C-peptide prevents glomerular hypertrophy and mesangial matrix expansion in diabetic rats

NDT Advance Access published January 21, 25 Nephrol Dial Transplant (25) 1 of 7 doi:1.193/ndt/gfh683 Original Article C-peptide prevents glomerular hypertrophy and mesangial matrix expansion in diabetic rats Bjo rn Samnegård 1,4, Stefan H. Jacobson 2, Georg Jaremko 3, Bo-Lennart Johansson 4, Karin Ekberg 4, Britta Isaksson 5, Linda Eriksson 5, John Wahren 4 and Mats Sjo quist 5 1 Department of Nephrology, Danderyd Hospital, Departments of 2 Nephrology, 3 Pathology and 4 Clinical Physiology, Karolinska University Hospital, Stockholm and 5 Department of Cell Biology, Biomedicum, Uppsala University, Sweden Abstract Background. There is accumulating evidence that C-peptide exerts beneficial renal effects in type-1 diabetes by reducing glomerular hyperfiltration, albuminuria and glomerular hypertrophy in the early stage of nephropathy. The aim of this study was to clarify further the effects of C-peptide on renal structural changes in type-1 diabetic rats. Methods. The effects of C-peptide or placebo on glomerular volume, mesangial expansion, glomerular basement membrane thickness, albuminuria and glomerular filtration rate (GFR) were studied in three groups of rats: a non-diabetic group (N, n ¼ 9) and two groups that, during 8 weeks of diabetes, were left untreated for 4 weeks and then given a subcutaneous infusion of either placebo (D, n ¼ 11) or C-peptide (DCp, n ¼ 11) during the next 4 weeks. Furthermore, GFR was studied after 4 weeks of diabetes in an additional diabetic group (D-early, n ¼ 9) and in an age-matched non-diabetic group (N-early, n ¼ 9). Results. After 4 weeks, GFR in the D-early group was 12% higher than in the N-early group. GFR after 8 weeks did not differ between the study groups. The D group presented with a 33% larger glomerular volume than the N group (P<.1), while glomerular volume in the DCp group was similar to that in the N-group. Total mesangial and mesangial matrix fractions were increased by 46% (P<.1) and 133% (P<.1), respectively, in the D group. The corresponding values in the DCp group did not differ from those for the non-diabetic animals. Neither the thickness of the glomerular basement membrane nor the level of albuminuria differed significantly between the study groups. Conclusions. C-peptide administration in replacement dose to streptozotocin-diabetic rats serves to limit or Correspondence and offprint requests to: Bjo rn Samnegård, MD, Department of Nephrology, Danderyd Hospital, SE-182 88 Stockholm, Sweden. Email: bjorn.samnegard@medks.ki.se prevent the glomerular hypertrophy and the mesangial matrix expansion seen in the post-hyperfiltration phase of early diabetic nephropathy. Keywords: C-peptide; diabetic nephropathy; glomerular basement membrane thickness; glomerular hypertrophy; glomerular hyperfiltration; mesangial expansion Introduction Patients with type-1 diabetes lack both insulin and C-peptide. The latter is synthesized in and released from the pancreatic b-cells in equimolar amounts with insulin. In recent years, it has been shown that, contrary to previous views, C-peptide has biological effects of its own. Thus, studies in type-1 diabetic patients with incipient nephropathy have revealed beneficial effects of C-peptide on glomerular hyperfiltration, a proposed risk factor for the development of diabetic nephropathy, and on urinary albumin excretion [1,2]. Furthermore, both an acute and a 2 week infusion of C-peptide to streptozotocin-diabetic rats have resulted in prevention of hyperfiltration and in reduced albuminuria [3,4]. Recent findings indicate that C-peptide and the angiotensin-converting enzyme inhibitor captopril may be equally effective in lowering glomerular hyperfiltration in type-1 diabetic rats [5]. The exact mechanism behind the beneficial effects of C-peptide on renal function in type-1 diabetes is not fully understood. Specific binding of C-peptide to cell membranes has been demonstrated for various cell types, including rat renal tubular and mesangial cells, and there are indications that the binding site is a G-protein-coupled receptor [6]. Binding leads to an increased intracellular Ca 2þ concentration and subsequently to stimulation of Na þ,k þ -ATPase [7] and endothelial nitric oxide synthase (enos) activities [8]. Nephrol Dial Transplant ß ERA EDTA 25; all rights reserved

2 of 7 B. Samnegård et al. However, the relationship between these events and the beneficial effects on renal function and glomerular morphology remain to be clarified. The morphological changes in diabetic nephropathy are characterized by mesangial matrix expansion and glomerular basement membrane (GBM) thickening [9]. Since little information is available on renal morphology during C-peptide treatment in diabetes, the main objective of the present study was to investigate the effects of C-peptide on glomerular volume, mesangial expansion, GBM thickening, albuminuria and glomerular filtration rate (GFR) in streptozotocin diabetic rats. Materials and methods measurements of the arterial blood pressure. The bladder was catheterized by a suprapubic approach. Thereafter, an infusion of isotonic saline containing [ 3 H]inulin (Pharmacia AB, Uppsala, Sweden) 4 mg/ml was started. After a bolus dose of 1 ml (5 mci), the infusion rate was maintained at 5 ml/h/kg body weight. At steady state, after 45 min, three urine samples were collected at 2 min intervals for analyses of urine volume, osmolality and concentrations of sodium, potassium and [ 3 H]inulin. At the midpoint of each of the three 2 min urine collection periods, plasma samples (6 ml) for analyses of [ 3 H]inulin were obtained to allow calculation of the GFR. When the GFR measurements had been completed, a blood sample for determination of the C-peptide concentration was taken. To ensure adequate renal perfusion, the sample volume was replaced by an equal volume of isotonic saline. Eight-week-old male Wistar rats (Møllegaard, Copenhagen, Denmark) with an initial weight of 25 g were divided into three groups and studied for 8 weeks: non-diabetic placebotreated rats (N group, n ¼ 9), diabetic placebo-treated rats (D group, n ¼ 11) and diabetic rats treated with rat C-peptide II (Genosys Biotechnologies, UK) (DCp group, n ¼ 11). In two additional groups, designated normal early group (N-early, n ¼ 9) and diabetic early group (D-early, n ¼ 7), the GFR was studied after 4 weeks for comparison with the three groups above. All animals had free access to tap water and standardized chow (R36, Lactamin, Ewos, So derta lje, Sweden) throughout the study period. Diabetes was induced by intravenous injection of streptozotocin 6 mg/kg body weight. Treatment with C-peptide (5 pmol/kg/min) dissolved in isotonic saline in the DCp group or with saline alone in the other two main study groups was initiated 4 weeks after onset of diabetes and administered as a continuous subcutaneous infusion for 4 weeks by an osmotic pump (type 22, Alzet, USA) placed in the subcutaneous tissue of the neck. The study protocol was reviewed and approved by the institutional animal ethics committee. Metabolic cages Once a week, the rats were put in metabolic cages (Techniplast Gazzada 371MO-, Buguggiate, Italy) for 24 h. Body weight, daily intake of water and food, and excretion of urine and faeces were measured individually. Urine samples were collected for analyses of albumin, sodium and potassium excretion and osmolality. Blood samples for blood glucose measurements were taken from the tip of the tail. Determination of GFR GFR was measured in the N-early and D-early groups after 4 weeks of diabetes. In the other study groups, GFR was measured after 8 weeks of diabetes, i.e. after 4 weeks without and 4 weeks with treatment with C-peptide or placebo. The rats were anaesthetized by an intraperitoneal injection of Inactin Õ (sodium 5-sec-butyl-5-ethyl-2-thiobarbiturate; RBI Natick, MA), 7 mg/kg body weight, and placed on a heating pad to maintain body temperature at 37.5 C. Spontaneous breathing was facilitated by a cannula inserted into the trachea. The femoral vein was cannulated for infusions. The femoral artery was cannulated for blood sampling and for Fixation of the kidney The left kidney, the renal vessels and aorta were exposed via a subcostal incision. Thereafter, the femoral artery catheter was adjusted to the level of the left renal artery and the aorta subsequently was ligated proximally and distally to the left renal artery. The left kidney was then perfused with 4% paraformaldehyde in.1 M phosphate buffer (ph 7.4) at a pressure adjusted to the mean arterial blood pressure recorded just before the perfusion was started. For electron microscopy, cortical tissue blocks were fixed with 2.5% glutaraldehyde in.1 M phosphate buffer (ph 7.4). Analyses Urine volumes were measured gravimetrically. Urinary sodium and potassium concentrations were determined by flame photometry (IL 543; Instrumentation Lab., Milano, Italy) and urine osmolality by a freezing point depression method (Model 3 MO; Advanced Instruments, MA). Urine albumin concentration was measured by nephelometry. Blood glucose concentrations were analysed by means of Accutrend Õ (Boehringer Mannheim GmbH, Mannheim, Germany) when obtained from the rat s tail and by a glucose oxidase method (Glucose Analyzer, Yellow Spring Instruments, USA) when obtained at the end of the final experiment. [ 3 H]Inulin in plasma and urine was determined by liquid scintillation counting (PW 47, Philips, The Netherlands). The sample (1 ml of urine or 1 ml of plasma) was mixed in 1 ml of water; thereafter 3 ml of scintillation fluid (Pico-Flour 4 TM, CiAB, Chemical Instruments AB, Lidingo, Sweden) was added. A radioimmunoassay technique (Linco Research Inc., USA) was used to measure the plasma concentrations of rat C-peptide. Light microscopy The mean glomerular volume was estimated by Cavalieri s principle as described in an earlier study [3]. Electron miscroscopy For electron microscopy, cortical tissue was post-fixed in 1% osmium tetroxide and embedded in Epon by standard procedures. Ultrathin sections were stained with

Effects of C-peptide on renal morphology 3 of 7 uranylacetate and lead citrate and studied in a JEM 1S electron microscope (Jeol, Tokyo, Japan). From each rat, 4 5 glomeruli were analysed. The reference space of the glomerular tuft was defined as in Osterby and Gundersen [1]. At 3, sets of 8 14 micrographs per glomerulus were taken in a systematic random manner by moving the specimen stage between predetermined points. The mesangial volume fraction and the mesangial matrix volume fraction were analysed by point counting, using a superimposed square lattice grid with 3 mm between the points at tissue level. The basement membrane thickness was estimated using the orthogonal intercept method of Jensen et al. [11] on a separate set of 6 9 micrographs taken from each glomerulus at 75. The final magnification was corrected using a grating grid with 216 lines per mm. group differed significantly from the normal group in all variables measured except for urine albumin excretion and urine osmolality. There were no significant differences in these variables between the D and DCp groups. The average blood glucose levels during weeks 1 4 (before C-peptide or placebo treatment) and during weeks 5 8 (during treatment) are shown in Table 1. The C-peptide treatment did not affect the glucose levels. In the DCp group, the plasma C-peptide concentration was 1.6±.2 nmol/l, not differing significantly from that in the non-diabetic group (1.7±.2 nmol/l; Table 1). In contrast, the diabetic group not receiving C-peptide showed a C-peptide level of only.1± nmol/l, as expected after streptozotocin treatment (Table 1). Statistical methods Differences in metabolic data, glomerular volume, renal size, GFR, mesangial volume and GBM thickness were evaluated by ANOVA followed by Tukey s post hoc test. Albuminuria (after the values had been logarithmically transformed) was compared with the basal state using ANOVA for repeated measurements within each group followed by Tukey s post hoc test. Statistical significance is defined as a P-value <.5. Data in text, figures and tables are presented as the mean±sem. Results Data from the metabolic cages Before streptozotocin treatment, there were no statistically significant differences between the study groups in body weight, water intake, urine volume, food intake, faeces excretion, urine osmolality or urinary sodium and potassium excretion. Table 1 shows the results from the metabolic cages recorded at 4 and 8 weeks after induction of diabetes. At 8 weeks, the DCp Table 1. Data from the metabolic cages and C-peptide levels Glomerular filtration rate The results indicate that after 4 weeks of diabetes, GFR was on average 12% greater in the diabetic compared with the non-diabetic group (Figure 1): D-early 5.23±.78 ml/min and N-early 2.59±.37 ml/min (P<.1). Results for the other groups show that after 8 weeks of diabetes, there was no longer a statistically significant increase in GFR, although GFR still tended to be higher in the diabetic untreated group (P<.51). Thus, GFR in the D group was 3.39±.24 ml/min and in the N group 2.47±.8 ml/min. GFR in the DCp group was 2.77±.31 ml/min, not significantly different from the N group. Glomerular volume The average glomerular volume was 1.34±.3 1 6 mm 3 in the D group, or 33% greater than in the N group (1.1±.5 1 6 mm 3, P<.1) (Figure 2). The corresponding value for the DCp group (1.9±.2 1 6 mm 3 ) was similar to that of the non-diabetic Group N D DCp Time Week 4 Week 8 Week 4 Week 8 Week 4 Week 8 Body weight (g) 316±7 362±7 257±7 c 25±7 c 25±6 c 241±7 c Water intake (ml/day) 2±2 21±2 167±9 c 15±8 c 156±1 c 13±12 c Urine volume (ml/day) 12±1 13±2 151±9 c 134±8 c 14±11 c 12±12 c Food intake (g/day) 14±1 12±1 33±1 c 3±1 c 31±1 c 27±2 c Faeces (g/day) 6±1 6±1 17±1 c 16±1 c 16±2 c 14±1 b U-osmolality (mosm/kg) 121±131 187±87 889±16 b 887±15 a 93±33 b 917±25 U-Na (mmol/day) 1.5±.1 1.3±.1 2.9±.2 c 3.1±.2 c 3.1±.3 c 2.9±.2 c U-K (mmol/day 1.9±.1 1.9±.1 3.5±.3 b 3.±.2 b 3.1±.3 a 2.7±.2 a U-albumin (mg/day) 19±23 239±67 546±133 a 591±212 441±126 698±113 Blood glucose (mmol/l) 5.7±.2 5.7±.2 24±.8 c 24.5±.4 c 25.4±.6 c 24.8±.4 c Plasma C-peptide (nmol/l) 1.7±.2.1± c 1.6±.2 Data from the metabolic cages 4 and 8 weeks after diabetes onset except for blood glucose, which is an average of the first 4 and last 4 weeks, respectively, and C-peptide, which is measured at the end of the study. Non-diabetic placebo-treated group (N, n ¼ 9), diabetic placebo-treated group (D, n ¼ 11), diabetic C-peptide-treated rats (DCp, n ¼ 11). ANOVA followed by Tukey s post hoc test was used for the statistical analyses. The superscripts a, b and c indicate significant differences from the N group at the same time point: a P<.5, b P<.1 and c P<.1.

4 of 7 B. Samnegård et al. animals and 2% lower than that of the D group (P<.1). Mesangial volume, mesangial matrix volume and interstitial tissue The mesangial volume expressed as a fraction of the glomerular volume was 12.3±.7% in the N group, 18.±.9% in the D group and 14.2±.7% in the DCp group (Figures 2 and 3). Thus, the D group showed a GFR (ml/min) 1 6 µm 3 7 6 5 4 3 2 1 1,5 1,5 Glomerular volume GFR N-early D-early N D DCp Fig. 1. Glomerular filtration rate (GFR) measured as the clearance of inulin. In the first two groups, i.e. the normal early (N-early, n ¼ 9) and the diabetic early (D-early, n ¼ 7) groups, GFR was measured 4 weeks after diabetes onset without any treatment. In the main study groups, i.e. the normal placebo-treated group (N group, n ¼ 9), the diabetic placebo-treated group (D group, n ¼ 11) and the diabetic C-peptide-treated group (DCp, n ¼ 11), GFR was measured 8 weeks after diabetes onset, thus, 4 weeks after start of treatment. Asterisks indicate statistical difference compared with the N-early group after 4 weeks and the N group after 8 weeks: P<.1. N D DCp % of glomerular volume 1 8 6 4 2 mesangial volume fraction that was 46% larger than in the N group (P<.1) and 27% larger than in the DCp group (P<.1). The DCp group did not differ significantly from the normal group. The relationships for the mesangial matrix volume fraction were similar to those for the total mesangial volume fraction. Thus, the mesangial matrix volume expressed as a fraction of the total glomerular volume was 3.±.3% in the N group, 7.±.5% in the D group and 4.4±.6% in the DCp group (Figure 2). The D group showed a mesangial matrix fraction that was more than twice that in the N group (P<.1) and 59% larger than in the DCp group (P<.1). The DCp group did not differ significantly from the N group. The mesangial cell volume fraction (data not shown) of the glomerular volume did not differ significantly between the study groups. There were no signs of interstitial expansion, fibrosis or tubular atrophy in any of the study groups. Glomerular basement membrane thickness The thickness of the GBM did not differ significantly between the study groups. Thus, GBM thickness was 147±2 nm in the N group, 156±6 nm in the D group and 153±2 nm in the DCp group (Figure 2). Albuminuria The urinary albumin excretion rate in the healthy N group after 4 and 8 weeks was 19±23 and 239± 67 mg/day, respectively (Table 1). In the D group, the albumin excretion was 546±133 mg/day (P<.5 vs the N group) after 4 weeks and 591±212 mg/day after 8 weeks (NS vs the N group). The DCp group showed Mesangial matrix volume N D DCp % of glomerular volume 2 15 1 5 Mesangial volume nm GBM thickness N D DCp N D DCp 2 15 1 5 Fig. 2. Glomerular volume, mesangial volume fraction, mesangial matrix volume fraction and glomerular basement membrane thickness after 8 weeks (4 weeks after treatment start) in each study group: the normal placebo-treated group (N group, n ¼ 9), the diabetic placebotreated group (D group, n ¼ 11) and the diabetic C-peptide-treated group (DCp, n ¼ 11). ANOVA followed by Tukey s post hoc test was used for the statistical analysis. Asterisks indicate statistically significant differences from the N group: P<.1.

Effects of C-peptide on renal morphology 5 of 7 Fig. 3. PAS-stained glomeruli of normal placebo treated rat (N group, A), diabetic placebo-treated rat (D group, B) and C-peptide-treated rat (DCp group, C). Note the widened mesangial areas (arrows) in the D rat (B) containing PAS-positive mesangial matrix. This matrix expansion is partially prevented in the DCp rat (C). an albumin excretion of 441±126 and 698±113 mg/day after 4 and 8 weeks, respectively (both NS vs the normal group). Discussion Evidence is accumulating that C-peptide is a bioactive peptide that can exert beneficial effects on the kidney in type-1 diabetes, resulting in prevention or retardation of diabetic nephropathy [1 5]. Previous studies have established that C-peptide can limit glomerular hypertrophy in type-1 diabetic rats [3]. The glomerular morphology in diabetic nephropathy is characterized primarily by mesangial expansion and GBM thickening. The effects of C-peptide on the specific renal structural components in type-1 diabetes have not been examined. Consequently, this study was undertaken to explore further the effects of C-peptide on diabetes-induced renal morphological changes. GFR was measured after 4 weeks in diabetic (D-early) and non-diabetic rats (N-early). Characteristic glomerular hyperfiltration was found in the diabetic group at this early stage. After 8 weeks, GFR in the placebo-treated group of diabetic rats (D group) was no longer statistically significantly increased compared with the non-diabetic group (N group), indicating that after 8 weeks of diabetes the hyperfiltration phase was over (Figure 1). This is the expected course of nephropathy in diabetes and at this stage the morphological changes usually become more established [12]. The absence of significant glomerular hyperfiltration in the diabetic rats after 8 weeks explains why C-peptide, in this study and in contrast to our earlier studies in diabetic rats [3 5], showed no statistically significant effect on the GFR. Glomerular hyperfiltration and glomerular hypertrophy usually occur simultaneously, but the hypertrophy can persist after the GFR has been normalized [13]. Accordingly, after 8 weeks and a substantial decrease in GFR, the D group still had significant glomerular hypertrophy. The glomerular volume in the D group was 33% larger than in the N group. In the DCp group, an increase in glomerular volume was prevented by the continuous infusion of physiological doses of C-peptide (Figure 2). This is in keeping with earlier findings in streptozotocin-diabetic rats [3]. In the present study, the increased glomerular size is explained in part by a 46% larger (P<.1) mesangial volume fraction in the D group compared with the N group (Figure 2). Specifically, electron microscopy of the glomeruli reveals that the mesangial expansion is due to an increased matrix volume (Figure 2) rather than an increased cell volume. In contrast, the C-peptidetreated group showed no or minimal mesangial expansion. This is a finding of interest since matrix accumulation is related to clinical manifestations of diabetic nephropathy and may cause occlusion of the

6 of 7 B. Samnegård et al. nephrons [14]. In this context, it is noted that a recent study in streptozotocin-diabetic mice has demonstrated that diabetes-induced upregulation of transforming growth factor-b (TGF-b) in glomeruli is prevented by C-peptide administration. Moreover, in vitro studies of mouse podocytes show that C-peptide dosedependently inhibits TGF-b-induced upregulation of collagen IV [15]. These observations may provide insight into the mechanism whereby C-peptide contributes to diminished mesangial matrix expansion in the diabetic state. The present findings are in keeping with the observation that after pancreas transplantation in humans, mesangial matrix expansion is reversed [16]. This amelioration has been ascribed to improved glycaemic control. However, it has also been suggested that the finding may be attributed to restoration of C-peptide levels after transplantation [17]. Furthermore, improved graft survival rates and diminished urinary albumin excretion have been observed after successful islet transplantation [17], again results that may be supported by earlier findings regarding C-peptide and albumin excretion [1,2]. The mechanism underlying the mesangial matrix accumulation in type-1 diabetes is likely to be multifactorial, including an imbalance between production and degradation of the matrix components collagen IV and fibronectin [18]. Growth factors such as plateletderived growth factor, TGF-b and vascular endothelial growth factor are likely to contribute to this process. Non-enzymatic glycosylation and formation of advanced glycation end-products tend to increase matrix protein deposition and render the glycated matrix component less susceptible to proteolysis [18]. To what extent C-peptide affects these variables is not known. GBM thickening is another hallmark of diabetic nephropathy, although it is not as closely related to the clinical findings as mesangial expansion [14]. In this study, GBM thickness was not significantly increased in the diabetic placebo-treated group compared with the non-diabetic group (Figure 2), and in the DCp group it did not differ significantly from the other study groups. It is conceivable that a longer study period is required in order to elicit GBM thickening in Wistar rats; the time point for its occurrence varies greatly in the literature (from 8 weeks to 6 months [19]). Interstitial expansion, fibrosis and tubular atrophy have been proposed as important factors in predicting subsequent progression to renal failure. However, in this material, there were no such changes in any of the study groups. Thus, at this early stage of experimental diabetes, glomerular changes are the only assessable indicators of renal damage. This supports the view that the interstitial changes in diabetic nephropathy are associated with a more advanced stage of glomerulopathy [2] than studied in the present report. In contrast to previous observations in both animals and humans [2 4], the albumin excretion rate in the diabetic groups was not significantly increased compared with the normal group, nor was there an effect of C-peptide on this variable. A possible explanation for these findings may be the large dispersion in the urinary albumin excretion data. In addition, the absence of C-peptide treatment during the first 4 weeks of diabetes may have caused damage to the filtration barrier that was too severe to be reversed in only 4 weeks of subsequent C-peptide therapy. Furthermore, in earlier animal studies of C-peptide effects in diabetes, Sprague Dawley rats rather than Wistar rats were used and the two strains may differ in the relationship between the morphological changes and albuminuria. In summary, administration of C-peptide to physiological levels for 4 weeks results in almost complete prevention of the glomerular hypertrophy and mesangial matrix expansion that otherwise occur in streptozotocin-diabetic Wistar rats. Conflict of interest statement. B.L.J., J.W. and K.E. own shares in Creative Peptides Sweden Inc., Stockholm, and J.W. and K.E. are employed by this company. References 1. Johansson B-L, Sjo berg S, Wahren J. The influence of human C-peptide on renal function and glucose utilization in type 1 (insulin-dependent) diabetic patients. Diabetologia 1992; 35: 121 128 2. Johansson B-L, Borg K, Fernqvist-Forbes E, Kernell A, Odergren T, Wahren J. Beneficial effects of C-peptide on incipient nephropathy and neuropathy in patients with type 1 diabetes a three-month study. Diabetic Med 2; 17: 1 9 3. Samnegard B, Jacobson SH, Jaremko G, Johansson BL, Sjoquist M. Effects of C-peptide on glomerular and renal size and renal function in diabetic rats. Kidney Int 21; 6: 1258 1265 4. Sjo quist M, Huang W, Johansson B-L. Effects of C-peptide on renal function at the early stage of experimental diabetes. Kidney Int 1998; 54: 758 764 5. Samnegard B, Jacobson SH, Johansson BL et al. C-peptide and captopril are equally effective in lowering glomerular hyperfiltration in diabetic rats. Nephrol Dial Transplant 24; 19: 1385 1391 6. Rigler R, Pramanik A, Jonasson P et al. Specific binding of proinsulin C-peptide to human cell membranes. Proc Natl Acad Sci USA 1999; 96: 13318 13323 7. Ohtomo Y, Aperia A, Sahlgren B, Johansson B-L, Wahren J. C-peptide stimulates rat renal tubular Na þ,k þ, ATPase activity in synergism with neuropeptide Y. Diabetologia 1996; 39: 199 25 8. Wallerath T, Kunt T, Forst T et al. Stimulation of endothelial nitric oxide synthase by proinsulin C-peptide. Nitric Oxide 23; 9: 95 12 9. Osterby R. Renal pathology in diabetes mellitus. Curr Opin Nephrol Hypertens 1993; 2: 475 483 1. Osterby R, Gundersen HJ. Fast accumulation of basement membrane material and the rate of morphological changes in acute experimental diabetic glomerular hypertrophy. Diabetologia 198; 18: 493 5 11. Jensen EB, Gundersen HJ, Osterby R. Determination of membrane thickness distribution from orthogonal intercepts. J Microsc 1979; 115: 19 33 12. Mogensen CE. How to protect the kidney in diabetic patients: with special reference to IDDM. Diabetes 1997; 46 [Suppl 2]: S14 111 13. Osterby R, Asplund J, Bangstad HJ et al. Glomerular volume and the glomerular vascular pole area in patients with insulin-dependent diabetes mellitus. Virchows Arch 1997; 431: 351 357

Effects of C-peptide on renal morphology 7 of 7 14. Mauer SM, Steffes MW, Ellis EN, Sutherland DE, Brown DM, Goetz FC. Structural functional relationships in diabetic nephropathy. J Clin Invest 1984; 74: 1143 1155 15. Maezawa Y, Yokote K, Sonezaki K et al. C-peptide can inhibit early glomerular changes in diabetic mice. Diabetes 24; 55 [Suppl 2]: A22 16. Fioretto P, Steffes MW, Sutherland DE, Goetz FC, Mauer M. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Engl J Med 1998; 339: 69 75 17. Fiorina P, Folli F, Zerbini G et al. Islet transplantation is associated with improvement of renal function among uremic patients with type I diabetes mellitus and kidney transplants. J Am Soc Nephrol 23; 14: 215 2158 18. Tsilibary EC. Microvascular basement membranes in diabetes mellitus. J Pathol 23; 2: 537 546 19. Macedo CS, Silva MD, Spadella CT et al. Effect of long-term treatment with insulin and/or acarbose on glomerular basement membrane thickening in alloxan-diabetic rats. Braz J Med Biol Res 1996; 29: 1329 1335 2. Fioretto P, Steffes MW, Sutherland DE, Mauer M. Sequential renal biopsies in insulin-dependent diabetic patients: structural factors associated with clinical progression. Kidney Int 1995; 48: 1929 1935 Received for publication: 26.5.4 Accepted in revised form: 1.12.4