹ who found lower thermal and/or mechanical withdrawal thresholds in diabetic ZDF rats compared with lean non-diabetic controls between nine and 16 weeks of age, and to those of Oltman et al.,^30,31 who showed an increase in thermal withdrawal thresholds in diabetic ZDF rats compared with lean non-diabetic controls at 24 and 28 weeks of age. However, Piercy et al.²⁹ as well as Oltman et al.³⁰ included a treatment group with rosiglitazone, another PPARγ agonist, and did not observe any difference between treated and untreated obese ZDF rats with respect to sensory thresholds. While Oltman et al. initiated rosiglitazone treatment after diabetes manifestation and so could not improve the glycaemic state of the ZDF rats, Piercy et al. started treatment at eight weeks of age in a prediabetic state, keeping the rats euglycaemic throughout the study period. Yet they saw no correction of thermal hypersensitivity in the treated rats. Thus they came, as we did, to the conclusion that altered nociceptive behaviour in obese ZDF rats is independent of their glycaemic status.

Nerve conduction measurements at the tail nerve are frequently used, because the nerve is easily accessible without preparation, and measurements can therefore be repeated over time. Several authors have demonstrated decreased maximum tail nerve conduction velocity in streptozotocin-induced diabetic rats,^32–34 and this decrease has been shown to parallel changes in the sciatic nerve.³⁵ In our study, maximum nerve conduction velocity of the tail nerve was not different between the groups, but amplitude of nerve action potential was. This was the only nerve parameter, where diabetic untreated rats were significantly different from non-diabetic lean controls, and treatment or prevention of diabetes resulted in a significant convergence towards values of non-diabetic controls. However, amplitudes were increased in diabetic untreated ZDF rats, while in neuropathic patients a decrease in action potential amplitudes was observed. The amplitude of a whole nerve action potential is a measure of the number of activated myelinated fibres and is influenced by synchrony of activity. Therefore, a slowing of conduction velocity of distinct fibre populations creating asynchrony will result in decreased peak amplitudes as well as a reduction in the number of activated fibres.³⁶ Accordingly, improved synchrony of conduction velocity or a higher number of myelinated nerve fibres could underlie an amplitude augmentation. Both scenarios would rather reflect improvements of pathologies than their worsening, e.g. in cases where collateral sprouting during nerve regeneration occurs.³⁷ Although the cause for the observed amplitude augmentation remains unclear, it seems improbable that it is a sign of neuropathy in the diabetic ZDF rats.

Summing this up, neither sensory nor electrophysiological data support the assumption that male diabetic ZDF rats develop neuropathy linked to their state of hyperglycaemia. If there are pathologically relevant alterations in the nerves of ZDF rats, which cannot be excluded on the basis of our data, they are independent of the diabetic phenotype of the animals and are therefore not diabetic late complications.

In conclusion, the present study has shown that male obese ZDF rats develop progressive nephropathy causally linked to their diabetic state, and can therefore be seen as a model for human diabetic nephropathy. However, no clear signs of neuropathy could be detected in these animals and no connection to hyperglycaemia was found, rendering the male ZDF rat inappropriate as a model of human diabetic neuropathy. To make sure results of future animal studies on diabetic neuropathy are transferable to the human situation, other models than the male ZDF rat should be used.

	Pioglitazone (n = 6)	Food restriction (n = 8)
Glomerulopathy	100	3.2	0	0	100	1.4
Tubulopathy	100	3.2	50	1.7	100	1.5
Armani-Ebstein cells	100	1.6	0	0	88	1.3

Footnotes

Acknowledgements

We thank Elke Kleinschmidt, Marion Stein, Dagmar Fenner-Nau, Anja Wille, Jörn Wandschneider and Siegmar Kille for excellent technical assistance. We also thank G Biemer-Daub and her collaborators for the analysis of blood and urinary parameters.

References

Friedman

, Friedman

. Is there really good news about pandemic diabetic nephropathy? Nephrol Dial Transplant 2007;22:681–3

Zelmanovitz

, Gerchman

, Balthazar

APS

, Thomazelli

FCS

, Matos

, Canani

. Diabetic nephropathy. Diabetol Metab Syndr 2009;1:10

Tesfaye

, Kempler

. Painful diabetic neuropathy. Diabetologia 2005;48:805–7

Skljarevski

, Malik

. Clinical diagnosis of diabetic neuropathy. In: Veves

, Malik

, eds. Diabetic Neuropathy: Clinical Management. Totowa, NJ: Humana Press Inc, 2007:275–92

The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86

Tesfaye

, Stevens

, Stephenson

, Prevalence of diabetic peripheral neuropathy and its relation to glycaemic control and potential risk factors: the EURODIAB IDDM Complications Study. Diabetologia 1996;39:1377–84

UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53

Akalin

, Berntorp

, Ceriello

, Intensive glucose therapy and clinical implications of recent data: a consensus statement from the Global Task Force on Glycaemic Control. Int J Clin Pract 2009;63:1421–5

Griffen

, Wang

, German

. A genetic defect in beta-cell gene expression segregates independently from the fa locus in the ZDF rat. Diabetes 2001;50:63–8

10.

Mizuno

, Sada

, Kato

, Koike

. Renoprotective effects of blockade of angiotensin II AT1 receptors in an animal model of type 2 diabetes. Hypertens Res 2002;25:271–8

11.

Schäfer

, Linz

, Bube

, Vasopeptidase inhibition prevents nephropathy in Zucker diabetic fatty rats. Cardiovasc Res 2003;60:447–54

12.

Schäfer

, Linz

, Vollert

, The vasopeptidase inhibitor AVE7688 ameliorates type 2 diabetic nephropathy. Diabetologia 2004;47:98–103

13.

Schäfer

, Schmidts

, Bleich

, Busch

, Linz

. Nephroprotection in Zucker diabetic fatty rats by vasopeptidase inhibition is partly bradykinin B2 receptor dependent. Br J Pharmacol 2004;143:27–32

14.

Figarola

, Loera

, Weng

, Shanmugam

, Natarajan

, Rahbar

. LR-90 prevents dyslipidaemia and diabetic nephropathy in the Zucker diabetic fatty rat. Diabetologia 2008;51:882–891

15.

Shimoshige

, Ikuma

, Yamamoto

, The effects of zenarestat, an aldose reductase inhibitor, on peripheral neuropathy in Zucker diabetic fatty rats. Metabolism 2000;49:1395–9

16.

Oltman

, Coppey

, Gellett

, Davidson

, Lund

, Yorek

. Progression of vascular and neural dysfunction in sciatic nerves of Zucker diabetic fatty and Zucker rats. Am J Physiol Endocrinol Metab 2005;289:E113–22

17.

Brussee

, Guo

, Dong

, Distal degenerative sensory neuropathy in a long-term type 2 diabetes rat model. Diabetes 2008;57:1664–73

18.

Smith

, Lister

, Toseland

CDN

, Buckingham

. Rosiglitazone prevents the onset of hyperglycaemia and proteinuria in the Zucker diabetic fatty rat. Diabetes Obes Metab 2000;2:363–72

19.

Bergeron

, Yao

, Woods

, Peroxisome proliferator-activated receptor (PPAR)-alpha agonism prevents the onset of type 2 diabetes in Zucker diabetic fatty rats: a comparison with PPAR gamma agonism. Endocrinology 2006;147:4252–62

20.

Ohneda

, Inman

, Unger

. Caloric restriction in obese pre-diabetic rats prevents beta-cell depletion, loss of beta-cell GLUT 2 and glucose incompetence. Diabetologia 1995;38:173–9

21.

Bergmeyer

. Methoden der enzymatischen Analyse. Weinheim: Verlag Chemie, 1974

22.

Silveiro

, Friedman

, Gross

. Glomerular hyperfiltration in NIDDM patients without overt proteinuria. Diabetes Care 1993;16:115–19

23.

Caramori

, Gross

, Pecis

, de Azevedo

. Glomerular filtration rate, urinary albumin excretion rate, and blood pressure changes in normoalbuminuric normotensive type 1 diabetic patients: an 8-year follow-up study. Diabetes Care 1999;22:1512–16

24.

Wiseman

, Saunders

, Keen

, Viberti

. Effect of blood glucose control on increased glomerular filtration rate and kidney size in insulin-dependent diabetes. N Engl J Med 1985;312:617–21

25.

Rigalleau

, Garcia

, Lasseur

, Large kidneys predict poor renal outcome in subjects with diabetes and chronic kidney disease. BMC Nephrol 2010;11:3

26.

Maeda

, Kiguchi

, Kobayashi

, Ozaki

, Kishioka

. Pioglitazone attenuates tactile allodynia and thermal hyperalgesia in mice subjected to peripheral nerve injury. J Pharmacol Sci 2008;108:341–7

27.

Oliveira

ACP

, Bertollo

, Rocha

LTS

, Nascimento

Jr , Costa

, Coelho

. Antinociceptive and antiedematogenic activities of fenofibrate, an agonist of PPAR alpha, and pioglitazone, an agonist of PPAR gamma. Eur J Pharmacol 2007;561:194–201

28.

, Abatan

, Kim

, Taurine reverses neurological and neurovascular deficits in Zucker diabetic fatty rats. Neurobiol Dis 2006;22:669–76

29.

Piercy

, Banner

, Bhattacharyya

, Thermal, but not mechanical, nociceptive behavior is altered in the Zucker Diabetic Fatty rat and is independent of glycemic status. J Diabetes Complications 1999;13:163–9

30.

Oltman

, Davidson

, Coppey

, Vascular and neural dysfunction in Zucker diabetic fatty rats: a difficult condition to reverse. Diabetes Obes Metab 2008;10:64–74

31.

Oltman

, Davidson

, Coppey

, Kleinschmidt

, Yorek

. Treatment of Zucker diabetic fatty rats with AVE7688 improves vascular and neural dysfunction. Diabetes Obes Metab 2009;11:223–33

32.

Andersen

, Nielsen

. Inability of insulin to maintain normal nerve function during high-frequency stimulation in diabetic rat tail nerves. Muscle Nerve 1994;17:80–4

33.

Hotta

, Kakuta

, Fukasawa

, Effect of a potent new aldose reductase inhibitor, (5-(3-thienyltetrazol-1-yl)acetic acid (TAT), on diabetic neuropathy in rats. Diabetes Res Clin Pract 1995;27:107–17

34.

Remuzzi

, Cornolti

, Bianchi

, Regression of diabetic complications by islet transplantation in the rat. Diabetologia 2009;52:2653–61

35.

Kasselman

, Veves

, Gibbons

, Rutkove

. Cold exposure exacerbates the development of diabetic polyneuropathy in the rat. Exp Diabetes Res 2009;2009:827943

36.

Arezzo

, Zotova

. Electrophysiologic measures of diabetic neuropathy: mechanism and meaning. Int Rev Neurobiol 2002;50:229–55

37.

Krarup

. Compound sensory action potential in normal and pathological human nerves. Muscle Nerve 2004;29:465–83

	Lean control (n = 5)		Obese control (n = 5)		Pioglitazone (n = 6)		Food restriction (n = 8)
	Incidence (%)	Score	Incidence (%)	Score	Incidence (%)	Score	Incidence (%)	Score
Glomerulopathy	0	0	100	3.2	0	0	100	1.4
Tubulopathy	0	0	100	3.2	50	1.7	100	1.5
Armani-Ebstein cells	0	0	100	1.6	0	0	88	1.3

Appropriateness of the Zucker Diabetic Fatty rat as a model for diabetic microvascular late complications

Abstract

Keywords

Materials and methods

Animals

Experimental design

Blood parameters

Urinary parameters

Analytical procedures of blood and urinary parameters

Sensory thresholds and nerve conduction

Kidney histology

Statistical analysis

Results

Metabolic parameters

Kidney function and histology

Nerve function

Discussion

Footnotes

Acknowledgements

References